as a PDF
Transcription
as a PDF
Volume 26, No. 6, 2005 Edited by • Édité par Dr. Johann Vallerand Université de Moncton, New Brunswick Strategy Stratégie Actes de Congres annuel de la section Stratégie de l’association des sciences administrative du Canada Proceedings of the Annual Conference of the Administrative Sciences Association of Canada, Strategy Division Toronto (Ontario) 28-31 mai, 2005 Toronto (Ontario) May 28-31, 2005 Des exemplaires de ces Actes peuvent être obtenus auprès de: Copies of these Proceedings may be ordered from: Dr. Nina Cole Faculty of Business Ryerson University 350 Victoria Street Toronto, Ontario M5B 2K3 Tel: 416-979-5000 x7558 Fax: 416-979-5266 Email: [email protected] ACKNOWLEDGEMENTS REMERCIEMENTS The proceedings and program of the Strategy Les actes de la division Stratégie de ASAC Division, ASAC 2005, are the result of many 2005 sont le résultat des efforts de plusieurs personnes. people’s efforts. Sincere thanks to the authors for the quality of their submissions and to the referees for their well-considered comments. Thank you also to those who agreed to serve as chairs and discussants at the conference. Je remercie les auteurs pour la qualité de leurs soumissions, et les évaluateurs pour leurs commentaires bien pensées. Merci aussi à ceux qui ont d’agir à titre de commentateur ou de président de session au congrès. I am grateful to the current (and several former) Strategy division officers for their guidance and encouragement, particularly Elizabeth Croft and Parshotam Dass. Aux officiers de la division présente et passée, tout particulièrement Elizabeth Croft et Parshotam Dass vos conseils et encouragement ont été d’une très grande valeur. I wish to acknowledge the support of the Business faculty, Université de Moncton, and Je tiens à remercier la Faculté d’administration the hard work and patience of my student de l’Université de Moncton pour leur soutien, et mon assistant étudiant Denis Ouellette pour assistant, Denis Ouellette. sa patience et son bon travail. Johann Vallerand Johann Vallerand Academic Reviewer Responsable de l’évaluation Faculté d’administration Faculté d’administration Université de Moncton Université de Moncton New Brunswick Nouveau Brunswick ii REVIEWERS-ÉVALUATEURS Samia Belaounia Sylvie Berthelot Frances Bowen Michele Bowring Shamsud Chowdhury Jacqueline Dahan Parshotam Dass Robert H. Desmanteau Carole Donada Karen Ekstein Salma Fardie Ellovze Denise Graham Irene Henriques Jack Ito Joyce Jin Mouloud Khelif Preeti Krishnan Juliano Lissoni Reg Litz Ebrahimi Mahrau Martin Martens Micheal Mayo Jane McKay-Nesbitt Agnes Meinhard Na Ni Marielle Payaud John Phillips Patrizia Porrin Laurent Renard Suhaib Riaz Glen Rowe Richard Sopranot Eric Stevens Carol Ann Tetrault Sirsly Norlaine Thomas Jacques Vallerand Johann Vallerand Huanglin Wang Taiyuan Wang Jaana Woiceshyn David di Zhang Zhou Zhang Rouen Graduate Business School Université de Moncton University of Calgary University of Manitoba Dalhousie University HEC Montreal University of Manitoba Université du Québec Montreal ESSEC, France York University Université El Manar, Tunisie Bristol Business School, UK/Uni. of Windsor York University University of Regina Queen’s University HEC Montreal University of Manitoba Universidade Federal de Santa Catarina University of Manitoba EURISTIK (IAE-Université de Lyon) Concordia University Athabasca University University of Manitoba Ryerson University University of Manitoba EURISTIK (IAE-Université de Lyon) University of Western Ontario Long Island University Université du Québec Montreal University of Western Ontario University of Western Ontario Groupe ESCEM Tours-Poitiers, France Groupe ESCEM Tours-Poitiers, France Concordia University University of Manitoba Université Laval Université de Moncton University of Western Ontario University of Western Ontario University of Calgary University of Manitoba University of Manitoba iii TABLE OF CONTENTS / TABLE DES MATIÈRES The Behavioral Theory of Strategic Decision Making: Building on Greve’s empirical studies................................................................................................................................. 1 Frances E. Bowen (University of Calgary) Mahdi Rostami (University of Calgary) Upward Ratcheting Of Ceo Compensation In Large Canadian Corporations: Alternative Explanations................................................................................................ 10 Shamsud D. Chowdhury (Dalhousie University) Mohammed F. Safa (Dalhousie University) Individual Success And The Strategy Process: An Exploratory Study ..................... 19 Jacqueline Dahan (HEC Montreal) STRATEGY DIVISION BEST ENGLISH PAPER (Meilleur papier anglais) Got rhythm? Testing a model of the rhythm of change in product diversity ........... 30 Parshotam Dass (University of Manitoba) Na Ni (University of Manitoba) DIVISION STRATEGIE MEILLEUR PAPIER FRANCAIS (Best French Paper) Les PME Biotechnologiques Sont-Elles Contre-Nature?............................................ 45 Carole Donada (ESSEC, France) Isabelle Dostaler (Concordia University) La Resistance D’un Fournisseur Aux Exigences De Flexibilite : L’impact Sur Son Slack Organisationnel..................................................................................................... 55 Carole Donada Isabelle Dostaler STRATEGY DIVISION BEST STUDENT PAPER ((Meilleur papier d’étudiant) Why do firms build network partnerships in hypercompetitive environments? An explanation from the resource-based view ................................................................... 65 Dev K. Dutta (University of Western Ontario) Organizational Responsiveness: A Conceptual Foundation And Research Agenda 75 Karen Ekstein (York University) Inside The Black Box: Understanding Dynamic Relations......................................... 86 Tamiko Hynes (York University) Approches De La Gouvernance Et Performance : Le Cas Des Organisations Du Secteur Public.................................................................................................................. 95 Mouloud Khelif (HEC Montreal) iv Entreprise Durable Et Responsable : Modélisation De La Stratégie Et Principes De Gouvernance.................................................................................................................. 107 AC Martinet (Euristik, IAE Université Lyon 3, France) Marielle A Payaud (Euristik, IAE Université Lyon 3, France) Are Alliances A Risky Proposition? Taking A Closer Look At Alliances From A Systematic-Risk Perspective ........................................................................................ 117 Patrizia Porrini (Long Island University) Are Investment Bankers To Blame For Acquisition Premiums? If Acquiring Proceed With Caution!................................................................................................................ 132 Patrizia. Porrini (Long Island University) Capacité Organisationnelle Internet Et Capacités Dynamiques : Le Cas D’une Organisation De L’industrie Du Tourisme Au Canada ............................................ 145 Laurent Renard (UQAM) Gilles St-Amant (UQAM) Can elephants dance? A look at the ‘mahouts’ for some answers............................ 161 Suhaib Riaz (University of Western Ontario) Sources Of Competitive Advantage And Above Normal Performance: An Empirical Test ................................................................................................................................. 172 W. Glenn Rowe (University of Western Ontario) Michael J. Rouse (University of Western Ontario) Suhaib Riaz (University of Western Ontario) Peut-on Opérationnaliser L’apprentissage Organisationnel ? Une Analyse Du Processus D’innovation De Services............................................................................ 190 Richard. Soparnot (Groupe ESCEM, France) Eric Stevens (Groupe ESCEM, France) Further On The Theory Of Stakeholder Salience: Defining Why And When Does The Who Really Counts................................................................................................ 209 Sujit Sur (Concordia University) Governance Beyond Agency Theory: The Tale Of Dual Class Firms ..................... 226 Sujit Sur (student) (Concordia University) Corporate Social Responsibility Initiatives: A First-Mover Theory........................ 243 Carol-Ann Tetrault Sirsly (Concordia University) Examining Pre-IPO Activity: Firm Growth By Acquisition .................................... 253 Melissa Toffanin (student) Martin L. Martens (Concordia University) Generic Strategy And Capital Structure: The Impacts Of Product Differentiation On Financial Leverage........................................................................................................ 268 Taiyuan Wang (student) (The University of Western Ontario) v How ‘Good Minds’ Think: A Study Of CEO Decision Making ............................... 277 Jaana Woiceshyn (University of Calgary) SUMMARIES / SOMMAIRES D’une Approche Deterministe A Une Approche Contingente Du Lien « Diversification –Performance » : Quel Impact Sur La Validite De Contenu Des Indicateurs « Classiques » De La Diversification ?.................................................... 286 Samia Belaounia (Rouen Graduate Business School, France) Dynamic Capabilities: The Strategy-Hrm Interest ................................................... 287 Denise Ghanam (Bristol Business School, UK/University of Windsor) The Dynamics Of Residual Revalorization: A Case Study In The Canadian Enterprises..................................................................................................................... 288 Jean D. Kabongo (Virginia State University) CEO Moral Capital And Strategic Leadership In Turbulent Times....................... 289 John R. Phillips (University of Western Ontario) An Analysis Of The Relationship Between Alliance Networks And Firm Performance ......................................................................................................................................... 290 Huanglin Wang (University of Western Ontario) vi ASAC 2005 Toronto, Ontario Frances E. Bowen Mahdi Rostami (student) Haskayne School of Business University of Calgary THE BEHAVIORAL THEORY OF STRATEGIC DECISION MAKING: BUILDING ON GREVE’S EMPIRICAL STUDIES In this paper we review Henrich Greve’s recent set of empirical tests of the behavioural theory of strategic decision making. We contextualise his work within broader behavioural theory and strategic management, and identify several theoretical and empirical critiques. Our paper builds an expanded behavioural model which addresses some of these critiques, and suggests future research directions. There has been a reinvigorated interest in the behavioural theory of the firm within strategic management, with several influential books and articles being published in the last few years (e.g. Augier & March, 2002; Bromiley, 2005). Prominent within this line of research have been a series of studies by Henrich Greve on strategic decision making in the radio station and shipbuilding industries (Greve, 1998, 2002, 2003a, b, c). Greve has made significant contributions to the behavioural theory of the firm by injecting risk and strategic decision making into the sometimes passive behavioural view, and by applying empirical techniques to a variety of strategic decisions such as asset expansion, product changes and innovation. While Greve’s models take us closer to an understanding of strategic decision making from a behavioural perspective, there remain several theoretical and empirical enhancements that need to be made to this line of research. While the behavioural approach can provide foundations for understanding strategic decisions because of its realistic assumptions (Bromiley, 2005; Bromiley & Papenhausen, 2003), it is not widely used within the field of strategic management. This is unfortunate, given behavioural theory’s ability to shed light on important issues such as the drivers of strategic change, the influence of managerial risk perceptions on strategic decision making, organizational learning from performance feedback and firm performance. We suggest that this relative lack of interest among strategy scholars is due to an over-reliance of previous behavioural theory on simulations rather than empirical tests based on “real world” strategic data (e.g. Cyert & March, 1963; Levinthal & March, 1981), and an overly passive view of decision makers. Greve’s work has begun to overcome these obstacles. The aim of our paper is to review Greve’s work and suggest future enhancements in this line of enquiry so that the behavioural view can take a more prominent place within strategic management. Our paper begins with a contextual overview of the origins of the behavioural theory of strategic decision making and recent developments in strategic management. We then go on to provide a summary and critique of Greve’s studies, and to suggest some modifications to his decision making model. We conclude by outlining some future challenges for a behavioural approach to strategic decision making research. The Behavioral Theory of Strategic Decision Making The behavioural theory of the firm is usually traced back to Cyert and March’s highly influential book A Behavioural Theory of the Firm. Its earlier intellectual origins can be seen in the work of 1 other members of the Carnegie School of Industrial Administration, particularly that of Herb Simon (March & Simon, 1958; Simon, 1955, 1957). Behavioural theory was born out of a frustration with the inability of economic theories of the firm to explain actual decision making behaviour within organizations. Proponents of this view aimed to inject behavioural realism into the stylised analytics of economic theories of the firm which were based on rational economic actors. The focus was to provide a more empirically grounded theory of organizational goals, expectations and choice. In moving away from assumptions of rationality and equilibrium, behavioural theory provides a more realistic foundation for strategic management research than conventional microeconomics (Bromiley, 2005). Core features of Cyert and March’s (1963) behavioural view were bounded rationality, imperfect environment matching and unresolved conflict (Cyert & March, 1992). Each of these tenets represented a serious challenge to previous economic theories of the firm, but has since to greater or lesser extent become widely accepted assumptions in strategic management. Cyert and March (1963) saw firms as an information-processing and decision-rendering system, and derived a coherent theoretical framework based on the new assumptions. Firms were conceived as “a coalition of individuals, some of them organised into sub-coalitions”, and “a complex system in which different decisions are made at different places in the organization” (Cyert & March, 1963: 31). Organizations are therefore characterised by the quasi resolution of conflict, uncertainty avoidance, problemistic search and organizational learning. Cyert and March’s (1963) theory was designed to explain core economic decisions such as output price and quantity, but over the years elements of behavioural theory have been used to analyse many other phenomena such as the evolution of organizational routines (Nelson & Winter, 1982), organizational learning (Greve, 2003c; Levinthal et al., 1981) and political behaviour (Bourgeois & Singh, 1983). Recently, the behavioural framework has been used to understand a range of strategic issues such as investment (Chen, 2003; Greve, 2003b), innovation (Geiger & Cashen, 2002; Greve, 2003a; Nohria & Gulati, 1996), corporate environmental management (Bowen & Sharma, 2005; Bowen, 2002) and managerial risk preferences (Miller & Chen, 2004). There is also an emerging line of enquiry within strategic management which attempts to integrate behavioral theory with the dominant resource-based view. Examples include bringing bounded rationality (Amit & Schoemaker, 1993; Bromiley et al., 2003; Kemmerer, Kellermanns, & Narayanan, 2004), bargaining and coalitions (Blyler & Coff, 2003), organizational slack (Bowen et al., 2005; Mishina, Pollock, & Porac, 2004) or the behavioral antecedents of investment decisions (Alessandri & Maritan, 2004; Moliterno & Wiersema, 2004) into the resource-based view. The most cited features of Cyert and March’s (1963) book are search, coalition and slack (Engwall & Danell, 2002). Among strategy researchers, however, the most useful concepts from within behavioural theory are bounded rationality, aspirations and search, routines and slack (Bromiley, 2005). Bounded rationality is an accepted assumption within several theories of the firm including transaction cost economics (Williamson, 1963), and the resource-based view (Barney, 2001). Routines have formed the basis of a whole school of strategic thought based on evolutionary views of the firm (Nelson et al., 1982), and are core to our understanding of dynamic capabilities (Eisenhardt & Martin, 2000; Teece, Pisano, & Shuen, 1997). Greve’s recent empirical studies have focused on the other two areas relevant for strategic management: aspirations and search, and slack, particularly as they relate to managerial risk taking. In the next section we review the recent set of empirical studies conducted by Henrich Greve, focusing on areas for future development. We use these critiques to suggest some adaptations to Greve’s decision making model. 2 Henrich Greve’s Empirical Tests of Behavioral Theory Henrich Greve employed the behavioural theory of the firm in empirical studies of US radio stations and Japanese shipbuilding. In a related series of studies (Greve, 1998, 2002, 2003a, b, c), he used an adapted behavioural model to explain strategic changes within firms such as product format change, R & D, innovation, and facility investment. He reoriented Cyert and March’s (1963) organizational approach more towards strategic management by highlighting the roles of managers in corporate strategy, and by emphasizing on “how managers set goals, evaluate performance, and determine strategic changes” (Greve, 2003c: III). A brief summary of Greve’s studies illustrates his diverse range of findings. Greve (1998) applies the behavioural theory of the firm to 160 US radio markets from 1984 to 1992. The dependent variable is format change, which is a significant strategic change for radio stations. The results showed that different format changes are organizations’ reactions to performance-aspiration in different ways. For instance, the innovative format mainly responds to the organizational historical aspiration, and does not react the social aspiration level. Greve (2003b) applies his behavioral model to the Japanese shipbuilding industry from 1971 to 1996. The results show that high performance decreases R&D expenditure, while higher absorbed slack increases it. Greve (2003a) also applied the model to production innovation among Japanese shipbuilding companies. He found that performance above the aspiration level reduces innovation launches in Japanese shipbuilding, and showed a decreasing probability of change trend (with a decreasing rate above the aspiration level) in response to higher performance. Application of the model to Japanese shipbuilding firms also revealed that they invest in their assets only in response to performance feedback and not to their slack resources. In addition, Greve (2003b) found that different measurements of investment in assets change the results. For example, production asset growth has a negative relationship with performance only above the historical aspiration level, while machinery growth has a positive relationship with performance below the historical aspiration level, and a negative relationship above it. Greve has made significant contributions to our understanding of strategic decision making by developing and refining Cyert and March’s (1963) original model. Firstly, unlike Cyert and March’s (1963) approach to the organizational aspiration level, in which they treat historical and social aspirations together as one single entity, Greve unpacked the concept of aspiration level. He developed empirical measures which allowed him to separate historical from social aspirations, and to observe the relative organizational reactions (manifested in organizational strategic change) to them (Greve 1998, 2003b). For instance, Greve (1998) shows that while some strategic changes (format changes) in radio stations are driven by both historical and social aspiration levels (e.g. new format and enter satellite), other strategies such as innovative format are initiated only by the historical aspiration level. Greve also raised the question of how managers decide on target aspiration when performance is between historical and social aspiration levels. He partially addressed this issue by proposing, “an organization near to mean performance of its competitors might pay attention to the social aspiration level, while an organization close to failure would pay attention to the survival point” (Greve, 2003c: 50). Greve also developed a model to explain the speed of updating the historical aspiration level by firms and its possible consequences for firm performance (Greve, 2002). He showed that in an uncertain environment, the slow updating of the historical aspiration improves firm performance. Secondly, Greve introduced a kinked curve which illustrated that the relationship between firm performance and the probability of strategic change has a tipping point at the aspiration level (Greve, 2003b: 1058; and 2003c: 63). The curve is downward sloping, showing an increasing 3 likelihood of change as performance decreases, and organizational inertia reduces the propensity to change below the aspiration level and leads to the kink in the curve. This kinked curve is a powerful graphical device in understanding the likelihood of strategic change. Although Greve (2003c) adapted the kinked curve into a more continuous inverse curvilinear probability function, there remain further modifications that might be made using this graphical approach to the theory. Thirdly, Greve incorporated theory on risk taking into his behavioral model. He employed managerial risk taking and prospect theory (Bromiley, 1991; Sitkin & Pablo, 1992; Tversky & Kahneman, 1981) to argue that managers take more risk when they fail to fulfil organizational goals. Finally, Greve also contributed to behavioural theory in terms of research design and analytical techniques. He designed his research mainly to use archival data, which is attractive to strategic management researchers as it leads to easier application by managers and replication by researchers. Greve also used a combination of time series and cross-section data which helps to show the causality between variables (Greve, 2003c). While Greve is clearly familiar with the simulation tradition of behavioural research (e.g. Greve, 2002), he brought behavioural theory closer to the mainstream of strategic management by conducting well designed empirical studies on publicly available data. Critiques and Extensions of Greve’s Empirical Studies While Greve contributed a series of well designed and executed tests of behavioural theory, there remain critiques and opportunities for the extension of his work. In this section, we outline three areas which could be enhanced within this theory, as a prelude to developing the model in the next section. Areas of Greve’s work which might be developed include the importance of managerial perceptions, the survival level as an additional reference point and interactions between slack and the performance-aspirations level. Cyert and March (1963) were aware of the importance of considering perceived firm performance as well as the firm’s actual performance. However, it seems that the “perceived approach” is missing from Greve’s empirical studies. Obviously under ceteris paribus perceived performance is equal to historical performance. But since managers are assumed to be boundedly rational within behavioural theory, they can only estimate perceived variables as environmental conditions, organizational capabilities, or managerial cognitive capacities change. Entering perceived performance can explain the paradoxical behavior of successful firms (with performance above their aspiration level) who undertake more change than would be expected according to behavioural theory. Some firms may more accurately perceive their performance and aspiration levels than others. Indeed, there may be companies with higher cognitive capacity that can predict future performance and prevent the decline of firm performance through the implementation of proactive strategic change. For example, Greve (2003c) reported that the pharmaceutical firm Eli Lilly’s increase in its R&D expenditure prior to a predicted decline in revenue does not conform to his performance feedback framework. As the behavioural theory of the firm only considers past performance, and reactive strategic change to it, adding perceived performance could capture the proactive firm’s behavior. The same reasoning holds for the slack and aspiration levels. Current performance, past performance, and change in environmental conditions, will shape managers’ perspectives about the future, enabling them to respond to perceived performance, slack, and aspiration like they respond to historical variables. Thus, risk taking behavior and consequently strategic choice are jointly determined by historical and perceived organizational performance, slack, survival, and aspiration levels. Greve’s Kinked-curve model is unable to explain the firm’s behavior especially below the aspiration level (1998, 2003a, and 2003b). For instance, Greve (2003c, p. 86) reported that firms with “several years of declining sales appeared to take greater risks when their performance 4 increased, contrary to the prediction.” Therefore it seems that when a firm approaches the survival point through declining performance, “assets are more important than performance.” This evidence supports March and Shapira (1992) who suggested that a firm responds to the survival level as well as the aspiration level. While Greve did mention the survival level in his book, this modification is not well developed in the book. Future research in this stream should incorporate recent tests of the firm’s survival level as an important reference point (e.g. Chen, 2003). Replacing the probability of performance which Greve used in his kinked curve, with the probability of strategic change (Greve, 2003c: 162) may provide a better explanation for firm behavior in response to its performance feedback. A continuous probability function could better take performance around the survival level into account. The slope of the curve would be influenced by different behaviors around and between the aspiration and survival reference points. Greve focused his work around the performance-aspiration level, hypothesising (but not finding conclusive evidence) that higher risk tolerance leads to more risk taking and a higher likelihood of strategic change as performance falls below the aspiration level. We suggest, however, that decision makers will only increase their risk tolerance to a certain extent. As the firm’s performance falls to near the survival level, decision makers become more sensitive to risk. The threat-rigidity perspective indicates that organizations make less strategic changes when their survival is threatened due to organizational inertia (Staw et al., 1981). This flip-side of prospect theory (Tversky et al., 1981) has been recognised within theory on managerial risk taking (Sitkin et al., 1992), but was not fully addressed by Greve. We would therefore expect an inverse curvilinear probability function for strategic change. Consistent with Greve (2003c) (but inconsistent with his kinked curve), managers will behave in line with the behavioral theory of the firm between the aspiration level and the maximum risk taking point, while they behave according to the threat-rigidity perspective between the survival and maximum risk taking point. Moreover, unlike Greve, and according to Rajagopalan and Spreitzer (1997), the maximum point on this curve, which is the firm’s maximum likelihood of choosing to enact a strategic change, is determined not only by organizational capabilities but also by the environmental conditions, organizational cognitions, and managerial cognitions as well as managerial actions. In Greve’s static model, slack and problemistic search are distinct, but complementary drivers of strategic change. We agree with Greve that achieving performance below the aspiration level triggers problemistic search. However, we also expect that it will affect the perceived level of organizational slack over time. Achieving performance below the aspiration level might lead to a perception of less slack being available, and to increased organizational inertia, and consequently might reduce the sensitivity (rate) of strategic change. We therefore suggest an interaction between a firm’s performance above or below the aspiration level and the intensity of slack search. Expanding Greve’s Behavioral Model of Strategic Decision making Figure 1 presents an expanded behavioural model of strategic decision making which takes several of the critiques of Greve’s work into account as well as Greve’s basic behavioural model. Highlighted boxes and dashed arrows in Figure 1 indicate refinements to Greve’s basic model. Expansions to the model are in three areas. Firstly, at the top of Figure 1, we recognise that Greve’s “observe performance feedback” is not as straightforward as it might appear. We suggest instead that decision makers “perceive performance feedback” based on the firm’s actual performance, the aspiration level and the decision maker’s interpretations of threats and opportunities, and cognitive capacity (as a moderator). Secondly, in the right hand side of the figure, we add the survival level as a significant reference point. As firms approach the survival level, they are expected to decrease their risk tolerance based on threat-rigidity theory. Including the two reference points – the 5 survival and aspiration levels – generates different expectations of risk tolerance in different regions of the strategic change curve, resulting in the expected inverse curvilinear relationship between performance and strategic change. Finally, near the centre of Figure 1, we recognise that there is an interaction between performance-aspirations and organizational slack. We also suggest the importance of perceived, as well as actual, slack in strategic decision making. Figure 1: An expanded behavioural model of strategic decision making Firm’s performance Perceived firm’s performance Managerial interpretations and cognitive capacity Perceived performance feedback Evaluation yes Is performance above the aspiration level? no Is performance close to the survival level? no yes no Increase problemistic search Search Deliver solutions to decision making Lower perceived slack Higher perceived slack no yes Slack search Increase risk tolerance Decision Decrease risk tolerance Decide based on risk tolerance, solutions, and problems Decrease risk tolerance Remaining Challenges within the Behavioural Approach to Strategic Decision making Our expanded behavioural model of strategic decision making addresses several of the critiques of Greve’s recent empirical work. However, several theoretical and empirical challenges remain for behavioural theory to become more attractive to strategic management researchers. Firstly, it could be argued that the even the expanded behavioural model of strategic decision making we have proposed suffers from myopia. The criticism presented by Levinthal and March (1993) of the organizational learning perspective is also valid for Greve’s work. The model suffers from “temporal myopia” which refers to the priority of the short run to the long run, as well as spatial myopia (locality of search in Greve’s case). The natural consequence of such myopia is the competence trap (Levitt & March, 1988), or the paradox of success (Audia, Locke, & Smith, 2000). Both terms refer to the “failure of successful organizations to search ways to improve” (Greve, 2003a: 3). Although Greve (2003: 157) argued that “while the myopia of problemistic search [as a core element of Greve’s model] seems to be inevitable, and perhaps also efficient result of bounded rationality, other aspects of search process can be modified to increase their effectiveness.” At the macro level and public policy perspective this myopia probably is not a problem since in the long-run, the competitive market will penalize organizations that chose incorrect strategies or that fall into the competence trap in response to performance feedback, and keep the efficient organizations. However, from a strategic management perspective, behavioural theory suffers as a managerial framework because myopia is capable of leading to the competence trap. Secondly, the challenge remains to test elements of the expanded model. The behavioral theory of the firm began with qualitative research (Cyert et al., 1963; Greve, 2004). It seems that more 6 qualitative studies are needed in order to build a better understanding of the effects of critical points (aspiration, survival, and maximum strategic change) on risk taking behavior and consequently on strategic decision making. As Greve (2002, p. 13) suggested, “a very useful triangulation of methods is afforded by the opportunity to test these propositions both by direct measure of aspiration levels (in either experiment or surveys) and by estimation from behavioral data.” Having more case studies and different tools to choose from to measure performance, slack, and risk taking constructs will improve our understanding of firm behavior. We would also encourage further quantitative tests of the theory examining various strategic decisions and empirical contexts. A wide range of strategic decisions could be investigated using this behavioural framework including divestments, mergers and acquisitions, foreign entry modes, product launches, etc. At first glance it may appear inconsistent to use optimization statistical techniques within the behavioral framework because the emergence of behavioural theory was itself a reaction to traditional optimisation economic theories. However, the decisions of boundedly rational actors can be understood using optimizing statistical techniques without inferring optimisation on the decision makers, and is a well accepted technique among behavioural scholars (Greve, 2003c; Miller et al., 2004; Nohria et al., 1996). While Greve made a valuable contribution by developing specific measurements of the social and historical aspiration level, several measurement challenges remain. There is a long standing debate on the measurement of organizational slack (see for example Bourgeois, 1981; Marino & Lange, 1983) which has still not been entirely resolved. While some valuable attempts have been made recently to directly measure managerial risk preferences (Miller et al., 2004), managerial interpretations of risks and opportunities (Chattopadhyay, Glick, & Huber, 2001) and the firm survival level (Chen, 2003; Miller et al., 2004), these could be developed further. The behavioural theory of strategic decision making has much to offer strategic management scholars: it makes realistic behavioural assumptions; it can be applied to a wide range of strategic situations; and it incorporates the pivotal strategic variables of firm performance and strategic change. Greve has enhanced behavioural theory’s attractiveness to strategic management scholars by incorporating managerial risk taking, developing empirical tests using secondary data and expanding the scope of strategic decisions to which behavioural theory can be applied. We have identified several further extensions to Greve’s work and provided an expanded behavioural model of strategic decision making. Our new model can better account for the behaviour of proactive firms, and may account for some of the Greve’s inconsistent empirical findings. Our hope is that others will join us in the challenge of developing behavioural theory, and in applying it to an increasing range of strategic problems. References Alessandri, T. M., & Maritan, C. A., “Evaluating investments in capabilities: a conceptual framework,” Paper presented at the Strategic Management Society, San Juan, Puerto Rico, 2004. Amit, R., & Schoemaker, P. “Strategic Assets and Organizational Rent,” Strategic Management Journal, 14, (1993), 33-46. Audia, P. G., Locke, E., & Smith, K. G. “The Paradox of Success: An Archival and a Laboratory Study of Strategic Persistence Following a Radical Environmental Change,” Academy of Management Journal, 43(5), (2000), 837-853. Augier, M., & March, J. G. “The Economics of Choice, Change and Organization” Cheltenham: Edward Elgar, 2002. Barney, J. B. “Is the Resource Based ‘View’ a Useful Perspective for Strategic Management Research? Yes,” Academy of Management Review, 26(1), (2001), 41-56. 7 Blyler, M., & Coff, R. W. “Dynamic Capabilities, Social Capital, and Rent Appropriation: Ties that Split Pies,” Strategic Management Journal, 24, (2003), 677-686. Bourgeois, L. J. “On the measurement of organizational slack,” Academy of Management Review, 6(1), (1981), 29-39. Bourgeois, L. J., & Singh, J. V. “Organizational Slack and Political Behaviour Among Top Management Teams,” Academy of Management Proceeding, 1983. Bowen, F., & Sharma, S. “Resourcing corporate environmental strategy: Behavioral and resource-based perspectives,” unpublished manuscript, University of Calgary, 2005. Bowen, F. E. “Organizational Slack and Corporate Greening: Broadening the Debate,” British Journal of Management, 13, 2002. Bromiley, P. “Testing a Causal Model of Corporate Risk Taking and Performance,” Academy of Management Journal, 34(1), (1991), 37-59. Bromiley, P. The behavioral foundations of strategic management. Oxford: Blackwell, 2005. Bromiley, P., & Papenhausen, C. “Assumptions of rationality and equilibrium in strategy research,” Strategic Organization, 1(4), (2003), 413-437. Chattopadhyay, P., Glick, W., & Huber, G. P. “Organizational actions in response to threats and opportunities,” Academy of Management Journal, 44(5), (2001), 937-956. Chen, W. Firms' technological search behaviors. Purdue University, 2003. Cyert, R. M., & March, J. G. A behavioural theory of the firm. Englewood Cliffs: Prentice-Hall, 1963. Cyert, R. M., & March, J. G. A Behavioural Theory of the Firm (2nd ed.). Oxford: Blackwell, 1992. Eisenhardt, K. M., & Martin, J. A. “Dynamic capabilities: what are they?” Strategic Management Journal, 21, (2000), 1105-1121. Engwall, L., & Danell, R. The Behavioural Theory of the Firm in Action. In M. Augier, & J. G. March (Eds.), The Economics of Choice, Change and Organisation: 27-47. Cheltenham, UK: Edward Elgar, 2002. Geiger, S. W., & Cashen, L. H. “A Multidimensional Examination of Slack and its Impact on Innovation,” Journal of Managerial Issues, 14(1), (2002), 68-84. Greve, H. R.. “Performance, aspirations, and risky organizational change,” Administrative Science Quarterly, 43(1), (1998), 58-86. Greve, H. R.. “Sticky aspirations: Organizational time perspective and competitiveness,” Organization Science, 13, (2002), 1-17. Greve, H. R. “A Behavioural Theory of R & D Expenditures and Innovations: Evidence from Shipbuilding,” Academy of Management Journal, 46(6), (2003a), 685-702. Greve, H. R.. “Investment and the behavioural theory of the firm: Evidence from Shipbuilding,” Industrial and Corporate Change, 12(5), (2003b), 1051-1076. Greve, H. R. Organisational Learning from Performance Feedback: A Behavioural Perspective on Innovation and Change. Cambridge, UK: Cambridge University Press, 2003c. Greve, H. R.. “Old wine in big bottles,” www.bi.no/upload/Info-avdeling/_nedlastingsfiler/ AF%20Filer/Greves%20presentasjon.ppt. [Accessed August, 2004]. Kemmerer, B., Kellermanns, F. W., & Narayanan, V. K. “Resource Assessment: How do we really evaluate resources,” Paper presented at the Strategic Management Society Annual Conference, San Juan, Puerto Rico, 2004. Levinthal, & March. “A Model of Adaptive Organizational Search,” Journal of Economic Behaviour and Organisation, 2, (1981), 307-333. Levinthal, D. A., & March, J. G. “The myopia of learning,” Strategic Management Journal, 14, (1993), 95-112. Levitt, B., & March, J. G. Organizational Learning. In W. R. Scott, & J. Blake (Eds.), Annual Review of Sociolog,y, (1988), 319-340. Palo Alto, CA: Annual Reviews. March, J. G., & Simon, H. A. Organizations. New York: Wiley, 1958. 8 Marino, K. E., & Lange, D. R. “Measuring Organizational Slack: A Note on the Convergence and Divergence of Alternative Operational Definitions,” Journal of Management, 9(1), (1983), 81-92. Miller, K. D., & Chen, W. “Variable organizational risk preferences: tests of the March-Shapira model,” Academy of Management Journal, 47(1), (2004), 105-116. Mishina, Y., Pollock, T. G., & Porac, J. F. “Are more resources always better for growth? Resource stickiness in market and product expansion,” Strategic Management Journal, 25, (2004), 1179-1197. Moliterno, T. P., & Wiersema, M. F. “Contextual determinants of firm-level strategic resource reallocation,” Paper presented at the Strategic Management Society, San Juan, Puerto Rico, 2004. Staw, B. M., Sandelands, L. E., & Dutton, J. E. “Threat-rigidity effects in organizational behavior: A multilevel analysis,” Administrative Science Quarterly, 26, (981), 501–524. Nelson, R. R., & Winter, S. G. An Evolutionary Theory of Economic Change. Cambridge, MA: Harvard University Press, 1982. Nohria, N., & Gulati, R. “Is slack good or bad for innovation?” Academy of Management Journal, 39(5), (1996), 1245-1264. Rajagopalan, N., & Spreitzer, G. “Toward a theory of strategic change: A multi-lens perspective and integrative framework,” Academy of Management Review, 22(1), (1997), 32-48. Simon, H. A. “A Behavioural Model of Rational Choice,” Quarterly Journal of Economics, 69, (1955), 99-117. Simon, H. A. Administrative Behaviour. New York: Free Press, 1957. Sitkin, S. B., & Pablo, A. “Reconceptualizing the determinants of risk behaviour,” Academy of Management Reivew, 17(1), (1992), 9-38. Teece, D. J., Pisano, G., & Shuen, A. “Dynamic capabilities and strategic management,” Strategic Management Review, 18(7), (1997), 509-533. Tversky, A., & Kahneman, D. “The framing of decisions and the psychology of choice,” Science, 30(53), (1981), 458. Williamson, O. E. A model of rational behaviour. In R. M. Cyert, & J. G. March (Eds.), A behavioural theory of the firm: 237-252. Englewood Cliffs: Prentice-Hall, 1963. 9 ASAC 2005 Toronto, Ontario Shamsud D. Chowdhury Mohammad F. Safa (Student) School of Business Dalhousie University UPWARD RATCHETING OF CEO COMPENSATION IN LARGE CANADIAN CORPORATIONS: ALTERNATIVE EXPLANATIONS Anecdotes, business press, and some empirical studies suggest that the 1993 OSC requirement for executive salary disclosure, coupled with traditional market forces, has led to information efficiency of executive labor market, resulting in an upward ratcheting of CEO compensation in large Canadian public corporations. Based on the tenets of managerial power, social comparison, and hidden action, we maintain that the information efficiency-based argument is partial, and that psychological, organizational, and social forces are even more powerful influencers of upward ratcheting. We present the case of a board’s compensation committee as an ideal context for supporting our argument. Over the last couple of decades, the relationship between CEO compensation and firm performance has generated a great deal of controversy. Besides becoming a topic of both interest and scrutiny in the business press, this relationship has also inspired a rich stream of debate among academic researchers in finance, organization theory, and strategy. A consistently high rate of growth in the average earnings of CEOs in Canada and the US over the recent past has stirred this controversy. Although the notion that rewards given to the heads of our public corporations should be linked to the earnings of the companies is consistent with theory, for the most part, academic researchers have been “stymied” in their search for a meaningful association between CEO pay and firm performance (Jensen & Murphy, 1990). Despite an ongoing debate surrounding the controversy, the effect of one critical influence on the process of setting CEO compensation—bidding-up—remains unexplored. Bidding-up refers to an upward “ratcheting” or shoot up of the executive pay over time due to the possibility of a comparison of compensation in the external executive labor market (Ezzamel & Watson, 1998; Finkelstein & Hambrick, 1988; Van Clieaf, 2004). According to anecdotes (e.g., McFarland, 2002, 2004; Van Clieaf, 2004) and some empirical research with indirect bearing on ratcheting (e.g., Chowdhury & Wang, 2004; Magnan, St-Onge, & Thorne, 1995; Park, Nelson, & Huson, 2001; Zhou, 2000), bidding-up has shot up the level of CEO compensation in large publicly held Canadian corporations. The 1993 Ontario Securities Commission (OSC) requirement that all publicly traded firms listed with the Toronto Stock Exchange (TSC) disclose the amount and composition of individual compensation of the five highest paid executives has contributed to an upward ratcheting of CEO pay. As public disclosure of executive compensation adds to information efficiency of the executive labour market (Fama, 1980), bidding-up appears to be an attendant corollary of such regulation. The US experience of upward ratcheting, which stems from a very strict disclosure regulation of the Securities Exchange Commission (SEC), is consistent with Fama’s explanation. We maintain that Fama’s (1980) is a partial explanation, and that there are other equally, or even more powerful social, psychological, and organizational forces that also influence the upward ratcheting of CEO pay both directly and indirectly. An exploration of the nature and interaction of these forces is important for both theoretical and practical reasons. Accordingly, this paper addresses itself to the exploration of these forces and to the way they interact and contribute to 10 upward ratcheting. In this bid, we draw on three theories—managerial power (Bebchuk, Fried, & Walker, 2002), social comparison (O’Reilly, Crystal, & Main, 1988), and hidden action (Holmstrom, 1979)—which seem equipped to encapsulate the intricate and interactive nature of the forces underlying upward ratcheting. Background and Rationale Effective October 1993, the OSC made it a requirement that firms traded on the TSE disclose the amount and composition of individual compensation of the five highest paid executives. Until October 1993, The OSC traditionally required only group disclosure of compensation data for individuals who are defined as executives. In the US, however, such disclosure became mandatory a year earlier. The SEC initiated changes in the Securities Exchange Act of 1934 and effective October 15, 1992 instructed the corporations to provide shareholders with easy, more understandable reports on executive compensation. Accordingly, corporations are required to disclose all compensation awarded to a CEO and four most highly compensated executive officers. This regulation came in response to demands from various stakeholders, especially stockholders and institutional activists, that executive pay packages be made a function of a corporation’s performance. Non-profit-organizations and social activists pressed for such a regulation as well. As the total CEO compensation is usually large, some find the amount to be “offensive” from a point of view of fairness and social justice Following the introduction of the disclosure requirement, directors have come under intense public pressure to make CEO compensation contingent upon firm performance (McFarland, 2002; Park et al., 2001). In reality, however, the opposite appears to be true, which could be attributed to two sets of reasons. The first set relates to a CEO’s obvious self-interests guided by risk aversion. Performance-based pay is riskier, more uncertain (Gray & Cannella, 1997; Hill & Phan, 1991), and generally unwelcome by risk-averse managers who have a significant part of their human capital tied to the company’s future (Muelbroek, 2001). Stockholders, on the other hand, have an obvious interest in opposing pay packages that are independent of performance, in that exorbitant CEO pay affects the ultimate profitability of a corporation. Therefore, to balance these opposing preferences, corporations generally divide total CEO compensation into two components—contingent and noncontingent. Both of these two forms of pay expose CEOs to different levels of risk (Daily, Johnson, Ellstrand, & Dalton, 1998). Contingent compensation induces real uncertainty in CEOs’ pay (Gray & Cannella, 1997), whereas noncontingent compensation provides them with a stable stream of income (Tosi & Gomez-Mejia, 1994). Stockholders prefer the former and CEOs prefer the latter; thus, some trade-off between the degree of risk sharing between the stockholders and the CEOs is likely to determine the proportion of contingent and noncontingent pay. The second set stems from imperfections in the design and execution of corporate governance and incentive mechanisms and their interactions with the market forces. Like any other remuneration, the classic justification for high CEO compensation is that it is driven by the market forces. With respect to market forces, Finkelstein and Hambrick wrote, “There are two perspectives from which this calibration can be considered. First, [CEO] pay may be viewed as a function of supply and demand. Second, it can be viewed as a function of how much the executive is expected to contribute to the performance of the firm. These are not mutually exclusive views, although they may require somewhat different assumptions” (1988: 546). Both perspectives are rooted in economics. If demand for CEOs outstrips their 11 corresponding supply, then, regardless of the performance of the firm, boards may be forced to pay their CEOs a compensation that may not be necessarily related to a firm’s performance. The second perspective is related to the concept of marginal productivity of labor, which simply states that a worker is paid his or her value to the firm. Because these two theoretical perspectives explain how CEO pay is determined in the market, according to the opponents of pay-for-performance, there should not be any controversy over a CEO’s pay. Therefore, argue these proponents, a firm’s board has to ensure that its CEO is paid at least the going rate, or a compensation level typically paid by similar firms to comparable individuals occupying similar posts. Going rate is the minimum level of compensation required to recruit and/or retain effective individuals (Ezzamel & Watson, 1998) to run the corporation. This phenomenon can be represented in a simple graph, which appears below. CEO Performance C2 Total Pay or Going Rate Contingent Pay C1 Non-contingent Pay Firm Performance In the graph, C1 represents noncontingent pay, which a CEO prefers, as it ensures him or her a stable stream of income independent of company performance. C2 represents total pay or going rate, which has to be awarded in order to retain the CEO. Any point along the contingent pay line represents a CEO’s pay-for-performance compensation, depending on the nature of compensation contract he or she has with the board. The corresponding space (against any point on the contingent pay line) between C2 and contingent pay line represents a CEO’s total pay, which is much inflated due to upward ratcheting. However, both perspectives are problematic, in that they add to the inefficiency of the executive labor markets. Imperfections in both demand and supply of available CEOs in the market make appropriate CEO pay questionable. A major inefficiency results from the ambiguities of CEO talent pool (Finkelstein & Hambrick, 1988). According to critics, on a more fundamental level, arguments for both shortage of supply and excessive demand for CEOs are not tenable. Based on discussion with experts in CEO compensation, McFarland (2004) noted that Canada has enough brilliant corporate leaders who are as good as the ones on corporate payrolls with very high amounts of salary. Similarly, on the demand side, according to McFarland, in some industries in Canada, banking being one, there are not many competitors to whom a CEO could turn for alternative employment. Similarly, the justification for CEO pay based on the concept of marginal productivity is questionable. A host of ambiguous factors, which are mostly perceptual in nature, make marginal productivity-based compensation complicated and arbitrary. Most important, how is the marginal productivity of a CEO measured? Besides hundreds and thousands of employees, a 12 CEO is aided by a top management team (TMT). How can, then, a CEO’s marginal productivity be delineated? Because a CEO is the individual responsible for the overall performance of the organization, there is good reason to suspect that a fair chunk of marginal productivity of other members of the TMT is lumped under that of the CEO. Therefore, a marginal productivity based explanation is far too simplistic and inaccurate to describe CEO compensation in the executive labor market (Iacobucci, 1996). In order to save on transaction costs stemming from sources, such as motivation, recruitment, and retention, added Iacobucci, boards pay CEOs a higher compensation, thus breaking the expectation consistent with the notion of marginal productivity of CEOs. Despite such inefficiencies, economic theories have been used in the study of executive compensation in such a way as if no inefficiency ever existed. Consistent with Finkelstein and Hambrick (1988) and O’Reilly et al. (1988), we maintain that theories from organizations, psychology, and sociology may help us present a more complete explanation of upward ratcheting. In this bid, we start with a special context—a board’s compensation committee—which we consider ideal for supporting our argument. Board of Directors The assessment of the performance of CEOs and the determination of their appropriate compensation packages is one of the central monitoring functions of a board of directors (Lorsch, 1989; Walsh & Seaward, 1990). Therefore, a careful utilization of the CEO compensation process should serve to align his or her interests to those of the stockholders of a firm (Jensen & Meckling, 1976; Tosi & Gomej-Mejia, 1994). As agency theory implies, a CEO’s compensation, at the least, must be consistent with the performance of the very corporation he or she serves. A failure on the board’s part to constrain excessive CEO compensation hurts corporate profitability, and thus, erodes stockholder wealth. The imperfections associated with the two market-driven forces, as outlined earlier, underscore the role of a board’s compensation committee in the determination of CEO pay. Consistent with agency theory, a widely held perception in many quarters and the business press is that independent compensation committees help to protect shareholder interests. However, the failure of the OSC’s mandatory disclosure requirement and the sales elasticity of CEO compensation, as outlined in the following paragraphs, indirectly justify why compensation committees, even with a large proportion of outside members, award their CEOs generous pay raises unrelated to firm performance. The purpose of the OSC’s legislated, mandatory disclosure of executive compensation to link pay to performance appears to have been defeated. With a sample of 80 firms, Park et al. (2001) found an increase in total executive pay in real terms following open disclosure. Because of mandatory executive salary disclosure, CEOs can easily determine their position among peers in Canada and can benchmark themselves rather easily, and may move to alternative corporations accordingly (Gelinas, Magnan, & St-Onge, 2004; Park et al., 2001; McFarland, 2004). As a result, this disclosure might have created ripple effects, initiating a tradition for other companies to match the CEO pay of a rival company. Besides finding out their pay position among peers in Canada, more importantly, CEOs have access to an even much larger and more attractive US executive labor market, which provides them with an opportunity to assess their options in terms of career moves. Certain provisions of the North American Free Trade Agreement (NAFTA) might have made such moves easier. Given these options, CEOs of large Canadian corporations now have effective bargaining chips to negotiate their employment contracts upward. Air Canada pilots illustrate this point. During their strikes on at least a couple of recent occasions, they have benchmarked their pay with their peers in the US, and have pressed for higher compensation. 13 Sales elasticity of executive compensation in Canadian firms tacitly supports why a bidding-up effect might be present in the Canadian executive labor market. Chowdhury & Wang (2004) found a clear association between the proportion of outside directors on a compensation committee and the level of CEO compensation. Similarly, Conyon and Peck (1998) found that the fraction of outside directors in a compensation committee is positively related to top management pay. Such findings suggest that independent compensation committee members are not there to set a CEO’s compensation level appropriately. The disclosure requirement, coupled with the sales elasticity argument, supports the presence of a upward ratcheting effect for CEO compensation in Canadian corporations. This situation might also explain why large pay raises to CEOs are unrelated to company performance. If compensation committees acted in the best interests of the stockholders, agency theory would suggest, CEO pay increases could not be so unrelated to firm performance. The above discussion suggests that there might be other explanations for an absence of the relationship between pay and performance. In other words, the elements of managerial power (Bebchuk, 2002), social comparison (O’Reilly et al., 1988), and hidden action (Holmstrom, 1979) work in a way that ultimately adds to the efficacy of market-driven forces in augmenting upward ratcheting of CEO pay. Managerial power. In large public corporations, the board of directors is in charge of compensating the CEO. A board is usually made up of both inside and outside members. For effectiveness in the trusteeship role of a board, according to agency theory, it should have a large proportion of independent outside directors on it. However, both anecdotes and survey evidence suggest that CEOs can handpick new board members from outside (Bebchuk et al., 2002.). This ability of a CEO formally ties managerial power to CEO compensation. Empirical evidence on suboptimal CEO compensation contracts also endorses managerial power. For example, Shivdasani and Yermack found that CEOs exert influence in the selection of new directors and that when CEOs are involved in the process, “directors predisposed to monitor the CEO are less likely to be appointed” (1999: 1830). Because of their long tenure or because of having a contract on the degree of influence in the selection of new directors (Bebchuk et al., 2002), boards appoint fewer independent outside directors and more gray outsiders (Shivdasani and Yermack, 1999). In such a situation, a board’s monitoring of the CEO deteriorates. Therefore, a compensation committee made up from such a board is more likely to protect the interests of a CEO, recommending an inflated pay package that other considerations would not normally justify. In formal terms, Proposition 1: All else being equal, when a CEO has discretion in the selection of new outside directors, a compensation committee made up of such outside directors is more likely to contribute to upward ratcheting of CEO compensation. Social comparison. The compensation committee of a board typically consists of a small number—three to five—of board members, all of whom are, by convention, not members of management (O’Reilly et al., 1988). In other words, the compensation committee typically consists of outside directors on the board. These individuals are CEOs of other firms (Kesner, 1988; O’Reilly et al., 1988), who are generally influential and well-compensated. O’Reilly et al. (1988) found that compensation committee members set CEO pay in relation to their own salaries. Outside directors, who also sit on the compensation committees of other companies, award CEOs as much compensation as they deem appropriate (Ezzamel & Watson, 1998; Kesner, 1988). Independent of other critical considerations, their basis of recommendation for CEO compensation is purely social comparison. These individuals anchor their judgments about appropriate CEO remuneration by comparing the CEO’s salary to their own (O’Reilly et al., 1988). In formal terms, 14 Proposition 2. All else being equal, in boards where a compensation committee is made up largely of independent outside directors, such a compensation committee is more likely to contribute to an upward ratcheting of CEO compensation. Hidden action. Several researchers highlight the importance of information regarding comparison levels of pay in setting the CEO compensation process, such as wage surveys and mandatory salary disclosure (Finkelstein & Hambrick, 1988; Ezzamel & Watson, 1998; Magnan et al., 1995; Park et al., 2001). It may be argued that this reliance on external information is an offshoot of what Holmstrom (1979) referred to as a “hidden action” model of CEO tasks. According to this model, the CEO is required to perform a series of activities to maximize the utility of the shareholders. However, the tasks of the CEO may not be observable to the investors. Similarly, the board may be unable to completely observe or understand the tasks a CEO performs. Although the CEO should receive higher compensation if the tasks are complicated and require greater skill or hard work, the opposite might be true as well. As this apparently hidden, unclear job role of a CEO can be likened to a black box, the contents of which are not really known, a compensation committee prefers to play safe. It aligns managerial incentives by tying CEO compensation to clear observable corporate characteristics, such as company size, performance, role titles, reporting lines, and capital structure (Holmstrom, 1977; O’Reilly et al., 1988; Van Clief, 2004), which might throw off the balance between pay and performance. There is another reason why compensation committees rely heavily on salary surveys and external compensation consultant reports. An overly inflated compensation package may create an “outrage” to the external constituencies of a firm, especially its stockholders and institutional activists. Outrage can be costly to the directors in terms of their reputation. Therefore, they need what Bebchuk et al. (2002) called “camouflage.” The recruitment of pay consultants and the use of their reports, coupled with other comparable statistics, legitimize exorbitant pay under managerial power approach. So, the interaction of hidden actions and managerial power through camouflage appears to be an important contributor to the inflation of a CEO’s compensation package. In formal terms, Proposition 3: All else being equal, when a compensation committee relies extensively on outside information, such as mandatory executive salary disclosure, independent consultant reports, and salary surveys, such a compensation committee is more likely to contribute to an upward ratcheting of CEO compensation. Conclusion and Implications Many investors consider CEO compensation as a proxy for board effectiveness. A lack of connection between CEO pay and firm performance signals a governance problem within the enterprise which, in turn, negatively affects the ability of a corporation to attract capital both at home and abroad. Excessive compensation packages for CEOs, mainly attributable to bidding-up, can “erode the respect of investors and employees, and even start to undermine faith in the whole free-market system” (McFarland, 2004: B7). Therefore, a complete understanding of the process of upward ratcheting of CEO pay is necessary. In this paper, we argued that executive labor market is not perfect, and that traditional economic theory inadequately explains the full scope of upward ratcheting. Rooted mainly in sociology, psychology, and organization, other equally, or even more powerful, forces explain why upward ratcheting of CEO pay occurs. As we have demonstrated in the context of a board’s compensation committee, which recommends a CEO’s pay package, managerial power, social comparison, and hidden actions also encourage, either singly or in combination, upward ratcheting. 15 More interestingly, these non-economic forces are found to foster upward ratcheting in combination with traditional market-driven forces. In Canada, the 1993 mandatory disclosure of executive pay is often blamed for this unintended consequence of upward shooting of CEO compensation. However, as we argued, the 1993 disclosure regulation should not be entirely blamed for this phenomenon. As a matter of fact, such a disclosure of executive pay provides a safeguard against looting by management in collusion with captive boards of directors (Jensen & Murphy, 1990). Ratcheting is largely an outcome of an imperfectly designed and poorly monitored CEO compensation package by a corporation’s board. As we demonstrated, there are other equally or more effective contributors to upward ratcheting to which attention needs to be paid. Due to an interaction of an array of both market and non-market forces, it may not be possible to establish an exact pay-for-performance package for CEOs, but the effects of upward ratcheting can be neutralized greatly. Therefore, determining the extent of ratcheting in the overall package of a CEO’s compensation appears to be the first crucial step. However, a precondition for this step is to understand the process of how a CEO’s compensation is determined. For example, of the corporate characteristics that form the basis of outside consultants’ recommendations for CEO pay, characteristics such as role titles, reporting lines, and a CEO’s locus in the overall structure are difficult to observe, and hence, an in-depth, first hand understanding of these characteristics is necessary. The second step involves a determination on the part of a board to undertake added responsibility for curbing excess. Added responsibility includes, but is not limited to, measures such as more truly independent, but professionally dedicated, outside members on the compensation committee (Vafeas, 2003); a detailed explanation in the proxy statement as to how compensation is awarded to a CEO and how it matches corporate performance (McFarland, 2004; Vafeas, 2003); restricting absolute reliance on compensation consultants (McFarland, 2004); measuring performance over a multiple year time period (Gray & Cannella, 1997; Van Clieaf, 2004); structure of compensation contracts (Perry & Zenner, 2001); and a separation of the total CEO compensation into contingent and non-contingent components (Daily, Johnson, Ellstrand, & Dalton, 1998; Tosi & Gomez-Mejia, 1994); and a restriction on the participation of a CEO in the board’s nomination committee. Once a complete understanding of how CEO pay is awarded is established, it will then be possible to model the effect of ratcheting on a CEO’s pay. References Bebchuk, L.A., Fried, J.M., & Walker, D.I., 2002. Managerial power and rent extraction in the design of executive compensation. The University of Chicago Law Review, 69, 751-846. Chowdhury, S.D., & Wang, E. 2004. CEO compensation: Turning conventional wisdom on its head (almost). Ivey Business Journal, November-December, 1-7. Conyon, M.J., & Peck, S.I. 1998. Board control, remuneration committees, and top management compensation. Academy of Management Journal, 41: 146-157. Daily, C.M., Johnson, J.L., Ellstrand, A.E., & Dalton, D.R. 1998. Compensation committee composition as a determinant of CEO compensation. Academy of Management Journal, 41: 209-220. Ezzamel, M., & Watson, R. 1998. Market comparison earnings and the bidding-up of executive cash compensation: Evidence from the United Kingdom. Academy of Management Journal, 41: 221- 231. Fama, E. G. (1980). Agency problems and the theory of the firm. Journal of Political Economy, 88: 16 280-307. Finkelstein, S., & Hambrick, D.C. 1988. Chief executive compensation: A synthesis and reconciliation. Strategic Management Journal, 9: 543-558 Gelinas, P., Magnan, M., & St-Onge, S. 2004. Increased compensation costs: An externality of mandatory executive compensation disclosure. Paper presented at the annual conference of the Administrative Sciences Association of Canada, Quebec City, Quebec. Gray, S.R., & Cannella, A.A. 1997. The role of risk in executive compensation. Journal of Management, 23: 517-540. Hill, C.W.L., & Phan, P. 1991. CEO tenure as a determinant of CEO pay. Academy of Management Journal, 34: 707-717. Holstrom, G. 1979. Moral hazard and observability. Bell Journal of Economics, 13: 234-340. Iacobucci, E. 1996. The law and economics of executive compensation. In E. Iacobucci (Ed.), Value for money: Executive compensation in the 1990s: 1-56. Toronto: C.D. Howe Institute. Jensen, M.C., & Meckling, W.H., 1976. Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3: 305-360. Jensen, M.C., & Murphy, K.J. 1990. Performance and top management incentives. Journal of Political Economy, 98: 225-264. Kesner, I.F. 1988. Directors’ characteristics and committee membership: An investigation of type, occupation, tenure, and gender. Academy of Management Journal, 31:66-84. Lorsch, J.W. 1989. Pawns or potentates: The reality of American corporate boards. Boston: Harvard Business School Press. Magnan, M.L., St-Onge, S., & Thorne, L. 1995. A comparative analysis of the determinants of executive compensation between Canadian and U.S. firms. Relations industrielles/Industrial Relations, 50(2): 297-318. McFarland, J. 2002. Boards rapped over CEO compensation. Globe and Mail, April 24. McFarland, J. 2004. CEOs rake it in as profits and stock prices rebound. Globe and Mail, April 30. Muelbroek, L. 2001. The efficiency of equity-linked compensation: Understanding the full cost of awarding executive stock options. Financial Management, 30: 5- 44. O’Reilly, C., Main, B., & Crystal, G. 1988. CEO compensation as tournament and social comparison: A tale of two theories. Administrative Science Quarterly, 33: 257-274. Park, Y. W., Nelson, T., & Huson, M. R. 2001. Executive pay and the disclosure environment. Canadian evidence. Journal of Financial Research, XXIV(3): 347-365. Perry, T., & Zenner, M. 2001. Pay for performance? Government regulation and the structure of compensation contracts. Journal of Financial Economics, 62: 453-488. Shivdasani, A., Yermack, D., 1999. CEO involvement in the selection of new board members: An empirical analysis. Journal of Finance, 54, 1829-1853. Tosi, H.L., & Gomez-Mejia, L.R. 1994. CEO compensation monitoring and firm performance. Academy of Management Journal, 37: 1002-1016. Vafeas, N. 2003. Further evidence on compensation committee composition as a determinant of CEO compensation. Financial Management, 32(2): 53-70) Van Clieaf, M. 2004. Are boards and CEOs accountable for the right level of work? Ivey Business 17 Journal, May-June, 1-12. Walsh, J.P., & Seward, J.K. 1990. On the efficiency of internal and external corporate control mechanisms. Academy of Management Review, 15: 421-458. Zhou, X. 2000. CEO pay, firm size, and corporate performance: Evidence from Canada. Canadian Journal of Economics, 33: 213-251. 18 Jacqueline Dahan (student)1 HEC Montréal ASAC 2005 Toronto, Ontario INDIVIDUAL SUCCESS AND THE STRATEGY PROCESS: AN EXPLORATORY STUDY2 Today’s business press questions the link—by no means automatic—between individual and organizational success. Top performers or mavericks (“straight shooters”) display idiosyncratic behaviour that, while it brings innovation, puts organizational success at risk. The study reveals that individual success is a social construct that is part of the process of strategy formation and stems from a corporatist attitude to official strategy. The business press regularly reports that individual professional success (the success of the manager or professional) is not an automatic component of organizational success (Bunker, 2002; Cliffe, 2001; Groysberg et al., 2004; Ludeman and Erlandson, 2004). In these articles aimed at practitioners, successful individuals (“star performers”), mavericks (McMahan, 1993) and “alpha males” (Ludeman and Erlandson, 2004) consistently display idiosyncratic behaviour that yields much innovation. However, such deviance from the norm, while bolstering the organization’s competitive edge on the one hand, may paradoxically prove harmful to organizational success on the other. A number of authors in the business press have raised the problem of the “abrasive personality” that star performers, so vital to organizational success, can nonetheless pose for the organizations that employ them (Alder et al. 1997; Bunker et al. 2002; Cliffe, 2001). Ludeman and Erlandson (2004) summarize the issue in their article “Coaching the Alpha Male”: “Bold, self-confident, and demanding, alpha males get things done. But the traits that make them so productive can also drive their co-workers crazy.” Similarly, Cliffe, in her article “What a Star—What a Jerk,” asks: “Sometimes an employee can be nasty, bullying, or simply hard-hearted. What should you do, though, when that person also happens to be a top performer?” Bunker et al. (2002) also refer to this dilemma: “[H]e’s a rising star. He’s also arrogant and unseasoned. Denying him that promotion might be the best thing you could do for his career—and your company (...) Unfortunately, most managers seek out smart aggressive people, playing more attention to their accomplishments than to their emotional maturity.” Faced with this quandary—that of the organization at the mercy of its maverick performers—Ludeman and Erlandson demonstrate that, for the good of the organization, it becomes imperative to “tame the Beast” and focus instead on group cohesion. 1 The author, Jacqueline Dahan, is currently working toward her Ph.D. in Management at HEC Montréal. Her thesis, which focuses on the meaning “professional success” holds for managers and professionals, sets out to examine managers’ interpretations of how such success is constructed, as well as any links they perceive between individual and organizational success. Ms.. Dahan is currently exploring how the process by which individual success is constructed relates to the strategy process. The data she has collected to date is from private sector Quebec manufacturers in the high-tech industry. 2 The author would like to thank Christiane Demers, Ph.D. and Full Professor at HEC Montréal, for her informed and insightful comments on the first draft of this article. 19 In the spirit of these articles and supporting the existence of such a quandary, the study conducted by Groysberg et al. (2004) on 1,000 senior and high-performance financial analysts in the finance industry unearthed some surprising findings. The authors drew the following conclusion: Top performers quickly fade out after leaving one company for another. The results were striking. After a star moves, not only does her performance plunge, but so does the effectiveness of the group she joins—and the market value of her new company. Moreover, transplanted stars don't stay with their new organizations for long, despite the astronomical salaries firms pay to lure them from rivals. Companies cannot gain a competitive advantage or successfully grow by hiring stars from outside. Firms shouldn't fight the star wars, because winning could be the worst thing that happens to them. Our reading of the business press led us to perceive this dilemma primarily as a strategy issue, one that appeared to emanate from the tension between, on the one hand, top performers who marked themselves out from the group (and the ensuing recognition of their success by their superiors), and on the other hand, organizational success which, by definition, implies group cohesion. We noted, too, that this issue faced by practitioners—the individual/group polarity and the tension it engendered—was a virtually non-existent topic in university research. Taking our lead from these findings, we set out to understand how the individual success of the manager or manager was constructed. In the summer of 2003, we conducted an exploratory study within a large Quebec parapublic organization. The choice of the parapublic domain was deliberate, and was based on the following premise: if individual success could be said to arise from the tension between deviant individualism and group cohesion, then any such divergence from the group ought to be that much more visible in the highly institutionalized character of the parapublic organization. Such a context would therefore provide us with an extreme case that would allow us to approach our research topic from an interesting perspective. The study was executed based on the idea that strategy, the core element of organizational success, also played a key role in the construction of individual success. Our starting point was thus the notion that a manager’s contribution to company strategy could have a significant positive effect on his or her individual success. More specifically, we wished to respond to the question, How does “success” come about for the manager, and how does this correspond to the strategy process? Research question This case study thus aims to uncover the link between professional success and company strategy. The question, while focused on individual success, is more particularly concerned with the process by which such “success” is constructed—a central tenet of our stance. To all intents and purposes, our starting point was the hypothesis that individual professional success is a process of social construction that transpires through the interactions between manager and peer or co-worker. Our question necessarily brings us to the field of strategy. There are two main reasons why we chose to conduct our research from this perspective. First, to return to the idea that the strategy process may serve an important function in the construction of individual success, we believe that managers distinguish themselves from their peers to the extent that they contribute differently or significantly to the strategy process. Underpinning this is our own belief in the local actor’s value-added contribution to strategy formation through micro actions, along with our subscription to the beliefs held by certain authors who study emergent strategy and strategy-in-action (strategizing) (Rouleau, 2002; Salvato, 2003; Jarzabkowski, 2003). These authors take their lead 20 from Bower and his classic work, Managing the Resource Allocation Process (1970), one of the first to cast light on the substantial contribution of the local actor to strategy and organizational success. The second reason for focusing on the field of strategy stems from the depiction of organizational success in both the business press and academic research. An apparent consensus exists among certain authors of university studies in strategy as to the near-tautological link between official corporate strategy (deliberate strategy), organizational success, and an organization’s enduring competitiveness (Barnett and Burgelman, 1996; Burgelman, 2002; Eisenhardt and Brown, 1998; Eisenhardt and Sull, 2001; Hamel and Prahalad, 1995; Kanter, 2002; Klein, 2002). As Klein (2002:318) writes, “[M]any authors (...) define competitive strategy as ‘the basis on which a SBU (strategic business unit) might achieve competitive advantage in its market,’ and in most of the definitions given to strategy, competitive advantage means nothing more than success.” Our research question thus entailed a two-pronged review of the literature: the first focusing on individual success; the second, on strategy and the local actor’s contribution to the strategizing process. A review of the literature Current literature on professional success is almost exclusively focused on the career; its authors implicitly endorse the simple idea that professional success is consistent with a successful career (Falcoz, 2001; Lee, 2003). In substance, career-centred literature falls into one of two camps. The first, academic and often prescriptive, takes an objectivist approach and mainly concerns career or performance management. This approach clearly has value for the practice insofar as it provides company directors and/or human resource management experts with a reference tool (Watson and Hassett, 2004). The second, equally prescriptive, comes from the business press. This type of literature addresses managers directly, much like a coach; urging readers to take their careers in hand and build sound individual career strategies, it essentially lays out the stepping-stones to professional success (Ibarra, 2004; Michaud, 2003). As we see it, both types of literature are informed by two dominant ideas: first, that professional success hinges on the degree to which a manager’s profile (education, experience, values, etc.) “fits” a given organizational culture (Anderson, 1997; Falcoz, 2001; Hall, 2004); and second, that success per se is the by-product of individual will. In this sense, the individual is an active and determining force in the construction of his or her own professional success and in achieving a good “fit” (Falcoz, 2001; Hall, 2004). Academic research on professional success is rooted in a positivist paradigm with an objectivist ontological foundation—which is to say, premised in an objective, not subjective, definition of success. Studies of this kind generally fall into the field of Organizational Behaviour (OB), and examine the correlation between variables. Objective success is sometimes the independent variable: the study by Weil and Kimball (1996) demonstrates the causal relationship between success and salary. Other positivist studies present success as a dependent variable; here, individual factors generally serve as the independent variables. For instance, the study by Zorn and Violanti (1996) deepens the causal relationship between an individual’s interpersonal skills and his or her professional success. However, less common positions in current academic research point to the importance of differentiating between “objective” and “subjective” success. While the first instance may be quantified through its factual indicators (promotions, salary scale, etc.), the second appears subject 21 to the interpretation given to it by the individual in question—the manager vis-à-vis his or her own career path. Subjective success (Lee et al., 2003), also known as “psychological success” (Hall, 2004; Lee et al., 2003), means that managers may still see themselves as a success even if they choose to eschew such traditional outward indicators as pay rises and promotions. Researchers’ interest in psychological success, apparently fleeting, becomes clear in a context where the proverbial “job for life” is increasingly rare, where organizations can no longer promise their managers endless promotions or guarantee employment until retirement. Construction of this kind of success, rather than being dependant on outward indicators, is fed by different experiences with different employers. This is precisely what certain authors are referring to when speaking of the “boundaryless career” (Ackah and Heaton, 2004; Hall, 2004; Lee et al., 2003), or of the multidirectional career path (Baruch, 2004; Hamori, 2003). The career characterized by frequent changes of employer invokes the “protean worker” (Hall, 2004; Reitman and Schneer, 2003). The analogy to the Greek god Proteus and his ability to take on any form points up the identity building process intrinsic to the career. In summary, we note that the bulk of these studies are positivist; the rare studies that take an interpretive approach centre their research objectives on the career, and frequently situate their arguments within the academic discourse around the “work/family balance.” As our methodology will make clear, our research delves instead into managers’ sense-making regarding professional success. To date, and to the best of our knowledge, no study has closely examined the ways in which managers perceive the process by which professional success is constructed; nor does any study closely examine the relationship between individual and organizational success. Because it is interpretive, and because it explores success from a strategy perspective, our study stands to reveal new insights into the subject. However, while individual success has never been a research topic per se in the field of strategy, the work of certain authors—in particular, Burgelman (2002), Kanter (1983) and Avenier (1997)—has nonetheless shed considerable light onto the subject, and should be mentioned. Burgelman argues that strategy formation is a collective process. In his work Strategy is Destiny (2002), Burgelman shows how the local actor can contribute to strategy in two different ways that are not mutually exclusive: by complying with the official strategy, and by diverging from it. Burgelman’s findings support the idea that individual success may be constructed around conforming to and/or deviating from the organizational context and the official (deliberate) strategy. In his famous study on Intel,1 Burgelman proposed a model of strategy that consists of the four processes inherent to strategy formation. These are variation, selection, retention and competition. Variation refers to the multiple ways in which actors or groups of actors experiment and the varied initiatives that allow them to reach personal fulfilment. Selection is a structured process consisting of a number of mechanisms (cultural, political, etc.) that act as regulators in sorting initiatives and in making resource allocation decisions. Retention refers to the fact that certain initiatives survive the jungle of local initiatives to be retained (by upper management). Lastly, competition is both internal and external by nature: internal insofar as internal projects must share the limited resources available; external in the extent to which organizational success turns on strategies that are formed with a view to prevailing over other organizations who display similar potential. It is worth noting that, with regard to these processes, Burgelman was strongly inspired by the evolutionary perspective—highly macro in nature—and brought it to an intra-organizational 1 Intel is an American corporation which started out in the semi-conductor industry and changed its positioning to the microprocessor industry. 22 level by applying the Darwinian model to the organizational system to explain the refinement of local initiatives (Burgelman, 1996). For Burgelman, these four processes fall within two generic types of strategic action: the Induced Strategy Process and the Autonomous Strategy Process. The first refers to the top-down process by which a strategy officially approved by upper management is implemented through subaltern levels. “Induced strategy” underpins the implementation of an official strategy, which evokes the organization’s past successes along with the successes to come. In this sense, induced refers to the idea that any such official strategy constitutes a strong frame of reference: the actors who implement the strategy not only share this frame of reference, but also fundamentally subscribe to the historic “essence” carried by the official strategy throughout the hierarchy; in this sense, “induction” is a kind of indoctrination. The second great strategic process, which Burgelman calls Autonomous Strategy, corresponds to “intrapreneurship.” Autonomous strategic action relates to the moment of emergence of local initiatives that leave the field of the official strategy. These initiatives may be described as forward-looking for the organization and/or its positioning within the industry. Autonomous strategic action never overlaps with induced strategic action in the target markets and/or use of technology (“strategic dissonance”). For Burgelman, the genotype or genetic code is an appropriate metaphor for induced strategy, while the metaphor of mutation most aptly describes autonomous strategy. Though far from being an interpretive approach, Burgelman’s model is an inspiring one for two main reasons. The first relates to our argument that strategy is a propitious place—a crucible—for the construction of individual success. Burgelman posits strategy as a social construct, a concept that parallels our ontological stance. The second reason is specific to the induced and autonomous processes that acknowledge the local actor’s contribution and enable individual success to be viewed from two perspectives: conformity on the one hand, and deviance from official strategy on the other. Avenier (1997), in turn, presents strategy from an even more micro perspective than Burgelman. Here, the literature treats strategy-in-action (strategizing), which Avenier (1997) calls “tentative” or “trial and error” strategy. The process referred to is iterative, complex and fluid; in it, the emerging and the deliberate, much like vision and action, come together and inform each other in a symbiotic exchange from which strategy materializes. Within the determining and limiting complexity of industry, the organizational actors (aiming, of course, for organizational performance) have to feel their way forward—a notion that perfectly illustrates the ebb and flow between emergent (induced, unexpected, involuntary) and deliberate (rational, deductive, voluntary) strategies. For our purposes, this “tentative” strategy or strategizing corresponds to strategic processes achieved on a micro level through daily emergent interactions that inevitably occur within the standardized context of organizational routine. These local emergent actions are only qualified as strategic in hindsight, i.e. when the members of upper management render an a posteriori judgment on their compliance with the overall vision (Mintzberg and Waters, 1985). What remains is that these ideas or projects for change which appear as emerging strategies for upper management are, in all likelihood, first considered at length by the local actor. The process by which success is constructed appears to be enmeshed with strategizing and emerging strategies. Another author whose research, along with that of Burgelman and Avenier, shed insight on our study is Kanter. In her work The Change Masters (1983), which examines the private American company, Kanter is credited with highlighting the local actor’s role in both the process of strategy formation and the organization’s capacity for change. For Kanter, all local actors are potential 23 agents of positive change regarding the organization’s competitiveness. Strongly loyal to the company, these actors are innovative subordinate managers whose role consists of anticipating the need for change and of steering this change. These are the “intrapreneurs” who allow the organization to reinvent itself. The organizations discussed by Kanter are from the private sector and see local initiatives as vital to their success. These initiatives are encouraged by upper management: “[I]nnovating companies provide the freedom to act, which arouses the desire to act” (Kanter 1983:142). An organization’s innovative capacity rests on its participative culture (culture of pride, culture of success) and its structural flexibility. However, while Kanter recognizes local actors’ contributions, she does not examine the visibility that the local actor might gain through his or her contribution to strategy and organizational success. In fact, for Kanter, the local actor does not stand out from the group, his or her initiatives notwithstanding: “[I]ndividuals disappear into collectives” (Kanter, 1983:284). When all is said and done, organizational success supplants individual success, even if the initiatives submitted by the local actor should contribute to major change. In short, the authors of works on strategy who influenced us did so because they acknowledged the local actor’s contribution to strategy. However, none of the authors delve into the formation of individual success from a micro enough perspective or an interpretive enough manner to broaden our understanding of how certain managers “break free from the herd” in the way that this is understood by organizational actors. Methodology Our exploratory study was carried out in the summer of 2003 in a large, parapublic quebec organization with over 1,000 unionized employees, of whom just over thirty could be called managers We gathered our data through nine semi-directed, semi-structured interviews, six of which were with directors who reported directly to the president, and three with department heads who reported to the directors. Each individual interview was recorded and transcribed solely by us. The deliberate choice of a parapublic organization may be explained by our initial surmise: a manager would stand out from his/her peers to the extent in which he or she contributed in an atypical manner to the organization’s strategy. It was our belief that, in such a heavily institutionalized context, any aberrance would be that much more apparent. In other respects, while our interests centred on the process by which success was constructed, there was no need for a longitudinal study because the process by which meaning is created is in itself retrospective1. Since our objective was to come to grips with a subject that was as yet undeveloped from an interpretive angle, this accounts for our use of the qualitiative methodology known as Grounded Theory. We did not wish to inventory the objective variables that confirm objective success. On the contrary, we wanted to understand the meaning the respondents gave to individual success and its construction within the process of strategy formation. Furthermore, the number of interviews was calculated in order to reach saturation. As Strauss and Corbin (1998:136) explain: “A category is considered saturated when no new information seems to emerge during coding, that is, 1 “The idea of retrospective sensemaking derives from Schutz's (1967) analysis of ‘meaningful lived experience.’ The key word in that phrase, lived, is stated in the past tense to capture the reality that people can know what they are doing only after they have done it..” (Weick, 1995:24) 24 when no new properties, dimensions, conditions, actions/interactions, or consequences are seen in the data.” Our results led us to believe that we had reached this level. Lastly, influenced by Glaser and Strauss (1967) and their work, The Discovery of Grounded Theory, our data analysis process was inductive and subject to a systematic comparison of the meaning our respondents gave to professional success and its construction. The results To begin with, the responses obtained showed a clear consensus regarding professional success and its construction. In essence, all the respondents agreed that professional success is a collective process that not only takes time, but depends narrowly on the manager’s interactions with his or her co-workers. Indeed, it is precisely through multiple local interactions that the manager is perceived as standing out from the group and his or her reputation established. More specifically, the meaning that respondents gave to success indicates that the process by which success is constructed feeds into the process by which others (peers, immediate superiors, upper management) establish recognition or the professional reputation. In short, success will not happen without collective approval, a recurring notion in our interviews: “Individual success depends on group success.” “Professional success is the reputation you earn from the people you work with (...) All of these people have to agree in saying, ‘That guy—he really scores in his field!’ It’s a little like a professional hockey player. His peers recognize his talent, his trainer recognizes his talent, the owner of the team will pay for his talent, the fans will see him as a good player... “ “An individual builds his credibility within the organization only through projects carried out with others—and the others are the ones who judge his potential or how successful he is when they see the results. Success emerges through team projects.” Secondly, this consensus deserves to be given nuance. Slightly less than half of the respondents spontaneously defined success in a way that matched the interpretation of the success-construction process as collective. However, others, while supporting the notion that success is constructed collectively, stated that a manager’s success occurred primarily through the fulfilment of personal ambition, whether purely professional in nature or of the “work/family balance” kind. In sum, for the second group of respondents, even though success also corresponded to a social construct, it could not be called true success if the individual who was experiencing it didn’t recognize his or her own active role in the process, particularly that of personal concerns and, above all, ambition. Moreover, the respondents who insisted that success is largely subordinate to personal satisfaction were naturally more inclined to couch their judgment in individual parameters (e.g. political skill, the ability to assemble peers around a common project, skill in presenting files, etc..), whereas the others framed their viewpoints by citing the role of conformity in the context of organizational strategy. Third, the responses supported the idea that success also rested on the manager’s propensity to take the helm and become deliberately involved in interactions through which his or her potential might be recognized. Such recognition was gained through a series of micro achievements that surpassed the actual requirements of the position: managers voluntarily marked themselves out from their peers through little ‘extras’ that fell outside normal responsibility. However, these micro achievements mainly serve to illustrate compliance with the established institutional 25 framework. More specifically, such extras bear similarities to the ideas for change that inform the strategy process: “For a manager to be successful, he or she has to be an agent of change. A manager who is happy to keep the status quo shouldn’t be a manager… However, it’s true that the public organization has its own set of values and ways of doing things (…) And, unfortunately for the middle manager, the public organization isn’t going to be the one to change or model itself according to what a middle manager thinks it ought to be (…) Public organizational structure and culture stifles its managers (…) It’s the manager who has to fit in, saturate himself (…)” Fourth, success translated into a value-added contribution to organizational strategy. Individual success that did not ultimately inform organizational success could not last. If it didn’t serve to sustain organizational success, individual success translated into an individualistic and/or careerist attitude that was diametrically opposed to recognition within the organization (“Success doesn’t just come about through major coups”; “It’s important not to cut corners if you want to establish credibility in the eyes of others”). Fifth, the results demonstrated that success is constructed from the bottom up; it begins with horizontal recognition of peers (those within range of interaction), followed by vertical recognition from upper management. Sixth, the results support the idea that success contains an element of ‘luck,’ which refers to timing and contextual contingencies (e.g. financial, human and/or material resources). Lastly, the results clearly indicate the existence of an element of taboo that is tied in with individual success. Talk of success did not happen without talk of those who had achieved it. To publicly name top performers was seen as problematic: it runs counter to team spirit. Discussion We wish to highlight three pivotal elements in our results. First, based on the meanings extracted from our data, individual success is essentially a process of social construction achieved through a series of daily actions. In this sense, our data concur with the strategy process described by Avenier and his concept of strategy-in-action or strategizing. Second, recognition, credibility or reputation was established through micro achievements that surpassed the stated requirements of a given occupation. In addition, these micro achievements corresponded to the micro ideas of change at the heart of strategy formation. However, these initiatives fed into the official strategy only in a marginal sense. This indicated that individual success was primarily constructed along the lines of the induced strategy process described in Burgelman’s model. Put in other terms, success emerged through compliance with the official strategy. Furthermore, this corporatist attitude to success is akin to the manager’s strong company loyalty discussed by Kanter. Whether initiatives represent marginal or radical change, Kanter’s thesis appears validated through our study’s findings that group success invariably supplants individual success. The ascent of group success to the detriment of visibility for the local actor/agent of change brings us back to the taboo surrounding individual success that was described by our respondents. The interdict emerging from our analysis gave us to understand that deviance or non-conformity is quite 26 simply off limits. This in turn allows us to imagine that the process by which success is constructed is first and foremost a reproduction of the organization’s identity, a process that supplants the individual and banishes the marginality of his or her ideas for change. In sum, the organization is essentially a greenhouse within which managers subscribe to the conformist attitude that, generated by the upper echelons, permeates the organizational context and culture. The creativity, projects for change and local initiatives that were most publicly supported by managers appeared to have been powerfully inspired by great strategic decisions which took place on a top-down level. Before they even had a chance to emerge, micro achievements were conditioned by the word from the top. As we noted, a manager’s success must first pass through a consensus on his or her worthiness to be recognized, credibility and reputation. Most importantly, this recognition, credibility and reputation that depends so narrowly on group approval is fundamentally or invariably enmeshed with a comparative process through which managers’ reputations are held up and measured against each other’s. The comparative process, based on the idiosyncratic recognition, credibility and reputation attributed to each manager, is ultimately where managers may distinguish themselves from their peers. Finally, our study confirms that success is not only a social construct, it is also the result of voluntary actions on the part of successful managers who know how to sell their ideas, or convince others of the good of their initiatives or projects. As such, success is also the result of political savvy. Put in other terms, our study does not contradict current wisdom in the field of Organizational Behaviour. Conclusion While not especially contradicting OB studies, our results uphold the concept of individual success as a social construct. Above all, our findings indicate that individual success is largely constructed through a manager’s conformist contributions to the strategy process. A manager thus stood out from his or her peers primarily through the inductive strategy process described in Burgelman’s model. While such visibility appears to transgress the boundaries of certain “taboos,” it doesn’t detract from the fact that individual success remains a well-established phenomenon within the organization. It emerges, reproduces, is shaped and modified in the shadows yet somehow in tandem with the strategy-in-action or strategizing that likewise enables official strategies to emerge and be implemented, shaped, and refined. The manager effectively stands out from his or her peers through multiple local initiatives. Our case study has encouraged us to further investigate the idea that a “favourable place” for the construction of professional success does indeed exist, and that this place might well be the process of strategy formation. As a final word, while the scope of the present study (which only concerned a single organization) is necessarily limited, we believe that future research—particularly research aiming to support the argument that individual success is a social construct—would do well to draw up a strategy that entailed a comparison of different organizational contexts. References Ackah, Carol & Heaton, Norma, “The relity of “new careers for men and for women”, Journal of European Industrial Training. Bradford, 28(2-4), (2004), 141-158 27 Anderson, Carl, “Values-based Management”, The Academy of Management Executive, 11(4), (November 1997), 25-47 Adler, Gordon, Vroom, Victor, Rokoff, June, Olsen, David, Burnham, David H.., “When your star performer can’t manage”, Harvard Business Review . 75(4), (Jul/Aug 1997), 22-27 Avenier, Marie-José, La stratégie chemin faisant, Paris, Economic, 1997 Barnett, William, P., Burgelman, Robert, A., “Evolutionary perspective on strategy”, Strategic Management Journal, 17, (Summer 1996), 5-20 Bartunek, Jean M.., “Changing Interpretive Schemes and Organizational Restructuring : The Example of a Religious Order”, Administrative Science Quarterly, 29(3), (September 1984), 355-373 Baruch, Yehuda, “Transforming careers: from linear to multidirectional career paths: Organizational and individual perspectives”, Career Development International, 9 (1), (2004), 58-68 Bower, Joseph, L., Managing the resource allocation process: a study of corporate planning and investment, Boston, Division of Research, Graduate School of Business Administration, Harvard University, 1970. Bunker, Kerry A., Kram, Kathy E. &, Ting, Sharon, “The Young and the Clueless”, Harvard Business Review, 80(12), (December 2002), 80-87 Burgelman, Robert, A., Strategy is Destiny, How Strategy-making shapes a Company’s Future, Free Press, 2002 Burgelman, Robert, A., “A process model of strategic business exit: Implications for an evolutionary perspective on strategy”, Strategic Management Journal, 17, (1996), 193-215 Cliffe, Sarah, “What a star – What a jerk”, Harvard Business Review, 79(8), (September 2001), 37-48 Cohen, Laurie & Mallon, Marie, “My brilliant career ? Using stories as a methodological tool in careers research”, International Studies of Management & Organization, 31(3), (2001), 48-68 Thomas J DeLong Thomas J.& Vijayaraghavan, Vineeta, “Let’s hear it for B players”, Harvard Business Review, 81(6), (June 2003), 96-104 Eisenhardt, Kathleen M & Brown Shona L. “Competing on the edge: Strategy as structured chaos”, Long Range Planning, 31(5), (October 1998), 786-790 Eisenhardt Kathleen M & Sull,Donald N.,, “Strategy as simple rules”, Harvard Business Review, 79(1), (January 2001), 106-116 Falcoz, Christophe, “Les cadres à haut potentiel ou l'obligation de réussite”, in Cadres: la grande rupture, Éditions La Découverte et Syros Paris, directed by Paul Bouffartigue, (2001), chap. 12, 221-239 Fiol, Marlene C., “Intraorganizational Cognition and Interpretation”, in Companion to Organizations, Éditions Blackwell, (2004) p. 119-137 Glaser, Barney G. & Strauss, Anselm L, The discovery of grounded theory; strategies for qualitative research, StraussNew York : Aldine, (1967) Groysberg, Boris, Nanda, Ashish & Nohria, Nitin, “The Risky Business of Hiring Stars”, Harvard Business Review, 82 (5), (May 2004), 92-152 Hamel, Gary & Prahalad, C.K., Competing for the future , Boston, Mass. : Harvard Business School Press, c1994. Hamel, Gary, Leading the Revolution, Boston : Harvard Business School Press, c2000. Hamori, Monica, “The impact of reputation capital on the career paths of departing employees”, Journal of Intellectual Capital, 4(3), (2003), 304-315 Hall, Douglas T. (2004) “The Protean Career: A quarter-century journey”, Journal of Vocational Behavior, 65(1), (August 2004), 1-13 Howell, Jane M. & Boies, Kathleen “Champions of Technological Innovation: The Influence of Contextual Knowledge, Role Orientation, Idea Generation, and Idea Promotion on Champion Emergence”, Leadership Quarterly, 15 (1), (Febuary 2004), 123-144 28 Ibarra, Herminia, “Network centrality, Power, and Innovation Involvement: Determinants of Technical and Administrative Roles”, Academy of Management Journal, 36(3), (June 1993), 471-502 Jarzabkowski, Paula, “Strategic Practices: An Activity Theory Perspective on Continuity and Change”, Journal of Management Studies, 40(1), (January 2003), 23-55 Jones, Nory B., Herschel, Richard T.& Moesel, Douglas D. “Using "knowledge champions" to facilitate knowledge management”, Journal of Knowledge Management, 7(1), (2003), 49-78 Kanter, Rosabeth Moss, The Change Masters, Simon & Shuster Ed., New York, c1983 Klein, Jeremy, “Behond Competitive Advantage”, Strategic Change,11(6), (September-October 2002), 317-327 Lant, Theresa K., “Organizational Cognition and Interpretation” in Companion to Organizations, Éditions Blackwell, (2004), 344-362 Lee, Mary D., Dohring, Pamela L. & Karakas, Farhi, “The Meaning of Career Sucess and Changing Career Conceptions Among Part-Time Professionals”, McGill, in the process of being published (2003) Ludeman, Kate & Erlandson, Eddie, “Coaching the Alpha Male”, Harvard Business Review, 83(5), (May 2004), 58-69 Luthans, F., Hodgetts, R.M., Rosenkrantz, S.A. (1988) Real Managers, Ballinger Publishing Maccoby, M. (1976) The Gamesman, Simon & Shuster Ed. Messmer, M. (2004) Recognizing Potential Stars by promoting from within, Strategic Finance, Vol. 86, I. 4, p. 9 Michaud, Pascale, La course, Comment les cadres talentueux font leur chemin dans la grande entreprise, Les Éditions Transcontinental inc., 2003 Mintzberg, Henry. (1983) Power in and around Organizations, Englewood Cliffs, Prentice-Hall Mintzberg, Henry & Waters, James A. “Of Strategies, Deliberate and Emergent”, Strategic Management Journal, 6, (1985), 257-273 Park, Jaesun & Chong, John K.S. “A comparison of absolute and relative performance appraisal system”, International Journal of Management, Vol. 17, I. 3, (september 2000), 423-438 Reitman, Freida & Schneer, Joy A. (2003) The promised path: A longitudinal study of managerial careers, Journal of Managerial Psychology, 18 (1/2), (2003), 60-76 Rouleau, Linda, “La formation des stratégies sous le microscope”, Canadian Journal of Administrative Sciences / Revue canadienne des sciences de l’administration, 19(1), (2002), 12-26 Salvato, Carlo, “The Role of Micro-Strategies in the Engineering of Firm Evolution”, Journal of Management Studies, 40 (1), (January 2003), 83-108 Sullivan, Sherry E., Martin, David F., Carden, William A. & Mainiero, Lisa A., “The Road Less Traveled: How to Manage the Recycling Career Stage, Journal of Leadership & Organizational Studies, 10 (2), (2003), 34-42 Watson, Douglas J. & Hassett, Wendy L., “Career Paths of City Managers in America's Largest Council-Manager Cities”, Public Administration Review, 64(2), (2004), 192-200 Weick, Karl, Sensemaking in Organizations, Sage Publications, 1995 Weil, Peter A. & Kimball, Peter A., “Determinants of salaries earned: Comparisons of black and white healthcare managers”, Journal of Organizational Behavior, 17 (6), (1996), 657-679 Zorn, Theodor E.& Violanti, Michelle T.,”Communication abilities and individual achievement in organizations”, Management Communication Quarterly, 10 (2), (1996), 139-168 29 ASAC 2005 Toronto, Ontario Parshotam Dass I. H. Asper School of Business University of Manitoba Na Ni (student) I. H. Asper School of Business University of Manitoba GOT RHYTHM? TESTING A MODEL OF THE RHYTHM OF CHANGE IN PRODUCT DIVERSITY In this paper, we test a model of the rhythm of change in product diversity, using pooled data from a random sample of 100 manufacturing firms over 1990-2001. The results support the relationships of the confluence of antecedents, the rhythm of change in product diversity, and subsequent adjusted return on assets. INTRODUCTION Most research on strategic management in general, and product diversification in particular, has focused on the magnitude of change and its consequences (Ramanujam & Venkataraman, 1989). It has largely ignored other dimensions of change such as rhythm, as well as antecedents of change. However, a few researchers have taken the lead and have shown that organizations may follow a gradual, incremental, and evolutionary or an irregular, quantum, and revolutionary model (Miller & Friesen, 1984). Organizations that change in a continuous, gradual fashion follow a more regular rhythm of change as compared to others that change in fits-and-starts. The rhythm a firm follows has consequences for its performance. Whereas some studies have shown the superiority of quantum change (e.g., Hoskisson & Galbraith, 1985; Miller & Friesen, 1984), others tend to portray the pre-eminence of continuous change in organizations (e.g., Brown & Eisenhardt, 1997; Vermeulen & Barkema, 2002). Nevertheless, most theory and research may indicate that the effectiveness of the rhythm of change may depend on its internal and external context, thereby leading to a contextual approach (e.g., Lamont, Williams, & Hoffman, 1994; Pettigrew, 1985; Pettigrew, Woodman, & Cameron, 2001). Moreover, if the rhythm that organizations follow is important for their success and survival, it is theoretically interesting and practically useful to documents its antecedents. The purpose of this paper is to empirically test a model of the rhythm of change in product diversity, including its antecedents and consequences. We take a contextual approach, and argue that the relationship of the rhythm of change with its antecedents as well as consequences may depend on their confluence or joint effects rather than independent effects. This approach is consistent with the classical schools of strategic management, e.g., design, planning and positioning, as well as the modern concepts of resources and competencies and incremental learning (e.g., Barney, 1991; Mintzberg, Ahlstrand, & Lampel, 1998). We draw from industrial organization (I/O) economics, transaction cost approach, and behavioral theory of the firm, among others, to build on the insights generated by previous research on strategic change (e.g., Aldrich, 1999; Brown & Eisenhardt, 1997; Ginsberg, 1988; Huff, Huff, & Thomas, 1992; Miller & Friesen, 1984; Rumelt, 1995; Sastry, 1997; Staw, Sandelands, & Dutton, 1981). In the following sections, first, we briefly present our model and hypotheses regarding the relationships involved. Next, we describe our research methods, followed by analysis and results. Finally, we present our discussion and conclude the paper, giving its limitations as well as 30 contributions. Given limitations of space, our focus in this paper is on empirical testing of the relationships. The theoretical background and development of the model as well as details of the hypotheses have been presented elsewhere in a conceptual paper. RHYTHM OF CHANGE IN PRODUCT DIVERSITY: MODEL AND HYPOTHESES Consistent with most schools of strategic management and integrating insights from I/O economics, transaction cost analysis, and behavioral theory of the firm, among other theoretical perspectives, we follow a contextual approach and conceptualize strategic change as the result of confluence of antecedents, including the internal context, external context and prior performance (Andrew, 1971; Ginsberg, 1988; Huff et al., 1992; Mintzberg et al., 1998; Pettigrew, 1985; Pettigrew et al., 2001; Porter, 1991; Sastry, 1997). Also, we propose that the strategic change within its context may lead to varying levels of subsequent organizational performance. To test our model, we choose three variables, one each from the internal, external and prior performance arenas, and examine their joint effects on the rhythm of change in product diversity. We also examine the combined influence of the rhythm of change and its contextual variables on subsequent organizational performance. Industrial organization economics began with its Bain-Mason paradigm of market structure-conduct-performance (Rumelt, 1974; Porter, 1991) with its emphasis on external pressures for and resistance to change. These external factors, such as technology, regulations, institutional rules, as well as industrial attributes may influence the market demand in terms of the well-known construct of industry munificence/scarcity (Dess & Beard, 1984; Keats & Hitt, 1988; Wiersema & Bantel, 1993). In this paper we selected industry scarcity/munificence as the external factor because: a) it has been shown to influence strategic change in product diversity (Wiersema & Bantel, 1993) as well as its subsequent performance (Palich, Cardinal, & Miller, 2000; Wan & Hoskisson, 2003); b) it presents a potential opportunity/threat to the vital interests of the firm (Staw et. al., 1981); and c) it could be reliably measured with well-developed measures in the literature using multiwave data available to us (Dess & Beard, 1984; Rogosa, Brandt, & Zimowski, 1982). Similar to the external pressures, internal pressure for or resistance to change may influence the organization’s willingness as well as capability to change. According to transaction cost theory—a theory that has guided considerable research on diversification and change—the nature of the assets in terms of their specificity increases the costs of unrelated change (Williamson, 1975, 1985). Related product diversification may not induce huge sunk cost, but it is not true for unrelated diversity. Additionally, existing explicit and implicit knowledge in the organization may prevent the organization from diversifying into some new unrelated industries. According to Williamson (1985), asset specificity is the most important construct in transaction cost theory. We chose asset specificity/flexibility as an indicator of organization’s internal context, because: a) it is expected to influence strategic change (Fombrun & Ginsberg, 1990), particularly product diversification (Chatterjee & Wernerfelt, 1991; Hoskisson & Hitt, 1990; Porter, 1991; Wan & Hoskisson, 2003); b) it is conceptually more distant from firm performance than is firm size, slack and risk and is therefore less likely to be correlated with firm performance; and c) it could be reliably measured using multiwave data available to us (Rogosa et al., 1982). Likewise, behavioral theory of the firm has highlighted the role of firm financial performance in strategic management, resulting rigidities and fluidities, problemistic search and learning (e.g., Cyert & March, 1963; Greve, 1998; Haveman, 1993; March & Sutton, 1997; Staw et al., 1981). Success can induce inertia in the organization (Boeker & Goodstein, 1991; Hannan & Freeman, 1984; Huff et al, 1992; Nelson & Winter, 1982; Oster, 1982; Romanelli & Tushman, 1994), while failure creates pressure for organizations to change (Duhaime & Grant, 1984; Kahneman & 31 Tversky, 1979). Ignoring the role of firm performance misses the role of feedback of internal and external stakeholders through learning, risk-taking, slack, and aspirations, thereby ultimately resulting in researchers’ use of “simple models of complex worlds” (March & Simon, 1997: 700). Hence, we consider firm’s financial performance as an antecedent as well as an outcome. In this paper, we integrate contributions from process as well as variance theories, thereby employing “variance type predictions” (Mohr, 1982: 69) to test a model of the rhythm of product diversity. We propose that industry scarcity, asset specificity, and prior firm performance set in motion a virtuous or vicious cycle, resulting into a regular or irregular rhythm of change, respectively. Therefore, each of the conditions is deemed necessary but not sufficient for strategic change. In other words, interactive effects of the constructs involved provide much stronger and compelling rationales for the likelihood of the rhythm of change. Hence, we eschew the independent effects as well as their two-way interactions in favor of a three-way interaction of the variables considered in the study. None of the conditions by themselves are likely to have a significant influence on the possibility of strategic change. However, they reinforce one another through positive feedback loops resulting into a significant effect over time. We improve upon previous attempts by testing the complex (three-way) interactions among the antecedents of the rhythm of change, going beyond the main effects of internal, external, and performance variables (Fombrun & Ginsberg, 1990) and their simple (two-way) interactions. We also examine the consequences of the rhythm of change by using complex interactions, going beyond simplistic models of continuous versus quantum change (Park, 2003; Vermeulen & Barkema, 2002). In the paragraphs below, we present two hypotheses of the relationships of the rhythm of change, its consequences as well as antecedents. Rhythm of Change and Subsequent Financial Performance A contextual approach implies that there is no one best strategy for change in all organizations; rather, the best strategy of change for an organization may depend on its internal and external context. For their continuous success or survival, organizations try to balance between two types of matches: a) the fit of the organization with its environments; and b) the fit of the internal organizational environment (e.g., Miller & Friesen, 1984). Any organization that focuses on either one of these two fits and ignores the other may do so only at its own peril. Accordingly, rather than raising the question, which rhythm, regular or irregular is best for the organizations, we ask: Under what conditions, is a regular (or irregular) rhythm good for organizations, in terms of their financial performance? For example, when industry scarcity is high, there may be fewer opportunities within the industry, and more firms may be desperate to escape their own industry to diversify into ‘greener pastures’ (Rumelt, 1974; Stimpert & Duhaime, 1997; Wan & Hoskisson, 2003). Therefore, they may be frantically searching for opportunities to do so. Flexible assets may enable them in this search process as well as lower their transaction costs to capitalize on windows of opportunity and increase fluctuations of changes. Such reactive moves may turn out to be more haphazard, accidental, episodic and opportunistic rather than deliberate, purposeful, systemic and strategic, thereby resulting in lower returns (Vermeulen & Barkema, 2002). In contrast, an organization that avoids the temptation to rush out with the herd—a reactive behavior—may be able to scrutinize its opportunities more fully on its own time and terms. Consistent with the design school approach (Andrew, 1971), this approach may enable it to match external opportunities with its own resources and competencies (Barney, 1991; Porter, 1991) to stick to its knitting (Peters & Waterman, 1982)—in terms of content (e.g., positioning school) and/or process (Mintzberg et al., 1998) because of underlying economies, culture, risks and competencies. Moreover, such circumstances may allow the firm to manage and digest the diversification in a timely manner without creating disruptions and diseconomies, given 32 constraints of internal procedures, management talent, cognitive limits of managers, absorptive capacity and learning, thereby resulting into higher returns (Cohen & Levinthal, 1990; Vermeulen & Barkema, 2002; Wan & Hoskisson, 2003). Therefore, Hypothesis 1. A regular rhythm of change in product diversity will lead to higher levels of subsequent return on assets than an irregular rhythm when industry scarcity and asset flexibility are high. Rhythm of Change and its Antecedents Consistent with our model, any organization that focuses its vision, strategy or change either on internal or external conditions rather than balancing them, including its performance feedback, may do so at its own risk. In other words, each of the selected three variables, industry scarcity, asset flexibility and prior financial performance are likely to influence the rhythm of change. However, none of them are expected to be significant enough to cause change by themselves, as revealed in several empirical studies (Birnbaum, 1984; Graham & Richards, 1979; Harrigan, 1981; Jauch, Osborne, & Glueck, 1980; Oster, 1982). Therefore, the effect of any one variable is expected to depend on the level of the other two variables. For example, Boeker and Goodstein (1991: 822) note that, "environment changes influence organizations only to the extent managers can act on them" through processes of problem recognition, signaling, analysis of deviations, and corrective action (Kiesler & Sproull, 1982). According to the I/O economics theory, industry scarcity may signal declining support for firm’s products and services as new entrants and increasing supply decrease their ability to appropriate rents (Porter, 1991). This may provide incentive for the firm to initiate experimentation for activities that precede change (Dutton & Freedman, 1985; Lawrence & Dyer, 1983) and increase the likelihood and/or urgency to escape its adverse industry environment into more attractive—related or unrelated—industries (Christensen & Montogomery, 1981; Rumelt, 1974; Stimpert & Duhaime, 1997; Wiersema & Bantel, 1993). However, it may be difficult to do so unless there are other internal pressures to lower inertia resulting from specific assets and from good firm financial performance. For instance, borrowing from the transaction cost theory, we argue that managers may have already invested into specialized assets due to a specific strategy to avoid opportunism by their partners (Williamson, 1975). They may escalate their commitment further (Staw & Ross, 1987), which is likely to constrain their choices and increase the transaction costs (Chatterjee & Wernerfelt, 1991; Montgomery & Wernerfelt, 1988; Porter, 1991; Williamson, 1975) to lower their discretion to change quickly in response to the rare windows of opportunity available in scarce markets. On the other hand, flexible assets may signal a prior general strategy, which may be considered more akin to organizational capital or core skills that is likely to facilitate diversification (Hoskisson & Hitt, 1990). In spite of facing adverse industry environment, managers may not have the incentives or legitimacy to use their flexible assets as long as the organization is functioning smoothly and performing well. According to behavioral theory of the firm, good performance may reinforce existing strategic processes and change may not be noticed enough to be part of the organizational discourse (Boeker & Goodstein, 1991; Hannan & Freeman, 1977; Huff et al., 1992; Johnson, 1988; Kiesler & Sproul, 1982; Nelson & Winter, 1982; Oster, 1981; Romanelli & Tushman, 1994). In contrast, failure may undermine confidence in current commitments, stimulate stakeholders to action, encourage managers to search for solutions, take risks or lose their positions, and provide impetus for quick, even unplanned change (Boeker & Goodstein, 1991; Kahneman & Tversky, 1979; March & Sutton, 1997; Sastry, 1997; Staw et al., 1981; Wiersema & Bantel, 1993). 33 In short, industry scarcity may exert external pressure to react and diversify irregularly, however, organizations with high financial performance would not have such momentum. Similarly, firms where asset flexibility is low would be constrained to capitalize on windows of opportunity for irregular diversification. On the other hand, firms facing scarce industries, low financial performance and high asset flexibility have greater pressure and much higher latitudes for irregular changes in product diversity. Hence, we propose a joint effect expressed in terms of a three-way interaction among these variables. Hypothesis 2. High industry scarcity, low financial performance and high asset flexibility will amplify the effect of one another to lead to a more irregular rhythm of changes in product diversity. METHOD Sample Data used to assess the hypotheses were obtained from the COMPUSTAT II database. We selected a random sample of 100 companies registered on U.S. stock exchanges in SIC codes 20-39 (manufacturing) that reported complete data on the variables of interest over the time period of the study from 1990 to 2001. The sample includes a wide range of small, medium and large organizations. For example, for the year 2001, the number of employees varied from 6 to 168,000, with an average of 14,499. Similarly, average assets of the firms in the sample for the year 2001 were $ 3.938 billion, ranging from $ .793 million to $78.863 billion. Likewise, net sales for the year 2001 ranged from $ .918 million to $ 93.744 billion, with an average of $ 3.832 billion. All variables are computed over three year period (Bergh & Fairbank, 2002; Dess & Beard, 1984; Rogosa et al., 1982). For example, subsequent performance for the year 2001 is computed using data from 1999 to 2001. Likewise, firm performance for the year 1992 is computed from data for years from 1990 to 1992. We computed data for each year and then pooled them. To test our two hypotheses, we used two regression equations. The first regression equation included rhythm of change as the dependent variable (1993 to 2001) and antecedents of rhythm—industry scarcity, financial performance, and asset flexibility—and control variables as exogenous variables (1992-2000). Data for this regression equation were pooled over nine years, giving a sample size of 900. The second regression equation used subsequent performance as the dependent variable (1994 to 2001), rhythm of change, industry scarcity and asset flexibility as the exogenous variables, including the controls such as prior performance (1993-2000). Data for the second regression equation were pooled over eight years, giving a sample size of 800. These time periods reflect realistic strategic planning horizons and have been used by other researchers (e.g., Boeker & Goodstein, 1991; Chang & Thomas, 1989; Finkelstein & Hambrick, 1990; Keats & Hitt, 1988; Miller & Friesen, 1984). In addition, we improve upon other studies (e.g., Grant et al., 1988) by ensuring that our findings are not favorably influenced by temporal overlap between the independent variables and the dependent variable. Measures Subsequent Firm Performance. Subsequent firm performance is measured by assessing adjusted return on assets (Bettis & Mahajan, 1985; Stimpert & Duhaime, 1997). Return is calculated as after tax profits plus interest received divided by total assets. The adjustment makes a firm's return on assets relative to the average returns for a firm with an identical diversification profile. Rhythm of change. Rhythm of change in related and unrelated product diversities are measured using entropy indexes (Palepu, 1985). The construct validity of these measures has been established by studies in the diversification area (e.g., Hoskisson, Hitt, Johnson, & Moesel, 1993). 34 Instead of using the difference score measure, we used the least squares measure (Bergh & Fairbank, 2002; Dess & Beard, 1984; Keats & Hitt, 1988; Rogosa et al., 1982; Wiersema and Bantel, 1993). The exact measure is the standard error of the regression coefficient, using sales as the dependent variable and time as the independent variable. Industry Scarcity. Industry scarcity (opposite of industry munificence) is measured following Dess and Beard (1984), Keats and Hitt (1988) and Wiersema and Bantel (1993). The munificence measure essentially assesses the growth of sales of an industry over time. The exact measure is the regression coefficient obtained for the firm's four-digit SIC code industry from aggregate sales divided by mean sales (to adjust for absolute industry size) of the industry. The coefficient is weighted for each firm on the basis of the percent of the firm’s sales accounted for by each four-digit industry in which the firm operates. It then is multiplied by –1 to assess industry scarcity (versus munificence). Prior Firm Performance. Prior firm performance is measured the same way as the subsequent firm performance, mentioned above. Asset Flexibility. Asset flexibility is measured as the ratio of financial assets to tangible (physical) assets. Financial assets are relatively flexible, whereas tangible assets are comparatively inflexible. The latter represent particular resource commitments to specific products or services (Chatterjee & Wernerfelt, 1991). Controls. Based on theory and prior research, we used a number of controls to avoid confounding the results. These included GDP, firm size, organizational slack, initial level of product and international diversities, and magnitude of change in international diversity (e.g., Ginsberg & Buchholtz, 1990; Lamont et al., 1994; Zajac & Kraatz, 1993). We used measures for these variables employed in previous research (Bettis & Mahajan, 1985; Bourgeois, 1981; Fiegenbaum & Thomas, 1988; Finkelstein & Hambrick, 1990; Hambrick & D'Aveni, 1988; Hill & Hansen, 1991; Hitt, Hoskisson, & Kim, 1997; Lubatkin & Chatterjee, 1994). All independent and control variables associated with related diversification are standardized with respect to firm's dominant 4-digit SIC codes, whereas those associated with unrelated diversification are standardized with respect to firm’s 2-digit SIC codes to render them comparable across industries within the set describing general manufacturing (SIC 2000-3999). ANALYSIS AND RESULTS Rhythm of Change and Subsequent Performance We examined the correlations (not shown) among the control variables, the rhythm of change in related and unrelated product diversity, and the subsequent performance. The highest correlation (r = .75; p < .001) is between prior firm performance and subsequent performance, as expected (Park, 2003). The second largest correlation coefficient is between firm size and initial international diversity. All other relevant correlation coefficients among the exogenous variables are less than .36. An inspection of the collinearity diagnostics revealed no multicollinearity problem. Using pooled data for regression analyses has its own strengths and challenges. We tested for the presence of heteroscedasticity and autocorrelation in the data and found evidence for the presence of both. Therefore, generalized least squares (GLS) instead of ordinary least squares (OLS) are considered appropriate for data analysis (e.g., Hill & Hansen, 1991). GLS involves correction for heteroscedasticity and autoregression before application of the OLS. We use PROC AUTOREG with the Yule-Walker method in SAS for correcting autocorrelation. The Yule-Walker method is similar to a sophisticated Cochrane-Orcutt method except that it retains the first observation. Judge and colleagues (1985) term this method as Estimated Generalized Least Squares (EGLS). Due to the transformations involved, statistics such as F and R square are not 35 considered reliable (Greene, 2000; Judge et al., 1985). The results of these analyses for the rhythms of change in related and unrelated diversities are presented in Table 1. Note that lower and higher levels of the rhythm of change in all regression analyses in this paper represent regular and irregular rhythms of change, respectively. ----------Insert Table 1 about here---------Hypothesis 1 proposed that a regular rhythm of change in product diversity will lead to higher subsequent adjusted return on assets than an irregular rhythm when industry scarcity and asset flexibility are high. To avoid confounding of these two three-way interactions, we entered the controls in Model 1, main effects of antecedents in Model 2, and two-way interactions in Model 3, followed by the three-way interactions in Model 4 (Table 1). All models were significant overall (p < .001). The three-way interactions were found to be statistically significant, thereby supporting Hypothesis 1 both in case of the rhythm of change in related and unrelated product diversities. We discuss the results of the controls before proceeding to the discussion of the hypothesized relationship of the rhythm of change and subsequent performance. Controls. We used a number of controls to avoid confounding the results of the relationships of principal interest in the study. The controls presented in Model 1 (Table 1) reveal some interesting relationships. First, there was no significant relationship between subsequent performance and magnitude of change in unrelated product diversity, which is not surprising. However, the effects of the magnitude of change in related product and international diversities, as well as firm risk were negative. Moreover, the relationship of firm size, slack, and GDP were not statistically significant. Main Effects. To further avoid confounding of results for Hypothesis 1, we entered industry scarcity, prior firm performance, and asset flexibility in Model 2. The positive relationship of prior performance with subsequent performance is obvious. We also entered rhythm of change in related and unrelated product diversities in this model as well. The results reveal that a regular rhythm of change in unrelated product diversity improves subsequent adjusted return on assets. The rhythm of change in related product diversity did not significantly influence subsequent firm performance. Two-way Interaction Effects. We entered the two-way interactions of rhythm of related and unrelated change with industry scarcity and asset flexibility, in addition to the industry scarcity x asset flexibility in Model 3 (Table 1) to partial out the influence, if any, of these interactions before entering the three-way interaction, which is of major interest in the study. The results show that a regular rhythm of change in related product diversity leads to a higher level of subsequent adjusted return on assets when industry scarcity is high. Similarly, a regular rhythm of change in unrelated product diversity improves subsequent adjusted return on assets when asset flexibility is high. Three-way Interaction Effect. As mentioned above, when two multiplicative terms reflecting the effect of rhythm of change in related and unrelated product diversities with industry scarcity and asset flexibility are added in Model 4 (Table 1), they make a statistically significant contribution to the subsequent adjusted return on assets. The regression coefficients of the three-way interactions for related and unrelated rhythms are -.07 (p < .05) and -.06 (p < .05), respectively. As predicted in Hypothesis 1, the influence of the rhythm of change in related and unrelated product diversities appears to be contingent on both industry scarcity and asset flexibility. Consistent with expectations, a regular rhythm of change in related and unrelated product diversities facilitates subsequent adjusted return on assets when both industry scarcity and asset flexibility are high. Therefore, these results strongly support Hypothesis 1. Rhythm of Change in Related and Unrelated Product Diversities A look at the correlations (not shown) reveals that the rhythms of change in related and unrelated product diversities are positively associated with initial levels of related and unrelated 36 product diversities, respectively. The highest correlation among the exogenous variables is the positive association between firm size and initial international diversity (r = .49, p < .001) followed by the positive association of rhythm of change in related product diversity and initial related product diversity (r = .35, p < .001). All other correlations among the exogenous variables were less than .27. An inspection of the collinearity diagnostics revealed no multicollinearity problem. Hypothesis 2 predicted a three-way interaction of industry scarcity, prior firm performance and asset flexibility on the rhythm of change in related as well as unrelated product diversities. As depicted in Table 2, to avoid confounding of our hypothesized results, we entered the controls in Model 1, main effects of antecedents in Model 2, two-way interactions in Model 3, and the three-way interaction in Model 4 (Table 2). All models were significant overall (p < .001). To test Hypothesis 2, three-way interaction was found to be statistically significant, thereby supporting it both in case of the rhythm of change in related and unrelated product diversities. We discuss the results of the controls before further discussion of the hypothesized results. ----------Insert Table 2 about here---------Controls. We used a number of controls to delineate the results of the relationships of prime interest in the study. The controls presented in Model 1 (Table 2) reveal two relationships that are similar across the rhythms of change in related and unrelated product diversities. First, firms with high initial product diversities make more irregular changes. Second, the effect of the magnitude of change in international diversity on the rhythm of change in both related and unrelated diversities seems to be statistically insignificant. All other relationships appear to vary across the rhythms of change in related versus unrelated product diversities. For example, the relationship of the magnitude and rhythm of change in product diversities varies with respect to related versus unrelated product diversity. Firms that make more change in the magnitude of related product diversity follow a more regular rhythm, whereas those that make more change in the magnitude of unrelated product diversity are likely to follow a more irregular rhythm. Similarly, higher firm slack and GDP lead organizations to change their related product diversity in a regular fashion. In contrast, these relationships are not significant with respect to unrelated product diversification. Likewise, firm size, risk, and initial level of international diversity are associated with the rhythm of change in unrelated product diversity but not in the case of related diversification. Firms that had higher variability of returns and initial international diversity seem to prefer a more irregular mode of change in unrelated product diversity. In contrast, firms with higher number of employees appear to follow a more regular rhythm of change in unrelated product diversity. Main Effects. To avoid confounding of results for Hypothesis 2, industry scarcity, firm performance, and asset flexibility are entered in Model 2. Though we expected that each will incline firms to change their rhythm in a regular or irregular fashion, we did not expect them to make significant influences by themselves, therefore, we did not hypothesize their main effects. Two of the variables—industry scarcity and prior firm performance—meet our expectation in related as well as unrelated product diversity. However, we find that asset flexibility has a main effect. Interestingly, the effect of flexibility is positive with respect to the rhythm of change in related product diversity (β = .08; p < .01), whereas it is negative with regard to the rhythm of change in unrelated product diversity (β = -.07; p < .01). In other words, whereas asset specificity (the opposite of flexibility) leads firms to change their related product diversity in a more regular pattern, asset flexibility facilitates firms to change their unrelated product diversity in a more regular fashion. Two-way Interaction Effects. We entered the two-way interactions of industry scarcity x firm performance, industry scarcity x asset flexibility, and firm performance x asset flexibility in 37 Model 3 (Table 2) to partial out the influence, if any, of these interactions before entering the three-way interaction, which is of primary interest in the study. Similar to the main effects, we did not expect two-way interactions to play a statistically significant role in the rhythm of change in related and unrelated product diversities. Four of the six two-way interactions met that expectation. However, we found that two interactions were statistically significant. These were industry scarcity x asset flexibility and industry scarcity x firm performance with respect to related and unrelated product diversification, respectively. Higher levels of industry scarcity and asset flexibility reinforced each other to lead to more irregular rhythm of change in related product diversity. Likewise, higher levels of industry scarcity and lower performance reinforced each other to stimulate a more irregular rhythm of change in unrelated product diversity. These interactions, though not hypothesized, act in the expected directions. Three-way Interaction Effect. As mentioned above, when a multiplicative term reflecting the effect of industry scarcity x financial performance x asset flexibility is added in Model 4 (Table 2), it makes a statistically significant contribution to the rhythm of change in related and unrelated product diversities. The regression coefficients for changes in related and unrelated diversities are -.08 (p < .01) and -.03 (p < .05), respectively. Consistent with Hypothesis 2, the effect of industry scarcity appears to be contingent on both financial performance and asset flexibility. As predicted, a higher level of industry scarcity leads to a more irregular rhythm of change when performance is low and flexibility is high. The nature is of this relationship is similar in the case of the rhythm of change in related and unrelated product diversities, though the relationships are more pronounced in the case of related diversification. Therefore, these results provide strong support for Hypothesis 2 (The plots of the interactions have been omitted due to limitations of space but will be available from the authors). DISCUSSION AND CONCLUSION The objective of this study was to examine the antecedents and consequences of the rhythm of change in product diversity. Using rationale from several theoretical approaches, particularly, I/O economics, transaction cost theory, and behavioral theory of the firm, and support of the extant empirical findings, we proposed a testable model of the rhythm of change. Using data from a random sample of 100 companies pooled from 1990 to 2001, we found strong support for our two hypotheses. The first hypothesis predicted that a regular rhythm of change in related and unrelated product diversities will lead to a higher subsequent adjusted return on assets when industry scarcity and asset flexibility were high. The second hypothesis proposed that a confluence of three antecedents of high industry scarcity, high asset flexibility and poor prior firm performance—expressed in terms of a three-way interaction—will lead to an irregular rhythm of change in related and unrelated product diversities. Data analyses using GLS that involved correction for heteroscedasticity and autocorrelation provided strong support for the hypothesized effects. This study supports developing organization and management theory and research on strategic change in several ways. First, it corroborates a contextual approach to strategic change (e.g., Pettigrew, 1985; Pettigrew et al., 2001) where change ought to be seen as a confluence of various external and internal factors as well as prior performance (Ginsberg, 1988; Greve, 1998; Huff et al., 1992; Kraatz & Zajac, 2001; Sastry, 1997). This necessitates avoiding an “either or” mentality and synthesizing a number of varied theoretical approaches that help understand a phenomenon, its antecedents and consequences for academic as well as practical insights (e.g., Aldrich, 1999). It reinforces a view of organizations as adaptive, co-evolving, learning systems and incorporating complex interactions (Brown & Eisenhardt, 1997; Lewin et al., 1999), which is consistent with a new philosophy of management developing in the 21st century. 38 More specifically, this study advances research on strategic change by extending it from the magnitude of related and unrelated product diversities (Chatterjee & Wernerfelt, 1991; Hoskisson & Hitt, 1990) to a relatively neglected dimension of change—the rhythm of change. Not only do we examine the consequences of the rhythm of change in related and unrelated product diversities, but we also analyze complex interactions of antecedents, controlling for a host of external as well as internal factors—from GDP to the magnitude of related/unrelated product and international diversity (Park, 2003). These findings generally support earlier research on the role of context and confluence (Greve, 1998; Huff et al., 1992; Kraatz & Zajac, 2001; Pettigrew et al., 2001; Sastry, 1997; Wan & Hoskisson, 2003). Specifically, our study confirms results from Vermeulen and Barkema (2002) about the efficacy of regular rhythm and continuous change in organizations (Brown & Eisenhardt, 1997), albeit with the caveat that the relationship may be contingent upon external and internal conditions such as industry scarcity and asset flexibility. These results can help make sense of the apparently contradictory prescriptions and findings on change research (e.g., Lamont et al., 1994; Miller & Friesen, 1984). In other words, the moral of the tale of the hare and the tortoise is relevant within its framework. Given proper context, being steady does win the race. However, it leaves plenty of space for other contexts and other successful modes and rhythms of change. Each research has strengths as well as limitations, which may point to avenues for further research. We examined a random sample of 100 diverse manufacturing firms from SIC codes 20-39 in the U.S. so the results may be internally valid and externally generalized to other manufacturing firms in the U.S. Further research and comparison with manufacturing firms in other countries, service firms and other business and non-business organizations in the U.S. and abroad may extend its findings. It may also be useful to study specific industries and sectors and compare them to test and strengthen its findings. We have utilized pooled data from 1990-2001 and have examined lagged models. Future researchers may extend the time period and use time-series analysis. Moreover, there is dearth of research on lags and this area may offer opportunities to examine different lags and how they may influence the relationships with antecedents and consequences of change. In addition, it may be theoretically challenging as well practically useful to explore relationships and interactions among various dimensions of change. Most of the earlier research on change has focused on the magnitude of change. For example, exploring how antecedents of change interact to influence magnitude, rhythm, and direction of change may help understand the multidimensionality of change. We hope, the results of this research will encourage others to advance these findings further to develop more contextual and rigorous theories and deeper understanding of strategic changes in organizations. REFERENCES Aldrich, H. E. 1999. Organizations evolving. Thousand Oaks, CA: Sage Publications. Barney, J. B. 1991. Firm resources and sustained competitive advantage. Journal of Management, 17: 99-120. Bergh, D. D & J. F. Fairbank. 2002. Measuring and testing change in strategic management research. Strategic Management Journal, 23: 359-366. Bettis, R., & Mahajan, V. 1985. Risk/return performance of diversified firms. Management Science, 31: 785-799. Birnbaum, P. H. 1984. The choice of strategic alternatives under increasing regulation in high technology companies. Academy of Management Journal, 27: 489-510. Boeker, W., & Goodstein, J. 1991. Organizational performance and adaptation: Effects of environment and performance on changes in board composition. Academy of Management Journal, 34: 805-826. Bourgeois, L. J. 1981. On the measurement of organizational slack. Academy of Management 39 Review, 6: 29-39. Brown, S. L., & Eisenhardt, K. M. 1997. The art of continuous change: Linking complexity theory and time-paced evolution in relentlessly shifting organizations, Administrative Science Quarterly, 42: 1-34. Chang, Y., & Thomas, H. 1989. The impact of diversification strategy on risk-return performance. Strategic Management Journal, 10: 271-284. Chatterjee, S., & Wernerfelt, B. 1991. The link between resources and type of diversification: Theory and evidence. Strategic Management Journal, 12: 33-48. Christensen, H. K., & Montgomery, C. A. 1981. Corporate economic performance: Diversification strategy versus market structure. Strategic Management Journal, 2: 327-343. Cohen, W. M., & Levinthal, D. A. 1990. Absorptive capacity: A new perspective on learning and innovation. Administrative Science Quarterly, 35: 128-152. Cyert, R., & March, J. G. 1963. A behavioral theory of the firm. Englewood Cliffs, NJ: Prentice Hall. Dess, G., & Beard, D. 1984. Dimensions of organizational task environments. Administrative Science Quarterly, 29: 52-73. Duhaime, I., & Grant, J. 1984. Factors influencing the divestment decision. Strategic Management Journal, 5: 301-318. Dutton, J. M., & Freedman, R. D. 1985. External and internal strategies: Calculating, experimenting, and imitating in organizations, In R. Lamb & P. Srivastava (Eds.), Advances in Strategic Management, vol. 3: 39-47. Greenwich, CT: JAI Press. Fiegenbaum, A., & Thomas, H. 1988. Attitudes toward risk and the risk-return paradox: Prospect theory explanations. Academy of Management Journal, 31: 85-106. Finkelstein, S., & Hambrick, D. 1990. Top-management-team tenure and organizational outcomes: The moderating role of managerial discretion. Administrative Science Quarterly, 35: 484-503. Fombrun, C., & Ginsberg, A. 1990. Shifting gears: Enabling change in corporate aggressiveness. Strategic Management Journal, 11: 297-308. Ginsberg, A. 1988. Measuring and modeling changes in strategy: Theoretical foundations and empirical directions. Strategic Management Journal, 9: 559-575. Ginsberg, A., & Buchholtz, A. 1990. Converting to for-profit status: Corporate responsiveness to radical change. Academy of Management Journal, 33: 445-477. Graham, K. R., & Richards, M. D. 1979. Relative performance deterioration: Management and strategic change in rail-based holding companies. Academy of Management Proceedings: 108-112. Grant, R. M., Jammine, A. P., & Thomas, H. 1988. Diversity, diversification, and profitability among British manufacturing companies, 1972-1984. Academy of Management Journal, 31: 771-801. Greene, W. H. 2000. Econometric Analysis, 4th Edition. Prentice Hall. Greve, H. R. 1998. Performance, aspirations, and risky organizational change. Administrative Science Quarterly 43: 58-86. Hambrick, D., & D'Aveni, R. 1988. Large corporate failures as downward spirals. Administrative Science Quarterly, 33: 1-23. Hannan, M., & Freeman, J. 1984. Structural inertia and organization change. American Sociological Review, 49: 149-164. Harrigan, K. R. 1981. Deterrents to divestiture. Academy of Management Journal, 21: 306-323. Haveman, H. A. 1992. Between a rock and a hard place: Organizational change and performance under conditions of fundamental environmental transformation. Administrative Science Quarterly, 37: 48-75. Hill, C. W., & Hansen, G. S. 1991. A longitudinal study of the cause and consequences of changes in diversification in the U.S. pharmaceutical industry 1977-1986. Strategic Management Journal, 12: 187-200. 40 Hitt, M., Hoskisson, R., & Kim, H. 1997. International diversification: Effects on innovation and firm performance in product-diversified firms. Academy of Management Journal, 40: 767-798. Hoskisson, R. E. & Galbraith, C. S. 1985. The effect of quantum versus incremental M-form reorganization on performance: A time series exploration of intervention dynamics, Journal of Management, 10: 55-70. Hoskisson, R. E., & Hitt, M. A. 1990. Antecedents and performance outcomes of diversification: A review and critique of theoretical perspectives. Journal of Management, 16: 461-509. Hoskisson, R. E., Hitt, M. A., Johnson, R. A., & Moesel, D. D. 1993. Construct validity of an objective (entropy) categorical measure of diversification strategy. Strategic Management Journal, 14: 215-235. Huff, J., Huff, A., & Thomas, H. 1992. Strategic renewal and the interaction of cumulative stress and inertia. Strategic Management Journal, 13: 55-75. Jauch, L. R., Osborn, R. N., & Glueck, W. F. 1980. Short term financial success in large business organizations: The environment-strategy connection. Strategic Management Journal, 1: 49-63. Johnson, G. 1988. Rethinking incrementalism. Strategic Management Journal, 9: 75-92. Judge, G., Griffiths, W. E., Hill, R. C., Lutkepohl, H., & Lee, T.,C. 1985. The Theory and Practice and Econometrics, New York: John Wiley. Keats, B., & Hitt, M. 1988. A Causal model of linkages among environmental dimensions, macro organizational characteristics, and performance. Academy of Management Journal, 31: 570-598. Kahneman, Daniel, & Amos Tversky 1979. Prospect theory: An analysis of decisions under risk. Econometrica, 47:263-292. Kiesler, S., & Sproull, L. 1982. Managerial response to changing environments: Perspectives on problem sensing from social cognition. Administrative Science Quarterly, 27: 548-570. Kraatz, M. S. & Zajac, E. J. 2001. How organizational resources affect strategic change and performance in turbulent environments: Theory and evidence. Organization Science, 12: 632-657. Lamont, B. T, Williams, R. J. & Hoffman, J. J. 1994. Performance during 'M-form' reorganization and recovery time: the effects of prior strategy and implementation speed. Academy of Management Journal, 37: 153-156. Lawrence, P. R., & Dyer, D. 1983. Renewing American industry, New York: Free Press. Lewin, A. Y., Long, C. P., & Carroll, T. N. 1999. The coevolution on new organizational forms. Organization Science, 10: 535-550. Lubatkin, M., & Chatterjee, S. 1994. Extending modern portfolio theory into the domain of corporate diversification: Does it apply? Academy of Management Journal, 37: 109-136. March, J.G., & Sutton, R. I. 1997. Organizational performance as a dependent variable. Organization Science, 8, 698-706. Miller, D., & Friesen, P. 1984. Organizations - A quantum view. Englewood Cliffs, NJ: Prentice-Hall. Mintzberg, H., Ahlstrand, B. & Lampel, J. 1998. Strategy safari, Prentice Hall. Mohr, L.B. Explaining organizational Behavior. San Francisco: Jossey-Bass, 1982. Nelson, R., & Winter, S. 1982. An evolutionary theory of economic change. Cambridge, MA: Belknap Press. Oster, S. 1982. Intraindustry structure and the ease of strategic change. The Review of Economics and Statistics, 64: 376-383. Palepu, K. 1985. Diversification strategy, profit performance and the entropy measure. Strategic Management Journal, 6: 239-255. Palich, L., Cardinal, L. & Miller, C. 2000. Curvilinearity in the diversification-performance linkage: An examination of over three decades of research. Strategic Management Journal, 21: 41 155-174. Park, C. 2003. Prior performance characteristics of related and unrelated acquirers. Strategic Management Journal, 24: 471-480. Peters, T., & Waterman, R. 1982. In search of excellence. New York, London: Harper & Row. Pettigrew, A. 1985. The Awakening giant: Continuity and change in ICI. Oxford, England: Basil Blackwell. Pettigrew, A. M., Woodman, R. W. & Cameron, K. S. 2001. Studying organizational change and development: Challenges for future research. Academy of Management Journal, 44: 697-713. Porter, M.E. 1991. Towards a dynamic theory of strategy, Strategic Management Journal, 12: 95-117. Ramanujam, V., & Varadarajan, P. 1989. Research on corporate diversification: A synthesis. Strategic Management Journal, 10: 523-551. Rogosa, D., Brandt, D. & Zimowski, M. 1982. A growth curve approach to the measurement of change. Psychological Bulletin, 92: 726-748. Romanelli, E., & Tushman, M. L. 1994. Organizational transformation as punctuated equilibrium. Academy of Management Journal, 37: 1141-1166. Rumelt, R. 1974. Strategy, structure, and economic performance. Boston, MA: Harvard Business School Press. Rumelt, R.P. 1995. Inertia and transformation. In Montgomery, C.A. (Ed.): Resource-Based and Evolutionary Theories of the Firm. Dordrecht, Netherlands: Kluwer Academic Publishers. SAS Institute Inc. 1993. SAS/ETS User's Guide, Version 6, Second Edition, Cary, NC: SAS Institute Inc. Sastry, A. 1997. Problems and paradoxes in a model of punctuated organizational change. Administrative Science Quarterly, 42: 237-275. Staw, B. M., Sandelands, L. E. & Dutton, J. E. 1981. Threat-rigidity effects in organizational behavior: A multilevel analysis. Administrative Science Quarterly, 26: 501-524. Stimpert, J. L., & Duhaime, I. M. 1997. Seeing the big picture: The influence of industry, diversification, and business strategy on performance. Academy of Management Journal, 40: 560-583. Vermeulen, F., & Barkema, H. 2002. Pace, rhythm, and scope: Process dependence in building a profitable multinational corporation, Strategic Management Journal, 23: 637-653. Wan, W. P. & Hoskisson, R. E. 2003. Home country environments, corporate diversification strategies and firm performance. Academy of Management Journal, 46: 27-45. Wiersema, M. F., & Bantel, K. A. 1993. Top management team turnover as an adaptation mechanism: The role of environment. Strategic Management Journal, 14: 485-504. Williamson, O. E. 1975. Markets and hierarchies: Analysis and antitrust implications. New York: Free Press. Williamson, O. E. 1985. The economic institutions of capitalism: Firms, markets, and relational contracting. New York: Free Press. Zajac, E. J., & Kraatz, M. S. 1993. A diametric model of strategic change: Assessing the antecedents and consequences of restructuring in the higher education industry. Strategic Management Journal, 14: 83-102. 42 TABLE 1 Results of Generalized Least Squares (GLS) Regression of the Subsequent Adjusted Return on Assets a, b Exogenous variables Control Variables Model 1 - .06 - .18*** .10** - .09*** .04 - .22*** .02 - .01 - .06*** .00 Initial related product diversity Initial unrelated product diversity Initial international diversity Magnitude of change in international diversity GDP Firm risk Firm size Firm slack Magnitude of change in related product diversity Magnitude of change in unrelated product diversity Model 2 Model 3 - .02 .03 .03 - .04* .01 - .02 - .02 - .04 - .01 .10*** - .02 .02 .02 - .05 .00 .00 - .02 - .04 .00 .09*** .75*** - .02 - .01 .02 - .07** .77*** .04 .004 .002 .11** .05*** .04 .03 .12*** .04 Model 4 - .06 .02 .01 - .06** .02 .00 - .02 - .06* .01 .11*** Study Variables Prior firm performance Industry scarcity Asset flexibility Rhythm of change in related product diversity Rhythm of change in unrelated product diversity Rhythm of change in related product diversity × Industry scarcity Rhythm of change in related product diversity × Asset flexibility Rhythm of change in unrelated product diversity × Industry scarcity Rhythm of change in unrelated product diversity × Asset flexibility Industry scarcity × Asset flexibility - Rhythm of change in related product diversity × Industry scarcity × Asset flexibility Rhythm of change in unrelated product diversity × Industry scarcity × Asset flexibility - .04 - .02 5.17*** 69.96*** 10, 789 15, 784 Intercept F value (OLS) Degrees of freedom (OLS) a b - .02 54.81*** 20, 779 N = 800. Standardized regression coefficients are reported and p values are for two-tail tests. † p < .10; * p < .05; ** p < .01; *** p < .001 (two-tailed). 43 - .78*** .09** .01 .02 .09** .12*** .01 .06* .05 .09** .07* .06* .04 47.65*** 23, 776 TABLE 2 Results of Generalized Least Squares (GLS) Regression of the Rhythm of Change in Related and Unrelated Product Diversity a, b Related product diversity Exogenous variables Unrelated product diversity Model 1 Model 2 Model 3 Model 4 Model 1 Model 2 Model 3 Model 4 .50*** .48*** .51*** .51*** .58*** .47*** .48*** .43*** .09** .05 .06* .27*** .30*** .24*** .04 .04* .06** .02 .01 Control Variables Initial related/unrelated product diversity Magnitude of change in related/unrelated product diversity - .20*** Initial international diversity - .05 - .04 - .04 - .03 Magnitude of change in - .01 - .01 .01 .01 .00 - .01 - .02 - .02 - .07*** - .04* - .02 - .02 - .03 - .02 - .01 .01 - .01 - .02 .02 - .01 - .07 - .04 international diversity GDP Firm risk .01 .01 Firm size .01 .13* Firm slack - .02* .04** .05** .19** .14** - .08* - .10* - .03* - .04*** - .04*** .02 .01 .00 - .004 - .03 .07** .03 - .01 .03 - .03 .02 Study Variables Industry scarcity Firm performance .02 .01 .01 Asset flexibility .08** .12*** .10*** Industry scarcity × Firm performance - .004 Industry scarcity × Asset flexibility - .03 .20*** Firm performance × Asset flexibility - .01 - .14* F value (OLS) 37.04*** Degrees of freedom (OLS) 8, 891 b - .02 .03 - .02 .004 - .04* .01 .03 - .02 - .08** ×Asset flexibility a .15*** - .01 Industry scarcity × Firm performance Intercept - .07** .004 .10*** - .03* - .03* .05 .02 .05 - .05 29.55*** 23.80*** 22.20*** 1.83*** 8.17*** 6.64*** 6.74*** 11, 888 14, 885 15, 884 8, 891 11, 888 14, 885 15, 884 N = 900. Standardized regression coefficients are reported and p values are for two-tail tests. † p < .10; * p < .05; ** p < .01; *** p < .001 (two-tailed). 44 .00 - .11* ASAC 2005 Toronto, Ontario Robert H. Desmarteau Département de Stratégie des Affaires Anne-Laure Saives Département de Management et Technologie École des sciences de la gestion - UQAM, Canada LES PME BIOTECHNOLOGIQUES SONT-ELLES CONTRE-NATURE1 ? Le cas des entreprises des bio-industries au Québec (Canada) Moult PME néo-technologiques résultent d’un processus complexe d’industrialisation des innovations fondées sur les sciences du vivant. Or, comment caractériser la spécificité de l’objet-PME néo-managériale ? Pour réponse, cet article propose une instrumentation du concept de modèle d’affaires pour scruter la diversité des PME, dans le cas des firmes technologiques des bio-industries au Québec (Canada). Introduction Dans le prolongement d’une réflexion sur les très petites entreprises (TPE) (Desmarteau et Saives, 2005), cet article ajoute aux interrogations de Torres (1997) sur la pertinence d’une réflexion renouvelée sur la définition et la spécificité contingente de la PME comme objet d’analyse, une réflexion sur l’instrumentation de cette définition au plan stratégique à l’aide du concept de modèle d’affaires. Partant d’un constat empirique éloquent sur le terrain des firmes valorisant les biotechnologies modernes au Québec (Canada) – la plupart des firmes y ont moins de 10 ans d’existence, 43% de ces firmes sont des TPE (moins de 10 salariés), 52% sont des PME de 10 à 250 employés; et plus de 36% de l’ensemble de ces firmes ne génèrent pas encore de chiffre d’affaires – cet article tente de répondre à la question de recherche suivante : comment caractériser la spécificité des PME biotechnologiques s’il en est une ? Notre réflexion s’articule en trois temps. D’abord, un cadre théorique est proposé pour caractériser les PME à partir d’une revue de littérature sur les critères de définition des PME et sur le concept de modèle d’affaires. Ensuite, nous faisons état de la méthodologie d’analyse statistique multi-factorielle classificatoire adoptée pour distinguer de façon constructiviste les différents comportements de PME technologiques sur le terrain choisi, la grappe bio-industrielle du Québec, avant, enfin, d’exposer quelques observations conclusives. 1. Cadre théorique Notre questionnement renvoie largement à deux questions récemment posées par Torres (1997) et Marchesnay (2003), respectivement : 1) « dans quelle mesure la conception traditionnelle de la PME élaborée à la fin des années 70 est-elle apte à intégrer certains phénomènes nouveaux » comme celui de l’émergence de petites et moyennes entreprises néo-managériales dans les secteurs modernes de haute-technologie ? et 2) comment qualifier la spécificité de la PME en scrutant sa diversité ? 1.1. De la PME traditionnelle à la PME « organisée » Plusieurs travaux récents (Torrès, 1997)1 s’accordent pour reconnaître la nécessité de repenser les critères de définition de la nature propre de la PME depuis les années 80. Julien et Marchesnay proposaient en 1987 une série de 6 critères (petite taille, centralisation de la gestion, spécialisation faible, système d’information informels et simples, stratégie intuitive et 1 Cf. aussi la recommandation de la Commission Européenne concernant la définition des PME (OJ, L 124/36, 20.5.2003) 45 peu formalisée, marché local) pour définir la PME. Plus récemment, pour mieux rendre compte de la diversité des PME – objet de recherche spécifique, Julien (1994) proposait une typologie sur continuum modulant ces critères1. En 2002, dans le cas des PME à forte croissance au Québec, P-A Julien observait également la pertinence d’intégrer à ces caractéristiques : le type de leadership (orientation dynamique donnée par la direction), le type de relation avec la clientèle (marquée par la proximité et l’échange de savoirs), le mode d’organisation de l’innovation et de l’apprentissage (organisation complexe, décentralisée et participative, décisions consensuelles, décisions stratégiques partagées, planification stratégique flexible), et enfin le recours à des réseaux de signaux forts et faibles (complémentarité de ressources), au sein du milieu local, ce que d’autres travaux sur le territoire (Saives, 2002) et les modes d’encastrement des firmes dans leur champ concurrentiel mondial (Torres, 2002) proposaient également d’investiguer. Marchesnay (2003) constate également la nécessité de ne plus ignorer la diversité des petites entreprises avec l’apparition d’entreprises plus « organisées » exploitant les résultats d’une veille stratégique permanente et affichant un plan d’activité mûrement réfléchi « de classe mondiale ». À l’instar de cet auteur (Marchesnay, 2003, p.110), en appelant à un programme de recherche « sur l’identité des petites et très petites entreprises, ce qu’elles sont et ce qu’elles font », nous proposons d’appréhender la réalité complexe des PME de biotechnologie 2 « par ce qu’elles font » à la lumière de leur modèle d’affaires, à l’ère de l’économie-réseau (Magretta, 2002). 1.2. Définir le modèle d’affaires de la PME de haute-technologie Devant la très grande variété de définitions du concept de modèle d’affaires en provenance des gestionnaires et des académiciens, en particulier depuis l’avènement de l’ère réseau (Magretta, 2002 ; Maître et Aladjidi, 1999), nous retenons la définition d’Hamermesh et al. (2002) qui définissent le concept comme une construction dans l’action, constituée d’une série de compromis négociés par la firme sur la base de compétences clés, orientés en fonction d’une stratégie et porteurs d’une finalité de profit. Dans la perspective où un bon modèle d’affaires est d’abord un bon outil de narration (good story) sur la façon dont fonctionne l’entreprise, Magretta (2002, p.4) évalue la performance d’un modèle d’affaires à sa capacité de répondre aux quatre questions canoniques de Drucker : 1) Qui est le client ? 2) Que valorise le client ? 3) Comment fait-on de l’argent dans cette activité ? et 4) Quelle est la logique économique sous-jacente qui explique que l’entreprise sait fournir de la valeur pour les clients à un coût approprié ? En combinant les réponses aux quatre questions précédentes, il s’agit de mettre en évidence les principales parties du véritable système d’affaires qu’est l’entreprise et la manière dont elles s’accordent dans un environnement donné pour former un tout cohérent porteur de valeur (Magretta, 2002). Or, force est de constater que, les modèles d’affaires, en particulier ceux qui portent sur les firmes de biotechnologie dans la littérature, apparaissent souvent relativement peu appareillés à ces questions fondamentales. Fisken & Rutherford (2002) identifient deux modèles d’affaires extrêmes dans les entreprises de biotechnologie essentiellement centrés sur les modalités de génération de revenus3 : d’une part, dans un modèle risqué de long terme, inspiré du modèle pharmaceutique intégré, la création de valeur peut être envisagée en 1 Critères : dimension brute (variable en nombre d’employés, quantité d’actifs, de chiffre d’affaires ou de ventes), secteur d’activité de traditionnel (produits finis) à moderne (ou de pointe, intégrant les produits intermédiaires), marché local et protégé ou bien ouvert et international, forme de contrôle (indépendante à liée) et organisation (centralisée à décentralisée), stratégie (d’intuitive et/ou de survie, à formalisée et/ou de croissance et à risque) et enfin technologie (de traditionnelle, issue d’une innovation faible, spontanée et incrémentale, à une technologie de pointe, issue d’une innovation organisée et radicale). 2 Définition de la biotechnologie (OCDE, 2002) : "The application of S&T to living organisms as well as parts, products and models thereof, to alter living or non-living materials for the production of knowledge, goods and services" (www.oecd.org). 3 . Avec l'apparition du nouveau paradigme scientifique fondé sur la génomique, en pleine crise de productivité de la recherche pharmaceutique (il faut plus de dix ans et plus de 800 M$ US (voire 1.7 Milliards de $) pour la recherche (fondamentale, pré-clinique et clinique de phases I, II, III et IV), le développement et la mise en marché (approbation) d’un nouveau composé thérapeutique (PhRMA, 2003)), a émergé un nouvel acteur dans le système bio-pharmaceutique : les firmes dédiées de biotechnologies modernes devenues partenaires incontournables des firmes pharmaceutiques traditionnelles. Nous qualifions pour partie ces firmes de « néo-technologiques » par contraste avec les firmes valorisant les biotechnologies classiques. 46 capitalisant sur des compétences de développement, plus ou moins internes ou sous-licence, d’un pipeline de produits thérapeutiques (« Product Business Model ») ; ou bien, d’autre part, dans un modèle datant de la fin des années 80, la création de valeur peut être envisagée à court terme par le développement de plate-formes et d’outils hautement technologiques et spécifiques (« Platform or Tool Business Model ») servant le processus de découverte et la chaîne de valeur pharmaceutique (bio-informatique, produits et réactifs de laboratoire, etc.). Du fait néanmoins d’un risque à moyen terme d’obsolescence technique et de banalisation dans ce second cas, Fisken & Rutherford (2002) constatent la convergence de ces deux modèles vers un modèle « hybride », fondé sur l’intégration verticale des deux modèles précédents en une plate-forme technologique capable de générer un pipeline de produits. La portée descriptive, explicative et référentielle d’un modèle unique orienté produit (Baker, 2003) pour des secteurs très différents (santé, nutrition, agriculture, environnement) et pour des stades de développement technologique et d’entreprise variés, devient limitée en proposant une « recette universelle » s’apparentant à un « One Best Way », centrée sur leur résultat plus que sur une description contingente de l’agencement des moyens mis en œuvre pour créer de la valeur à la fois pour l’entrepreneur, l’investisseur et le client. Aussi, comme Hamdouch & Depret (2003) proposant une approche contingente des modèles de gouvernance des firmes de biotechnologie en Europe, nous retenons la nécessité d’opérationaliser plus avant la description des modèles d’affaires des PME à partir d’une analyse empirique contingente des PME valorisant les biotechnologies au sein d’une grappe bio-industrielle. Pour cela, nous ne pouvons ignorer les fruits des réflexions récentes portant sur la gestion de l’innovation dans les secteurs de haute-technologie, en particulier biotech. 1.3. Vers un « Open Business Model » Dans les secteurs de haute-technologie ou encore fondés sur l’innovation technologique (« technology-based » et « science-driven », Saives et al., 2005), et selon l’approche évolutionniste knowledge-based de la firme, une ressource stratégique des entreprises de biotechnologie est le savoir, ou plus précisément la capacité de créer, d’absorber (Cohen & Levinthal, 1990) et de monitorer le savoir au sein de réseaux de collaboration et d’innovation (Powell, 1998). Pour penser davantage la dimension réticulaire en termes stratégiques, Chesbrough (2003) propose le concept « d’Open Innovation », où la dynamique de l’innovation “ouverte” est bidirectionnelle : « Open Innovations means that valuable ideas can come from inside or outside the company and can go to market from inside or outside the company as well. » (Chesbrough, 2003, p. 43). Dans la foulée de Pisano (1994) et Powell (1998), qui ont montré que le lieu de l’innovation réside dans l’échange des savoirs entre les membres d’un réseau dédié, et considérant la dimension réticulaire de la création de valeur (Venkatraman et al., 2002 ; cf. aussi Catherine et al. (2003) dans le cas des firmes de biotechnologies en Europe), nous suggérons qu’instrumenter un modèle d’affaires « ouvert » suppose de caractériser cette logique réticulaire de la création de valeur faisant fructifier l’innovation et les compétences clés qui y sont associées sur une base interne et externe à différents stades d’industrialisation, à partir de connaissances et recherches propres ou développées sous-licences. Aussi, largement inspirés par Chesbrough (2003, p.64), nous caractériserons un modèle d’affaires à partir de la liste des fonctions suivantes et notamment les trois dernières rarement intégrées ailleurs : 1. Articuler spécifiquement une proposition pour le client, fondée sur des bénéfices porteurs de valeur. 2. Identifier un marché cible, c’est-à-dire les consommateurs pour lesquels la technologie, le produit ou encore le service est utile, et ce, en précisant l’objet de l’utilisation. 3. Définir la structure de la chaîne de valeur de la firme, c’est-à-dire les compétences distinctives requises pour créer et distribuer son offre et déterminer les actifs complémentaires nécessaires pour soutenir la position de la firme dans cette chaîne. Baker (2003) insiste sur les quatre compétences clés des firmes de biotechnologie dans le paradigme (bio)scientifique actuel : la capacité de détection l’innovation, la capacité de 47 4. 5. 6. 7. gérer (en interne ou sous-licence) son porte-feuille de produits, la capacité de gérer les alliances et la capacité de gérer les relations avec les investisseurs. Spécifier les mécanismes de génération de revenus, estimer la structure des coûts et prévoir les marges de production de l’offre étant donné la proposition de valeur formulée et la structure de la chaîne de valeur choisie. Les sources de revenus peuvent varier, par exemple selon que l’entreprise intègre, orchestre en partie et/ou licence l’innovation (Andrew et al., 2003). Suivant les enseignements de Parolini (1999), proposer une vision systémique des activités créatrices de valeur, décrire la position de la firme dans un réseau de valeur (« Value Network ») liant fournisseurs, clients, partenaires d’alliances et de collaboration, et incluant l’identification des compétiteurs potentiels. Formuler une stratégie compétitive ou encore une stratégie de marché grâce à laquelle la firme innovante développera des avantages sur ses rivaux. La formulation d’une stratégie n’est pas toujours intégrée au modèle d’affaires (Magretta, 2002). Néanmoins, elle lui fournit, selon nous, sa portée dynamique ou encore son ancrage dans l’action. Enfin, dans la perspective d’entreprises post-modernes ou encore néo-managériales (Marchesnay, 2003), plus autonomes qu’indépendantes sous l’égide d’investisseurs externes variés, il faut aussi organiser la gouvernance, c’est-à-dire l’ensemble des systèmes de surveillance destiné à l’exécution de la stratégie dans une organisation porteuse de valeur économique et de légitimité sociale. Le modèle d’affaires “ouvert” d’une entreprise peut alors être défini comme le design d’un système stratégique qui articule une cabine de pilotage initiée et des compétences distinctives dans une chaîne de valeur en réseau avec des partenaires choisis (universités, firmes, fournisseurs) pour offrir des produits et/ou des services valorisés par les clients dans un marché donné à des coûts appropriés. À partir de cette définition, nous nous pencherons donc sur les questions de recherche suivantes : (Q1) : Quels sont les modèles d’affaires des PME biotechnologiques (que font-elles ?) ? (Q2) : Sont-ils révélateurs d’une spécificité de l’objet-PME (néo)technologique ? 2. Méthodologie Les firmes du système bio-industriel du Québec, représente la principale grappe industrielle d’entreprises valorisant les biotechnologies au Canada avec Toronto et Vancouver (Niosi et al., 2002). Pour analyser les PME de ce système, à partir du cadre théorique synthétisé jusqu’ici, nous utiliserons quelques unes de nos données tirées d’une vaste enquête par entretiens semi-directifs menée auprès de 124 entreprises des bio-industries du Québec (soit plus de 52% de la population estimée dans la province) par une équipe de chercheurs spécialisés sur le management des entreprises de biotechnologie, associés à l’École des sciences de la gestion de l’UQAM (Niosi et al., 2002). Au cours de ces travaux, plus d’une cinquantaine de variables quantitatives et qualitatives ont été reconstruites pour coder les contenus des entretiens semi-directifs menés auprès des dirigeants d’entreprises de toutes tailles à partir d’une batterie de critères propres à caractériser les entreprises de biotechnologie, mise au point par J. Niosi (2003) dans une enquête précédente au Canada en concertation avec StatCan et largement utilisée également en France par les équipes spécialisées sur le sujet (Boissin et al., 2002). Plus spécifiquement, une série de variables qualitatives nominales qualifiant les dimensions d’un modèle d’affaires, dans une perspective d’innovation « ouverte », ont été utilisées pour documenter les comportements de 58 petites et moyennes entreprises (PME) interrogées dans la grappe bio-industrielle du Québec, et mener une analyse multi-factorielle en correspondances multiples (AFCM) et surtout une classification ascendante hiérarchique (CAH) de données dans une perspective typologique exploratoire. Ces variables (listées dans le tableau 1 ci-après) visent à qualifier les 7 dimensions précédentes d’un modèle d’affaires dont : la proposition de valeur (activité technologique, orientation produit et/ou service), le marché ciblé (secteur (nutrition, santé humaine, agriculture ou environnement), et géographie des 48 marchés (exportation)), la compétence distinctive dans la chaîne de valeur (possession d’une plate-forme technologique, effort de R&D, obstacles à la croissance, fidélité des employés et stabilité des équipes), les modèles de génération de revenus (brevets, licences, etc.), le réseau de valeur (origine de l’essaimage, réseau d’alliances et de collaborations, objectifs des alliances et collaborations), permettant l’identification d’une stratégie compétitive, et enfin les éléments de gouvernance reliés à la vision stratégique (perspectives à 5 ans), au milieu d’encastrement (facteurs de localisation et d’ancrage) et aux mécanismes de contrôle inhérents aux moyens de financement (fonds publics de recherche, « seed-money », capital-risque, IPO). Les variables nominales ont été complétées à titre illustratif par une série de 17 variables continues comprenant entre autres des variables démographiques classiques (taille, âge). Nous ne procéderons pas ici à la justification a priori de chacune de ces variables. Nous privilégierons plutôt une approche plus constructiviste de notre objet en décrivant au fur et à mesure les observations découlant de leurs agencements différenciés. 3. Observations résultantes Intéressant de constater tout d’abord que les deux axes factoriels les plus structurants de l’AFCM 1 relèvent essentiellement de deux dimensions : le premier axe distingue essentiellement les entreprises selon leur compétence distinctive; selon qu’elles possèdent où non une plate-forme technologique. Le deuxième recouvre essentiellement des variables relatives au réseau de valeur des firmes. Il oppose essentiellement les firmes selon qu’elles possèdent ou non des réseaux d’alliances et collaborations non seulement universitaires mais aussi inter-firmes. La classification ascendante hiérarchique nous permet plus précisément de distinguer au moins quatre groupes2 de PME, décrits dans le tableau 1 ci-après. - La classe 1 (50% du nombre total des PME interrogées) rassemble des « découvreurs-développeurs », des entreprises « néo-technologiques » ciblant le marché de la santé humaine, et développant à la fois des plates-formes de technologies et de propositions de « produits » pharmaceutiques (biomatériaux, produits de diagnostic, candidats et cibles thérapeutiques potentielles). Ces jeunes entreprises possèdent des compétences distinctives de recherche et développement inscrites dans des plates-formes à forte intensité et complexité technologique, le plus souvent essaimées des institutions publiques de recherche, ainsi que dans des capacités organisationnelles cognitives (équipes de R&D). Dans ce modèle, les firmes ciblent des niches innovantes sur les différents marchés intermédiaires et thérapeutiques de la santé humaine, en exploitant des innovations nées à l’externe (licences reçues) comme à l’interne (importance des brevets) à partir de capacités technologiques complexes (avance technologique de leurs plate-formes reconnue, légitimée dans des publications, et créditée par leurs investisseurs en capital-risque) dont la valeur est entretenue par le biais d’un réseautage actif avec les laboratoires de recherche des universités. Au sein d’un réseau d’alliances de R&D universitaires et inter-firmes, ces firmes construisent leurs capacités distinctives à innover (importance des budgets et des effectifs consacrés à la R&D), notamment génératrices de revenus d’invention (tirés « d’outputs de recherche ») à court et moyen terme sous licence à des tiers (importance du nombre de brevets par firme). - La classe 2 (13.8% des firmes) regroupe une série d’entreprises dont la proposition de valeur pour des clients finaux divers repose sur la fourniture de prestations de services spécifiques (design et fourniture de produits et réactifs de laboratoire, bio-procédés de l’environnement). Ces entreprises disposent d’un savoir-faire technologique reconnu quant à leur capacité de design à la demande, de solutions intégrées, clés en main (séquences de matériel génétique, produits spécifiques de laboratoire, solution intégrées de dépollution par exemple). Ces entreprises ne s’intègrent dans un réseau de valeur qu’à titre marchand, ne pratiquent pas d’alliances et poursuivent plutôt une stratégie de niche sur des créneaux spécialisés et « sur mesure » à court terme, et d’envergure de marchés géographiques à plus long terme. Leur situation au sein du milieu bio-industriel québécois facilite surtout leur accès à 1 2 Pourcentage d’inertie sur l’axe 1 : 12.7%, sur l’axe 2 : 10.7%, total sur cinq axes : 43.6%. Quotient (inertie intra / total) : 30.4% 49 la clientèle (les laboratoires des universités, le milieu industriel local), y-compris à l’exportation nord-américaine. - Les entreprises de la classe 3 (15.5% des firmes) sont positionnées sur la fourniture efficiente de prestations de services génériques (recherche clinique, manufacture à contrat). Ces entreprises, les plus « anciennes » relativement aux autres dans la grappe bio-industrielle (près de 13 ans d’âge en moyenne), ne disposent pas de plate-formes technologiques hautement spécifiques mais d’un savoir-faire reconnu quant à leur capacité de standardisation et à l’efficience des services fournis, à l’aide d’une main d’œuvre flexible et bon marché (grâce aux crédits fiscaux). Dans ce modèle également, les entreprises ne s’intègrent dans un réseau de valeur qu’à titre marchand, ne pratiquent pas d’alliances et poursuivent plutôt une stratégie d’avantage par les coûts sur des marchés spécialisés (recherche (pré)-clinique, manufacture de produits de santé naturels) à court terme, et d’envergure de marchés géographiques à plus long terme. Leur situation au sein du milieu bio-industriel québécois facilite aussi leur accès à la clientèle (laboratoires des universités, entreprises pharmaceutiques environnantes), y-compris à l’exportation nord-américaine. 50 Tableau 1 : Typologie des PME en 4 classes1 GOUVERNANCE RÉSEAU /SYSTÈME DE VALEUR MODÈLE DE REVENU CHAÎNE DE VALEUR DE LA FIRME PROPOSITION MARCHÉ de VALEUR CIBLÉ (1) Développeurs néo-technologiques Dimensions du modèle d’affaires 50% des firmes Santé humaine*** Marché (Santé végétale) Exportation Activité technologique Orientation Plate-forme technologique % dépenses RD / Ventes Compétence distinctives (exemples) Comp. Distinctive : technol. Comp. Distinctive : équipe Obstacle à la croissance Age Nb total employés % d'employés RD / total Nb gestionnaires Turn-over total (%) Chiff.d'aff. / employé(M$) Brevets Nb brevets Nb licences accordées Nb licences reçues Publications Essaimage universitaire (ou de laboratoire public) Alliances Alliances avec des universités Alliances avec des firmes Alliances de recherche Alliances commercialisation Allia. Partenaires étrangers Allia. Partenaires canadiens Produit (72.4%) O*** 46.7% O (62%) O (66%) Service*** N*** 16.2% Moy. Échan totale -tillon Nutrition humaine*** N (50%) Extraction biologique d’ingrédients fonctionnels Formulation N*** 17.4% O(100%) N(50%) N (44%) O (67%) N N RH (67%) Coût / tps approbation réglementaire *** RH (75%) 8.9 34.5 35.2% 2.9 4.5% 0.197 O (100%) 2.9 0 6.4 O (75%) 12.7 75.5 24% 3.8 12.6% (19% RD) 0.143 N (67%) 1.11 0.44 0.22 N (67%) 10.8 35.75 23.6% 2.9 6.2% 0.299 N (67%) 2 0.08 0 N*** O *** N (100%) N (89%) N (100%) O*** N N*** O (92%) O (66%) N (75%) N (67%) O (75%) O*** O *** N O*** O (93%) N (63%) N*** N*** N (56%) N (100%) N (89%) O (92%) O (92%) O *** N (75%) O (100%) Pdt en marché (45%) Croissance interne (31%) Fusion (31%) à long terme Service (50%) O (100%) 21.6% CRO*** CMO (4) Fabricants spécialistes 20.7% des firmes 7 36.3 54.2% 3.5 2% 0.266 O (72.4%) 5.43 0.44 1.31 O (90%)*** Vision A court terme Bioprocédé*** Produits et réactifs Capital (48%) Technologie (20%) Accélérat. innovation*** Financement *** Complémentarité (72.4%) Obtention de Capital-Risque Capital Risque (M$) Compagnie publique (3) Fournisseurs génériques 15.5% des firmes Santé humaine (67%) Environnement*** Nutrition hum. (33%) O (88%) O(78%) 13% 32.6% Fermentation bactérienne Fermentation Protocoles standardisés Inhibition d’enzymes bactérienne (probiotiques) de recherche clinique Hydrolyse enzymatique et Sélection génétique Extraction, purification, (analyses biochimiques, nanoséparation Traitements biologiques stabilisation de biostatistique, monitoring) Amplification de la (air, sol, polluants) bio-ingrédients Manufacture normalisée phytosynthèse, criblage PCR, Clonage, Formulation et efficiente à façon Contrôle de réplication séquençage à façon suppléments Résultats obtenus des alliances et partenariats Facteur de localisation STRATEGIE DE MARCHE Rech. de cibles in silico et in vitro*** Biomatériaux / DDS Pdt diagnostic (2) Designers spécifiques 13.8% des firmes N (88%) N (50%) 10.31 8.9 82.97 42 20.4% 38.2% 3.89 3.4 3.9 4.9% 0.535 0.228 3.4 0.3 1.51 3.67 0.3 1.58 Capacités de commercialisation *** Réponse consos *** Mise au pt nvx pdts** Pdt en marché (50%) Alliances (67%)*** Croissance interne (67%) Croissance interne (67%) O (72.4%) O (88%) N (56%) N*** 8.62 7.75 3.98 n/a 4.79 O (28%) N (100%) O (22%) N (100%) Infrastructures Avant. fiscaux (79.3%) Laboratoires publics Avantages fiscaux (25%) (56%) services (50%) Inf. universi. (48%) Niche ou focalisation Niche ou focalisation Niche ou focalisation Niche ou focalisation Envergure de produits, licences, marchés Envergure de Avantage par les thérapeutiques, marchés coûts Envergure de produits possiblement création géographiques (et Envergure de et domination de sectoriels) marchés marché 1 7.57 - Modalités : O (oui), N (non). Les variables nominales actives significative(s) : *** au seuil de 1%, ** seuil de 5%, * de 10%. - % entre parenthèses : pourcentage de population correspondant aux modalités citées dans la classe 51 Tableau 2 : Typologie des modèles d’affaires des PME technologiques Dimensions du modèle d’affaires PME « technologique » PME néo-managériale “néo-biotechnologique” PROPOSITION DE VALEUR Bénéfices génériques Bénéfices spécifiques MARCHÉ CIBLÉ Sectoriel et géographique Multi-sectoriel et glocal COMPÉTENCES DISTINCTIVES Organisation scientifique du travail Organisation professionnelle du travail MECANISMES DE GENERATION DE REVENUS Efficience des opérations Valorisation des actifs intellectuels (PI) VALUE NETWORK Réseau de contacts commerciaux Réseau ‘expertal’ de connaissances STRATÉGIE +/- Rationnelle Adaptative Part de marché Création de valeur Monocéphale Bicéphale Objectif stratégique GOUVERNANCE En effet, les PME de biotechnologie peuvent articuler des propositions de valeur alliant des bénéfices génériques pour le client (cas de la classe 3) ou de plus en plus spécifiques et spécialisés (classe 4, 2 puis 1). Le marché ciblé est souvent sectoriel et géographique dans le cas des PME technologiques classiques, il peut prendre une envergure multi-sectorielle et glocale, de « classe mondiale » (Torres, 2002) dans le cas des PME néo-technologiques valorisant des plates-formes de technologies et de produits. Il est aussi possible de distinguer des firmes organisées selon les principes classiques d’inspiration taylorienne de division du travail et gestion scientifique des intrants, des tâches, et/ou des fonctions versus des firmes néo-managériales ou prime l’organisation professionnelle du travail (Drucker, 1980) (la bureaucratie de professionnels au sens de Mintzberg) et une division cogni-productive des savoirs entre « knowledge workers » à l’ère de l’économie du savoir. Dans le premier cas, les mécanismes de génération des revenus s’appuient sur la valorisation efficiente d’actifs, le plus souvent tangibles alors que l’art des secondes (PME néo-technologiques) relèvent de la gestion d’un portefeuille d’actifs intellectuels (brevets, licences sur les outputs de recherche à tous les stades de développement, avance technologique, secret industriel). Plus déterminant encore s’avère le réseau de valeur dans lequel s’insère la PME. Les PME technologiques s’appuient souvent sur des réseaux de contacts essentiellement marchands quand les PME néo-technologiques valorisent un véritable capital social (Desmarteau, 2000), à savoir un réseau de connaissances et d’expertises tout autant académiques qu’industrielles, etc. La stratégie de ces PME souvent rationnelle est plus encore adaptative dans le cas des PME néo-technologiques où l’environnement est particulièrement instable et où la sérendipité des découvertes requiert la flexibilité stratégique. Enfin, la dimension de la gouvernance peut-être très distinctive entre PME technologiques et néo-technologiques où elle est monocéphale dans les premières, entre les mains d’un dirigeant autonome, et bicéphale dans les secondes, au sens de Drucker1 puisque incarnée à la fois par un conseil d’administration et un conseil scientifique, chargés de limiter les risques techno-scientifiques et financiers associés à la stratégie de savoir de l’entreprise. 1 « Population dynamics are rapidly transforming traditional organization into a « double-headed monster » in which autonomous managerial and autonomous professional organizations live together in symbiotic tension. » (Drucker, 1980, p. 78). 52 Conclusion Dans l’esprit de Catherine et al. (2003b) qui distinguent 4 modèles d’entreprises dédiées de biotechnologies, essentiellement discriminés par l’axe technologique de l’intensité de l’innovation et l’orientation produit ou service adoptée par les entreprises1, nous avançons 4 modèles d’affaires distincts au sein d’une grappe bio-industrielle, discriminés selon les fonctions contingente d’un business model que sont la proposition de valeur pour le client et le marché ciblé (santé, nutrition, agriculture, environnement), les mécanismes de génération de revenus (d’innovation) et la stratégie adoptés à court, moyen et long terme mais surtout les compétences distinctives de la firme (valeur interne) et la valeur de réseau. Là où la littérature fournit souvent en guise de business model, un modèle « économétrique » de développement orienté vers la « croissance » des firmes dans des cycles vertueux de financement (rondes successives de capital-risque puis IPO pour financer la recherche interne « fermée »), nous décrivons différents modèles d’organisation « stratégiques » orientés vers la « performance », intégrant la valeur stratégique du réseau de valeur pour capter l’innovation sur une base interne ou externe dans une logique coopérative, et selon la rationalité nouvelle de l’open innovation, avancée par Chesbrough (2003). Ces PME sont-elles contre-nature dans un contexte aussi instable et complexe que celui de la biotechnologie ? Nous répondons non en avançant par contre leur spécificité par la compréhension des « pourquoi ? » qui légitiment la manifestation des « comment ?» de ces formes d’organisation. Pour valider nos conclusions et pallier aux limites idiosyncratiques de ce travail monographique, une étude à la fois longitudinale et comparative des PME de différents biopoles européens (Edinburg, Nantes, Milan, etc.) est en cours. 1 Catherine et al. (2003) distinguent les entreprises de produits (innovation moindre, orientation produit), les firmes de « biosanté » (innovation importante, orientation produit), les prestataires de service (innovation moindre, service à façon) et les entreprises plate-forme technologique (service à haute valeur ajoutée, innovation importante). 53 Références Andrew, James P. & Sirkin, Harold L., “Innovating for Cash”, HBR (September 2003), 9p. Baker, Ann, “Biotechnology’s growth-innovation paradox and the new model for success”, Journal of Commercial Biotechnology, 9(4), (June 2003), 286-288. Boissin, J-P., et al., « Structure de gouvernance dans les PME high-tech ; caractéristiques des dirigeants, structure de contrôle et croissance de l’entreprise », rapport au Commissariat Général au Plan, convention 1999/9, (Décembre 2002). Catherine, David, Corolleur, Frédéric & Coronini, Roger (coll.), « Les fondateurs des nouvelles entreprises de biotechnologies et leurs modèles d'entreprise », RIPME, 15(2), (2003), 63-92. Chesbrough, Henry, Open Innovation, Boston : Harvard Business School Publishing, 2003. Cohen, Wesley M., & Levinthal, Daniel A., “Absorptive capacity: A new perspective on learning and innovation”. Administrative Science Quarterly, 35, (1990), 128-152. Desmarteau Robert H., «Le point: La valeur stratégique du capital social», ASAC-IFSAM, 21 (6), (2000), 47-60. Desmarteau, Robert H., & Saives, Anne-Laure, « Les TPE de biotechnologie sont-elles contre-nature ? », RIPME, à paraître, (2005). Drucker, P., Managing in turbulent times, New-York: Harper & Row, 1980. Fisken, Jane & Rutherford, Jan, “Business models and investments trends in the biotechnology industry in Europe”, Journal of Commercial Biotechnology, 8 (3), (Winter 2002), 191-199. Hamdouch, Abdelillah, & Depret, Marc-H., “La gouvernance des jeunes entreprises innovantes : un éclairage analytique à partir du cas des sociétés de biotechnologies », à paraître F.C.S (Juin 2003). Hamermesh, Richard G., Marshall Paul W., & Pirmohamed, Taz, “Note on Business Model Analysis for the Entrepreneur, Harvard Business School, January 22nd, 2002. Julien P-A. (dir)., Les PME : bilan et perspectives. Paris : Economica. 1994. Julien P-A. (dir)., Les PME à forte croissance. L’exemple de 17 gazelles dans 8 régions du Québec. Collection PME & Entrepreneuriat, Presses de l’Université du Québec. 2002. Magretta, Joan, “Why Business Models Matter”, HBR, (May 2002). Marchesnay M., « La petite entreprise : sortir de l’ignorance », RFG, 29(44), (2003), 107-118. Niosi, J., "Alliances are not enough. Explaining rapid growth in biotechnology firms", Research Policy, 32 (5), (2003), 737-750. Niosi, Jorge E., Cloutier, Martin L., Lejeune, Albert. (Coord.), Biotechnologie et Industrie au Québec, Montréal: Éditions Transcontinental, 2002. Parolini, Cynthia, The Value Net, Chichester, England : John Wiley & Sons, 1999. Powell, Walter W., «Learning From Collaboration: knowledge and Networks in the Biotechnology and Pharmaceutical Industries», California Management Review, 40 (3), (1998), 228-240. Saives, Anne-Laure, Territoire et compétitivité de l’entreprise, L’Harmattan, 2002. Saives, Anne-Laure, Desmarteau Robert H. & Seni, Dan, « Vers une conception des bio-industries? », à paraître, Économies et Sociétés, « Dyn. agroalimentaires », 27, 2005. Torres, O., « Pour une approche contingente de la spécificité de la PME », RIPME, 10(2), (1997). Torres, O., « Small firm, glocalization strategy and proximity », ECSB – Research in Entrepreneurship and Small Business – 16th Conference (Barcelona, 21-22 november 2002). Venkatraman, N., & Subramaniam, M., « Theorizing the Future of Strategy: Questions for Shaping Strategy Research in the Knowledge Economy”, in Pettigrew A., Thomas H., Whittington R. (ed), Handbook of Strategy and Management, Sage, London, (2002). 461-473. 54 ASAC 2005 Toronto, Ontario Carole DONADA ESSEC Business School of Paris France Isabelle DOSTALER John Molson School of Business Concordia University LA RESISTANCE D’UN FOURNISSEUR AUX EXIGENCES DE FLEXIBILITE : L’IMPACT SUR SON SLACK ORGANISATIONNEL Cette recherche examine quatre antécédents de la résistance d’un fournisseur aux exigences de flexibilité réclamées par son client. Elle explore également les effets de cette résistance sur le niveau de slack du fournisseur. L’étude empirique porte sur 178 entreprises françaises fournisseurs des industries automobile et aéronautique. Introduction La flexibilité est définie comme la capacité à s’adapter facilement aux circonstances. Elle joue un rôle d'amortisseur des contraintes de l'environnement (Miner et al., 1990). Elle permet de répondre aux menaces et opportunités qui se posent aux entreprises (Grewal et Tansuhaj, 2001) et constitue pour cela un facteur clef de succès (Volberda, 1999 ; Sanchez, 1995). La flexibilité doit donc être recherchée, développée et protégée (De Meyer et al., 1989 ; Gerwin, 1993). Les études sur l’importance de la flexibilité pour les entreprises clientes qui externalisent leurs activités sont nombreuses. Leurs résultats montrent que la flexibilité augmente la qualité des échanges (Brown et al., 2000), leur continuité (Cannon et Homburg, 2001) et la satisfaction des clients (Morgan et Hunt, 1994). Les conséquences de la flexibilité pour les entreprises fournisseur sont en revanche peu étudiées et les rares études qui traitent de ce sujet suggèrent que la flexibilité contraint et diminue l’efficience économique du fournisseur ; qu’elle représente une obligation à laquelle il faut savoir résister (Moritsen,1999). Ainsi, dans la lignée des recherches de Kotabe et al (2003) ou de Subramani et Venkatraman (2003)) qui s’intéressent aux entreprises fournisseurs ne possédant a priori aucun pouvoir de négociation, notre recherche explore les sources de la résistance des fournisseurs vis-à-vis de leurs clients. Nous apportons des éléments de réponse aux deux questions : comment des fournisseurs dépendants de leurs clients parviennent-ils à leur résister et à dire non aux demandes de flexibilité qui les contraignent ? Quelles sont les conséquences de cette résistance aux exigences des clients sur le slack des fournisseurs ? L’article est structuré en deux parties. La première présente le cadre conceptuel de la recherche. Nous nous appuyons sur les théories du pouvoir et de la dépendance et sur les approches relationnelles pour identifier quatre déterminants de la résistance ainsi que sur les travaux sur le slack organisationnel. La deuxième partie présente l’étude empirique. 55 Les antécédents conceptuels de la résistance à la flexibilité Résistance et dépendance Résister, c’est exercer un pouvoir. Selon Pfeffer et Salancik (1978), une organisation exerce son pouvoir pour agir sur son environnement en vue d’acquérir et maintenir dans son portefeuille les ressources dont elle dépend. Dans cette approche, pouvoir et dépendance sont indubitablement liés car la dépendance d’un acteur sur un autre détermine les conditions d’exercice de son pouvoir (Emerson, 1962). Crozier et Friedberg (1977) ont illustré cette relation et montré comment des acteurs, a priori exclus des sphères du pouvoir, parviennent à développer des stratégies d’influence afin de limiter l’asymétrie de leur dépendance et sortir des contraintes qui les limitent. De son coté, Porter (1980) a montré que le lien entre le pouvoir de négociation des acteurs d’une filière et leur dépendance vis-à-vis de leurs ressources découlent directement de la structure de l’industrie dans laquelle ils se trouvent. Les propositions de ces approches par le pouvoir ont souvent été validées dans les études empiriques qui prennent le point de vue d’un client dominant vis-à-vis duquel des partenaires externes n’exercent pas leur pouvoir pour résister (Lusch et Brown, 1996 ; Provan et Gassenheimer, 1994 ; Kumar, Scheer et Steenkamp, 1995). Dans cette recherche, nous nous intéressons à la situation inverse et suggérons un lien négatif entre la dépendance d’un fournisseur et sa capacité de résistance vis-à-vis de son client : H1a : La dépendance d’un fournisseur vis-à-vis d’un client est négativement associée à sa résistance aux exigences de flexibilité du client Résistance et coûts de changement Les coûts de changement d’un fournisseur représentent l’ensemble des dépenses nécessaires à un client pour changer de fournisseur. Selon Williamson (1985), les coûts de changement résultent essentiellement du petit nombre de fournisseurs sur le marché ainsi que de la spécificité des actifs indispensables aux échanges et détenus par ces fournisseurs. Lorsqu’un fournisseur dispose d’actifs spécifiques et qu’il est difficile pour le client de trouver des sources alternatives d’approvisionnement, le fournisseur peut prendre le client en otage et exercer son pouvoir sur lui (Klein, Crawford et Alchian, 1978). Porter (1980) appréhende également cette situation en suggérant que le pouvoir de négociation des fournisseurs dépend positivement de leur taux de concentration, de l’absence de produits de remplacement ou des coûts induits par leur changement. D’où l’hypothèse : H1b : Les coûts de changement d’un fournisseur (supportés par un client) sont positivement associés à la résistance du fournisseur face aux exigences de flexibilité du client Résistance et autorité French et Raven (1959) suggèrent que l’expertise et la reconnaissance constituent les deux sources du pouvoir. En effet, une entreprise disposant d’une expertise rare, non-substitutable et difficile à imiter, possède une ressource stratégique (Barney, 1991) qui légitime sa capacité d’influence. De plus, si cette entreprise est reconnue et si elle fait référence dans son domaine, elle détient une autre base de pouvoir vis-à-vis de ses partenaires externes. S’appuyant sur ces propositions Frazier (1984) rassemble ces deux sources de pouvoir sous la notion d’autorité. L’auteur définit l’autorité comme l’ensemble des ressources qui légitiment une organisation et lui donnent le droit d’en influencer d’autres. En adaptant la proposition de Frazier au contexte des 56 relations client-fournisseur et, plus spécifiquement, au problème de la capacité de résistance des fournisseurs dépendants, nous suggérons que : H1c : L’autorité d’un fournisseur vis-à-vis d’un client est positivement associée à sa résistance face aux exigences de flexibilité du client Résistance et normes relationnelles Selon Crozier et Friedberg (1977), l’exercice d’une résistance dépend de la puissance de ses sources mais aussi du contexte dans lequel elle s’exprime. Dans le cadre d’une relation client-fournisseur, la qualité du contexte relationnel est définit par les normes partagées entre les partenaires (Heide et John, 1992). Ces normes sont des valeurs relationnelles communes qui structurent les échanges, caractérisent leur nature et influencent les conditions de coopération (Macneil, 1980). Elles reposent sur la solidarité, la flexibilité et la communication des partenaires ainsi que sur leur attente de continuité de la relation. Sans les normes relationnelles, les relations client-fournisseur ne sont rien d’autre que des échanges distants exempts de réciprocité (Jap et Anderson, 2003; Noordewier et al., 1990). La flexibilité étant une norme relationnelle, un fournisseur respectueux des normes partagées avec son client doit logiquement accepter les exigences de flexibilité de ce dernier. D’où l’hypothèse : H2a : L’importance des normes relationnelles dans la relation diminue la résistance du fournisseur face aux exigences du client Par ailleurs, la proposition de Crozier et Friedberg (1977) nous conduit à formuler une hypothèse supplémentaire liant l’effet des normes relationnelles à toutes les sources de pouvoir du fournisseur. De nombreuses études empiriques ont validé que plus une relation est basée sur le respect des normes relationnelles, plus les partenaires coopèrent et moins ils exercent leur pouvoir (Aulakh et Kotabe, 1996 ; Hitt et al., 1998 ; Subramani et Venkatraman, 2003 ; Wathne et Heide, 2004 ; Zajac et Olsen, 1993). Dans la lignée de ces travaux, nous suggérons l’hypothèse modératrice : H2b: Les liens entre la dépendance du fournisseur (H1a), les coûts de changement supportés par le client (H1b), l’autorité du fournisseur (H1c) et la résistance du fournisseur face aux exigences du client sont modérés lorsque les normes relationnelles de la relation sont importantes. Les hypothèses précédentes permettent d’identifier des déterminants de la résistance d’un fournisseur. Nous explorons maintenant les effets de cette résistance sur le niveau de slack du fournisseur. Lien entre la flexibilité et le slack du fournisseur Ne connaissant pas à l’avance les exigences de flexibilité de leurs clients, les fournisseurs gardent en réserve des ressources de flexibilité appelées aussi ressources de slack (Gerwin, 1983). Selon Cyert et March (1963:36), le slack correspond aux ressources qui sont stockées dans l'organisation, mais qui ne sont pas strictement nécessaires pour son fonctionnement normal. Autrement dit, le slack résulte de la différence entre les ressources dont dispose l’entreprise et les moyens nécessaires pour maintenir la coalition. Les chercheurs distinguent généralement deux types de slack : le slack engagé (ou dédié) qui correspond aux ressources supplémentaires ne pouvant pas facilement être réutilisées, et le slack non-engagé lié aux ressources facilement réutilisables (Daniel et al., 2004 ; Sharfman et al., 1988 ; Singh, 1986 ; Tan et Peng, 2003). Ces deux types de slack illustrent la distinction entre : (1) des ressources qui jouent le rôle de tampons externes et qui aident à répondre aux demandes changeantes des partenaires d’échange et (2), des ressources supplémentaires qui ne dépendent pas directement des partenaires externes et qui sont à la disposition l’organisation qui les gère. Le slack d’un fournisseur se compose donc à la fois de 57 slack engagé (qui dépend de ses contraintes de flexibilité vis-à-vis de ses clients) qui menace sa performance, et de slack non-engagé qui découle de ses choix de gestion et l’autorise à se créer de nouvelles opportunités. Dès lors, un fournisseur qui peut résister aux demandes de flexibilité d’un client a besoin de moins de slack engagé qu’un fournisseur incapable de résister. Il peut notamment réduire ses stocks de produits dédiés et ajuster au mieux son outil de production à ses objectifs de productivité. En outre, ce fournisseur peut disposer d’un slack non-engagé (par exemple un surplus de liquidités) qu’il peut utiliser à des fins de développement d’innovations ou à la mise en œuvre de nouvelles orientations stratégiques. D’où les deux hypothèses : H3a. La résistance du fournisseur aux exigences de flexibilité du client a un lien négatif avec son niveau de slack engagé. H3b. La résistance du fournisseur aux exigences de flexibilité du client a un lien positif avec son niveau de slack non-engagé. Etude empirique Méthodologie Pour opérationnaliser les variables de notre modèle, nous avons repris des échelles déjà testées par des chercheurs et nous les avons adaptées au contexte de l’étude. Le tableau 1 présente les variables retenues et indique les références théoriques qui ont servi à la construction de chaque échelle de mesure. Trois variables de contrôle ont été prises en compte dans l’étude : le secteur (automobile et aéronautique), la taille des entreprises (mesurée par le log du nombre d’employés) et le degré de personnalisation des produits commandés par le client (variable de « customization » mesurée sur une échelle de likert en 5 points selon Heide et John, 1992). Les répondants devaient considérer les relations qu’ils avaient avec leur client principal (le client X). Le questionnaire a été administré en France lors de salons professionnels auprès des représentants d’entreprises fournisseurs de l’industrie automobile et aéronautique. 85 entreprises automobile et 93 entreprises aéronautique constituent notre base de données de 178 entreprises (soit 33,6% de la population totale des entreprises fournisseurs de ces secteurs1). Tableau 1 – Variables, items et références2 Items Dépendance (DEP) Le fournisseur ne peut pas facilement remplacer le client X Le pourcentage des ventes vers le client X Le client X contribue fortement au chiffre d’affaires du fournisseur Le fournisseur est dépendant du client X Autorité (AU) Le fournisseur a l’expertise qui lui permet d’exiger Le fournisseur est perçu comme étant suffisamment expert pour que d’autres entreprises veuillent s’en rapprocher Les capacités spécifiques du fournisseur sont reconnues à l’extérieur Coûts de changement (SWC) Les coûts de changement du fournisseur pour le client X 1 Références Cool et Henderson (1998), Frazier et Rody (1991) Frazier (1984), Kholi (1989) Cool et Henderson (1998), Sur les 22053 entreprises de plus de 20 personnes opérant en France (base EAE du ministère de l’industrie), 418 entreprises composent le secteur des fournisseurs automobile (code NAF 343Z, 316A et 361A) et 112 entreprises composent le secteur des fournisseurs aéronautiques (code NAF 353). Nos échantillons représentent respectivement 20, 34% et 83% des entreprises de ces secteurs. 2 Les questions ont été formulées en anglais. 58 Le remplacement du fournisseur pour ce produit particulier entraînera des dépenses et des délais importants pour le client X Il est facile pour le client X de remplacer le fournisseur Normes relationnelles (NOR) La relation avec le client X est basée sur l’entente mutuelle et la confiance Il est convenu que le fournisseur et le client X communiquent toutes les informations concernant le produit Il est convenu que les problèmes survenant sont traités conjointement plutôt qu’individuellement Résistance (RES) Le fournisseur ne s’ajuste pas en cas en changements provoqués par le client X Le fournisseur n’accepte pas les changements une fois l’accord signé Le fournisseur n’accepte pas d’être flexible sur les exigences du client X Slack engagé (SE) Le fournisseur opère en dessous de ces capacités productives Le client X mobilise des ressources qui pourraient être redéployées autrement par le fournisseur Stocks/ventes Créances clients/ventes Slack non-engagé (SNE) Working capital/current net asset Curent ratio Le fournisseur dispose de ressources supplémentaires non encore dédiées Wathne et Heide (2004), Frazier et Rody (1991) Gundlach, Achrol et Mentzer (1995), Berthon et al (2003) Canon et Homburg (2001), Berthon et al (2003), Lusch et Brown (1996), Noordewier et al (1990) Tan et Peng (2003), Nohria and Gulati (1996), Daniel (2004) Daniel (2004) Analyse et discussion des résultats Une analyse factorielle (rotation varimax) des variables liées à la résistance fait apparaître 6 facteurs orthogonaux correspondant aux quatre antécédents de la résistance des fournisseurs et aux deux types de slack (tableau 2). La validité des échelles de mesure de chaque variable est mesurée par l’alpha de cronbach ainsi que par les contributions des items sur les axes factoriels. La validité de l’échelle mesurant la résistance d’un fournisseur aux exigences de flexibilité de son client est assurée. L’échelle proposée est unidimensionnelle (les coefficients des items sur l’axe factoriel sont de .861, .854, .812) et l’alpha de cronbach est de 79,6. 59 Tableau 2 – Analyse des variables (analyse factorielle avec rotation varimax) Facteurs SNE SE DEP SWC NOR AU SNE1 .945 .120 -.019 .003 .020 -.048 SNE2 .895 .010 -.011 .036 .111 -.028 SNE3 .891 .094 -.010 -.047 .023 -.049 SNE4 .868 .189 .074 -.049 .037 -.111 SE1 .138 .886 .075 -.089 .085 -.050 SE2 .038 .869 .067 -.108 .171 -.019 SE3 .135 .839 .222 -.144 .190 -.132 SE4 .146 .806 .160 -.191 .167 -.031 DEP1 -.026 .187 .854 -.085 -.003 -.007 DEP2 .005 -.050 .830 -.102 .091 .019 DEP3 .024 .171 .826 -.043 -.050 .140 DEP4 .024 .143 .821 -.168 -.064 -.007 SWC1 -.009 -.195 -.158 .847 -.079 -.057 SWC2 .021 -.179 -.118 .825 .051 -.036 SWC3 -.054 -.053 -.092 .815 -.045 -.137 NOR1 .025 .026 .021 -.037 .846 -.036 NOR2 .134 .224 .006 -.075 .846 -.059 NOR3 .020 .334 -.058 .049 .786 .015 AU1 -.095 -.057 .062 -.083 -.058 .889 AU2 -.140 -.026 .053 -.077 -.053 .800 AU3 .030 -.078 .008 -.055 .027 .795 Valeur propre 3.356 3.285 2.923 2.219 2.198 2.151 Alpha de Cronbach 93.1 91.8 86.9 81.6 81.4 78.9 Les scores factoriels des variables ont ensuite été utilisés dans les différents modèles de régression linéaire (Tableaux 3 et 4). Variable RES (Constante) DEP SWC NOR AU NOR*DEP NOR*SWC NOR*AU taille custom secteur R² Tableau 3 – Régressions sur la variable résistance Modèle 1 Modèle 2 Bêta t sig Bêta t sig .000 1.0 .000 1.00 -.199 -2.87 .00 -.216 -3.11 .002 .188 2.713 .00 .189 2.718 .007 -.257 -3.72 .00 -.258 -3.73 .000 .175 2.525 .01 .176 2.528 .012 .097 1.365 .174 .121 1.684 .094 .011 .151 .880 17.2 19.9 Modèle 3 t -2.77 -.192 -3.12 .208 3.286 -.137 -2.16 .115 1.845 .128 2.038 .110 1.716 .024 .390 -.143 -2.23 .015 .232 .458 6.962 39 Bêta sig .006 .002 .001 .032 .067 .043 .088 .697 .026 .817 .000 Le modèle 1 teste les quatre hypothèses H1a, H1b, H1c et H2a qui associent les variables explicatives -dépendance, coût de changement, autorité, normes relationnelles- à la résistance des fournisseurs. Les résultats de la régression valident ces hypothèses. Ils montrent que les fournisseurs (même des fournisseurs aussi dépendants des clients que les fournisseurs automobile et aéronautique) peuvent accroître leur capacité de résistance en s’affranchissant de leurs clients. Ils doivent pour cela gérer au mieux l’asymétrie de leur dépendance, asseoir leur autorité dans leur domaine et augmenter les coûts pour les clients qui voudront changer de fournisseurs. Ces résultats éclairent aussi la prédominance de la variable « norme relationnelle » confirmant ainsi les 60 propositions théoriques de l’approche de l’échange social (Macneil, 1980) et les conclusions des travaux empiriques d’Andaleeb (1995), de Boyle et al. (1992) ou de Kim (2000). Notre étude apporte toutefois une contribution supplémentaire à ces travaux en examinant des entreprises fournisseurs et non plus des clients tout puissants. Le deuxième modèle teste l’hypothèse H2b qui suggère que le poids des normes relationnelles structurant le contexte des échanges modère les effets directs des antécédents de la résistance des fournisseurs. La validation de cet effet doit confirmer la méfiance de certains fournisseurs vis-à-vis des relations partenariales et de tout ce qu’elles impliquent en termes de suivi des normes relationnelles. Les résultats de la régression ne valident pas l’hypothèse L’entente de deux partenaires sur le respect des normes relationnelle impacte donc la résistance du fournisseur aux exigences de flexibilité du client mais n’affecte pas ses sources de pouvoir. Le troisième modèle introduit les variables de contrôle illustrant le secteur, la taille des entreprises et le degré de personnalisation des produits commandés par le client. La régression présente trois résultats significatifs. Nous observons tout d’abord une relation négative et significative entre la taille et la résistance, suggérant que les grandes entreprises résistent moins que les plus petites. Deuxièmement, l’introduction de la variable « taille » a activé la significativité de la variable modératrice (NOR*DEP) mesurant l’effet des normes relationnelles sur la relation dépendance-résistance. Enfin, nous observons que la variable « secteur » a l’influence la plus forte dans la régression. Ce résultat nous encourage donc à tester les hypothèses H3a et H3b en distinguant les entreprises automobile des entreprises aéronautique. Les hypothèses sur les liens entre la résistance et le slack du fournisseur ont été testées dans deux régressions supplémentaires isolant les deux types de slack. Les résultats montrent que la résistance permet au fournisseur de diminuer significativement ses ressources de slack engagé. La validation de H3a, apporte ainsi un argument statistique aux résultats qualitatifs de Moritsen (1999). Si nous prolongeons ce résultat par l’argument de Jensen (1986) qui voit dans la détention de slack la diminution de la performance de l’entreprise, nous pouvons suggérer que les fournisseurs qui résistent le mieux aux exigences de flexibilité de leurs clients devraient être aussi les plus performants. Contrairement à nos hypothèses, nous n’observons pas de lien positif entre la résistance des fournisseurs et leur niveau de slack non-engagé et, pour les entreprises du secteur automobile, la régression n’est pas significative. Enfin, le signe négatif du coefficient illustrant la taille laisse penser que les entreprises de petite taille ne peuvent se permettre d’avoir du slack non engagé. Conclusion Depuis une quinzaine d’années le discours dominant des entreprises clientes porte sur le respect des normes relationnelles comme première condition de succès de leurs relations avec leurs fournisseurs. Parallèlement, les plaintes des fournisseurs contraints à la flexibilité, à la transparence des informations et à la solidarité avec leurs clients se font de plus en plus entendre. Dans l’industrie automobile, certains analystes prédisent un prochain renversement des rapports de pouvoir à l’image de celui qui s’est opéré dans l’industrie informatique sous l’impulsion des fournisseurs de microprocesseurs. Force est de constater que ces fournisseurs sont encore structurellement dominés par leurs clients. Leur seule marge de manœuvre est de résister aux clients en refusant de se soumettre à leurs exigences. Pour ces fournisseurs, il ne s’agit pas de résister par plaisir mais de dire non aux exigences qui pénalisent leur performance et leur survie. Depuis vingt ans, beaucoup de sous-traitants automobile ont déposé leur bilan parce que leur niveau de slack engagé pour des clients plombait leur rentabilité. De même beaucoup ont engagé des ressources trop importantes pour eux dans le seul but de “suivre leurs clients” et répondre à leurs exigences de délocalisation ou d’internationalisation. Ces fournisseurs ont investi des sommes considérables dans des implantations de nouvelles unités de production et ont supporté 61 tous les risques économiques financiers consécutifs. Ils n’avaient pas le choix et leur décision d’investissement résultait davantage d’une obligation à être flexible que d’un calcul positif de risque-performance. Dans l’industrie automobile, les effets ne sont pas aussi marqués mais la logique qui les accompagne reste équivalente. Dans ce cadre, notre recherche peut apporter des arguments supplémentaires à ceux qui s’interrogent sur les conséquences de la flexibilité. D’un point de vue théorique, nous pensons également contribuer aux réflexions sur le slack. Le concept de slack n’a jamais cessé d’interroger les chercheurs comme en témoignent les derniers travaux sur le sujet (Daniel, 2004, Tan et Peng, 2003). Notre recherche éclaire de nouvelles variables explicatives du slack en établissant un lien entre l’approche de l’échange social, l’approche du pouvoir et de la dépendance et les théories du slack. Le lecteur pourra regretter l’absence d’un lien final entre la résistance, le slack et la performance, sujet auquel nous nous attachons désormais. Références Andaleeb S.S., Dependence relations and the moderating role of trust: implications for behavioral intentions in marketing channels, International Journal of Research in Marketing, 12, 1995, 154-172. Aulakh, P. S., & M. Kotabe, Trust and Performance in Cross-Border Marketing Partnerships: A Behavioural Approach, Journal of International Business Studies, 27:5, 1996, 1005-1032. Barney, Jay B., “Firm resources and sustained competitive advantage,” Journal of Management, Vol. 17, 1, 1991, 99-120. Boyle B., Dwyer F.R., Robicheaux R.A.& Simpson J.T., Influence strategies in marketing channels : measures and use in different relationship structures, Journal of Marketing Research, 29, November, 1992, 402-473. Brown J.R., Dev C.S. & Lee D.J., Managing marketing channel opportunism : the efficacy of alternative governance mechanisms, Journal of Marketing, 64, April, 2000, 51-65. Cannon J.P. & Homburg C., Buyer-supplier relationships and customer firm costs, Journal of Marketing, 65, January, 2001, 29-43. Cool, K., & J. Henderson, Power and Firm Profitability in Supply Chains: French Manufacturing Industry in 1993, Strategic Management Journal, 19:10, 1998, 909-927. Crozier, M. & Friedberg, E.,, L’Acteur et le système, Paris: Éditions du Seuil, 1977 Cyert, R. M., & J. G. March, A Behavioral Theory of the Firm, Englewood Cliffs: Prentice-Hall, 1963 Daniel F., Lohrke, F.T., Fornaciari, C.J., & Turner, R.A.,, Slack Resources and Firm Performance: a meta-analysis, Journal of Business Research, 57, 2004, 565-574. De Meyer, A., Nakane, J., Miller, J.G. & Ferdows, K.,, Flexibility: The next competitive battle; The manufacturing futures survey,” Strategic Management Journal, 10, 1989, 135-144. Emerson R. M., Power-Dependence Relations, American Sociological Review, 27, 1962, 31-41. Frazier, G.L., “The interfirm power-influence process within a marketing channel,” Research in Marketing, J.N. Sheth (ed.), Greenwich, Connecticut: JAI Press, 7, 1984, 63-91. Frazier, G. L., & R. C. Rody, The Use of Influence Strategies in Interfirm Relationships in Industrial Product Channels, Journal of Marketing, 55:1, 1991, 52-69. French, J.R.P., & B. Raven, The Bases of Social Power, in: D. Cartwright, Studies in Social Power, Ann Arbor: University of Michigan Press, 1959, 150-167. Gerwin, D., Manufacturing flexibility: A strategic perspective, Management Science, 39, 1993, 395¯410. Grewal R., & Tansuhaj, P.,, Building organizational capabilities for managing economic crisis: the role of market orientation and strategic flexibility, Journal of marketing, 65, 2001, 67-80. Gundlach, G. T., R. S. Achrol, & J. T. Mentzer, The Structure of Commitment in Exchange, 62 Journal of Marketing, 59:1, 1995, 78-92. Heide, J. B., & G. John, Do Norms Matter in Marketing Relationships?, Journal of Marketing, 56:2, 1992, 32-44. Hitt, M. A., Keats, B. W., & DeMarie, S. M., Navigating in the new competitive landscape: Building strategic flexibility and competitive advantage in the 21st century. Academy of Management Executive, 12, 1998, 22-42. Jap S. & Anderson E., Safeguarding Interorganizational Performance and Continuity Under Ex Post Opportunism, Management Science, 49, 2003, 1684 -1701. Jensen, M.C., Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers, American Economic Review, 76:2, 1986, 323-329. Kim K., On interfirm power, channel climate, and solidarity in industrial distributor-supplier dyads, Journal of the Academy of Marketing Science, 28, 3, 2000, 388-405. Klein, B., Crawford, R.G. & Alchian, A.A. . Vertical Integration, Appropriable Rents, and the Competitive Contracting Process, Journal of Law & Economics, vol.21, n°2, 1978, 297-326. Kohli, A., Determinants of Influence in Organizational Buying: A Contingency Approach, Journal of Marketing, 53:3, 1989, 50-65. Kotabe, M., X. Martin, & H. Domoto, Gaining From Vertical Partnerships: Knowledge Transfer, Relationship Duration, and Supplier Performance Improvement in the U.S. and Japanese Automotive Industries, Strategic Management Journal, 24:4, 2003, 293-316. Kumar N., Scheer L.K.& Steenkamp J., The effects of supplier fairness on vulnerable resellers, Journal of Marketing Research, 32, February, 1995, 54-65. Lusch R.F. & Brown J.R., Interdependency, contracting and relational behavior in marketing channels, Journal of Marketing, 60, 1996, 19-38. MacNeil, I.R., The New Social Contract: An Enquiry Into Modern Contractual Relations, New Haven: Yale University Press, 1980. Miner, A. S., T. L. Amburgey, & T. M. Stearns, Interorganizational Linkages and Population Dynamics: Buffering and Transformational Shields, Administrative Science Quarterly, 35:4, 1990, 689-713. Morgan R.& Hunt S., The commitment-trust theory of relationship marketing, Journal of Marketing, 58, 1994, 0-38. Mouritsen, J.,, The flexible firm: Strategies for a subcontractor's management control, Accounting, Organizations and Society, 24, 1999, 31-55. Nohria, N., & R. Gulati, Is Slack Good or Bad For Innovation?, Academy of Management Journal, 39:5, 1996, 1245-1265. Noordewier, T. G., G. John, & J. R. Nevin, Performance Outcomes of Purchasing Arrangements in Industrial Buyer-Vendor Relationships, Journal of Marketing, 54:4, 1990, 80-93. Pfeffer, J., & G. R. Salancik, The External Control of Organizations. A Resource Dependence Perspective, New York: Harper & Row, 1978 Porter, M. E., Competitive Strategy, New York: Free Press, 1980 Provan, K.G. & Gassenheimer, J.B., Supplier commitment in relational contract exchanges with buyers: A study of interorganizational dependence and exercised power, Journal of Management Studies, Vol. 31, 1, 1994, 55-68. Sanchez R., Strategic Flexibility in Product Competition, Strategic Management Journal, 16, 1995, 135-159. Sharfman, M. P., G. Wolf, R. B. Chase, & D. A. Tansik, Antecedents of Organizational Slack, Academy of Management Review, 13:4, 1988, 601-614. Singh, J. V., Performance, Slack and Risk-Taking in Organizational Decision Making, Academy of Management Journal, 29:3, 1986, 562-586. Subramani, M. R., & N. Venkatraman, Safeguarding Investments in Asymmetric Interorganizational Relationships: Theory and Evidence, Academy of Management Journal, 46:1, 63 2003, 46-62. Tan, J., & M. W. Peng, Organizational Slack and Firm Performance During Economic Transitions: Two Studies From an Emerging Economy, Strategic Management Journal, 24:13, 2003, 1249-1264. Volberda, HW, Building the Flexible Firm: How to Remain Competitive, Oxford: Oxford University Press, 1999 Wathne K.H.& Heide J.B., Relationship governance in a supply chain network, Journal of Marketing, 68, 2004, 73-89. Williamson, O.E., The economic institutions of capitalism, New York: Free Press, 1985. Zajac, Edward J. & Olsen, Cyrus P., From transaction cost to transaction value analysis: Implications for the study of interorganizational strategies, The Journal of Management Studies, Vol. 30, 1, 1993, 131-145. 64 ASAC 2005 Toronto, Ontario Dev K. Dutta (student) Richard Ivey School of Business University of Western Ontario WHY DO FIRMS BUILD NETWORK PARTNERSHIPS IN HYPERCOMPETITIVE ENVIRONMENTS? AN EXPLANATION FROM THE RESOURCE-BASED VIEW D’Aveni (1994) suggests that the only way for firms to survive in a hypercompetitive environment is to relentlessly compete against each other. Yet, we notice that intense competitive conditions lead firms not only to compete but also cooperate. Using the RBV as a theoretical lens, we examine why firms tend to form cooperative network partnerships in hypercompetitive environments. We suggest that by doing so, firms not only reduce environmental uncertainty but also derive performance benefits. Introduction The changing nature of competition and its implications for business strategy has been well documented (Ghemawat, 2002; Tsaliki and Tsoulfedis, 1998). As competition intensifies and evolves from static to dynamic, firms demonstrate greater involvement in managing the competition process (Barney, 1986). There is a growing body of literature on one specific kind of competitive environment termed hypercompetition (D’Aveni, 1994). It is suggested that under hypercompetition, the notion of sustained competitive advantage (SCA) has no meaning. In such environments, firms are continuously engaged in competitive moves and counter-moves, as they attempt to create and break temporary advantages in the marketplace. Given the high levels of competitive intensity that prevail in a hypercompetitive environment, firms would be wary of each other and refrain from forging any kind of collaborative partnerships between themselves. Yet, paradoxically enough it is seen that under hypercompetition firms engage in a plethora of network partnerships – sometimes with other firms that happen to be their closest competitors. To understand what motivates firms to engage in collaborative behavior in hypercompetitive environments, we use the Resource-Based View (RBV) and explain why firms would engage in such network partnerships. In doing so, we identify the reasons behind such behavior of firms and indicate how firms benefit from engaging in these partnerships. Resource-Based View of the Firm and SCA According to the RBV, sustained competitive advantage (SCA) accrues to the firm from its access to and ownership of heterogeneous, immobile, rare, hard-to-imitate and non-substitutable resource bundles and capabilities (Barney, 1991; Rumelt, 1984; Wernerfelt, 1984). Firm resources are broadly defined to include assets, capabilities, organizational processes, information and knowledge – anything that is owned and controlled by the firm and that allows it to create SCA, i.e. implement a value creating strategy that cannot be simultaneously created or its benefits duplicated by the firm’s current or potential competitors (Barney, 1991). The link between a firm’s resources and sustained competitive advantage is complex. It is not easy to be replicated by its competitors. 65 Imperfect inimitability and imperfect substitutability of resource bundles possessed by a firm make them non-tradable and lead to sustained competitive advantage (Peteraf, 1993). This happens because of both ex ante and ex post limits to competition arising out of possessing these resources. On the one hand, ex ante limits to competition ensure that the firm that first develops or secures access to the resource bundle does so in the absence of competition from others. Only this firm has the foresight and/or access to certain amenable conditions that make it the possessor of this resource. Yet, acquisition of the resource bundle also acts as a limiting factor on ex post competition (or competition faced by the firm after its acquisition of the specific resource bundles), making it difficult for these resources to be weaned away from the firm. As a result, once it has been developed, SCA remains within the firm. Having developed SCA, the firm is able to engage in activities that provide it an ability to capture economic rents (Amit and Schoemaker, 1993). Not only do the unique nature of these resources make it difficult for other firms to replicate them, but given that a firm acquires a specific resource bundle as a consequence of its past managerial decisions and action provides it with a unique opportunity set for securing advantage in the marketplace in the future (Lockett and Thompson, 2001). These resource bundles are not static, however. They are shaped and reshaped continually based on the firm’s organizational processes, specific asset position, paths available to it and its managerial action (Teece, Pisano and Shuen, 1997). Similarly, resources need not always be tangible. Intangible resources – both assets and competencies – enable a firm to create competency differentials vis-à-vis its competitors along functional, cultural, positional and regulatory dimensions, thus securing for itself competitive advantage (Hall, 1993, 1992). Knowledge forms the critical element in developing intangible and dynamic capabilities within the firm, given that this is essentially an evolutionary process (Grant, 1996b). Building dynamic capabilities involves experience accumulation, knowledge articulation and codification and the process is shaped by co-evolution of these learning mechanisms (Zollo and Winter, 2002). At the same time, to protect knowledge from expropriation and imitation, firms use institutional capabilities such as incentive alignment, employment contract, job design, and reorder rewards (Liebeskind, 1996). The RBV provides an internally focused view on building SCA through heterogeneous firm resources and capabilities by evaluating them for their value, rareness, inimitability and organizational orientation (Barney, 1991). At the same time, subsequent research has expanded the coverage to incorporate the role of context and the wider industry into the RBV. There are limits to the extent that SCA can be developed based on firm-level resources and capabilities. The context of the firm plays a significant role in deciding whether SCA will accrue to the firm or will it be invalidated by higher-order capabilities that a competitor secures access to (Collis, 1994). Firm capabilities and industry co-evolve and acquisition of a competitive position by the firm is based not only on the firm’s internal resources and capabilities, but also driven by market position and external relationships (Levinthal and Myatt, 1994). Barney and Zajac (1994) note that there appears to be a linkage between competition and competence, in that as firms learn to handle specific competitive challenges, they develop potentially valuable resources and capabilities that, in turn, provide firms with competitive advantage that enable them to face competitive conditions in future. In other words, though the RBV begins with the assumption that SCA is an endogenous characteristic of the firm, it acknowledges the existence of an inextricable link between firm and industry and suggests that competitive advantage and competitive positioning affect each other on a dynamic basis (Fahy, 2000). In their study of comparison of firm versus industry specific factors on SCA, Spanos and Lioukas (2001) note that both effects are important and explain different aspects of firm performance. Thus, firm performance is affected not only by access to resources but also by 66 competitive rivalry. Even industry structure characteristics are partially endogenously determined by firms’ specific strategic action. A question that arises is what happens to the notion of SCA when competitive pressures on the firm intensify. In the next section of the paper, we describe the competitive context that emerges around the firm in dynamic or fast changing environments. We focus on a specific type of the environment, viz. under hypercompetition. From Competition to Hypercompetition – Challenges for the Firm According to the classical view, competitive behavior on the part of firms arises as a response to neutralizing inequalities associated with deployment of capital and labor in the marketplace. It is “… a process characterized by free mobility of capital and labor, and with the concomitant tendential inter-industry equalization of the rate of profit, as well as of supply and demand” (Tsaliki and Tsoulfidis, 1998, p. 189). The classical view of competition is cognizant not only of price and quantity adjustments by the firm to achieve equilibrium but is also concerned with overall uncertainty and disequilibrium in the industry, which must necessarily be set right through competitive actions. Contrasted to this, the neo-classical view adopts a static view of competition and suggests that firms freely enter and exit from the market. Therefore, the intensity of competition prevailing at any point in time – from monopolistic to perfect competition – depends on the actual number of buyers and sellers active in the market at that point in time (Tsaliki and Tsoulfidis, 1998). Thus, the main issue of concern in the neo-classical view of competition is whether or not there is any persistence of profits above normal, which would indicate that there is a kind of market imperfection and a degree of monopolistic behavior. If competition is depicted in terms of its intensity along a continuum varying from monopoly through oligopoly to perfect competition, then hypercompetition falls in between oligopoly and perfect competition (D’Aveni, 1994). Characterized by rapid competitive maneuvers in the marketplace, hypercompetition occurs when a firm intensifies the level of competition in the marketplace by continuously generating new competitive advantage and destroying, neutralizing or making obsolete competitors’ advantage such that other firms have no option but to retaliate. Under conditions of hypercompetition, firms are engaged in redefining the so-called parameters of competition based on cost-quality balance, creation of know-how and moving into new markets or geographical territories as a first mover, and deep pockets of resources available (D’Aveni, 1994). By competing simultaneously on one or more of these parameters, firms create a series of short-term, discontinuous competitive advantage that always keeps them one step ahead of the competition. Craig (1996) distinguishes factors leading to hypercompetitive by classifying them as those that produce a “round” of hypercompetition versus those that lead to a broader “state” of hypercompetition. A round of hypercompetition can be initiated by a sudden competitive move on the part of one firm, which automatically leads to a series of retaliatory moves on the part of the others, thus creating tension and discontinuity in the marketplace. On the other hand, a state of hypercompetition is reached when “competitive change is continual and multiple rounds succeed one another with little respite in between” (p.162). It is brought about by a major shift in technology and innovation in the industry, which turns the rules of the game upside down as it were, and leads to a series of follow-up hypercompetitive shifts on the part of all players in the industry, not just the player that introduced the innovation. 67 Even though hypercompetition primarily is a firm-induced phenomenon, it is also affected by industry-level parameters. In their study of 352 firms in ten economic sectors between 1977 and 1984, Fombrun and Ginsberg (1990) find that strategic activity in terms of corporate aggressive in changing environments is conditioned by firm-level and sector level forces that both enable change as well as disable it. The extent of hypercompetition in an industry depends on the level of dynamic resourcefulness in the industry, i.e., the innate propensity of an industry to create new strategic assets (Thomas, 1996). Unstable market conditions brought about by hypercompetition actually result in firms building their survival strategies based on internal capabilities rather than based on cues received from served markets (Grant, 1996a). At the same time, however, firms may find it increasingly difficult to develop all the required capabilities in-house. One reason why this may be so is because of the bounded rationality of managers, which makes it difficult for them to accurately predict the content and pace of environmental change and prepare the organization to adequately respond to these forces. Chakravarthy (1997) suggests that firms do not really create and/or “manage” turbulence on an ongoing basis; instead they cope with it in the best possible way they can, sometimes through offensive postures vis-à-vis competition and sometimes through exit mechanisms. The environment now becomes a major source of uncertainty (Elenkov, 1997). Not only is it characterized by changes that managers must make sense of, but also managerial action is dependent upon their perceptions and interpretation of the environment (Schneider and De Meyer, 1991). Managers interpret and “know” the environment in one of three ways: (i) as an objective environment out there, given, completely known and with the organization embedded in it, (ii) a perceived environment, different from the objective environment in that managers are constrained by their bounded rationality and can, therefore, never completely know the environment in its entirety, and (iii) an enacted environment, which suggests that the organization and environment are enacted (or jointly created) as an outcome of managers’ social interaction processes (Smircich and Stubbart, 1985). Thus, managerial action is directly affected by how managers interpret the environment and make sense of signals that they perceive as emanating from the environment and the ways in which they can affect the firm. Environmental uncertainty is also perceived as falling into one of three categories: (i) state uncertainty, or uncertainty relating to the conditions emanating in the perceived environment, (ii) effect uncertainty, or uncertainty relating to predicting what the nature of a future state of the environment will be or how a future environmental change will affect the organization, and (iii) response uncertainty, or uncertainty associated with understanding the range of response options an organization has in dealing with the environment and what the value or utility of each option might be (Milliken, 1987). In other words, managerial perception of uncertainties associated with the environment is compounded not only by differences relating to categorization of the environment but also differences associated with understanding of environmental uncertainty. Under hypercompetition as competitive pressures build up in the environment and competitive moves intensify in terms of attributes such as disruption, speed, surprise and dissonance, it becomes increasingly difficult for managers to make sense of the environmental uncertainty and take appropriate strategic action. This is when managers are compelled to look beyond the boundaries of their own organization and look for potential coping mechanisms that may exist or be created in the wider industry network. 68 Meeting the Challenges of Hypercompetition – Benefits of Network Partnerships Given the high levels of environmental uncertainty, the firm is faced with a gap in the knowledge it requires to meet emerging market requirements. Hypercompetitive environments demand of the firm an increased ability to integrate knowledge and constantly combine and recombine its capabilities. However, to do so firms face a practical constraint – a burgeoning knowledge gap or what can be termed as a “knowledge trap”. We suggest that firms engage in network partnership with close competitors in order to get out of this knowledge trap, even though it involves easing out some of the competitive intensity that is otherwise required. In other words, in deciding to emphasize collaboration in areas of significance for knowledge gathering with close competitors, firms make a conscious choice that helps them reduce environmental uncertainty over creating intensely disruptive competitive conditions in the market. Appleyard (1996) finds that even in fiercely competitive industries firms use a plethora of mechanisms, both public and private, for the purpose of reducing environmental uncertainty through knowledge sharing. For instance, under Schumpeterian competition (Schumpeter, 1942) firms not only compete against each other but also collaborate with select competitors in order to earn above normal rents by (i) preventing new firms from entering the industry and sharing the factors of production, (ii) dissuading infusion of substitute technology, (iii) reducing overall environmental uncertainty at least in niches where the firms are collaborating, and (iv) agreeing with rivals in order to more clearly define the boundaries of collaboration and competition (Garud, 1994). Network ties between organizations allow for learning between partners and this helps in reducing environmental uncertainty and facilitating organizational adaptation (Kraatz, 1998). It has also been suggested that inter-firm networks improve a member firm’s competitive position by allowing for cost minimization of transactions even while maintaining flexibility (Park, 1996). Therefore, in intensely competitive environments it is beneficial for firms to engage in network partnerships, which lead to reduction in environmental uncertainty. In contrast to this, the requirement of forming network partnerships will be less in static environments, since the level of environmental uncertainty in such environments is relatively low. Therefore, we suggest: Proposition 1: In hypercompetitive environments, firms tend to engage in network partnerships in order to reduce environmental uncertainty. Again, under high levels of environmental uncertainty the firm may not be able to plan for and internally develop its valuable, rare, inimitable and non-substitutable resource bundles. In fact, it may even be that some of the unique resources required by the firm are available only outside its organizational boundaries. This possibility may especially arise when the environment is fast changing and organizational boundaries get blurred. McEvily and Zaheer (1999) suggest that sources of combinative capabilities of firms can lie outside their organizational boundaries or more specifically in their network resources comprising bridging ties and relationships with other firms in the external environment where the firm is embedded. Specifically, in industries characterized by open systems architecture and network technologies (such as Sun Microsystem’s Open Systems Strategy) there is evidence that collaborative partnerships and sharing of technical knowledge with competitors and even facilitation of market entry by them helps the firm in question in the long term (Garud and Kumaraswamy, 1993). This is a clear example of the benefit of network partnerships towards preserving SCA in markets that are otherwise highly competitive. Combs and Ketchen (1999) find that firms tend to go for cooperative mechanisms to build resources in the 69 long-term despite the unfavorable exchange conditions it might bring in the short run. In fact, this occurs because many of the resources available and used by a firm may not strictly lie within the firm boundaries. Powell, Koput and Smith-Doerr (1996) suggest that in industries characterized by a complex and rapidly expanding knowledge base and with dispersed nodes of expertise (e.g. biotechnology), the locus of innovation is found within networks of firms rather than in individual firms. This does not mean that these firms are not competitors in the traditional sense of the term. What this indicates, however, is that to build SCA firms make choices, which enable them to balance external competitive pressures vis-à-vis needs for developing capabilities or building competence. In fact, it has been suggested that a firm’s position in the network and the overall network create resource asymmetries by influencing the flow of assets, information and status, thus leading to continuous configuration and reconfiguration of competitive positions (Gnyawali and Madhavan, 2001). Network relationships facilitate firms to obtain relatively easily and quickly resource bundles required by them. For instance, firms build capabilities using a socio-cognitive approach; human and organizational resources combine to form group capabilities, which, in turn, lead to creation of SCA (Ginsberg, 1994). Similarly, firms form alliance relationships with other firms, in order to ensure resource complementarity and achieve higher standards of business performance (Harrison, Hitt, Hoskisson and Ireland, 2001). In their longitudinal study of 1903 competitive moves undertaken in the software industry, Young, Smith and Grimm (1996) find that as the number of horizontal cooperative arrangements a firm engages in goes up, the firm performance improves. In other words, it is natural that as firms engage in consolidating their competitive advantage vis-à-vis other players in the industry, they will take recourse to occupying network positions and/or involve in network arrangements with other players. In effect, firm resources such as relation-specific assets, knowledge-sharing routines, complementary resources/capabilities and facilitating network governance structures will increasingly be found outside of the firm (Dyer and Singh, 1998). We posit that in hypercompetitive environments, given the uncertainties associated with resource requirements, intangibility and dynamic capabilities expected out of resources, and the higher need for creating influence in the market, will lead to firms looking out into the network as a possible source for providing such resources. Therefore, under hypercompetition firms make a conscious choice to forge selective network partnerships that help them to derive superior rent and preserve their SCA. Therefore, we suggest: Proposition 2: In hypercompetitive environments, firms that engage in network partnerships perform better than firms that do not do so. Discussion and Implications In this paper, we have argued that though it seems paradoxical why firms would build network partnerships in the face of relentless competition in a hypercompetitive environment, there are strong reasons for firms to do so. We have noted that there is empirical support of firms engaging in cooperative behavior in fast changing environments. Using the RBV as our theoretical lens, we have explored the reasons why firms happen to be intense competitors in a hypercompetitive environment would find merit in developing collaborative partnerships. In doing so, we have provided an argument for why firms engage in both collaborative behavior in an intensely competitive environment, such as hypercompetition. We used evidence from previous literature to suggest that not only does SCA accrue to a firm based on its dynamic capabilities and 70 intangible resource bundles but that it is also subject to competitive pressures, especially in fast changing or hypercompetitive environments. In building our theoretical arguments, our paper makes three major contributions. Firstly, we suggested that environmental uncertainty is a critical challenge the firm must deal with in a hypercompetitive environment. A relentlessly shifting environment faced with high competitive intensity impinges upon the firm high levels of environmental uncertainty. Multiple notions of the environment as well as of uncertainty in the sensemaking schema of managers, coupled with volatile signals emanating out of the environment, create conditions that make it difficult for managers to develop coherent strategic action plans that can adequately address the environmental uncertainties. In addition, the increased focus on knowledge and the requirement of knowledge integration across all competitive activities makes it more difficult for firms to sustain competitive advantage and avoid lapsing into a competency trap. Secondly, we show how network partnerships help firms to tackle environmental uncertainty - by sourcing appropriate knowledge and capabilities that may be available in the wider network and achieve performance benefits. In doing so, we provide a logic for why firms engage in seemingly conflicting pursuits of simultaneous cooperative and collaborative relationships with other players in environments that demonstrate high levels of uncertainty and competitive pressures. In elaborating our argument on simultaneous cooperative and competitive behavior on the part of firms, we support similar notions that have emerged in recent literature on firm behavior in fast-changing environments, e.g. coopetition (e.g. Bengtsson and Kock, 2000) or co-evolution (e.g. Hinterhuber, Handlbauer, Matzler and Valdani, 1998). The third and foremost contribution of our paper is in the assertion that firms employ choice in balancing competitive pursuits versus building collaborative arrangements in order to reduce environmental uncertainty and retain SCA. We suggest that firms make this decision as a part of their overall strategic choice and action. In drawing this inference, we provide an explanation of why firms engage in apparently conflicting cooperative and competitive behavior in hypercompetitive markets. Our research has several implications for managers. Firstly, it provides a theoretical explanation for managerial action noticed in hypercompetitive environments, which finds firms often engage in collaborative partnerships with their close competitors. Secondly, by indicating that there are strong reasons for engaging in simultaneous cooperative and competitive moves in the marketplace, our research exhorts managers to consider this as a powerful strategic alternative to engaging in simply cooperative or competitive behavior. By drawing a reference to network resources and the possibility that a firm’s critical resources may lie outside the organizational boundaries, our research suggests to managers to critically examine their firm’s resource position and source of competitive advantage and take strategic action vis-à-vis competitors in line with that assessment. Finally, our research offers directions to managers on ways of conceptualizing hypercompetition as a complex phenomenon. We suggest that not only is hypercompetition always unavoidable or difficult to deal with, but that it can provide certain new advantages to firms provided managers unleash the creative forces of hypercompetition and build network partnerships for mutual benefit. 71 References Amit, R. and Schoemaker, P.J.H., “Strategic assets and organizational rent,” Strategic Management Journal, 14, (1993), 33-46. Appleyard, M.M., “How does knowledge flow? Interfirm patterns in the semiconductor industry,” Strategic Management Journal, 17, (Winter Special Issue 1996), 137-154. Barney, J., “Firm resources and sustained competitive advantage,” Journal of Management, 17 (1), (1991), 99-120. Barney, J.B., “Types of competition and the theory of strategy: Toward an integrative framework,” Academy of Management Review, 11 (4), (1986), 791-800. Barney, J.B. and Zajac, E.J., “Competitive organizational behavior: Toward an organizationally-based theory of competitive advantage,” Strategic Management Journal, 15, (1994), 5-9. Bengtsson, M. and Kock, S., “’Coopetition’ in business networks – to cooperate and compete simultaneously,” Industrial Marketing Management, 29, (2000), 411-426. Chakravarthy, B., “A new strategy framework for coping with turbulence,” Sloan Management Review, (Winter, 1997), 69-82. Collis, D.J., “How valuable are organizational capabilities?” Strategic Management Journal, 15, (1994), 143-152. Combs, J.G. and Ketchen, D.J, Jr. 1999. “Explaining interfirm cooperation and performance: Toward a reconciliation of predictions from the resource-based view and organizational economics,” Strategic Management Journal, 20 (9), (1999), 867-888. Craig, T., “The Japanese beer wars: Initiating and responding to hypercompetition in new product development,” Organization Science, 7 (3), (1998), 302-321. D’Aveni, R. A., Hypercompetition: Managing the dynamics of strategic maneuvering. New York: The Free Press, 1994. Dyer, J.H. and Singh, H., “The relational view: Cooperative strategy and sources of interorganizational competitive advantage,” Academy of Management Review, 23 (4), (1998), 660-679. Elenkov, D.S., “Strategic uncertainty and environmental scanning: The case for institutional influences on scanning behavior,” Strategic Management Journal, 18 (4), (1997), 287-302. Fahy, J., “The resource-based view of the firm: Some stumbling blocks on the road to understanding sustainable competitive advantage,” Journal of European Industrial Training, 24 (2, 3, 4), (2000), 94-104. Fombrun, C.J. and Ginsberg, A., “Shifting gears: Enabling change in corporate aggressiveness,” Strategic Management Journal, 11, (1990), 297-308. 72 Garud, R., “Cooperative and competitive behaviors during the process of creative destruction,” Research Policy, 23, (1984), 385-394. Garud, R. and Kumaraswamy, A., “Changing competitive dynamics in network industries: An exploration of Sun Microsystems’ open systems strategy,” Strategic Management Journal, 14, (1993), 351-369. Ghemawat, P., “Competition and business strategy in historical perspective,” Business History Review, 76, (Spring, 2002), 37-74. Grant, R.M., “Prospering in dynamically-competitive environments: Organizational capability as knowledge integration,” Organization Science, 7 (4), (1996a), 375-387. Grant, R.M., “Toward a knowledge-based theory of the firm,” Strategic Management Journal, 17 (Winter Special Issue, 1996b), 109-122. Ginsberg, A., “Minding the competition: From mapping to mastery,” Strategic Management Journal, 15 (1994), 153-174. Gnyawali, D.R. and Madhavan, R., “Cooperative networks and competitive dynamics: A structural embeddedness perspective,” Academy of Management Review, 26 (3), (2001), 431-445. Hall, R., “A framework linking intangible resources and capabilities to sustainable competitive advantage,” Strategic Management Journal, 14, (1993), 607-618. Hall, R., “The strategic analysis of intangible resources,” Strategic Management Journal, (13), (1992), 135-144. Harrison, J.S., Hitt, M.A., Hoskisson, R.E. and Ireland, R.D., “Resource complementarity in business combinations: Extending the logic to organizational analysis,” Journal of Management, 27, (2001), 679-690. Hinterhuber, H.H., Handlbauer, G., Matzler, K. and Valdani, E., “The new rules of competition: From evolution to co-evolution,” CEMS Business Review, 2, (1998), 167-175. Kraatz, M.S., “Learning by association? Interorganizational networks and adaptation to environmental change,” Academy of Management Journal, 41 (6), (1998), 621-643. Levinthal, D. and Myatt, J., “Co-evolution of capabilities and industry: The evolution of mutual fund processing,” Strategic Management Journal, 15, (1994), 45-62. Liebeskind, J.P., “Knowledge, strategy, and the theory of the firm,” Strategic Management Journal, 17 (Winter Special Issue, 1996), 93-107. Lockett, A. and Thompson, S., “The resource-based view and economics,” Journal of Management, 27, (2001), 723-754. McEvily, B. and Zaheer, A., “Bridging ties: a source of firm heterogeneity in competitive capabilities,” Strategic Management Journal, 20, (1999), 1133-1156. 73 Milliken, F.J. 1987. “Three types of perceived uncertainty about the environment: State, effect, and response uncertainty,” Academy of Management Review, 12 (1), (1987), 133-143. Park, S.H., “Managing an interorganizational network: A framework of the institutional mechanism for network control,” Organization Studies, 17 (5), (1996), 795-824. Peteraf, M.A., “The cornerstones of competitive advantage: A resource-based view,” Strategic Management Journal, 14, (1993), 179-191. Powell, W.W., Koput, K.W. and Smith-Doerr, L., “Interorganizational collaboration and the locus of innovation: Networks of learning in biotechnology,” Administrative Science Quarterly, 41, (1996), 116-145. Rumelt, R., “Towards a strategic theory of the firm,” In Lamb, R. (Ed.): Competitive strategic management, Englewood, NJ: Prentice-Hall, 1984. Schneider, S.C. and De Meyer, A., “Interpreting and responding to strategic issues: The impact of national culture,” Strategic Management Journal, 12 (4), (1991), 307-320. Schumpeter, J.A., Capitalism, socialism and democracy, New York: Harper and Row, 1942. Smircich, L. and Stubbart, C., “Strategic management in an enacted world,” Academy of Management Review, 10 (4), (1985), 724-736. Spanos, Y.E. and Lioukas, S., “An examination into the causal logic of rent generation: Contrasting Porter’s competitive strategy framework and the resource-based perspective,” Strategic Management Journal, 22, (2001), 907-934. Teece, D., Pisano, G. and Shuen, A., “Dynamic capabilities and strategic management,” Strategic Management Journal, 18 (7), (1997), 509-533. Thomas, L.G., III, “The two faces of competition: Dynamic resourcefulness and the hypercompetitive shift,” Organization Science, 7 (3), (1996), 221-242. Tsaliki, P. and Tsoulfedis, L., “Alternative theories of competition: evidence from Greek manufacturing,” International Review of Applied Economics, 12 (2), (1998), 187-204. Wernerfelt, B., “A resource based view of the firm,” Strategic Management Journal, 5, (1984), 171-180. Young, G., Smith, K.G. and Grimm, C.M., “‘Austrian’ and industrial organization perspectives on firm-level competitive activity and performance,” Organization Science, 7 (3), (1996), 243-254. Zollo, M. and Winter, S.G., “Deliberate learning and the evolution of dynamic capabilities,” Organization Science, 13 (3), (2002), 339-351. 74 ASAC 2005 Toronto, Ontario Karen L. Ekstein Schulich School of Business York University ORGANIZATIONAL RESPONSIVENESS: A CONCEPTUAL FOUNDATION AND RESEARCH AGENDA The strategy literature is silent about organizational responsiveness and its role in organizational outcomes. This paper defines organizational responsiveness as an important concept for strategy researchers. Initial steps are taken in understanding organizational responsiveness, particularly in dynamic environments. Challenges facing top managers tasked with creating responsive organizations are specifically noted. How firms respond in light of their environments can have implications for firm outcomes. Works on contingency theory suggest the necessity to create alignments between firms and the environments; these alignments have implications for various outcomes, including performance and survival (Morgan, 1997). In his early works, Schumpeter (1942) considered industry-level change, highlighting that periodic waves of creative destruction flow through industries, changing their rules and, essentially, change the shape of industries by determining who stays in and who is squeezed out. The implication of such works is that firms which are sensitive to their environments and build capabilities to respond to their environments and its changes will be more likely to benefit competitively and to soldier through tumultuous periods. It would seem that organizational responsiveness is a particularly relevant notion in the current competitive environment – an environment often described as high-velocity, fast-changing and dynamic (Bennis, Spreitzer & Cummings, 2001; Davenport & Prusak, 1998; Drucker, 1999; Eisenhardt & Martin, 2000; Hesselbein, Goldsmith & Beckhard, 1997; Kotter, 1995; Ray & Rinzler, 1993; Seely Brown, 1997). It would seem that consideration of how organizations should respond within such environments would be an important area for organizational researchers to consider, both to remain insightful and relevant to practice, and to understand the influences on firm-level competitiveness, and firm survival. Though scholarly works provide clues about organizational responsiveness, they remain fairly silent about it as an important strategic management concept and in terms of how it might operate, and become a firm-level capability. As such, a variety of practical considerations arise. First, how should organizational responsiveness be defined and conceptualized? Second, how should firms assess their organizational responsiveness and develop it or create it? And, how should these questions be approached, assuming the underlying dynamism and change characterizing the current competitive environment? Approaching these questions is an enormous, important and complex task which has the 75 potential to pave the way for a new branch of relatively untouched strategic management research and theory. This paper takes a modest first step in understanding organizational responsiveness in dynamic contexts by attending to it at the firm level of analysis. Particularly, it is suggested that enabling organizational responsiveness at the firm level presents challenges to top managers, having implications for the creation and development of organizational competences to support it. It is only by understanding organizational responsiveness that it is possible to work towards understanding how organizational responsiveness might be conceived of and operate as an organizational capability. The paper contributes to the literature in the following way. First, it establishes that indeed, organizational responsiveness is a noteworthy concept within the strategic management literature, particularly amidst organizational dynamism. Second, a framework is established with which future research and theory might be guided and from which a variety of research questions can be seen to emerge. Third, this paper considers complexities surrounding how top managers might prepare to conceive of organizational responsiveness. And fourth, how competences should be developed in light of these complexities is suggested. Finally, the implications of this framework are described, primarily in terms of shaping a research agenda. Foundations for Studying Organizational Responsiveness The notion of organizational responsiveness and its role in organizational outcomes is suggested in various works within the strategic management field. Fundamental to the competitive signaling literature is the suggestion that firms exist within environments characterized by information asymmetries. Thus, organizational decision makers rely heavily on interpreting signals within their environments in order to make decisions regarding investments in other firms (Fombrun & Shanley, 1990) and competitive moves (Dollinger, Golden & Saxton, 1997). Much of the literature is focused, however, on how signals can be used or created to attract investment and to motivate purchase decisions (cf. Weigelt & Camerer, 1988; Grossman & Helpman, 1999; Deutsch & Ross, 2001; Lounsbury & Glynn, 2001; Stuart, Hoang & Hybels, 1999). Within this literature, it is clear that how signals are interpreted has implications for decision making and organizational action. A precise focus on whether, when and how signal interpretation leads to organizational responsiveness is lacking. Similarly, the issue interpretation literature provides some insights into organizational responsiveness. These works focus on whether and how issues are labeled as strategic in organizations (Dutton, 1986; Dutton & Jackson, 1987). These works further suggest that there are implications for whether and how issues are labeled as strategic for how top managers respond in light of these issues; this leads to the suggestion that issue interpretation leads to differences in responsiveness to the same events across firms (Jackson & Dutton, 1988). In some works, it is specifically shown that the labeling of issues as strategic, and whether a strategic issue is labeled as an opportunity or threat, has influences on how responses are undertaken within an organization and particularly 76 mandated by its leaders (Sharma, 2000; Sharma, Pablo & Vredenburg, 1999). Clearly, the issue interpretation literature provides an idea as to how issues, if and when they are labeled as strategic and how they are labeled as strategic (opportunities versus threats) has implications for motivating organizational responsiveness. These works are only two examples from the strategic management literature that provide the impetus for focusing more specifically on organizational responsiveness within the strategic management literature. It might be noted, however, that neither of these works focus specifically on organizational responsiveness despite the concept’s centrality to the works. These works provide initial groundwork and an impetus whereupon to consider the role of organizational responsiveness vis-à-vis firm-level factors and outcomes. An initial framework delineating the relationships suggested in the literature is shown in Figure 1. --------------------------------Insert Figure 1 About Here --------------------------------As shown in Figure 1, organizational responsiveness can be seen to involve a relationship between a firm and its environment. The literature that forms the basis whereupon the framework is developed focuses primarily on the interpretive factors surrounding top managers and their influence on the relationships between firms and their environments. Clearly, there is substantial room in which to elaborate on the framework. It should also be particularly noted that within the framework presented, organizational responsiveness is seen to lead to firm outcomes rather than performance; this is deliberate, and based on recognition in recent years that dominant measures of firm performance do not necessarily capture the range of outcomes that firms might work towards (i.e. works on triple bottom line, stakeholder theory, works critical of performance measures). With regards to responsiveness, goals might indirectly be performance-related, but directly are contingent on the targets of responsiveness (i.e. the customers, employees and industry trends to which the firm seeks to respond, etc.). It is feasible that the target of responsiveness defined by a firm has implications for how organizational responsiveness is characterized and manifested at the firm-level of analysis, how success or failure of organizational responsiveness is assessed and evaluated and how differences in organizational responsiveness may be manifested across firms. Organizational Responsiveness and Dynamic Contexts A range of factors and trends within organizational environments have led to characterization of the business context as high-velocity, changing and dynamic. There are various forces which have led to such changes within business environments, among which are forces of technology, globalization and economic transformation (Bennis, Spreitzer & Cummings, 2001; Davenport & Prusak, 1998; Drucker, 1999; Eisenhardt & Martin, 2000; Hesselbein, Goldsmith & Beckhard, 1997; Kotter, 1995; Ray & Rinzler, 1993; Seely Brown, 1997). Among the effects of such changes is a suggested need for managers to shift assumptions about management (Drucker, 1999), to focus their attention carefully 77 (Davenport & Beck, 2001), and to seek methods of achieving current strategies with a mind towards the future (Christensen, 2004). Thus there is an urgent need for academics and managers alike to begin to understand the need for organizational responsiveness within firms and its relationships with organizational outcomes, competitiveness and survival. It is necessary not only to begin to understand what organizational responsiveness means in an academic sense, but also to understand what is expected of it and the top managers responsible for fostering it amidst the cited dynamism. Demands on Organizational Responsiveness In setting the stage for developing organizational responsiveness, top managers play a crucial role. It is necessary for them to recognize that developing competence around organizational responsiveness is necessary for competitiveness and survival. It is also necessary for them to recognize that preparing their firms for organizational responsiveness requires them to undergo certain changes in mindset and grapple with certain paradoxes. It is necessary for them to realize how contemplating these paradoxes and undergoing changes in mindset affects whether and how organizational responsiveness is fostered within and manifested by their organizations. Three paradoxes that top managers must understand, consider and deal with in the current context are: balancing current and future strategies, balancing internal and external attention and balancing proactive and reactive responsiveness modes. Balance between current and future strategic orientations The challenge of balancing current and future orientations has received dominant attention within various works on innovation. It is suggested that firms seek mechanisms with which to ensure that decision makers are able to work towards the fulfillment of current strategic objectives while also remaining sensitive to the need to think long-term. Balance between proactive and reactive responsiveness modes Modes of responsiveness, as seen here, refer to how firms relate with their environments. Though commonly assumed to be equated with reaction, responsiveness accommodates the consideration of proactive thinking and behavior. While a reactive mode of responsiveness may be seen as involving responding to that which has already occurred, a proactive mode of responsiveness entails sensing that which might occur and devising methods with which to respond prior to possible occurrences. It is in this vein that another challenge is presented to decision makers within the current environment, as they seek to be both reactive and proactive in responding to their environments. Balance between internal and external attention What is an organizational environment? What is or are the organizational environments to which decision makers must attend? The concept of organizational environment, as it is commonly conceived of within the strategic management literature, refers to the external environment. Yet, the internal organizational environment has also 78 received some attention, as leaders of firms seek to develop mechanisms with which to develop an orientation toward the future. Thus, in understanding the challenges of focus of attention, it is necessary for managers to develop a focus of attention that is both internally and externally oriented and sensitive to features of each. Organizational Responsiveness: An Initial Framework Understanding demands for organizational responsiveness in dynamic environments is not only a conceptual or contemplative activity; rather, it has important implications for how organizational responsiveness is facilitated and operates. The Organizational Responsiveness Framework is presented in Figure 2. --------------------------------Insert Figure 2 About Here --------------------------------As can be seen from the framework, there are various tasks with which a top manager is charged with seeking to foster organizational responsiveness within their organizations. It is necessary for these top managers to develop an underlying organizational responsiveness mindset as well as to ensure that appropriate mechanisms are developed within four dimensions of organizational responsiveness. It is in this way that top managers prepare their organizations for organizational responsiveness in a manner that enables him or her to envision the relationships between organizational responsiveness, organizational factors, processes and activities and organizational outcomes. The underlying mindsets and four dimensions are thus considered. Underlying Mindset of Organizational Responsiveness Within the current dynamic business context, the mindset that managers are required to adopt for responsiveness is one which is oriented towards both the achievement of current strategies as well as future ones. In this sense, top managers are required to plan for today with the future in mind. And, as might be imagined, within such a mindset, top managers must not only recognize, but also attempt to balance both orientations. Four Dimensions of Organizational Responsiveness In addition to the overarching mindset that appreciates both an orientation towards fulfilling current strategy and working toward future strategies, it is also necessary for organizations to develop organizational capabilities within the four organizational responsiveness dimensions. The dimensions, depicted in Figure 2, are: external/reactive, external/proactive, internal/reactive and internal/proactive. The External/Reactive Dimension The external/reactive dimension involves an orientation to the firm’s external environment and reactive responsiveness mode. It is suggested within the literature that while a capability in this dimension is necessary within a firm, it is not necessarily useful to be or remain competitive and survive industry upheaval. 79 The External/Proactive Dimension The external/proactive dimension involves an orientation to a firm’s external environment combined with a proactive responsiveness mode. This dimension is suggested to lead to value within the current economy, for it presupposes that firms seek to sense and analyze trends, patterns and to create linkages to motivate proactive responsiveness to these ideas. The Internal/Reactive Dimension The internal/reactive dimension involves an orientation inward, to an organization’s internal environment, combined with a reactive responsiveness mode. This dimension is oriented toward detecting and solving problems and barriers to operations and strategic management as they occur. The Internal/Proactive Dimension The internal/proactive dimension involves an orientation inward, to an organization’s internal environment, combined with a proactive responsiveness mode. This dimension is oriented toward detecting and solving issues, trends and patterns that might indicate emergent problems, seeking to deal with these before they become problems. From Mindsets and Dimensions to Competence Further, in addition to preparing oneself cognitively for organizational responsiveness, it is also necessary for top managers to begin to prepare their organization for responsiveness. This involves auditing the organization with recognition of the precise factors, processes, activities and mechanisms operating within each quadrant, ensuring that these support responsiveness in that quadrant. Examples of such factors, processes, activities and mechanisms are provided in Table 1. --------------------------------Insert Table 1 About Here --------------------------------As can be seen within the table, there are three dimensions upon which each quadrant is evaluated. For the purposes of analysis, these have been described with regards to information collection methods, modes of inquiry and analytical methodology and examples. Information collection refers to the ways in which the organization seeks information for responsiveness within each quadrant. Modes of inquiry and analytical methodology include approaches to information, for example, which questions are asked? How the information is approached? Examples are provided to illustrate how each quadrant relates with responsiveness, and particularly can be seen to contribute to (or detract from) successful responsiveness. The framework can also be used by top managers as an auditing mechanism. The dimensions noted within each quadrant can also be seen as dimensions upon which each quadrant might be audited within organizations. For example, how information is collected 80 within a firm and whether such methods are working or appropriate might be considered. In assessing modes of inquiry and analytical methodology it is possible for top managers to question whether the right questions are being asked in viewing information and in order to facilitate the goals of responsiveness within that quadrant. Further, specific examples of when and how mechanisms have and have not worked might also be considered within each quadrant. These analytical processes can lead top managers to ask necessary questions about whether the methods of information collection within each quadrant support the mode of responsiveness and what barriers stand in the way of responsiveness. It is on this basis that they might seek to correct these and develop measures of success around responsiveness. From Relevant Competences to Organizational Capability Organizational responsiveness is a competitive necessity in dynamic environments. Thus, while firms might have and develop competences within each quadrant, this does not necessarily mean that they have a capability which will yield superior outcomes. The question that thus arises is how might an organizational responsiveness capability be conceptualized? In order to understand this, it is necessary to first contemplate what comprises a capability according to the literature. Grant (1991) describes capabilities as the: “capacity for a team of resources to perform some task or activity” (p. 119). Collis (1991) describes capabilities as “socially complex routines that determine the efficiency with which firms physically transform inputs into outputs” (p. 145). Notable, however, is that the value of resources is seen to be derived from their value, rarity, inimitability and non-substitutability (Barney, 1991). As Barney (1986) suggests, in the case that a capability can be devised by using a framework, it is replicable by other firms. Thus, for organizational responsiveness to become a capability this implies something grander than a series of competences within the four dimensions outlined. Such investigations are out of the scope of the current paper, but might be seen to involve investigations into whether and how firms develop super-competence with regards to one quadrant or superior alignments with environments as based on patterns of dependency between firm and environment. Other insights into how firms create organizational responsiveness capabilities might involve assessing how competences do or should work together, across the framework; this involves recognizing that each quadrant does not work independently, but rather, requires an overarching coordination mechanism whereby the quadrants work simultaneously and movement between these quadrants is facilitated. Refining the Research Agenda The question for any work based within a relatively untouched area of research is how to scope the initial forays into that research area in a manageable fashion. It is in this sense that it is possible to turn to creating a research agenda within this area of research. While the current work begins to present and establish responsiveness as a critical strategic management concept with particular importance in dynamic environments, it does only represent a first step. 81 Yet, this work asks almost as many questions as it provides answers, as it identifies a multitude of research areas that might be exploited in future works. The current work presents a framework to guide such inquiries by proposing constructs and relationships, though the need for these constructs, relationships and elements of context to be elaborated is acknowledged. The framework does provide a foundation and springboard for future works on and inquiries into organizational responsiveness. This work also specifies the role of top managers in organizational responsiveness, since they are often cited in the strategic management literature to play a dominant role in setting the stage for organizations to work in a particular strategic direction. These are the people who presumably lead to organizational responsiveness decisions and actions and set the framework wherein organizational responsiveness will occur. As such, the current work focuses pointedly on the paradoxes that face top managers as they seek to develop responsive organizations. Three paradoxes have been identified and it is suggested that whether and how top managers seek to overcome these paradoxes and how they create mechanisms to deal with these paradoxes has implications for both organizational responsiveness and organizational outcomes. It has been demonstrated that the organizational responsiveness framework might be used to ground further studies on organizational responsiveness as well as the basis upon which top managers might launch audits into the competences and their applicability for organizational responsiveness to both occur and help them to achieve organizational goals. Initial insights are provided into how organizational responsiveness might be conceptualized as a capability, though it is suggested that to be viewed as a capability, organizational responsiveness is somewhat more than organizational competences within the quadrants in the framework. Some recommendations for what comprises an organizational responsiveness capability and how this might be investigated are provided. There is clearly substantial room for contributions to the notion of and understanding of organizational responsiveness and its role in organizational outcomes and in strategic management in general. This paper lays the groundwork for conceptual, theoretical and empirical developments in this relatively untouched but critical area of research. 82 Figure 1: Relationships Between Environments and Outcomes Organizational Responsiveness Environments Outcomes ? Competitive Signals Issue interpretation Figure 2: Organizational Responsiveness Framework Environmental Orientation Modes of Responsiveness Reactive Proactive Internal/Reactive Internal/Proactive External/Reactive External/Proactive Internal External Table 1: Mechanisms within the Dimensions Information Collection Mode of inquiry and analytical methodology Examples of dimension in action Internal/Reactive Through surveys, surveys, monthly reports (e.g. on turnover, absenteeism, defect rates) Most likely shared at periodic management meetings, deductive approach. Reacts to problems after they are problems. Introduction of a comment box for employees. Internal/Proactive Best practices in operations strategy Inductive, based on frameworks like value chain Seeks to identify patterns and trends before they become problems or to seek Feedback sessions with management, continual collection and assessment of data 83 External/Reactive External/Proactive Signals, investment reports, news characteristically. On trends, patterns, uncreated linkages and fusion Often directed information searches – searching for particular information Deductive, interpretive Browsing, fusion of information across documents, organizational units, etc. Inductive, based on frameworks like value chain, and comparison to competitors and other environmental factors Initiating a retaliatory price cut. Disruptive technology, innovation JIT, Dell, Amazon, Walmart References Barney, J. 1986. Strategic factor markets: Expectations, luck and business strategy. Management Science, 32(10), 1231-1241. Barney, J. 1991. Firm resources and sustainable competitive advantage. Journal of Management, 77, 99-120. Bennis, W., Spreitzer, G.M. and Cummings, T.G. Eds. 2001. The future of leadership. Jossey-Bass: New York. Christensen, C.M. 2004. Seeing what’s next. Harvard Business School Press: Boston, Massachussetts. Collis, D.J. 1991. The resource-based analysis of global competition: The case of the bearings industry. Strategic Management Journal, 12, 49-68. Davenport, T.H. and Beck, J.C. 2001. The attention economy. Harvard Business School Press: Boston, Massachusetts. Davenport, T.H. and Prusak, L. 1998. Working knowledge. Harvard Business School Press: Boston, Massachusetts. Deutsch, Y. and Ross, T.W. 2001. You are known by the directors you keep. Unpublished Manuscript. Dollinger, M.J., Golden, P.A. and Saxton, T. 1997. The effect of reputation on the decision to joint venture. Strategic Management Journal, 18(2), 127-140. Drucker, P.F. 1999. Management challenges for the 21st century. HarperBusiness: New York. Dutton, J.E. 1986. The processing of crisis and non-crisis strategic issues. The Journal of Management Studies, 23(5), 501-518. Dutton, J.E. and Jackson, S.E. 1987. Categorizing strategic issues: Links to organizational action. Academy of Management Review, 12(1), 76-90. Eisenhardt, K. M. and Martin, J.A. 2000. Dynamic capabilities: What are they? Strategeic Management Journal, 21, 1105-1121. Fombrun, C. & Shanley, M. 1990. What’s in a name? Reputation building and corporate strategy. Academy of Management Journal, 33(2), 233-258. Grant, R.M. 1991. The resource-based theory of competitive advantage: Implications for strategy formulation. California Management Review, 114-135. Grossman, G.M. and Helpman, E. 1999. Competing for endorsements. The American Economic Review, 89(3), 501-524. Hesselbein, F., Goldsmith, M. and Beckhard, R. Eds. 1997. The organization of the future. Jossey-Bass Publishers: San Francisco. Jackson, S.E. and Dutton, J.E. 1988. Discerning threats and opportunities. Administrative Science Quarterly, 33, 370-387. Kotter, J.P. 1995. The new rules. The Free Press: New York. Lounsbury, M. and Glynn, M.A. 2001. Cultural entrepreneurship: Stories, legitimacy, and the acquisition of resources. Strategic Management Journal, 22, 545-564. Morgan, G. 1997. Images of organization. Sage Publications, Inc.: California Ray, M. and Rinzler, A. Eds. 1993. The new paradigm in business. G.P. Putnam’s Sons: New York. Schumpeter, J. 1942. Capitalism, socialism and democracy. Harper: New York., 82-85. 84 Seely Brown, J. 1997. Seeing differently: Insights on innovation. Harvard Business School Press: Boston, Massachusetts. Sharma, S. 2000. Managerial interpretations and organizational context as predictors of corporate choice of environmental strategy. Academy of Management Journal, 43(4), 681-698. Sharma, S., Pablo, A.L. and Vredenburg, H. 1999. Corporate environmental responsiveness strategies: The importance of issue interpretation and organizational context. The Journal of Applied Behavioral Science, 35(1), 87-108. Stuart, T.E., Hoang, H. & Hybels, R.C. 1999. Interorganizational endorsements and the performance of entrepreneurial ventures. Administrative Science Quarterly, 44(2), 315-349. Weigelt, K. and Camerer, C. 1988. Reputation and corporate strategy: A review of recent theory and applications. Strategic Management Journal, 9(5), 443-454. 85 ASAC 2005 Toronto, Ontario Tamiko Hynes (student) Schulich School of Business York University INSIDE THE BLACK BOX: UNDERSTANDING DYNAMIC RELATIONS AS A SOURCE OF SUSTAINABLE COMPETITIVE ADVANTAGE1 Some resource based view proponents suggest people are a source of competitive advantage that cannot be easily managed. Others in strategy and HR propose managing people more strategically. Through micro analysis, a typology of dynamic relations helps clarify this tension by opening the invisible ‘black box’ contributing to competitive advantage. “No person is an island.” (Donne, 1975) The importance of people to organizational success has gained prominence in various fields in the past two decades, particularly in the field of strategic management. The resource-based view (RBV) proposes people-based resources are key sources of competitive advantage yet concludes their social-complexity prevents understanding their link to competitive advantage (Barney, 1996). Other approaches in strategy and strategic human resource management (SHRM) match people to strategy but do not focus on relational dynamics among people (e.g. Eisenhardt & Martin, 2000; Lepak & Snell, 2002). Although a micro analysis approach would enable a closer look and deeper understanding of social complexity and the role of people and relations as competitive resources, such an approach has not been used. Targeting this gap, this paper integrates strategy and micro analysis approaches to explore interpersonal connecting activities and reveal how people resources are linked to competitive advantage. It argues that interactions at work are understandable and that some of these, not just people, are invisible assets fundamental to sustainable competitive advantage. Building on existing literatures in strategic management, social capital, strategic human resource management and organizational behaviour, the paper develops an integrative perspective of employees as critical organizational resources. To do this, this paper conceptualizes dynamic relations between employees that develop in daily work. Dynamic relations are defined as dyadic interactions (with internal and external organizational members) that have organizational benefit and occur among proactive (versus passive) employees, yet are largely unnoticed in organizations. As opposed to more static, already-developed networks, the concept of dynamic relations reflects employees’ active initiation of and engagement in relations that occur largely unnoticed in organizations. 1 The author thanks Ellen Auster for her helpful comments on drafts of this paper. 86 Theoretical Background Strategic Views of People and Social Complexity Strategy recognizes the role of employees in organizational success. The RBV focuses on idiosyncratic, difficult-to-copy resources (Wernerfelt, 1984) that if valuable, rare, inimitable and organizationally exploited can be sources of sustainable competitive advantage which cannot be competitively eliminated (Barney, 1991; 1996). Resources are valuable if they reduce costs or raise revenues; rare if not commonly found and fewer firms possess them than are necessary for perfect competition and inimitable if they or their benefits cannot be copied. Inimitability stems from causal ambiguity or difficulty identifying invisible or taken-for-granted resources and their link to competitive advantage and social complexity or difficulty shaping such resources (Barney, 1996) due to various intricate social connections. Thus the RBV argues socially complex resources, including interpersonal relationships such as unique employee-client relationships, may be organizationally invisible and incredibly hard for any organization to copy. Because how people and competitive advantage connect and the means to force these emergent, path-dependent connections is unclear, this view necessarily assumes the preexistence of such invisible resources. Despite discussions about its deficiencies (see Barney, 2001; Priem & Butler, 2001a & 2001b for tautological debate), the RBV is influential in strategy and other fields such as Strategic Human Resource Management (SHRM). SHRM argues people are both discretionary and a stock of capital to be motivated individually and managed collectively (e.g. knowledge base) for sustainable advantage (see Wright et al., 2001 for a review). Some SHRM and strategy approaches tie strategy and people, suggesting the ability to fit HR practices to groups of varying importance (Lepak & Snell, 1999; 2002) or configure human resources to fit strategy (Eisenhardt & Martin, 2000; Galunic & Rodan, 1998; Teece et al., 1997) gives strategic advantage. Considering employees collectively or as configurable groups enables managing observable aspects of people resources to rationally fit strategy yet does not consider the relevance of hidden activities such as securing a colleague critical for a project. Individual employee management practices such as work design (Wright et al., 2001) may miss how employees interact with others to shape their jobs. In other words, managing people as solitary assets or bundles of resources may overlook hidden interpersonal connecting behaviour and potentially destroy relational processes. (Notably, a simultaneous individual and collective focus does not coincide with the RBV’s emphasis on relational complexity and competitive advantage). Existing Relationship-Oriented Research Relevant relationship literature considers relationships as a “whole” or as having a productive function. Social capital literature focuses on relationship networks as valuable resources (Nahapiet & Ghoshal, 1998) while network theory emphasizes the structure (see Burt, 1992) of relationship networks. As with the RBV, these approaches expose the value of 87 relationships and focus on formed relationships versus relational dynamics. In a recent work, Lengnick-Hall and Lengnick-Hall (2003) suggest relationships are important and focus on conventional or productive relationships, such as those formed during training rather than less typical relational activities lacking mutual goals or seemingly work-unrelated. Their relationship-typology does not emphasize taken-for-granted interactions that are not visibly tied to performance or innovation but that are perhaps more unique and competitively relevant assets. These approaches – strategy, SHRM and social capital - clearly recognize people and relationships as resources but suggest resources cannot be decoded, migrate people according to strategy and potentially separate hidden resources, or consider relationships statically. A micro analysis approach revealing behaviour at a lower level of analysis, i.e. individual and interactional, uncovers people-based, invisible, relational activities that create competitive advantage. Converging strategy and organizational behaviour, this paper builds on current approaches and argues that focusing on dynamic relations helps us understand and perhaps avoid separating invisibly connected resources – important sources of competitive advantage. Dynamic Relations Although interpersonal relations are said to contribute to competitive advantage there is a lack of knowledge on how relations develop - perhaps because research often assumes work relationships are instrumental, pursued strategically and identifiable. The paper now explores dynamic relations invisible at a macro level to reveal their role in creating competitive advantage. Types of Dynamic Relations “…our relationships with individuals and groups constitute the environment in which we live our professional lives” (Gersick et al., 2000: 1026). Dynamic relations are aptly described via dimensions of people and work relevance (Figure 1). A typology may conceal the relations’ active nature but offers a tidy map of these relations. The first dimension is individual focus, i.e. whether a relational activity is intended to benefit “self” or “other(s)”. The second is work focus, i.e. whether an activity is “job” or “work environment” related. These generate four types of dynamic relations: job-self (“creating”), job-other (“collecting”), environment-self (“connecting”) and environment-other (“caregiving”). “Creating” dynamic relations are self-benefiting and job specific while “collecting” dynamic relations are other-focused and job specific. “Connecting” dynamic relations focus on self and the work environment while “caregiving” relations are environment and other-focused. The paper now considers each with empirical examples. It is important to note that while relational work is necessary to develop and maintain relationships (Jacques, 1992), this concept contrasts with dynamic relations which are not necessarily focused on relationship-building. Creating. Through job-self or “creating” activities, employees initiate interactions with colleagues or outside members to create or define their role or do their job. Such interactions help 88 individuals meet needs for control over their jobs and define their work roles. These relational activities appear similar to role making activities (Graen & Scandura, 1987) and role innovation (Van Maanen & Schein, 1979) however the former reflects a more formal model of how individuals create their roles while the latter is mainly limited to reactive versus proactive behaviour. “Creating” interactions are more transparent to theorists and organizations because they are not readily associated with creating a role, self image or doing a job yet through work-shaping efforts, some employees self-define their roles (Wrzesniewski & Dutton, 2001). There are several empirical illustrations. Gersick et al. (2000) reveal collegial work where individuals initiated relations with colleagues regarding tasks or projects, producing valuable work. Here, the initiator actively contributed to and, less obviously, gained satisfaction from the collegiality of the interaction. Other types of collegial interaction might involve asking a colleague with whom the individual does not have a relationship for advice on a project. Fletcher’s (1998) qualitative study of engineers in a hi-tech firm surfaces additional examples of relational activity. Through relational theory, Fletcher (1998) describes how engineers respectfully asked for help, shared information and handled upset employees in ways allowing them to do their jobs and help career progression. Or, engineers empathically coached coworkers thereby sharing in the learning experience and benefiting from having to explain only once (Fletcher, 1998). These illustrations that show how employees create and manage their jobs are not typically associated with job functions and consequently organizational effectiveness. Although they might remain largely unnoticed at an organizational level and even to initiating employees themselves, such interactions are intertwined with these employees’ job activities. Figure 1 Dimensions and Types of Dynamic Relations Individual Focus Work Focus Self Beyond Self (Other) Job Creating Collecting Beyond Job (Work Environment) Connecting Caregiving Collecting. Through “collecting” activities, some employees selflessly draw in others to accomplish a job or task and thus build and sustain bridges to support work. Illustrations reveal employees collecting input from external and internal members. Benner et al. (1996) and Jacques (1993) describe how nurses interacted with patients’ families to gather and communicate nonquantifiable and nonmedical information that improved patient care. Fletcher (1998) describes how certain individuals secured input from employees critical to but not visibly connected to a project by expressing thanks. Considering this integral to an effective job, these individuals had less regard for others who took less care, implying that only some engage in collecting activities. The study also uncovers selflessly attending to a project versus potential career gains. Having exposed a problem to her manager, an employee allowed this manager to 89 take a visible role to resolve a problem and keep the project on track. Common in these work-critical if not work-visible relational activities is the "indirect nature and apparent invisibility of …activities assumed to be characteristic of their effectiveness" (Fletcher, 1998: 170). Some collecting examples seem work-unrelated yet smooth interpersonal conflict, supporting coworkers and enabling continued task-specific input. Star and Strauss describe how computer technicians’ “articulation work” that gets things back “on track” is invisible to rational work models (1999: 10). What might be perceived as idle chit chat is actually smoothing a conflict, creating a relaxed, trusting relationship or screening unnecessary interruptions - which support other employees and help do a job. Similarly, Fletcher’s (1998) illustrations of smoothing communication are relation-repairing, other-supportive and job-focused activities in which employees tracked down coworkers they had offended or made an effort to talk to someone with whom they had disagreed. Other collecting activities contributing to long-term work outcomes involve teaching coworkers in a way that does not harm their egos (Fletcher, 1998) and supporting their skills development (Gersick et al., 2000). Common in these illustrations are relational activities accomplishing a job while supporting or benefiting important other(s). They reveal that some employees’ active, job-oriented or skill-specific relational activities which might not be visibly instrumental to work do facilitate work deliverables and provide unquantifiable but definite organizational benefits. These might be overlooked by employees themselves as they disregard social activities as work, even though work would not get done without this interaction (Nardi & Engestrom 1999). Connecting. The two remaining types of dynamic relations focus on the work environment or climate. Through “connecting” activities, employees relate to other(s) and create an emotionally supportive, socially engaging work atmosphere; thus “connecting” relations simultaneously benefit the initiating employee and target the work environment. People desire connection with others (Baumeister & Leary, 1995) and need for social interaction can motivate joining the workforce (Feldman & Doerpinghaus, 1992). Baumeister and Leary (1995) suggest social interaction with “a long-term intimate” offers a “sense of belonging that would not be available in interactions with strangers or new acquaintances” (500). This focuses on the need for frequent or regular social contact, stability in and the emotional element of interactions. Gersick et al. show how employees felt emotional support from relational activities that were often “casual and spontaneous”, such as “having light-hearted fun” with people at and outside work (2000: 1037). Informal work arrangements (Bishop, 1999); informal, undocumented interaction such as humour and storytelling (Nardi & Engestrom, 1999) and talking about a new haircut and sharing personal problems (Tschan et al., 2004) demonstrate a need to connect to others. These self-benefiting activities create a supportive social work environment that supports organizational functioning through improved job satisfaction, performance and well being (Tschan et al., 2004). Employees also change their “relational boundaries” or interact with others to compose their social work environment (Wrzesniewski and Dutton, 2001: 180). Managing interaction with others, the employee supports his or her own work identity that fits with a desired self image (Wrzesniewski & Dutton, 2001). By proactively interacting with patients, showing visitors around, brightening another person’s day and interacting with nurses more often, hospital cleaners benefited from social interaction while enhancing their work identity (Dutton et al., 2003) and improving the work climate for coworkers and patients. 90 In sum, connecting interactions are fluid and their work role is often overlooked. Yet they are fundamental for meeting a need for social interaction and creating a desired work environment. Such connecting relations can lead to “safe harbour” relationships that further promote and facilitate relational work (Kahn, 1998). This is distinguished from the earlier-described research which views relationships as a goal rather than an extra benefit of interaction. Caregiving. Finally, “caregiving” also has a work environment or climate focus but here, employees’ activities care for other(s). One caregiving approach involves helping others feel appreciated, heard, and understood by verbally and nonverbally affirming and empathically listening to another (Fletcher, 1998), creating a supportive atmosphere. Fletcher’s study reveals caregiving activities including encouraging comments, maintaining eye contact and genuinely listening to a senior colleague describe “the good old days” (1998: 173). This supportive behaviour is likely to contribute to a “zest for interaction and connection” (Miller 1986, c.f. Fletcher 1998) and the “background conditions in which group life can flourish” (907). Other caregiving activities such as using collaborative confrontational language (“what I like about Dave’s idea is”) create a positive group atmosphere and a more cooperative (Fletcher, 1998: 174) and coworker-supportive environment. By publicly supporting others, employees’ caregiving activities promote team spirit and embed outcomes in others and in social relations, making outcomes inseparable from people and relational interactions and ultimately improving organizational effectiveness and decision making. Kahn’s (1993) study of “organizational caregiving” also reveals examples of behaviours involving daily interactions that are work environment and other focused. Some caregiving behaviour Kahn identifies involves interactions that support another, such as asking for information that serves another’s emotional needs, showing “emotional presence” and speaking warmly and expressing affection (1993: 546). Thus, these specific caregiving activities are not specific to a task or job; rather they help create a trusting, consistent, supportive work climate that encourages more caregiving toward others. Insights and Implications “What is the glue that holds the organization together?” (Lancaster, 1994: B1) People matter to organizations. Socially complex relationships and connections are said to lead to superior performance but in indecipherable ways. Consideration of relationships as static and primarily career-instrumental concepts does not reveal the complex, unnoticed relational dynamics that are work-supporting and may or may not be relationship-building. Strategic human resource reconfigurations could unknowingly wipe out value-creating, cost-reducing dynamic relations and create competitive disadvantage versus advantage. Rationalizing and restructuring work functions can lessen opportunities for invisible, informal work interactions and create “worker widgets” (Bishop, 1999: 123) who are incapable of participating in relational activities. Building on organizational-level insights about strategic advantage, this micro approach uncovers dynamic relations - unnoticed when only traditional ties and relationships are considered - yet part of social complexity linked to competitive advantage. While employees use relationships as exchange mechanisms to transfer such things as knowledge, some may also engage in relational activities to mold their work and meet human needs. These dynamic relations create sustainable competitive advantage (Figure 2) because they are valuable, rare inimitable and exploited. 91 Figure 2 Inside the Black Box: Dynamic Relations as a Source of Competitive Advantage Individual: Self-Other Work: Creating Collecting Connecting Caregiving Coach others Task innovation Gather information Involve others in project Smoothing talk Storytelling, chatting, brighten person’s day Supportive communication SCA Valuable Rare Inimitable Exploited Dimensions Types Activities Competitive Advantage -----------------------------Dynamic Relations--------------------------------Through relational activities, proactive employees interact with others to shape their roles so they and others more effectively perform tasks (creating); draw in others to accomplish a job or task, building bridges to support work (collecting); fulfill their needs to connect with others by creating a supportive work atmosphere (connecting); and provide care and a supportive environment to others (caregiving). This behaviour directly or indirectly can avoid costs of poor customer service, project delays and employee burnout and instead grow revenue by attracting clients and increasing new-project capacity - creating unseen value for organizations. Dynamic relations are elemental to a proactive employee’s daily work as not all employees will necessarily be motivated to go to such lengths to support coworkers or clients, do an outstanding job or create personally meaningful work. Hence dynamic relations are not common among all employees and are rare. These relational activities occur unpredictably with outcomes embedded in unpredictable social relations, creating higher-level social complexity and extreme replication difficulty – they are inimitable. Finally, because they are unrecognizable as an invisible asset, certain organizations allow these relational resources to flourish (perhaps unknowingly) because their structure allows interaction and avoids relation-destroying practices. Exploited dynamic relations are a socially complex resource contributing to sustainable competitive advantage. Research calls for greater knowledge of sources of superior performance (Mahoney, 1995; Rouse & Daellenbach, 1999). This paper helps characterize and explain the role of dynamic relations in competitive advantage, leading to important research and practical implications. While calling for greater relationship orientation (Lengnick-Hall & Lengnick-Hall, 2003) targets obvious organizational benefits of relationships, such approaches focus on “visible” activities and miss the subtle complexities of daily work containing the richness of and key insights into social interaction, particularly interaction that is unique and hard to reproduce due to its invisibility. Instead, examining relational dynamics by focusing on interactional “fluidity” uncovers hidden organizational benefits. Qualitative, rich descriptions (Miles & Huberman, 1994) open the door to an inside-view of organizations and a grounded understanding of relational phenomena. Since factors in organizations provide advantage, understanding this advantage must be gained in organizations (Rouse & Daellenbach, 1999). Research conducted at dyadic levels reveals relations residing among employees. Further, considering dynamic relations as explanatory variables may resolve variance unexplained in studies. Using integrative approaches as in this 92 paper benefits from “detailed comparative data about organizational processes” and increases knowledge of sources of competitive advantage (Rouse & Daellenbach, 1999: 491). Important managerial and practical implications are also evident. Management practices must enable but avoid micromanaging these relational activities, recognizing that employees invisibly build interconnections leading to employee and organizational benefit. Managing employees as individuals or groups by offering employees increased autonomy can destroy dynamic relations and overlook the work and employee relevance of these relational activities. Likewise, shifting employees for strategic purpose may disrupt relational work and employees’ needed stable interaction. Finally, the paper has research limitations. The need to develop a deeper understanding of dynamic relations requires an approach of depth and thus the study utilizes primarily localized illustrations of dynamic relations. However, work contexts cannot be ignored in conceptualizing dynamic relations. A more comprehensive review of dynamic relations in organizations would result from examining multiple internal and external perspectives. Finally, individual factors and culture that may be relevant in conceptualizing people dynamics are not considered here. Concluding Remarks Existing strategy literature recognizes that employees are fundamental to organizational success and competitive advantage. Building on existing views, this paper has tried to delve into employees’ engagement of others in their daily work to understand how employees matter and reveal invisible activities that contribute to social complexity performance advantage. Through further integrative research we will continue to advance knowledge of socially complex resources and the illusive ‘black box’ that contributes to sustainable competitive advantage. References Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17(1), 99-120. Barney, J. B. (1996). Gaining and sustaining competitive advantage. Reading, MA: Addison-Wesley. Barney, J. B. (2001). Is the resource-based "view" a useful perspective for strategic management research? Yes. Academy of Management Review, 26(1), 41-56. Baumeister R.F. & Leary M.R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117(3), 497-529. Benner P., Tanner C.A. & Chesla C.A. (1996). Expertise in nursing practice. New York: Springer. Bishop L. (1999). Visible and invisible work: The emerging post-industrial employment relation. Computer Supported Cooperative Work, 8, 115-126. Burt, R. (1992). Structural Holes. Cambridge, MA: Harvard University Press. Donne, J. (1975). Devotions upon emergent occasions. Montreal, QC: McGill Queens University Press. Dutton J.E., Debebe G. & Wrzesniewski A. (2003). Interpersonal sensemaking and the meaning of work. Research in Organizational Behaviour, 25, 93-135. Eisenhardt K.M. & Martin J.A. (2000). Dynamic capabilities: What are they? Strategic Management Journal, 21, 1105-1121. 93 Feldman D. & Doerpinghaus H.I. (1992). Missing Persons no longer: Managing part-time workers in the 90's. Organizational Dynamics, 21(1), 59-73. Fletcher J.K. (1998). Relational practice: A feminist reconstruction of work. Journal of Management Inquiry, 7(2), 163-186. Galunic D.C. & Rodan S. (1998). Resource Recombinations in the firm: Knowledge structures and the potential for Schumpeterian innovation. Strategic Management Journal, 19, 1193-1201. Gersick C.J.G., Bartunek J.M. & Dutton J.E. (2000). Learning from academia: The importance of relationships in professional life. Academy of Management Journal, 43(6), 1026-1044. Graen G.B. & Scandura T.A. (1987). Toward a psychology of dyadic organizing. In Research in Organizational Behavior (9, 175-208). Jacques R. (1992). Re-presenting the knowledge worker: A post-structuralist analysis of the new employed professional. Unpublished doctoral dissertation, University of Mass. at Amherst. Jacques R. (1993). Untheorized dimensions of caring work: Caring as structural practice and caring as a way of seeing. Nursing Administration Quarterly, 17(2), 1-10. Kahn W.A. (1993). Caring for the caregivers: Patterns of organizational caregiving. Administrative Science Quarterly, 38, 539-563. Kahn, W. A. (1998). Relational systems at work. In Staw, B. and Cummings, L.L. (Eds.), Research in Organizational Behavior. Greenwich, CT. (vol. 20: 39-76) Lancaster H. (Dec. 5, 1995). A new social contract to benefit employer and employee. The Wall Street Journal, B1. Lengnick-Hall M.L. & Lengnick-Hall C.A. (2003). HR's role in building relationship networks. Academy of Management Executive, 17(4), 53-63. Lepak D.P. & Snell S.C. (1999). The human resource architecture: Toward a theory of human capital allocation and development. Academy of Management Review, 24(1), 31-43. Lepak D.P. & Snell S.C. (2002). Examining the human resource architecture: The relationships among human capital, employment and human resource configurations. Journal of Management, 28(4), 517-543. Mahoney, J. (1995). The management of resources and the resource of management. Journal of Business Research, 33, 91-101. Miles. M. B. & Huberman A.M. (1994). Qualitative Data Analysis: An expanded sourcebook (2nd). Thousand Oaks, CA: Sage Publications, Inc. Miller J.B. (1986). What do we mean by relationships?. Unpublished manuscript, Work in Progress #22, Stone Center, Wellesley College, Wellesley, MA. Nahapiet J. & Ghoshal S. (1998). Social capital, intellectual capital, and the organizational advantage. Academy of Management Review, 23(2), 242-266. Nardi B.A. & Engestrom Y. (1999). A web on the wind: The structure of invisible work. Computer Supported Cooperative Work, 8, 1-8. Priem R. & Butler J. (2001a). Is the resource-based view a useful perspective for strategic management research? Academy of Management Review, 26, 22-40. Priem R. & Butler J. (2001b). Tautology in the resource-based view and the implications of externally determined resource value: further comments. Academy of Management Review, 26, 57-66. Rouse M.J. & Daellenbach U.S. (1999). Rethinking research methods for resource-based perspective: Isolating sources of sustainable competitive advantage. Strategic Management Journal, 20, 487-494. Star S.L. & Strauss A. (1999). Layers of silence, arenas of voice: the ecology of visible and invisible work. Computer Supported Cooperative Work: The Journal of Collaborative Computing, 8, 9-30. Teece D.J., Pisano G. & Shuen A. (1997). Dynamic capabilities and strategic management. Strategic Management Journal, 18(7), 509-533. Tschan F., Semmer N.K. & Inversin L. (2004). Work related and "private" social interactions at work. Social Indicators Research, 67, 145-182. Van Maanen J. & Schein E.H. (1979). In Toward a theory of organizational socialization: 1. Research in Organizational Behavior (Research in Organizational Behavior, 209-264). Wernerfelt, B. (1984). A resourced based view of the firm. Strategic Management Journal, 5, 171-180. Wright P.M., Dunford B.B. & Snell S.A. (2001). Human resources and the resource based view of the firm. Journal of Management, 27, 701-721. Wrzesniewski A. & Dutton J.E. (2001). Crafting a job: revisioning employees as active crafters of their work. Academy of Management Review, 26(2), 179-201. 94 ASAC 2005 Toronto, Ontario Mouloud Khelif (student) HEC Montréal APPROCHES DE LA GOUVERNANCE ET PERFORMANCE : LE CAS DES ORGANISATIONS DU SECTEUR PUBLIC Nous présentons différentes caractéristiques du système de gouvernance d’une organisation, en dépassant la vision traditionnelle portée par la théorie de l’agence. L’exemple des organisations relevant du secteur public et des dimensions clés qui permettent de les distinguer des autres secteurs (privé, non lucratif) nous amène à approfondir la démarche en nous basant sur plusieurs théories pertinentes en management stratégique (coûts de transaction, théorie de l’agence, droits de propriété). Nous examinons enfin plusieurs voies de recherche sur la gouvernance avec des implications pour les praticiens. Introduction Le système de gouvernance et son lien à la performance de l’organisation est un sujet d’une grande importance pour les chercheurs en management stratégique, mais qui intéresse également et à juste titre les praticiens. Charreaux (1997) définit la gouvernance de l’organisation comme l’ensemble des mécanismes organisationnels et institutionnels ayant pour effet de délimiter les pouvoirs et d’influencer les décisions des dirigeants, autrement dit, qui « gouvernent » leur conduite et définissent leur espace discrétionnaire. Dans la littérature en stratégie, et dans l’esprit des praticiens, force est de constater que la vision dominante de la gouvernance des entreprises reste financière et actionnariale (Shleifer et Vishny, 1997; Daily et al., 2003), basée sur les théories contractuelles de la firme (Gomez, 1996) et leur corollaire, la théorie de l’agence (Eisenhardt, 1989; Jensen et Meckling, 1976). Globalement, le « système de gouvernance » fait référence au dispositif institutionnel et comportemental qui régit l’ensemble des relations entre les dirigeants de l’organisation au sens large et les parties concernées (stakeholders): actionnaires, propriétaires, créanciers, salariés, et plus généralement agents ou institutions intéressées aux activités de l’entreprise, et en premier lieu ceux qui détiennent des droits jugés « légitimes » sur celle-ci (Pérez, 2003). Pour une entreprise privée, il fait en général référence aux processus et mécanismes institutionnels et organisationnels qui délimitent le pouvoir et définissent la marge de manœuvre des dirigeants de manière à accroître la valeur de l’entreprise. Au centre du concept de système de gouvernance de l’organisation se trouve toujours une catégorie d’acteurs-clés : les dirigeants de l’organisation, que cette catégorie soit réduite à une personne, à une équipe dirigeante ou encore à un ensemble d’organes de suivi, d’évaluation et de contrôle (conseil d’administration, conseil de surveillance, État-actionnaire, etc…). Ces acteurs sont sensés faire preuve de leadership organisationnel tout en étant les véritables architectes de la stratégie de l’organisation (Andrews,1971). Une autre perspective peu abordée de la gouvernance repose sur l’approche des ressources et des compétences, selon laquelle l’organisation est un portefeuille de ressources et de compétences (Barney, 1991). La source de l’avantage concurrentiel de l’organisation réside alors dans la capacité dynamique de combiner, protéger et renouveler ces ressources et ces compétences (Teece et al., 1997). L’idée ici est que la gouvernance elle-même peut être appréhendée comme une capacité dynamique (dynamic capability), propre à l’organisation, et qui lui permet de combiner des ressources, d’arbitrer et de résoudre des demandes souvent conflictuelles des acteurs internes et externes. 95 Sous l’impulsion de la mondialisation et des vagues de privatisations successives, le modèle de l’État-providence a été remis en cause dans de nombreux pays développés et moins développés. Les entreprises du secteur public ont tout naturellement vécu une période de grand changement technologique et idéologique qui a affecté leurs méthodes de gestion, leur activité stratégique et leur gouvernance. Si l’on se place dans le contexte spécifique du secteur public, une perspective de la gouvernance qui semble assez répandue est que si en théorie ces organisations sont contrôlées par le public (le citoyen), en pratique le lien n’est pas direct. L’État, Les gouvernements et leurs employés, en tant que garants de l’intérêt collectif, sont sensés contrôler, voire guider les gestionnaires des entreprises du secteur public, et ainsi assurer l’atteinte des objectifs et la performance voulue. Nous commençons d’abord par examiner le lien de base entre gouvernance et performance organisationnelle selon différentes approches, pour ensuite nous intéresser plus particulièrement au cas des organisations relevant du secteur public, c'est-à-dire où l’État (et ses représentants) est l’acteur majeur, en tentant d’examiner ce que plusieurs théories peuvent nous apprendre. Une conclusion nous amène enfin à ouvrir plusieurs voies de recherche qui nous paraissent prometteuses. Système de Gouvernance et Performance Organisationnelle : Les Bases On peut distinguer au moins deux niveaux du système de gouvernance : le « contenu institutionnel » et les « comportements ». Le premier concerne la firme et le second les agents, personnes physiques dont le comportement est influencé par ce contenu institutionnel. Ce dernier comprend en fait des structures et des procédures. Les structures peuvent être variées : certaines sont propres à l’organisation (conseil d’administration, assemblée générale etc..), alors que d’autres sont externes et interviennent de manière ponctuelle (cabinets d’audit, agences de notation, organismes de régulation etc…). Les procédures sont également diverses et généralement explicitées dans des codes s’imposant aux acteurs concernés (Pérez, 2003). A l’image de tout système humain, les « bonnes pratiques » et à l’inverse les déviations jouent un grand rôle dans la mesure de l’efficacité et de la performance des systèmes de gouvernance. Gouvernance : Domination de la Théorie de l’Agence Comme le soulignent Daily et al. (2003), la théorie de l’agence domine à la fois dans la recherche en gouvernance et dans le monde des praticiens et des entreprises. Selon les mêmes auteurs, cela est probablement du à la simplicité de la théorie de l’agence (Eisenhardt, 1989; Jensen et Meckling, 1976) : les organisations sont réduites à deux catégories d’acteurs, le principal et l’agent. D’autre part, l’idée que les individus se comportent généralement selon leur propre intérêt est largement répandue depuis longtemps. Pourtant, le système de gouvernance d’une organisation peut aussi être vu comme une ressource, voire même une compétence clé qui contribue à générer un avantage concurrentiel durable, dans le sens de la théorie des ressources (Barney, 1991; Wernerfelt, 1984, 1995). Le conseil peut donc contribuer au système d’innovation de l’entreprise et avoir une conséquence positive en terme de création de valeur. Il est en effet utile de lier la gouvernance à l’innovation, puisque l’innovation, dans ses différentes phases (formulation, validation, diffusion) est un processus complexe qui implique plusieurs parties constituantes de l’organisation. Le Cas du Conseil d’Administration : Organe de Contrôle ou Ressource En effet si on s’intéresse au cas du conseil d’administration, c’est l’organe de gouvernance que la théorie de l’agence (Jensen et Meckling, 1976) considère comme un organe et un outil nécessaire pour réduire l’asymétrie informationnelle. Il permet de superviser et contrôler le comportement du management (agent) et en théorie, d’éviter les déviances. Le conseil est responsable de l’adoption de mécanismes de contrôle (évaluation, rémunération, composition, 96 comités, séparation entre CEO et Chairman etc…) de manière à s’assurer que les décisions des dirigeants (surtout du CEO pour une grande entreprise) soient en conformité avec les intérêts des propriétaires de l’organisation (Fama et Jensen, 1983). Pourtant, vu avec d’autres lunettes, le conseil d’administration peut avoir d’autres rôles que celui administratif et coercitif. Quant à eux, Kose et Senbet (1998), estiment que d’une manière générale, l’efficacité d’un conseil d’administration est déterminée par sa taille, sa composition et son indépendance. Ainsi, au titre de sa composition (nombre de membres externes et indépendants par rapports aux membres internes), de la qualité et de l’expertise de ses membres (en d’autres termes ce qu’ils représentent et ce qu’ils peuvent apporter à l’organisation) un conseil d’administration peut aussi être caractérisé comme une ressource de l’organisation (plutôt intangible). La théorie de l’agence paraît donc insuffisante pour tenir compte de l’impact de la composition, du comportement et du rôle des membres du conseil d’administration sur la création de valeur et la performance de l’organisation. Par exemple, la théorie standard de l’agence ne tient pas compte du fait que les individus peuvent aussi dans une certaine mesure exhiber de l’honnêteté et de la confiance (Hendry, 2002). Par ailleurs, si l’on suit le raisonnement de la théorie de l’agence, les membres externes du conseil (réputés indépendants), tout comme les managers, vont agir uniquement (et égoïstement) de manière à préserver leur capital ‘notoriété’ et ‘réputation’ (dont découle leur valeur marchande en tant qu’administrateur), c'est-à-dire en pratique dans le sens des intérêts des actionnaires. N’oublions pas qu’en réalité, ces administrateurs externes sont souvent directement choisis ou nommés par le dirigeant exécutif (CEO) de l’organisation, ce qui peut les amener à servir ses intérêts personnels au détriment de ceux des actionnaires, ou inversement, agir de manière honnête même si leur compétence peut être remise en cause. De plus, si on s’intéresse à l’évidence empirique de l’efficacité de la présence de membres externes par exemple, les résultats sont peu concluants (Shleifer et Vishny, 1997) voire même négatifs (Agrawal et Knoeber, 1996). Mais plus important encore, la théorie de l’agence n’informe pas sur le rôle du conseil comme ressource et stratège (Daily et al., 2003). Le conseil d’administration est situé à la frontière entre l’organisation et son environnement (économique, légal, institutionnel etc…). Il est l’une des parties de l’organisation qui joue le rôle de passerelle et de lien (linkage role) entre cet environnement et les préoccupations internes. Dans cet ordre d’idées, le conseil d’administration, notamment par la ressource « expérientielle », « relationnelle » et « réputationnelle » qu’il peut constituer, a la capacité d’identifier les changements qui se produisent dans l’industrie, le secteur, la société dans son ensemble, et plus particulièrement ceux qui vont avoir un impact majeur sur l’organisation. Il est sans doute le plus à même de déceler les risques encourus par l’organisation, identifier les occasions et même contribuer à développer, reconfigurer et renouveler les ressources, compétences et capacités dynamiques existantes, de manière à contribuer à la dynamique d’innovation et de création de valeur. En envisageant une coopération entre les membres internes du conseil, qui ont eux une position avantageuse en terme d’information et de compréhension de l’organisation à l’interne et ses membres externes, qui constituent à la fois une ressource stratégique (Barney, 1991), un moyen de sécuriser des ressources non disponibles à l’interne (Pfeffer et Salancik, 1978), et un levier stratégique pour accroître la légitimité de l’organisation (Dacin et al., 2002; DiMaggio, 1988; Suchman, 1995). En utilisant de multiples approches, on peut ainsi concevoir un organe de gouvernance qui contribue à l’innovation et à la création de valeur. Le sujet de la rémunération des dirigeants est également de grand intérêt quand on s’intéresse à la gouvernance et à la performance d’une organisation. En effet, pour qu’il y ait innovation dans une organisation et création de valeur, les managers doivent accepter un certain degré de risque. Or, si l’on se réfère à la théorie de l’agence, les dirigeants auront plus tendance à prendre des risques s’ils sont eux-mêmes actionnaires de l’entreprise (Miller et al., 2002). L’autre manière de lier la compensation des dirigeants à la prise de risques et à la performance à long terme de l’entreprise (synonyme de création de valeur essentiellement par innovation interne) est 97 l’incitation de type « long term pay » (Markman et al., 2001), c'est-à-dire l’utilisation assez répandue de contrats basés sur les résultats à plus ou moins long terme (Eisenhardt, 1989), qui combinent rémunération fixe (salaire) et variable (stock options et bonus). Un autre lien entre gouvernance de l’organisation et sa performance est le processus d’évaluation du CEO. Young et al. (2000) ont examiné les facteurs associés à l’adoption par le conseil d’administration d’un processus d’évaluation de la performance du CEO dans le domaine hospitalier. En se basant sur les apports de la théorie de l’agence (Fama et Jensen, 1983; Jensen et Meckling, 1976) et de la perspective institutionnelle (DiMaggio et Powell, 1983; Meyer et Rowan, 1977), ils en retirent que les initiatives nécessaires pour améliorer l’efficacité du système de gouvernance doivent prendre en compte l’environnement institutionnel spécifique au conseil d’administration. Le processus par lequel des ressources sont développées et utilisées est central à la dynamique de l’innovation et de la création de valeur dans les entreprises. Selon O’Sullivan (2000), les organisations génèrent de l’innovation et de la valeur grâce à des structures de gouvernance qui leur permettent de soutenir l’intégration organisationnelle et de contrôler « l’argent et le savoir ». Ces structures de gouvernance varient non seulement selon le type d’activités, mais aussi dans le temps, le processus d’innovation étant par essence, dynamique. On peut être tenté de relier cette idée au concept de ‘innovation games’ développé par Miller et Floricel (2002). Le ‘jeu d’innovation’ est un niveau inter-organisationnel (un peu comme le champ organisationnel de DiMaggio et Powell) caractérisé notamment par une logique dominante de création et d’appropriation de la valeur. Les modes de gouvernance adaptés à la création de valeur pourraient alors être contingents au jeu d’innovation auquel l’entreprise participe. Gouvernance : la Spécificité des Organisations Relevant du Secteur Public Au fur et à mesure que le modèle de l’État-providence est remis en cause dans de nombreux pays (Bernier et Burlone, 2000), les méthodes de gestion du secteur public semblent évoluer vers une recherche d’une plus grande efficience (Roy et Séguin, 2000) et tendent à se rapprocher de plus en plus à celles du secteur privé. Ainsi, le fonctionnement et la composition des conseils d’administrations des entreprises publiques attirent l’intérêt des chercheurs et praticiens (Bernier et Burlone, 2000), cela d’autant plus qu’aujourd’hui, les entreprises dont l’état est le propriétaire continuent de représenter une bonne part de l’économie des pays industrialisés et des pays en développement. Dans un contexte où les organisations appartenant à l’état, entreprises hybrides par excellence, doivent souvent réconcilier des impératifs commerciaux avec les objectifs socio-économiques plus ou moins clairs de leur propriétaire, l’étude des dispositifs de gouvernance trouve tout son intérêt (Bégin, 1998). D’autre part, si l’étude du cas particulier des entreprises publiques dont les modes de gouvernance ont connu une évolution sensible sous l’impulsion de nouveaux rapports entre l’État et le marché est un axe de recherche intéressant, il est assez peu traité (Hafsi et al.,1987; 1988). Les travaux de recherche sur la gouvernance des organisations publiques trouvent aussi leur source dans divers questionnements, selon les disciplines des sciences sociales invoquées et les orientations des chercheurs : distinction public/privé et son utilité dans la recherche en théorie des organisations (Perry et Rainey, 1988), droits de propriété et privatisations (Vining, Boardman,1992), performance comparée des secteurs publics et privés (Boardman, Vining, 1989; Heracleous, 2001), rôle et composition des conseils d’administration des entreprises publiques (Bernier, Burlone, 2000), évolution de la relation entre le gouvernement et la société d’état (Hafsi, 1988) etc… Cette vision est soutenue par des travaux basés essentiellement sur les théories économiques de la gouvernance d’entreprise, à savoir la théorie de l’agence, les droits de propriété et la séparation de la propriété et du contrôle (Alchian et Demetz, 1972; Fama et Jensen, 1983; Shleifer et Vishny, 1997). Ainsi, les tenants de cette approche affirment que les arguments pour 98 justifier l’existence d’entreprises contrôlées par l’état ne tiennent pas la route et que les entreprises publiques ne semblent pas mieux servir l’intérêt public que les entreprises privées : “ …with a few exceptions of activities where the argument for state ownership carries the day, such as police and prisons, the reality of state ownership is broadly inconsistent with this efficiency argument.” (Shleifer et Vishny, 1997: 767). L’autre aspect caractéristique de la gouvernance dans le cadre de la propriété étatique de l’organisation tient au fait que si ces entreprises sont en théorie contrôlées par le public (citoyens), en pratique le lien n’est pas direct (Laurin, 1998). En effet, si on suit la logique de la théorie de l’agence, il devrait y avoir des coûts d’agence (monitoring, detecting etc…). Mais comme l’explique Aharoni (1986), le problème c’est que ces coûts ne peuvent être clairement définis car le principal n’est pas directement identifiable. Les entreprises détenues par l’État sont donc ‘des agents sans principal’. Les gouvernements, agissant en tant que représentants des citoyens et garants de l’intérêt collectif, sont effectivement sensés contrôler, voire guider les gestionnaires des entreprises publiques, ce qui pose au moins deux problèmes d’agence, l’un au sens des mécanismes de contrôle interne et l’autre externe (Fama et Jensen, 1983; Dharwadkar et al., 2000). Premièrement, au niveau interne, les gouvernements poursuivent des objectifs parfois contradictoires, confus, dictés à la fois par des considérations de nature sociale, politique et économique (Laurin, 1998). Le contrôle effectif est souvent entre les mains de bureaucrates nommés et non élus dont l’objectif n’est pas forcement le bien collectif mais plutôt leur propre intérêt (Shleifer et Vishny, 1997). Deuxièmement, l’État agissant comme propriétaire entraîne de facto une protection et une isolation de l’entreprise vis à vis des interventions du marché (hostile takeovers, LBO, protection des droits des actionnaires minoritaires…) comme mécanisme de contrôle externe (Laurin, 1998). Enfin, les entreprises publiques oeuvrant souvent dans des situations de monopole « naturel » (énergie, eau, télécoms), la discipline que le marché et la concurrence peuvent imposer sur les gestionnaires ne se fait pas sentir de la même manière. Pourtant, contrairement à la perspective dominante, dans certains pays comme le Canada, la France ou Singapour, des entreprises publiques présentent des résultats, des profits (parfois une efficience) et une performance comparables voire supérieurs à des entreprises privées similaires. Dans son étude de cas sur Singapore Telecom, Heracleous (2001) propose une explication à cette « anomalie ». En adoptant une perspective de management stratégique basée sur le « strategic choice » (Child, 1972) et sur la distinction entre stratégies délibérées, réalisées et émergentes (Mintzberg, 1985), l’auteur dépasse le débat entre propriété privée et publique. Prenant le cas de l’entreprise publique de Télécoms de Singapour, il nous donne un exemple clair où propriété et contrôle de l’État peuvent être associés, sous certaines conditions (privatisation limitée, relative autonomie, politique claire de l’État, stratégie de la firme), à une performance supérieure : « …draws attention to the fact that performance is ultimately rooted in strategies that are successfully realized, whether intended or emergent. In this view, what matters to performance is primarily the strategies chosen and implemented, rather than who owns the organization per se. » (Heracleous, 2001 : 74). Si le contexte du cas et l’environnement de Singapour peuvent nous paraître trop spécifique, il n’en reste pas moins que l’idée est ici suggérée que les organisations étatiques de nombreux pays sont relativement inefficientes, non pas à cause de la nature de la propriété, mais plutôt à cause d’un déficit en gouvernance, résultant notamment de la confusion dans les objectifs et les politiques à mener de la part du propriétaire. Dimensions-Clés de la Gouvernance dans le Secteur Public En terme de gouvernance, dès que l’on s’intéresse plus particulièrement à des entreprises du secteur public, aux diverses structures hybrides ainsi qu’aux organismes à but non lucratif, on peut voir émerger au moins quatre facteurs importants, basées sur les théories de l’approche contractuelle (droits de propriété, agence et coûts de transaction) et l’approche institutionnelle 99 (recherche de légitimité, accès aux ressources) : la propriété, le contrôle que l’on peut aussi associer au degré de concurrence/régulation auquel l’organisation fait face dans son environnement, les objectifs de l’organisation, et le type de financement ou plus généralement la provenance des ressources. Pour illustrer notre propos, dans la figure 1 ci-dessous, nous avons choisi deux dimensions, le degré de régulation/contrôle et le type de propriété dans un continuum. On voit bien apparaître de nombreuses combinaisons possibles, menant à divers degrés d’hybridations. Figure 1 : Public Vs Privé – 2 Dimensions-clés Privé pur Entreprise privée régulée P r o p r i é t é Entreprise privée dans un environnement concurrentiel Contrôle/Régulation Contrôle/Régulation pur Concurrence pure Entreprise d’État comme monopole Société d’État commerciale dans un environnement concurrentiel Public pur Le tableau 1 ci-dessous reprend de manière synthétique trois dimensions énoncées pour différents types d’organisations. Ainsi, pour les organisations qui relèvent purement du secteur public, l’objectif est souvent multiple, voire divergent (Aharoni, 1986), l’État étant propriétaire et détenteur des ressources, alors que dans le cas une organisation hybride comme une entreprise commerciale détenue par l’État (par exemple la Société des Alcools du Québec), si l’État est l’unique « actionnaire », le financement et la provenance des ressources en général est assurée par le privé (marché), ce qui indique des objectifs organisationnels plus clairs (profit, rente versée à l’État, disponibilité du produit aux citoyens) mais qui peuvent tout de même être variables et fluctuer dans le temps. 100 Tableau 1 : Les 3 dimensions selon le type d’organisation PUBLIC PRIVÉ HYBRIDE NON LUCRATIF OBJECTIF Multiple, Variable, Divergent (Collectivité) Clair (actionnaires) Actionnaires (État) Variable Fluctuant Membres, Société PROPRIÉTÉ État Privée Peut être Mixte Privée RESSOURCES État Marché Mixte Membres, État L’Apport des Théories Institutionnelles de la Gouvernance au cas du Secteur Public Nous allons voir de manière succincte comment les perspectives des théories institutionnelles, droits de propriété, de la théorie des coûts de transaction et de l’agence peuvent être ajustées pour faire sens dans le contexte particulier des organisations du secteur public et rendre compte des enjeux qui les concernent. Théorie institutionnelle La dynamique qui prévaut dans le secteur public de nombreux pays (réduction des effectifs, recherche de manières d’augmenter la productivité, innovations organisationnelles, réingénierie de l’État) fait que les organisations du secteur subissent des pressions normatives et cognitives. En tant qu’acteurs du champ, elles doivent impérativement rentrer dans cette dynamique pour maintenir ou accroître leur légitimité vis-à-vis de l’État (politique) et de la société dans son ensemble (sociale), s’assurer d’obtenir les ressources et l’autonomie nécessaires, et finalement survivre. Il s’ensuit, selon les termes de DiMaggio et Powell (1983, 1991) des comportements isomorphiques, à la fois de nature normative et mimétique, puisque les organisations publiques qui ont une forte légitimité politique et sociale sont le modèle à suivre. Droits de propriété La théorie des “droits de propriété” (Coase, 1939; Alchian et Demsetz, 1972) affirme que ces derniers sont une limite à l’autonomie des acteurs. Ils sont aussi, subjectifs, exclusifs et cessibles (marché). Comment appliquer cela dans un secteur public où par définition les droits de propriété n’existent pas, les organisations faisant partie du bien collectif? Ainsi, les dirigeants des organisations du secteur public sont peu enclins à prendre des risques car ils ne peuvent ‘jouir’ de leurs résultats. Doit-on alors supposer que les dirigeants qui obtiennent de bons résultats le font de manière complètement désintéressée? On peut aussi tenter d’adapter la théorie au contexte du secteur public (et à but non lucratif aussi d’ailleurs) de la manière suivante. On peut considérer que les managers du secteur public ont ‘intérêt’ à créer de la valeur et à innover dans leur organisation car, par leurs réalisations et leurs initiatives, ils acquièrent de facto un droit sur la valorisation de ce résultat. Les notions de prestige, de notoriété et d’expérience acquise peuvent ici être invoquées 101 pour élargir la théorie. Ces ‘récompenses’ individuelles encouragent donc les gestionnaires publics à expérimenter, car ils vont pouvoir en disposer et même éventuellement les monnayer (augmentation de la mobilité) sur le marché du travail. Théorie de l’agence Le problème d’agence ((Eisenhardt, 1989; Jensen et Meckling, 1976) existe entre un (ou plusieurs) principal et un agent lorsqu’il y a une divergence d’intérêt entre les deux parties et qu’un contrat complet ne peut être écrit du fait de l’asymétrie de l’information. Le contrat entre propriétaire et managers donne ainsi une grande discrétion aux managers. La théorie de l’agence identifie trois coûts liés à la relation d’agence : les coûts de surveillance (Monitoring), d’obligation (Bonding) et les pertes résiduelles (Residual Losses). Dans le contexte de gestionnaires du secteur public, nous avons vu que principal ne peut être clairement identifié. Ces dernières années, la transparence accrue de la gestion des organismes d’État liée aux développements des technologies de l’information ont entraîné une baisse significative des coûts de surveillance. Par ailleurs, pour les agents du secteur public, on note une diminution notable des coûts d’obligation qu’ils supportent vis-à-vis de leur employeur. Enfin, la remise en cause systématique de l’emploi garanti pour les employés des nouvelles structures issues du secteur public a pour conséquence de réduire de manière significative les pertes résiduelles. Tableau 2 : Approches théoriques institutionnelles adaptée au contexte du secteur public PUBLIC PRIVÉ Droits de Propriété Valorisation des résultats, Appropriation du Prestige, Valeur marchande Propriété exclusive et cessible Coûts de Transaction Coûts de coopération Réduction des coûts (Objectif Politiciens) Autonomie et Impartition Rationalité limitée et Opportunisme Agence Réduction des coûts (surveillance, obligation et pertes résiduelles) Surveillance, Incitatifs, Protection légale Pression institutionnelle : Politique (État) et social (Société) Encouragement fort à expérimenter et innover Pression institutionnelle : Champ organisationnel Stratégie des acteurs contrainte par l’environnement Théorie Institutionnelle Coûts de transaction L’Économie des Coûts de Transaction reconnaît le marché et la firme comme deux formes alternatives essentielles de gouvernance de la production et des échanges (Williamson, 1979, 1996) 102 et les incitatifs du marché et le contrôle managérial comme les deux formes de contrat (Coase, 1937; Alchian and Demsetz, 1972). Comme nous l’avons vu par ailleurs dans la partie 2, cette approche est basée sur la transaction comme unité d’analyse, et sur les hypothèses de rationalité limitée des acteurs et d’opportunisme. La structure de gouvernance est sensée s’aligner en économisant les coûts de transaction. Dans le cadre du secteur public, les structures bureaucratiques sont de plus en plus remises en cause. Une priorité des politiciens ces dernières années a été la réduction des coûts par tous les moyens, la création d’agence plus petites et relativement autonomes, le recours à l’impartition et l’encouragement du partenariat public/privé ou public/organisme-à-but-non-lucratif, ce qui est en accord avec les exigences d’économies de coûts dictés par la théorie des coûts de transaction. Puisque les dirigeants du secteur public sont encouragés à faire preuve d’innovation organisationnelle, tous ces développements vont dans le sens d’une plus grande action managériale, ce qui augmente sensiblement leurs possibilités d’initiatives et leur marge de ‘discrétion’. Finalement, toutes ces approches adaptées ont en commun l’idée que pour les gestionnaires du secteur public une augmentation potentielle de leur marge de manœuvre, de leur espace décisionnel et une incitation à expérimenter et à contourner les « rigidités du système ». Discussion et Conclusion Comme nous l’avons vu, le concept de gouvernance rassemble plusieurs concepts, idées et réalités, selon les niveaux d’analyse, les époques, perspectives, le type d’organisation, les contextes et parfois les auteurs. Parler alors d’une seule « théorie » de la gouvernance qui tiendrait dans un paradigme intégrateur parait bien dérisoire. L’étude approfondie et la recherche sur les organisations au contexte spécifique comme les entreprises du secteur public ou les organismes à but non lucratif est d’un grand intérêt, car ces formes hybrides et complexes montrent les limites inhérentes aux modèles théoriques les plus répandus de la gouvernance et sont donc une occasion d’élargir, d’adapter et de tester de nouvelles approches. Au vu des divers mécanismes internes, enjeux externes, questionnements qui regroupent et définissent la notion de gouvernance, une approche multi-théorique et multi-niveaux parait être la plus appropriée. Il s’agit, en s’appuyant sur les théories de base invoquées en management stratégique (théories institutionnelles notamment) de développer un cadre de référence qui soit suffisamment général pour mériter le qualificatif de théorique, tout en étant suffisamment précis pour coller aux réalités et aux spécificités des organisations. A partir de là, on pourra s’aventurer à suggérer des propositions en terme de management stratégique, ce qui nous ramène ensuite à la possibilité de tester des hypothèses sur le terrain du système de gouvernance. La théorie des ressources et ses corollaires (Barney, 1991, 2001; Eisenhardt et Martin, 2000; Teece et al., 1997) permettent d’entrevoir la gouvernance comme une capacité dynamique source de compétences distinctive pour les organisations relevant de l’État. Mieux comprendre la nature et la dynamique de la gouvernance des entreprises d’État, surtout si elle sont en situation de relative autonomie (entreprises commerciales, régies), et comment cette gouvernance détermine leur performance devrait intéresser autant les chercheurs en management stratégique que les gestionnaires. Les ressources et capacités spécifiques de la firme mises en avant par l’approche des ressources (Barney, 1991) peuvent aussi être combinées avec les contraintes formelles et informelles (générées par un environnement institutionnel donné) auxquelles font face les acteurs qui prennent des décisions stratégiques (Oliver, 1997; Scott, 2001). Cela permet de donner un éclairage pertinent à la question du choix stratégique, particulièrement si l’on s’intéresse aux questions stratégique et de gouvernance dans les contextes autres que ceux des économies de pays développés. Le rôle du conseil d’administration, comme mécanisme représentatif de la structure de gouvernance semble un sujet d’étude important, dans la mesure où il est à véritablement à la frontière entre l’organisation et les forces de l’environnement, là ou le ‘phénomène’ devient 103 intéressant. A ce titre, la complexité de son fonctionnement, ses rôles multiples doivent être mis sous la lumière de divers approches théoriques, en allant au-delà des classiques de la théorie de l’agence, de la théorie des parties-prenantes ou de la dépendance des ressources. L’approche des ressources-compétences-capacités parait être prometteuse, surtout si elle est combinée avec d’autres perspectives comme la théorie institutionnelle en sociologie et notamment l’idée de légitimité comme un capital que les organisations peuvent gérer et développer. Une autre voie qui parait aussi prometteuse, que ce soit pour les chercheurs en management stratégique ou pour les praticiens est celle de la relation entre modes de gouvernance et stratégies d’innovation des organisations. Certains pourraient notamment explorer l’aspect contextuel de la gouvernance en fonction du système d’innovation mis en place par l’entreprise ou imposé par l’environnement institutionnel, et en prenant un cadre dynamique, intégrer plusieurs facteurs comme le temps (facteur essentiel dans le processus d’innovation), le degré de dynamisme de l’environnement (modéré vs high velocity) ou encore la notion récente de ‘jeux d’innovations’ (Miller et Floricel, 2002) qui tente de mieux prendre en compte les enjeux de la création de valeur. Références Bibliographiques Agrawal A., Knoeber C., (1996), "Firm Performance and Mechanisms to Control Agency Problems Between Managers and Shareholders." Journal of Financial and Quantitative Analysis, 31, 377-397. Aharoni, Y., (1986), "The Evolution and Management of State Owned Enterprises", Ballinger, Cambridge, Mass. Alchian A., Demetz H. (1972), "Production, information costs and economic organization". American Economic Review, 62. Andrews, K. A., (1971/1987), The Concept of Corporate Strategy. 3rd edition, Homewood, IL: Irwin. Barney J.,(1991).Firm Resources and Sustained Competitive Advantage. Journal of Management, 17, 1, 99-120. Barney, J. (2001). "Is the Resource Based “View” a Useful Perspective for Strategic Management Research? YES. Academy of Management Review, 26(1) pp. 41-56. Bégin, L., (1998), "À Différentes Organisations, Différentes Conditions de Gouvernance". Gestion, 23, 3, 23-29. Bernier, L., Burlone N., (2000), “Qu'est devenu l'intérêt public? Les conseils d'administration des sociétés d'État québécoises à l'époque des privatisations”. Gestion, vol. 24, 4, 57-62. Boardman A. E., Vining A. R., (1989), Ownership and performance in competitive environments : A comparison of the performance of private, mixed and state-owned enterprises. Journal of Law and Economics, vol. 32, 1-33. Child, J. (1972), "Organizational Structure, Environment and Performance: The Role of Strategic Choice". Sociology, 6, 1, 1-22. Coase R., (1939), The nature of the firm: Meaning. Journal of Law, Economics and Organization, 7, 129-132. Charreaux G., (1997). “Vers une théorie du gouvernement des entreprises”, in G. Charreaux, Le Gouvernement des entreprises. Paris: Economica, 421-469. Dacin, M.T., Goodstein, J., Scott, W.R., (2002), "Institutional Theory and Institutional Change: Introduction to the Special Research Forum". Academy of Management Journal, 45, 1, 45-57. Daily C. M., Dalton, D. R., Cannella, A. A. Jr., (2003), "Corporate Governance: Decades of Dialogue and Data". Academy of Management Review, 28, 3, 371-382. Dharwadkar R., George G., Brandes P., (2000), Privatization in Emerging Economies : An Agency Theory Perspective. Academy of Management Review, vol. 25, 3, 650-669. 104 DiMaggio P., Powell W., (1983), The Iron Cage revisited: Institutional Isomorphism and Collective Rationality in Organizational Field. American Sociological Review, 48, april, 147-160. DiMaggio P., (1988), “Interest and Agency in Institutional Theory”, in L. Zucker, Institutional Patterns and Organizations: Culture and Environments. Cambridge, MA: Ballinger, 3-21. Eisenhardt, K. M., (1989), "Agency Theory : An Assessment and Review". Academy of Management Review, vol. 14, 1, 57-74. Eisenhardt K. M., Martin, J. A., (2000). "Dynamic Capabilities: What Are They? ". Strategic Management Journal, 21, 10/11, 1105-1121. Fama E., Jensen M., 1983), "Separation of Ownership and Control". Journal of Law and Economics, vol. 26, 301-325. Gomez P. 1996), Le Gouvernement d’Entreprise : Modèles Économiques de l’Entreprise et Pratiques de Gestion. Paris : InterÉditions/Masson Hafsi T., Kiggindu M. N., Jorgensen J. J., (1987), "Strategic Apex Configurations in State-Owned Enterprises". Academy of Management Review, vol. 12, 4, 714-730. Hafsi, T., Koenig, C., (1988), "The State-SOE Relationship: Some Patterns". The Journal of Management Studies, vol. 25, 3, 235-250. Hendry, J., (2002), "The Principal’s Other Problems: Honest Incompetence and the Specification of Objectives". Academy of Management Review, 27, 1, 98-113. Heracleous L., (2001), "State Ownership, Privatization and Performance in Singapore: An Exploratory Study from a Strategic Management Perspective". Asia Pacific Journal of Management, vol. 18, 69-81. Jensen M., Meckling W., (1976), "Theory of the firm: managerial behavior, agency costs and ownership structure". Journal of Financial Economics, 3, 301-360. Kose, J., Senbet, L.W., (1998), "Corporate Governance and Board Effectiveness". Journal of Banking and Finance, 22, 4; 371-403. Laurin C., (1998), "La Réaction des Marchés Boursiers au Changement dans les mécanismes de régie d’entreprise : le cas des privatisations". Gestion, vol. 23, 3, 48-54. Markman, G.D., Balkin, D.B., Schjoedt, L., (2001), "Governing the Innovation Process in Entrepreneurial Firms". Journal of High Technology Management Research, 12, 273-293. Meyer J., Rowan B., (1977), "Institutionalized Organizations: Formal Structure as Myth and Ceremony". American Journal of Sociology, 83, 340-363. Miller, J. S., Wiseman, R.M.,Gomez-Mejia, L.R., (2002), "The Fit between CEO Compensation Design and Firm Risk". Academy of Management Journal, 45, 4, 745. Miller, R., Floricel S., (2002), "Value Creation and Innovation Games. Managing R&D for Business Growth". Working Paper. Ecole Polytechnique de Montreal. Mintzberg, H., Waters, J., (1985), "Of Strategies, Deliberate and Emergent". Strategic Management Journal, 6, 257-272. Oliver, C. (1997), "Sustainable Competitive Advantage: Combining Institutional and Resource-Based Views". Strategic Management Journal, 18, 697-713. Perez, Roland, La Gouvernance de l’Entreprise, Paris: Collection Repères – La découverte 2003. Peteraf, M. A. (1993), "The Cornerstones of Competitive Advantage: A Resource-Based View". Strategic Management Journal, 14, 179-191. Pfeffer, J., Salancik, G.R., (1978), The External Control of Organizations: A Resource-Dependence Perspective. New York: Harper & Row. Scott R., (2001), Institutions and Organizations – Second Edition, Thousand Oaks, CA: Sage Publications. Shleifer A., Vishny R. W., (1997). "A Survey of Corporate Governance". The Journal of Finance, 52, 2, 737-783. Suchman, M.C., (1995), "Managing Legitimacy: Strategic and Institutional Approaches". Academy of Management Review, 20, 3, 571-611 105 Teece D.J., Pisano G., & Shuen A., (1997). "Dynamic Capabilities and Strategic Management". Strategic Management Journal, 18,7,509-533. Vining A. R., Boardman A. E., (1992), "Ownership versus Competition : Efficiency in Public Enterprise". Public Choice, 73, 205-239. Wernerfelt B. (1984), "A Resource-Based View of the Firm". Strategic Management Journal, 5, 2, 171-180. Wernerfelt, B., (1995), "The Resource-based View of the Firm: Ten Years After". Strategic Management Journal, 16, 3, 171-174. Williamson O. E., (1979), "Transaction-cost Economics: the Governance of contractual relations". Journal of Law and Economics, 22, 233-261 Williamson O. E., (1996), The Mechanisms of Governance. Oxford University Press. 106 ASAC 2005 Toronto, Ontario A.C. Martinet Professeur des Universités [email protected] M.A. Payaud Maître de Conférences [email protected] EURISTIK (IAE – Univ. Lyon3) 15 quai Claude Bernard 69007 Lyon FRANCE ENTREPRISE DURABLE ET RESPONSABLE : MODÉLISATION DE LA STRATÉGIE ET PRINCIPES DE GOUVERNANCE La communication s’inscrit dans la nécessaire réarticulation de 3 champs de recherche -le gouvernement d'entreprise, la responsabilité sociale et le management stratégique- dès lors que l'on s'intéresse au caractère durable du développement de l'entreprise ("sustainable development"). Dans une telle optique, la politique générale cherche en permanence les voies et moyens susceptibles d'éviter les risques de ruptures tant financières -à l'encontre des trop nombreuses "aventures" récentes- qu’économiques, sociales ou écologiques. Dans des environnements turbulents, la responsabilisation des dirigeants et des managers constitue une condition à la construction d’une entreprise durable. Middle managers, compétence, plan, apprentissage, savoirs, relations, variation. La construction d'une entreprise durable suppose la mise en cohérence de stratégies locales fines, contextualisées, congruentes avec les territoires dans lesquelles elles s'inscrivent d'une part, d'une stratégie globale ("corporate strategy") soucieuse de la création de valeurs partenariales ("stakeholders theory") et d'une gouvernance qui incite les dirigeants à celle-ci, d'autre part. La communication présente dans un premier temps un modèle synthétique et heuristique de la formation des stratégies élaboré à partir de recherches cliniques dans de grandes entreprises de services déployées territorialement (EDF, GDF, Adecco…) qui sont particulièrement illustratives de la double nécessité de stratégies globales très intégrées et "mondialisées" et de stratégies différenciées, ancrées durablement dans les territoires avec lesquels elles doivent coopérer économiquement, socialement et écologiquement. Ce modèle voit l'entreprise comme entité écologique guidée intentionnellement par le noyau stratégique et enrichi par l'organisation, il s'inscrit dans une "systémique ago-antagoniste" où il s'agit de réguler des processus en tension constante : plan/apprentissage, délibéré/émergent, exploration/exploitation, innovation/continuité… Il identifie les dispositifs, les variables d'action et les conditions pour que se concrétise un double mouvement : « l'écologisation » de stratégies plus congruentes avec leur milieu et la « stratégisation » de l'organisation par le rôle de stratège conféré aux « middle managers ». Dans un second temps, la communication montre comment cette modélisation guide la recherche d'une responsabilisation accrue des managers : les cadres intermédiaires d'une part, appelés à concrétiser et à déployer territorialement les impératifs du développement durable, c'est-à-dire co-construire des connaissances et des relations avec les parties prenantes locales ; les hauts dirigeants d'autre part, incités, rémunérés et sanctionnés selon la contribution effective de leurs actions à la triple performance économique, sociale et écologique de 107 l'entreprise ainsi qu'à l'évitement des risques et des dangers selon les deux principes de précaution et de responsabilité. La communication débouche, en conclusion, sur la réforme des cadres institutionnels et des pratiques professionnelles que peut inspirer la recherche ainsi renouvelée sur la gouvernance : accroissement du rôle des différentes parties prenantes, renforcement des règles de prudence, exigences à l'égard des analystes financiers, des agences de notation et des cabinets d'audit, réforme des règles comptables, refonte du fonctionnement des conseils d'administration… dont on a vu, a contrario, au cours de la dernière décennie, l'influence sur l'horizon, la teneur et la robustesse des stratégies de très nombreux groupes de par le monde. 1. Une modélisation de la formation des stratégies de la grande entreprise D’une manière générale, la relation complexe des éléments constitutifs de la stratégie présuppose une réflexion a priori, anticipée, afin que l’entreprise soit capable d’une action dans la durée et non pas d’une soumission aux évènements. Dans le cadre de l’entreprise de services de réseau, un autre enjeu apparaît, celui d’intégrer les émergences locales. Une description sommaire – un siège administratif et des unités dispersées sur le territoire— appellent une réflexion, illustrée dans le schéma page 4 : ce type d’entreprise nécessite d’articuler, de faire dialoguer des systèmes ago-antagonistes (Bernard-Weil, 1988), fondamentaux pour faire face aux injonctions paradoxales liées non seulement à son activité (les services) mais aussi à la répartition de ses infrastructures. Si une stratégie et une ligne directrice sont identifiées (planning) pour une durabilité de l’action et du sens de l’action, il nécessite également que les subtilités et singularités des territoires soient prises en considération ainsi que leurs dimensions sociologiques, écologiques et économiques (learning). Effectivement, l’intensité d’investissement et le système d’offre de l’entreprise de services de réseau justifient une stratégie délibérée planifiée. En effet, pour s’implanter, l’entreprise de services de réseau doit pouvoir disposer d’un assez grand nombre d’unités qui se traduit par des coûts fixes élevés. Le retour sur investissement ne pouvant pas se parier sur une absorption aléatoire des évènements, le chemin à prendre, la destination à atteindre et les moyens pour y parvenir doivent être explicités et diffusées à l’ensemble de l’organisation. Egalement, le consommateur ou le client souhaite pouvoir (re)trouver le service d’une enseigne identique quel que soit l’endroit où il se manifeste. Dès lors, l’entreprise homogénéise l’offre, la manière de la construire, de la délivrer, autrement dit son « système d’offre » et ce, afin qu’une qualité de service soit irréprochable sur l’ensemble du territoire. Cette homogénéisation ou « organisation systématique et cohérente » (Eiglier et Langeard, 1987) est également le fruit d’une délibération, laquelle constitue la stratégie globale et les compétences globales. En même temps, l’entreprise de services de réseau, de par sa dispersion géographique est confrontée à des marchés multiples et mouvants. La diversité, la variété et la dynamique de la demande requièrent de la part de l’entreprise de l’ambition, une capacité de perception, d’adaptation, de cohésion et de cohérence. Au-delà et en plus des lignes directrices planifiées et délibérées, les unités territoriales doivent être capables de capter, d’absorber les émergences de leur site, tout en les fondant aux exigences requises par le système d’offre. L’adaptation des pratiques de l’entreprise aux émergences nées localement peut déboucher sur des nouvelles offres, des nouvelles pratiques, lesquelles sont susceptibles d’être étendues soit à d’autres territoires, soit à l’ensemble de l’entreprise, soit l’un après l’autre suivant un processus évolutionniste. Ainsi, il apparaît des logiques le plus souvent opposées mais que nous pensons compatibles et plus encore, qu’il est nécessaire d’articuler et de faire coexister dès lors que l’entreprise se veut robuste, responsable et durable. L’ago-antagonisme planning/learning sous-entend à la fois un processus descendant et un processus ascendant. Le premier signifie qu’une équipe managériale inspirée des expériences précédentes, d’une connaissance sectorielle, structurelle et conjoncturelle délibère de nouvelles stratégies (suivant une logique téléo) lesquelles favorisent une absorption de l’incertitude macro (Martinet et Payaud, 2004). 108 Ces nouvelles stratégies globales sont alors adaptées localement par les middle managers tenus responsables de l’adéquation de leur unité au système global. Le second processus, ascendant, traduit le fait que les équipes déployées sur le territoire et menées par les middle managers en adaptant localement les stratégies globales absorbent l’incertitude micro lors de négociation et de co-création (« single-loop » d’Argyris et Schön, 1978), elles participent aux stratégies globales lorsque leurs contributions locales sont portées à un niveau organisationnel (« double loop » d’Argyris et Schön, 1978). Cet apprentissage suit les phases du processus évolutionniste : variation, sélection, rétention (Nelson et Winter, 1982). Les équipes en développant un système de relations et de savoirs avec les parties prenantes locales se mettent en situation de faire émerger des idées. Une zone d’inspiration, désignée zone de « variation » par les évolutionnistes, où se construisent problématique et ébauche de réponses. Se créent ce que Nonaka et alii (2000) appellent des « Ba » : des lieux partagés de relations émergentes, transactionnelles et interactionnelles où peuvent se déployer des processus d’apprentissage situé. A la connaissance des stratégies, compétences et offres globales, s’ajoute une connaissance située et contextualisée du marché et des acteurs qui le composent. La connaissance construite et produite est ensuite soumise et évaluée pour être retenue ou rejetée dans les phases respectives de sélection et de rétention. Le système ago-antagoniste planning/learning fait résonner deux autres antagonismes, à savoir les logiques téléo/écolo et la dialectique continuité/changement. Telle que nous l’avons décrite, l’entreprise de services de réseau est vue comme un système d’entités interdépendantes qui créent, partagent et appliquent des connaissances (écolo) pour poursuivre les objectifs stratégiques et économiques (téléo) (Lovas et Ghoshal, 2000). Une telle conception permet d’abandonner la distinction entre la formulation de la stratégie et sa mise en œuvre, pour préférer une formation de la stratégie en temps et lieux qui concilie la continuité et le changement. Des systèmes ago-antagonistes fondamentaux planning/learning, téléo/écolo, continuité/changement, découlent des ago-antagonismes dits « dérivés » qui donnent corps au référentiel de l’entreprise durable (Martinet, 2002c). Celle-ci privilégie non pas les seuls résultats économiques, mais son efficacité passe par sa responsabilité et responsabilisation, et la performance économique ne saurait être sans les performances sociale et écologique. Cette nouvelle définition de l’entreprise et de sa performance passe par une nouvelle stratégisation : une formation (≠ de la dichotomie formulation/mise en œuvre) de la stratégie, laquelle autorise les dialectiques délibéré/émergent, global/local, savoirs/relations, facteurs/acteurs, technico-économique/socio-politique. Chaque pôle n’inhibe pas l’autre, mais entretient une relation structurante. Une coexistence des pôles et des paradoxes permet ainsi l’intégration des subtilités des situations et donc de leur complexité. 109 Logique Téléo Evolution des Incertitude Stratégies Globales Rétention TOP MANAGEMENT = Absorption Macro VIRTUALISATION OPEN LEARNING / STRAT THINKING Nouvelles Stratégies Compétences Globales PLANNING Sélection MIDDLE MANAGEMENT Adaptation Locale des Stratégies Globales Contributions Locales aux Stratégies Globales ACTUALISATION Ba2 LEARNING Double Compétences Indiv. Locales Ba1 Absorption d’incertitude micro + Savoirs Single Loop PARTIES PRENANTES LOCALES Logique Ecolo 110 Relations Négociation - stabilisation ACTION LEARNING Co-création 2. Responsabilisation des managers et gouvernance Quatre principes pour une entreprise responsable. La modélisation générale qui précède constitue un cadre conceptuel susceptible de guider une grande variété d’actions et de décisions, toutes plus ou moins contributrices à la construction de l’entreprise responsable. Il peut permettre de mieux piloter des processus responsabilisants tant au niveau des hauts dirigeants que des managers intermédiaires. Reformulons les principes fondamentaux qui fondent leur cohérence et, pourtant, font que l’entreprise responsable augmente ses chances d’être aussi une entreprise durable. P1 : Le Principe Ethique La revendication d’a-moralité de la gestion n’est plus guère recevable tant elle a pu, notamment ces dernières années, servir de paravent à l’immoralité de nombreuses pratiques des affaires. Au plan de la recherche, elle a surtout favorisé un scientisme positivisme éloignant la discipline de sa raison d’être praxéologique contribuant à la production non questionnée d’un « ordre gestionnaire ». L’éthique ne peut être mise en marge ni des pratiques, ni de la recherche en gestion. Elle ne peut, a contrario, être seulement au service d’une critique globale et radicale du « système capitaliste ». Elle doit leur être pleinement intégrée tant elle est critique pour le devenir des sociétés libres. P2 : Le Principe Politique Reconnaître la dimension politique du management, créer les conditions de son plein exercice sont le corollaire de fait du pouvoir désormais détenu par les grandes entreprises non seulement dans la réalisation de leurs missions économiques mais dans la structuration du monde. Le processus de fixation des buts, les indicateurs d’évaluation et de mesure des performances, l’intégration ou l’exclusion de telle ou telle dimension ou partie prenante, la récompense ou la sanction des responsabilités ou des irresponsabilités managériales… sont tellement déterminants qu’ils ne peuvent rester impensés ou traités de façon purement technique. Si l’on a pu constater la diffusion massive et fulgurante de « l’ebitda »1 comme indicateur privilégié de performance de l’entreprise et comme assiette des dirigeants, a-t-on étudié la façon dont il a façonné nombre de décisions stratégiques ? La prise en compte explicite du politique dans et alentour de l’entreprise conduit à interroger le style politique ou, le mode de gouvernement pratiqués ; dans chaque cas sous examen : Quelles doses de commandement, de négociation ou d’animation ? Quels équilibrages entre l’imposition ou l’influence ?... P3 : Le Principe Organisationnel La grande entreprise actuelle ne peut plus s’en remettre à la seule hiérarchie, à la délibération au sommet pas plus qu’elle ne peut se satisfaire d’un pilotage à vue et d’un court-termisme sans cesse remis en cause. Une structuration souple est nécessaire propice au développement de l’organisant (« organizing ») plutôt qu’à la cristallisation de l’organisation. Plus la ressource critique devient la connaissance, plus la politique trouve sa pleine expression, plus l’autonomie mais aussi la coordination des acteurs s’avèrent déterminants, plus la structuration fonctionnelle et le pilotage des processus doivent favoriser les apprentissages dynamiques. P4 : Le Principe Systémique 1 Ebitda : earnings before interest, taxes, depreciation and amortization. 111 Il s’agit ici d’un principe de bouclage de nature évidemment épistémologique. La conviction affichée est que la responsabilité de l’entreprise se construit dans la durée et par une multitude d’inter-actions. Dès lors, des effets de levier doivent être recherchés ainsi que la mise en cohérence de spirales de développement dont a pu montrer le caractère généralement ago-antagoniste. La recherche permanente de la composition de l’autonomie et de l’interdépendance marque, au premier chef, ce principe systémique. Figure 2 : Quatre principes pour une entreprise responsable Principe Ethique Ethique Politique Principe Organisationnel Principe Politique Style Politique Organizing > Organization Autonomie / Interdépendance Principe Systémique Des middle managers responsables Selon la perspective relationnelle de Dyer et Singh (1998), une entreprise s’assure un avantage concurrentiel soutenable lorsqu’elle développe des relations avec d’autres acteurs. Dyer et Singh (1998) définissent la rente relationnelle comme un bénéfice substantiel produit conjointement lors d’un échange et qui ne peut pas être généré par l’une ou l’autre société seule. Elle résulte de contributions idiosyncrasiques communes relevant d’une alliance spécifique. Les auteurs considèrent d’ailleurs que les membres de l’échange, de la relation, constituent la source la plus importante des nouvelles idées et d’information. Ainsi, si la relation dégage une rente, l’entreprise a tout intérêt à développer et maintenir un tissu de relations au niveau local, qui contribuent à la performance de l’entreprise. Une personne à la frontière des mondes internes et externes, un « gatekeepers », intervient : le « middle manager ». 112 Les middle managers, alors représentants de l’entreprise, de ses ressources et objectifs, doivent les concilier avec les logiques des territoires, leurs ressources et objectifs. Les middle managers se trouvent impliqués dans des relations complexes entre des organisations diverses qui, tout en poursuivant leurs buts dans des logiques qui leur sont propres, cherchent à construire un intérêt commun, et réciproque. Dès lors, le middle manager repère et participe à la construction du projet commun, et est en situation d’identifier les acteurs parties prenantes, pour ensemble, construire une responsabilité sociale partenariale et non pas de la seule entreprise. Des précédents travaux (Payaud, 2004) se sont attachés à la problématique des rôles, fonctions et responsabilités du middle manager dans la formation des stratégies de la grande entreprise de services de réseau. Nous avons vu que si la structure de l’entreprise l’autorisait et l’encourageait, les relations inter- et intra-organisationnelles du middle manager lui procuraient une position favorable pour participer à la co-production de l’offre, à la co-construction des compétences et donc à la co-évolution de la stratégie. Nos recherches ont alors montré que ses relations inter- et intra-organisationnelles augmentaient ses savoirs micro et macroscopiques et ainsi, alimentaient les modèles mentaux (individuels et organisationnels) nécessaires à des processus d’apprentissage complet. De fait, les relations inter et intra et les savoirs micro et macro le placent comme un acteur au centre du processus évolutionniste (Nelson et Winter, 1982) : source de variation, déterminant de la sélection et catalyseur de rétention. Dès lors, le réseau, comme le territoire, en établissant des relations dans la durée assurent une évolution de leurs organisations et une progression, un enrichissement de la cause commune. Les actions économiques et comportements sociologiques des uns sont enchâssés dans ceux des autres. La notion d’ « embededness » de Granovetter, le plus souvent traduite par « encastrement », est décisive dès lors que l’on veut lier les relations sociales aux faits économiques. En effet, les échanges inter-organisationnels développés par le middle manager mobilisent et participent à la formation de réseaux sociaux, sources de variation. La construction de réseaux sociaux passe par les échanges inter-organisationnels. Les middle managers sont en position théorique de reconnaître, d’identifier les signaux susceptibles de conduire à des opportunités ou des menaces de l’environnement qui peuvent influencer soit la soutenabilité ou la désuétude des compétences actuelles, soit de révéler des futures compétences pour lesquelles il conviendrait d'investir. Les middle managers doivent être encouragés à développer et entretenir des relations avec l’ensemble des partenaires de leur environnement local, afin de pouvoir capter ces signaux. Des dirigeants, stratèges et gouvernants responsables Une demande de responsabilisation accrue des middle managers ainsi devenus « stratèges » ordinaires serait évidemment scandaleuse et obscène si elle n’était orientée, couverte et amplifiée par une (re)responsabilisation forte, incitée et sanctionnée des managers supérieurs et des hauts dirigeants. Cette exigence implique « ipso facto » l’interdépendance entre la stratégie et le gouvernement d’entreprise, tant au plan des pratiques que des champs théoriques. En effet, et comme l’ont très bien clarifié Pérez (2002) et le collectif dirigé par Igalens (2004), les trois interrogations essentielles – de quoi est-on responsable ? Qui est responsable ? Devant qui ? – se voient apporter des réponses, implicites ou explicites, inconscientes ou conscientes, lors de chaque décision ou acte managériaux. La qualité de ces réponses est évidemment façonnée par le « modèle » et la doctrine auxquels se réfère, de façon plus ou moins, le décideur. Nous avions stylisé et opposé en les tendant au maximum, les deux référentiels fondamentaux qui peuvent être dégagés de la littérature économique et gestionnaire, que nous avons nommés « financier » et « durable » (Martinet, 2002a et 2002b). Pour le présent propos, ils se superposent bien aux modèles de la dissociation et de l’intégration retenus par Pérez. La place manque pour les détailler ici. Relevons seulement que le référentiel « durable » se nourrit de l’approche « stakeholders » (Freeman, 1984 ; Martinet, 1984), de la création de valeurs partenariales (Charreaux et Desbrières, 1998), d’une conception managériale dialogique acteurs/facteurs 113 (Martinet, 1984) ou relations/savoirs (Hatchuel, 2000), d’une vision de l’entreprise en « société » (Perroux, 1973) ou « encastrée » (Granovetter, 2000) tendue par un projet englobant (« affectio societatis ») bien au-delà du nœud de contrats, qui se forme de façon organique dans la durée. Ce référentiel donne sens à la notion de responsabilité globale ou élargie où les dirigeants sont récompensés et sanctionnés sur l’atteinte, lors d’une durée suffisante, d’objectifs économiques, sociaux et écologiques et non sur la seule maximisation de la valeur actionnariale lors de jeux courts et reconduits de trois mois en trois mois. En matière stratégique, ce référentiel conduit à réexaminer, de façon critique, les injonctions du référentiel financier qui ont eu tendance à s’imposer au cours des années 90. c’est ainsi que le triptyque « métier unique, rachat de ses propres actions, rentabilité permanente de 15% » et ses implications directes – suspicion de l’autofinancement et de l’enracinement des dirigeants – doit donner lieu à des diagnostics et des recommandations beaucoup plus contingentes ; c’est précisément l’objet de la stratégie bien comprise que de moduler, selon les espaces x temps, les degré de diversification, de dilution, de rentabilité, d’autofinancement et d’enracinement qu’il est souhaitable de conduire dans une optique de développement soutenable dans la durée. La stratégie redécouvre alors les principes fondamentaux d’autonomie – dans ses différentes dimensions et, notamment, à l’égard des marchés financiers – de responsabilité (Jonas, 1990) et de précautions (Godard, 2002) pourvu qu’on en fasse pas une lecture intégriste ou radicale mais raisonnable. Les décisions stratégiques qui font la vie des sociétés cotées ne sont intelligibles qu’à condition de réarticuler le contexte stratégique, les critères de gouvernement et les menées personnelles des hauts dirigeants. En ce sens, elles contiennent inexorablement une dimension morale (ou a-morale ou immorale) et politique puisqu’il s’agit toujours d’arbitrages et de négociations entre les intérêts divergents. La réduction de la raison à la rationalité de l’homo oeconomicus dont les vertus épistémologiques sont bien connues – simplicité, lisibilité externe, standardisation, calculabilité, abstraction mimétisme – détruit l’objet même et la raison d’être du gouvernement et de la stratégie d’entreprise (Williamson, 1991) en même temps qu’elle évacue la dimension politique (Revault d’Allones, 1999). Sauf à définir cette dernière comme le serviteur exclusif de la maximisation de l’utilité économique de l’acteur, quel qu’il soit : consommateur, actionnaire, salarié, citoyen, riverain… dont les interdépendances sont nulles ou, en totalité, réglées par des marchés. Même la finance « bien comprise » – c’est-à-dire une dialogique effective entre les deux référentiels retenus – passe obligatoirement par l’acceptation des rationalités multiples, c'est-à-dire de la raison. N’est-il pas révélateur qu’un économiste peu suspect d’antipathie à l’égard de la finance, intitule justement son Xème commandement : « Que tout explose, tu empêcheras » et de conclure : « [l’économie financière] implique plus de conscience et de clairvoyance de chacun, plus de contrôle et de temps passé pour en lire les résultats, des sanctions plus nettes pour les mauvaises pratiques, les abus et les mensonges (…). Cette économie financière qui s’étend, avec ses nouvelles responsabilités et ses risques, se doit d’intégrer de nouveaux objectifs. L’entreprise peut ainsi être éthique en ne faisant pas travailler des prisonniers ou en n’intervenant pas dans des pays à moralité douteuse ; ou responsable, en s’engageant sur des achats de qualité, en prenant en compte l’environnement… » (Betbèze, 2003, page 294). Reconnaissance implicite de l’absolue nécessité de la stratégie et de son épistémologie – complexité, autonomie, irréductibilité au calcul, contextualisation, singularité...—. CONCLUSION Le corpus de la stratégie d’entreprise ne s’est explicitement développé et institutionnalisé qu’à partir des années 60 avec les travaux séminaux de Chandler, Andrews et Ansoff de nature historique et praxéologique. Sa scientifisation s’est traduite par un fractionnement et des disjonctions glissantes. Ainsi, la disjonction initiale entre stratégie et management a été largement 114 comblée à partir de 1973, année de naissance académique du management stratégique. A contrario, la liaison politique-stratégie, centre de gravité du corpus à ses débuts, s’est affaiblie jusqu’à dissocier les champs de réflexion de la stratégie et du gouvernement d’entreprise. Dissociation aujourd’hui intenable pour les deux champs. Pour les sociétés et les groupes cotés, les choix stratégiques deviennent incompréhensibles si l’on n’intègre pas les conditions de gouvernement – disons pour simplifier avec Gomez (2002), la latitude discrétionnaire des dirigeants et le mécanisme d’alignement « ex post » par le marché ou « ex ante » par la structure, la composition et le fonctionnement du conseil d’administration—. Symétriquement, et sauf à adhérer à une conception orthodoxe de la théorie de l’agence où le gouvernement d’entreprise est le résultat technique d’une optimisation financière entre les coûts de surveillance et les coûts de contrôle et le profit que les dirigeants sont incités à produire, toute avancée sur le gouvernement passe par la prise en compte de formation des stratégies ; ou, dit de façon imaginée, de ce qui se passe dans la tête et les schémas mentaux des managers comme dans l’organisation. Nous avons suggéré par ailleurs (Martinet, 2002b) que ces schémas mentaux, comme la légitimité politique de leur mise en acte, c'est-à-dire dans le langage de Weick le contenu et l’orientation des « enacting » et « sensemaking », étaient largement façonnés par les institutions et, plus encore, par les technologies invisibles qu’elles promeuvent. Les systèmes de notation, les indicateurs de performance, les procédures et les normes jouent aujourd’hui un rôle déterminant. La phase de capitalisme financier qui prévaut depuis 1990 a imposé une prescription marquée par cinq caractéristiques : sophistication technique, quantification, vélocité, prescription oligopolistique et mimétisme autoréférentiel. On connaît aujourd’hui les effets pervers, les excès, les malhonnêtetés que ces mécanismes ont favorisés ou, à tout le moins, n’ont pas su endiguer. Dès lors, et si l’on se situes dans un paradigme organisationnel du gouvernement, de la stratégie et des finances (Charreaux, 1997), on ne peut se satisfaire de quelques modifications formelles de la composition des conseils d’administration. Si la séparation des fonctions de président et de directeur général, la présence d’administrateurs indépendants… sont probablement nécessaires, de telles mesures doivent s’inscrire dans une re-conception politique de l’entreprise où la participation effectives des parties prenantes, la négociation sur les voies et moyens du développement, l’accord sur les critères de performance et d’évaluation… redeviennent premiers. 115 BIBLIOGRAPHIE Argyris C. & Schön D., (1978), Organizational Learning : a Theory of Action Perspective, Addison Wesley Publishing Company. Bernard-Weil E., (1988), Précis de Systémique Ago-Antagonisme, Limonest, L’interdisciplinaire. Betbèze J.P., (2003), Les Dix Commandements de la Finance, Paris, Odile Jacob. Charreaux G., (1997), Le Gouvernement des Entreprises, Paris, Economica. Charreaux G., Desbrières Ph., (1998), « Gouvernance des entreprises : valeur partenariale contre valeur actionnariale », Finance, Contrôle, Startégie, vol.1(2), pp.57-88. Dyer J.H. et Singh H., (1998), " The relational view : cooperative strategy and sources of interorganizational competitive advantage", Academy of Management Review, vol.23(4), pp.660-679. Eiglier P. & Langeard E., (1987), Servuction : le Marketing des Services, Paris, Mc Graw Hill. Floyd SW. & Wooldridge B., (2000), Building Strategy from the Middle, Foundations for Organizational Science, Sage. Freeman R.E., (1984), Strategic Agent : a Stakeholder Approach, Pitman. Godard et alii, (2002), Traité des Nouveaux Risques, Paris, Gallimard-Folio. Gomez P.Y., (2002), « Le gouvernement des entreprises et incidence sur son développement », in Durand R., dir., Développement de l’Organisation, Paris, Economica. Granovetter M., (2000), Le Marché Autrement, Paris. Hatchuel A., (2000), « Quel horizon pour les sciences de gestion ? vers une théorie de l’action collective » in Les nouvelles fondations des sciences de gestion coord. par A.David, A. Hatchuel et R. Laufer, Paris, Ed. Vuibert-Fnege, 216 pages. Igalens J., (dir) (2004), Tous Responsables, Paris, Ed. d’Organisation. Jonas H., (1990), Le Principe de Responsabilité, Paris, Cerf. Lovas B. & Ghoshal S., (2000), « Strategy as guided evolution », Strategic Management Journal, vol.21(10), pp.875-896. Martinet A.C., (1984), Management Stratégique, Organisation et Politique, Paris, Mc Graw Hill. Martinet A.C., (2002a), « Eco-management et stakeholders », Congrès Européen de Systémique, Heraklion, sept. Martinet A.C., (2002b), « L’actionnaire comme porteur d’une vision stratégique », RFG, n°141, pp.57-76. Martinet A.C., (2002c), « La formation de la stratégie : entre gouvernement et management », Colloque CREPA-EURISTIK : « le travail de direction générale : stratégie, organisation, contrôle ». Martinet A.C. & Payaud M.A., (2004), « Absorption de l'incertitude, contextualisation de la stratégie et middle managers », 4ème Journées Internationales de l'Association Tunisienne des Sciences de Gestion (ATSG), Hammamet, Tunisie, 26 pages. Nelson R.R. et Winter S.G., (1982), An Evolutionary Theory of Economic Change, Cambridge, Harvard University Press. Nonaka I. et alii, (2000), “ SECI, Ba and Leadership : a Unified Model of Knowledge Creation ”, Long Range Planning, vol.33, pp.5-30. Payaud M.A., (2004), « Le middle manager dans la formation de la stratégie : repère théorique et précision empirique », in Perspectives en Management, sous la direction de B. Quélin et L.Mezghani, Paris, Ed. EMS. Pérez R., (2002), « A propos de responsabilité globale en management », Communication à la IXème Journée François Perroux, Lyon, 10 décembre 2002. Perroux F., (1973), Pouvoir et Economie, Paris, Bordas. Revault d’Allones M., (1999), Le Dépérissement de la Politique, Paris, Champs-Flammarion. Williamson (1991), « Strategizing, economizing and economic organization », Strategic Management Journal, vol.12, pp.75-94. 116 ASAC 2005 Porrini Torronto, Ontario Patrizia Long Island University ARE ALLIANCES A RISKY PROPOSITION? TAKING A CLOSER LOOK AT ALLIANCES FROM A SYSTEMATIC-RISK PERSPECTIVE This paper examines whether alliances decrease firms’ systematic risk. Alliances are popular modes by which firms share resources and capabilities. Alliances can reduce firms’ risk by decreasing firms’ susceptibility to their environmental uncertainty, allowing for risk sharing, and allowing for reversibility of resource-commitments. Thus alliances can limit firms’ exposure to risk as alliances help firms avoid being locked out of new technologies, and help firms defend and extend their strategic positions. The study examines 407 alliances where all partners are public US-based firms and finds that alliances decrease firms’ systematic risk. Specifically, systematic risk decreases by an average of 17% and 25% for one-year and two-year windows of Beta surrounding alliance announcements, respectively. Furthermore, different alliance types have different risk-reduction benefits. These findings have important implications for theory and practice. Alliances have become popular forms of exchange relationships by which firms share and develop resources, capabilities and core competencies (Balakrishnan & Koza, 1993; Kogut, 1988a; Mowery, Oxley & Silverman, 1996; Teece, 1992). Thus it is not surprising that firms engage in alliances at increasing rates. Over 20% of the revenue generated from the top 2000 U.S. and European is generated from alliances (Booz, Allen, & Hamilton, 2003). In fact, 82% of polled executives believe alliances will be a prime vehicle for future growth (Accenture, 1999). As alliances have become increasingly popular researchers have examined whether alliances create value by examining changes in stock prices during short windows surrounding alliance announcements (Neill, Pfeiffer, and Young-Ybarra, 2001; Anand and Khanna, 2000a; Chan, Kensinger, Keown and Martin, 1997; Das, Sen and Sengupta, 1998; McConnell and Nantel, 1985). These studies have shown that share prices react immediately to alliance announcements with most finding that alliances do create value. However because the window around the alliance announcement date is usually short the impact of an alliance on value creation appears short-lived. Although studies have explored immediate stock price movements in response to strategic decisions, short-term stock price movements do not inform of longer-term risk adjustments in response to strategic decisions. This study examines whether value created by alliances has a more enduring impact on firms and their share prices by examining whether alliances decrease firms’ systematic risk. Alliances can have an enduring impact on firms and firms’ share prices and this can result in a change in firms’ systematic risk. A firm’s systematic risk is the covariance of the returns of a security with those of the market, divided by the total variance of the portfolio of stocks that represent the market portfolio (Markowitz, 1952). A firm’s systematic risk is higher when the firm’s stock is more volatile and hence riskier than the market portfolio. Hence a firm’s systematic risk characterizes its relationship with the market over a period of time. When all other factors are 117 held constant, lower systematic risk implies a lower required rate of return, which increases the value of the firm (Van Horne, 1980: 68). Thus corporate managers can increase the wealth of their stockholders by pursuing actions that lower firms’ systematic risk. Alliances can help firms to deal with environmental trends such as the increasing convergence of high-technology industries, the increasing presence of global competitors, and the speed and cost of developing new technologies, and environmental uncertainties (Hitt, Hoskisson and Ireland, 2003). These pressures create uncertainties that require speed, flexibility and innovation for firms to remain competitive. By enabling firms to share resources and capabilities, alliances can help firms combat these external pressures. Alliances have the potential to have a more enduring effect on firms as alliances create the potential for firms to share resources in pursuit of new products, technologies, and strategies to initiate or adapt to competitive change (Volberda, 1996). Thus alliances may enable firms to create value for a longer term and can ultimately shape their systematic risk. This study examines whether alliances create longer-term benefits by examining changes in firm’s systematic risk. An investigation of whether alliances decrease firms’ systematic risk contributes to literature on alliances in several ways. First, as mentioned above previous research on alliances has found that alliances create value for firms immediately following their announcement. However excess returns over a short event window are not necessarily indicative of whether alliances affect firms in the longer-term. Risk is a more permanent attribute of a firm’s stock price behavior whereas returns measure the immediate impact of the event on the firm’s stock price. Studies in the alliance literature have found that alliances create learning effects by examining returns, but can alliances reduce systematic risk? Second, researchers have discussed systematic risk within the context of corporate strategy and diversification and have argued that managers are not able to influence a firm’s systematic risk, thus to some extent management theory diverges from financial theory, which states that only systematic risk matters as unsystematic or firm-specific risk can be diversified away by investors. By exploring whether alliances can influence firms’ systematic risk the study contributes to this discussion by investigating whether systematic risk may be managed by strategic choices such as alliances. Finally, within the literature on diversification there are discussions of relatedness and their effect on firms’ performance and risk. This study contributes to this discussion by investigating whether relatedness of alliance partners can lead to differences in risk reduction. The next section discusses hypotheses about alliances and changes in firms’ systematic risk. Ensuing sections describe methodology and results and the final section draws conclusions and discusses implications. THEORY AND HYPOTHESES Strategic Management and Systematic Risk The Capital Asset Pricing Model (CAPM) implies that managers should focus on reducing their firm’s systematic risk (Beta) and not be concerned with unsystematic (firm-specific) risk. One reason for this is that investors can diversify away unsystematic risk, thus the market will only reward systematic risk. However, Chatterjee, Lubatkin and Schulze (1999: 556) point out that the CAPM poses a strong challenge to the field of strategy in that the field of strategy is based on the assumption that management matters yet management supposedly cannot influence systematic risk. Chatterjee, Lubatkin, and Schulze (1999) claim that one issue with CAPM is that reducing systematic risk requires that managers do what they cannot as it requires that managers reduce investors’ exposure to macroeconomic uncertainties at a cost lower than what it would cost investors to do so on their own by altering their investment portfolio. The role of managers is to protect their revenues from market forces in ways that are valuable to investors through strategic actions. Thus some researchers conclude that the CAPM goes against strategic management theory 118 since it implies that managers should focus on that which they cannot influence (Chatterjee, Lubatkin, and Schulze, 1999; Bettis, 1983). However, to date researchers have not considered the effect of alliances on firms’ systematic risk. Alliances are intermediate contracts that are frequently used by firms to limit their exposure to uncertainties by sharing risks as well as resources with alliance partners (Gommes-Casseres, 2000). Consequently, alliances may offer a way for managers to reduce systematic risk as managers may be able to limit firms’ exposure to macroeconomic uncertainties at a lower cost than what investors can transact on their own. For example, investors aren’t able to form inter-firm alliances on their own nor can they gain access to or use information and resources that firms can access and use via alliances. Although investors may form a portfolio of the stocks of most firms in an alliance network if the firms are public, doing so cannot simulate the value added by inter-firm relationships. Firms’ alliances can provide risk reduction benefits by providing proprietary access to new technologies, information, and other firms’ resources. In this regard, alliances may reduce risk by limiting investors’ exposure to macroeconomic uncertainties at a lower cost than what it would cost investors to do so on their own by altering their investment portfolio. The next section discusses how alliances can create value by reducing firms’ systematic risk. Alliances and Systematic Risk Researchers have found that alliances can provide many benefits for firms. Alliances allow firms to learn from each other by providing opportunities whereby firms absorb information, knowledge, capabilities and skills from partners (Hamel, 1991; Kale, Singh, & Perlmutter, 2000; Khanna, Gulati, & Nohria, 1998). Alliances also help partners to learn how to manage the collaboration process and learn how to manage alliances better leading them to build alliance capabilities which enable firms to access resources from its environment (Amburgey & Dacin, 1996; Arino & de la Torre, 1998; Doz, 1996; Anand & Khanna, 2000a; Dyer & Singh, 1998; Kale & Singh, 1999; Lyles, 1988). Exposure to partners’ resources may increase firms’ sensitivity and ability to detect new opportunities (Cohen & Levinthal, 1990; Lane & Lubatkin, 1998). In all, the benefits of alliances may enable firms to reduce their risk. Risk is characterized by the uncertainty related to an outcome. Firms face uncertainty from their technological, market, and competitive environments (Gommes-Cassares, 2000). Uncertainty and volatility from firms’ environments can increase firms’ riskiness because firms’ products, technologies, and core competencies may become uncompetitive or obsolete. To reduce the ratio of the firms’ covariance with the market to the total variance of the market, alliances must reduce firms’ susceptibility to environmental changes such as that resulting from economic downturns or technological obsolescence. One way that alliances may help firms’ reduce their systematic risk is by providing firms opportunities to reduce the uncertainty in their immediate environment (Hennart, 1988; Kogut, 1988b). Firms’ access to partners’ skills, capabilities, core competencies, technologies, and markets not only provide opportunities for learning (Anand & Khanna, 2000a), but have longer-term benefits that may shape firms’ risk. Alliances give partners the ability to gain tacit resources such as capabilities, competencies, technologies, and know-how, among many others (Hamel, 1991; Kale, Singh, & Perlmutter, 2000; Khanna, Gulati, & Nohria, 1998). Furthermore, alliances allow partners to access information about competitors’ strategic direction or forthcoming technologies, making firms’ competitive environment less unpredictable (Badaracco, 1991). Alliances also allow firms to access technologies that firms do not currently have in house thereby helping firms react more quickly to new innovations (Porter and Fuller, 1986). By providing these benefits alliances act as a buffer between the firm and the environment. Thus alliances help to decrease environmental uncertainty by increasing opportunities, increasing resources for dealing 119 with environmental threats, and decreasing firms’ susceptibilities to environmental shocks. In this regard alliances help firms to reduce risk by reducing uncertainties related to competitors’ strategic directions, forthcoming technologies, and outcomes of strategic investments, actions or decisions. A second way that alliances may help firms’ reduce their systematic risk is by providing opportunities for risk sharing. The cooperative nature of alliances allows for risk sharing by allowing firms to co-develop, co-market or co-fund new projects or innovations. By allowing for partial commitments alliances allow firms to leave a portion of their resources available for other investments or projects thereby making firms less susceptible to the repercussions of relying on one or few major projects. In this regard alliances allow firms to invest a fraction of their net worth while leaving funds available for additional projects that may become quite crucial to their success. For example, biotechnology allows for faster development of drugs relative to traditional pharmaceutical methodologies. This has lead many pharmaceuticals to form alliances with biotech firms. Although these alliances do not cost much, they offer pharmaceuticals large potential gains by making pharmaceuticals less susceptible to the threat that biotechnology will emerge as the primary way to engineer drugs. Similarly, Toyota and GM entered into an alliance to develop alternative-powered vehicles, which protects each firm’s future competitive position with respect to this upcoming technology, as well as protects each firm from bearing the total loss of the technological investment not paying off (Ball, 1999). Thus through alliances firms can benefit from future gains by securing an investment in a potentially beneficial technology while minimizing firms’ susceptibilities to repercussions that come with strategic advancement. In this regard, alliances protect firms’ future positions, allow firms to share losses, and allow firms to invest only a fraction of their net worth into new projects that may benefit or compromise their future. A third way that alliances may help firms reduce their systematic risk is by allowing firms to make incremental and reversible commitments to unfolding strategies. Reversibility is very important for helping firms avoid being stuck in expensive irreversible commitments that may create additional risks for the firm in the future. For example, a firm may make an acquisition to acquire needed technologies. Yet, making an acquisition may lead a firm to pay a sizeable premium, raise external capital, and create problems associated with integration of the acquired firm (Haspeslaugh and Jemsion, 1987). Alliances can help firms to secure access to resources or capabilities that may be useful for their future competitiveness, without creating risks of high exit costs. The open-ended nature of alliances allows firms to exit alliances with relatively fewer costs than more permanent arrangements like acquisitions or greenfields, which may lead to divestiture costs or costs and risks of building new operations from the ground up. Thus firms that need to change strategic direction or build an international presence to remain competitive can use alliances to ease into new areas or new markets rather than make irreversible commitments or commitments that are costly to reverse. In this regard, alliances can help firms to hedge risks as they allow firms to access new technologies that can help to change firms’ strategic positions, while minimizing the potential repercussions. In conclusion, alliances can reduce firms’ systematic risk by creating opportunities for firms to reduce uncertainty, increasing firms’ opportunities to pursue emerging technological trends with minimum exposure or investment, allowing firms to share risks with other firms by avoiding full commitments to strategies and reducing exit costs. These benefits can ultimately reduce firms’ total exposure to market-wide changes and events. By minimizing firms’ exposure to environmental uncertainties and decreasing firms’ risk exposure, alliances can reduce firms’ systematic risk. H1: Alliances decrease firms’ systematic risk. Although, the main premise of the study is that alliances can decrease systematic risk, the inclusion of data on different types of alliances can lead to more insight. Different types of alliances may have different benefits for risk reduction because different types of alliances allow firms to 120 exchange specific types of resources. For example, research and development (R&D) alliances allow firms to share their research and development resources and can enable the exchange of explicit and tacit know-how. A manufacturing alliance allows firms to exchange know-how related to manufacturing specific products. A technology transfer alliance allows firms to exchange knowledge relating to a specific technology. A licensing alliance allows members to manufacture and sell a firm’s products. A marketing alliance allows firms to co-market a product or service. Alliances of different types increase a firm's exposure to diverse areas of expertise and its ability to recognize and value new information and resources in those areas and apply them to its current operations (Cohen & Levinthal, 1990). Each of these different types of alliances allows firms to gain access to specific aspects of partners’ resources providing partner-specific and type-specific information, resources, and competencies, which can facilitate different levels of risk reduction. H2a: Cross border alliances decrease firms’ systematic risk. H2b: Licensing alliances decrease firms’ systematic risk. H2c: Manufacturing alliances decrease firms’ systematic risk. H2d: Marketing alliances decrease firms’ systematic risk. H2e: Research and development alliances decrease firms’ systematic risk. H2f: Supply side alliances decrease firms’ systematic risk. H2g: Technology transfer alliances decrease firms’ systematic risk. Similarly, prior work has found distinctions between alliance types and levels of learning (Anand and Khanna, 2000a; Anand and Khanna, 2000b). To date, a main factor of distinction among learning effects from diverse alliances types has been the type of alliance contract. For example, Anand and Khanna (2000a) found that learning from alliances is greater in less specific types of alliances where there is greater contractual ambiguity or relatively less precise criteria available to guide the alliance. Anand and Khanna (2000b) also found empirical evidence that there is a clear relationship between the extent of ambiguity in codifying knowledge and the choice of contract in interfirm alliances. Specifically, licensing alliances are significantly more frequently employed than other types of alliances in contexts where it is relatively easy to establish property rights over knowledge and where ambiguity is low (Anand and Khanna, 2000b). Licensing alliances are very clear and have well-specified dimensions on the calculation of royalty payments, exclusivity clauses, territorial restrictions and many other variables (Caves, Crookell, & Killing, 1983; Parr & Sullivan, 1988). Contract types that are less specific than licensing contracts provide broader access to partners’ resources because there are relatively fewer specifications and restrictions. Consequently, it is not surprising that prior studies have found learning effects are weaker in licensing alliances than in other less specific contract-types such as R&D, manufacturing, technology transfer and marketing alliances for firms in manufacturing (Anand and Khanna, 2000a). Broader access to information and experience can increase the level of learning from alliances and this may increase the possibility that what is learned is applicable to future experiences (Cohen and Levinthal, 1990). Similarly, different types of alliances ought to have different effects on the level of risk reduction. Specifically in more specific types of alliance contracts such as licensing agreements, risk reduction ought to be lower than risk reduction from other alliance types. Alliance types that provide broader access to partners’ resources ought to be more beneficial to acquisition performance than those that are more specific and restrictive. Broader access to partners’ resources allows firms to gain more from alliances. Although licensing alliances may also allow firms to obtain access to firms’ resources, the access may be narrower than that obtained through other types of alliances such as R&D, manufacturing, supply side, technology transfer or marketing alliances because of the more specific nature of licensing agreements. As a result, access to information that can act to reduce uncertainty and can lead to risk sharing is likely to be weaker in 121 licensing alliances due to narrower exposure to partners’ resources. Thus, the less specific is the type of alliance contract the greater the amount of partner-specific information that the firm can obtain and the greater the possibility for risk reduction. H3: Risk reduction is greater in cross border, R&D, marketing, manufacturing, supply side, and technology transfer alliances than for licensing alliances. A widely held belief among many strategic management theorists is that resource combinations among related firms or by firms with high business commonality ought to be more advantageous than combinations by firms with less related businesses (Chatterjee, 1986; Michel and Shaked, 1984; Salter and Weinhold, 1979; Lubatkin, 1983; Bettis and Hall, 1982; Porter, 1985; Singh and Montgomery, 1987). Some reasons that have been used to support this belief are that combinations by firms in similar businesses provide opportunities for specialization, improvements in market power, reduction of costs or enhancement of quality by providing opportunities to exploit scale and scope economies in tangible areas such as manufacturing, research, and distribution. In addition combinations by firms in similar businesses allow for greater administrative know-how and the possibility of obtaining monopolistic influence on pricing or purchasing inputs. Relatedness among firms has been hypothesized to be beneficial to firms as firms have a common language by which to exchange resources (Cohen and Levinthal, 1990). Similarly, in alliances business commonality among alliance partners may also ensure the applicability of one firm’s resources to a partner’s resources. Firms operating in similar businesses usually have similar resource bases because they use similar technologies, compete in similar markets, cater to similar customer needs, and serve similar market segments (Bettis and Prahalad, 1995; Lane and Lubatkin, 1998; Prahalad and Bettis, 1986. Although alliances among related firms are likely to reduce firms’ risk, from a risk reduction standpoint alliances among unrelated firms are likely to be more advantageous for risk reduction. Alliances are greatly beneficial when resource endowments among partners are complementary but not identical or highly similar (Grant, 1991; Hamel, 1991; Khanna, Gulati, and Nohria, 1998; Lane and Lubatkin, 1998; Mowery, Oxley, and Silverman, 1996). Many theorists have recognized that collaboration among firms with complementary resources affords firms opportunities to exploit new opportunities. Dussauge, Garrette, and Mitchell (2000) find that alliances in which partners contribute different capabilities lead to greater levels of learning and capability acquisition than do alliances in which partners contribute similar capabilities. As such, combinations via alliances by unrelated firms may offer greater opportunities from a risk reduction standpoint in the areas of uncertainty reduction and risk sharing (Amit and Livnat, 1988). For example, combinations among firms with related resources may result in overly narrow focus, lack of diversification, and lack of market breadth making firms more sensitive to their environment. Firms often try to reduce their exposure to risks by diversifying technologies or pursuing new innovations, to recover from quickly eroding competitive positions. Thus specialization may further increase firms’ exposure to risk. In this regard, alliances among partners in unrelated businesses may offer an element of resource diversification by offering know-how and potential for unique strategic development. Furthermore, access to a broader information and opportunity set may decrease firms’ susceptibility to environmental occurrences and increase the potential for risk reduction. H4: Risk reduction is greater when members of the alliance operate in unrelated businesses. DATA AND METHODOLOGY Data The sample consists of alliances announced between January 1, 1996 and December 31, 1998. The study includes only alliances in which all of the firms in each alliance are U.S. participants and are publicly held so that stock price data would be available for all alliances and each partnaer. Prior to 1990, SDC did not initiate systematic and comprehensive data collection on alliances and even 122 between 1990 and 1993 not all deals were tracked due to corporate reporting requirements so the year 1996 was a good starting point for this sample. The sample also eliminates any firms who had a previous alliance or made an acquisition within a three-year window before and after the alliance to ensure clean estimates of Beta. The total sample includes 407 alliances, after removing 111 firms which either had missing financial information or had an alliance within a three-year period before or after the date of this alliance announcement. Descriptive statistics of the sample of the 111 alliances that were excluded from the statistical analyses were not statistically different from the sample of 407 alliances used in the statistical analyses. The study includes alliances in from several industries in manufacturing: software, computer manufacturers, electronics, communications, defense, food, tobacco, textile, wood, paper, rubber, leather, stone, and metal. The data mainly come from SDC’s Alliance/JV Database. Anand and Khanna (2000a) find that SDC data on alliance types and alliance SIC codes is quite accurate based of their extensive comparison of SDC data to data from non-SDC sources. However, Anand and Khanna (2000a) do find that in many cases the date of the alliance is misstated and thus alliance dates were verified using additional sources such as Lexis Nexis, newspapers, and trade publications. Stock price data was obtained from CRSP and firm-specific balance sheet and income statement data was obtained from COMPUSTAT and verified using Annual Reports. Partner similarity represents the level of business commonality between the participants in the alliance. The sum the primary three-digit SIC codes in common among the members of the alliance is divided by the total number of members in the alliance. Alliances among members that operate in diverse businesses may have more benefit for systematic risk reduction. Methodology To estimate the change in Beta for each firm in the alliance I estimated the firm’s Beta from a standard asset pricing model normally used to predict returns (Brown and Warner, 1985). Change in Beta is the difference between the post-alliance Beta of the alliance partner and the pre-alliance Beta of the alliance partner. Daily data for each publicly listed firm was used. It has been conventional in finance to compute Beta based on a time-period of 254 days, approximately one year of trading days. The one-year post-alliance Beta is estimated based on days +60 to +314 and the pre-alliance Beta is estimated based on days -314 to -60, with day 0 being the day of alliance announcement (two-year Betas are calculated using days -568 to -60 and +60 to +568). Given that this study examines changes in Betas due to alliance announcements, it omits a period of -60 to +60 days around the day of alliance announcement. This is to avoid the inclusion of volatile periods of adjustment surrounding the day of alliance announcement and exclude temporary reactions in the measurement of Beta. The Center for Research in Security Prices’ (CRSP) value-weighted index with dividends was used to represent the market return. The return on security i for day t, Rit, is: Rit = (Pit - Pit-1 )/ Pit-1 (1) where Pit represents the price of security i, for day t. The Beta of stock i, Bi, is: Bi = ∑ t =1 ..253 {(Rit - Ravgit) (Rmt - Ravgmt)}/ ∑ t =1 ..253 (Rmt-Ravgmt)2 (2) where Rmt represents the return on the market portfolio for day t, Rit represents the return on security i on day t, and Ravg represents the average of the returns on security i for days 1 through t. However, to allow for continuous compounding when calculating Beta, ln(1+R) is used in place of R. 123 Studies in the finance literature have demonstrated that traditional estimates of Beta can be biased (Scholes and Williams, 1977; Dimson, 1979). Many securities are traded infrequently and few securities are traded so actively that their prices are recorded almost continuously. As a result, the use of daily returns to estimate Betas introduces the possibility of downward bias in Beta estimates. Scholes and Williams (1977) introduced a method to correct for nonsynchronous securities data, and this study uses their estimation method. Beta is estimated by running a regression of the stock's return on the return on the market index, Rit = α + BKitMit+k + e k = -1, 0, 1 (3) where R represents the return on security i on day t, BKit represents the estimated Beta for security i, and Mit+k represents the CRSP value-weighted index with dividends. The corrected Beta, BC, is the sum of the Beta coefficients estimated by a synchronous, lagged, and leading market return. BC = ∑ BC / (1+2ρ) (4) The corrected (hereafter Scholes-adjusted) Beta is then divided by one plus twice the autocorrelation, ρ, of the market returns. RESULTS Table 1 shows, the descriptive statistics for 407 alliances in the study. The average difference for Beta for the alliances was -0.24 and -0.37 for the one-year and two-year Scholes-adjusted Betas, respectively. Furthermore, 57% of firms experienced a decrease in Scholes-adjusted one-year Betas and 62% of firms experienced a decrease in Scholes-adjusted two-year Betas. --------------------------------Insert Table 1 about here. --------------------------------A standard test statistic is used in determining the statistical significance of whether the change in Beta is not equal to zero. The key test statistic for this study is the difference score which is the change in one-year and two-year Beta surrounding the alliance announcement. If the alliance causes a change in the firms’ systematic risk the change in Beta should be different than zero and this difference should be statistically significant. Table 2 shows the results of these tests. --------------------------------Insert Table 2 about here. --------------------------------Hypothesis 1 predicted that alliances decrease firms’ systematic risk and the results show that the mean differences for the changes in one-year and two-year Scholes-adjusted Betas were each negative, and statistically and significantly different from zero (p < 0.001 for one-year and two-year Betas). Hypotheses 2a to 2e predicted that each of the different alliance types would decrease firms’ systematic risk. Each of the average differences for one-year and two-year Betas for each of the alliance types are negative, statistically significant and different from zero. Furthermore, the average difference in Beta for all alliance types indicate that partners experience a greater decrease in systematic risk for participating in the specific type of alliance than for not participating in the specific type of alliance. 124 More specifically, the results support hypothesis 2a, where participating in a cross border alliance results in a decrease in systematic risk. Furthermore, the difference coefficients for engaging in a cross border alliance show a mean change that is large, negative and statistically different than zero. Furthermore, engaging in a cross border alliance results in a sizable decrease in systematic risk and this decrease is statistically significant from the decrease in systematic risk that occurs from not engaging in a cross border alliance for one-year Betas (p < 0.01). The results also support hypothesis 2b, where participating in a licensing alliance results in a decrease in systematic risk. All the difference coefficients for engaging in a licensing alliance show a mean change that is large, negative and statistically different than zero. Furthermore, engaging in a licensing alliance results in a sizable decrease in systematic risk and this decrease is statistically significant from the decrease in systematic risk that occurs from not engaging in a licensing alliance for one-year Betas and two-year Betas (p < 0.01, and p < 0.05). The results also support hypothesis 2c, where participating in a manufacturing alliance results in a decrease in systematic risk. All the difference coefficients for engaging in a manufacturing alliance show a mean change that is large, negative and statistically different than zero. However engaging in a manufacturing alliance does not result is a statistically distinguishable decrease in systematic risk from the decrease in systematic risk that occurs from not engaging in a manufacturing alliance. The results also support hypothesis 2d, where participating in a marketing alliance results in a decrease in systematic risk. All the difference coefficients for engaging in a marketing alliance show a mean change that is large, negative and statistically different than zero. Furthermore, engaging in a marketing alliance results in a sizable decrease in systematic risk and this decrease is statistically significant from the decrease in systematic risk that occurs from not engaging in a marketing alliance for one-year Betas (p < 0.10). The results also support hypothesis 2e, where participating in a research and development alliance results in a decrease in systematic risk. All the difference coefficients for engaging in a research and development alliance show a mean change that is large, negative and statistically different than zero. Furthermore, engaging in a research and development alliance results in a sizable decrease in systematic risk and this decrease is statistically significant from the decrease in systematic risk that occurs from not engaging in a research and development alliance for two-year Betas (p < 0.01). The results also support hypothesis 2f, where participating in a supply side alliance results in a decrease in systematic risk. All the difference coefficients for engaging in a supply side alliance show a mean change that is large, negative and statistically different than zero. Furthermore, engaging in a supply side alliance results in a sizable decrease in systematic risk and this decrease is statistically significant from the decrease in systematic risk that occurs from not engaging in a supply side alliance for one-year Betas and two-year Betas (p < 0.001, and p < 0.001). The results also support hypothesis 2g, where participating in a technology transfer alliance results in a decrease in systematic risk. All the difference coefficients for engaging in a technology transfer alliance show a mean change that is large, negative and statistically different than zero. However engaging in a technology transfer alliance does not result is a statistically distinguishable decrease in systematic risk from the decrease in systematic risk that occurs from not engaging in a technology transfer alliance. Hypothesis 3 predicted that risk reduction is greater in cross border, R&D, marketing, manufacturing, supply side, and technology transfer alliances than for licensing alliances. 125 Hypothesis 3 was not supported as none of the differences of decreases in Beta between the alliances types are statistically distinguishable from each other. Hypothesis 4 predicted that alliances among partners in unrelated businesses will decrease firms’ risk more than alliances among related businesses. The results show that the mean differences for the changes in one-year and two-year Scholes-adjusted Betas were negative and different from zero and these differences were each statistically significant. Furthermore a statistical test for the difference between two means shows that the mean for alliances among partners with business commonality and the mean for alliances among partners with no business commonality for two-year Betas are statistically different from each other (p < 0.10). Discussion In summary, the findings indicate that on average alliances decrease firms’ systematic risk. These findings hold true whether testing average one-year or two-year Betas. These findings suggest that alliances can indeed help firms minimize their risk relative to the market portfolio. Alliances may provide opportunities for firms to reduce their vulnerability to uncertainty from their environment, share risks with other firms, and pursue advantageous opportunities with lower exit costs. By minimizing firms’ systematic risk alliances have the potential to be enduring value-creating events with risk reduction occurring for a period of two years subsequent to the alliance. Furthermore, the findings also show that different types of alliances reduce firms’ systematic risk and facilitate different levels of risk reduction. Specifically, the average difference in Beta for each alliance type indicates that partners experience a greater decrease in systematic risk for participating in each of the specific alliance types than for not participating in the specific alliance type. Finally, firms experience greater risk reduction when they ally with partners in unrelated businesses than when they ally with partners in related businesses, indicating that risk reduction is more likely when firms diversify their resources. Overall, the results indicate that alliances can reduce firms’ systematic risk. This study also has some limitations that if acknowledged can improve the applicability of the results. One potential limitation is that the study does not consider the duration of the alliances that were included or account for informal ties, which may also be important when examining changes in Beta due to alliances (Coleman, 1990). One issue with using data on the duration of alliances or informal ties is that data on the duration of the alliances and informal ties is not regularly available for all alliances. Thus including such data would create bias in the results because the study would include only the few alliances that accurately report alliance duration and informal ties. Second while the study tries to limit the influence of other events on the Beta estimation by controlling for previous alliances and acquisitions, there may be events that are not reported by firms that can influence Beta estimates. Third, the study does not consider the quality of the alliance relationship and its effect on Beta, nor does it consider the impact of the alliance on the firm’s total operations. For example, trust between partners, previous ties and many other factors that have been linked to the quality of alliance relationships may play a role in Beta changes. This may be a venue that future studies may further explore. Lastly, the results are specific to alliances by firms in manufacturing and these results may or may not replicate for alliances by firms in services or other industries. In all the study contributes to research on alliances by finding that alliances decrease firms’ systematic risk. First, the results of this study indicate that alliances can create value by reducing firms’ riskiness, and the risk-reducing benefit of alliances extends beyond a short event window surrounding announcement. The study shows that risk reduction benefits continue for at least two-years beyond alliance announcement. Second, managers’ strategic decisions such as the decision to form strategic alliances can influence a firm’s systematic risk. Management theory may 126 not entirely diverge from financial theory as systematic risk may be managed by strategic choices such as alliances. Finally, relatedness can also play a role in alliance benefits and the relatedness of alliance partners ought to be a consideration for managers and may warrant further investigation. On a more practical level, firms’ systematic risk reduction is an important factor for both managers and investors. First it is important for managers to be aware of their potential to decrease firms’ systematic risk via alliances and how reducing risk can benefit a firm. Risk is an important attribute of a public firm. For example, firms that investors perceive as being risky incur higher costs when raising capital. Additionally, riskier firms sell equity at lower prices or pay higher prices when incurring debt and borrowing at higher rates may limit a firm’s ability to pursue some of its opportunities. Higher costs of capital may put firms at a competitive disadvantage when compared with competitors with lower costs of capital. Thus systematic risk can have an adverse effect on a firm’s strategic opportunity set (Copeland & Weston, 1991). Furthermore because higher risk impacts firms’ cost of capital, this can lead to valuations that discount future earnings more severely than they would be discounted at lower costs of capital, which may ultimately place more emphasis on firms’ short-term earnings. Ultimately systematic risk can impact firm value, as the lower is the systematic risk of the firm the lower the required rate of return and the higher the value of the firm (Van Horne, 1980: 68). From a managerial perspective, firms with lower systematic risk present less managerial risk for firms’ managers. In conclusion it seems beneficial for firms to pursue strategic actions that lower their systematic risk and thus managers ought to pursue actions that lower the level of their firm’s risk in the eyes of the financial community. REFERENCES Amburgey, T. L. & Dacin, T. 1996. As the Left Foot Follows the Right? The Dynamics of Strategic and Structural Change. Academy of Management Journal, 37: 6: 1427-1452. Amit, R. & Livnat, J. 1988. Diversification, capital structure and systematic risk: An empirical investigation. Journal of Accounting, Auditing, and Finance, 3:1: 19-43. Anand, B. N., & Khanna, T. 2000a. Do firms learn to create value? The case of alliances. Strategic Management Journal, 21: 295-315. Anand, B. N., & Khanna, T. 2000b. The structure of licensing contracts. Journal of Industrial Economics, 48: 103-135. Arino, A., & de la Torre, J. 1998. Learning from failure: Towards an evolutionary model of collaborative ventures. Organization Science, 9: 306-325. Badaracco, J. 1991. The knowledge link: How firms compete through strategic alliances. Cambridge, MA: Harvard Business School Press. Balakrishnan, S., & Koza, M. P. 1993. Information asymmetry, adverse selection and joint ventures: Theory and evidence. Journal of Economic Behavior and Organization, 20: 99-117. Ball, J. 1999. To define the future car, GM, Toyota say bigger is better, The Wall Street Journal, April 20, B4. Bettis, R. A. 1983. Modern financial theory, corporate strategy, and public policy: Three conundrums. Academy of Management Review, 8: 549-563. Bettis, R. A. & Hall, W. K. 1983. Diversification strategy, accounting determined risk, and accounting determined return. Academy of Management Journal, 45: 2: 254-264. Booz, Allen, & Hamilton. 2003. Alliance Strategy. Brown, S. J., & J. B. Warner. 1985. Using daily stock returns. Journal of Financial Economics, 14: 3-31. Caves, R., Crookell, H., & Killing, J. P. 1983. The imperfect market for technology licenses. Oxford Bulletin of Economics and Statistics, 45: 249-267. Chan, S., Kensinger, J., Keown, A., & Martin, J. 1997. Do strategic alliances create value? Journal of Financial Economics, 46, 199-221. 127 Chatterjee, S. & Lubatkin, M. 1990. Corporate mergers, stockholder diversification, and changes in systematic risk. Strategic Management Journal, 11: 4: 255-268. Chatterjee, S. 1986. Types of synergy and economic value: The impact of acquisitions on merging and rival firms. Strategic Management Journal, 7, 119-139. Chatterjee, S., Lubatkin, M. H., & Schulze, W.S. 1999. Toward a strategic theory of risk premium: Moving beyond CAPM. Academy of Management Review, 24: 3: 556-567. Cohen, W., & Levinthal, D. 1990. Absorptive capacity: A new perspective on learning and innovation. Administrative Science Quarterly, 35: 128-152. Coleman, J. 1990. Foundations of social theory. Cambridge, MA: Harvard University Press. Copeland, T. E., & Weston, J. F. 1991. Financial theory and corporate policy. Reading, MA: Addison-Wesley. Copeland, T. E., & Weston, J. F. 1991. Financial theory and corporate policy. Reading, MA: Addison-Wesley. Das, S., Sen, P., & Sengupta, S. 1998. Impact of strategic alliances on firm valuation. Academy of Management Journal, 41, 1: 27-41. Dimson, E. 1979. Risk measurement when shares are subject to infrequent trading. Journal of Financial Economics, 7, 197-226. Doz, Y. 1996. Specialized supplier networks as a source of competitive advantage: Evidence from the auto industry. Strategic Management Journal, 17: 271-292. Dussage, P., Garrette, B., & Mitchell, W. 2000. Learning from competing partners: Outcomes and durations of scale and link alliances in Europe, North America, and Asia. Strategic Management Journal, 21: 99-126. Dyer, J. H., & Singh, H. 1998. The relational view: cooperative strategy and sources of interorganizational competitive advantage." Academy of Management Review 23: 660-679. Gommes-Casseres, B. 2000, May 9. Alliances and risk: Securing a place in the victory parade. Financial Times, pg 6-7. Grant, R. M. 1996. A knowledge based theory of inter-firm collaboration. Organization Science, 7: 375-387. Hamel, G. 1991 Competition for competence and inter-partner learning within international alliances. Strategic Management Journal, 12: 83-103. Hamel, G. 1991. Competition for competence and inter-partner learning within international strategic alliances. Strategic Management Journal, Summer Special Issue, 12: 83-103. Haspeslagh, P.C. & Jemison, D. B. 1991. Managing Acquisitions. Free Press, New York. Hitt, M. A., Ireland, R. D., & Hoskisson, R. E. 2003. Strategic management: Competitiveness and globalization, 6th edition, South Western. Kale, P., & Singh, H. 1999. Building alliance capabilities: A knowledge-based approach. Academy of Management Best Paper Proceedings, Chicago, IL. Kale, P., Singh, H., & Perlmutter, H. 2000. Learning and protection of proprietary assets in strategic alliances: Building relational capital. Strategic Management Journal, 21: 217-237. Khanna, T. R., Gulati, R., & Nohria, N. 1998. The dynamics of learning alliances: Competition, cooperation and relative scope. Strategic Management Journal, 19: 193-210. Kogut, B. 1988a. A study of the life cycle of joint ventures. In F. K. Contractor & P. Lorange (Eds.), Cooperative strategies in international business: 169-186. Lexington, MA: Lexington Books. Kogut, B. 1988b. Joint ventures: Theoretical and empirical perspectives. Strategic Management Journal, 9:4: 319-332. Lane, P. J., & Lubatkin, M. 1998. Relative absorptive capacity and interorganizational learning. Strategic Management Journal, 19: 461-477. Lane, P. J., & Lubatkin, M. 1998. Relative absorptive capacity and interorganizational learning. Strategic Management Journal, 19: 461-477. 128 Lubatkin, M. 1983. Mergers and performance of the acquiring firm. Academy of Management Review, 8: 218-225. Lyles, M. A. 1988 'Learning among joint venture-sophisticated firms.' In F. K. Contractor and P. Lorange (eds.), Cooperative Strategies in International Business. Lexington Books, Lexington, MA, pp. 301-316. Markowitz, H. 1952. Portfolio selection. Journal of Finance, 7: 77-91. McConnell, J., & Nantel, T. 1985. Corporate combinations and common stock returns: The case of joint ventures. Journal of Finance, 40: 2: 519-536. Michel, A. & Shaked, M. 1984. Does business diversification affect performance? Financial Management, 13: 4: 18-25. Mowery, D. C., J. E. Oxley & Silverman, B. S. 1996. An empirical investigation of joint venture dynamics: Evidence from US-Japan joint ventures. International Journal of Industrial Organization, 14: 521-541. Neill, J. D., Pfeiffer, G. M., & Young-Ybarra, C. E. 2001. Technology R&D alliances and firm value. Journal of High Technology Management Research, 12: 227-237. Parr, R., & Sullivan, P. 1988. Technology licensing: Corporate strategies for maximizing value. New York: Wiley. Porter, M. 1985. Competitive advantage. Free Press, New York. Porter, M., & Fuller, M. 1986. Coalitions and global strategy. In: M. Porter (Ed.), Competition in global industries (pp. 315-344). Boston, MA: Harvard Business School Press. Salter, M. & Weinhold, W. 1979. Diversification through acquisition. Free Press, New York. Scholes, M., & Williams, J. 1977. Estimating betas from nonsynchronous data. Journal of Financial Economics. 5, 309-327. Singh, H., & Montgomery, C. 1987. Corporate acquisition strategies and economic performance. Strategic Management Journal, 8, 377-386. Teece, D. J. 1992. Competition, cooperation and innovation. Journal of Economic Behavior and Organization, 3: 537-556. Van Horne, J. C. Financial management and policy, 5th Edn, Prentice-Hall, Englewood Cliffs, NJ, 1980. Volberda, H.W. 1996. Toward the flexible form: How to remain vital in hypercompetitive environments. Organization Science, 7: 4: 359-374. www.Accenture.com, Accenture, 1999. 129 Table 1. Descriptive Statistics Pre-alliance Scholes-Adjusted 1-year Beta Post-alliance Scholes-Adjusted 1-year Beta Difference Scholes-Adjusted 1-year Beta Percentage Decrease in 1-year Beta Pre-alliance ScholesAdjusted 2-year Beta Post-alliance ScholesAdjusted 2-year Beta Difference ScholesAdjusted 2-year Beta Percentage Decrease in 2-year Beta All Alliances Mean Median Variance Total alliances 1.46 1.18 1.25 407 1.23 1.13 0.88 407 -0.24 -0.20 1.23 407 -16.71% -16.74% 1.52 1.28 1.17 407 1.17 1.05 0.75 407 -0.37 -0.26 1.16 407 -24.58% -20.38% Basic-tech Alliances Mean Median Variance Total alliances 1.22 0.98 1.27 117 0.89 0.84 0.43 117 -0.33 -0.17 1.16 117 -27.09% -17.42% 1.24 1.08 1.17 117 0.84 0.81 0.30 117 -0.41 -0.18 1.14 117 -33.06% -16.96% High-tech Alliances Mean Median Variance Total alliances 1.56 1.30 1.22 290 1.37 1.27 1.00 290 -0.21 -0.20 1.26 290 -13.44% -15.25% 1.63 1.44 1.14 290 1.30 1.19 0.87 290 -0.36 -0.27 1.17 290 -22.00% -18.99% 130 Table 2. T tests of Difference in One-Year and Two-Year Scholes-Adjusted Betas Difference in Scholes-Adjusted 1-year Beta Difference in Scholes-Adjusted 2-year Beta Variance Scholes-Adjusted 1-year Beta Variance Scholes-Adjusted 2-year Beta Number Included All firms -0.24*** -0.37*** 1.23 1.16 407 Partner Business Commonality No -0.27*** -0.41*** 1.21 1.10 290 Partner Business Commonality Yes -0.18* -0.29** 1.30 1.33 117 Cross Border No -0.24*** -0.37*** 1.24 1.18 397 Cross Border Yes -0.48* -0.45* 0.71 0.69 10 Licensing Agreement No -0.16** -0.31*** 1.05 0.95 255 Licensing Agreement Yes -0.39*** -0.48*** 1.51 1.51 152 Manufacturing Agreement No -0.21*** -0.36*** 1.27 1.17 277 Manufacturing Agreement Yes -0.31*** -0.4*** 1.15 1.16 130 Marketing Agreement No -0.21*** -0.36*** 1.30 1.25 309 Marketing Agreement Yes -0.34*** -0.42*** 0.99 0.89 98 Research and Development No -0.24*** -0.35*** 1.21 1.16 367 Research and Development Yes -0.33* -0.55*** 1.42 1.14 40 Supply Agreement No -0.24*** -0.37*** 1.24 1.17 398 Supply Agreement Yes -0.53** -0.63** 0.36 0.67 9 Technology Transfer No -0.24*** -0.36*** 1.25 1.17 332 Technology Transfer Yes -0.26* -0.43*** 1.16 1.15 75 † p < .10 , * p < .05, ** p < .01, *** p < .001 131 Difference in 1-Year Betas Between Participants and Non-Participants Difference in 2-Year Betas Between Participants and Non-Participants -0.09 -0.12† 0.24** 0.08 0.23** 0.17* 0.1 0.04 0.13† 0.06 0.09 0.20** 0.29*** 0.26*** 0.02 0.07 ASAC 2005 Toronto, Ontario Patrizia Porrini Long Island University ARE INVESTMENT BANKERS TO BLAME FOR ACQUISITION PREMIUMS? IF ACQUIRING PROCEED WITH CAUTION! Since the 1980s, the number of acquisition deals has increased as has the use of investment bankers. Among their many strategic and financial advisory roles, investment bankers negotiate acquisition premiums. Studies have found support for agency conflicts between acquirers’ bankers and their acquiring clients, resulting in bankers being associated with acquirers’ payments of higher acquisition premiums. This study investigates the transaction-specific attributes under which acquirers’ bankers are associated with greater acquisition premiums and whether bankers’ first-tier status attenuates the premium paid by acquirers when they use bankers. The study examines a sample of 481 acquisitions completed between 1988 and 1998 and finds the presence of acquirers’ bankers along with transaction-specific attributes can lead acquirers to pay greater acquisition premiums. These findings have implications for theory and practice. Since the 1980s, the number of acquisition deals has increased, as has the use of investment bankers. Between 1981 and 2002, the use of investment bankers on acquisitions went from 78% in 1981 to a peak of 88% in 1999. Among their many strategic and financial advisory roles, investment bankers negotiate acquisition premiums. An acquisition premium is the percentage increase between the share price the acquirer pays for the target and the target's share price four weeks before the acquisition announcement. Between 1990 and 2002, acquisition premiums averaged 53.2% with 40% of acquirers paying premiums of over 50% and 10% of acquirers paying premiums over 100%. If markets are efficient and investors and managers have all publicly available information then any premium that is above what the acquirer needs to pay in order to secure the deal is likely too much. Through their advisory roles, investment bankers can influence premiums as they advise acquirers and targets during the acquisition process. Specifically, targets hire bankers to get the highest premium for their firm, while acquirers hire bankers to pay the lowest premium for the target. However, fee structures motivate acquirers’ bankers to increase acquisition premiums because bankers’ fees are a percentage of deal size, and deal size increases with acquisition premium. Acquirers’ goals of minimizing premiums inherently conflict with acquirers’ bankers’ goals, which include increasing their payoff. Accordingly, it is not surprising that researchers have expressed skepticism towards acquirers’ investment bankers’ advisory roles on acquisition transactions (Kesner, Sharma, and Shapiro 1994; Lax and Sebenius, 1986; McLaughlin; 1990). Studies have found support for agency conflicts between acquirers’ bankers and their acquiring clients. For example, Kesner, Sharma and Shapiro (1994) find that premiums correlate positively with bankers’ fees—for both targets and acquirers, leading them to the conclusion that there is misalignment between acquirers’ bankers’ interests and acquirers’ interests. Furthermore, studies that examine whether the presence of investment bankers influence performance have also found that cumulative abnormal returns to acquirers that hire bankers are lower than those to acquirers that use in-house acquisition teams (Servaes and Zenner, 1996). Another concern of researchers has 132 been that overpayment can have serious implications on acquisition performance. For example, studies that have correlated variables with acquisition performance have found a statistically significant negative correlation between premium and performance (Lubatkin, 1983; Sirower, 1997; Varaiya and Ferris, 1987). In all, acquisition premiums represent difficult financial hurdles for acquirers to overcome given the many other pressures on performance in the post-acquisition phase. Large premiums can take cash away from other potentially profitable projects and raising debt to cover premiums places unnecessary burden on the combined firm post-acquisition. Acquisition premiums are important aspects of deals thus it is important to investigate the conditions under which bankers can cause acquirers to pay greater acquisition premiums. This study re-examines prior findings that there is misalignment of interests between acquirers’ bankers and acquirers by testing whether the presence of acquirers’ bankers correlates positively with acquisition premiums. Furthermore, the study extends this premise by exploring which transaction-specific attributes allow experts such as investment bankers to inflate premiums most and also explores whether bankers’ first-tier status attenuates the premium paid by acquirers when using bankers. The study contributes to existing literature by exploring whether clients’ reliance on experts and experts’ influence is greater under conditions of transactional uncertainty. Servaes and Zenner (1996) find that firms are more likely to hire bankers in situations where there is greater transactional ambiguity and information asymmetry. However, from their findings, it is not clear to what extent bankers along with transaction-specific attributes play a role in inflating premiums. The study also contributes to existing literature by further investigating the role of professional experts as advocated by Hayward (2003: 797). It is of importance for researchers and practitioners to know whether and when investment bankers have most influence over acquisition premiums. The next section discusses hypotheses regarding acquirers’ investment bankers, transaction attributes, and their effect on acquisition premiums. Ensuing sections describe methodology and results. The final section draws conclusions and discusses implications. THEORY AND HYPOTHESES Acquirers and targets often employ investment bankers for their expert knowledge about acquisition transactions, skills in structuring acquisitions, access to potential buyers and sellers, strategic advice, ability to negotiate premiums, or because they lack the resources to complete deals in-house (Benston and Smith, 1976; Hayward, 2003; Hunter and Walker, 1990; Smith 1986; Titman and Trueman, 1986). Investment bankers can provide increased legitimacy and signal deal quality to stakeholders by their presence (DiMaggio and Powell, 1983; Meyer, 1994; Meyer and Rowan, 1977; Pfeffer, 1981; Zucker, 1977). In all, clients use bankers to gain professional advice and improve performance from their decisions (Eccles and Crane, 1988; Hayward, 2003; Maister, 1993). However, researchers have found that there is misalignment between acquirers’ bankers' interests and acquirers' interests (Kesner, Sharma, and Shapiro, 1994). Researchers have posited that fee structures and asymmetric information may be reasons for misaligned interests between acquirers and their bankers (Kesner, Sharma, and Shapiro, 1994; Hayward, 2003; Servaes and Zenner, 1996). For example, bankers’ fees are based on deal size, which increases as premiums increase. Thus, acquirers’ bankers are motivated to increase premiums in spite of their clients’ interests of paying less for targets. Furthermore, bankers’ payoffs and reputations are not influenced by the long-term performance of the deals they advise. If extant theory and research holds true, acquirers’ bankers should correlate positively with acquisition premiums as payment structures and asymmetries in expertise and information create incentives for acquirers’ bankers to allow acquirers to pay more for targets to increase their own payoff. H1: Acquirers’ investment bankers will correlate positively with acquisition premiums. 133 Given bankers’ expert power, clients are more likely to rely on bankers’ expertise under conditions of informational and experiential disadvantage. Often transaction-specific attributes create the need for acquirers to use experts over their in-house acquisition teams. Accordingly, Servaes and Zenner (1996) find that firms are more likely to hire bankers in situations where there is greater transactional ambiguity and information asymmetry in the form of acquirers purchasing unrelated targets, targets that operate in many business areas, and in non-cash financed acquisitions. However, it is not clear to what extent these transaction-specific attributes play a role in increasing the potential for agency conflicts, resulting in acquirers paying higher premiums. Specifically, when acquirers are at an informational disadvantage, they are likely to be more reliant on bankers’ professional contributions (Servaes and Zenner, 1996). Similarly, in situations where there is greater transactional ambiguity, ambiguity about the value of targets, or ambiguity about financing acquisitions with stock, acquirers are more likely to rely on their bankers’ judgments than their own. In this regard, disparity in the form of information, knowledge and experience between bankers and acquirers, may give bankers expert power or undue influence over their clients. Acquirers’ may unquestioningly rely on bankers’ advice because of bankers’ expert or referent power (French and Raven, 1968; Pfeffer, 1981). Conversely, in situations where acquirers have greater information about valuing targets than do their bankers, they are more likely to consider their own judgment as well as or more than their bankers’ judgments. Consequently, if acquirers’ bankers’ interests are misaligned with those of their acquiring clients, the inflation in premiums should relate to transaction-specific attributes that cause information deficiency, which may increase acquirers’ reliance on bankers. Researchers have speculated that the degree of disparity in information, knowledge, or experience between acquirers and their bankers offers greater potential for bankers to act in their own best interests rather than acquirers’ best interests (Kesner, Sharma, and Shapiro, 1994; Servaes and Zenner, 1996). Acquirers routinely hire bankers for their expert advice and studies have shown that clients are likely to trust professional experts’ advice (Hayward, 2003). Thus in situations where there are greater disparities in information, knowledge, or experience acquirers are more likely to trust investment bankers. Relatedness of Targets The greater the degree of business commonality between acquirers and targets, the more information acquirers have about targets’ resources and value, and the less asymmetry between bankers and acquirers (Coff, 1999; Servaes and Zenner, 1996). Similarly, the greater the acquirer’s familiarity with the target’s business areas, the less is the acquirer’s needs to rely on experts’ opinions regarding the target’s value. For example, acquirers of targets in different areas of expertise may not be able to value the targets’ resources accurately and precisely (Servaes and Zenner, 1996). Thus, if acquirers’ bankers have misaligned interests, which results in acquirers paying higher premiums, buying a target in unrelated businesses creates the potential for actual or perceived informational asymmetry on the part of the acquirer. This actual or perceived informational disadvantage can increase the acquirer’s reliance on bankers’ advice regarding the premium necessary to secure the acquisition. H2: Acquirers will pay more for targets when they hire investment bankers and targets operate in unrelated businesses than when acquirers hire investment bankers and targets operate in related businesses. Targets’ Business Areas When targets operate in many business areas, it is likely more difficult and complex for acquirers to value targets (Servaes and Zenner, 1996). Targets that operate in many business areas have greater complexity in that acquirers must accurately asses the combined value of many business areas and it is less likely that the acquirer has detailed knowledge of the operations in all business areas. Accordingly, there is greater uncertainty with respect to evaluating the financial and strategic 134 benefits of such targets. Acquirers are more likely to use the professional valuation of investment bankers to assist their decision-making (Servaes and Zenner, 1996). Similarly, if acquirers’ bankers are associated with increasing the premiums paid by acquirers, the informational disadvantage of the acquirer that may result from valuing a target in multiple business areas creates the potential for bankers to inflate the premium paid for the target. H3: Acquirers will pay more for targets when they hire investment bankers and targets operate in many business areas than when acquirers hire investment bankers and targets operate in relatively fewer business areas. Non-Cash Financing Researchers have associated the need for investment bankers on acquisition transactions with non-cash financing. Stock-financed acquisitions are more complex transactions than cash-financed acquisitions transactions as they require greater use of bankers’ professional expertise (Brealey and Meyers, 1996; Hayward, 2003). Stock-financed acquisitions not only require valuation of the targets’ stock but also require valuation of the acquirer’s stock and determination of an exchange rate that convinces the target’s shareholders to give up their stock for a set portion of the acquirer’s stock. Exchanging stock also requires acquirers to choose whether they will set a fixed exchange rate or fixed value pricing (see Hayward (2003) for a more complete discussion). Because stock prices change, especially after acquisition announcements, it is more difficult to value stock-financed acquisitions than those using cash. Researchers have found that there is greater risk transfer from the acquirer to the target when there is stock financing rather than cash financing. Thus it is no surprise that studies have found that non-cash financing is positively associated with the need for investment bankers as banks can more extensively apply their expert knowledge of equity securities in stock financed acquisitions (Brealey and Meyers, 1996; Hayward, 2003; Servaes and Zenner, 1996). As non-cash financing requires the financial expertise of bankers, there is likely greater reliance on experts’ opinions. In situations where acquirers are better able to evaluate the financial aspects of funding the transaction they are less likely to rely exclusively on bankers’ advice. As such, if acquirers use bankers on non-cash acquisitions, they are more likely to rely on bankers professional assessments in non-cash deals, and if there is misalignment of interests, they are likely to pay a greater premium. H4: Acquirers will pay more for targets when they hire investment bankers and use non-cash financing, than when acquirers hire investment bankers and use cash to finance acquisitions. Intermediaries' First-Tier Rank Investment bankers’ first-tier status can affect their ability to influence acquirers’ decisions regarding acquisition premiums. Investment bankers’ first-tier status is determined from league tables which rank investment banks by number of deals or deal value1. For example, popular press sources often publish league tables and/or cite investment banks' relative standing and often the focus of attention is on the top five M&A advisory firms. League table standing is an important aspect of reputation and studies have found that wealth gains are greater to acquirers or targets that use first-tier bankers (Bowers and Miller, 1990). Investment bankers' reputations for advising success can signal negotiation ability and access to networks of targets', acquirers', investment banks, information that can be useful to the deal, financial expertise and in-house services that are useful to deal making. Acquirers and targets may associate greater professional expertise with bankers’ first-tier rankings. For example, Servaes and Zenner (1996) ran analyses of clients' likelihood to choose first-tier (top five firms in the league table) or second-tier investment bankers (firms other than top five) and found that first-tier bankers are more likely to be chosen over second-tier bankers with larger 1 Rankings only include the number of deals that were completed. 135 transactions completed by acquirers with little acquisition experience1. Acquisitions advised by first-tier intermediaries may receive more attention by investors and the intermediaries' peers and thus first-tier intermediaries ought to be more conscientious of the quality of their advice. Additionally, their limelight status may create incentives for top-tier intermediaries to preserve their reputations (Roberts and Dowling, 2002). Thus, although transaction-specific attributes may intensify the effect of misaligned interests on the part of bankers, when investment bankers' reputations are at stake, the incentive to uphold reputation may minimize agency conflicts. Their public status may make first-tier investment bankers more conscientious of their advisory roles. Thus, even though acquirers may associate greater expertise with first-tier bankers, acquirers’ first-tier bankers may be less likely to pursue self-interest and demonstrate their negotiation expertise by not recommending higher premiums at the expense of client interest. H5: The presence of acquirers' first-tier investment bankers correlate negatively with acquisition premium. METHOD The sample consists of public acquisitions of public targets in manufacturing completed between January 1, 1988 and December 31, 1998. The sample was limited to public acquirers and public targets so that financial data would be available on both firms. The total sample consisted of 607 acquisitions before the exclusion of 126 acquisitions due to missing data, leaving an effective sample of 481 acquisitions. Of the acquisitions removed, either targets or acquirers had missing data or acquirers were themselves acquired within one year of the acquisition. The data are from SDC’s Merger and Acquisition Database, Compustat, CRSP, Annual Reports, Lexis Nexis, and Datastream. Data were primarily collected using the SDC databases and then checked using two additional sources from those listed above. If there was a discrepancy, I used an additional source. The dependent variable, acquisition premium, is the percentage increase between the price paid for the target’s stock and the target's stock price four weeks before the acquisition announcement. Acquisition premium is a percentage representing the fraction of overpayment relative to the starting price and is an accurate way to capture overpayment (Cotter & Zenner, 1994; Hayward and Hambrick, 1997; Jerrell and Poulson, 1989; Jerrell, Brick, and Netter, 1988; Kesner, Shapiro & Sharma, 1994; Sharma, Shapiro, and Kesner, 1991). For example, a $5 million premium on a $50 million market value of equity is much different than a $5 million premium on a $10 million market value of equity. In the former, the premium is 10% of the market value of equity whereas in the latter the premium is 50% of the total equity value. The independent variables in the study focus on the use of investment bankers on acquisition transactions. Acquirer investment banker indicator is coded 1 if the acquirer used an investment banker and 0 otherwise (Servaes and Zenner, 1996)2. Target investment banker is similarly measured. When the investment banker variable takes the value of zero, acquirers or targets are likely using in-house advisory teams that perform due diligence and advisory roles that bankers usually perform. Thus, firms that use in-house counsel do not face the agency risks that firms that use outside advisors face.3 Further investigation revealed that none of the investment bankers included in this study served both sides of the same transaction. Acquirer first-tier investment banker indicator is coded 1 if any of the investment bankers used by the acquirer were among the top five bankers by value of the transaction in league tables for the year prior to the transaction year 1 This study controls for both of these factors. I also ran regressions using the number of investment bankers. The results were analogous to those presented here. 3 The average premium for the portion of deals in this sample where acquirers do not use a banker was 50.65 and the average premium for acquirers that used bankers was 51.02. These groups are not statistically different, and this lends support to the conclusion that the results are not spurious or endogenous in that it does not seem that bankers are used on deals that are likely to require bankers larger premiums to close deals. 2 136 (Bowers and Miller, 1990; Servaes and Zenner, 1996). Target first-tier investment banker is measured similarly.1 Transaction-Specific Variables SIC Relatedness between the acquirer and the target is the number of three-digit Standard Industrial Classification codes that the acquirer and the target have in common (Servaes and Zenner, 1996). The level of relatedness between acquirer and target can affect information asymmetry, the acquirer’s need to rely on bankers’ valuations, and the acquirer’s valuation of the target. Target number of SIC codes is a count variable representing the number of business areas that the target operates in. Acquiring, integrating, and valuing targets that operate in multiple business areas is more complex than acquiring targets that operate in fewer business areas (Servaes and Zenner, 1996). Method of payment is a variable coded 0 if a cash offer is made and 1 for any other form of payment such as cash and stock (Hayward and Hambrick, 1997; Servaes and Zenner, 1996). Non-cash financing is more likely to require the use of outside advisors. Acquirers who have enough cash on hand to buy the target may be more willing to pay a higher premium to prevent or in response to a bidding war because they do not have to finance the payment. Control Variables Value of the Transaction is the total dollar value of the transaction (Servaes and Zenner, 1996). Larger transactions may be more complex and may be more likely to require the help of bankers. Challenged deal is an indicator variable coded 1 if the acquisition was challenged by another bidder, and 0 otherwise. Bidding wars usually require the winning bidder to pay a larger premium to outbid other bidders (Varaiya, 1987). Attitude is an indicator variable coded 1 if the acquisition was friendly and 0 otherwise (hostile). Non-friendly acquisitions may lead targets to implement defense measures that cause higher premiums in an effort to prevent acquirers from buying them. Target is a high-tech firm is an indicator coded one if a target is a high-tech firm. Acquirers buying high-tech targets may have greater uncertainty about targets’ value as their resources are largely tacit as well as explicit, likely requiring bankers’ advice (Coff, 1999). Acquirer’s debt to assets ratio is the debt to assets ratio of the acquirer averaged over a three year period prior to the acquisition. Acquirers’ debt ratio may affect their willingness to pay large premiums as well as their ability to pursue non-cash financing. Defense measure is an indicator variable coded 1 if the target has defense measures in place and 0 otherwise. Targets employ defense measures such as poison pills and lockup options can make the target more expensive to acquirers (Varaiya, 1987). Defense measures have been associated with higher premiums. Acquirer's and target's acquisition experience is the number of acquisitions completed during the four-year period prior to the acquisition (Kesner, Shapiro & Sharma, 1994; Haleblian and Finkelstein, 1999; Hayward, 2003). Premiums may be affected by prior acquisition experiences. Furthermore acquirers’ and targets’ acquisition experience may affect their decision to hire bankers. Relative size of acquirer to target represents the ratio of the acquirers’ sales to targets’ sales (Hayward and Hambrick, 1997; Servaes and Zenner, 1996). Relative size may be an important aspect of whether firms choose to hire bankers. I controlled for macroeconomic effects by entering the years as a set of indicator variables. I created indicator variables for 10 of the 11 years in the sample, with 1998 as the omitted year. Sensitivity analysis revealed that omitting any other year would have produced similar results. The indicator variables were not statistically significant and the results were substantially unchanged when I removed them from the analysis. To keep the results simple, the year of acquisitions indicator variables are not included in the tables that follow. 1 Note that these variables function as interaction effects or moderators in that if there are no investment bankers, they take the value of zero, and if they are first-tier the variable takes the value of one. 137 RESULTS Table 1 shows the descriptive statistics and correlations among the variables for the acquisitions in this study. On average, the acquisition premium is 50.86, which is consistent with the average premium reported by other studies (Hayward and Hambrick, 1997; Varaiya and Ferris, 1987). I used ordinary linear regression to analyze the data and Table 2 shows the results of the regression analyses. Model 1 includes only control variables and yields an R-square of 9.2% (p < 0.002). Model 2 includes control variables and the independent variables representing whether the acquirer and the target used investment bankers and yields an R-square of 9.8% (p < 0.002). Hypothesis 1 predicted that acquirers’ bankers will correlate positively with acquisition premium and the coefficient was positive and statistically significant (p < 0.01). The actual value of the coefficient was 6.91, which is 13.6% of the average premium (50.86). ------------------------------Insert Table 1 about here. Insert Table 2 about here. ------------------------------Models 3 through 5 include variables that represent transaction-specific attributes.1 Hypothesis 2 predicted that acquirers pay less for targets when they hire investment bankers and targets operate in related businesses than when acquirers hire investment bankers and targets operate in unrelated businesses. The coefficient for this interaction effect is negative and statistically significant (p < 0.001) confirming that as acquirers hire investment bankers and targets have business commonality, premiums decrease. Hypothesis 3 predicted that acquirers pay more for targets when they hire investment bankers and targets operate in a larger number of businesses than when acquirers hire investment bankers and targets operate in relatively fewer businesses. The coefficient for H3 was positive and statistically significant (p < 0.10). Interestingly the interaction of targets’ investment bankers and targets’ number of SIC codes is positive and statistically significant as well. Hypothesis 4 predicted that acquirers pay more for targets when they hire investment bankers and acquirers use non-cash financing than when acquirers hire investment bankers and use cash to finance the acquisition. The coefficient for the interaction effect testing H4 was positive and statistically significant indicating that as acquirers hire investment bankers and use non-cash financing their premiums increase. Model 6 includes the first-tier variables. Hypothesis 5 predicted that the presence of first-tier investment bankers would correlate negatively with acquisition premium and the coefficient was not statistically significant. Surprisingly, the opposite is true in that when acquirers use investment bankers and those bankers have first-tier status, this further inflates premiums (p < 0.001). Exploratory Analyses As an experiment, I ran exploratory regressions including the presence of legal advisors. Research on principal-agent relationships has led researchers to advise the use of monitors to prevent abuse of power by agents (Bazerman, Neale, Valley, Zajac, and Kim, 1992; Eisenhardt, 1989; Sharma, 1997). Legal advisors are paid on an hourly basis so acquirers’ legal advisors have less incentive to recommend higher premiums however they too are experts and potential monitors, yet their presence may reduce adverse behavior by acquirers’ bankers. Acquirers’ and targets’ legal advisors are coded 1 if the acquirer used a legal advisor and 0 otherwise. The coefficient for acquirers’ legal advisor is –2.709 and is statistically significant (p < 0.01). Interestingly, the presence of acquirers’ legal advisors can lower the inflationary effect that acquirers’ bankers have on premiums. Discussion 1 I ran separate models to avoid bias in the results resulting from potential inter-correlations among the interaction variables, as they were all based on investment bankers’ presence. 138 The results of this study extends results of prior studies by finding that when acquirers do use bankers, transaction-specific attributes are associated with greater acquisition premiums and bankers’ first-tier status inflates the premium paid by acquirers. Specifically this study finds that acquirers’ investment bankers correlate positively with acquisition premiums, even when controlling for the presence of targets’ investment bankers. Controlling for targets’ investment bankers is important as targets hire bankers to increase the premium obtained by targets—however prior studies have not controlled for the presence of targets’ bankers even though both acquirers and targets hired bankers in 68.1% of the deals in this study. Furthermore, the results of this study support hypotheses that assert premiums are greater in situations where transaction-specific attributes create informational disadvantages for acquirers, relative to their bankers. Specifically, acquirers pay more for targets when they use investment bankers, and targets operate in unrelated businesses, targets operate in many business areas, and acquirers use non-cash financing to complete deals. Acquirers also pay more for targets when they use investment bankers and the bankers have first-tier status. However, based on the results of the exploratory analysis, it appears that the presence of expert monitors such as legal advisors, can alleviate the aforementioned inflammatory effects of bankers and transaction-specific attributes on acquisition premiums. This study extends current knowledge by finding that investment bankers and transaction-specific attributes have an inflammatory effect on acquisition premiums in two ways. First, Servaes and Zenner (1996) find that transaction-specific characteristics such as acquirers’ relatedness with targets, targets’ multiplicity of businesses, and type of acquisition financing are all positively associated with the choice to use investment bankers. However, the results of this study show that the use of bankers, combined with transaction-specific attributes, correlates with acquirers paying greater premiums. It appears that the same factors that cause informational disparities causing acquirers to use investment bankers in the first place are the factors leading acquirers to pay more for targets—even though acquirers hire bankers for their expertise and to negotiate better deals for them so that they ca pay less for targets. Second, Hayward (2003) concludes that as managers defer to professional advice, their firms’ decisions become more representative of professional biases. He states that by leading clients toward complex decisions that apply professional expertise, professional firms gain greater control over such decisions and clients may risk overestimating the efficacy of professional expertise. Thus, Hayward calls for more attention to the role of professional advisors on these decisions (2003: 797). This study both gives more attention to the role of professional advisors on acquisition decisions and also finds that in decisions where professional expertise is retained by acquirers, transaction-specific attributes appear to increase the premium paid when managers hire professionals such as bankers. Furthermore, when acquirers’ hire bankers and their bankers have first-tier status there in an inflationary effect on acquisition premium and this may lend support to Hayward’s theory that clients may overestimate the efficacy of professional expertise—especially as first-tier advisors are symbolically more experienced. Conclusively, conditions of informational disparity may increase acquirers’ reliance on experts and can be detrimental. This study has some limitations that should be acknowledged. First, the study examined a time-period where there were a large number of acquisitions. Perhaps high acquisition activity led to higher premiums becoming more acceptable as firms scrambled to maintain scale relative to competitors. Therefore, it is not clear that these findings would replicate in other time-periods. Second, the study includes only public deals as financial information is not available for non-public deals. Third, the study examines only completed deals. Although, it is difficult to investigate the effect of investment bankers on acquisitions for deals that were not completed, perhaps acquirers’ bankers’ roles on deals that were not completed is different from their role on completed deals, as bankers may have advised firms not to make the acquisitions. It is also important to recognize that 139 the findings apply to deals in manufacturing. Acquisitions in services may have different characteristics than those in manufacturing and may require different services from bankers. This study has several practical implications. For one, it is of value for investors and managers to be aware of how the presence of bankers and transaction-specific attributes can cause informational and experiential disparities for acquirers and may lead to inflated acquisition premiums. The use of professional experts in acquisition transactions is not necessarily beneficial, thus it is important for managers to question experts’ advice in order to decide what is best for their firm. It may also be beneficial for one-time or repeat acquirers to form an in-house acquisition team that can monitor, supplement, or substitute for investment bankers. Second, acquirers and targets can conduct their own due diligence to reduce informational disadvantages and thus reduce their reliance on bankers’ advice. One way would be for acquirers to form a strategic alliance with a target prior to acquiring so they can gather information and gain experience that can enable more objectivity about the benefits of the acquisition, more informed target choice, and more accurate target valuation. Third, the use of legal advisors as monitors or neutral third parties in acquisition negotiations may be very important because mediators can more objectively evaluate deal attributes leading to decisions that serve the long-term best interests of acquirers, targets, shareholder, and other stakeholders. These findings should also appease acquirers' and targets' concerns regarding the use of in-house acquisition teams. Some firms may be skeptical to use their own in-house acquisition team, thinking that they will be at a professional, informational, and experiential disadvantage but this may not be the case. In all, these findings should motivate researchers and managers to further examine the role of professional experts such as investment bankers and legal advisors on acquisition transactions. REFERENCES Bazerman, M. H., Neale, M. A., Valley, K. L., Zajac, E. J., and Y. M. Kim. 1992. The effects of agents and mediators on negotiation outcomes. Organizational Behavior and Human Decision Processes, 53: 55-73. Beatty, R. P., and J. R. Ritter. 1986. Investment banking, reputation, and the underpricing of initial public offerings. Journal of Financial Economics, 15: 213-232. Benston, G., and C. W. Smith. 1976. A transaction costs approach to the theory of financial intermediation. Journal of Finance, 31: 215-231. Bowers, H. M., and R. Miller. 1990. Choice of investment banker and shareholders wealth of firms involved in acquisitions. Financial Management, 19: 34-44. Brealey, R. A. and S. Meyers. 1996. Principles of Corporate Finance. McGraw-Hill: New York. Coff, R. W. 1999. How Buyers Cope with Uncertainty when Acquiring firms in Knowledge-intensive industries: Caveat emptor. Organization Science, 10: 2: 144-161. Cotter, J. F., and M. Zenner. 1994. How managerial wealth affects the tender offer process. Journal of Financial Economics, 35: 63-97. DiMaggio, P. J. and W. W. Powell. 1983. The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48: 147-160. Eccles, R. G. and D. B. Crane. 1988. Doing Deals: Investment Banks at Work. Harvard Business School Press: Boston, MA. Eisenhardt, K. M. 1989. Agency theory: An assessment and review. Academy of Management Journal, 14: 1: 57-74. French, J. R. P. and B. Raven. 1968. The bases of social power. In Group Dynamics, Cartwright D, Zander A (eds). Harper and Row: New York. Haleblian, J., and S. Finkelstein. 1999. The influence of organizational acquisition history on acquisition performance: A behavioral learning perspective. Administrative Science Quarterly, 44: 29-56. 140 Haunschild, P. 1994. How much is that company worth?: Interorganizational relationships, uncertainty, and acquisition premiums. Administrative Science Quarterly, 39: 3: 391-411. Hayward, M. L. A. 2003. When do firms learn from their acquisition experience? Evidence from 1990-1995. Strategic Management Journal, 23: 21-39. Hayward, M. L. A., and D. C. Hambrick. 1997. Explaining the premiums paid in large acquisitions: Evidence of CEO hubris. Administrative Science Quarterly, 42: 103-127. Hunter, W. C. and M. B. Walker. 1990. An empirical examination of investment banking merger fee contracts. Southern Economic Journal, 56, 1117-1130. Jerrell, G., and A. Poulson. 1989. Stock trading before the announcement of tender offers: Insider trading or market anticipation? Journal of Law, Economics and Organization, 5: 225-248. Jerrell, G., Brickley, J. A., and J. M. Netter. 1988. The market for corporate control: The empirical evidence since 1980. Journal of Economic Perspectives, 2: 49-68. Kerr, S. 1975. On the folly of rewarding A while hoping for B. Academy of Management Journal, 18: 4: 769-785. Kesner, I. F., Shapiro, D. L. and A. Sharma.1994. Brokering mergers: an agency theory perspective on the role of representatives. Academy of Management Journal, 37: 3: 703-721. Lax, D. A. and J. K. Sebenius. 1986. The manager as negotiator: Bargaining for cooperative and competitive gain. New York: Free Press. Lubatkin, M. 1983. Mergers and the performance of the acquiring firm. Academy of Management Review, 8: 218-225. Maister, D. 1997. Managing the professional service firm. Free Press: Copenhagen. McLaughlin, R. M. 1990. Investment banking contracts in tender offers: An empirical analysis. Journal of Financial Economics, 7: 209-231. Meyer, J. W. 1994. Rationalized environments. In Institutional Environments and Organizations: Structural Complexity and Individualism, Scott W. R., Meyer, J.W. (eds). Sage: Thousand Oaks, CA: 28-54. Meyer, J. W. and B. Rowan. 1977. Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 8: 340-363. Pfeffer, J. 1981. Power in Organizations. Ballinger: Cambridge, MA. Roberts, P. W., and G. R. Dowling. 2002 Corporate reputation and sustained superior financial performance. Strategic Management Journal, 23: 1077-1093. Servaes, H. and M. Zenner. 1996. The role of investment banks in acquisitions. The Review of Financial Studies, 9: 3: 787-815. Sharma, A. 1997. Professional as agent: knowledge asymmetry in agency exchange. Academy of Management Review, 22: 3: 758-798. Sharma, A., Shapiro, D. L., and I. F. Kesner. 1991. Targets or mergers: Applying a negotiations perspective to predict degree of resistance. International Journal of Conflict Management, 2:2:117-138. Sirower, M. L. 1997. The Synergy Trap, The Free Press, New York. Smith, C. W. 1986. Investment Banking and the Capital Acquisition Process. Journal of Financial Economics, 15: 3-29. Titman, S. and B. Trueman. 1986. Information quality and the valuation of new issues. Journal of Accounting and Economics, 8: 159-172. Varaiya, N. P., and K. R. Ferris. 1987. Overpaying in corporate takeovers: The winner’s curse. Financial Analysts' Journal, May-June: 64-70. Zucker, L. G. 1977. The role of institutionalization in cultural persistence. American Journal of Sociological Review, 42: 726-743. 141 Table 1 Descriptive Statistics and Correlations Among the Variables 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Value of Transaction Challenged Deal Friendly Acquisition Target is a High-Tech Firm Acquirer Debt/Assets Ratio-3yr Defense Measure Relative Size Acquirer's Acquisition Target's Acquisition Experience # 3-Digit SIC Codes in Common Target Number of SIC codes Cash Transaction Target Investment Banker (IB) Acquirer Investment Banker Target IB*Relatedness Acquirer IB*Relatedness Target IB*Cash Payment Acquirer IB*Cash Payment Target IB *Target # of SIC Codes Acquirer IB *Target # of SIC Target First-tier IB Acquirer First-tier IB Acquirer Legal Advisor Target Legal Advisor Premium * p < 0.05, ** p < 0.01 Mean 704.02 0.13 0.91 0.53 0.49 1.80 0.78 0.27 0.07 0.70 2.87 0.67 0.68 0.67 0.53 0.57 0.48 2.47 2.07 0.47 0.25 0.26 0.71 0.66 50.77 Std. 2470.18 0.33 0.35 0.50 0.20 0.86 4.23 0.66 0.26 0.46 2.33 0.47 0.47 0.47 0.50 0.50 0.50 2.47 2.51 0.50 0.43 0.51 0.46 0.48 42.67 N 481 481 481 481 481 481 481 481 481 481 481 481 481 481 481 481 481 481 481 481 481 481 481 481 481 1 1.00 -0.03 -0.05 -0.03 0.08* 0.01 0.03 0.09* 0.05 0.05 0.23** 0.07 0.14** 0.14** 0.13** 0.11** 0.14** 0.25** 0.28** 0.00 0.21** 0.27** 0.15** 0.15** -0.06 2 3 4 1.00 -0.30** -0.18** 0.07 -0.21** -0.02 -0.04 0.32** -0.11** 0.20** -0.13** -0.14** -0.15** -0.09* -0.11** -0.13** 0.10** 0.09* -0.16** -0.06 0.02 -0.12** -0.13** 0.14** 1.00 0.01 -0.06 0.23** 0.01 -0.02 -0.63** 0.12** -0.10** 0.14** 0.05 0.05 0.06 0.13** 0.10* -0.03 -0.07 -0.01 0.08* -0.06 0.07 0.03 -0.11** 1.00 -0.31** 0.01 0.00 0.08* -0.09* 0.14** -0.19** 0.15** 0.10* 0.06 0.14** 0.16** 0.16** -0.14** -0.11** 0.88** -0.02 -0.01 0.01 0.09* -0.03 142 5 6 1.00 -0.07 1.00 0.11** -0.03 -0.08 0.04 0.13** -0.30** -0.08* 0.15** 0.27** 0.06 -0.21** 0.74** -0.01 0.14** 0.04 0.17** -0.07 0.14** -0.16** 0.63** -0.12** 0.52** 0.24** 0.12** 0.20** 0.14** -0.26** 0.01 0.07 0.03 0.14** 0.06 0.07 0.12** 0.00 0.12** 0.07 -0.02 7 8 9 1.00 -0.03 0.00 -0.03 0.07 -0.04 0.00 -0.04 -0.03 -0.05 -0.06 0.06 0.06 0.00 -0.06 -0.05 -0.05 0.00 0.02 1.00 -0.01 0.12** 0.08* 0.10** 0.11** 0.07 0.15** 0.10** 0.13** 0.05 0.09* 0.10** 0.12** 0.04 0.09* 0.09* -0.02 1.00 -0.15** 0.14** -0.15** -0.15** -0.11** -0.11** -0.16** -0.13** 0.01 0.02 -0.06 -0.11** 0.05 -0.14** -0.13** 0.06 10 11 1.00 0.09* 1.00 0.06 -0.04 0.24** 0.10** 0.18** 0.05 0.69** 0.14** 0.15** 0.00 0.19** 0.05 0.17** 0.88** 0.20** 0.79** 0.16** -0.15** 0.12** 0.10** 0.12** 0.18** 0.23** 0.09* 0.24** 0.11** 0.03 0.01 Table 1 Descriptive Statistics and Correlations Among the Variables (continued) 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Cash Transaction Target Investment Banker (IB) Acquirer Investment Banker Target IB*Relatedness Acquirer IB*Relatedness Target IB*Cash Payment Acquirer IB*Cash Payment Target IB *Target # of SIC Codes Acquirer IB *Target # of SIC Target First-tier IB Acquirer First-tier IB Acquirer Legal Advisor Target Legal Advisor Premium * p < 0.05, ** p < 0.01 12 1.00 0.06 0.08* 0.10** 0.80** 0.67** -0.05 0.02 0.11** 0.03 -0.03 -0.01 0.04 -0.09* 13 14 15 16 17 18 19 20 1.00 0.53** 0.72** 0.28** 0.63** 0.29** 0.57** 0.22** 0.39** 0.29** 0.51** 0.95** 0.00 1.00 0.36** 0.31** 0.35** 0.26** 0.32** 0.20** 0.23** 0.35** 0.72** 0.55** 0.05 1.00 0.26** 0.51** 0.26** 0.47** 0.23** 0.31** 0.25** 0.38** 0.69** 0.00 1.00 0.75** 0.17** 0.15** 0.30** 0.12** 0.07 0.27** 0.33** -0.04 1.00 0.14** 0.34** 0.23** 0.25** 0.13** 0.29** 0.58** -0.10* 1.00 0.85** -0.02 0.17** 0.25** 0.32** 0.33** 0.03 1.00 -0.04 0.27** 0.30** 0.33** 0.55** 0.04 1.00 0.03 0.04 0.19** 0.27** 0.02 143 21 22 1.00 0.29** 1.00 0.22** 0.30** 0.36** 0.31** 0.01 0.07 23 24 25 1.00 0.57** 0.03 1.00 0.00 1.00 Intercept Value of Challenged Friendly Target is a Acquirer Defense Relative Size Acquirer's Target's # 3-Digit SIC Target Cash Target Acquirer Target Acquirer Target Acquirer Target IB Acquirer IB Target Acquirer Target Legal Acquirer Legal Model 1 Coeff. S.E. 57.911*** 5.363 -0.002*** 0.000 18.971*** 3.138 -12.150*** 3.119 -0.471 1.937 13.383** 5.055 1.737 5.192** -0.001** 0.000 -0.162 1.515 -1.219 4.885 2.547† 1.965 -0.487 0.447 -7.190** 2.990 R-Squared: 0.092 F-statistic: 2.215 p-value 0.002 Increase in † p < 0.10, * p < 0.05, Table 2 Results of Ordinary Linear Regression Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Coeff. S.E. Coeff. S.E. Coeff. S.E. Coeff. S.E. Coeff. S.E. Coeff. S.E. 47.676*** 6.116 40.765*** 6.367 53.692*** 6.427 51.429*** 6.184 50.614*** 6.207 49.127*** 6.364 -0.002*** 0.000 -0.002*** 0.000 -0.002*** 0.000 -0.002*** 0.000 -0.002*** 0.000 -0.002*** 0.000 17.877*** 3.215 17.896*** 3.210 16.891*** 3.220 18.162*** 3.209 17.522*** 3.219 18.981*** 3.248 -10.554*** 3.154 -10.320*** 3.153 -10.147*** 3.170 -10.976*** 3.151 -11.100*** 3.170 -11.313*** 3.172 -0.012 1.953 0.198 1.969 0.346 1.973 -16.733*** 4.294 0.062 1.975 -0.620 1.988 ** ** ** 13.714 5.074 13.977 5.067 13.473 5.078 14.792** 5.106 12.731** 5.092 14.647** 5.090 5.300** 1.737 5.340** 1.737 5.687*** 1.737 4.757** 1.740 4.838** 1.754 5.602*** 1.742 ** ** ** ** ** -0.001 0.000 -0.001 0.000 -0.001 0.000 -0.001 0.000 -0.001 0.000 -0.001** 0.000 -0.065 1.527 0.223 1.523 -0.183 1.523 -0.471 1.525 -0.079 1.528 -0.138 1.524 -2.483 4.904 -2.385 4.895 -2.329 4.963 -3.974 4.929 -4.160 4.953 -2.138 4.931 2.203 1.983 2.665† 1.994 2.098 1.995 1.689 1.985 2.059 1.985 2.645† 2.002 † † -0.482 0.447 -0.516 0.447 -1.868 1.187 -0.424 0.448 -0.583 0.449 -0.414 0.448 -7.361** 3.003 3.788 4.473 -8.338** 3.012 -5.994* 3.013 -6.849* 3.012 -8.231** 3.035 -0.178 2.307 9.958** 3.373 -9.484** 3.413 -5.081* 2.992 -1.729 2.441 0.968 2.387 ** ** *** 6.911 2.292 6.738 2.402 7.792 2.425 5.295* 2.338 5.446* 2.360 10.220*** 2.654 -0.161 3.622 -16.353*** 4.157 -1.697* 1.034 3.891*** 1.000 13.487*** 4.153 7.718* 4.104 2.018 2.290 5.267** 2.077 -1.596 3.161 -2.709** 1.054 0.098 2.076 0.002 0.070 ** p < 0.01, *** p < 0.001 0.106 2.060 0.002 0.147 0.105 2.048 0.002 0.141 144 0.106 2.078 0.002 0.155 0.102 1.978 0.003 0.106 0.102 1.973 0.003 0.103 ASAC 2005 Toronto, Ontario Laurent Renard Gilles St-Amant Département Management et Technologie Université du Québec à Montréal CAPACITÉ ORGANISATIONNELLE INTERNET ET CAPACITÉS DYNAMIQUES : LE CAS D’UNE ORGANISATION DE L’INDUSTRIE DU TOURISME AU CANADA On présente les résultats d’une recherche qualitative menée auprès d’une entreprise canadienne de l’industrie du tourisme qui a mis en œuvre différents projets de commerce électronique de type entreprise à consommateurs (B2C) et interentreprises (B2B), de 1998 à 2004, dans le cadre de sa stratégie Internet. L’objectif de la recherche est de comprendre le processus de création et de développement de la capacité organisationnelle Internet en caractérisant son cycle de vie (Helfat et Peteraf, 2003) et le rôle joué par les capacités dynamiques dans ce processus (Eisenhardt et Martin, 2000; Teece et al. 1997; Winter, 2000; Zollo et Winter, 2002). L’industrie du tourisme illustre très bien les changements induits par l’avènement des technologies Internet 1 . En effet, bien que le produit touristique final prenne la forme d’un assemblage de différents éléments, l’information y occupe un rôle de premier plan. Cette particularité a pour conséquence que les technologies Internet vont jouer un rôle important dans la transformation structurelle qui est observée dans l’industrie du tourisme canadienne (Buhalis, 2003; Werthner et Klein, 1999). Ces technologies Internet permettent l'adoption de nouveaux standards transactionnels autorisant des innovations à dominante technologique, commerciale, organisationnelle et institutionnelle. Toutefois, Porter (2001) rappelle que les technologies Internet en soi apportent rarement un avantage concurrentiel, mais que leur utilisation rend encore plus importante la nécessité d’avoir une stratégie Internet 2 . D’autre part, la mise en œuvre d’une stratégie Internet pose également le problème de la manière dont une organisation pourra intégrer ces technologies au cœur de ses activités. En d’autres termes, l’organisation devra posséder une capacité organisationnelle particulière que l’on appelle capacité organisationnelle Internet et que l’on définit comme étant une habileté pour déployer, combiner et coordonner des ressources (notamment les technologies Internet) et des compétences à travers différents flux de valeur (donc des processus de commerce électronique) pour mettre en œuvre sa stratégie Internet. Cependant, cette capacité organisationnelle Internet n’est pas présente la plupart du temps dans une organisation qui n’est pas née « point com », ce qui implique qu’elle devra être créée et développée. D’où la première question de recherche : comment une organisation qui œuvre dans l’industrie du 1 Les technologies Internet sont une sous-catégorie des technologies de l’information et de communication. Elles comprennent d’une part, un ensemble d’éléments techniques matériels et logiciels constitutifs des réseaux Internet, intranet, extranet et d’autre part, les applications informatiques qui utilisent ces réseaux pour faire transiter de l’information nécessaire aux activités d’affaires. 2 On considère la stratégie Internet tout autant sous l’angle de sa formation, de son implantation que de son évolution puisque les résultats obtenus, c’est-à-dire la stratégie réalisée et les retours d’expérience qui y sont associés, vont transformer la stratégie initiale. La stratégie Internet de Zêta est « tâtonnante » (Avenier, 1997) dans la mesure où elle repose sur la mise en œuvre de projets de commerce électronique qui sont des actions délibérées, qui reposent sur une rationalité procédurale et sur une dialectique continuelle entre les fins visées et les moyens utilisés au sein de situations émergentes qui imposent une révision constante de la stratégie initialement formée. 145 tourisme crée-t-elle et développe-t-elle sa capacité organisationnelle Internet? Ce processus de création et de développement de la capacité organisationnelle Internet nécessite la présence au sein de l’organisation de capacités dynamiques (Eisenhardt et Martin, 2000; Teece et al., 1997; Zollo et Winter, 2002). Les capacités dynamiques vont déterminer la façon dont l’organisation va changer et se transformer dans le temps étant donné sa dépendance de sentier et les positions de marché qu’elle occupe (Teece, et al., 1997). Elles vont permettre d’identifier et d’exploiter les opportunités d’affaires qui s’offrent à l’organisation en ce qui concerne l’utilisation des technologies Internet, d’une part, et de réaliser l’intégration, la reconfiguration et l’acquisition des ressources nécessaires à la création et au développement de la capacité organisationnelle Internet, d’autre part. Les capacités dynamiques sont exprimées dans les différents processus de gestion (Eisenhardt et Martin, 2000), qui se retrouvent tant au niveau stratégique qu’opérationnel1. Ceci amène à poser la seconde question de recherche : quelles sont les différentes capacités dynamiques qui sont mobilisées et quel est leur rôle dans le processus de création et de développement de la capacité organisationnelle Internet? Le texte est organisé en quatre parties. Dans une première partie, on présente le cadre conceptuel de la recherche, dans une deuxième, le cadre méthodologique, dans une troisième, les résultats. On termine par une conclusion. Modèle théorique de la capacité organisationnelle Internet et des capacités dynamiques On aborde cette recherche en faisant référence à un modèle théorique qui puisse rendre compte de la dynamique du changement et l’on s’est inspiré de la démarche contextualiste (Pettigrew, 1987) qui se compose de trois grandes dimensions interreliées entre elles : le contenu, le contexte et le processus (Pettigrew, 1987). Le modèle théorique a émergé de façon inductive de la recherche. Contexte Le contexte externe renvoie à l’environnement concurrentiel (Amit et Schoemaker, 1993; Porter, 1991) et technologique (Cummings et Doh, 2000) qui va déterminer les conditions du jeu concurrentiel dans lequel évolue l’organisation de même que la nature des ressources et capacités organisationnelles (Internet et fonctionnelles) qui vont s’avérer être des actifs stratégiques (Amit et Schoemaker, 1993; Peteraf, 1993). Quant au contexte interne, il renvoie d’une part à la stratégie générale de l’organisation et à sa stratégie en systèmes d’information et d’autre part, aux différentes capacités organisationnelles qui sont disponibles à l’organisation pour mettre en œuvre les capacités dynamiques qui vont permettre la création et le développement de la capacité organisationnelle Internet. Ce modèle repose sur une vision systémique d’une organisation qui est envisagée comme un ensemble de capacités organisationnelles interreliées entre elles. On postule ici à l’existence de relations entre les capacités organisationnelles de l’organisation et les capacités dynamiques, les premières servant de « toile de fond » aux secondes (Zollo et Winter, 2002). Les éléments du contexte qui viennent d’être définis vont jouer un rôle d’inhibiteur ou de facilitateur dans la mise en œuvre de la stratégie Internet et dans le processus la création et de développement de la capacité organisationnelle Internet. 1 On adopte ici la position de Eisenhard et Martin (2000) en ce qui concerne la définition des capacités dynamiques. Pour ces auteurs : “ First, dynamic capabilities consist of specific strategic and organizational processes like product development, alliancing, and strategic decision making that create value for firms within dynamic markets by manipulating resources into new value-creating strategies. Dynamic capabilities are neither vague nor tautologically defined abstractions. Second, these capabilities, which often have extensive empirical research streams associated with them, exhibit commonalities across effective firms or what can be termed ‘best practice’. (p.1106)”. 146 Contenu La dimension contenu fait référence à la capacité organisationnelle Internet, que l’on a défini précédemment comme étant l’habileté pour déployer, combiner et coordonner des ressources et des compétences à travers différents flux de valeur (donc des processus de commerce électronique) pour mettre en œuvre une stratégie Internet de type B2C ou B2B. Cette habileté peut être évaluée en regard de l’atteinte ou non des objectifs de performance initialement définis. D’un point de vue analytique, il est raisonnable de penser qu’il existe un niveau minimum de résultats ou niveau de performance, sous lequel les aspirations de l’organisation ne peuvent plus être rencontrées (Winter, 2000). L’évolution de la capacité organisationnelle Internet peut être représenté selon un modèle du cycle de vie de la capacité organisationnelle (Helfat et Peteraf, 2003). Figure 1 Modèle théorique de la recherche - Environnement concurrentiel Stratégie de l'organisation - Contexte Externe Contexte Interne + + Environnem ent technologique Stratégie Systèm e d'information Ressources et com pétences Capacités organisationnelles Fonction SI de n Pr de o c e la s s s t us ra d té e g i fo e rm In t e a t io rn n et Capacité organisationnelle Internet ti o A la pp s t r en ra t t é is s gi a e ge In te e t rn év e t ol u Niveau stratégique Fonctions m étiers Projet de commerce électronique de Pr o de c e la s s u st s ra d t é ' im gi p e la In n t t e at r n io et n n te t io In u a ie al é g év t d' ra s st s u la es d e oc ts P r lt a su re Niveau opérationnel s rn et Processus On s’intéresse aux capacités dynamiques qui sont mobilisées par l’organisation pour maintenir son adaptation aux conditions changeantes de son environnement qui découlent de l’arrivée des technologies Internet et de leur utilisation par des organisations concurrentes. En effet, les actions des organisations concurrentes vont venir modifier la dynamique concurrentielle qui prévaut dans l’industrie du tourisme, ce qui nécessite de la part de l’organisation de réagir pour contrecarrer ces menaces et pour saisir les opportunités stratégiques qu’elle découvre. Dans ce cas précis, la problématique de l’organisation à l’étude va être de former, d’implanter, d’exploiter et de faire évoluer sa stratégie Internet, ce qui nécessite de créer et de développer sa capacité organisationnelle Internet. Pour parvenir à cet objectif, l’organisation devra réaliser quatre types d’activités spécifiques, qui se retrouvent dans des processus de niveau stratégique et opérationnel. Ces processus expriment et mettent en œuvre les capacités dynamiques de l’organisation. On précise que l’organisation, au départ, n’a pas de stratégie Internet, et en corollaire, qu’elle ne possède pas de capacité organisationnelle Internet. Elle devra donc: 1) former sa stratégie Internet, c’est-à-dire qu’elle devra identifier les opportunités stratégiques qui s’offrent à elle en ce qui concerne l’utilisation des technologies Internet en définissant un problème stratégique qui va nécessiter d’élaborer un plan d’action stratégique; 2) implanter sa stratégie Internet pour apporter une réponse au problème stratégique, c’est-à-dire qu’elle devra réaliser l’intégration, la 147 reconfiguration et l’acquisition des ressources et des compétences nécessaires à la création et au développement de sa capacité organisationnelle Internet ce qui va lui permettre par la suite d’exploiter sa stratégie Internet en assurant le passage entre le « vouloir faire », « le pouvoir faire » et le « faire » de l’organisation; 3) Évaluer la performance atteinte par l’exploitation de sa stratégie Internet, c’est-à-dire qu’elle devra colliger des informations formelles et informelles en ce qui concerne le pilotage du projet, les problèmes rencontrés, les opportunités découvertes et la performance atteinte. Ces informations sont également essentielles pour évaluer le niveau d’apprentissage de la capacité organisationnelle Internet et décider des modifications et ajustements à apporter en conséquence. 4) Faire évoluer sa stratégie Internet, c’est-à-dire qu’elle devra réaliser des activités d’apprentissage stratégique qui porte sur l’intégration des retours d’expérience qui proviennent de l’évaluation de la performance en évitant de refaire les mêmes erreurs et en prenant en compte les transformations qui se sont déroulées au niveau du contexte interne et externe de l’organisation. C’est à ce moment que l’organisation choisit de faire évoluer sa stratégie Internet et de continuer ou non son apprentissage de la capacité organisationnelle Internet. On précise également que la manière dont l’organisation réalise ces quatre processus est contributive de la « qualité » de la capacité organisationnelle Internet qui sera créée et développée. En effet, ces processus peuvent être plus ou moins formalisés et structurés, et utiliser différentes méthodologies de gestion (Eisenhardt et Martin, 2000). De même, la réalisation de ces processus va se traduire par un ensemble d’apprentissages (Zollo et Winter, 2002), qui permettront d’en améliorer leur réalisation dans une étape subséquente de développement de la capacité organisationnelle Internet. Cadre méthodologique Pour répondre aux deux questions de recherche, on réalise une étude de cas (Yin, 1994) en profondeur d’une organisation de l’industrie du tourisme canadienne. L’organisation choisie, que l’on nomme Zêta, occupe une position dominante dans son industrie. À l’intérieur de ce cas, l’unité d’analyse retenue est tous les projets de commerce électronique B2C et B2B qui découlent de la stratégie Internet. Trois grandes sources d’évidence ont été employées: la documentation relative à l’organisation, les données archivées et les entrevues. 30 entrevues semi-directives ont été réalisées auprès de 20 membres de l’organisation. Les répondants ont été choisis en regard de deux grandes dimensions : selon leur position hiérarchique et fonctionnelle dans l’organisation, selon la nature de leur participation à la stratégie Internet et aux projets de commerce électronique. Dans un premier temps, et comme phase préparatoire, une étude de cas a été réalisée. Dans un deuxième temps, à partir de l’étude de cas préliminaire, on est passé à une étape supplémentaire dans l’agrégation des données et leur synthèse en recourant à une approche de représentation graphique de type organigramme (visual mapping) (Langley, 1999). La dernière étape consistait à comparer les résultats obtenus avec le cadre d’analyse provisoire et produire une synthèse théorique. Plusieurs principes ont été respectés durant la recherche pour assurer sa validité et sa fiabilité (Yin, 1994). Premièrement, différentes sources d’évidence ont été utilisées ce qui a permis une triangulation des données pour permettre de corroborer le phénomène à l’étude. En second, on a créé une base de données de l’étude de cas de façon à la rendre disponible pour tout autre chercheur. Troisièmement, on a utilisé la technique de la répétition du codage après trois semaines. Quatrièmement, les résultats de la recherche ont été validés par la personne ressource de l’organisation à l’étude. Finalement, la fiabilité a été assurée par une documentation des procédures qui ont été suivies durant tout le déroulement de la recherche. 148 Résultats On présente tout d’abord, l’organisation à l’étude, puis, en second, la stratégie Internet qui a été poursuivie par Zêta. Par la suite, on présente le cycle de vie de la capacité organisationnelle Internet et le rôle joué par les capacités dynamiques. Présentation de l’organisation Zêta est une organisation qui se spécialise dans l’organisation, la commercialisation et la distribution de produits touristiques, notamment des voyages vacances. Elle a vu le jour, il y a une quinzaine d’années, avec la fusion de deux voyagistes expéditifs et la création d’une compagnie aérienne spécialisée dans les vols nolisés. Depuis ses débuts, elle a connu un développement continu en adoptant une stratégie d’intégration verticale (voir tableau 1 et 2 en annexe). Elle réalise des revenus, pour l’année 2003, de plus de deux milliards de dollars canadiens, et elle intègre une vingtaine de filiales qui lui appartiennent soit en pleine propriété soit en partie. Cette organisation est structurée autour de trois grandes activités qui composent sa chaîne de valeur, à savoir des activités de distribution de produits touristiques, des activités de création de produits touristiques et des activités de transport aérien. On précise toutefois que son cœur de métier est composé des activités des voyagistes expéditifs et réceptifs. Stratégie Internet et chemin d’évolution de la capacité organisationnelle Internet La stratégie Internet de Zêta est étudiée sur une période de temps de quelque 6 années, c’est-à-dire, de 1998 à 2004 (voir figure 1 en annexe). Zêta met d’abord en œuvre un projet de commerce électronique B2C qui verra la création de l’agence de voyages virtuelle Oméga (1998-2002). Les technologies Internet vont donc être exploitées localement c’est-à-dire uniquement au sein de cette agence de voyages virtuelle (Venkatraman, 1995). À partir de 2002, la stratégie Internet va évoluer progressivement vers une stratégie de distribution multicanal qui va miser sur les synergies entre le canal de distribution électronique (sites Web transactionnels et agence de voyages virtuelle), physique (l’agence de voyages ayant pignon sur rue), et un centre d’appels, et ce, pour soutenir les activités de distribution de ses différentes filiales (réseaux d’agences de voyages, compagnie aérienne, voyagistes). Les technologies Internet vont donc être intégrées à l’interne (Venkatraman, 1995), toutefois, on précise que cette intégration ne vient pas bouleverser les processus opérationnels de ces filiales. À partir de 2003, la stratégie de Zêta évolue de façon plus importante puisque l’organisation va mettre en œuvre un projet de commerce électronique B2B orienté distribution. L’objectif sera de réviser les processus de distribution des produits touristiques entre les voyagistes, qui sont les producteurs de produits touristiques et les réseaux de distribution physiques et électroniques qui appartiennent ou non à Zêta. En d’autres termes, ce projet marque une étape importante puisque les technologies Internet sont dorénavant utilisées au sein des activités des voyagistes qui composent son cœur de métier. Elles deviennent un des moteurs de la révision des processus de distribution avec l’objectif d’en améliorer l’efficacité et l’efficience. On assiste donc à une recomposition des processus de gestion de Zêta à l’aide des technologies Internet (Venkatraman, 1995). 149 Figure 2 : Chemin d'évolution de la stratégie Internet Étendue Fort Degré de transformation de l'organisation 4. Reconception du réseau de gestion: CE interentreprises orientés fournisseurs (à venir) 3. Reconception des processus : CE interentreprises orientés agences de voyages 2. Intégration interne: distribution multicanal Niveaux révolutionnaires Niveaux évolutifs Locale 1. Exploitation en local : agence de voyages virtuelle Faible Faible Capacité organisationnelle Internet Redéfinition des objectifs de l'organisation (une entreprise en réseau à venir) Fort Éventail des bénéfices potentiels de l'utilisation des technologies Internet De façon générale, on retiendra qu’avant la création de son agence de voyages virtuelle Oméga, Zêta n’a aucune expérience dans le domaine du commerce électronique : sa capacité organisationnelle Internet est donc inexistante. Sur une période de quelque six années, cette capacité organisationnelle va être créée et développée à mesure que cette organisation met en œuvre différents types de projets de commerce électronique. Ce processus d’évolution témoigne également de la place qu’occupe progressivement la capacité organisationnelle Internet par rapport aux autres capacités organisationnelles de Zêta. Cette capacité organisationnelle Internet, qui, au départ, n’est utilisée qu’à la marge des activités traditionnelles de Zêta, c’est-à-dire au sein de l’agence de voyages virtuelle Oméga, devient de plus en plus importante en évoluant progressivement vers une capacité organisationnelle cardinale (Hamel et Prahalad, 1995). En effet, elle se trouve dès lors mobilisée dans de nombreux processus critiques de l’organisation, tant au niveau des processus de distribution vers les clients finaux que des processus de distribution vers les différents réseaux de distribution de Zêta. Elle participe directement à l’atteinte des objectifs stratégiques qui sont poursuivis en étant intégrée au modèle d’entreprise (Rapport annuel, Zêta, 2003). Cette figure porte également l’emphase sur les apprentissages qui sont réalisés par Zêta en ce qui concerne la création et le développement de sa capacité organisationnelle Internet et le rôle joué par les capacités dynamiques dans ce processus. Ces apprentissages témoignent du développement d’une compréhension collective partagée en ce qui concerne la stratégie Internet et l’utilisation des technologies Internet au sein de Zêta (Zollo et Winter, 2002) Cycle de vie de la capacité organisationnelle Internet et capacités dynamiques On utilise le cadre d’analyse présenté précédemment pour étudier le cycle de vie de la capacité organisationnelle Internet et le rôle joué par les capacités dynamiques. On présente le contexte interne et externe qui définissent les conditions de l’action stratégique ainsi que les différents processus qui expriment et mettent en œuvre les capacités dynamiques de Zêta. 150 Figure 3 Cycle de vie de la capacité organisationnelle Internet de Zêta (1998-2004) Étapes du développem ent de la capacité organisationnelle Internet Étape de la création et du développement Étape du redéploiement et de la réplication Étape du déploiement Étape de la maturité Com merce électronique d'entreprise à consom m ateurs 1ière période 1ière période - concurr entiel - Stratégie de l'organisation Ressources Conte xte Inte rne + Envir onnement Stratégie Système d'infor mation Fonction SI de or ganisationnelle Inter net ntis Stratégie de comm erce électronique Interentreprises Inte atio rne n t la str App re gie sag e Inte et rne évo t luti on Niveau stratég iq u e Pro de ce la ssu stra s té de gie form Capacité Environnement technologique Inte nta rne tion t Pro de ces la sus str até d'im gie pla Fonction SI rne métiers ion Inte uat ie val tég d'é tra s et com pétences or ganisationnelles sus la de ces ts Pro ulta Niveau opérationnel + Str atégie Systèm e d'information res Fonctions + s Ressources Capacités Conte xte Exte rne C onte xte Inte rne de - 3 ièm e période et compétences Projet de c ommerc e électronique concurr entiel 2 ième période technologique or ganisationnelles métiers até Capacités Fonctions + Conte xte Exte rne Capacité cardinale Internet Environnement Str atégie de l'organisation ? Comm erce électronique interentreprises 2 ièm e période Envir onnement ? Étendue Étape de la création Environnement technologique évo luti on Projet de c ommer ce élec tr onique de Niveau opérationnel de évo luti on t la t EÀF ? rne Ap pre stra ntis té sag gie e Inte et rne ion Inte uat gie val até d'é str sus la de ces ts Pro ulta res Fonction SI Pro de ces la sus stra tég de ie form Inte a tion rne t Capacité organis ationnelle Internet Inte nta rne tion t s Niveau stratég iq u e Place de marché ? Pro de ces la sus stra tég de ie form Inte a tion rne t Capacité organis ationnelle Internet et com pétences métiers Pro de ces la sus str até d'im gie pla Ressources Capacités or ganisationnelles Fonctions t sag e Inte et rne + Str atégie Système d'information la + ie Conte xte Exte rne C onte xte Inte rne Niveau stratég iq u e - Ap pre stra ntis tég - concurr entiel Str atégie de l'organisation de t Envir onnement Projet de c ommerc e électr onique de Pro de ces la sus s tra d'im tég ie pla Inte nta rne tion t tion Inte ua ie val tég d'é tra s sus la de ces ts Pro ulta res Niveau opérationnel s rne t Stratégie de comm erce électronique d'entreprise à consomm ateur - concur rentiel - Stratégie de l'or ganisation + Conte xte Exte rne Conte xte Inte rne + Fonction SI de Capac ité ntis la str Ap pre gie sag e Inte et rne évo luti on Niveau stratég iq u e Pro de ce la ssu stra s té de gie form or ganisationnelle Inter net Inte atio rne n t Pr ojet de c ommerc e élec tronique Inte nta rne tion t d'é sus la de ces ts Pro ulta res Pro de ces la sus str até d'im gie pla des ion Inte uat gie val até str Niveau opérationnel Envir onnement technologique Stratégie Système d'infor mation et compétences organisationnelles métier s até Capacités Fonctions Sites Web transactionnels pour les filiales EÀD Stratégie de distribution multicanal rne Capacité organisationnelle Internet Ressour ces t Environnement t Agences de voyage virtuelles Agences de voyage virtuelles à rabais 1995 Locale Sites informationnels 2000 2003 2002 Simple 2005 2006 Complexe Projet Création et de développement de la capacité organisationnelle Internet (1998-2002) Il s’agit d’une étape importante durant laquelle Zêta va commencer son processus d’accumulation des ressources et compétences qui sont nécessaires à l’exploitation de sa stratégie Internet qui renvoie à l’exploitation d’une agence de voyages virtuelle. Ces ressources et compétences vont avoir une influence sur les étapes suivantes d’évolution de la capacité organisationnelle. Durant cette étape un certain nombre de connaissances vont être créées et cristallisées dans le système technique qu’est l’agence de voyages virtuelle et qui définit un archétype de connaissances (Nonaka et Takeuchi, 1997). Contexte externe et interne. En 1998, les concurrents traditionnels de Zêta n’ont pas encore de stratégie Internet. L’environnement technologique qui prévaut à cette époque est dominé par des systèmes d’information légués qui ne sont pas ou peu compatibles entre eux. Il existe donc peu de normes technologiques communes pour les communications à part dans le domaine de la distribution électronique via les Systèmes Mondiaux de Distribution (Werthner et Klein, 1999). De nouveaux joueurs apparaissent au Canada à partir de 1998, qui capitalisent sur les technologies Internet pour proposer une nouvelle façon de distribuer les produits touristiques au moyen de sites Web transactionnels et ils sont considérés comme une menace par Zêta (ex : Expedia; Travelocity, etc.). Durant la période qui s'échelonne de 1997 à 2001, la stratégie d'intégration verticale de Zêta arrive à maturité (voir tableau n°2 en annexe). On assiste au prélude d’une transformation structurelle de première importance qui se traduit par l’embauche de nouveaux gestionnaires d’expérience au niveau du siège social pour seconder la haute direction du groupe. L’objectif est d’introduire les « meilleures pratiques » de gestion ce qui va se manifester par une amélioration progressive des capacités organisationnelles métiers et en gestion des SI (voir tableau n°3 en annexe) et, en corollaire, de passer progressivement d’une culture entrepreneuriale à une culture corporative qui reflète davantage le niveau de développement du chiffre d’affaires de Zêta. 151 Formation de la stratégie Internet de commerce électronique d’entreprise à consommateurs autour d’une agence de voyages virtuelle. Ce processus débute par une série de discussions informelles entre le chef de la direction des systèmes d’information qui a été embauché en 1997 et le vice-président exécutif en charge de la distribution des produits touristiques de Zêta. Ces gestionnaires prennent alors conscience de l’importance stratégique de la distribution en direct des produits touristiques du groupe de même que les bouleversements induits par l’avènement des technologies Internet. Rapidement, la réflexion se structure et focalise sur la création d’une agence de voyages virtuelle qui serait une nouvelle entité organisationnelle autonome pour vendre des produits touristiques, notamment des forfaits vacances, en provenance des filiales de Zêta et des organisations concurrentes en adoptant une position de premier entrant (Lieberman et Montgomery, 1988). Cette stratégie Internet particulièrement ambitieuse et innovatrice en regard de l’inexpérience de Zêta dans le domaine du commerce électronique B2C est caractéristique d’une part de sa culture entrepreneuriale qui prévaut à cette époque et d’autre part, de la croyance partagée, qui va d’ailleurs s’avérer sans fondement par la suite, que les technologies Internet sont des technologies perturbatrices (Christensen, 1997). Ce processus de formation de la stratégie Internet est un processus structuré qui va nécessiter la recherche, l’organisation, la diffusion et la discussion de nouvelles connaissances pour alimenter la réflexion stratégique puisque l’organisation n’a pas d’expérience dans le domaine. Progressivement un univers de discours propre au domaine du commerce électronique se développe au sein de Zêta. Processus d’implantation de la stratégie de commerce électronique bâtie autour d’une agence de voyages virtuelle. Ce processus qui se trouve à l’interface des niveaux stratégiques et opérationnels va exprimer ce que l’organisation sera capable de faire, donc son « pouvoir faire », en ce qui concerne l’implantation de la stratégie Internet qui a été formée précédemment. Ce processus correspond à la phase de cristallisation des connaissances dans un système technique qui est un archétype de connaissances (Nonaka et Takeuchi, 1997). Toutefois, ce processus d’implantation de la stratégie Internet ne se fait pas sans heurts ce qui révèle l’inexpérience et le manque de maturité de Zêta au niveau de sa capacité organisationnelle de gestion de projet qui est insuffisamment maîtrisé par la fonction SI. En d’autres termes, on note un décalage très important entre le « vouloir faire » de Zêta et son « pouvoir faire » en matière de stratégie Internet. Processus d’exploitation de l’agence de voyages virtuelle : mise en œuvre de la capacité organisationnelle Internet. L’agence de voyages virtuelle est officiellement mise en ligne le 1er février 2000 mais elle connaît des difficultés techniques très importantes qui mettent en lumière les différentes erreurs de conceptualisation qui ont été commises au niveau de l’architecture du moteur de recherche et de réservation sur lequel repose le site Web transactionnel. Même si les attentes en termes de performance de la capacité organisationnelle Internet ne sont pas atteintes, les gestionnaires de Zêta ne décident pas d’abandonner purement et simplement ce projet stratégique (Helfat et Peteraff, 2003). Au contraire, ils vont relancer un nouveau cycle d’apprentissage, qui va consister à résoudre les problèmes techniques qui sont apparus au niveau du moteur de recherche et de réservation sur lequel repose le site Web transactionnel. D’autre part, conscient des lacunes de la fonction SI en ce qui concerne la gestion de projet, Zêta va choisir d’impartir le développement de cette seconde version du site auprès d’un fournisseur. Le nouveau site Web transactionnel et le moteur de recherche sont mis en ligne le 1er septembre 2000. Cette première étape de création et de développement de la capacité organisationnelle Internet témoigne de la nature des capacités dynamiques de Zêta qui sont exprimées et mises en œuvre dans les différents processus de gestion que l’on a présentés précédemment et des difficultés rencontrées par Zêta pour implanter sa stratégie Internet. Plusieurs éléments vont permettre d’expliquer ces difficultés. Au niveau du contexte externe, on mentionne le manque de maturité des technologies Internet et la difficulté de trouver des partenaires qui les maîtrisent à cette époque. Au 152 niveau du contexte interne, en premier lieu, on trouve les pressions liées à la volonté d’être un premier entrant sur le marché. En second lieu, si l’arrivée de nouveaux gestionnaires d’expérience au niveau du siège social, qui ont été engagés pour améliorer les pratiques de Zêta, a permis de rehausser le niveau de compétence de la haute direction, leur impact au niveau opérationnel est encore faible. En conséquence, il se crée un déséquilibre entre le processus de formation et le processus d’implantation de la stratégie Internet qui exprime un manque de cohérence entre le « vouloir faire » et le « pouvoir faire » de l’organisation. En d’autres termes, si Zêta a pu rapidement définir une stratégie Internet, son implantation va mettre en relief le manque de maturité de la capacité organisationnelle de gestion de projet. Cette situation témoigne également du niveau de développement insuffisant des capacités organisationnelles de la fonction SI qui sont critiques dans toute stratégie technologique. Cependant, la résolution de ces problèmes sont des sources d’apprentissage et donc de création de nouvelles connaissances qui vont être incorporées au stock de connaissances que possède déjà Zêta, ce qui va permettre d’améliorer la réalisation de ce processus critique de gestion de projet dans la phase suivante de développement de la capacité organisationnelle Internet. Redéploiement et réplication de la capacité organisationnelle Internet (2002-2004) Après avoir réalisé la création et le développement de la capacité organisationnelle Internet qui permet l’exploitation de l’agence de voyages virtuelle, les gestionnaires vont donc être confrontés à un choix qui porte sur la suite à donner à ce processus de développement (Hamel et Prahalad, 1995). Plusieurs options s’offrent alors à eux : soit redéployer, renouveler ou recombiner la capacité organisationnelle, soit la répéter, soit l’abandonner de façon immédiate ou progressive (Helfat et Peteraf, 2003). Plusieurs éléments vont influencer la prise de décision concernant la relance ou non de l’apprentissage de la capacité organisationnelle : 1) l’analyse de la performance obtenue par l’agence de voyages virtuelle Oméga, à un moment donné, en fonction de l’atteinte ou non d’objectifs définis dans des plans, des cibles et des seuils de succès (Winter, 2000). 2) l’analyse de l’évolution du contexte interne et externe dans lequel l’action stratégique s’est produite. On va voir dans la suite du texte que Zêta va choisir de redéployer sa capacité organisationnelle d'une part, mais aussi de la répliquer, d’autre part. Contexte externe et interne1.Le marché du tourisme au niveau mondial et au niveau des marchés de Zêta se dégrade singulièrement de la fin de l’année 2001 jusqu’en 2003 ce qui va nécessiter une réaction rapide de la part de Zêta pour maintenir son adaptation au contexte externe. Zêta va alors accélérer la mise en œuvre des changements qu’elle avait débutés dans la période précédente, notamment au niveau de la révision de sa structure organisationnelle et de ses processus pour obtenir des gains d’efficience et d’efficacité. La transformation de la stratégie générale va avoir aussi un effet sur la stratégie Internet qui doit être adaptée à ce contexte de rareté des ressources financières. Le marché canadien du commerce électronique B2C dans le domaine touristique continue à se développer à un rythme soutenu. Les principaux joueurs qui se posent en concurrents directs d’Oméga sont bien installés sur le marché (Expédia, Travelocity, Destina). En ce qui concerne l’environnement technologique, les technologies Internet sont de mieux en mieux maîtrisées au niveau des organisations. Les fournisseurs de technologie les intègrent dorénavant au cœur de leurs applications commerciales et proposent des solutions complètent pour faire du commerce électronique, pour améliorer la gestion des inventaires ou pour supporter la réalisation des forfaits vacances. 1 La présentation du contexte externe et interne vaut également pour la stratégie Internet qui met en oeuvre un projet de commerce électronique interentreprises orienté distribution. 153 Processus d’évaluation de la stratégie Internet bâtie autour de l’agence de voyages virtuelle Oméga. Du lancement de la deuxième version de son site Web transactionnel, qui a lieu en septembre 2000, jusqu'à l'automne 200, l’agence de voyages virtuelle Oméga est quand même exploitée. Les gestionnaires d’Oméga vont procéder à l’évaluation des premiers résultats d’exploitation en regard des objectifs de performance qui étaient consignés dans le plan d’affaires. Zêta visait à s'accaparer 50% du marché du voyage en ligne au Canada et devenir ainsi le principal distributeur de produits touristiques du groupe. Cependant, ses premiers résultats d’exploitation sont modestes par rapport aux objectifs stratégiques. Entre septembre 2000 et février 2001, l'agence de voyages virtuelle a fait voyager 15 000 personnes. Alors que le plan d'affaires initial avait prévu des revenus de 6,7 millions de dollars à la fin de l'année 2000, les revenus de commission n'auront été que d'un million de dollars. À la fin de l'année 2001, les revenus de commissions vont atteindre 2,7 millions alors que les prévisions étaient de 29 millions de dollars dans le plan d'affaires. Processus d’évaluation et d’évolution de la stratégie Internet, formation de la stratégie de distribution multicanal. Le processus d’évaluation de l’exploitation de la stratégie Internet a permis la création des connaissances « de qualité » qui sont maintenant disponibles pour les gestionnaires de la haute direction ce qui permet de faire baisser l’ambiguïté et l’incertitude dans leur prise de décision. Ils décident alors d’opérer une transformation de la stratégie Internet initiale afin de l’adapter au nouveau contexte externe et interne. Celle-ci va évoluer vers une stratégie de distribution multicanal de type « clic, talk, walk ». Zêta abandonne donc l'idée de faire d'Oméga la marque phare du groupe dans le domaine du commerce électronique d’entreprise à consommateurs puisqu’elle ne peut pas soutenir la compétition avec les grands joueurs de ce domaine, notamment Expedia et Travelocity étant donné le manque de moyens marketing dont elle dispose. Du point de vue de Zêta, il semble plus avantageux de capitaliser sur l’image de marque des autres filiales qui sont déjà bien établies sur leur marché respectif, qui deviennent des ressources complémentaires (Teece et al., 1997). Processus d’implantation de la stratégie de distribution multicanal . Le processus d’implantation de la stratégie se fait en suivant ces deux grands axes. On précise que les apprentissages réalisés lors de la mise en œuvre de l’agence de voyages virtuelle ont pour conséquence que cette implantation de la stratégie se déroule sans problème. En premier lieu, on assiste à la restructuration et au repositionnement de l’agence de voyages virtuelle Oméga qui va devenir une agence de voyages virtuelle qui vend à rabais des produits touristiques de dernière minute, ce qui va nécessiter de redéployer la capacité organisationnelle Internet dans un nouveau couple produit-marché (Helfat et Peteraf, 2003). En second lieu, on assiste à la mise en oeuvre, auprès de certaines filiales de Zêta d’un ensemble de sites Web transactionnels, ce qui nécessite dans ce cas-ci de répliquer la capacité organisationnelle Internet. Cela signifie que celle-ci va être utilisée pour produire un même type de produit mais dans un autre marché que celui qui était visé au départ (Helfat et Peteraf, 2003). La réplication nécessite d’abord une phase d’exploration (Helfat et Peteraf, 2003). Cette phase d’exploration va être réalisée au niveau du premier site Web transactionnel de la compagnie aérienne canadienne qui va être créé à partir du moteur de recherche d'Oméga au début de l'hiver 2002. Les gestionnaires de Zêta vont alors décider par la suite de mettre en œuvre les autres sites Web transactionnels au niveau de leurs différents réseaux de distribution et de leurs principaux voyagistes. Le processus d’implantation devient alors un processus routinier (Zollo et Winter, 2002) qui se fait de manière progressive jusqu’à l’automne 2003. Processus d’exploitation de la stratégie de distribution multicanal. À la fin de l’année 2003, les processus de restructuration et de repositionnement de l’agence de voyages Oméga sont achevés et on entre dans une phase de routinisation de son exploitation. L’agence de voyages virtuelle va évoluer maintenant sans faire appelle aux capacités dynamiques de Zêta. L’exploitation des sites 154 Web transactionnels devient également une activité routinière pour les différentes filiales de Zêta. L’exploitation des sites Web transactionnels permet aussi d’exposer les gestionnaires et les employés de ces filiales aux technologies Internet et à la problématique du commerce électronique d’entreprise à consommateurs. La capacité organisationnelle Internet dans le domaine du commerce électronique d’entreprise à consommateur arrive alors à maturité (Helfat et Peteraff, 2003). En ce qui concerne les capacités dynamiques qui sont mobilisées durant cette phase, on remarque des différences par rapport à la phase précédente de création et de développement de la capacité organisationnelle Internet. Les transformations observées dans le contexte interne et externe vont avoir une influence sur la manière dont sont réalisés les différents processus qui expriment et qui mettent en œuvre les capacités dynamiques de Zêta. En premier lieu, le processus de formation de la stratégie multicanal implique moins de participants et il est moins structuré. Il repose davantage sur des routines de résolution de problème qui dépendent des compétences individuelles puisque les gestionnaires possèdent des connaissances qui proviennent directement de l’exploitation de l’agence de voyages virtuelle Oméga ce qui améliore la qualité de leur prise de décision. Également, le contexte externe va nécessiter des actions stratégiques rapides ce qui explique la nécessité de recourir à des routines de résolutions de problème plus simples (Eisenhardt et Martin, 2000). En second lieu, le processus d’implantation de la stratégie Internet est un processus itératif et répétitif qui est réalisé au niveau de la fonction SI. Ce processus routinier repose, lui aussi, sur les apprentissages qui ont été faits durant la première période notamment en ce qui concerne la gestion de projet qui est mieux maîtrisé maintenant et sur la connaissance des fournisseurs de technologie qui ont en charge le développement du moteur de recherche et de réservation. D’autre part, la fonction SI a connu une amélioration significative de la maturité de ses capacités organisationnelles fonctionnelles qui sont exprimées et mises en œuvre dans les différents processus que l’on retrouve au niveau de la gouvernance et de la gestion des ressources informationnelles (voir tableau 2 en annexe). En troisième lieu, à mesure de l’exploitation de la stratégie Internet, les connaissances opérationnelles sont crées et diffusées et elles viennent renforcer le cadre mental des gestionnaires de Zêta dans ce domaine qui est maintenant institutionnalisé (Montealegre 2000). On en retrouve des traces dans les rapports annuels qui brossent à grand trait les activités réalisées durant l’année financière. Renouvellement de la capacité organisationnelle Internet vers le commerce électronique interentreprises (2001-) La stratégie Internet de type commerce électronique interentreprises est envisagée comme le prolongement de la stratégie de commerce électronique d’entreprise à consommateurs dans la mesure où Zêta va capitaliser sur la capacité organisationnelle Internet précédemment développée pour mettre en œuvre ses différents projets de commerce électronique interentreprises. Cependant, cette stratégie Internet va nécessiter le renouvellement de la capacité organisationnelle Internet (Helfat et Peteraf, 2003) ce qui nécessite la relance d’un nouveau cycle d’apprentissage. En d’autres termes, le passage du commerce électronique B2C au commerce électronique B2B marque une étape importante dans l’utilisation des technologies Internet chez Zêta et donc en corollaire, en ce qui concerne la capacité organisationnelle Internet. Formation de la stratégie Internet de type commerce électronique interentreprises. La formation de cette stratégie se déroule selon deux grandes périodes. La première période débute à l’hiver 2001 et se termine à l’été 2003. On assiste aux prémices d’une réflexion stratégique sur ce que pourrait être le commerce électronique interentreprises chez Zêta mais sans que cela ne débouche sur des projets à réaliser d’une part, parce que la situation du contexte externe et la nécessité de s’y adapter est un enjeu prioritaire pour les gestionnaires, d’autre part, par ce que ces 155 derniers n’ont pas encore une compréhension très fine des enjeux stratégiques qui découlent de l’utilisation des technologies Internet. La fonction SI va alors réaliser durant cette période un travail « d’éducation » auprès de ces gestionnaires, en diffusant des connaissances pertinentes. La deuxième période débute en novembre 2003 et se termine en décembre 2003. Un comité stratégique est créé pour réfléchir à ce type de stratégie Internet qui est alignée maintenant avec la stratégie générale de l’organisation. L’objectif est de procéder à une réduction des coûts d’opération et d’améliorer l’efficience et l’efficacité des processus opérationnels. Deux types de commerce électronique interentreprises sont alors définis selon qu’ils peuvent être orientés distribution ou fournisseur. Dans un premier temps, les gestionnaires choisissent de mettre en œuvre un projet de commerce électronique interentreprises orienté distribution. En résumé, durant cette période, la stratégie de commerce électronique interentreprises est formée. Le concept est défini précisément de même que les objectifs à atteindre et les projets qui vont avoir la priorité. Implantation de la stratégie Internet de type commerce électronique interentreprises. (cette phase d’implantation est toujours en cours). Le coût relativement élevé d’un tel projet et ses conséquences sur les modes opératoires des processus de l’organisation nécessitent que ce processus d’implantation de la stratégie soit très bien structuré et maîtrisé. Les membres de l’équipe projet vont donc utiliser une nouvelle méthode de gestion de projet qui a été élaborée par la fonction SI. Dans un premier temps, l’objectif va être de réaliser une étude de faisabilité en précisant l’architecture technologique et les modalités du développement technologique, l'identification des besoins d'affaires. Un comité « aviseur » va être constitué pour chapeauter l'ensemble du projet. Son rôle est de s’assurer que les décisions prises dans le cadre de ce projet soient en ligne avec les besoins actuels et futurs du marché. On remarque que les processus de gestion qui expriment et qui mettent en œuvre les capacités dynamiques de l’organisation ont gagné d’une part en maturité et d’autre part en structuration, surtout si on les compare aux deux périodes présentées précédemment. Le déséquilibre qui existait entre le processus de formation de la stratégie et celui de l’implantation s’est estompé en raison des efforts qui ont été entrepris pour améliorer les capacités organisationnelles de la fonction SI et des différentes fonctions métiers. Également, une pression supplémentaire s’ajoute à la nécessité d’améliorer la réalisation de ces processus. Elle provient du fait que les transformations qui vont être occasionnées par la mise en œuvre de ce type de stratégie Internet sont beaucoup plus importantes que les précédentes. La gestion du changement est alors une problématique de première importance pour Zêta puisque la mise en œuvre de cette stratégie Internet va toucher aux processus opérationnels de son cœur de métier et ceux également du réseau de distribution. Conclusion De façon générale, Zêta fait preuve de clairvoyance, car dès la fin 1998 elle débute l’apprentissage de la capacité organisationnelle Internet avant certains de ces concurrents traditionnels. En effet, Si les organisations concurrentes n’ont pas débuté à temps leur processus d’accumulation des ressources, il leur sera difficile de combler rapidement leur retard étant donné les déséconomies liées au temps nécessaire pour les accumuler. En effet, la problématique de l’apprentissage ne peut être éludée par les organisations concurrentes. Le temps nécessaire à l’accumulation des ressources confère alors un avantage à l’organisation qui est la première à les avoir acquis (Lieberman et Montgomery, 1988). Une fois développée, la capacité organisationnelle devient un actif stratégique qui peut être utilisé dans différents contextes et pour différentes stratégies Internet. Cependant, il y a dépendance de sentier dans la mesure où les phases liminaires de création et de développement, et l’accumulation des ressources et compétences qui en résultent 156 vont contraindre le développement ultérieur de la capacité organisationnelle Internet et donc la stratégie Internet La stratégie Internet débute par un projet de commerce électronique de type entreprise à consommateur qui marque l’étape de création et de développement de la capacité organisationnelle Internet. L’objectif est de proposer un nouveau canal de distribution électronique aux consommateurs canadiens de produits touristiques. Zêta va alors exploiter une agence de voyages virtuelle qui se trouve en marge des filiales traditionnelles du groupe. Par la suite, cette stratégie va évoluer vers une stratégie de distribution multicanal de type « clic, talk, walk » qui marque l’étape de redéploiement et de réplication de la capacité organisationnelle Internet. En effet, Zêta, en même temps qu’elle repositionne son agence de voyages virtuelle qui devient spécialisée dans la vente à rabais de produits touristiques de dernière minute, va mettre en œuvre plusieurs sites Web transactionnels pour soutenir les activités de distribution de ses différentes filiales. La stratégie Internet va connaître une évolution plus radicale par la suite, puisque l’organisation va mettre en œuvre un projet de commerce électronique interentreprises orienté distribution qui renvoie à la phase de renouvellement de la capacité organisationnelle Internet. L’objectif visé est de réduire les coûts d’opération ainsi que l’efficience et l’efficacité des processus de distribution des produits touristiques de Zêta vers les différents réseaux d’agences de voyages qui lui appartiennent ou non. Le processus de création et le développement de la capacité organisationnelle Internet est un processus complexe, itératif et cumulatif, qui renvoie aux différentes dynamiques organisationnelles à l’œuvre au sein de l’organisation à l’étude. Ce processus nécessite la présence des capacités dynamiques qui sont exprimées et mises en œuvre dans quatre différents processus de gestion qui sont respectivement : les processus de formation, d’implantation, d’évaluation et d’évolution de la stratégie Internet. Ces processus sont soumis à l’apprentissage et donc, chacun à leur manière, lors de leur mise en œuvre, ils vont permettre la création de connaissances. La qualité de la réalisation de ces processus est influencée par plusieurs éléments; 1) par le contexte externe qui définit la nature du défi stratégique à relever par l’organisation de même que la rapidité pour le relever; 2) par le contexte interne qui définit la maturité des capacités organisationnelles fonctionnelles qui servent de « toile » de fond aux capacités dynamiques; 3) de la cohérence affichée entre les processus qui expriment et qui mettent en œuvre les capacités dynamiques. Bibliographie Amit, R. and Schoemaker, P.J. (1993). Strategic Assets and Organizational Rent. Strategic Management Journal, 14, 33-46. Buhalis, D., (2003). e-Tourism, Information technology for strategic tourism management, Harlow, Prentice-Hall Financial Times. Davies, A., Brady, T. (2000). Organisational capabilities and learning in complex products systems:Towards repeatable solutions, Research Policy, 29, 931-942 Barney, J.B. (1991). Firm resources and Sustain of Competitive Advantage. Journal of Management, 17, 99-120. Christensen, C.M. (1997). The Innovator's Dilemna, When New Technologies Cause Great Firms to Fail. Cambridge, Harvard Business School Press. Durand, T., (1996) Savoir, savoir-faire, savoir être, repenser les compétences de l’organisation, White paper, non publié, 37p. Einsenhardt, K. and Martin, J.A. (2000). Dynamic Capabilities: What are they? Strategic Management Journal, 21, 1105-1121. Girod-Seville, M., Perret, V. (1999). Fondements épistémologiques de la recherche, p.13-33 in Méthodes de Recherches en Management, Thiétard, T., et coll., Dunod Paris, 5335 p. 157 Grant, R.M. (1991). The Resource-Based Theory of Competitive Advantage: Implication for Strategy Formulation. California Management Review, 33, 114-135. Helfat, C., Raubitschek, R. (2000). Product sequencing: Co-evolution of knowledge, capabilities and products, Strategic Management Journal, 21, 961-975. Helfat, C., Peteraf, M., (2003). The Dynamic Resource-Based View: Capability Lifecycles, Strategic Management Journal, 24, 997-1010. Holbrook, D., et al. (2000). The nature, sources, and consequences of firm differences in the early history of the semiconductor industry. Strategic Management Journal, 21, 1017-1029. Kuwada, K. (1998). Strategic Learning: The Continuous Side of Discontinuous Strategic Change. Organization Science, 9, 719-736. Langley, A., (1999). Strategies for theorizing from process data, The Academy of Management Review, 24, 4, 691-710. Lejeune, A., Roehl, T (2003). Hard and Soft Way to create value from information flows: Lessons from the Canadian financial service industry, Working Paper, ESG, UQAM. Lieberman, M.B. and Montgomey, D.B. (1988). First-Mover Advantages. Strategic Management Journal, 9, 41-58. Lorino, P. (2001). Méthodes et Pratiques de la Performance, Paris, Éditions d’Organisation, 551 p. Montealagre, RA., (2002). Process Model of Capability development : Lessons fom the Electronic Commerce Strategy at Bolsa de Valores de Guyaquil, Organization Science, 13, 514-531. Nonaka, I. and Takeuchi, H. (1997). La connaissance créatrice, la dynamique de l'entreprise apprenante. Bruxelles, DeBoeck. Oliver, C. (1997) Sustainable Competitive Advantage; Combining Institutional and Resource-Based Views. Strategic Management Journal, 18, 697-713. Peteraf, M.A. (1993) The Cornerstones of Competitive Advantage: A Resource-Based View. Strategic Management Journal, 14, 179-191. Pettigrew, A. (1987), Context and Action in the Transformation of the Firm, Journal of Management Studies, 24, 6, 649-670. Prahalad, C.K. and Hamel, G. (1990). The Core Competence of Corporations. Harvard Business Review, 68, 3, 79-91. Porter, M.. (1991). Towards a Dynamic Theory of Strategy. Strategic Management Journal, 12, 95-117. Porter, M. (2001). Strategy and the Internet, Harvard Business Review, March, 63-78. Rosenbloom, R.S. (2000). Leadership, Capabilies, and Technological Change: The Transformation of NCR in the Electronic Era. Strategic Management Journal, 21, 1083 Teece, D.J. et al. (1997). Dynamic Capabilities and Strategic Management. Strategic Management Journal, 18, 509-533. Tripsas, M., Giovanni, G. (2000). Capabilities, Cognition, and Inertia: Evidence from Digital Imaging, Strategic Management Journal, 21, 1147-1167 Wernerfelt, B. (1984). A Resource-based View of the Firm. Strategic Management Journal, 5, 171-180. Werthner, H., Klein, S., (1999). Information Technology and Tourism, A Challenging Relationship, Springer, New York. Winter, S.G. (2000). The Satisficing Principle in Capability Learning. Strategic Management Journal, 21, 981-996. Yin, R.K., (1994). Case Study Research, Thousand Oaks: Sage. 170 p. Zander U., Kogut, B (1995). Knowledge and the Speed of the Transfer and Imitation of Organizational Capabilities: An Empirical Test. Organization Science, 6, 76-92. Zollo, M., Winter, S., G. (2002). Deliberate Learning and the Evolution of Dynamic Capabilities. Organization Science, 13, 339-351. 158 Tableaux Tableau 1 Périodes et axes stratégiques suivis par Zêta, (1986-2004) Périodes Cycle de vie du modèle stratégique 1986-1989 Prémices de la stratégie d'intégration verticale 1998- 2001 Maturité de la stratégie d’intégration verticale 2002-2004 Révision de la stratégie d’intégration verticale Entrepeunariale 1989-1997 Développement de la stratégie d’intégration verticale :atteinte de la masse critique Entrepeunariale Culture dominante Marchés Entrepeunariale/corporative Corporative Québec/France Canada/France Canada/France Canada/France (Europe/Etats-Unis) Marché de masse (+++) Marché de masse (+++) Marché de masse (+++) Marché de niche (+) Conquête de marchés Processus Croissance par acquisition (++) Diversification et conquête de marchés Diversification et conquête de marchés Marché de masse (+++) Marché de niche (+) Croissance par acquisition (+++) Croissance par acquisition (+) Révision du positionnement des marques Stabilisation de la croissance Diversification des activités le long de la chaîne de valeur Diversification des activités le long de la chaîne de valeur Optimisation de la structure Recherche de l’efficience et de l’efficacité pour réduire les coûts d’opération et améliorer la rentabilité 159 Périodes Approche Tableau 2 Évolution de la fonction SI (1987-2003) - 1997 1997-2000 2000-2003 Répondre aux Améliorer les Améliorer les besoins capacités capacités techniques des organisationnelles de organisationnelles de filiales selon la fonction SI et son la fonction SI et la une approche arrimage sur les qualité des services empirique besoins des clients. Participer à la restructuration des processus de Zêta TIC Systèmes hérités Gouvernance des SI Gestion décentralisée au niveau des filiales Pas de fonction SI au siège social Pas d’alignement stratégique entre fonctions métiers et SI Gestion des SI comme un centre de coûts Orientation technique Systèmes hérités/technologies Internet Gestion centralisée des SI au niveau du siège social de Zêta Organisation par directions informatiques filiales Création du poste de VP exécutif et chef de la direction des SI Prémices à la gestion de l’alignement stratégique métiers et SI Mise en place de la relation : client/prestataire Systèmes hérités/technologies Internet Gestion centralisée de la fonction au niveau du siège social de Zêta Organisation par domaines d’activités Création du poste de VP-SI Prémices à la gestion de l’alignement stratégique métiers et SI Relation client/prestataire Relation de partenaire Pérenniser les équipes Supporter les applications 160 2003Améliorer les capacités organisationnelles de la fonction SI et réduire les coûts de fonctionnement Participer à la restructuration des processus de Zêta Systèmes hérités/technologies Internet Gestion centralisée de la fonction SI/TI au niveau du siège social Organisation par services partagés et domaines de compétences Favoriser l’alignement stratégique métiers Relations de partenariat entre fonction SI et fonctions métiers Gestion de la gouvernance et contrôle financier (Comité de gouvernance des SI) ASAC 2005 Toronto, Ontario Suhaib Riaz (Student) Richard Ivey School of Business University of Western Ontario CAN ELEPHANTS DANCE? A LOOK AT THE ‘MAHOUTS’ FOR SOME ANSWERS The impact of outsider versus insider CEO successions is considered in the context of turnarounds, and TMT changes are discussed as a possible moderator of the CEO changes – performance relationship. The paper suggests that turnaround situations are best handled with an insider/outsider mix of CEO change and TMT change combinations. Turnarounds have been a subject of fascination for business executives and a frequent source of enthusiasm for investors (Furman & McGahan, 2002). The phenomenon is of wide interest as many organizations go through major performance declines at some point during their evolution and have to deal with the issue of how to get back to leadership positions (Barker & Barr, 2002). I use the word ‘turnaround’ here in its broadest sense, as representing a period of severe performance decline followed by a lasting performance improvement. The existing literature on the subject employs a variety of terms for closely-related phenomena, such as organizational change (Mintzberg & Westley, 1992), strategic change (Boeker, 1989; Boeker, 1992; Boeker, 1997a, b), organizational transformation (Romanelli & Tushman, 1994), organizational renewal (Barr, Stimpert, & Huff, 1992; Hurst, Rush, & White, 1989), etc. These terms have overlapping content and are also considered to be finer refinements of a broader concept that can be thought of as an organizational ‘turnaround’. For definitional purposes, following earlier studies, a turnaround can be thought of as a severe and prolonged downturn in performance followed by a lasting upturn; for example: four consecutive years of declining performance followed by four consecutive years of improved performance (Hambrick & Schecter, 1983). Turnarounds and Top Management: Can elephants dance? 1 In other words, can large, established organizations with well-set routines / capabilities change with the times and get back to leadership positions after going through periods of severe performance downturns? I argue that elephants can indeed dance, but we need to look at the mahouts2 for some of the key answers. The mahout that keeps and drives an elephant possesses deep knowledge about the elephant and shares an intricate relationship with it. The elephant could learn to dance, but it would take the understanding and trust of the current mahout, and some help from outside, to accomplish this. In other words, I will argue that to understand organizational turnarounds, we have to consider the role of both current top executives who have been with an organization for a reasonable length of time and possess knowledge about the ‘dance routines’ of the organization, and newer top executives who can bring change and teach the organization a few new ‘dance routines’. In general, top management and its relationship with organizational performance has been a much researched area in studies of organizations (Lieberson & O'Connor, 1972; Smith, Carson, & Alexander, 1984; Thomas, 1988; Weiner & Mahoney, 1981). Researchers have also argued that top management assumes great importance in organizational renewal (Hurst et al., 1989) and that it is the executive leadership of an organization that initiates and directs turnarounds (Barker et al., 2002). In this vein, some researchers have considered top management related issues in organizations going through turnaround type situations (Barker et al., 2002; Barker & Patterson, 1 I use the metaphor from Louis Gertsner’s book on IBM’s turnaround “Who says elephants can’t dance?” 2 Mahout: keeper and driver of an elephant (from Sanskrit “maha’matra”: one having great measure) 161 1996; Boeker, 1997a, b; Boeker & Goodstein, 1993; Goodstein & Boeker, 1991; Mueller & Barker, 1997; Schwartz & Menon, 1985; Virany, Tushman, & Romanelli, 1992). In this paper, I focus on changes in the top management of organizations and place this issue in the context of turnarounds. Changes in top management have been discussed as an important response to problems for organizations going through performance downturns (Daily & Dalton, 1995). Empirical studies have shown that changes in top management were far more prevalent in distressed organizations than in healthier ones (Hambrick & D'Aveni, 1992; Schwartz et al., 1985). While a large stream of literature exists on top management changes and CEO succession (Cannella & Lubatkin, 1993; Cannella & Shen, 2001; Shen & Cannella, 2002), there are relatively fewer works that look at CEO succession and top management changes with a focus on turnarounds and the impact these changes might have on success or failure of turnaround attempts. My research question is thus: What impact do CEO succession and top management changes have on organizational turnarounds? In studying this, I will develop arguments related to CEO succession and top management changes in the context of turnarounds, using the robust theoretical lenses of resource-based view (RBV) and institutional theory. I will argue that RBV logic implies rebuilding the organization through making use of the capabilities already established over a period of time, and that current top management has a vital role to play in this effort, since it has intimate knowledge of existing organizational capabilities. However, in later sections, I complement the RBV logic with theoretical insights from institutional theory and discuss how changes to top management could provide fresh blood to the organization, which is much needed for effecting successful turnarounds. Finally, I bring these arguments from RBV and institutional theory together to formulate propositions about CEO change, TMT change and firm performance for firms attempting turnarounds. Turnarounds and Top Management: An RBV Analysis The resource-based view (RBV) of the firm developed through the works of several authors over a period of time (Barney, 1986, 1991; Barney & Arikan, 2001; Peteraf, 1993; Rumelt, 1984; Wernerfelt, 1984), and has emerged as the major theoretical counterpoint to the earlier IO based framework (Caves & Porter, 1977; Porter, 1981) in strategic management. The key assumptions of RBV are that firms may have resource heterogeneity i.e. possess different bundles of resources, and that these resources may be immobile i.e. the resource differences may persist. In addition, similar to other organizational economics theories, RBV assumes that firms are profit-maximizing entities and that managers in firms are boundedly rational (Barney et al., 2001). According to RBV logic, firms acquire or develop their resources (tangible and intangible assets) in strategic factor markets (Barney, 1986) to conceive of and implement their strategies (Barney et al., 2001). Since resources could be inelastic in supply in imperfectly competitive (Barney, 1986) or incomplete (Dierickx & Cool, 1989) strategic factor markets, resources that are valuable (V), rare (R), inimitable (I) and can be exploited by the organization (O), are a source of sustained competitive advantage (Barney, 1991; Barney, 2002). While much debate exists on dynamic capabilities in the literature (Eisenhardt & Martin, 2000; Teece, Pisano, & Shuen, 1997), following Barney (Barney et al., 2001), I consider the dynamic capabilities literature as an expansion of the RBV domain and essentially included in the RBV logic. I thus argue considering RBV / Dynamic Capabilities as an integrated literature base for the purpose of this paper. RBV studies have considered managers as important, building on Penrose’s idea that managerial resources are one of the most critical for the firm (Penrose, 1959). Researchers have argued that organizational rent stems from the discretionary decisions of top managers to deploy selected resources and capabilities (Amit & Schoemaker, 1993). Top managers constitute an important firm resource that, as part of the entire bundle of resources of the firm, could enable the it to generate rents (Castanias & Helfat, 1991, 2001). Recent research has shown that ‘dynamic managerial 162 capabilities’ embodied in the top management of the firm are associated with time-varying corporate effects and are related to firm performance, where firms face the same external environments (Adner & Helfat, 2003). This is particularly relevant to the case of turnarounds where firms going through performance declines face the same external environment as their competitors, and suggests that top level managerial decisions could have an impact on firm performance over time in such situations. Drawing upon related RBV based studies (Barney & Tyler, 1992; Berman, Down, & Hill, 2002; Castanias et al., 1991, 2001; Wright, Smart, & McNahan, 1995), I discuss in more detail below the importance of existing top managers to the firm, and how they can be a source of improved performance. I thus consider the question of VRIO (Barney, 1991) with respect to the existing top management of the firm. The question of value of a resource has been discussed in the RBV literature, particularly with respect to the value being exogenously or endogenously defined with respect to RBV analysis (Barney, 2001; Priem & Butler, 2001). Following Barney (Barney, 2002), I consider this question in terms of whether the top managers can enable the firm to respond to external demands. Top management decisions operate on the resource and capability base of an organization (Adner et al., 2003), and it takes time for newer managers to fully understand the organization-specific resource and capability base (Hambrick & Fukutomi, 1991; Shen et al., 2002) or learn about management control systems and formal information-based routines in order to change patterns of firm behavior (Simons, 1994). Existing top managers, who have intricate knowledge of the existing organizational capabilities, therefore have value in terms of enabling the organization to respond to external demands in the context of turnarounds. It could also be argued that managers who have served in top roles in a particular organization would be rare, particularly since they are the only ones who posses important firm-specific skills, as distinct from industry-specific or generic skills (Harris & Helfat, 1997), and typically top levels of a particular organization comprise of a handful of individuals. Top managers possessing firm-specific skills are therefore a rare resource for the organization. In addition, the important question of inimitability can also be argued in favor of existing top management in an organization. Top management and its relations within the organization and with external constituents comprise a socially complex phenomena (Barney et al., 1992) that emerges over time, subject to time-compression diseconomies (Dierickx et al., 1989). Since top management acts on the resource and capability base of the organization and is intricately woven with them, issues related to interconnectedness of asset stocks also become important (Dierickx et al., 1989). Finally, the development of relationships within the top management and their relationship with organizational and external constituents is subject to the complexities of human behavior, suggesting the causal ambiguity surrounding the phenomenon (Barney, 1991; Dierickx et al., 1989). In addition to the above, the issue of ‘tacit knowledge’ with respect to the top management has also been raised in the literature. Idiosyncratic knowledge that accrues to top management of an organization is both articulate and tacit, and the tacit knowledge component that is of vital importance (Athanassiou & Nigh, 1999, 2000) could be lost in case of top management changes. The concept top management based tacit knowledge is therefore another aspect that adds to the inimitability of the top management. Barney also suggests that the organization’s other policies and procedures should be organized in order to effect full exploitation of the valuable, rare and costly to imitate resource (Barney, 2002). I argue here however, following earlier research related to top management within the RBV stream (Adner et al., 2003; Amit et al., 1993), that as top managers effect changes in the organization’s existing resource and capability base, the question of ‘O’ (other resources / capabilities being organized) is largely dependent upon the top management itself. Top management then, could be considered a special case of resources as they can be argued to be ultimately responsible for the 163 other resources / capabilities of the organization (Adner et al., 2003; Amit et al., 1993). This also avoids the problem of endlessly recursive capabilities that has been discussed in RBV literature (Collis, 1994). In sum, RBV arguments suggest that as a ‘VRI’ resource, existing top management has an important role to play in enabling the organization to achieve improved performance, which is the need of the hour for an organization going through a performance decline and attempting a turnaround. Turnarounds and Top Management: Complementary Analysis using Institutional Theory While RBV essentially developed from the organizational economics stream, institutional theory brings in a more sociological analysis to the study of organizations (Scott, 2001). Institutional theory holds that organizations operate as part of a wider social framework comprising norms, values and taken-for granted assumptions about what is acceptable or not-acceptable behavior. Organizational success is incumbent not only upon strictly economic factors, but also on conformity to social expectations. Conformance to institutional pressures occurs through institutional isomorphism processes that could be coercive, mimetic or normative (DiMaggio & Powell, 1983). This conformance is achieved through ‘myths’ and ‘ceremonies’ that get institutionalized as organizational structures and processes over time (Meyer & Rowan, 1991). This conformance to institutional pressures confers legitimacy to the organization (DiMaggio et al., 1983). However, organizations can also strategically respond to institutional pressures (Oliver, 1991) and manage their legitimacy (Suchman, 1995). Internal rigidities: Recent work by Oliver has placed institutional theory as complementary to RBV and has provided a combined analysis of the two theories with respect to competitive advantage (Oliver, 1997). Arguments from institutional theory apply to RBV in terms of providing the context in which individual, firm and interfirm processes take place. While resource-based determinants suggest that managerial choice would be influenced by economic rationality, institutional determinants suggest that normative rationality might also influence this choice (Oliver, 1997), and that over time, the nature of the decision process related to acquiring, maintaining and combining resources could become habitual, unreflective, and embedded in norms and traditions. These arguments parallel the research on the development of ‘rigidities’ in organizations which suggests that core organizational capabilities, over a period of time could become ‘core rigidities’ (Leonard-Barton, 1992). In particular, in organizations going through severe performance declines, inertia takes root (Tripsas & Gavetti, 2000), and external threats do little to shake off rigidities (Staw, Lance, & Dutton, 1981). Patterns of activities become routinized and cannot always be easily changed (Nelson & Winter, 1982). While these rigidities could be thought of as existing across various levels of the organization, my focus in on rigidities at the top management level, particularly in terms of how these affect top management decision making. As suggested by Oliver, where top management decision processes operate within historical and normative contexts with the objective of justifying prior resource choices, suboptimal resources allocations result and there is resistance to change (Oliver, 1997). The rigidity argument thus applies to top management decision making and could be a major source of problems for the firm attempting turnaround. Unless top management mental models change, organizations can go further into decline (Barr et al., 1992). However, troubled firms operating in hostile environments do break away from past behaviors and are able to accumulate innovative resource bundles and re-build industry leadership (Baden-Fuller & Stopford, 1994). Efficiency increases based on cutback actions might not be enough and strategic reorientation is needed (Barker & Mone, 1994; Pearce & Robbins, 1994; Robbins & Pearce, 1992). Given that top management decisions ultimately determine resource allocations of strategic importance (Adner et al., 2003; Amit et al., 1993), important change essentially has to 164 start with top management in the organization. Changes in top management are one way of overcoming organizational rigidities, since new management could effect resource allocations independent of historical and normative contexts, without having to justify prior resource choices (Oliver, 1997). Paralleling these arguments, earlier research suggests that top management changes are likely to lead to organizational transformations (Romanelli et al., 1994) by bringing in new ideas to counter internal cognitive biases and consensus (Weitzel & Jonsson, 1989). The discussion here thus suggests that top management changes could be of key importance for effecting successful turnarounds. Signal to external constituents: While countering internal rigidities is important from a long term perspective, for organizations undergoing performance downturns and hoping to turnaround, immediate actions also assume a symbolic importance (Hambrick et al., 1992). Creating positive impressions of the organization to both internal and external constituents is important (Flynn & Staw, 2004), and social construction activities are needed to influence stakeholders to accept that a change is underway (Gioia & Chittipeddi, 1991). External constituents need special attention during turnaround attempts and changes to the top management team could be a symbol of major change underway (Weitzel et al., 1989). Top management changes could have an important impact on external constituents, particularly stockholders (Friedman & Singh, 1989; Lubatkin, Chung, Rogers, & Owers, 1989), in effect impoving the chances of the firm to manage a successful turnaround. These arguments parallel research that suggests a relationship between top management and legitimacy of the organization (Hambrick et al., 1992; Lynall, Golden, & Hillman, 2003; Zajac & Westphal, 1996). Thus, I argue here that these top management changes can be an important signal of conformance to external constituents and could help with legitimacy related issues that the organization faces in times of severe performance declines. In sum, top management changes would help improve the chance of a successful turnaround by improving the organization’s legitimacy with external constituents at a critical time. However, while signals to external constituents could be important in the short-term, the turnaround attempts must also be substantive (Weitzel et al., 1989), and therefore the arguments related to dispensing with unwanted rigidities in the earlier section remain very important from the long-term perspective. Propositions I have argued above that in the current scenario being analyzed i.e. an organization attempting turnaround, arguments from RBV logic could suggest retaining existing top management as an important and unique resource, whereas arguments from institutional theory suggest the development of rigidities that can be overcome through top management changes. Further, these top management changes are needed as a signal to external constituents. Based on these competing arguments from the two theoretical lenses built in the preceding sections, I will develop testable hypotheses in this section related to top management change and organizational turnaround. The preceding discussion has referred to ‘top management’ of the organization in general. Here I introduce a refinement of ‘top management’ into CEO (Chief Executive Officer) and (TMT) Top Management Team. This refinement does not subtract from the theoretical arguments developed above for ‘top management’ in general as similar arguments could be built for both the CEO and TMT separately along the above lines. I have omitted separate arguments for the sake of brevity. The CEO position is explanatory, but the TMT requires definition. TMTs: I define TMTs as those reporting directly to the CEO, as these are likely to constitute an important part of the dominant coalition (Cyert & March, 1963) and are responsible for most organizational decision making (Hambrick & Finkenstein, 1987; Hambrick & Mason, 1984). In this paper, I use the term TMT to mean top management excluding the CEO. TMT changes: A top management team is considered changed when either (i) a new member has 165 been inducted in addition to the existing members, or (ii) a new member has been inducted as a replacement for an earlier team member (as a follow-up to a non-voluntary dismissal of the earlier team member). Cases such as voluntary departures and retirement situations are excluded as they are generally of less theoretical interest (Boeker, 1992; Frederickson, Hambrick, & Baumrin, 1988) and are of less relevance to our arguments here. Earlier researchers have argued for distinguishing between CEO change and top management change (Shen et al., 2002; Tushman & Rosenkopf, 1996; Virany et al., 1992). In keeping with these arguments, I distinguish CEO changes from TMT changes and study these as two separate components. Insiders versus outsiders: In discussing CEO changes, I also introduce a refinement in terms of ‘insider’ versus ‘outsider’ succession or changes. Insider replacements are those where members are changed from within the firm, and outsider replacement pertains to changes made by inducting individuals from outside the firm. Earlier studies have considered the insider versus outsider distinction as important (Bailey & Helfat, 2003; Lubatkin et al., 1989; Shen et al., 2002). In particular, this distinction has been considered important in the context of organizations with performance downturns (Datta & Guthrie, 1994) and is therefore relevant to our focus on turnarounds. While the insider and outsider distinction can be drawn for TMTs as well, for purposes of simplicity in this paper, I consider outsider TMT changes versus ‘insider TMT change / no TMT change’. In other words, TMT promotions from within are considered together with no TMT change situations and these are together contrasted with outsider TMT change situations. This does not conflict with our logical arguments under RBV or institutional theory sections, but rather, helps to draw a sharper contrast than if we had considered outsider versus insider TMT changes, as the distinctions between outsider TMT changes and no TMT change situations are expected to be more stark, based on our earlier logic. Performance: Firm performance is a multidimensional construct and has been measured with both operational and market based measures (Shen et al., 2002). Earlier studies focused on top management changes have looked at either operational performance (Shen et al., 2002) or market-based performance (Friedman et al., 1989; Lubatkin et al., 1989). However, in the context of top management changes for turnaround firms, both operational and market-based measures would be important. Financial markets have not always accurately identified turnarounds (Furman et al., 2002), and a focus on only operational measures might not capture all effects relevant to my arguments above, such as the impact of signals to external constituents. Lubatkin and Shrieves make similar arguments for reconciling market based performance measures with strategic management research (Lubatkin & Shrieves, 1986). In the context of turnarounds, it is important to see performance changes at both an operational level inside the firm, and also through market-based measures. This study would be among the first to consider CEO / TMT changes for turnaround firms and consider both market and operational performance measures to understand the phenomena better. My concern thus is to study the impact of CEO change (insider versus outsider) and TMT changes (insider / no change versus outsider) on both market based and operational firm performance measures for turnaround firms. Consider CEO changes made to a firm undergoing a performance downturn. From arguments discussed based on RBV logic above, a change in CEO would mean the loss of a valuable and unique resource that can have an important impact on helping the firm turnaround at this crucial time. However, following earlier arguments under the RBV section on social complexity, causal ambiguity, tacit knowledge, etc. related to the top management, if the CEO change is made from inside the firm rather than outside, the impact would be less disruptive as the new CEO possesses some of the intricate knowledge related to established organizational capabilities and control systems that would be needed to effect a turnaround. Parallel arguments by earlier researchers also suggest that insider CEO changes result in less disruption (Boeker et al., 1993; Wiersema & Bantel, 1992) and thus, following my earlier arguments for preserving some continuity based on RBV logic, I expect that, ceteris paribus, an insider CEO would be more likely to effect a successful 166 turnaround than an outsider CEO. Similarly, one could argue based on the same RBV logic, that insider TMT change / no change would be less disruptive, as opposed to outsider TMT changes, and would thereby be more likely to effect performance improvements leading to a turnaround. However these arguments are countered by the institutional theory perspective discussed earlier. According to the arguments under the institutional theory section, the need is to get rid of internal rigidities (Oliver, 1997; Weitzel et al., 1989), and therefore changes to CEO or TMT through outsiders are more likely to achieve better performance and effect successful turnarounds. Clearly, these two opposing arguments suggest that while some change is certainly needed to overcome rigidities (institutional theory based arguments), some continuity is also necessary to enable the top management to perform the necessary turnaround on an organization that they know best (RBV based arguments). Based on these two divergent arguments from RBV logic and institutional theory logic, I posit that a mixed approach to the two priorities would achieve the best results rather than an exclusive approach. In other words, situations where the CEO is changed from outside, but the TMT has low outsider representation; or the CEO is changed from inside, but the TMT has high outsider representation, are most likely to lead to successful turnarounds. I discuss these scenarios in further detail below. Consider the first scenario: While an outside change to the CEO ensures the loss of debilitating internal rigidities and a fresh cognitive outlook (Tushman et al., 1996; Virany et al., 1992), continuity in the TMT through high representation of inside members ensures that vital tacit knowledge (Athanassiou et al., 1999, 2000), firm-specific skills (Harris et al., 1997), and an intricate understanding of the firm’s unique resources and capabilities (Simons, 1994) is not completely lost. This argument is similar to Barney’s call that firms are most likely to succeed by first looking at and understanding what they already possess (Barney, 1986; Barney et al., 2001). In other words, based on knowledge of the organization’s exiting resources and capabilities (through insider TMT members), and ensuring that a new perspective is also in place (outsider CEO), the organization is likely to achieve successful turnaround. In other words, an outsider CEO under the condition of low outsider representation on the TMT, is more likely to achieve turnaround than an outsider CEO under the condition of high outsider representation on the TMT. In the latter case, the outsider perspective dominates, and as discussed above, vital firm specific knowledge is lost. The above discussion is essentially an argument for the moderating effect of outsider TMT representation on the relationship between CEO change and firm performance during a turnaround. In line with my earlier discussion, I examine effects on both operational and market based performance. In the long term, the impact of undoing internal rigidities while utilizing the organization’s unique resource and capabilities would be expected to have positive impact on both operational and market based performance. These discussions lead to an interesting set of propositions: P1a. Firms undergoing turnarounds are more likely to improve operational firm performance through outsider CEO succession than insider CEO succession, under condition of low outsider changes to the TMT. For market based performance measure, we would investigate: P1b. Firms undergoing turnarounds are more likely to improve market based firm performance through outsider CEO succession than insider CEO succession, under condition of low outsider changes to the TMT. The second scenario in which the CEO is changed from inside while the TMT has a high representation from outside is similar to the first scenario in terms of achieving the ‘best of both worlds’: continuity and change, and arguments similar to the first scenario would apply. Thus: P2a. Firms undergoing turnarounds are more likely to improve operational firm performance through insider CEO succession than outsider CEO succession, under condition of high outsider changes to the TMT. 167 For market based performance measure, we would investigate: P2b. Firms undergoing turnarounds are more likely to improve market based firm performance through insider CEO succession than outsider CEO succession, under condition of high outsider changes to the TMT. How do the two scenarios (P1a, b and P2a, b) compare? I do not offer propositions separately on this for the sake of brevity, but do mention that this would be an interesting question to investigate. Broadly, I expect the first scenario (P1a, b) to be more effective in achieving turnarounds compared to the second (P2a, b), considering that the first one provides a radical change at the topmost level (CEO) while preserving a varied set of knowledge related to the organization through the TMT. The moderation effects would be graphed (revealing whether the moderation is ordinal or disordinal) to clearly show which CEO / TMT change combination is the most effective recipe for a turnaround. We also investigate the legitimacy based argument discussed under the institutional theory section. From the arguments under ‘signal to external constituents’ section, outsider CEO succession is likely to send a more positive signal to external constituents, and would therefore have a more positive impact on market based performance than insider succession. This would be manifested in short term market based measures of performance (Friedman et al., 1989). Thus, in the short term: P3. Firms undergoing turnarounds are more likely to improve market-based firm performance through outsider CEO succession than insider CEO succession. Note that several other propositions are possible based on the logic developed in this paper (for example: CEO insider change versus no CEO change), but for purpose of brevity, I have confined the discussion to the interesting propositions outlined above. This research could help improve our understanding of the turnaround phenomenon and have important practical implications for how CEO / TMTs are changed during turnaround situations. References Adner, R., & Helfat, C. E. 2003. Corporate effects and dynamic managerial capabilities. Strategic Management Journal, 24: 1011-1025. Amit, R., & Schoemaker, P. J. H. 1993. Strategic assets and organizational rent. Strategic Management Journal, 14: 33-46. Athanassiou, N., & Nigh, D. 1999. The impact of U.S. company internationalization on top management team advice networks: A tacit knowledge perspective. Strategic Management Journal, 20: 83-92. Athanassiou, N., & Nigh, D. 2000. Internationalization, tacit knowledge and the top management teams of MNCs. Journal of International Business Studies, 31(2): 471-487. Baden-Fuller, C. W. F., & Stopford, J. M. 1994. Creating corporate entrepreneurship. Strategic Management Journal, 15: 521-536. Bailey, E. E., & Helfat, C. E. 2003. External management succession, human capital, and firm performance: An integrative analysis. Managerial and decision economics, 24: 347-369. Barker, V., & Barr, P. S. 2002. Linking top manager attributions to strategic orientation in declining firms attempting turnarounds. Journal of Business Research, 55: 963-979. Barker, V., & Mone, M. A. 1994. Retrenchment: Cause of turnaround or consequence of decline? Strategic Management Journal, 15: 395-405. Barker, V., & Patterson, P. W. 1996. Top management team tenure and top manager causal attributions at declining firms attempting turnarounds. Group and organization management, 21(3): 304-336. Barney, J. 1986. Strategic factor markets: Expectations, luck, and business strategy. Management Science, 32(10): 1231-1241. 168 Barney, J. 1991. Firm resources and sustained competitive advantage. Journal of Management, 17(1): 99-120. Barney, J. 2001. Is the resource-based "view" a useful perspective for strategic management research? Yes. Academy of Management Review, 26(1): 41-56. Barney, J. B. 2002. Gaining and sustaining competitive advantage (Second ed.). Upper Saddle River, NJ: Prentice Hall. Barney, J. B., & Arikan, A. M. 2001. The resource based view: origins and implications. In J. S. Harrison (Ed.), Handbook of strategic management: 124-188. Oxford UK: Blackwell publishers. Barney, J. B., & Tyler, B. 1992. Top management team attributes and sustained competitive advantage. Advances in Global High-Technology Management, 2: 33-47. Barr, P. S., Stimpert, J. L., & Huff, A. S. 1992. Cognitive change, strategic action and organizational renewal. Strategic Management Journal, 13: 15-36. Berman, S. L., Down, J., & Hill, C. W. L. 2002. Tacit knowledge as a source of competitive advantage in the national baseball association. Academy of Management Journal, 45(1): 13-31. Boeker, W. 1989. Strategic change: The effects of founding and history. Academy of Management Journal, 32(3): 489-515. Boeker, W. 1992. Power and managerial dismissal: scapegoating at the top. Administrative Science Quarterly, 37(3): 400-421. Boeker, W. 1997a. Executive migration and strategic change: The effect of top manager movement on product-market entry. Administrative Science Quarterly, 42: 213-237. Boeker, W. 1997b. Strategic change: The influence of managerial characteristics and organizational growth. Academy of Management Journal, 40(1): 152-170. Boeker, W., & Goodstein, J. 1993. Performance and successor choice: The moderating effects of governance and ownership. Academy of Management Journal, 36(1): 172-186. Cannella, A. A., & Lubatkin, M. 1993. Succession as a sociopolitical process: Internal impediments to outsider selection. Academy of Management Journal, 36(4): 763-793. Cannella, A. A., & Shen, W. 2001. So close and yet so far: Promotion versus exit for CEO heirs apparent. Academy of Management Journal, 44(2): 252-270. Castanias, R. P., & Helfat, C. E. 1991. Managerial Resources and Rents. Journal of Management, 17(1): 155-171. Castanias, R. P., & Helfat, C. E. 2001. The managerial rents model: theory and empirical analysis. Journal of Management, 27: 661-678. Caves, R. E., & Porter, M. E. 1977. From entry barriers to mobility barriers: Conjectural decisions and contrived deterrence to new competition. Quarterly Journal of Economics, 41: 241-261. Collis, D. J. 1994. How valuable are organizational capabilities. Strategic Management Journal, 15: 143-152. Cyert, R. M., & March, J. G. 1963. A behavioral theory of the firm. Englewood Cliffs NJ: Prentice-Hall. Daily, C. M., & Dalton, D. R. 1995. CEO and director turnover in failing firms: An illusion of change? Strategic Management Journal, 16: 393-400. Datta, D. K., & Guthrie, J. P. 1994. Executive succession: Organizational antecedents of CEO characteristics. Strategic Management Journal, 15: 569-577. Dierickx, I., & Cool, K. 1989. Asset stock accumulation and sustainability of competitive advantage. Management Science, 35(12): 1504-1514. DiMaggio, P. J., & Powell, W. W. 1983. The iron cage revisited: Institutional isomorphism and collective rationality in organization fields. American Sociological Review, 48: 147-160. Eisenhardt, K. M., & Martin, J. A. 2000. Dynamic capabilities: What are they? Strategic Management Journal, 21(10/11): 1105-1121. Flynn, F. J., & Staw, B. M. 2004. Lend me your wallets: The effect of charismatic leadership on external support for an organization. Strategic Management Journal, 25: 309-330. Frederickson, J., Hambrick, D. C., & Baumrin, S. 1988. A model of CEO dismissal. Academy of 169 Management Review, 13: 255-270. Friedman, S. D., & Singh, H. 1989. CEO succession and stockholder reaction: The influence of organizational context and event content. Academy of Management Journal, 32(4): 718-744. Furman, J. L., & McGahan, A. M. 2002. Turnarounds. Managerial and decision economics, 23: 283-300. Gioia, D. A., & Chittipeddi, K. 1991. Sensemaking and sensegiving in strategic change initiation. Strategic Management Journal, 12: 433-448. Goodstein, J., & Boeker, W. 1991. Turbulence at the top: A new perspective on governance structure changes and strategic change. Academy of Management Journal, 34(2): 306-330. Hambrick, D. C., & D'Aveni, R. A. 1992. Top team deterioration as part of the downward spiral of large corporate bankruptcies. Management Science, 38(10): 1445-1466. Hambrick, D. C., & Finkenstein, S. 1987. Managerial discretion: a bridge between polar views on organizations. In B. M. Staw (Ed.), Research in Organizational Behavior: 359-406. Greenwich CT: JAI Press. Hambrick, D. C., & Fukutomi, G. D. S. 1991. The seasons of a CEO's tenure. Academy of Management Review, 16(4): 719-742. Hambrick, D. C., & Mason, P. A. 1984. Upper Echelons: The organization as a reflection of its top managers. Academy of Management Review, 16: 193-206. Hambrick, D. C., & Schecter, S. M. 1983. Turnaround strategies for mature industrial-product business units. Academy of Management Journal, 26(2): 231-148. Harris, D., & Helfat, C. E. 1997. Specificity of CEO human capital an compensation. Strategic Management Journal, 18: 895-920. Hurst, D. K., Rush, J. C., & White, R. E. 1989. Top management teams and organizational renewal. Strategic Management Journal, 10(special issue): 87-105. Leonard-Barton, D. 1992. Core capabilities and core rigidities: A paradox in managing new product development. Strategic Management Journal, 13: 111-125. Lieberson, S., & O'Connor, J. F. 1972. Leadership and organizational performance: a study of large corporations. American Sociological Review, 37(2): 117-130. Lubatkin, M., Chung, K. H., Rogers, R. C., & Owers, J. E. 1989. Stockholders reactions to CEO changes in large corporations. Academy of Management Journal, 32(1): 47-68. Lubatkin, M., & Shrieves, R. E. 1986. Towards reconciliation of market performance measures to strategic management research. Academy of Management Review, 11(3): 497-512. Lynall, M. D., Golden, B. R., & Hillman, A. J. 2003. Board composition from adolescence to maturity: a multi-theoretic view. Academy of Management Review. Meyer, J. W., & Rowan, B. 1991. Institutionalized organizations: Formal structure as myth and ceremony. In P. J. DiMaggio (Ed.), The new institutionalism in organizational analysis: 41-62. Chicago: University of Chicago Press. Mintzberg, H., & Westley, F. 1992. Cycles of organizational change. Strategic Management Journal, 13: 39-59. Mueller, G. C., & Barker, V. L. 1997. Upper echelons and board characteristics of turnaround and nonturnaround declining firms. Journal of Business Research, 39: 119-134. Nelson, R. R., & Winter, S. G. 1982. Evolutionary theory of economic change. Cambridge, Massachusetts: The Belknap Press of Harvard University Press. Oliver, C. 1991. Strategic responses to institutional pressures. Academy of Management Review, 16(1): 145-179. Oliver, C. 1997. Sustainable competitive advantage: Combining institutional and resource based views. Strategic Management Journal, 18(9): 697-713. Pearce, J. A., & Robbins, K. D. 1994. Retrenchment remains the foundation of business turnaround. Strategic Management Journal, 15: 407-417. Penrose, E. T. 1959. The theory of the growth of the firm: Oxford. Peteraf, M. A. 1993. The cornerstones of competitive advantage: a resource-based view. Strategic 170 Management Journal, 14: 179-191. Porter, M. E. 1981. The contributions of industrial organization to strategic management. Academy of Management Review, 6(4): 609-620. Priem, R. L., & Butler, J. E. 2001. Is the resource-based "view" a useful perspective for strategic management research? Academy of Management Review, 26(1): 22-40. Robbins, K. D., & Pearce, J. A. 1992. Turnaround: Retrenchment and recovery. Strategic Management Journal, 13: 287-309. Romanelli, E., & Tushman, M. L. 1994. Organizational transformation as punctuated equilibrium: An empirical test. Academy of Management Journal, 37(5): 1141-1166. Rumelt, R. 1984. Toward a strategic theory of the firm. In R. Lamb (Ed.), Competitive strategic management. Englewood Cliffs, NJ: Prentice-Hall. Schwartz, K. B., & Menon, K. 1985. Executive succession in failing firms. Academy of Management Journal, 28(3): 680-686. Scott, W. R. 2001. Institutions and Organizations. Thousand Oaks CA: Sage. Shen, W., & Cannella, A. A. 2002. Revisiting the performance consequences of CEO succession: The impacts of successor type, postsuccession senior executive turnover, and departing CEO tenure. Academy of Management Journal, 45(4): 717-733. Simons, R. 1994. How new top managers use control systems as levers of strategic renewal. Strategic Management Journal, 15: 169-189. Smith, J. E., Carson, K. P., & Alexander, R. A. 1984. Leadership: It can make a difference. Academy of Management Journal, 27(4): 765-776. Staw, B. M., Lance, S. E., & Dutton, J. E. 1981. Threat-rigidity effects in organizational behavior: A multilevel analysis. Administrative Science Quarterly, 26: 501-524. Suchman, M. C. 1995. Managing legitimacy: strategic and institutional approaches. Academy of Management Review, 20(3): 571-610. Teece, D., Pisano, G., & Shuen, A. 1997. Dynamic capabilities and strategic management. Strategic Management Journal, 18: 509-534. Thomas, A. B. 1988. Does leadership make a difference to organizational performance? Administrative Science Quarterly, 33: 388-400. Tripsas, M., & Gavetti, G. 2000. Capabilities, cognition, and inertia: Evidence from digital imaging. Strategic Management Journal, 21: 1147-1161. Tushman, M. L., & Rosenkopf, L. 1996. Executive succession, strategic reorientation and performance growth: A longitudinal study in the U.S. cement industry. Management Science, 42(7): 939-953. Virany, B., Tushman, M. L., & Romanelli, E. 1992. Executive succession and organization outcomes in turbulent environments: An organization learning approach. Organization Science, 3(1): 72-91. Weiner, N., & Mahoney, T. A. 1981. A model of corporate performance as a function of environmental, organizational, and leadership influences. Academy of Management Journal, 24(3): 453-470. Weitzel, W., & Jonsson, E. 1989. Decline in organizations: A literature integration and extension. Administrative Science Quarterly, 34: 91-109. Wernerfelt, B. 1984. A resource-based view of the firm. Strategic Management Journal, 5: 171-180. Wiersema, M. F., & Bantel, K. A. 1992. Top management team demography and corporate strategic change. Academy of Management Journal, 35(1): 91-121. Wright, P. M., Smart, D. L., & McNahan, G. C. 1995. Matches between human resources and strategy among NCAA basketball teams. Academy of Management Journal, 38(4): 1052-1074. Zajac, E. J., & Westphal, J. D. 1996. Who shall succeed? How CEO/Board preferences and power effect the choice of new CEOs. Academy of Management Journal, 39: 64-90. 171 ASAC 2005 Toronto, Ontario W. Glenn Rowe Michael J. Rouse Suhaib Riaz (Student) Richard Ivey School of Business The University of Western Ontario SOURCES OF COMPETITIVE ADVANTAGE AND ABOVE NORMAL PERFORMANCE: AN EMPIRICAL TEST In this paper, we isolate a critical organizational resource – the general manager – and empirically demonstrate its linkage with value generation using a disaggregated dependent variable. In the process, we make a contribution to the literature on a number of methodological issues that have plagued empirical work in RBV: level of analysis, selection and specification of the dependent variable, controlling for industry effects and unobserved heterogeneity, and the call for longitudinal studies. The results of the study suggest that general managers with different organizations have been a potential source of competitive parity, competitive disadvantage or competitive advantage which in turn lead to normal, below normal and above normal performance, respectively. The resource-based view of the firm (RBV), while theoretically compelling, has not received empirical support for the linkage between critical resources and the generation of value (Peteraf & Barney, 2003). The RBV has been difficult to test empirically precisely because, theoretically, the tacit dimensions of resources that are valuable, rare, inimitable, and non-substitutable are, methodologically, difficult to operationalize (Barney, 1997; Barney, Wright & Ketchen, 2001). The very nature of the resources of interest makes it difficult to test using large, multi-industry samples from secondary data sources to isolate potential sources of advantage (Rouse & Daellenbach, 1999; 2002). Indeed, “…the effects of the resource-level have not been analyzed directly…” (Peteraf & Barney, 2003: 312, italics ours). In this paper we tackle this issue. We isolate a critical organizational resource and empirically demonstrate its linkage with value generation. We begin by briefly stating the core of the RBV and its empirical support focusing particularly on the empirical difficulties which our study seeks to address. We then discuss the utility and relevance of using professional sports businesses for our analysis. Next we turn our attention to the general manager of the organization as a theoretically valuable, heterogeneous resource and the focus of our attempt to isolate that resource as a potential source of organizational advantage/disadvantage/parity. The next sections set out our methodology and discuss the results. Finally, we offer some conclusions and implications of our empirical findings for a resource-based view of the firm. In summary, we examine a specific organizational resource – the general manager – which we isolate and explicitly link to performance. In so doing, we make four substantial contributions to the RBV literature: 1) empirically test the linkage between a critical organizational resource and performance; 2) demonstrate the importance of selecting and operationalizing a disaggregated dependent variable at the resource level; 3) provide support for controlling for industry effects and unobserved heterogeneity in RBV studies; and 4) underscore the importance of using longitudinal studies. 172 The resource-based view of the firm predicts that organizations generate competitive advantages by using valuable, inelastic resources (Wernerfelt, 1984; Barney, 1986, 1991; Peteraf, 1993; Ray, Barney & Muhanna, 2004). From a central insight of Penrose (1959) that organizations are heterogeneous bundles of resources, organized within an administrative framework, the RBV developed into a major analytical perspective (Wernfelt, 1984; Barney, 1991; Peteraf, 1993). The theoretical status of RBV continues to generate considerable debate (Godfrey & Hill, 1995; Barney et al., 2001; Priem & Butler, 2001; Foss & Knudsen, 2002), yet there is general agreement that the RBV is a useful perspective for strategic management research (Barney, 2001). Methodological Issues in RBV Studies While empirical tests of the major tenets of the RBV have been growing (Barney & Arikan, 2001), few stand out as rigorous RBV studies; exceptions being Henderson and Cockburn’s (1994) research in the pharmaceutical industry, and Makadok’s (1999) money market mutual fund study. However, a number of empirical issues continue to plague RBV research. Below we describe four issues that are key to the RBV that, to-date, have been problematic for researchers: 1. level of analysis 2. selection and specification of the dependent variable 3. industry effects and unobserved heterogeneity 4. need for longitudinal studies. Level of analysis Few studies have been able to isolate potential sources of competitive advantage at the resource level (Peteraf & Barney, 2003). This issue is important in any attempt to test the RBV empirically. We know that performance is generated at multiple levels: industry (Porter, 1980), strategic group (Dranove, Peteraf & Shanley, 1998) and firm level (Rumelt, 1991; Peteraf & Barney, 2003). Research also suggests that firm level effects account for a large portion of performance variation (Rumelt, 1991). Since the RBV predicts that resources within business units generate value, it is important to examine the value generated by specific, critical resources, yet these effects have not been analyzed directly (Peteraf & Barney, 2003). Precisely because the RBV asserts that the resources of interest are characterized as valuable, rare, inimitable and inelastic, research methods need to be able to tap into the resource level for analysis. Since the very nature of the resources of interest makes it difficult to test using large, multi-industry samples from secondary data sources to isolate potential sources of advantage, Rouse and Daellenbach, (1999; 2002) called for research directed explicitly at the resource-level. They further suggested that intrusive field studies within organizations might be the only way to get at and begin to understand those resources and how they might be linked to performance. On the other hand, Levitas and Chi (2002) suggest that in order to avoid some of the problems inherent in intrusive methods, researchers should strive for creativity in operationalizing constructs and empirically measuring value generation predicted by the RBV. Peteraf and Barney (2003) go so far as to suggest that no studies have been able to rise to the challenge. We believe that in this study we have risen to the challenge with some interesting results. Dependent variable Closely linked to the level of analysis is consideration of the dependent variable. Ray, Barney and Muhanna (2004) make a strong case for careful selection and operationalization of the dependent variable. Using a highly aggregated dependent variable such as firm performance may obscure 173 the contribution of a particular resource to competitive advantage. For example, one resource that has the potential to generate competitive advantage may be offset by another that has negative effects. Conceivably there are numerous other scenarios in which the potential of valuable resources is not realized due to countervailing, neutralizing, appropriation (e.g., Coff 1999), or obscuring effects. What this means is that it is theoretically more valid to operationalize the dependent variable at the same level as the resource of interest and, further, to disaggregate that dependent variable so that the potential impact of other resources and processes are controlled for (Barney 2001; Henderson & Cockburn, 1994; Priem & Butler, 2001; Rouse & Daellenbach, 1999). In this study we disaggregated the dependent variable by controlling for as many performance effects as possible in order to maximize the likelihood that the performance we capture was linked with general manager effects. Controlling for industry and unobserved heterogeneity Firm performance is influenced by industry (Dess et al., 1990). Cross-industry studies, therefore, that do not control for industry effects may lead to misleading resource effects (Makadok, 1999). An easy way to control for industry effects is to use a single industry study approach, though this may have an impact on generalizability of results. Even though Barney (2001) explicitly avoids tackling the complex issue of defining industry boundaries, partly on the basis that the resource-based logic takes the firm as its unit of analysis, industry plays an important role in organizational performance. Industry attributes have an effect on strategy decisions (Mascarenhas & Aaker, 1989) and affect organizational culture (Huff, 1982; Gordon, 1991). Hrebiniak and Snow (1980) show that perceptions of environmental uncertainty, intra-organizational influence and degree of structural decentralization vary by industry. Organizations compete within industries and share strategic factor markets (Barney, 1986). Taken together, these and similar studies suggest the importance of controlling for industry effects. Not only does industry impact organizations and their performance, but the value of a resource may be more generally context dependent (Powell, 1992; Barney, 2001; Priem & Butler, 2001). If context is considered more broadly than just industry, other controls may be required. According to Rouse & Daellenbach (1999), there may be implications of membership in a strategic group. In this study we have taken these issues on board and have limited, therefore, our study to one industry (professional sports) and one strategic group, the National Hockey League (NHL) in order to control for possible industry level and strategic group level effects. A similar argument holds for unobserved heterogeneity (Levitas & Chi, 2002). Henderson & Cockburn (1994) note that it is notoriously difficult to identify and to measure resources and other effects. To control for team level effects and unobserved heterogeneity, we included 29 organization (team) variables. Longitudinal research design Time is not only a factor in the path-dependent generation or development of resources (Dierickx & Cool, 1989), but there may be a time-lag between resource utilization and performance (Priem & Butler, 2001). Barney (2001: 51) agrees that, "dynamic research – where the conditions under which resources are developed or acquired in one period have implications for the strategic advantages of firms in subsequent periods – is particularly important…" Our study is longitudinal. It utilizes data for the population of organizations (NHL teams) for 60 years: from the 1942/43 season to the 2001/2002 season. We also measure potential for competitive disadvantage, competitive parity, and competitive advantage during the NHL regular season and associated 174 performance during the Stanley Cup playoffs, the "second season" as playoffs are called in professional sports. On the Use of Sports Teams in Organizational Analysis The empirical examination of the linkage between an isolated critical resource with value generation requires a sample that meets three criteria. First, we wanted a single industry from which to select a definable, valid strategic group in order to control for as much context dependent variation as possible. Research at the level of strategic groups controls for some variation in strategies available to competitors. By using a strategic group composed of professional sports teams, that control is arguably greater. All teams are regulated by the same set of game and other rules (e.g., player trades and selection) which are closely monitored by a powerful, neutral authority and by competitors, their fans and pundits. This, in effect, generates an enforced, shared consensus about the types of actions and strategies that can be used (Wright, Smart & McMahan, 1995). Second, sports teams have the advantage of sharing fundamental objectives – maximize games won and win the Stanley Cup – to an extent that is not necessarily so for businesses which may have more complex goals. Third, the study demanded a comprehensive, longitudinal data set, that was fine grained enough both to provide data at the resource level and that would enable us to disaggregate the dependent variable. Finally, professional sports constitute a recognizable industry. Single industry studies such as those in airlines (Gimeno & Woo, 1996), banks (Fox-Wolfgramm, Boal & Hunt, 1998), hotels (Ingram & Inman, 1996; Baum & Ingram, 1998), and trucking (Silverman, Nickerson & Freeman, 1997), have made positive contributions to general management research. The Role of the General Manager The general manager of an NHL team occupies a position central to the fortunes of the team. The general manager is responsible for building the actual team that competes in its respective division each year. This involves selecting and hiring (and firing) the coach as needed, determining which players to select in the annual draft, trading players and negotiating the terms of contract with each player. The general manager is thus the senior person responsible for making and implementing the strategic and operational decisions that are critical to the operation of the team in the NHL. While earlier studies focusing on sports settings have looked at various position holders within sports teams, the role of the general manager in sports settings, despite its obvious importance, has been the most under-researched. The above explication of the role of the NHL general manager suggests that the role is conceptually comparable to the top management / CEO of other types of business organizations, e.g., directing and organizing resources and activities, and making and implementing strategic and operational decisions (Castanias & Helfat, 1991). Further, the CEO / general manager has the potential to generate rents through his or her activities (Castanias & Helfat, 1991). In the case of NHL teams specifically, such an executive is clearly the general manager of the team. The general manager can therefore be conceptualized as a managerial resource that theoretically has been identified as a key contributor to the organization's ability to generate rents (Castanias & Helfat, 2001). 175 The managerial resource would be expected to vary across teams / firms in terms of quality because of variation in the type and degree of skill and skillfulness (Castanias & Helfat, 1991; 2001), and they may differ in terms of the specific combination of skill types and levels they possess (Bailey & Helfat, 2003). Since differences in managerial skills theoretically leads to differences in the variability of firm performance (Castanias & Helfat, 2001), in the sense that a particular resource is a source of performance that is aggregated into firm performance, we expect that general managers of NHL teams would have a differential impact on team performance, especially after firm performance is disaggregated into performance associated with the particular resource. It is the differential disaggregated impact of this organizational resource – the general manager – that we investigate in this paper. Methods As stated earlier, the purpose of this study is to isolate a critical resource and study its impact on a disaggregated dependent variable, and thereby, test the assertion in Barney (1997, 2002) and Peteraf and Barney (2003) that resources which are sources of competitive disadvantage lead to below normal performance, resources which are sources of competitive parity lead to normal performance, and resources which are sources of competitive advantage (temporary or sustained) lead to above normal performance. To test this assertion we needed to operationalize competitive disadvantage, competitive parity and competitive advantage. In this section we describe the methodology we used to achieve this. We started with the observation that if a resource creates value less than that expected by the owners of that resource, it is a source of competitive disadvantage. In addition, if a resource creates value equal to that expected by the owners of a resource then that resource is a source of competitive parity. Finally, if a resource creates value greater than that expected by the owners of the resource it is a source of competitive advantage (Barney, 1997, 2002). We decided to use the residuals from the model described below as our starting point to operationalize value. To obtain these residuals, we regressed the number of points gained divided by the number of points available during the regular season (our dependent variable) on the independent variables described below. We considered that the residuals from this analysis closely approximate value created versus expected value. Similar to proponents of the RBV, we believe that value created is similar to actual values and expected value is similar to predicted values. Therefore, using the residual is a good approximation of value creation. As mentioned, we used the number of points gained divided by the number of points available during the regular season as our measure of value. In addition, we used playoff performance, or how far a team went in the post-season, as our measure of performance. Note that even though NHL teams are business organizations, we have purposely not used "firm" performance in terms of some profitability metric precisely because a disaggregated dependent variable is required to disentangle critical resource effects (Peteraf & Barney, 2003). We used playoff performance as our measure of performance since this ‘second season’ is generally accepted as the ultimate measure of performance for NHL teams and, further, would seem to be a more valid outcome of interest since, on the face of it, it seems more directly linked to the general manager resource than an aggregated measure of organizational performance. This is consistent with Peteraf and Barney’s (2003) argument that a firm’s competitive advantage is an “indicator of the firm’s potential [italics in original] to best its rivals in terms of rents, profitability, market share, and other outcomes of interest.” For us the regular season demonstrates the potential for an outcome of interest – performance in the Stanley Cup playoffs. 176 As mentioned earlier, the resources we focus on are the general manager (GM) of each NHL team from the 1942/1943 season to the 2001/2002 season. We chose the general manager because he (no woman has ever served as the general manager of a hockey team so we use the male pronoun in our paper) is the senior person in each team responsible for hockey operations. The purpose of our analyses, therefore, was to develope residual measures for the regular season and post-season after including as many explanations as possible for performance in the regular season and in the post-season. We wanted to disaggregate the impact of the general manager on each team’s regular season potential for value creation and on each team’s post-season performance. General Managers in the NHL During the sixty seasons of our data set, there were 122 general managers and 166 general manager spells (a spell is a measure of time viz. GM tenure/season). The spells ranged from a partial season (coded as 0.5) to 28.5 seasons. The average spell was 5.61 seasons. The average total time as a general manager was 7.64 seasons. One general manager, Bobby Clarke, had the most spells as a GM in that he filled this position for Philadelphia twice (1984/1985 to 1989/1990 and 1994/1995 to present), for Dallas once (1990/1991 to 1991/1992) and for Florida once (1993/1994) for a total of four spells. There were eight GMs who had three spells each, twenty-seven who had two spells each and eighty-four who had one spell each. No GM worked for more than three different teams. There were six GMs, including Bobby Clarke, who worked for three teams, twenty-five who worked for two teams and ninety-one who worked for one team. We considered that GMs with less than three full, contiguous years as a GM would only allow us to average residuals over two years and that it would be better to have at least three full, contiguous years to average for each GM. There were 80 GMs who had a total of 96 GM spells and who met this stringent criterion. One GM, Emile Francis had three stints of eleven seasons, seven seasons and six seasons with three different teams. There were another fourteen GMs who had two stints of three or more years and sixty-five who had one stint of three or more years. It was this group of 80 GMs we used to test the assertion that competitive disadvantage leads to below normal performance, competitive parity leads to normal performance and competitive advantage leads to above normal performance. The Sample As already mentioned, we examined teams from the National Hockey League, observed continually, from the 1942–1943 through to the 2001–2002 seasons. We collected the data from Total Hockey: The Official Encyclopedia of the National Hockey League (2nd Edition), published in 2000, the annual NHL Official Guide & Record Books, and Total NHL (2003) all published by Dan Diamond and Associates, Inc. Where information was not available from these publications, we sourced data from the NHL website, and from individual teams. Data were collected from The Internet Hockey Database (www.hockeydb.com) to develop number of assists (an assist is a league-standard measure for a player who helped another player score a goal for the team) and team member turnover variables. We dropped the Oakland Seals team (entered the NHL in the 1967-1968 season) from the analysis because it folded after the 1977–1978 season. We subsequently included thirty teams in the study, even though only six were present in the 1942–1943 season. When team names were changed or teams changed cities or both, we treated them as the same team. During the 60-year 177 period, there were 878 team-season observations. We dropped eleven when we dropped the Oakland Seals. In addition, we dropped 24 team-seasons because they were first-year teams. This left 843 team-season observations in our data set. The data were ordered in alphabetical order starting with the Anaheim Mighty Ducks and finishing with the Washington Capitals. For each team the data were ordered from 1942–1943 to 2001–2002. For all teams except the “Original Six,” the data were ordered from the second year they entered the NHL. The Analyses We conducted two analyses. In the first analysis, we regressed the number of points gained by a team versus the number of points available (we called this actual value created) during the regular season on several variables to explain as much variance as possible. The independent variables controlled for regression to the mean, the external environment, and internal team attributes. In the second analysis, we regressed how far each team progressed in the playoffs on the same variables we used in our first analysis with one exception. In the second analysis we used lagged playoff performance not lagged actual value created. Actual value: Our dependent variable was measured as the number of wins times two plus the number of tied games with the total divided by the number of games played times two. We used this measure because the number of games played changed from 50 in 1942/1943 to 82 in 2001/2002. There were only 48 games in 1994/1995 because of an owners’ lockout and prior to this season teams played 84 games for several seasons. This measure intrinsically gauges the level of performance, relative to other teams in the same season, and is consistent with other sports-team studies (Brown, 1982; Cannella & Rowe, 1995; Pfeffer & Davis-Blake, 1986; Rowe et al., 2005). Lagged actual value: This variable was measured the same as the dependent variable in the first analysis and was included to control for regression to the mean. Playoff performance: For each season we coded this variable as follows. When a team won the Stanley Cup we gave this variable a value of 4; when a team lost in the Stanley Cup finals, we gave this variable a value of 3; when a team made it into the playoffs but did not make it into the final series, we gave this variable a value of 2; and when a team did not make it into the playoffs we gave this variable a value of 1. We treat this variable as a continuous variable because it is univariate normal (see Table 1). Lagged playoff performance: This variable was measured the same as the dependent variable for the second analysis and was included to control for regression to the mean. League reorganization: The league expanded from six teams to thirty teams over the period of our study. Therefore, each season teams were added we coded reorganization as 1; otherwise it was coded as 0. This was important as we could not use the first year teams that were added to the league in our study as we needed to use the data from each new team’s first year to develop several of our variables. This meant that in these years the average points gained in the regular season relative to those available of the older teams left in our study would be higher than in other years when new teams were not added. In addition, we expected that newer teams would be likely to achieve worse performance by not advancing as far into the playoffs. Location: During the 60 seasons of our data set, several teams changed cities. When they did, we treated these as the same teams. However, we controlled for change of location by coding all observations for each team after a location change as one and all observations for those teams and all other teams as zero. 178 Intensity of rivalry: Sports teams compete in a manner that makes the level of rivalry among teams an important concern. We define rivalry in our study as the degree to which competitors in the same division have equal capabilities. In the NHL, as in other professional sports leagues, teams are arranged into divisions for league play, and, although teams sometimes play other teams outside of their division, competition within the division is the most important factor for determining the final number of points achieved. We measured intensity of rivalry in the following manner. For each division, each season, we calculated the standard deviation of performance for all teams in the division. Then we divided this value by the idealized standard deviation as recommended by Quirk and Fort (1992). The idealized standard deviation presumes that each team has playing strength equal to all other teams. It equals 0.5 divided by the square root of the number of games played during the regular season. The number varied from 50 games played in the 1942–1943 season to 84 in the early 1990s. By the end of the 2001–2002 regular season each team played 82 games. Our measure of rivalry estimates the equality of proficiency across teams that play in the same division and are, therefore, in direct competition. The intensity of rivalry in our sample varied across divisions as well as across seasons. All stars: The quality of players can affect the number of points a team gains during the regular season and how far they progress in the playoffs. Consequently, we coded this variable as the number of players voted to the all-star teams by the Professional Hockey Writers’ Association at the end of each current regular season from each team. Player turnover: We coded player turnover in the following manner. We counted the number of players on each team at the end of each season. We then assessed the number of players from this group no longer on the team at the beginning of each current season. We divided the latter by the former to determine the percentage of player turnover from the end of each previous season to the beginning of each current season. Teamwork: Teamwork was measured as the number of assists divided by the number of goals scored for each team for each current season. We considered that teams with better teamwork would score more points during the regular season and progress further in the playoffs. Coach ability: We measured coach ability at the end of each previous season as the proportion of regular season points gained divided by the regular season points available for the total number of games coached during his entire career. In the instances where a coach had no previous NHL head coaching experience, we assigned a value of 0.500. New Coach: When we assigned a value of 0.500 because the coach was new to the NHL, we included a dummy variable to indicate the coach’s first season in the NHL. A "1" indicated that the coach was in his first season, and a "0" indicated otherwise. Coach tenure: Coach tenure was operationalized as the number of games coached with the same team during that current coaching spell, and was measured at the end of each previous season. GM and coach same person: For some teams in some seasons, the coach and general manager was the same person. When this was the case, we coded this dummy variable as "1" and "0" otherwise. 179 Succession variables: We included six succession variables in our study to control for any impact of succession on points gained and progression in the playoffs. For teams where the coach at the end of the previous season was different from the coach at the beginning of the previous season, coach previous-season succession was coded as "1" otherwise it was coded as "0". When the coach changed during the period from the end of the previous season to the beginning of the current season we coded coach between-season succession as "1", otherwise it was coded as "0". When the coach changed during the current season, we coded coach within-season succession as "1", otherwise it was coded as "0". The same procedure was followed for general manager previous-season, general manager between-season, and general manager within-season successions. There were 155 within-season, 216 between-season, and 158 previous-season coach successions in our data set. In addition, we found 36 within-season, 102 between-season, and 32 previous-season GM successions. Team: To control for team-level effects and unobserved heterogeneity we included 29 team variables with the Buffalo Sabres as our reference group. We randomly chose the Sabres as the fixed-effects procedure requires that the number of dichotomous variables used be one less than the number of teams. These 29 team variables were not included in our tables to conserve space. Time: We considered that it was necessary to control for calendar time (operationalized as 1 to 60 for each year in the data set). Consequently, we adjusted the data set for the effect of calendar time prior to running the two analyses. All continuous variables were regressed on calendar time. When the association between each of these variables and calendar time was significant we used their residuals for our two analyses. We did this for two reasons: to reduce the variance in the data set due to calendar time, and because there was an increase in the number of teams in the league during our observation period. We found that seven of the ten continuous variables related significantly to calendar time. These were playoff performance, lagged playoff performance, intensity of rivalry, all stars, player turnover, teamwork, and coach ability. Actual value created, lagged actual value created and coach ability were not associated with time. Analytical methodology: We used a panel data and regression methodology to analyze our data. The data were structured as an unbalanced, pooled, cross-sectional time series with the data sequenced by team and within team, by season. By controlling for calendar time we were able to treat all 843 observations as independent and homoscedastic and use OLS regression as our analytical methodology. We did not find multicollinearity to be a problem in our two analyses (Aiken & West, 1991) as the variance inflation factors ranged from 1.0 to 3.0. This is well below 10.0, the accepted maximum. A visual check for heteroscedasticity revealed that this was not a problem. Finally, the Durbin-Watson for our multivariate analyses revealed that autocorrelation was not a problem. In the two analyses, the Durbin-Watsons were 1.991 and 2.108, respectively, indicating no first-order autocorrelation. With these three key problems (autocorrelation, heteroscedascity, and multicollinearity) resolved, we felt confident in using OLS regression for our analyses. Between analysis one and analysis two: After analysis one, we saved the residuals as the operationalization of actual value created minus value expected to be created. Then we identified the GM for each team for each season and averaged the residuals for each GM with three or more full contiguous seasons with the same team. For example, Harry Sinden was the GM for the Boston Bruins for 28.5 seasons. We averaged the residuals for the 28 full seasons that he was Boston’s GM. In this manner we associated each of our 96 GM’s spells that were three full seasons or longer with an average residual value. We then compared each of these averages to the average of all other 180 residual values. For example, we compared the average of Harry Sinden’s 28 seasons to the average of the other 817 residual values. Where a spell was three seasons, we compared the average of the three seasons to the average of the other 840 residual values. We used the appropriate one-tailed t-test to make the comparison after assessing whether the variance for each of the two groups in each t-test was equal or unequal. When a GM’s residual average was significantly less than the residual average of all other observations, we coded all of his years as one signifying that he was a source of competitive disadvantage. We coded five GM spells this way for a total of 35 team-season observations. When the GM’s average residual equalled the residual average of all other observations, we coded all of his years as two signifying that he was a source of competitive parity. We coded 84 GM spells this way for a total of 616 team-season observations. When the GM’s average residual was significantly greater than the residual average of all other observations, we coded all of his years as three signifying that he was a source of competitive advantage. We coded seven GM spells this way for a total of 77 team-season observations. The 115 observations that were left were coded as zero and signified all observations associated with GM spells of less than three full contiguous seasons. Analysis two: As we mentioned above, we regressed playoff performance on lagged playoff performance and all of the other independent variables described above with the exception of lagged actual value. From this analysis we saved the residuals as we consider them to be a close approximation of performance as described by Barney (1997, 2002) in that low residuals are similar to below normal performance, medium residuals are similar to normal performance and high residuals are similar to above normal performance. After analysis two: We used a oneway analysis of variance test to assess whether the residuals from analysis two were significantly different for groups one, two and three as identified in analysis one. For groups two and three, the variances were unequal so we used Dunnett’s T3 test. For groups one and two, the variances were equal so we used Scheffe’s and Bonferroni’s tests. Results Table 1 reports the mean, standard deviation, skewness and kurtosis for all continuous variables used in the two multivariate analyses. As mentioned earlier, seven variables were significantly related to calendar time and the standardized residuals saved from regressing each of these variables on calendar time were used in the multivariate analyses. We also report the skewness and kurtosis for all seven standardized residual variables. The values for skewness and kurtosis in Table 1 indicate that each continuous variable was univariate normal. -------------------------------------------------Insert Tables 1-3 about here -------------------------------------------------- 181 Table 2 contains the results from our two multivariate analyses. We report unstandardized betas. Both models were significant with the model from analysis one explaining 61.9 percent of the variance in actual value created and the model from analysis two explaining 34.3 percent of the variance in playoff performance. To conserve space we do not comment on these models as they are not the intent of this study. Table 3 presents the results of the oneway analysis of variance. We report the number of observations, means, standard deviations and whether groups were a source of competitive disadvantage, parity or advantage. In addition, we report the results of the contrast tests and post hoc tests used to ascertain whether the residualized playoff performance (after controlling for lagged playoff performance, several environmental and organizational variables and time) was significantly better for Group 2 than Group 1 and for Group 3 than Group 2. Using two-tailed significance tests we see that Group 2’s average residualized playoff performance is better than Group 1’s and that Group 3’s is better than Group 2’s. Discussion What do these results suggest about the relationship between a critical organizational resource that is a source of competitive disadvantage, parity or advantage, and associated performance? Looking at Table 4, these results support the argument made by Barney (1997, 2002) and Barney and Peteraf (2003) that a source of competitive disadvantage will be a source of below normal performance, a source of normal performance will be a source of competitive parity, and a source of competitive advantage will be a source of above normal performance. This is an important finding. It supports the heretofore assertion of the RBV that – at the resource-level – critical resources have the potential to generate competitive advantage and be a source of above normal performance. The results also suggest that it is uncommon for a GM to be either a source of competitive disadvantage or a source of competitive advantage. In the case of competitive advantage particularly, it suggests that GMs who generate competitive advantage are relatively rare. These results also suggest that the competitive advantage generated by these GMs may be non-transferable. This is suggested from Table 3 where we see that none of the GMs who were a source of competitive advantage for their team were a source of competitive advantage for another team. In the case of four of them (Allen, DelVellano, Pollock,and Schmidt) they only had a GM spell with one team. The other three had multiple GM spells (Bowman – St. Louis and Buffalo; Fletcher – Calgary and Toronto; and, Francis – New York Rangers, St. Louis and Carolina). This may suggest that in the context of the NHL it is very difficult for a GM to transfer the tacit set of skills, knowledge and abilities that made him a source of competitive advantage and above normal performance with one team. Another interesting note from Table 3 is that Frank Selke is found to be a source of competitive disadvantage even though he is widely considered as one of the great GMs in NHL history. During his 18 year tenure as GM, Montreal won six Stanley Cups and was in the finals another five times. In addition, he led Montreal to eight first place finishes and six second place finishes in a tough six team league in regular season play. However, being a great GM is different from being a source of competitive advantage, competitive parity or competitive disadvantage (Barney & Tyler, 1991). He was found to be a source of competitive disadvantage in our study because, on average, the actual value created was significantly less than the expected value. His average residual regular season value was negative. This indicates that during regular season play 182 he achieved less with the players and coaches he brought to Montreal than he should have achieved as predicted by the model in analysis one. Some limitations of our study should be recognized. Our study and its findings only addressed the linkage between the GM as an example of a critical organizational resource and performance within the context of the NHL. This means that generalizability to competitive advantages of CEOs of large corporations and other organizations is limited. Internal validity is, however, a necessary condition for external validity and limiting our study to a single industry and strategic group provides many advantages in terms of internal validity: objective measures of disaggregated performance, longitudinal data, and the ability to control for many potentially confounding and alternative explanations of performance variation. Given these strengths, we can make very strong inferences based on our results and while we may not be able to generalize to other kinds of organizations, we can generalize back to theory (Stinchcombe, 1968), thereby claiming empirical support for a core theoretical concept of the RBV. The RBV has been difficult to test empirically because of methodological issues associated with untangling resource effects at the resource-level and linking those to a theoretically appropriate measure of performance (Barney, 1997; Barney et al., 2001; Peteraf & Barney, 2003). Following the advice of Rouse and Daellenbach (1999), we used a longitudinal data set confined to a sole strategic group within a single industry to control for as much extraneous variation as possible. However, we stopped short of Rouse and Daellenbach's (1999) full methodology and as suggested by Levitas and Chi (2002), tried to be creative in operationalizations and in measuring theorized outcomes in order to avoid intrusive research methods. The results suggest that we were successful even though our data are unique in many respects and such data are not available for most industries. Our results suggest that we were successful in demonstrating empirically the link between a critical resource and that resource being a source of either below normal, normal or above normal performance. Our results underscore the importance of selecting and operationalizing a disaggregated dependent variable that is theoretically relevant at the resource-level. The results allowed us to make some inferences because we were able to control for industry and strategic group effects as well as unobserved heterogeneity. In the end we think that our results provide, perhaps the first empirical support, for the theoretical assertion that some resources have the potential to generate competitive advantages, and by extension to explain inter-firm performance differences, ceteris paribus, that are directly attributable to differences in resources. References Aiken, L. S., & West, S. G. 1991. Multiple Regression: Testing and Interpreting Interactions. Newbury Park, CA: Sage. Allen, M. P., Panian, S. K., & Lotz, R. E. 1979. Managerial succession and organizational performance: A recalcitrant problem revisited. Administrative Science Quarterly, 24: 167-180. Bailey, E. E., & Helfat, C. E. 2003. External management succession, human capital and firm performance: An integrative analysis. Managerial and Decision Economics, 24: 347-369. 183 Barney, J. B. 1986. Strategic factor markets: Expectation, luck, and business strategy. Management Science, 32(10): 1231-1241. Barney, J. B. 1991. Firm resources and sustained competitive advantage. Journal of Management, 17(1): 99-120. Barney, J. B. 1997. Gaining and Sustaining Competitive Advantage. Upper Saddle River, NJ: Prentice Hall. Barney, J. B. 2001. Is the resource-based "view" a useful perspective for strategic mnagemetn research? Yes. Academy of Management Review, 26(1): 41-56. Barney, J. B., & Arikan, A. M. 2001. The resource-based view: origins and implications. In Hitt, M. A., Freeman, R. E., & Harrison, J. S. (eds), Handbook of Strategic Management, Blackwell: Oxford: 124-188. Barney, J. B., Wright, M., & Ketchen Jr., D. J. 2001. The resource-based view of the firm: Ten years after 1991. Journal of Management, 27: 625-641. Baum, J. A. C., & Ingram, P. 1998. Survival-enhancing learning in the Manhattan hotel industry. Management Science, 44(7): 996-1016. Brown, M. 1982. Administrative succession and organizational performance: The succession effect. Administrative Science Quarterly, 27: 1-16. Cannella, Jr., A. A. & Rowe, W. G. 1995. Leader capabilities, succession and competitive context: A baseball study. The Leadership Quarterly, 6(1): 69-88. Castanias, R. P., & Helfat, C. E. 1991. Managerial resources and rents. Journal of Management, 17(1): 155-171. Castanias, R. P., & Helfat, C. E. 2001. The managerial rents model: Theory and empirical analysis. Journal of Management, 27:661-678. Coff, R. W. 1999. When competitive advantage doesn't lead to performance: The resource-based view and stakeholder bargaining power. Organization Science, 10: 119-132. Dess, G., Ireland, R. D., & Hitt, M. A. 1990. Industry effects and strategic management research. Journal of Management, 16: 7-27. Dierickx, I., and Cool, K. (1989) Asset stock accumulation and sustainability of competitive advantage. Management Science, 35(12): 1504-1511. Dranove, D., Peteraf, M., & Shanley, M. 1998. Do strategic groups exist? An economic framework for analysis. Strategic Management Journal, 19: 1029-1044. Eitzen, D. S., & Yetman, N. R. 1972. Managerial change, logevity, and organiztional effectiveness. Administrative Science Quarterly, 17: 110-116. Foss, N. J., & Knudsen, T. 2002. The resource-based tangle: Towards a sustainable explanation of competitive advantage. Managerial and Decision Economics, 24: Fox-Wolfgramm, S. J., Boal, K. B., & Hunt, J. G. 1998. Organizational adaptation to institutional change: A comparative study of first-order change in prospector and defender banks. Administrative Science Quarterly, 43(1): 87-126. Gamson, W., & Scotch, N. 1964. Scapegoating in baseball. American Journal of Sociology, 70: 69-76. 184 Gimeno, J., & Woo, C. Y. 1996. Hypercompetition in a multimarket environment: The role of strategic similarity and multimarket contact in competitive de-escalation. Organization Science, 7(3): 322-341. Godfrey, P. C., & Hill, C. W. L. 1995. The problem of unobservables in strategic management research. Strategic Management Journal, 16: 519-533. Gordon, G. W. 1991. Industry determinants of organizational culture. Academy of Management Review, 16(2): 396-415. Henderson, R., & Cockburn, I. 1994. Measuring competence? Exploring firm effects in pharmaceutical research. Strategic Management Journal, 15(1): 63-84. Hrebiniak, L. G. & Snow, C. C. 1980. Industry differences in environmental uncertainly and organizational characteristics related to uncertainty. Academy of Management Journal, 23(4): 750-759. Huff, A. S. 1982. Industry influence on strategy reformation. Strategic Management Journal, 3: 119-131. Ingram, P., & Inman, C. 1996. Institutions, intergroup competition, and the evolution of hotel populations around Niagra Falls. Administrative Science Quarterly, 41(4): 629-658. Khanna, N., & Poulsen, A. B. 1995. Managers of financially distressed firms: Villians or scapegoats. The Journal of Finance, L(3): 919-940. Levitas, E., & Chi, T. 2002. Rethinking Rouse and Daellenbach's rethinking: Isolating vs. testing for sources of sustainable competitive advantage. Strategic Management Journal, 23: 957-962. Lewis, M. 2004. Moneyball. New York: Norton. Makadok, R. 1999. Interfirm differences in scale economies and the evolution of market shares. Strategic Management Journal, 20(10): 935-952. Mascarenhas, B. & Aaker, D. A. 1989. Strategy over the business cycle. Strategic Management Journal, 10: 199-210. Penrose, E. T. 1959. The Theory of the Growth of the Firm. New York: Wiley. Peteraf, M. A. 1993. The cornerstones of competitive advantage: A resource-based view. Strategic Management Journal, 14(3): 179-192. Peteraf, M. A., & Barney, J. B. 2003. Unraveling the resource-based tangle. Managerial and Decision Economics, 24: 309-323. Pfeffer, J. & Davis-Blake, A. 1986. Administrative succession and organizational performance: How administrator experience mediates the succession effect. Academy of Management Journal, 29: 72-83. Porter, M. E. 1980. Competitive Strategy: Techniques for Analyzing Industries and Competitors. New York: Free Press. Powell, T. C. 1992. Strategic planning as competitive advantage. Strategic Management Journal, 13: 551-558. Priem, R. L. & Butler, J. E. 2001. Is the resource-based "view" a useful perspective for strategic management research? Academy of Management Review, 26(1): 22-40. Quirk, J., and Fort, R. D. 1992. Pay Dirt: The Business of Professional Team Sports. Princeton, NJ: Princeton University Press. 185 Ray, G., Barney, J. B., & Muhanna, W. A. 2004. Capabilities, business processes, and competitive advantage: Choosing the dependent variable in empirical tests of the resource-based view. Strategic Management Journal, 25(1): 23-37. Rouse, M. J., & Daellenbach, U. 1999. Rethinking research methods for the resource-based view: Isolating sources of sustainable competitive advantage. Strategic Management Journal, 20: 487-494. Rouse, M. J., & Daellenbach, U. 2002. More thinking on research methods for the resource-based perspective. Strategic Management Journal, 23(10): 963-969. Rowe, W. G., Cannella, Jr., A. A., Rankin, D., & Gorman, D. 2005. Leader succession and organizational performance: Integrating the common-sense, ritual scapegoating and vicious-circle succession theories. Leadership Quarterly, (Forthcoming). Rumelt, R. P. 1991. How much does industry matter? Strategic Management Journal, 12: 167-185. Schroeder, R. G., Bates, K. A., and Junttila, M. A. 2002. A resource-based view of manufacturing strategy and the relationship to manufacturing performance. Strategic Management Journal, 23(2): 105-117. Silverman, B. S., Nickerson, J. A., & Freeman, J. 1997. Profitability, transactional alignment, and organizational mortality in the U. S. trucking industry. Strategic Management Journal, 18: 31-52. Stinchcombe, A. 1968. Constructing Social Theories. Chicago: University of Chicago Press. Wernerfelt, 1984. A resource-based view of the firm. Strategic Management Journal, 5(2): 171-180. Wright, P. M., Smart, D. L., & McMahan, G. C. 1995. Matches between human resources and strategy among NCAA basketball teams. Academy of Management Journal, 38: 1052-1074. 186 Table 1 Means, Standard Deviations, Skewness and Kurtosis for all Continuous Variables Variable Actual value (t+1) Actual value (t-1) (R) Playoff performance (t+1)1 (R) Playoff performance (t-1)1 (R) Rivalry (t+1)1 (R) All stars (t+1)1 (R) Player turnover (t)1 (R) Teamwork (t+1)1 (R) Coach ability (t-1)1 Coach tenure (t-1) Mean .506 .502 1.900 1.890 1.959 .850 .321 1.626 .513 138.6 Std Dev .113 .116 .813 .815 .707 1.249 .114 .097 .085 169.3 Skew -.100 -.147 .990 .998 .397 1.706 .371 -2.432 -.273 2.111 Kurtosis -.075 .026 .918 .923 .224 2.712 .333 8.732 .731 5.174 Skew2 Kurtosis2 .854 .863 .422 1.258 .289 -1.280 .783 .786 .201 1.732 .158 5.297 1.852 4.646 1) (R) means that the variable was regressed on calendar time and the standardized residual used in the multivariate analysis. 2) The skewness and kurtosis for all residualized continuous variables after they were regressed on calendar time. This means that all variables used in the analyses had acceptable levels of skewness (< 2) and kurtosis (<7). 187 Table 2 OLS Regressions Variables Intercept Lagged actual value (R) Lagged playoff performance League reorganization Location (R) Intensity of rivalry (R) All stars (R) Player turnover (R) Teamwork Coach ability Coach new (R) Coach tenure GM Coach same person Coach previous-season Coach between-season Coach within-season GM previous-season GM between-season GM within-season R2 Adjusted R2 F Durbin-Watson Analysis One Dependent Variable Actual value (t+1) B Sig. .285 .000 .405 .000 ----.015 .009 .024 .069 .002 .351 .040 .000 .003 .378 .004 .122 .080 .032 .014 .136 .003 .359 .003 .737 .016 .026 .002 .799 -.048 .000 .023 .086 -.003 .676 -.038 .003 .639 .619 30.7 1.993 .000 188 Analysis Two Dependent Variable (R) Playoff performance (t+1) B Sig. -.918 0.001 ----.096 .010 -.002 .977 .252 .101 .023 .432 .307 .000 -.056 .088 .026 .387 1.766 .000 -.007 .953 .100 .007 .015 .871 -.080 .323 .149 .117 -.319 .000 .108 .482 -.029 .747 -.264 .075 .379 .343 10.6 2.108 .000 Table 3 The Oneway Analysis of Variance Group N 115 35 616 77 Mean -.059 -.489 -.019 .463 Std Dev .718 1.042 .965 1.157 Contrast Tests Groups 1 & 2 Groups 2 & 3 Variance Equal Unequal Value of Contrast .470 .482 t-test 2.824 3.505 Sig (2 Tail) .005 .001 Post Hoc Tests Groups 1 & 2 Groups 2 & 3 Variance Equal Unequal Scheffe – Sig .047 --- Bonferroni - Sig .029 --- Dunnett T3 – Sig --.004 0 1 2 3 189 CDA, CP or CA CDA CP CA ASAC 2005 Toronto, Ontario Richard Soparnot Groupe ESCEM Tours-Poitiers Mail : [email protected] Eric Stevens Groupe ESCEM Tours-Poitiers Mail : [email protected] PEUT-ON OPÉRATIONNALISER L’APPRENTISSAGE ORGANISATIONNEL ? UNE ANALYSE DU PROCESSUS D’INNOVATION DE SERVICES Le concept d’apprentissage se situe à la confluence de nombreux champs d’étude (la gestion de projet, le changement organisationnel, l’innovation, le management des savoirs, le benchmarking…). Il apparaît comme un concept « carrefour ». Situé à l’interface de nombreuses pratiques de management, il est difficile de saisir l’influence du phénomène d’apprentissage sur ces pratiques. De même, l’abondance des écrits sur l’apprentissage rend peu aisée une délimitation précise du concept. Afin de mieux cerner l’objet et évaluer son influence sur les pratiques managériales, une réflexion sur son opérationnalisation nous paraît essentielle. Dans cette perspective, nous recourons à « l’observation » de situations d’apprentissage : les processus d’innovation constituent de tels « moments organisationnels ». De là, nous formulons une série de propositions sur l’opérationnalisation de l’apprentissage dans la perspective des processus d’innovation de services. Introduction Le concept d’apprentissage a fait l’objet d’une littérature particulièrement abondante (Leroy, 1998). Pour reprendre l’expression de Koenig (1996) à propos du changement, l’apprentissage se caractérise par un ensemble de travaux « dont la richesse et la variété sont parfois confondantes ». Même s’il renvoie à la Resource Based View (RBV) et notamment la branche des Dynamic Capabilities (DC) (Durand 1 , 2001), le concept d’apprentissage se situe à la confluence de nombreux champs d’étude (la gestion de projet, le changement organisationnel, l’innovation, le management des savoirs, le benchmarking…). L’apprentissage apparaît ainsi comme un concept « carrefour », situé à l’interface de nombreuses pratiques de management. Il est souvent difficile de saisir l’influence du phénomène d’apprentissage sur ces pratiques. De même, l’abondance des écrits sur l’apprentissage rend difficile une délimitation précise du concept (Leroy, 1998). Tout d’abord, la nature immatérielle de cette aptitude la rend insaisissable, difficilement contrôlable, voire parfaitement évanescente. Le manque de clarté sur l’objet théorique justifie cette contribution. Afin de mieux cerner l’objet et évaluer son influence sur les pratiques managériales, une réflexion sur son opérationnalisation nous paraît essentielle. Cela contribuerait à stabiliser l’objet théorique et à le rendre plus actionnable. Dans cette perspective, nous recourons à « l’observation » de situations d’apprentissage : les processus d’innovation constituent de tels « moments organisationnels ». Le projet apparaît alors comme un support à l’apprentissage. Afin 1 L’auteur note que « le courant des ressources dynamiques tente de repenser à la fois les origines de la compétitivité et les moyens de gérer le futur. L’apprentissage, tant individuel qu’organisationnel, est à la base de la création des ressources et des aptitudes stratégiques » (Durand, 2001, p.150). 190 de réaliser un « premier repérage » des variables opératoires du phénomène, nous nous attachons, donc d’abord, à clarifier les contours théoriques du concept d’apprentissage. Puis, nous soulignons la nature des liens qu’entretiennent innovation et apprentissage en nous fondant sur le corpus théorique. De là, nous formulons une série de propositions sur l’opérationnalisation de l’apprentissage dans la perspective des processus d’innovation de services. I L’apprentissage : quels contours théoriques ? La variété théorique du concept d’apprentissage en rend l’opérationnalisation particulièrement difficile. Les premiers pas vers son opérationnalisation nécessitent de mettre en évidence le sujet et l’objet d’apprentissage (qui apprend quoi ?), les modes d’apprentissage (comment apprendre ?) et les déclencheurs de l’apprentissage (quand apprendre ?). Dans un dernier temps, nous verrons si le concept d’organisation « apprenante » permet une avancée dans l’agrégation des différentes dimensions du concept. Ce recours à la littérature nous permet de mieux saisir les différentes facettes du concept. 1.1 Les sujets et objets d’apprentissage : de l’individu à l’organisation, de la connaissance à la compétence L’apprentissage aborde la question essentielle de l’acquisition des connaissances et compétences (Kim, 1993 ; Mack, 1995 ; Schein, 1993). Il met en jeu le capital cognitif, son potentiel d’enrichissement et de renouvellement et la régénération des schémas de l’action. L’apprentissage concerne deux sujets principaux : l’individu et l’organisation. L’apprentissage individuel est le processus par lequel les individus acquièrent des connaissances nouvelles. Deux courants principaux, le béhaviorisme et le cognitivisme, ont étudié la manière dont les individus apprennent (Leroy, 1998 ; Fillol, 2004). Pour les béhavioristes, l’apprentissage se fonde sur le concept de stimulus-réponse et se produit lorsque les comportements effectifs sont ajustés. Pour les cognitivistes, l’apprentissage n’est pas exclusivement comportemental. Piaget (1959), auteur représentatif de cette école, met en exergue deux processus d’apprentissage. L’assimilation consiste à intégrer une information et enrichir les schémas de pensée. L’accomodation est un renouvellement des modèles mentaux qui intervient lorsque l’assimilation n’est plus possible. L’apprentissage, dans une perspective cognitiviste, conduit alors à un enrichissement des connaissances et/ou à une modification des systèmes de croyances. Les deux dimensions de l’apprentissage, comportementales et cognitives, se complètent toutefois plus qu’elles ne s’opposent. En effet, un apprentissage comportemental, fondé sur la répétition et les automatismes, risque de demeurer superficiel et limité s’il n’intègre pas une connaissance poussée de ce qui dicte ce même comportement. Il semble plus approprié d’utiliser les termes d’apprentissage à dominante cognitive ou comportementale 1 (Leroy et Ramanantsoa, 1997 ; Leroy, 1998). Concernant l’apprentissage organisationnel, Koenig le définit comme « un phénomène collectif d’acquisition et d’élaboration de compétences, qui plus ou moins profondément, plus ou moins durablement, modifie la gestion des situations et les situations elles-mêmes » (Koenig, 1994, p.78). On considère ici que l’organisation apprend. Cette conception est pour le moins problématique car les organisations n’apprennent pas par elle-même, seuls ses membres apprennent. « L’organisation est composé d’individus et donc l’apprentissage individuel est 1 Cette réflexion souligne le caractère plus ou moins actionnable de ce qui est appris. Leroy note ainsi : « l’apprentissage à dominante cognitive concerne en priorité les compétences axées sur la connaissance dont le contenu est articulé ou articulable. Il passe par une explicitation ou même une formalisation de la connaissance. L’apprentissage à dominante comportementale apparaît plutôt quand la connaissance n’est pas articulée ou articulable mais peut cependant être enseignée, de façon moins directe et moins explicite » (Leroy, 1998, p.18). 191 nécessaire à l’apprentissage organisationnel ; cependant, l’organisation est capable d’apprendre indépendamment de chaque individu mais non de l’ensemble des individus » (Fillol, 2004, p.37). L’apprentissage organisationnel ne saurait donc être considéré comme la somme des apprentissages individuels. S’il est admis que les organisations apprennent (Kim, 1993), il est essentiel d’expliciter comment s’opère le passage entre les deux sujets d’apprentissage car il ne suffit pas qu’un individu ait appris pour que la compétence de la firme s’en trouve améliorée. Pour Argyris et Schön (1978), de l’apprentissage individuel dépend celui de l’organisation. C’est ainsi que les auteurs se fondent sur les théories de l’action1 pour soutenir leur théorie de l’apprentissage. Selon nous, pour saisir la tension entre l’apprentissage individuel et l’apprentissage organisationnel, il convient de s’interroger sur son objet. En effet, l’individu peut apprendre de nouveaux savoirs (connaissances) et raisonnements (programme maître ou représentations) et de nouvelles routines2 comportementales (compétences). Ces résultats d’apprentissage alimentent le niveau supérieur : les connaissances et compétences des individus peuvent enrichir celles de l’organisation (profiter aux autres acteurs), qui à leur tour enrichissent celles des individus. Les routines sont, à cet égard, très révélatrices. Elles correspondent à des savoir-faire, précisément à un « modèle d’activité répétitif pour l’organisation toute entière » car « ce qui est central pour la performance organisationnelle dans la production est la coordination ; ce qui est central dans la coordination est que les individus connaissent leur travail, interprètent et répondent correctement aux messages qu’ils reçoivent » (Nelson et Winter, 1982, in Coriat et Weinstein, 1995, p.116). La firme se dote ainsi d’un répertoire ou un portefeuille de réponses, de solutions prêtes pour l’action qui acquiert un caractère « d’automaticité » pour les acteurs. « Bien formés et entraînés, ils puisent spontanément dans le répertoire de réponses dont ils disposent pour fournir la réponse correcte, sans être nécessairement capables d’expliquer ni leurs choix, ni en quels savoir-faire particuliers ces choix consistent » (Coriat et Weinstein, 1995, p.117). Ces propos soulignent bien le « chemin » de ce qui est appris entre ceux qui apprennent. Ce sont les relations entre les individus qui constituent ce chemin entre l’apprentissage individuel et collectif. Pour mieux comprendre ce « chemin », émanant de la complémentarité entre sujets (et objets) d’apprentissage, le groupe constitue une unité d’observation intéressante. Un espace restreint de l’organisation peut apprendre et le contenu d’apprentissage se diffuser. Emerge alors une « géographie » de l’apprentissage. En effet, apprendre localement ne signifie pas apprendre globalement. Pour que l’apprentissage soit véritablement organisationnel et confère à l’entreprise la maîtrise d’une capacité nouvelle, il doit se répandre dans le tissu organisationnel. Ce « voyage » n’est en aucun naturel. « Il ne suffit pas de mener des expériences locales jugées comme des réussites par ceux qui les ont approchées pour croire qu’elles vont faire école : ces jugements ne sont-ils pas partisans ? Les conditions de la réussite sont-elles transférables ? Les critères d’évaluation sont-ils bons ? Ce qui est valable à échelle réduite le reste t-il lorsqu’on le généralise ? » (Midler, 1993, p.192). Démultiplier les dynamiques locales implique pour les fonctionnels de devenir les promoteurs de leurs initiatives, mais cette multiplication doit être organisée par des dispositifs de globalisation des apprentissages. Se référant à la gestion de projet « concourante », Midler montre comment les apprentissages issus du projet Twingo se sont diffusés dans d’autres espaces de l’organisation. « L’apprentissage de la logique projet chez 1 « Relève d’une routine défensive toute action, toute politique ou toute pratique, qui évite aux membres d’une organisation d’éprouver un embarras ou de ressentir une menace et les empêche en même temps d’en découvrir les causes » (Argyris, 1995, p.70). Ces routines sont surprotectrices, elles contribuent au maintien du statu quo essentiellement parce que, face aux situations embarrassantes et menaçantes, les individus pratiquent l’esquive (et la dissimulation de l’esquive). Les acteurs deviennent ainsi d’habiles joueurs qui pratiquent deux théories d’action. Les théories professées sont celles « dont les gens rendent compte ou qu’ils décrivent ». Les théories d’usage sont celles que « les gens utilisent en fait pour élaborer et mettre en œuvre leur action » (Argyris 1995, p.83). 2 Les routines correspondent à « des modèles d’interaction qui constituent des solutions efficaces à des problèmes particuliers » (Dosi, Teece et Winter (1990) cité par Coriat et Weinstein, 1995). Elles sont des résolutions aux problèmes récurrents. 192 Renault a ainsi mis en jeu divers agents de capitalisation. Ainsi […], le responsable de la formation au management a été l’instigateur du Club de Montréal […]. Enfin, un groupe de responsables de hauts niveaux, le groupe Delta, a été constitué, précisément, pour catalyser, systématiser et démultiplier les avancées diverses réalisées chez Renault dans la conduite de projet » (Midler, 1993, p.192). In fine, les sujets d’apprentissage entretiennent des relations que l’objet d’apprentissage nous permet de mieux cerner. L’individu, le groupe et l’organisation sont susceptibles d’apprendre selon des modalités dont nous avons tracé les grandes lignes. La diffusion des savoirs s’effectue ainsi par échange et partage, c'est-à-dire par socialisation. La « spirale du savoir » chez Nonaka (1991) reposant sur la conversion de la connaissance1 en est une parfaite illustration. La question des modalités d’apprentissage est toutefois jusqu’ici resté en suspend. En effet, quels processus caractérisent l’apprentissage ? Y a-t-il des types ou niveaux d’apprentissage ? Autant de questions auxquelles nous tentons d’apporter quelques éléments de réponse. 1.2 Les modes d’apprentissage ou comment les firmes apprennent-elles ? A partir de Koenig, nous avons posé que l’apprentissage « modifie la gestion des situations et les situations elles-mêmes » (Koenig, 1994, p.78). Cependant comment se produit le renouveau dans la gestion des situations et les situations ? Deux phénomènes expliquent la régénération des pratiques et de leur contexte d’émergence : l’accumulation d’expérience et l’intelligence d’expérimentation. Par l’expérience, la répétition, ou « l’exploitation des régularités » (Koenig, 1994, p.77), les individus et les organisations apprennent en améliorant constamment les pratiques. L’expérience correspond à un mode d’apprentissage lié à la révision et l’interrogation des pratiques courantes. Lorsque les résultats s’éloignent des objectifs, la détection des écarts conduit à adopter des aménagements, à réviser la pratique afin de la perfectionner. Ce processus correctif, fait d’ajustements réguliers, amène à les retoucher en fonction de la capacité à réduire l’écart entre le résultat et l’objectif. L’organisation apprend dans une succession d’erreurs et d’essais, par réplication. Cet apprentissage correspond à un apprentissage en simple boucle, ou opérationnel, dans lequel sont modifiées les stratégies d’action (Argyris, 1995). Il répond à une logique d’expérience car les fondements des stratégies d’action ne sont pas remis en cause ; les valeurs qui permettent de construire les théories d’action demeurent inchangées. En apprenant à modifier ses stratégies d’action, l’organisation consolide les processus d’exploitation de ses activités. Par l’expérimentation, les individus et les organisations modifient leur mode d’interprétation du monde, les schémas dominants, le cadre cognitif au sein duquel s’élaborent les réponses. Il s’agit d’une « philosophie pratique où les connaissances sont celles produites dans et par l’acquisition d’un savoir-faire et la réflexion dans l’action » (Koenig, 1994, p.77). L’expérimentation traduit un apprentissage dans lequel les valeurs directrices qui conduisent à l’adoption des pratiques organisationnelles sont interrogées et mènent au renouvellement du paradigme managérial. En s’en éloignant, cette forme d’apprentissage permet la mise en place de pratiques innovantes car il s’agit d’une réinterprétation et d’une reconstruction des modalités de l’action. Comme le précise Midler, « d’un coté, l’action est structurée par les savoirs existants ; mais d’un autre, elle constitue une mise à l’épreuve de l’applicabilité et de la légitimité de ce dispositif cognitif » (Midler, 1994, p.347). Cet apprentissage est en double boucle car il modifie les pratiques organisationnelles grâce à l’interrogation des valeurs directrices qui les déterminent (Argyris, 1995). Ainsi, les organisations qui expérimentent, en adoptant des pratiques qui se situent à l’extérieur du paradigme managérial, enrichissent leur portefeuille de réponses aux sollicitations de l’environnement. Thiétart souligne « le besoin 1 Cette conversion s’effectue par le passage du tacite à l’explicite par l’étape de l’extériorisation et par celui de l’explicite en tacite par l’intériorisation. 193 qu’ont les organisations de développer de nouvelles réponses afin de pouvoir réagir à des situations que l’on ne peut prévoir. De même, Weick (1977) et March (1981) suggèrent que les activités qui ne sont pas directement liées à la mission de l’organisation sont des moyens pour améliorer sa capacité de réponse à la complexité et aux conditions changeantes » (Thiétart, 2001, p.367). Ces deux formes d’apprentissage peuvent sembler paradoxales 1 . L’expérience invite à exploiter les régularités, l’expérimentation à s’en éloigner. La gestion de l’apprentissage consiste alors à concilier les exigences contradictoires des logiques d’expérience et d’expérimentation. Il est donc question de gérer deux forces en tension ; l’ordre et la stabilité qui se situent dans le registre de l’opération et le désordre et l’instabilité qui se situent dans celui du projet (Mbengue, 1997, Thiétart, 2001). En effet, l’ordre et la stabilité reposent sur des dispositifs de planification, de contrôle et de structure. Ils permettent aux organisations d’assumer leur mission, aux acteurs de se positionner au sein d’une structure et de réduire une complexité problématique pour l’être humain aux capacités cognitives limitées, voire de créer une certitude imaginaire mais rassurante, permettant de mieux affronter les aléas du jeu des affaires. A l’opposé, l’instabilité et le désordre résultent des initiatives individuelles. Les espaces de liberté permettent l’innovation par l’expérimentation de l’inconnu et la nouveauté. L’organisation se construit un catalogue de réponses qui pourra être mobilisé lorsque des demandes, inconnues aujourd’hui, les solliciteront demain. Des modes de travail expérimentés dans le passé peuvent devenir dominants dans le futur. L’antagonisme de l’apprentissage par l’expérience et l’expérimentation se retrouve, par exemple, dans le paradoxe de la mémoire. « La mémoire peut être la meilleure et la pire des choses. D’un côté, elle permet de gagner en efficience par l’exploitation des régularités et des répétitions de l’histoire. D’un autre côté, elle peut être un obstacle à la découverte de modalités d’action plus efficaces, comme elle est susceptible d’empêcher une appréciation judicieuse de ce qui est véritablement nouveau » (Koenig, 1994, p.81). La mémoire résultant de l’expérience accumulée par l’organisation permet d’éviter de répéter les erreurs du passé, réduit les coûts de réflexion et de mise en œuvre de solutions et constitue un puissant guide pour l’action. De tels atouts peuvent conduire à l’inertie, au maintien de la stabilité par aveuglement et à l’incapacité à sortir du paradigme (Laroche, 1997). La gestion des tensions entre stabilité et instabilité, ordre et désordre, expérience et expérimentation, exploitation et exploration, se trouve au cœur des modalités de l’apprentissage. Il s’agit alors de maintenir le dialogue entre ces attracteurs et d’assurer entre eux une dialectique permanente. In fine, il est question de gérer les tensions inhérentes à l’exploitation et l’exploration (March, 1991). Si expérience et expérimentation doivent se concilier pour produire des apprentissages, il convient de s’interroger sur leur déclenchement. Comment l’exploitation et l’exploration naissent-elles ? Quels sont les déterminants d’un processus d’apprentissage ? 1.3 Les déterminants de l’apprentissage : occasions et déclenchement du processus Leroy et Ramanantsoa (1997) proposent d’identifier les déclencheurs d’apprentissage à partir de leur source. En effet, la firme peut apprendre dans le cadre des relations avec l’environnement, de soi-même et de ses partenaires. Ces sources constituent des occasions d’apprentissage auxquelles correspondent des déclencheurs, des stimuli. Ainsi, la firme peut apprendre de son 1 Pour autant, ces deux formes d’apprentissage ne sont pas mutuellement exclusives. « Faire ce que l’on a toujours fait est nécessaire dans l’adaptation à court terme. Faire ce que l’on n’a jamais fait est nécessaire dans l’adaptation à long terme et les deux sont nécessaires simultanément » (Weick, 1977 cité par Thiétart et Forgues, 1993, p.14). 194 milieu dès lors que des changements environnementaux se produisent et exigent une adaptation. De même, la firme peut apprendre d’elle-même lorsqu’elle connaît des échecs sur certains projets. En pratiquant le retour d’expérience, elle ajuste ses méthodes et déclenche un apprentissage. Enfin, les relations avec les partenaires sont sources d’apprentissage. Lorsque des différences organisationnelles sont repérées, l’un des partenaires peut chercher à les comprendre et à modifier sa propre organisation pour bénéficier des avantages identifiés. Le tableau suivant propose une synthèse des sources et déclencheurs principaux de l’apprentissage. Tableau 1 : Les sources et déclencheurs d’apprentissage Apprendre à partir de ses Apprendre de soi-même Apprendre dans le relations avec le milieu cadre des partenariats Différence Changements dans Echecs Déclencheurs organisationnelle Dysfonctionnements l’environnement d’apprentissage Compétences Mauvaise performance Erreurs recherchées Crise expliquée par l’externe Créativité Source : Inspiré de Leroy et Ramanantsoa (1997). Source d’apprentissage Cependant s’interroger sur les déterminants de l’apprentissage ne saurait se limiter à en identifier les déclencheurs. Si la crise, l’erreur… constituent des stimuli d’apprentissage (et de désapprentissage), la saisie des occasions doit retenir notre attention. En effet, il n’est possible de comprendre ce qui déclenche le processus sans appréhender le contexte dans lequel il prend place. C’est plus l’interprétation du signal qui déclenche l’apprentissage que le signal lui-même. Extraire l’occasion d’apprentissage de son contexte consisterait à éliminer le rôle de l’activité de fabrication de sens. Or l’environnement de l’action n’est pas parfaitement objectif et extérieur (Daft et Weick, 1984), « il est dans l’organisation ». C’est donc plus l’activité d’interprétation du réel qui produit la saisie de l’occasion d’apprentissage que le réel lui même. De même, cette interprétation doit être suffisamment collective pour pouvoir enclencher une réponse organisationnelle. L’absence de consensus dans l’interprétation limite la production d’un comportement correctif ou de recherche de nouveaux savoirs. Mais alors comment permettre que les membres de l’organisation saisissent ces occasions et déclenchent des apprentissages ? Selon Pascale (1990), c’est la capacité d’interprétation et d’initiatives qui est au cœur du déclenchement de l’apprentissage. Or ces deux aptitudes relève d’un même phénomène : le maintien de la diversité. Celle-ci se caractérise par « l’incitation faîte aux employés de remettre en cause la routine et la statu quo ; un degré d’autonomie et de délégation d’autorité suffisant à tous les niveaux de l’organisation ; l’absence d’une domination trop marquée d’un cadre de référence au travers duquel se règlent les décisions ; la capacité à intégrer et prendre en compte une grande diversité d’informations externes… » (Pascale, 1990, p.236). De même, selon Thiétart (2000, p.7-8), « la décentralisation, l’autonomie, la liberté d’initiative et l’acceptation de déviances doivent être recherchées et favorisées ». La diversité interne est donc le résultat de l’autonomie, elle-même favorisée par l’espace de liberté dont disposent les acteurs, leur donnant à la fois le droit et le devoir de résoudre les problèmes auxquels ils sont confrontés dans l’action. Selon Alter (1999), le management doit faire « trembler » et réagir l’organisation pour entretenir la diversité. « En tentant de programmer et de contrôler la « culture de l’entreprise », le management verrouille les négociations de fond. Il interdit, par la neutralisation du conflit, l’apparition et le développement de logiques nouvelles. A l’inverse de cette perspective, l’entreprise pourrait positivement s’ouvrir à tous ceux qui, par « manque de conformité » sont mis à l’écart du 195 système social et même de l’organisation » (Alter, 1999, p.170). Le fameux « work-out » de General Electric illustre une telle volonté. Il réunit le personnel et les dirigeants à tous les échelons pour critiquer, débattre, échanger et suggérer. Le débat ouvert permet de discuter des normes établies, de les éprouver et de les modifier (Pascale, 1990). « L’exclusion des conflits n’est en effet pas plus salutaire que la volonté de fusion culturelle : le conflit représente un moyen de mise en œuvre de l’innovation » (Alter, 1999, p.184). Le conflit est alors vertueux car il amène les individus à agir conformément aux théories professées. En corollaire, l’organisation apprend lorsque ses membres prennent conscience de la part de responsabilités qui leur incombe et modifient alors leur façon de raisonner. Comme le note Hodgson (1988, p.128), « l’utilisation habituelle de concepts particuliers et de modes de pensée peut masquer les hypothèses cachées et les axiomes qui ont été employés. Nous pouvons apparaître « persuadés » par l’évidence ou par un argument sans se demander si leur utilisation est légitime ». Dans cette optique, le rôle du dirigeant est de « fabriquer » des valeurs fortes axées autour du débat et de la confrontation. Le manager incite ses salariés à ne pas sombrer dans la répétition et la routine paralysante. Il garantit les expérimentations, les valorise et les symbolise. Il ne doit pas être le gardien des croyances organisationnelles, lui plus que tout autre est celui qui s’éloigne des pratiques conservatistes sans les ignorer. Il est un manager « passeur » ; il sert de « guide dans l’évolution des représentations et des identités » (Laroche, 1997, p.188). Il connaît les représentations dominantes, déploie des dispositifs d’apprentissage aptes à les renouveler de manière progressive et permet l’expression des diversités d’interprétation des problèmes et de variabilité des solutions proposées. Un tel management induit chez les individus l’acquisition d’une capacité d’interrogation des prémisses qui servent de base à leurs décisions : ils questionnent la logique sous-jacente à leurs actes. Les déterminants de l’apprentissage dépendent donc plus de la capacité d’interprétation des occasions que des occasions elles-mêmes. En effet, les déclencheurs offrent une explication trop réductrice des déterminants de l’apprentissage. Leur saisie s’avère un objet d’étude plus éclairant mais aussi plus complexe à appréhender. Elle trouve son origine dans le contexte de l’organisation et en particulier dans le maintien et la recherche de la diversité interne. Jusque là, le concept d’apprentissage apparaît de manière fragmentée formant une mosaïque théorique. Ses contours théoriques sont épars et sa dimension opérationnelle et instrumentale demeure limitée. Cependant, le « concept » d’organisation apprenante semble prometteur. Il laisse espérer l’agrégation des connaissances relatives au phénomène d’apprentissage et entrevoir un objet théorique aux contours délimités. 1.4 L’organisation « apprenante » : un concept intégrateur ? Le concept d’organisation apprenante laisse penser qu’il existe une approche théorique stable du phénomène. Qu’en est-il exactement ? L’organisation « apprenante » apparaît comme le concept fédérateur des différents angles d’attaque (sujet, objet, processus et déterminants) de l’apprentissage pour former un objet théorique stable. Il permet d’approcher le phénomène dans sa complexité. Pour Senge, une organisation apprenante est un lieu « où les personnes augmentent continuellement leurs capacités de créer les résultats qu’ils désirent vraiment, où de nouveaux modèles de pensée sont développées, où les aspirations collectives sont encouragées et où les individus apprennent continuellement comment apprendre ensemble » (Senge, 1990, p.1). Les principes fondateurs d’une organisation apprenante sont alors les suivants : • les salariés sont engagés dans une dynamique permanente d’auto-apprentissage, 196 • l’organisation est capable de générer des nouveaux comportements en mettant à l’épreuve ses modèles mentaux et cadres d’interprétation, • il existe un ensemble de normes communes autour d’une vision, • la démarche d’analyse des problèmes est systémique. Cette conceptualisation est pour le moins fragile (Baumard, 1995 ; Leroy, 1998). On attendrait une agrégation harmonieuse des connaissances relatives à l’apprentissage, on se retrouve avec une série de principes dont on saisit mal à la fois la nature et les connexions. D’une part, les principes restent assez vagues et relève même d’une forme d’idéalisme managérial (un one best way organisationnel). Leur contenu opérationnel reste à préciser, son pouvoir d’instrumentation est faible. S’agissant du second principe, il est légitime de s’interroger sur la manière dont la firme parvient à la mise à l’épreuve des modèles mentaux. D’autre part, les liens entre les principes sont inexistants, or ils permettent d’agréger les connaissances pour former un objet théorique stable. La mesure du phénomène demeure donc problématique car ses déterminants et ses composants sont nombreux, imbriqués et entretiennent des relations complexes dans le mécanisme de structuration (l’objet pouvant être structurant et structuré). A titre d’exemple, la tension entre les deux sujets d’apprentissage est esquivée, de même que l’articulation entre l’exploitation et l’exploration. L’approche de l’entreprise apprenante montre des insuffisances manifestes et suscite des espoirs vite déçus. En tout cas, elle témoigne de l’existence d’un champ de recherche à explorer. Il semble donc difficile de parler d’un concept fédérateur, tout au plus s’agit-il d’un concept en devenir, d’une esquisse théorique. Incontestablement, le pouvoir instrumental de l’organisation « apprenante » est limité. Dans cette première partie, nous avons défini les contours de notre objet théorique. L’apprentissage revêt quatre dimensions dominantes : le sujet (qui apprend ?), l’objet (qu’apprennent les sujets ?), les modalités et processus (comment les sujets apprennent-ils ?) et les déterminants (quand les sujets apprennent-ils ?). Le caractère multiforme et multidimensionnel de ce concept le rend particulièrement problématique à saisir. Le recours à « l’organisation apprenante » s’est avéré infructueux pour l’appréhender de façon plus opérationnelle. Or, ce détour par l’instrumentalisation nous semble indispensable pour le rendre actionnable. Les premiers pas vers cette opérationnalisation procèdent, selon nous, de l’étude de situations d’apprentissage. Dans ce qui suit, après avoir souligné que les processus d’innovation constituent des moments d’apprentissage intéressants, nous dégageons les premiers éléments d’opérationnalisation du concept. II L’innovation comme situation d’apprentissage collectif S’il existe un relatif consensus autour de l’innovation, c’est bien dans la reconnaissance de l’incertitude inhérente au concept lui-même. Dans la perspective ouverte par Schumpeter (1942), la capacité à élaborer de « nouvelles combinaisons » entre les ressources de l’entreprise, leur rapport au marché et l’usage social qui en établit l’adoption, doit être conçue comme à la fois destructrice et créatrice, l’un et l’autre mouvement étant générateurs d’incertitudes. Deux dimensions caractérisent alors la situation de l’innovateur : l’incertitude par rapport au résultat final et l’incertitude quant aux moyens à mettre en œuvre pour atteindre cet objectif (Pearson, 1991). Comme l’a illustré Callon (1988, p.5): « pour avoir une idée de l’extrême complexité du processus d’innovation, il faudrait imaginer une fusée pointée vers une planète à la trajectoire inconnue, et décollant d’une plateforme mobile, aux coordonnées mal calculées ». L’incertitude des organisations face à l’innovation est donc élevée et se matérialise par un déficit d’information 197 de deux natures. Il y a déficit d’information sur ce que devraient être des choix se rapportant à une solution qui n’existe pas encore. Il y a également un déficit d’information sur les règles de fonctionnement que l’organisation devrait adopter pour produire cette solution. L’existence d’un niveau élevé d’incertitude consubstantielle à l’innovation pose la question de la rationalité du processus. Choisir, parmi les multiples options de développement, celles qui conduiront à l’optimum se révèle, à l’examen des situations managériales, un idéal impossible à atteindre. Comme le souligne Alter (2003, p.26) : « l’acteur ne peut jamais utiliser toutes les informations disponibles : ou parce qu’il ne peut les traiter pour cause d’incompétence, ou parce qu’elles sont trop coûteuses, ou parce qu’il est difficile d’en connaître l’usage, ou parce que d’autres font de la rétention d’information. La rationalité est donc objectivement limitée parce que l’information est rarement parfaite, même par rapport à un état donné du savoir ». Dans cette perspective formulée par Simon (1955, 1991), où le choix ne peut être défini a priori, l’innovateur aura nécessairement recours à des heuristiques dites de « rationalité limitée » ; l’acteur recherche non pas une solution parfaite mais une solution satisfaisante, ou de « rationalité procédurale » ; l’acteur construit sa décision en découvrant les difficultés et les ressources qui se présentent à lui. Le processus d’innovation devient un processus de réduction d’incertitude. Au fur et à mesure de l’avancement, l’acteur découvre tout à la fois les problèmes, les moyens de réponses et les solutions elles-mêmes. Les opérations d’interprétation des situations, les « épreuves » ou tests des représentations formées, la formalisation des options et les choix, les essais-erreurs sont tout à la fois des étapes d’apprentissage et des étapes potentielles du projet. Faire avancer une innovation, c’est faire progresser les connaissances relatives au projet de telle sorte que les impasses aient pu être évitées. Ce processus est en outre collectif. Comme le soulignent Lenfle et Middler (2003), un succès est le résultat de la combinaison de nombreux paramètres : « il s’agit toujours d’un compromis entre les logiques différentes des marchés, des études, de la recherche et de la production » (Lenfle et Middler, 2003). A l’incertitude quant aux objectifs et aux moyens se rajoute le manque de lisibilité des objectifs de chacun des acteurs associés au processus, la difficulté d’imaginer a priori leur compétences et contributions potentielles, et enfin l’interprétation qu’ils feront des projets qui vont leur être soumis. L’apprentissage en situation d’innovation est à poser dans un ensemble intra et inter organisationnel. Progresser dans le projet d’innovation, c’est donc réduire le niveau d’incertitude relatif aux processus de production, aux réactions du marché, aux comportements des acteurs du processus, en dégageant, par apprentissages successifs, les voies qui permettront la réalisation de solutions acceptables par tous. Tout au long des deux dernières décennies, un flux continu de travaux empiriques a progressivement confirmé la pertinence de ce cadre d’analyse. Etudiant les processus d’innovation, Maidique et Zirger (1985) constatent que l’échec d’un projet d’innovation, parce qu’il entraîne une dynamique d’apprentissage renouvelée, conduisait au succès du projet suivant. Situant l’apprentissage au cœur du projet innovant, Madhavan et Groover (1998) observent que : « chaque individu apporte dans une situation ses compétences, connaissances et actions qui la transforment et sont transformées en retour. L’interaction des individus dans les situations de gestion se traduit par la performance cognitive du groupe ». Les travaux de Nonaka et Takeuchi (1995) démontrent comment les processus de création de nouveaux produits sont indissociables de la création de connaissances. L’expérimentation, la formalisation, le partage et l’internalisation des connaissances sont les processus qui vont aboutir directement ou indirectement à la transformation de combinaisons productives et à leur mise en marché (Nonaka, 1994). Les travaux de Callon et al. démontrent bien l’importance de « l’enrôlement », c'est-à-dire de l’association et de la prise d’un rôle de chacun des acteurs de l’innovation au sein du projet auquel ils doivent contribuer (Callon, Latour et al., 1998). Enfin, Lenfle et Middler (2003) soulignent que « l’épreuve », c'est-à-dire la vérification de la pertinence des représentations, des actions et des solutions, permet tout au long du processus à la fois de valider les choix, de créer des connaissances et de coordonner l’action des acteurs. 198 L’ensemble de ces travaux conduit à penser que les phénomènes d’apprentissage et de partage de connaissances se trouvent au cœur des processus de création de l’offre. L’identification des variables d’opérationnalisation de l’apprentissage organisationnel peut ainsi passer par l’observation empirique des processus de création de l’offre. Nous nous proposons donc de travailler dans cet objectif sur le cas particulier des innovations de service. 2.1 L’intérêt des travaux sur l’innovation de service A la suite de Gallouj (2003), on peut affirmer que la faible reconnaissance de l’innovation au sein des services provient plus d’une méconnaissance des spécificités de ces activités que d’un réel déficit d’innovation. Les entreprises de services innovent et contribuent, comme dans les secteurs industriels, à la création d’avantages concurrentiels et de valeur ajoutée ainsi qu’au développement de la compétitivité de la firme. Quand bien même la protection de l’innovation de service est faible, l’étude longitudinale d’un secteur démontre les bénéfices tangibles pour les entreprises innovantes (Roberts et Amit, 2003). La faible visibilité des travaux sur l’innovation de service, qui trouve sans doute ses sources dans une tradition de recherche centrée sur les entreprises industrielles, provient également du nécessaire renouvellement des cadres d’analyse de l’innovation, rendu inévitable par trois caractéristiques très différenciantes de ces activités. Tout d’abord, les activités de service se définissent à travers le caractère intangible de l’output, poussant les auteurs à la métaphore autant qu’à la définition. Berry (1980) les décrit comme « des actes, des processus ou des représentations1 ». Le caractère intangible de l’activité de service pose problème. Ainsi, Quinn, Baruch et al. (1987) proposent-ils de définir les services comme : « toutes les activités économiques dont le résultat final n’est ni un produit ni une construction, qui est généralement simultanément produit et consommé, et qui produit une valeur ajoutée à son acheteur sous une forme intangible (comme la praticité, le confort, le gain de temps, la santé) ». Le caractère intangible des services pose le délicat problème de l’identification de l’innovation tant dans sa description, comme son degré de nouveauté par exemple, que dans l’évaluation de ses effets économiques. Au-delà des questions de l’identification elle-même, l’intangibilité induit l’impossibilité de dissocier l’acte de production de l’acte de consommation. Les conséquences de ce simple constat, rapportées à l’innovation, sont de trois ordres. Callon et al. (1995) le soulignent : « cette affirmation conduit à identifier trois différences majeures entre innovation de produit et de service: la première est qu’il y a en même temps innovation de produit et de processus, la seconde est qu’on ne peut opérer de séparation entre innovation de produit et innovation organisationnelle, et la troisième est qu’il ne peut y avoir de distinction entre la création de l’offre et l’activité de production et de commercialisation ». Innover dans le domaine des services consiste, de par la nature de l’output, à modifier les processus d’interaction par lequel est obtenu la satisfaction du client, et de le faire en modifiant l’organisation productrice de ce processus. L’objet de l’innovation n’est donc pas d’intégrer une technologie, ou un savoir scientifique, dans une organisation existante mais de considérer l’organisation comme la technologie elle-même, et sa modification comme l’élaboration d’une nouvelle technologie qui permettra de renouveler la valeur créée. A ce constat, qui évoque la nécessaire rupture avec les modèles d’innovation technologiques, il faut ajouter la position bien particulière du client au sein même du processus de production de la firme. Lors du délivrement d’un service, le client participe à l’achèvement de certaines séquences d’action au point que celles-ci ne sauraient se dérouler sans sa présence (Zeithmal et Bitner, 2003). Poussant l’analyse au-delà de la simple présentation des flux d’interaction, Gallouj et Weinstein 1 Traduction de « deeds, process and performances ». Le terme représentation renvoie au vocabulaire du théâtre ou du film. On pourrait également traduire dans ce contexte par le mot scénario. 199 (1997) montrent que dans la production d’opération de conseil, c’est l’articulation des compétences des clients et de celles apportées par les consultants qui génère l’utilité produite au client. La prestation de service est alors définie comme : « la mobilisation simultanée de caractéristiques techniques (matérielles et immatérielles) et de compétences (internes et externes) pour produire les caractéristiques de service » (Gallouj et Weinstein, 1997, p.118). On le voit, les caractéristiques particulières de l’offre de service mettent les questions d’apprentissage organisationnel au cœur du processus de création de nouvelles offres. Nous nous proposons donc de restituer les travaux portant sur l’innovation de service et d’utiliser ceux-ci dans un deuxième temps comme révélateur des dimensions mesurables de l’apprentissage organisationnel. 2.2 Développement d’offre de service et apprentissage organisationnel Dans une logique managériale, le principal objectif des premiers travaux de recherche vise à l’identification des facteurs pouvant induire le succès ou l’échec dans le développement de nouveaux services. Le travail de Cooper et al. (1994) est représentatif d’un ensemble de travaux effectués pour la plupart dans le domaine des services financiers (Brentani, 1989, 1993; Easingwood et Storey, 1995; Edgett et Parkinson, 1994). Les méthodologies quantitatives visent à identifier, par comparaison de succès et d’échec, les facteurs positivement ou négativement corrélés au résultat, puis à comparer les résultats obtenus pour les services avec ceux obtenus pour les produits. Au-delà des limites méthodologiques de l’approche, les résultats communs à l’ensemble des travaux révèlent que les synergies (synergies marketing, managériales et financières) établies entre les offres existantes et les innovations, ainsi que l’intégration inter fonctionnelle (communication interne, gestion du projet) se révèlent être fortement corrélées au succès. Dans une perspective « Resource Based View », non développée par les auteurs, ces résultats peuvent s’expliquer aisément. Plus les nouveaux services font appel aux répertoires de connaissances individuels et organisationnels existants, plus l’incertitude et les risques sont réduits. Les facteurs qui apparaissent corrélés au succès soutiennent une explication du succès par les compétences de l’organisation et du client. A ces premiers résultats, qui ne font pas référence explicite à un cadre d’apprentissage organisationnel, viennent s’articuler des travaux empiriques qui cherchent à donner un contenu au processus de développement lui-même et des travaux plus quantitatifs portant sur les processus de communication interne. Hart et Service (1993) ont cherché à analyser le processus de développement lui-même. En conduisant une recherche action au sein d’une entreprise de distribution de composants électromécaniques et électroniques, ils mettent en évidence les composantes organisationnelles de l’innovation de service. D’une part, les divergences de vision entre les différents départements de l’entreprise peuvent avoir un effet considérable sur les durées de développement. D’autre part, l’innovation induit un changement dans les processus de délivrement, cette transformation entraînant à son tour un changement organisationnel. Les auteurs suggèrent alors qu’une approche des processus par « intégration fonctionnelle » forte, reposant sur une communication inter fonctionnelle riche et informelle, sur le partage d’information, et sur du partage de décision doit conduire à plus d’efficience. Dans une perspective similaire, les travaux de Raesfeld et al. (1996), s’appuyant sur une étude de cas longitudinale, analysent les flux de communication en démarrage et en fin du processus d’innovation sur deux dimensions : la fréquence de communication entre les membres de l’organisation et le degré de similarité des affirmations relatives au projet. Les résultats montrent un changement profond dans les flux de communication : la fréquence de communication et la similarité des perceptions sont profondément altérées au point de voir apparaître, en fin du processus, des personnes littéralement exclues des réseaux établis, celles-ci jouant un rôle déterminant dans la production des nouvelles offres. Les auteurs concluent en justifiant de l’échec du projet par le déficit de « sens partagé » résultant lui-même de l’absence de communication. 200 Faisant suite à l’ensemble de ces travaux, Lievens et al. (Lievens et Moenaert, 2000; Lievens, Blazevic et al., 2003) adoptent une perspective similaire en lui ajoutant une méthodologie de mesure quantitative. La communication inter et intra organisation est conceptualisée comme permettant de réduire l’incertitude relative au client, aux concurrents et à la technologie. « La qualité de l’ajustement entre la communication liée au projet et le champ d’incertitude rencontré induit l’efficacité des flux de communication sur la réduction de cette incertitude » (Lievens et Moenaert, 2000). En communiquant tout au long du projet, les équipes transfèrent les constats, idées et apprentissages en tous points de l’organisation, permettant ainsi une meilleure coordination intra/inter fonctionnelle. En ce sens, il est possible de démontrer, comme le font les auteurs, que la dynamique de communication est directement facteur de succès ou d’échec dans la création et mise en marché de nouvelles offres. Ces premiers travaux ne suffisent cependant pas à fournir un cadre théorique intégrateur. En effet, les flux de communication ne permettent pas en soi de réduire l’incertitude. Pour cela il faut que, en amont et en aval du transfert d’information, il y ait eu apprentissage, c'est-à-dire production d’une première interprétation, de la « mise à l’épreuve » de celle-ci, et enfin de la formalisation qui permettra le partage. Ce constat conduit à considérer le processus d’innovation sous l’angle de l’apprentissage organisationnel. Dans cette lignée, les travaux de Stevens (2002), Stevens et Dimitriadis (2002, 2004) centrés sur l’observation empirique et longitudinale de processus d’innovation, mettent en évidence tout à la fois le rôle de la communication, mais également des opérations de transformation des représentations et de l’organisation impliquée dans le projet. Les observations peuvent être regroupées en quatre principaux constats. Tout d’abord, l’initiation du processus repose sur la perception de menaces ; remise en cause des recettes existantes créant de fait une tension à l’action justifiant l’ouverture d’un nouveau projet. Le management utilise alors sciemment un vocabulaire flou, permettant à chacun de construire une représentation individuelle de la solution au problème évoqué. L’absence de directives précises, la tension générée, la réappropriation par les individus de la problématique dans leur propre perspective vont permettre de déclencher le démarrage plus ou moins formel du projet, lancer la production d’interprétation des situations et des solutions à apporter et un premier flux d’échanges informels autour de « l’objet » en cours de construction. Puis, en parallèle, tant par des processus formels qu’informels, les acteurs vont mettre les premières représentations à l’épreuve suivant des modalités très diverses. De la simple conversation à l’essai grandeur nature, chacune des parties impliquées par l’énoncé du problème va chercher à partager son interprétation de la façon la plus large possible. En ce sens, l’offre finale est façonnée par ses multiples allers et retours entre imagination et réalité, entre premières formalisations et prototypes. Il y a, bien sûr, mise à l’épreuve par le client, mais également par les autres services impliqués dans la nouvelle offre. A ce stade, la communication a pour fonction de permettre l’intégration des apprentissages multiples, localisés en de nombreux lieux, en un tout cohérent. Loin d’être un processus linéaire, ce flux de confrontations à la réalité, permet de faire les apprentissages dont émerge progressivement, par touches successives, une image de plus en plus nette de ce que pourrait être l’offre. Ce processus s’accompagne d’une transformation organisationnelle. En effet, choisir de transformer une séquence de production pour pouvoir produire un output différent suppose d’adapter le travail d’un département, d’en modifier les contours, d’en adapter les compétences et le rôle de chacun. En retour, les résultats des apprentissages sont formalisés, « figés » dans les bibles de procédures, les logiciels informatiques, l’agencement des surfaces de vente et progressivement routinisés. Sans plasticité organisationnelle, la production de l’innovation est rendue impossible. Sans modification de l’organisation, de ses structures et de ses routines, l’apprentissage ne sera pas mémorisé sur le moyen terme. Enfin, il faut noter que la reproduction de l’innovation de service sur des sites multiples conduit à la multiplication des apprentissages. Quand bien même le projet est dessiné au niveau 201 national pour l’ensemble des magasins, son implantation sur chaque lieu induit un nouvel apprentissage : celui des hommes qui s’approprient un objet complexe, celui de la mise en place dans un espace de contraintes différentes (géographie des magasins, systèmes d’information non homogènes, répertoires de compétences différentes, interaction avec l’ensemble des autres projets existants), celui des clients qui doivent redécouvrir les nouveaux contours du service proposé. In fine, les processus d’innovation et particulièrement d’innovation de services sont, comme nous venons de le démontrer, des situations d’apprentissage. Pour autant, dans quelle mesure est-il possible d’opérationnaliser l’apprentissage organisationnel ? L’ensemble de ces travaux montre à l’évidence qu’une mesure de l’apprentissage par l’output s’avérerait vaine. Définir a posteriori s’il y a eu ou non apprentissage est difficile voire impossible sans connaître le processus, ou plutôt les étapes du processus qui ont été mobilisées pour aboutir au résultat. En observant les actions qui ont été menées sur un processus et en observant l’atteinte ou non des objectifs qui ont émergé du processus, on peut conclure au succès de l’apprentissage. L’opérationnalisation du concept, si complexe soit il, est alors envisageable. III Les premiers pas vers l’opérationnalisation du concept d’apprentissage L’ensemble de ces observations permet de constater que les apprentissages, loin d’être une fin, sont conçus comme un moyen de réduire l’incertitude. On peut regrouper les principales axes d’opérationnalisation autour de quatre dimensions principales : l’implication dans les projets d’innovation, l’adoption d’une gestion systémique des projets, le degré d’ouverture et d’expérimentation et le partage et l’intégration des connaissances acquises. 3.1 L’implication dans les projets Cette dimension est sans doute délicate à mesurer. Elle se situe dans les phases amont des projets. Elle consiste clairement en l’identification d’un problème ou d’une difficulté provenant soit de l’environnement ; les compétiteurs innovent ou gagnent des parts de marché, soit de l’interne ; les secteurs/produits ne produisent pas assez de résultats par rapport au reste du magasin. A ce stade, il n’y a pas encore de projet et tous les constats de problème ne déboucheront pas nécessairement sur un projet. Cependant, dans les cas étudiés (Stevens, 2002), lorsqu’un projet émerge en tant que tel, il repose sur le premier partage des constats qui justifient un investissement. Sans préjuger de sa légitimité, la formulation et le partage de la problématique de départ constituent un premier critère qui permettra de juger de la dynamique d’apprentissage collectif. Tout aussi délicate est l’ouverture d’un processus d’interprétation. Il faut entendre par là la possibilité donnée aux acteurs du projet de sortir des routines organisationnelles pour initier l’offre à venir. Il ne suffit pas en effet d’annoncer l’ouverture d’un projet pour qu’un processus d’apprentissage soit déclenché. Il s’agit tout en initiant le projet de laisser chacun des acteurs libre de produire une/des solutions, basées sur une interprétation personnelle du problème. Cette production repose sur la formulation d’une règle explicite ou implicite : les acteurs du projet doivent produire des solutions, mais la nature des solutions à produire ne peut être formulée au stade du démarrage du projet. L’existence de règles tacites ou explicites, reconnues par les acteurs en situation, et permettant ou facilitant l’expérimentation est à ce titre une des dimensions permettant d’opérationnaliser le concept. La formulation et le partage d’un problème ainsi que l’ouverture formelle d’un projet, par lequel l’organisation peut répondre, conduit à la création de réseaux d’acteurs, formels et informels qui, dans une logique d’apprentissage, vont permettre de gérer l’ensemble des réponses organisationnelles à apporter au projet. 3.2 L’adoption d’une gestion systémique du projet 202 La gestion du projet doit permettre de couvrir l’ensemble des dimensions organisationnelles du problème abordé. L’analyse des raisons d’échec met en relief une pratique ou une vision compartimentée du projet. L’action localisée, non coordonnée au reste de la structure, génère de l’échec par déficit d’intégration. Un projet qui n’intégrerait pas toutes les dimensions de l’organisation va générer un apprentissage localisé, valable pour quelques individus, mais qui se révélera insuffisant pour porter la pratique de l’organisation dans son ensemble. Il y a donc dans l’ambition du projet, le champ des questions traitées, la position des acteurs impliqués le moyen de matérialiser la dimension systémique de celui-ci. Une vision d’ensemble du projet, partagée par les acteurs impliqués dans celui-ci doit conduire à un apprentissage intégrateur de l’ensemble des contraintes et non à une vision partielle et réduite des problèmes posés. La conduite systémique du projet va soutenir un ensemble d’expérimentations et l’élaboration de solutions tirées de l’expérience. 3.3 Le degré d’ouverture et d’expérimentations Le degré d’ouverture et d’expérimentation peut se mesurer sur trois dimensions principales : le climat d’ouverture, le nombre et la nature des expérimentations, la diversité des compétences associées au projet. Le climat d’ouverture est nécessaire en ce que tout apprentissage remet en cause les schémas et modes de fonctionnement existants. Il se traduit par le fait que les acteurs acceptent de remettre en cause les modes de fonctionnement existants, prennent en compte les expériences, remarques et avis de l’ensemble des acteurs, acceptent le décentrement des points de vue. Sans ce qu’il faut appeler un état d’esprit réaffirmé implicitement ou explicitement dans la gestion du projet, il n’est pas d’expérimentation concevable. Celle-ci requiert, en effet, d’admettre l’erreur comme source d’apprentissage et, par conséquent, de tolérer le risque. Enfin, l’expérimentation signifie la mise en place d’actions concrètes et mises à l’épreuve des solutions imaginées. Le nombre et la nature des actions d’apprentissage constituent, sans doute, la partie la plus visible du processus. Il s’agit sur cette dimension d’observer les moyens, formels et informels, par lesquels les acteurs mettent à l’épreuve les affirmations, hypothèses et solutions élaborées tout au long du processus. La fréquence des observations et la diversité des méthodologies doivent logiquement être corrélées au niveau d’apprentissage. En questionnant les acteurs du projet sur les modalités d’acquisition d’informations et sur les modalités de mise à l’épreuve des hypothèses et des solutions imaginées, il est possible de reconstituer une dimension importante de l’apprentissage. La mesure des actions d’apprentissage ne serait cependant pas complète sans l’analyse des compétences associées au projet. L’apprentissage peut, en effet, résulter de l’apport de compétences nouvelles requises pour une tâche ou une mission particulière. Dans ce cas, les connaissances ne sont pas créées par test d’une hypothèse mais par adjonction de nouveaux répertoires de savoir, par extension interne ou externe du réseau d’acteurs associés au projet. Le choix d’équipes multifonctionnelles, et plus précisément le recrutement des membres de l’équipe basé sur une analyse des compétences attendues doit permettre l’apport des connaissances préalablement existantes sur un sujet donné. Dans ce dernier cas en particulier, mais aussi de façon plus générale, les mises à l’épreuve et l’apport de compétences ne résulteraient pas dans la production de nouvelles connaissances sans le partage de celles-ci et leur transformation en actions de l’organisation. 3.4 Le partage et l’intégration des connaissances Les travaux, tant théoriques qu’empiriques, mettent l’accent sur la socialisation comme l’un des facteurs clef de réussite des projets d’innovation. Un projet a d’autant plus de chance de 203 réussite s’il est porté par un réseau d’acteurs en communication étroite, et capables de transformer les échanges et expérimentations en décisions et actions. Ce constat se comprend d’autant mieux que l’on adopte la perspective de l’apprentissage organisationnel comme cadre intégrateur. Le concept même repose sur une extension des savoirs créés au niveau de l’ensemble de l’organisation. Il induit dans son objet même le partage des connaissances. Nous pouvons cependant distinguer plusieurs niveaux de partage. En premier lieu, la formalisation et l’échange contribuent à l’élaboration d’hypothèses. Par la discussion ou l’échange d’informations, les équipes confrontées à un problème cherchent à construire une première interprétation du problème posé en des termes qui permettent sa résolution. En second lieu, la communication permet de réaliser un test informel par la discussion. Par un simple échange, l’individu va pouvoir faire valider une première idée à moindre coût. Par partages et adoptions successives, la communication va permettre de tester l’idée, puis de la « tangibiliser ». En effet, la décision d’engager des ressources à plus grande échelle va reposer sur une décision collective d’avancement. L’ensemble de ces ajustements mutuels va permettre l’élaboration de décisions partagées puis la mise en œuvre des solutions, leur routinisation. Le fonctionnement efficace des nouvelles solutions repose en effet sur la reproduction en tous les points de délivrement de l’offre des actions décidées suite aux apprentissages. Sans mise en œuvre résultant des opérations précédemment décrites, on ne peut parler d’apprentissage organisationnel. La transformation des processus d’interactions avec le client, la réécriture des programmes informatiques, la modification des tâches de chacun, la réorganisation des départements font partie du processus d’apprentissage lui-même. La capacité de l’organisation à traduire des expérimentations en comportement ainsi que les moyens utilisés pour obtenir les comportements attendus doivent être considérés comme étant le processus d’apprentissage à part entière. Ces quatre pistes, identifiées à partir de la littérature sur l’innovation de services et l’apprentissage, constituent un premier pas vers l’opérationnalisation du concept. Elles permettent de mieux saisir le contenu théorique de l’apprentissage en relation avec les pratiques d’innovation. Le tableau suivant reprend nos quatre propositions et les met en relation avec les facettes de l’apprentissage traitées dans la première partie. 204 Tableau 2 : Quatre pistes d’opérationnalisation de l’apprentissage en regard de ses dimensions théoriques Sujets et objets : qui ? Implication Les individus et groupes fonctionnels dans un dans les contexte de sollicitations projets organisationnel Les individus et groupes par Gestion cheminement de systémique le du projet l’apprentissage du local vers le global Modalités : comment ? Déterminants : quand ? Par une réflexion liée à un projet d’amélioration potentiel préalable à son déclenchement Par une vision d’ensemble du projet dans une perspective intégrative En Formulant et partagent la problématique pour initier un processus d’interprétation En faisant se confronter les positions et contraintes des acteurs impliqués dans le projet pour une intégration respective En valorisant la diversité interne, la recherche de l’erreur et l’acceptation du risque. En assurant l’adjonction de compétences nouvelles par greffe. Les individus et groupe Par la recherche des Degré remarques et avis sur le d’ouverture fonctionnels problème et les solutions, et la promotion des d’expérimen initiatives, leur mise à tations l’épreuve, soutenus par des moyens formels et informels (réunions, discussions, débats, rapports…) Partage et Les individus en position de Par la mise à l’épreuve, la la intégration transformer les échanges en validation, décisions et actions. « tangibilisation » et la des diffusion des connaissance connaissances et s compétences nouvelles. En organisant le partage et l’échange d’informations et de solutions par l’existence de liens de communication Conclusion Dans ce travail, nous avons souligné les différentes manières d’approcher l’objet théorique qu’est l’apprentissage organisationnel. Aussi avons-nous mis en évidence l’absence d’un cadre théorique intégrateur. Dès lors, il est relativement difficile de donner une dimension opérationnelle au concept. En conséquence, au niveau instrumental, l’apprentissage s’avère d’une portée limitée. Afin de mieux cerner l’influence de l’apprentissage sur les pratiques de gestion et de clarifier le contenu opérationnel du concept, l’analyse des travaux sur l’innovation de services a été particulièrement éclairante. Nous avons à la fois démontré l’intérêt de l’apprentissage pour expliquer les logiques de succès et d’échecs des processus d’innovations et mis en perspective quelques pistes pour opérationnaliser le concept. Les apports se situent tant dans le champ du processus d’innovation que dans celui de l’apprentissage. Les retombées de ces mesures pour l’évaluation des succès et échecs dans le développement de nouvelles offres de service est à l’évidence l’un des champ d’application du concept. L’opérationnalisation des variables, en permettant de mesurer l’apprentissage sur ses principales dimensions, doit permettre aux gestionnaires en charge des développements de l’offre 205 d’améliorer la probabilité de succès. Sur le plan théorique, la mesure des variables permettra également clarifier le rôle joué par l’apprentissage. Son influence sur le processus d’innovation est explicite. Dans le champ de l’apprentissage organisationnel, ces premiers pas vers l’opérationnalisation visent à le rendre plus actionnable. L’instrumentaliser constitue une avancée managériale car les managers pourraient mieux le gérer. En en connaissant les finalités, ils pourraient accéder à une pratique dont les effets bénéfiques sur nombre de pratiques d’entreprises sont réels. Mais ces premiers pas vers l’opérationnalisation constituent une modeste avancée vers l’intégration théorique d’un concept particulièrement complexe à cerner. L’Eden de l’organisation apprenante est à ce prix. Bibliographie Alter, N., L’innovation Ordinaire, Paris, Presses Universitaires de France, 2003. Alter N., « Peut-on programmer l’innovation ? », Revue française de gestion, (mars-avril-mai 1995), 78-86. Alter N., La gestion du désordre en entreprise, Paris, L’Harmattan, Logiques Sociales, 1999. Argyris C., Savoir pour agir : surmonter les obstacles à l’apprentissage organisationnel, InterEditions, 1995. Argyris C., Schön D.A., Apprentissage organisationnel : théorie, méthode et pratique, De Boeck Université, 2002 (traduction de l’édition américaine de 1978). Baumard P., « Des organisations apprenantes ? Les dangers de la « consensualité » », Revue française de gestion, (septembre-octobre 1995), 49-57. Berry L.L., Service marketing is different, Business, May-June 1980. Brentani U.D., “Success and failure in new industrial services”, Journal of product innovation management, Vol 6, (1989), 239-258. Brentani U.D., “The New product process in financial services: strategy for success”, International Journal of Bank Marketing, Vol 11 n°3, (1993), 15-22. Callon M., La science et ses réseaux, Paris, La découverte, 1988. Callon M., Laredo P., Rabeharisoa V., « Que signifie innover dans les services : une triple rupture avec le modèle de l’innovation industrielle », La Recherche, n°282, (Décembre 1995), 78-83. Callon M., Latour B. et al., « A quoi tiennent les innovations ? », Gérer et comprendre Annales des Mines, (1998), 4-29. Cooper R.G., Easingwood C.J. et al., “What distinguishes the top performing new products in financial services”, Journal of product innovation management, Vol 11, (1994). Coriat B., Weinstein O., Les nouvelles théories de l’entreprise, Edition Le Livre de Poche, 1995. Daft R.L., Weick K., “Toward a Model of Organizations as Interpretation System”, Academy of Management Review, Vol.9, n°2, (1984), 284-295. Durand R., « Théorie évolutionnistes et management stratégique », in Martinet A.C., Thiétart R.A., Stratégies, Actualité et futurs de la recherche, Editions Vuibert, Collection FNEGE, (2001), 135-165. Easingwood C.J., Storey C., “Determinants of new product performance: a study in the financial services sectors”, International Journal of Service Industry Management, Vol 7, n°1, (1995), 32-55. Edgett S., Parkinson S., “The development of new financial services: identifying determinants of success and failure”, International Journal of Service Industry Management Vol 5, n°4, (1994), 206 24-38. Fillol C., « Apprentissage et systémique, une perspective intégrée », Revue française de gestion, (mars-avril 2004), 33-49. Gallouj F., Innovation dans une économie de service, Encyclopédie de l’innovation, in Mustar P., Penan H. Paris, Economica, 2003. Gallouj F., Weinstein O., “Innovation in services”, Research Policy, n°2, (1997), 537-556. Hart S.J., Service L.M., “Cross Functional integration in the new product introduction process: an application of action science in service”, Journal of Service Management, Vol 4 n°3, (1993). Hodgson G.M., Economics and institutions. A manifesto for a modern institutional economics, Grande Bretagne, Edition Polity Press, 1988. Kim D.H., “The link between individual learning and organizational learning”, Sloan Management Review, vol. 35, n°1, (Fall 1993), 37-50. Koenig G., « L’apprentissage organisationnel : repérage des lieux », Revue française de gestion, (janvier-février 1994), 67-85. Koenig G., Management stratégique, Paradoxes, interactions et apprentissages, Nathan, Paris, 1996. Laroche H., « L’entreprise close : une approche cognitive », in P. Besson, Dedans, Dehors, Editions Vuibert, Collection Entreprendre, (1997), 171-190. Lenfle S., Midler C. « Management de projet et Innovation », Encyclopédie de l’innovation, Paris, Economica, (2003), 49-69. Leroy F., « L’apprentissage organisationnel : une revue critique de la littérature », Acte de la VIIè conférence de l’AIMS, (1998). Leroy F., Ramanantsoa B., « Dimensions cognitives et comportementales de l’apprentissage organisationnel : un modèle intégrateur », in Perspectives en management stratégique, Tome IV, Edition Economica, (1997). Lievens A., Blazevic V., et al., “Antecedent of Project Learning and Time to Market during new Mobile Service Development”, International Journal of Service Industry Management, Vol 14, n°1, (2003), 120-147. Lievens A., Moenaert R.K., “Project team communication in Financial Service innovation”, Journal of Management Studies, Vol 37 n°5, (2000), 733-766. Lievens A., Moenaert R.K., “New service team as information processing systems: reducing Innovating Uncertainty”, Journal of Service Research, Vol 3 n°1, (2000), 46-65. Mack M., « L’organisation apprenante comme système de transformation de la connaissance en valeur », Revue française de gestion, (septembre-octobre 1995), 43-48. Madhavan R., Grover R., “From embeddded knowledge to embodied knowledge: New product development as knowledge management”, Journal of Marketing, n°62, (1998), 1-12. Maidique M.A., Zirger B.J., “The new product learning cycle”, Research Policy, Vol 14, (1985), 299-313. March J., “Exploration et exploitation in organizational learning”, Organization Science, No 2, Vol 1, (1991), 71-87. Mbengue A., « Le fonctionnement dual des organisations », Revue française de gestion, (juin-juillet-août 1997), 27-37. Midler C., « Evolution des règles de gestion et processus d’apprentissage », in Orlean A., Analyse économique des conventions, Paris, PUF, (1994). Midler C., L’auto qui n’existait pas. Management des projets et transformation de l’entreprise, Edition Dunod, 1998. Nonaka I., “The Knowledge creating company”, Havard Business Review, (november-december 1991), 96-104. 207 Nonaka I., “The dynamic theory of Organizational Knowledge Creation”, Organization Science, Vol 5, (1994), 14-37. Nonaka I., Takeuchi H., The Knowledge Creating Company, Oxford University Press, 1995. Pascale R., Managing on the Edge, Touchstone, 1990. Pearson A.W., Managing Innovation: an uncertainty reduction process, London, 1991. Piaget J., La naissance de l’intelligence chez l’enfant, Edition Delachaux et Niestlé, 1959. Quinn J.B., Baruch J.J. et al., “Technology in services”, Scientific American, Vol 6 n°257, (1987), 50-58. Raesfeld M.A., Ruyter K.D., et al., “Cooperation in new service development: a social dynamic approach”, Advances in Services Marketing and Management, Vol 5, (1996), 193-214. Roberts P.W., Amit R., “The dynamics of innovative activity and competitive advantage: The case of Australian retail banking, 1981 to 1995”, Organization Science, Vol 14 n°2, (2003), 107-128. Schein E.H., “How can organizations learn faster? The challenge of entering the green room”, Sloan Management Review, vol. 34, n°2, (Winter 1993), 85-92. Schumpeter J.A, Capitalism, Socialism and Democracy, London, Allen and Unwin, 1942. Senge P., The Fifth Discipline, New York, Doublebay, 1990. Shostack L.G., “Designing services that deliver”, Harvard Business Review, (January-February 1984). Simon H.A., “A Behavioural Model of Rational Choice”, Quarterly Journal of Economics, Vol 69, (1955), 99-118. Simon H.A., “Bounded Rationality and Organisational Learning”, Organization Science, Vol 2, (1991), 125-134. Stevens E., Managing the interpretations and learning while innovating: evidences from two longitudinal case studies, Newcastle upon Tyne, University of Newcastle upon Tyne, 2002. Stevens E., Dimitriadis S., Investigating the new service development process: towards a systemic dynamic model, 7 th International Research Seminar in Service Management, La Londe des Maures, France, 2002. Stevens E., Dimitriadis S., “New service development through the lens of organizational learning: preliminary evidence from longitudinal case studies”, Journal of Business Research, Vol 54, (2004), 1074-1084. Thiétart R.A., « Gérer entre l’ordre et le chaos », Cahier de Recherche DMSP, n°283, (mai 2000). Thiétart R.A., « Management et complexité : concepts et théories », in Martinet A.C. et Thiétart R.A., Stratégie, Actualité et futurs de la recherche, Vuibert FNEGE, (2001), 361-375. Thiétart R.A., Forgues B., « La dialectique de l’ordre et du chaos dans les organisations », Revue française de gestion, (mars-avril-mai 1993), 5-15. Zeithmal V.A, Bitner M.J., Services Marketing: Integrating Customer Focus across the Firm, Boston, Mc Graw Hill, 2003. 208 ASAC 2005 Toronto, Ontario Sujit Sur (student) John Molson School of Business Concordia University FURTHER ON THE THEORY OF STAKEHOLDER SALIENCE: DEFINING WHY AND WHEN DOES THE WHO REALLY COUNTS1 It is proposed that firms identify and revise their salient stakeholders (thus their institutional structure) contingent on the interests of (and the firm’s interest in) their salient stakeholders. Culture, industry, life-cycle stage and firm size determine this process, and this process is preceded by changes in cooperative agreements with stakeholders. Ever since Freeman (1984) made a persuasive case for attention to stakeholder interests as critical to firm success overriding the erstwhile dominant shareholder model, a debate has raged consistently about which aspect is central to the stakeholder theory, and more importantly, what might be the process of stakeholder identification (Berman et al., 1999). The recent increase in awareness of corporate governance issues led the proponents of stakeholder theory to focus primarily on the normative model which is concerned with how should the managers of firms deal with corporate stakeholders (Collins & Porras, 1994; Freeman, 1994; Paine 1994, Donaldson & Preston, 1995), neglecting the instrumental and descriptive realms which address the why and when the managers deal with such issues. The normative stakeholder model is widely accepted by researchers, however some questions remains as to how do firms identify their stakeholders, and what factors determine the institutional structure they adopt to reflect their interests in these stakeholders. Furthermore, once adopted, is this structure immutable or is there a convergence amongst the different institutional structures; and finally what factors cause this modification and what actions predict this process. This paper strives to address these questions by arguing that a firm initially adopts its institutional structure based on its circumstances and thereafter modifies its structure contingent on the changes in those circumstances. It also asserts that the identification and salience of stakeholder groups by a firm is dependent on its circumstances and its strategic choice. It also proposes that the modification of institutional structure by firms is not unidirectional as claimed by convergence theorists (Boyer, 1996; Coffee 1999; McCahery 2002), and the direction and extent of modification is unique to each firm. The factors that might shape these circumstances are explored and based on the relationship between the firm and its circumstances, some tentative propositions predicting the direction and the process of the modification are offered. This study thus is an important first step in furthering the understanding the stakeholder phenomenon in three distinct ways. Firstly, it builds a theory explaining the how, why and when firms chooses one structure over the other. Secondly, it advances a tentative rationale for the mixed 1 Inspired by Mitchell, Agle and Wood’s (1997) “Towards a Theory of Stakeholder Identification and Salience: Defining the Principle of Who and What Really Counts” and aspiring to “further” their theory of stakeholder salience. The invaluable mentoring by Dr. Jean McGuire and Dr. Rick Molz is gratefully acknowledged and has greatly enhanced this paper - all persisting errors of course, are my creation. 209 results of the studies by the convergence researchers. And finally, it explores the causality of the process of modification of institutional structure and attempts a prediction of this process. Institutional Structure: Defining the domain Historically, strategic literature has been grounded in industrial organizational economics. Sloan’s strategy-structure-systems trilogy, like the Taylorian managerial system, of breaking down complexity into simple components to make them systematic and predictable, served firms well from the 1920s, through the conglomerate diversification of the 60s, up to the start of globalization in the 70s and 80s. The theories and doctrines of this era essentially focused on the firm as the central institution (Berle and Means, 1932; Cyert and March, 1963; Porter, 1980). Everything else as viewed merely as its sub-components, either within or outside of the firm. ‘Objectives’ of firms were perceived as purely economic and the ‘responsibilities’ were secondary, modifying and constraining social influences (Ansoff, 1965).This was an era of tremendous growth especially in the Anglo-American firms; however as growth slowed, some of the inherent flaws in this doctrine started to become apparent (Ghoshal, Bartlett and Moran, 1999). The growing impact of the ‘extraneous’ elements on a firm led to a more encompassing view of a firm. There was also a movement towards including outside ‘clientele groups’ (King and Cleland, 1978) and stakeholders (Mason and Mitroff, 1982) as integral to the firm’s objectives. Despite some dissenting views (Friedman, 1970), maximizing profit was no longer perceived as the sole issue for firms. Management researchers started to seek more holistic approaches towards strategy and its implications in firms (Ackoff, 1974; Davis and Freeman, 1978; Trist, 1981). This resulted in the view of a firm as an institution within an overarching institution or a system. Since Freeman (1984) seminal work on ‘stakeholder theory’, this model was devoutly embraced and in spite of sporadic calls for rejections of the stakeholder model (Weyer, 1996; Key 1999, Gregg, 2001), this theory became entrenched in management literature. However, the large body of work that this interest produced often led to diverse and contradictory evidences and arguments on the understanding of the stakeholder concept. Donaldson and Preston (1995) asserted that the normative aspect of stakeholder theory was the most relevant, and as this view was also of most interest to the ethics and corporate governance researchers, the instrumental and descriptive aspects were largely ignored. The normative stakeholder concept was intriguing, the appeal across disciplines and the tenets juxtaposed to the erstwhile dominant shareholder model; thus the ensuing multi-disciplinary literature on the issue saw widely conflicting and sometimes contradictory views on both models (Jones, 1995; Wicks & Freeman, 1998; Kaler, 2003). Despite a few attempts at ‘merging of the two’ (Weaver & Trevino 1994; Blair, 1995; Shankman, 1999), stakeholder-shareholder centered structures were commonly understood as dichotomous edifices (Milgrom & Roberts, 1990; Charreaux & Desbrieres, 2001). This led to convergence theorists researching worldwide movement towards a single institutional structure amongst the two. Zingales (2000) asserts that as firms evolved from the capital intensive, manufacturing focused stage to the human capital and service industry stage, the focus shifted from the shareholders to the stakeholders. Of late the research points to convergence from stakeholder centered model to the shareholder centered one (Trevino & Weaver, 1999; Yoshikawa & Phan, 2001). Becht et al. (2003) find that the direction of ‘convergence’ is in fact towards whichever is the dominant economy of the period – towards stakeholder model in the 80s, to the shareholder model in the 90s and recently, back towards the stakeholder model. The results thus, so far have been mixed at best, and there seems to be no clear resolution to the debate in sight, at least not in the near future (Macharzina, 2000; McGuire and Dow, 2003). Thus the causes, direction and extent of changes in the stakeholder orientation and consequent institutional structure of firms remain unclear. 210 This paper asserts that this polarized view is a consequence of accepting the normative aspect as central to the stakeholder theory. The drawback of accepting stakeholder-shareholder as dichotomous and the resulting unidirectional convergence rationale is that acceptance of one structure meant overlooking its limitations and the strengths of the other. For example, if we accept that the normative stakeholder model as the only ‘viable’ model, we face the problem of identifying all relevant stakeholders, and such an aspirational standard will fail to provide managers and directors with any meaningful guidance, and is considered difficult to implement and enforce (Sternberg 1999). Such a model also poses a threat to private-property rights (Barry 1998) and distorts traditional notions of accountability in the firm (Gregg 2001). It might also lead to ‘rent seeking’ attempts by some to capture the wealth created by others on a grand scale (Barry 2002). Key (1999) also attacked the normative stakeholder model for its insufficient attention to the system within which business operates and the levels of analysis within the system. Taken to its extreme, the stakeholder model might end up being a ‘collectivist’ approach, thereby creating an untenable artificial market like those of the communist states (Weyer 1996). Clearly the normative stakeholder model or the IO shareholder model by themselves fails to provide either practitioners or researchers with a satisfactory resolution of the issues. Attempts at a convergent (normative & instrumental) stakeholder theory (Jones and Wicks, 1999) was strongly contested by others (Freeman, 1999; Gioia, 1999; Trevino et al., 1999; Donaldson, 1999) The few studies that argued for the instrumental or descriptive stakeholder model (Mitchell, Agle and Wood, 1997; Berman, Wicks, Kotha and Jones, 1999; Jawahar, and McLaughlin, 2001), though furthering the understanding of stakeholder management, were still unable to resolve some of the key issues. Berman, Wicks, Kotha and Jones (1999) conducted an empirical study that found no support for the normative (intrinsic) stakeholder model, while their proposed strategic stakeholder management model was seen as a robust and predictive model of firm performance. This study thus clearly demonstrated the importance of stakeholder orientation in the selection of institutional structure, however, the issues of stakeholder identification, variation in stakeholder salience, and more importantly, the cross cultural and industry level effects on this process were not addressed. Fact remains that firms do pay differing attention on their constituents. Thompson (1967) proposed that when the organization cannot show improvement on all dimensions, it seeks to show its improvement on the elements of interest to those on which it is most dependent and emphasize scoring well on those criteria which are most visible to important task-environment elements. Davis (2003) illustrates this point as follows: “In 1998 when the Coca Cola Company had achieved its historical high share price and shareholder dominance was at its zenith, its mission statement was: “We exist to create value for our share owners on a long-term basis by building a business that enhances the Coca-Cola Company’s trademarks.” In 2003, with its price at half the apex it had reached, it was “The Coca-Cola Company exists to benefit and refresh everyone who is touched by our business.” Also at its 1998 price peak, “Sara Lee Corporation’s mission is to build leadership brands in consumer packaged goods markets around the world. Our primary purpose is to create long-term stockholder value.” In 2003, “Sara Lee Corporation’s mission as a company is to feed, clothe and care for consumers and their families the world over.” The oft cited study by Mitchell, Agle and Wood (1997) addresses this issue of stakeholder salience and offers a model wherein the stakeholder salience is related to the cumulative number of stakeholder attributes of power, legitimacy and urgency. They propose that the process of stakeholder salience is dynamic in nature and is caused by the changes in these three attributes. Thus the stakeholder with more attributes will be perceived as more salient. However, as they themselves admit as a limitation of their study, the attribute of legitimacy might be subsumed 211 within the attribute of power, thus drastically reducing the explanatory value of the model. Oliver (1991) recognizes that these opposing forms of organizational behavior, passive conformity (legitimacy) and active resistance (power) actually represent two ends of the same continuum. Also, the Mitchell et al. (1997) model does not address the issue of managerial action with the stakeholders who vary in terms of salience (Jawahar et al., 2001). Furthermore, the fact that the relations with some of the stakeholders (e.g. special interest groups, environmental groups, and media) are based upon ill-defined and changing expectations, implicit contracts, and culturally held normative beliefs makes assessment of power, salience, and urgency more difficult than Mitchell et al. (1997) might imply. Jawahar and McLauglin (2001) better address these issues by proposing a descriptive stakeholder theory based on the life-cycle stages of a firm. They draw on resource dependence and prospect theory to propose that depending on the firm’s life-cycle stage, the firm will choose either a gain frame, i.e. risk averse (all stakeholders oriented) strategy; or a loss frame, i.e. risk seeking (only primary stakeholder oriented) strategy. They utilize Clarkson’s (1991, 1995) RDAP (reactive, defensive, accommodating, and proactive) scales of strategy for stakeholder management, and propose that firms adopt proactive or accommodating strategies with their salient stakeholders and reactive and defensive towards the others. This study lends support to the concept of firms’ strategically identifying their salient stakeholders and also encompasses the dynamic nature of variation in stakeholder salience, and furthermore, it also explains the managerial action towards these salient stakeholders. However, as this study limits itself to the life-cycle stage of the firm as the sole determinant factor for stakeholder salience, it does not address the relevant issues beyond the firm level. As Roberts and Greenwood (1997) proposed in their constrained-efficiency framework, strategic choice and organizational transformation are complicated by institutional contexts. Davis’ (2003) illustration of Coca Cola’s and Sara Lee’s drastic change in focus on stakeholder groups within a relatively short period also suggests involvement of macro-environmental factors beyond the firm level. Thus Jawahar et al. (2001) model will be unable to explain variation in stakeholder orientation of firms that might be at the same life-cycle stage but in different industries and nationalities (or cultures), or the similarity between firms at differing life-cycle stage but in the same industry or culture. This paper attempts to address this gap in the literature and offers a descriptive model that integrates both firm level and institutional level forces, and offers possible factors that determine the process of identification and variation in salient stakeholder groups across these levels. The ‘Contingent’ Stakeholder Model This paper argues that the salience of the stakeholder, and the resulting institutional structure that a firm adopts to reflect its interest in (or the interests of) this stakeholder group, is contingent on its circumstances as determined by the interaction between the following factors: Life-cycle stage of the firm Industry life-cycle stage Culture (including regulations) Industry type It is also argued that firms modify their structure depending on its emergent circumstances that results from a change in any of the determining factors. This process of modification is preceded by changes in the cooperative agreements between the firm and its stakeholder groups. This model is asserted to be better representative and descriptive of firms’ institutional structure and offers a rationale for the adoption and modification of its institutional structure. This model builds on the instrumental stakeholder aspect of the stakeholder theory to explore why and when different 212 stakeholders gain in salience for firms. The premises for this model are developed further in the following propositions section. -----------------------------------insert figure 1 here ------------------------------------ Propositions Industry and organizational life cycle stage as a determinant factor: Chandler (1962) right at the very introduction of ‘stages of life-cycle’ model notes that as stages changed, so did firm’s strategies and structure. Further theory and research suggest that the pressures, threats, and opportunities in the environment of a firm vary with its life cycle stages (Anderson & Zeithaml, 1984; Dodge & Robbins, 1992). Jawahar et al. (2001) argue that at any given organizational life-cycle stage, certain stakeholders become more important than others. Their propositions state the nature of the changing importance of stakeholder groups to an organization at every stage of its life-cycle. While in full agreement with this assertion, we argue that this process is applicable for industries also. An illustrative example can be the Computer and Information Technology industry. At the onset, this industry were focused primarily on gathering financial resources and competitive analysis and thus were near exclusively shareholder oriented due to cognitive and resource limitations (Bourgeois & Eisenhardt, 1988), much like the neoclassical Input–Output (shareholder) model firms. No environmental impact of such firms was even envisaged at that stage. At its intermediate stage, this industry pioneered stock options in view of the invaluable contribution of its employees, and thus the industry norm became inclusive of employees as a major ‘shareholder’ or a highly salient stakeholder. The industry also thereafter actively involved its suppliers in its strategies and operations (Burt, 1989). As the industry further matured, the unforeseen problem of environmental pollution caused by the exponentially growing obsolete computers led these firms to hitherto include the EPA and other bodies into consideration (Knemeyer et al., 2002). Another example is the automobile industry. This was a classical Taylorian industry at its inception; in fact most of the early management models were based on the Ford assembly line functionality, wherein the most primary of stakeholders – the firm’s employees, were considered as replaceable parts. However, with this industry approaching its mature stage, automobile companies like GM are one of the only few corporations offering ‘lifetime employment’ to its employees (Zellner & Berstein, 1987). We build on the rationale and propositions of Jawahar et al. (2001) by including the industry life-cycle stage as a determining factor, and also utilize the same Clarkson (1995) RDAP scale to reflect stakeholder salience (The proactive strategy is to anticipate responsibility, accommodative is to accept responsibility, defensive is to admit responsibility but fight it, and reactive is to deny responsibility). We also expand the ‘environmental group’ as described by Jawahar et al. (2001) to ‘special interest group’ so as to include the stakeholder groups of ‘media’, ‘political groups’ and ‘environmental groups’ as defined by Freeman (1984). Thus we propose: Proposition 1a: Firms and industries in the startup stage of their life-cycle will tend to be proactive towards shareholders, creditors and customers; accommodative towards employees and suppliers; reactive towards trade associations and special interest groups; and defensive towards government and community stakeholders. Proposition 1b: Firms and industries in the growth stage of their life-cycle will tend to be proactive towards creditors, employees, suppliers and trade associations; accommodative towards shareholders, customers, government, special interest groups and community stakeholders. 213 Proposition 1c: Firms and industries in mature stage of their life-cycle will tend to be proactive towards shareholders, employees, suppliers, customers, trade associations, government, special interest groups and community stakeholders; and accommodative towards creditors. Proposition 1d: Firms and industries in decline/ transition stage of their life-cycle will tend to be proactive towards shareholders, creditors and customers; accommodative towards employees and suppliers; reactive towards trade associations and special interest groups; and defensive towards government and community stakeholders. Culture as a determinant factor: The cultural influence on the institutional structure adopted by corporations in a country is well researched (Nakane, 1970; Kitano, 1970; Kumagai, 1992). Many scholars have found support for the notion that society is a composition of institutional logic and this is the underlying basis for the organizations therein (Galvin, 1999). These institutions in the firm’s external environment place pressures and constraints upon organizational behavior by supporting the enforcement of rules, laws, values, and social norms (North, 1990; Scott, 1990; Coffee, 2001). Leontiades (1990) and Nonaka (1990) pointed out differences in national orientations towards addressing even the overseas subsidiaries by a firm. Thus clearly all the institutional structures adopted by a firm are dependent on the national culture of the firm. The national regulatory structures that govern firms are also derived from the national culture per se (Dammann, 2003). Dooley (1995) further argues that in the individualistic cultures like US, maximizing shareholder wealth is by far the most important goal of US corporate law. Others too support this notion (Bainbridge, 1993; Black & Kraakman, 1996). Similarly, there is ample support for the notion that in the collectivist cultures like Japan (Nagai & Bennett, 1953; Bhappu, 2000) and Germany (Schmidt, 1997; Bradley, 1999; Cunnigham, 1999), the corporate law emphasizes the maximization of stakeholder values, especially the interests of employees, suppliers and creditors (Becht, 2003). Hart (1992) also finds that Japanese corporations are firmly enmeshed with the government, trade associations (Keidanren), political groups (MITI) and the community. In view of the above, it is proposed that: Proposition 2a: Firms in individualist cultures will tend to be proactive towards shareholders, creditors and customers; accommodative towards employees and suppliers; reactive towards trade associations and special interest groups; and defensive towards government and community stakeholders. Proposition 2b: Firms in collectivist cultures will tend to be proactive towards employees, suppliers, government, and community stakeholders; and accommodative towards shareholders, creditors, customers, trade associations and special interest groups. Type of Industry as a determinant factor: Brammer and Millington (2002, 2003) in their empirical study of UK firms found that the choice of institutional structure by a firm reflected distinct industry characteristics. Fineman and Clarke (1996) also found that firms in different industries encounter different intensities and types of stakeholder pressures. Adams and Mehran (2003) also found firms in the financial industry showed distinctly different institutional structure vis-à-vis those in the Manufacturing industry. Also firms in ‘politically sensitive’ industry (e.g. nuclear power generation firms, mining, and firms in heavy effluence producing industries) are perforce required to pay particular attention to the Environment Protection and Community Protection groups. McLarney (2002) found that community action groups not only targeted ‘defaulting’ firms in the canning industry (those who were dumping effluents into a lake), but also included the ‘non-defaulting’ firms. Thus the action was industry wide and not firm specific. Likewise, Heugens, et al. (2002) found that the Dutch food industry as a whole tried to establish structural linkages with the national government bodies and Consumer League to preserve its 214 autonomy in biotechnology-related (for genetically modified food products) affairs. These ‘additional’ pressures are not felt by firms in more ‘mainstream’ industries as their narrower focus and impact perforce reduces the extent of pressure from their relevant stakeholders. Firms achieve this by means of professional associations that bargain with governmental agencies on standards of service and accountability (Oliver, 1991). Likewise, firms also exert important effects on governmental structures by means of petitioning, lobbying and “expert” testimonies (Dacin, Goodstein & Scott, 2002). Therefore, it is proposed: Proposition 3a: Firms in ‘sensitive’ industries will tend to be proactive towards government, trade associations, special interest groups, customers and community stakeholders; and accommodative towards shareholders, creditors, employees and suppliers. Proposition 3b: Firms in ‘mainstream’ industries will tend to be proactive towards shareholders, creditors and customers; accommodative towards employees and suppliers; reactive towards trade associations and special interest groups; and defensive towards government and community stakeholders. Interaction between the determinant factors. Each of the abovementioned determinant factors by itself will merely indicate the tendency of the firm’s stakeholder orientation. It is argued that the actual structure that a firm will adopt will be determined by the interaction among these factors. As Rowley (1997) stated, organizations do not simply respond to each stakeholder individually; they respond to the interaction of multiple influences from the entire stakeholder set. Tylecote et al. (1998) found differences among Britain and Japan in the financial systems, and amongst industries. Dodge et al. (1994) also found interaction between organizational life cycle and size of firms that dictated the institutional structure of the firm. Ely & Pownall (2002) found similar stakeholder-related incentives among Japanese firms co-listed in US, when matched by industry and size. Gioia (1999) asserted that the firm’s strategic choices are affected by the sociopolitical stakeholder network, and the network is changed over time as a result of the firm’s strategic interaction with sociopolitical actors. In other words, both the external influences and the strategic choice of which stakeholder group to target, influences the adoption of its institutional structure by a firm. Thus, it is proposed that: Proposition 4: Firms adopt its institutional structure contingent on its strategic choice and the interactions between the determinant factors. Modification of institutional structure. It is further argued that firms modify their adopted structure dependent on the emergent changes in its determinant factors. Convergence theory considers this modification as unidirectional – in the mid 80s as towards shareholder model (Harbison, 1959; Dunphy, 1986), 70s to mid 90s as towards the stakeholder model (Vogel, 1979; Ouchi, 1981) and of late back towards the shareholder model (Yoshikawa & Phan, 2001). A similar finding is reported by Becht et al. (2003). However, the evidence for such a unidirectional movement is mixed at best (McGuire & Dow, 2003; Gifford & Kudrle, 2003). This paper argues that this movement is not unidirectional and is in fact is a dynamic process, unique to each firm, caused due to the series of ongoing modification because of the variation in the stakeholder salience caused by the interaction between the above mentioned factors. It is thus proposed: Proposition 5: Firms modify their institutional structure and the extent and direction of this modification is dependent on the variation in stakeholder salience caused by the emergent changes and interactions within the determinant factors 215 Cooperative arrangements as precursors of modification. Downsizing literature (Morris et al., 1999; Laabs, 1999; Beylerian & Kleiner, 2003) is replete with studies finding firms moving towards greater profitability by downsizing despite a booming economy (indicating a shareholder orientation) and this process involved systematically reducing cooperative arrangements with erstwhile partners. -------------------------------insert figure 2 here --------------------------------On the other hand, Deri (2003) argues that today’s firms need to ‘befriend’ all relevant stakeholders in order to sustain growth by means of ‘enlisting and engaging multiple partners and perspectives’ in the proposed movement towards a more stakeholder oriented model. Dyer and Singh (1998) also asserted that cooperative strategy adopted by firms led to a relational view that allowed interorganizational competitive advantages over the traditional resource based firms (shareholder model). Savage et al. (1991) while asserting that effective strategy requires consensus from a plurality of key stakeholders, proposed the following diagnostic typology for identifying and managing stakeholders. Such a typology implies that as a firm moves towards a stakeholder oriented institutional structure, it will collaborate or involve its salient stakeholder groups; and as it moves towards a shareholder centered structure, it will adopt a defensive strategy with low cooperation with the stakeholder group. Nasi et al. (1997) assert that issues are driven by stakeholder groups (and their alliances) that are constantly negotiating for support from external stakeholder groups and from within firms, and that firms manage the issues by a process of negotiation and communication to satisfy critical stakeholder group as a means of reducing the legitimacy gap. Wolfe & Putler (2002) suggest options such as coalition building to form priority based clusters of stakeholders. Thus it is proposed that: Proposition 6a: Modification towards stakeholder oriented institutional structure will be preceded by increase in cooperative arrangements with stakeholder groups that are increasing in salience Proposition 6b: Modification toward shareholder oriented institutional structure will be preceded by decrease in cooperative arrangements with stakeholder groups that are decreasing in salience Research and Managerial Implications The dyadic shareholder-stakeholder oriented institutional structures can be argued to be interrelated – the differences being conceptual in design, and artificial in practice. For example, maximizing stakeholder loyalty is with the goal of improving business performance (Wheeler and Sillanpaa, 1997). Berman et al. (1999) also assert that firms view stakeholders as part of an environment that must be managed in order to assure returns to shareholders. Freeman’s (1984) model included the ‘investor’ stakeholder group. Similarly, the shareholder model also did encompass stakeholders - although as constraining (Ansoff, 1965), conflicting or competing forces (Gregg, 2001). The model as presented in this paper enables clarity in comprehending of a firm’s institutional structure vis-à-vis its shareholder-stakeholder orientation. We assert that the firm strategically chooses its institutional structure to reflect its interest in (or the interests of) its salient stakeholders while working within its institutional environment, or in other words, the structure of a firm is not only based on endogenous factors (life-cycle stage), but is also embedded in its multilayered institutional context (culture, industry, and industry life-cycle). This paper thus 216 furthers the dynamic stakeholder identification and salience model proposed by Mitchell et al. (1997), and Jawahar et al. (2001), by including the institutional and industrial-level factors that are extraneous to the firm, but nevertheless impact its stakeholder orientation. Thus this model offers a comprehensive method of identifying the salient stakeholders for any given firm. This model also meets Gioia (1999) call for “adequately representing the complex social and organizational realities that management faces”; and offers a tool to enable managers, special interest groups, industry associations, and regulators to effectively target their interventional efforts towards the appropriate stakeholders. A major managerial implication of this study is the fact that managers are not solely responsible of choosing the stakeholder orientation, and thus the institutional structure of their firm – in contrast to the exhortations of the proponents of normative stakeholder theory; and thus influential stakeholders groups like government, political groups, special interest groups and media also need to bear responsibility for shaping the stakeholder orientation and thus the institutional structures of the firms within their domain. For researchers, a possible research interest may be exploring additional factors that determine the firm’s stakeholder orientation. The determining factors as proposed are by no means claimed to be exhaustive. Of special interest might be the role and effect of (i) ownership and (ii) competitive dynamics between firms, on salient stakeholder identification and the adoption and subsequent modification of institutional structure because of these factors. We opine that competition dynamics might very well be a major influential factor for the change in stakeholder orientation of Coca Cola and Sara Lee, as illustrated by Davis (2003) between 1998 and 2003. Also, further research investigating and refining the model, and operationalizing these factors to empirically study the model might be of interest. By understanding that the process of identification of salient stakeholder operates on multiple levels, and thus utilizing appropriate statistic tools like hierarchical linear modeling, we may be able to distinguish between the firm, industry and organization level effects and hence better understand the interaction between these factors. Limitations All the arguments in this paper are based on the basic premise that the stakeholder orientation of a firm defines the institutional structure adopted by the firm and not vice versa. However, if the institutional structure is viewed as a consequence of the ownership structure of the firm, then the ideology and norms and beliefs of the dominant owner or ownership group might become the most important deterministic factor. In such a case, the existing corporate ‘morality’ and ethics will define the strategic choice of the firm, and thus the normative (intrinsic) aspect of stakeholder theory will be salient. However, Berman et al. (1999) found no support for such a normative stakeholder model. Further research investigating firm ownership as a factor determining institutional structure is certain to lead to a better understanding of such a relationship, if indeed it exists. Another limitation of this paper is the fact that the relationship between firm size and its institutional structure has not been considered. It might well be that the firm size will moderate the stakeholder orientation – institutional structure relationship. However, it is also possible that this effect might be the result of the interaction between firm size and firm life-cycle stage, which are generally closely related. Further analysis of this issue is called for and this again might be a useful focus of further research studies. References Abe, Etsuo: “The development of modern business in Japan”. Business History Review; Summer 1997; 71, 2 217 Adams R.B., and Mehran, H. 2003. Is Corporate Governance Different for Bank Holding Companies? Economic Policy Review. 123-143 Abrams, F.: “Management Responsibilities in a Complex World” in T. Carroll (ed.) Business Education for Competence and Responsibility. Chapel Hill: Univ. of NC Press. 1954 Ackerman, R. W. 1975. The Social Challenge to Business. Cambridge, MA: Harvard University Press Anderson C.R. & Zeithaml C.P.1984. Stage of product life cycle, business strategy, and business performance. Academy of Management Journal. 27: 5-24 Ansoff, I.1965. Corporate Strategy. New York: McGraw Hill Inc. Ackoff, R.L.1974. Strategic Management. New York: John Wiley and Sons. Bainbridge S.M.1993. In Defense of the Shareholder Wealth Maximization Norm: A Reply to Professor Green. Washington & Lee L. Rev 50. 1423, 1423-25. Barry, N. 1998. Business Ethics. London: Macmillan. Barry, Norman. 2002. The Stakeholder Concept Of Corporate Control Is Illogical And Impractical. Independent Review, Vol. 6, Issue 4, 1086-1653 Becht, M., Bolton, P., & Roell, A. 2003. Corporate Governance and Control. Handbook of the Economics of Finance, Edited by G.M. Constantinides, M. Harris and R. Stulz. Elsevier B.V Berle, A. and Means G. 1932. The Modern Corporation and Private Property. New York: Commerce Clearing House Berman S.L., Wicks A.C., Kotha S., Jones T.M.1999. Does stakeholder orientation matter? Academy of Management Journal. 42,5;pg. 488 Bernard , C. 1938. The function of the Executive. Cambridge: Harvard University Press Beylerian M.; Kleiner B. H. 2003. The downsized workplace. Management Research New. 26, 2-4 Bhappu A. D. 2000. The Japanese family: An institutional logic for Japanese corporate networks and Japanese Management. Academy of Management Review. 25, 2. Pg. 409-415. Bird, Allan: Book Review of “Hybrid Factory: The Japanese Production System in the United States”,”The Japanese Firm:Sources of competitive strength” and “Does ownership matter”.Academy of Management. The Academy of Management Review; Apr 1996; 21, 2; Black, Bernard & Reinier Kraakman. 1996. A Self-Enforcing Model of Corporate Law. Harvard Law Review 109. 1911-1921 Blair Margaret M. 1995. Ownership and Control: Rethinking Corporate Governance for the Twenty-First Century, Washington. The Brookings Institution Bourgeois, L. J., III; Eisenhardt, Kathleen M. 1988. Strategic Decision Processes in High Velocity Environments. Management Science, 34, 7 Boyer, R. 1996. The convergence hypothesis revisited: Globalization but still the century of nations? National Diversity and Global Capitalism. S. Berger and R. Dore. Ithaca, Cornell University Press. Bradley, M., schipani, C., Sundaram, A., and Walsh, J. 1999. The Purpose and Accountability of the Corporation in Contemporary Society: Corporate Governance at a Crossroads. Law and Contemporary Problems, 62: 9-86 218 Brammer S, Millington A. 2002. The Composition of Organizational Slack and the Nature of corporate Community Involvement. Presented at Academy of Management Meeting, Denver. Brammer S, Millington A. 2003. The effect of stakeholder preferences, organizational structure and industry type on Corporate Community Involvement. Journal of Business Ethics, 45, 3 Burt D. N. 1989. Managing suppliers up to speed. Harvard Business Review, 67, 4 Chandler A.D.1962. Strategy and Structure. Cambridge, MA: MIT Press. Charreaux, Gerard; Desbrieres, Philippe. 2001. Corporate Governance: Stakeholder Value versus Shareholder Value. Journal of Management & Governance. 5, 2 Clarkson M. B. E. 1991. Defining, Evaluating, and Managing Corporate Social Performance: The Stakeholder Management Model. In L.E. Preston (Ed.), Research in Corporate Social Performance and Policy; 12, 331-358. Greenwich, CT. JAI Press. Clarkson M. B. E. 1995. A Stakeholder Framework for Analyzing and Evaluating Corporate Social Performance. Academy of Management Review; 20, 92-117 Clegg Stewart; Kono Toyohiro. 2002. Trends in Japanese management: An overview of embedded continuities and disembedded discontinuities. Asia Pacific Journal of Management, 19, 2, 3 Coffee, J. C. J. 1999. The Future as History: The Prospects for Global Convergence in Corporate Governance and its Implications. Northwestern University Law Review 93(3): 641-708., 1999. Coffee, John C. 2001. Do Norms Matter?: A Cross-Country Examination of the Private Benefits of Control. Columbia Law and Economics Working Paper No. 183. http://ssrn.com/abstract=257613 Coles Jerilyn W, McWilliams Victoria B, Sen Ninaljan. 2001. An examination of the relationship of governance mechanisms to performance. Journal of Management, 27 (2001) 23–50 Cunningham L.A. 1999. Commanalities and Prescriptions in the Vertical Dimension of Global Corporate Governance. Cornell Law Review, 84: 1133 Cyert, R. and J. March. 1963. A behavioral Theory of the Firm. Englewood Cliffs: Prentice Hall Inc. Dacin Tina M.; Goodstein Jerry; Scott Richard W. 2002. Institutional Theory and Institutional Change: Introduction to the Special Research Forum. Academy of Management Journal; 45, 1, 45-57 Dammann J. C. 2003. The future of codetermination after Centros: Will German corporate law move closer to the US Model. Fordham Journal of Corporate & Financial Law; 8, 2. Davis Gerald. 2003. Organizations in Action: Social Science Bases of Administrative Theory, reprint ed. Administrative Science Quarterly; 48, 3, p. 502 Davis, P. and Freeman E. 1978. Technology Assessment and Idealized Design in Evaluating new Telecommunications Services. 325-344. New York: Plenum Press Deri, Chris. 2003. Make alliances, not war, with crusading external stakeholders. Strategy & Leadership; 31, 5 Derkinderen, F. and R. Crum. 1979. Project Set Strategies. Boston: Martinus Nijhoff Publishing. Dodge H.J. & Robbins J.E.1992. An empirical investigation of the organizational life cycle model for small business development and survival. Journal of small Business Management.; 30(1): 27-37 219 DodgeH.J., Fullerton S., & Robbins J.E. 1994. Stage of the organizational life cycle and competition as mediators of problem perception for small businesses. Strategic Management Journal; 15: 121-134 Donaldson, T, and Preston L.E. 1995. The Stakeholder Theory of the Corporation. Concepts, Evidence, and Implications. Academy of Management Review, 20, 1: 65-91 Dooley M.P. 1995. Fundamentals of Corporation Law 97. Dunphy D 1986. An Historical Review of the Literature on the Japanese Enterprise and its Management. In S R Clegg, D Dunphy, and S G Redding (eds.), The Enterprise and Management in East Asia. Hong Kong: Centre of Asian Studies: University of Hong Kong. 334-686 Dunphy D. 1987. Convergence/Divergence: A Temporal Review of the Japanese Enterprise and Its Management.Academy of Management Review; 12, 3 Dyer, Jeffrey H; Singh, Harbir. 1998. The relational view: Cooperative strategy and sources of interorganizational competitive advantage. Academy of Management Review; 23, 4, 660-680 Ely, Kirsten M; Pownall G. 2002. Shareholder- versus stakeholder-focused Japanese companies: Firm characteristics and accounting valuation. Contemporary Accounting Research; 19, 4 Fineman, S, Clarke K. 1996. Green stakeholders: Industry interpretations and response. Journal of Management Studies. Oxford: 33, 6; pg. 715 Freeman R.E. 1984. Strategic Management: A stakeholder approach. Boston: Pitman Freeman, H. E. 1999. Making Stakeholder Theory Whole. Academy of Management Review, 24, 2 Friedman M. 1970. The Social Responsibility of Business is to Increase its Profits. In T. Beauchamp & N. E. Bowie (Eds.), Ethical Theory and Business. 55-60. Englewood Cliffs, NJ: Prentice-Hall Galbraith J.R. 1973. Designing Complex Organizations. Addison-Wesley, Reading, MA. Galvin, T. L. 1999. Contradiction, contestation, and change: An institutional reinterpretation of US healthcare: 1890-1980. Working paper, University of Texas at Dallas. Gerschenkron A.: Economic Backwardness in Historical Prespective. Cambridge 1962. Chapter1 Ghoshal S, Bartlett CA, Moran P. 1999. A New Manifesto for Management. Sloan Management Review. Cambridge.40, 3; pg. 9 Gifford Daniel J; Kudrle Robert T. 2003. European Union Competition Law and Policy: How Much Latitude For Convergence With The United States. Antitrust Bulletin; 48, 3; Gioia, Dennis A. 1999. Response: Practicability, Paradigms, and Problems in Stakeholder Theorizing. Academy of Management Review, 24, 2 Goodstein J., Gautam K., & Boeker W. 1994. The effect of board size and diversity on strategic change. Strategic Management Journal. 15, 241-250 Gregg Samuel. 2001. ‘Stakeholder' theory’. Policy, 10326634, 17, 2 Harbison F.: Management in Japan. 1959. In F. Harbison & C A Myers (Eds.), Management in the industrial world: An international analysis. New York: McGraw Hill; 249-264 Hart Jeffrey. 1992. Rival Capitalists: International competitiveness in the United States, Japan, and Western Europe. Ithaca, NY: Cornell University Press 220 Heugens, P M A R Pursey; Van Den Bosch, Frans A J; Van Riel, Cees B M. 2002. Stakeholder Integration. Business and Society; 41, 1 Hillman Amy J; Keim Gerald D; Luce Rebecca A. 2001. Board composition and stakeholder performance: Do stakeholder directors make a difference. Business and Society, 40, 3, 295 Jones T.M., Wicks, A.C. 1999. Convergent stakeholder theory. Academy of Management Review. 24 Jawahar, I M; McLaughlin, Gary L. 2001. Toward a descriptive stakeholder theory: An organizational life cycle approach. Academy of Management Review; 26, 3 Kaler J. 2003. Differentiating stakeholder Theories. Journal of Business Ethics; 46,1; 71. Key Susan. 1999. Towards a new theory of the firm: A Critique of Stakeholder Theory. Management Decision. London. 37, 4; 317. Kitano S. 1970. Dozuku and kindred in a Japanese rural society. In R. Hill & R. Konig (Eds.), Families in East and West: Socialization process and kinship ties. 248-269. The Hague: Mouton. Knemeyer, Michael A.; Ponzurick, Thomas G; Logar, Cyril M. 2002. A qualitative examination of factors affecting reverse logistics systems for end-of-life computers. International Journal of Physical Distribution & Logistics Management; 32, 6 Kochan Thomas A; Rubinstein Saul A. 2000. Toward a stakeholder theory of the firm: The Saturn partnership. Organization Science; 11, 4; Kumar Krishna B., Rajan Raghuram G., ZingalesLuigi. 2001. What Determines Firm Size? NBER working paper, University of Chicago, NBER. http://gsbwww.uchicago.edu/fac/finance/ papers/size.pdf Kumagai F. 1992. Research on the family in Japan. In The changing family in Asia: 159-237. Bangkok: UNESCO. Laabs J. 1999. Has downsizing missed the mark? Workforce. Leontiades M. 1990. Scenarios for the Nineties. Paper presented at the Tenth Annual Strategic Management Society Meetings, Stockholm, Sweden. Mahon, John F., Sandra A Waddock. 1992. Strategic Issues Management: An Integration of Issue Life Cycle Perspectives. Business & Society. 31, 1 Mason, R. and I. Mitroff. 1982. Challenge Strategic Planning Assumptions. New York: John Wiley and Sons. Macharzina, Klaus. 2000. The Japanese model--out of date? Management International Review; 40, 2 McGuire J. and Dow S. 2003. The Persistence And Implications Of Japanese Keiretsu Organization. Journal of International Business Studies. Washington; 34, 4; 374 McLarney C. 2002. Stepping Into the Light: Stakeholder Impact on Competitive Adaptation. Journal of Organizational Change Management; 15, 3 Modis T. 1994. Forecasting the rise and fall of almost anything. The Futurist. September/October, 20-24 Milgrom P., and Roberts, J. 1990. The Economics of Modern Manufacturing: Technology, Strategy, and Organization. American Economic Review, 80, 3: 511-528 221 Millon D. 1993. Communitarians, Contractarians, and the Crisis in Corporate Law. Washington & Lee Law Review. 1373-1390 Morris J.R., Cascio W.F., and Young C.E. 1999. Downsizing after all these years: Questions and answers about who did it, how many did it, and who benefited from it. Organizational Dynamics. Meyer J, Rowan B. 1977. Institutional Organizations: Formal structure as myth and ceremony. American Journal of Sociology; 83: 340 -363 Nagai M & Bennett J.W.1953. A summary and analysis of ‘The Familial Structure of Japanese Society’ by Takeyoshi Kawashima. Southwestern Journal of Anthropology. 9: 239-250. Nakane, C.1970. Japanese Society. Berkeley: University of California Press. Juha Nasi; Salme Nasi; Nelson Phillips; Stelios Zyglidopoulos. 1997. The evolution of corporate social responsiveness. Business and Society; 36, 3 Nonaka I. 1990. Managing Globalization as a self-renewing process: Experiences of Japanese MNCs. In Bartlett CA, Doz Y, and Hedlund G. Managing the Global Firm. London: Routledge. North D.C. 1990. Institutions, Institutional Change and Economic Performance. Cambridge University Press, Cambridge. Oliver C. 1991. Strategic Responses to Institutional Processes. Academy of Management Review; 16, 1 Orlitzky M.: “A meta-analysis of the relationship between corporate social performance and firm financial performance”. Unpublished doctoral dissertation. The university of Iowa, Iowa city, IA; UMI no. AAt 9904332; DAI-A 59/09, p. 3527. March 1999 Peng Mike W. 2004. Outside Directors and Firm Performance during Institutional Transitions. Strategic Management Journal; 25: 453-471 Pfeffer J.; Salancik G. 1978. The External Control of Organizations: A Resource Dependence Perspective. New York. Harper & Row. Pinkston T.S. & Carroll A.B. 1993. An investigation of the relationship between organizational size and corporate social performance. IABS Proceedings. 1993, 109-114 Porter M.E. 1980. Competitive Strategy, New York: Free Press, 1980. Rajagopalan N., Spreitzer Gretchen M. 1996. Towards a Theory of Strategic Change: A Multi-lens Perspective and Integrative Framework. Academy of Management Review. 22, 1 Roe, Mark J., J. Mark Ramseyer, & Roberta Romano. 1993. Some differences in corporate structure in Germany, Japan, and the United States. Yale Law Journal, 102(8): 1927-50. Rowley T.J. 1997. Moving beyond dyadic ties: A network theory of stakeholder influences. Academy of Management Review; 22: 887-910 Roberts, P.W., and Greenwood, R. 1997. Integrating Transaction Cost and Institutional Theories: Towards a Constrained-Efficiency Framework for Understanding Organizational Design Adoption. Academy of Management Review, 22, 2: 346-373 Savage, Grant T., Nix, Timothy W., Whitehead, Carlson J., Blair, John D., 1991. Strategies for assessing and managing organizational stakeholders. Academy of Management Executive, 5 2 Scott W.R.: “Technology and Structure: An Organizational Level Prespective” in PS Goodman and LS Sproull (eds.), Technology and Organizations. Jossey-Bass Inc. , San Francisco, CA, 1990 222 Schmidt, R.H. 1997. Corporate Governance: The Role of Other Constituencies. In Pezard & Thiveaud (ed.) Workable Corporate Governance:Cross-Border Perspectives, Montchrestien, Paris Sethi, Prakash S. 1979. A conceptual framework for environmental analysis of social issues and evaluation of business response patterns. Academy of Management Review. 4 Scott W.R. 1990. Technology and Structure: An Organizational Level Prespective in PS Goodman and LS Sproull (eds.), Technology and Organizations. Jossey-Bass Inc. , San Francisco, CA Shankman, Neil A. 1999. Reframing the debate between agency and stakeholder theories of the firm. Journal of Business Ethics; 19, 4 Smith, Michael P. 1996. Shareholder Activism by Institutional Investors: Evidence from CalPERS. Journal of Finance, 51, 1: 227-252 Stanwick P.A. & Stanwick S.D. 1998. The relationship between corporate social performance, and organizational size, financial performance and environmental performance: An empirical examination. Journal of Business Ethics; 17; 195-204 Sternberg, Elaine. 1999. The Stakeholder Concept: A Mistaken Doctrine. London: Foundation for Business Responsibilities. Suchman Mark C. 1995. Managing Legitimacy: Strategic and Institutional Approaches. Academy of Management Review, 20, 3 Takahashi Takuma. 2002. Who declares the demise of the Japanese main banking system? Journal of International Banking Regulation; 4, 2 Thompson James D. 1967. Organizations in Action: Social Science Bases of Administrative Theory. New York. McGraw-Hill. Trevino, Linda Klebe; Weaver, Gary H. 1999. The Stakeholder Research Tradition: Converging Theorists -Not Convergent Theory. Academy of Management Review, 24, 2, p222 Trist , E. 1981. The Socio-Technical Perspective. in Perspectives on Organizational Design and Behavior. 19-75. New York: John Wiley and Sons Tylecote A.; Cho, Yong Doo; Zhang W. 1998. National Technological Styles Explained In Terms Of Stakeholding Patterns, Enfranchisement and Cultural Differences: Britain And Japan. Technology Analysis & Strategic Management; 10, 4 Weaver G.R & Trevino L.K. 1994. Normative and Empirical Business Ethics: Separation, marriage of Convenience, or Marriage of Necessity? Business Ethics Quarterly. 4; 129-143 Welcomer Stephanie A. 2001. Firm-stakeholder networks. Business and Society, 41, 2, 251 Weyer, Martin Vander. 1996. In an ideal world... Management Today. London: pg. 34, 5 pgs Wheeler D. and Sillanpaa M. 1997. The Stakeholder Corporation: A Blueprint for Maximizing Stakeholder Value. Pitman, London. Wolfe, R. and Putler, D. 2002. How Tight Are the Ties that Bind Stakeholder Groups? Organizational Science. 13: 64-80. Wood, D. 1991. Corporate Social Performance Revisited. Academy of Management Review, 16, 4, 691 Yermack D. 1996. Higher Market Evaluation of Companies with A Small Board Of Directors. Journal of Financial Economics, 40, 185-211 223 Yoshikawa, Toru, and Phan, Phillip H. 2001. Alternative corporate governance systems in Japanese firms: Implications for a Shift to Stockholder Centered Corporate Governance. Asia Pacific Journal of Management Zellner W. & Berstein A. 1987. How GM and the UAW Kept from Butting Heads. Business Week, October 26; 32 Zingales, L. 2000. In search of new foundations. Journal of Finance. 55,4, 1623-1653 224 Figure 1: The Contingent Stakeholder Model Life Cycle Stage of the Firm Firm Size Type of Industry Stakeholder Orientation Institutional Structure Industry Life Cycle Stage Culture (including regulations) Figure 2: Diagnostic Typology of Organizational Stakeholders1 1 Savage et al. Academy of Management Executive. Vol. 5 Issue 2 (1991) 225 ASAC 2005 Toronto, Ontario Sujit Sur (student) John Molson School of Business Concordia University GOVERNANCE BEYOND AGENCY THEORY: THE TALE OF DUAL CLASS FIRMS This paper looks beyond agency theory and utilizes multiple theoretical lenses to explore governance in firms adopting dual class share structure. A model is offered that proposes dual class share structure as a contingent governance mechanism. Governance is asserted to operate at individual, industry and national levels under different rationales. This paper attempts to investigate the theoretical underpinnings for the reasons why firms frequently adopt the dual class equity structure, i.e. a capital (or ownership) structure based on the issuance of shares with differential voting rights (DeAngelo et al. 1985). Such an ownership structure is justified by some researchers as a defensive structure adopted by the directors and managers of a firm to prevent hostile acquisitions and takeovers by corporate raiders (Grossman and Hart, 1988; Harris and Raviv, 1988), and as a means of wealth enhancement of all shareholders (Alchain & Demsetz, 1972; Gromb 1993; Zingales, 1995). Many researchers however, assert that dual class shares are a mechanism by which insiders expropriate value from the minority shareholders, extract private benefits of control and ensure managerial entrenchment (Jarrell and Poulson, 1988; Gompers et al., 2004). The rationale for both of these assertions, as for most corporate governance mechanisms, is based on the principles of agency theory (Shleifer & Vishny, 1997; Becht et al., 2003). However, numerous studies have pointed out inconsistencies in the dual class ownership mechanism that cannot be explained by agency theory (Lease, 1984; DeAngelo and DeAngelo, 2000; Faccio and Lang, 2002). Thus the governance models based solely on the tenets of agency theory based does not seem to suffice in explaining many of the relevant issues of the dual class firms. Understanding the governance issues in ownership structures that enable controlling shareholders voting/control rights significantly in excess of their cash flow rights (and thus the issues of control within ownership), is considered to be of importance as such structures are common in firms globally (La Porta et al, 1999) and are the predominant structure outside of US and UK (Gompers et al., 2004). However, most researchers continue to focus on separation of ownership and control for studying governance mechanisms issues of these firms. This paper addresses this gap in literature and looks beyond agency theory to seek alternative rationales specifically for the adoption of dual class ownership structure by firms, with a focus on exploring the theoretical underpinnings of rationales for control issues within ownership. We argue that firms adopt dual class ownership as a contingent governance mechanism, and this phenomenon can be best understood only when examined by a multi-theoretic lens. We assert that the contingent governance mechanism operates at multiple levels and at each level differing attributes are salient. The propositions explore how and why firms adopt dual class structure as a governing mechanism. Such a framework for understanding control issues within ownership is opined to be generalizable beyond dual class structure and thus has strong implications for other corporate governance mechanisms as well. This study thus furthers understanding of the issues of dual class shares in three distinct ways. Firstly, it advances a tentative rationale for explaining the contradictory findings in literature on dual class shares by the researchers. Secondly, it builds a model explaining how and why firms adopt dual class capital structure as a contingent governance mechanism. And finally, it offers an alternative model for understanding corporate governance beyond agency theory, especially in case of firms with concentrated ownership. 226 Dual Class Equity: Defining the Domain Dual class firms offer researchers a unique opportunity to separate the effects of voting (control) incentives from cash flow incentives (Gompers et al., 2004), and thus enables a fine-grained examination of principal-principal governance issues. In one of the oft cited studies on dual class shares, DeAngelo and DeAngelo (1985) study managerial stock holdings in dual class firms and find that corporate officers and their families hold voting stock in excess of common stock. They also find significant family involvement, and document explicit acquisition premiums paid for superior voting shares. This premium, examined solely in terms of agency theory, is assessed as private benefit of control. Subsequent studies report similar findings (Bergstrom and Rydqvist, 1990; Becht et al., 2003; Gompers et al., 2004). Implicit in such a view is the assumption that the owner-managers or controlling shareholders as the ‘other principal’, because of their additional managerial function or voting control, will usurp ownership ‘rights’ of the non-controlling shareholders. This private ‘benefit of control’ is reflected as the premia paid for the voting shares. Even when viewed in terms of such Principal-Principal Agency (PPA) – a term first used only in 2000, numerous studies report confounding and contradictory findings that can not explained by agency theory including the PPA perspective. Prior to exploring these findings in detail, we first revisit agency theory and its underlying premises. Agency Theory Revisited Simply put, agency theory is a principal-agent control conflict, and corporate governance is the mechanism by which owners minimize agency costs Daily et al. (2003b). Wide acceptance of this theory as per them is thus: “First, it is an extremely simple theory, in which large corporations are reduced to two participants - managers and shareholders, and the interests of each are assumed to be both clear and consistent. Second, the notion of humans as self-interested and unwilling to sacrifice personal interests for interests of others is both age old and widespread”. Once thus entrenched in the very definition of corporate governance, agency theory became the de facto underlying principle of all corporate governance mechanisms. Shleifer and Vishny (1997) categorically state that ‘‘Corporate governance deals with the agency problem: the separation of management and finance, or – in more standard terminology – of ownership and control”. They then go on to illustrate this point by stating: “An entrepreneur, or a manger, raises funds from investors either to put them to productive use or to cash out his holdings in the firm. The financers need the manager’s specialized human capital to generate returns on their funds. The manager needs the financer’s funds, since either he does not have enough capital of his own to invest or else wants to cash out his holdings.” The dominant model of the 90s perhaps was an entrepreneurial manager seeking ‘seed capital’, and this might explain equating entrepreneur with manager in the above illustration, however, in such a case of an existing owner raising capital, the governance issue at best will be that of control between principal owners and not of “ownership and control” between principal and agent – the basic tenet of agency theory. Most firms prior to the 90s might have had limited managerial ownership, thus the governance problems might have been seen as issues of ‘ownership and control’ but using this ‘agency’ lens to examine the issues of control within ownership is puzzling. This inconsistency might not be of major consequence if a typical firm reflects the Berle-Means’ basic premise of diffused ownership, whereby the majority of the governance issues might still concern the ‘principal-agent’ problem. However, of late there have been recent challenges to this Berle-Means perspective (La Porta et al., 1999; Faccio et al., 2002), who argue 227 that the Berle-Means firm with widely dispersed ownership is the exception rather than the rule for industrialized and industrializing economies, US and UK being the notable exceptions. However, corporate governance in all firms worldwide were still viewed in terms of ‘principal–agent’ agency issues, and agency theory based on the basic premise of separation of ‘ownership and control’ was used to explain control issues within ownership. It was only in 2000, that Dharwadkar et al (2000) proposed the term “principal-principal agency” (PPA) problem to define problem centered on conflict between two types of principals in the context of post-privatization emerging economies. Faccio et al (2001) thereafter utilized PPA problem to address expropriation of minority shareholders’ value by controlling shareholders. Defined in terms of PPA, agency theory thus might have some explanatory powers for governance issues of dual class firms; however numerous studies have found inconsistent and contradictory findings even from this perspective. In the following section we examine the accepted justifications for adoption of dual class share structure, and contrast the findings that support agency theory (inclusive of PPA problems), to those that stand in contradiction. The Justifications and the Inherent Contradictions As expropriation of minority shareholder returns: In accordance with Jensen and Meckling (1976): With increase in control, the private benefit of control rises proportionately. It also lead to hidden action (moral hazard) by a controlling shareholder with a low capital investment, as he has incentives to make decisions which are favorable to himself but which will lower the value of the firm and the returns of minority shareholders. Viewed only from this perspective, firms with increasing concentrated ownership (like those with dual class capital structure) will increasingly expropriate non-voting shareholders returns, freeze out the minority, or sell control. Some researchers do find supporting evidence for such behavior by firms with dual class share structure (Bebchuk, Kraakman, and Triantis, 2000; Amoako-Adu and Smith, 2001). However, Faccio and Lang (2002) find no extra benefit for controlling shareholders, while Cornett and Vetsuypens (1989) find that outside shareholder value actually increases when firms adopt dual class share structure. Dimitrov and Jain (2004) also find shareholder value enhancement when firms recapitalize to dual class. Bergstrom and Rydqvist (1990) find that large shareholders own much more equity than required for control, inconsistent with the expropriation hypothesis. Thus overall, there is mixed evidence on dual class structure as an expropriation mechanism. As firm value enhancing: Proponents of concentrated ownership (Jensen and Meckling, 1976; Leland and Pyle (1977) also utilize agency theory to state that the value of the firm will increase with the controlling shareholder’s ownership fraction of the firm. As per this perspective, benefits from holding a large equity fraction should induce the controlling shareholder to invest in more shares than required merely for control, and thus the controlling shareholder should not be interested in creating dual-class shares for the purpose of holding control with a minimum of equity. However, owners in dual class firms continue to hold sizable equity in these firms. The empirical literature is also quite mixed regarding the effect of dual class share structures on firm performance. Proponents of agency theory are divided whether dual class structure leads to increase or decrease in firm valuation. Agency theory might explain Harris and Raviv (1988) assertion that issuing dual class of shares will increase firm valuation, and Claessens et al. (2002) finding that firm value increases with the cash flow ownership of the largest shareholder but decreases when the voting ownership exceeds the cash flow ownership, or even Lins (2003) finding of decreasing firm value with increasing voting ownership to cash flow ownership. However, agency theory cannot explain Morck et al. (1988) finding of non linear relationship between voting rights and market valuation, nor Gompers et al. (2004) finding that firm value (as measured by Tobin’s Q) increases in cash flow ownership and decreases in voting ownership in a non linear fashion. Furthermore, Scott and Zutter (2003) find no differences in post-IPO revenue growth 228 between dual and single-class firms, however stocks of dual class firms traded at lower valuations relative to fundamentals than single class firms. Dalton et al. (2003) also find no support for the agency theory prescribed relationship between equity ownership and firm performance. Neither inside nor outside equity ownership was found to be related to firm financial performance in that study. As defense against takeover: Dual class recapitalization during 1984-1986 in the US, garnered much interest from researchers. Most researchers utilizing agency theory principles argue that firms use concentrated ownership afforded by dual class structure as a defense against takeovers and to ward off corporate raiders (Alchian & Demsetz, 1972; DeAngelo & Rice, 1983; Easterbrook & Fischel, 1983; Shultz, 1988). Smart and Zutter (2003) also find that single-class firms become acquisition targets within five years of the IPO more frequently than dual-class firms do. However, Shum (1995) finds that firms recapitalized to dual class structure even when they were not targets of takeover attempts. Becht et al. (2003: 50-51) finds very few and decreasingly rarity of takeovers in US (highest at 1.5% in 1986 and steeply decreasing thereafter), however Gompers et al. (2004) find stability (rather a slight increase) in the number of firms with dual class structure in the US from 1970 to 2001. Thus adoption of dual class structure is not merely a takeover defense mechanism. Furthermore, there is lesser consensus if takeover defense mechanisms are beneficial or detrimental for firms. As per the dictates of agency theory, hostile takeover bids result in significant wealth gain for shareholders and dismissal of inefficient managers, thus anti takeover defenses should result in wealth loss of ‘outside’ shareholders (Grossman & Hart, 1988; Jarrell and Poulson, 1988). Partch (1987) studied the wealth effects of 44 recapitalizations from 1962 through 1984, and concluded that the creation of dual classes of common stock with unequal voting rights does not harm shareholders. Lauterbach (1991) also finds shareholders are not harmed by adoption of anti-takeover amendments. Jarrell and Poulson (1988) however find significant wealth loss of ‘outside’ shareholders for recapitalizing firms, especially during 1986-1988. Pound (1988) finds takeover attempts as value reducing for firms and leading to further entrenchment of management. However, Kabir et al., (1997) find mixed results in the case of Dutch firms. Harris & Raviv (1988), Gromb (1993) and Zingales (1995) contest that some deviation from one-share-one-vote structure will in fact lead to wealth gain in cases of takeovers. Alchain & Demsetz (1972) suggest non-voting shares will actually benefit outside shareholders as it deters takeovers that do not enhance wealth and reduces managerial allocation of resources towards anti-takeover efforts. Stulz (1988) finds that management actually forces acquirers to pay higher premiums to gain control when management's stake is higher, and this sometimes leads to an increase in the target firm's ex ante value, thus resulting in gains for outside shareholders. Burkart et al. (1998) further propose that dual-class capitalization helps majority holders achieve higher bid prices for the firm. Daines (2001) finds that anti-takeover protection is used to protect management when takeovers are most likely and management performance is most transparent, but do not find any evidence of managerial desire to protect unduly high private benefits. Becht et al. (2003) assert that that the main beneficiaries of hostile takeovers are target company shareholders and the main losers are acquiring company shareholders and target management in case of dual class firms. Thus, there is no consensus if dual class structure is a takeover defense mechanism, and is such a structure harmful or beneficial for the firm if examined solely in terms of agency theory. As Private Benefit of Control: As per traditional financial theory, there should be no price differential between the voting and non voting class of shares if they have the same pecuniary payoff (Levy, 1983). Jensen and Meckling (1976) explain this paradox as the controlling ‘management’ may increase its welfare by receiving non-pecuniary benefits that are not enjoyed by outside shareholders and thus the voting shares will be worth more. However some European firms have had negative differential between the voting and non-voting shares in the 90s. This discounting of voting shares cannot be explained in terms of agency theory. Most researchers 229 however, address only acquisition premium paid for voting stock, and attribute it as the ‘private benefit of control’ (DeAngelo et al., 1985; Rydqvist, 1992; Zingales, 1995; Shum et al., 1995; Nicodano, 1998; Gompers et al., 2004), and this ‘benefit’ is generally viewed as a means of expropriating rights or as a takeover defense mechanism by the proponents of agency theory. In a related study, Dyck and Zingales (2004) find that higher private benefits of control are associated with less developed capital markets and more concentrated ownership. Thus dual class firms, as per agency theory should have uniformly high premia because of the high private benefits of control. However, Faccio et al. (2002), Dimitrov et al. (2004) find no support for private benefit of control for dual class firms. During 1990-1993, Norway and Denmark in fact had negative valuation for the voting shares, while other Scandinavian countries have declining premia (Becht et al., 2003: 57). Neumann (2003) addresses the discounts specifically and asserts that the price differential in stock classes is more likely to reflect investors' liquidity risks than benefits of control in the case of Danish firms. In any case, fact remains that there are significant variations in the premia (discount) between countries1 that cannot be fully explained by agency theory. Nanova (2003) finds that legal environment variables explain 75% of the cross-country variation in the value of control ‘benefits’; however, Becht et al. (2003) assert that time series data shows significant variations within countries2 that cannot be fully explained in terms of only legal environment. They also caution that the measures are still incomplete and call for further research for understanding the valuation of premia (discount). In fact Gompers et al. (2004) conclude by speculating that the most plausible explanation for some firms adopting dual-class structures is when their original owners are reluctant to cede control – and perhaps not to extract private benefit of control or any of the other such agency theory prescribed costs. As means to managerial entrenchment: Most researchers assert that the dual class structure results in managerial entrenchment in firms (Morck, Shleifer and Vishny, 1988; Gompers et al., 2004). However, Dimitrov et al. (2004) finds no evidence for entrenchment. DeAngelo & DeAngelo (2000) state managerial replacement in the Times Mirror dual class case study. Moreover, many researchers find managerial entrenchment even in widely dispersed ownership and even block owned single class firms (Grossman and Hart, 1980; Black, 1990; Becht et al., 2003). Furthermore, Scott and Zutter (2003) find factors other than management entrenchment influence ownership structure, most notably risk and growth prospects. Also there is no consensus if entrenched management benefits or harms a firm. While most researchers imply entrenchment is detrimental for firms (Coles, 2001), DeAngelo and DeAngelo (1985) argue that dual class structures can encourage managers to invest in firm-specific human capital adding value to the firm. Thus agency theory cannot explain effect of entrenchment on dual class firms. As a ‘principal-agent’ governance mechanism. As per the dictates of agency theory, concentrated ownership should lead to better governance. Because of the majority of voting rights are controlled by insiders, dual class firms should have minimal or nil ‘principal – agent’ agency problems and monitoring costs. However, DeAngelo & DeAngelo (2000) study of the Times Mirror (TM) Company is an illustrative example of agency theory’s inability to explain the instance of poor governance in such a dual class firm. Four aspects of the study stand out in stark contradiction to agency theory – (1) Interests of the family (controlling owner) and the outside 1 Canada, 8–13%; France, mean 1986–1996 51.4%; Germany, mean 88–97 26.3% in 2000 50%; Israel, 45.5% ; Italy 82% ; Korea, 10% ; Norway, −3.2–6.4% ; Sweden, 12% ; Switzerland, 18% ; UK, 13.3% ; USA, 5.4% , mean 1984–90 10.5%, median 3% (Becht et al., 2003: 57) 2 While premia have been rising from 20% in mid-1998 to 54% in December 1999 in Germany, in Finland they have dropped from 100% in the 1980s to less than 5% in 2003. Similarly in Sweden premia have declined from 12% in the late 1980s to less than 1% today, 129 and in Denmark from 30% to 2%. In Norway the differential was actually negative in 1990–1993, but has risen to 6.4% in 1997 (Becht et al., 2003: 58) 230 manager were never divergent or misaligned; (2) First reaction to the sale was positive, and then negative when full information was available, (3) Management was replaced by controlling shareholder even when management met their interests, and (4) Though eventually the Chandler family’s pressure led to improved performance of TM, the path to this outcome was slow and circuitous, thus the disciplinary forces were weaker than agency theory typically implies. Thus agency theory prescribed ‘principal-agent’ governance mechanism does not work as stipulated. However, Brown and Caylor (2004) find dual class firms score relatively high on their 54 variable ‘Gov-score’. Also, Gompers et al. (2004) find that dual class firms score significantly lower than single class firms as per the 24 variable ‘Governance Rating’ developed by Gompers et al. (2003), signifying low managerial powers and thus higher governance. Both these scales were developed to assess the relationship between corporate governance and firm performance. Though neither of these scales has been rigorously tested or is widely accepted as a reliable measure of corporate governance as yet, these indicate dual class share structure seems to work as a corporate governance mechanism, however not as prescribed by agency theory. This might well be true of other concentrated ownership structures also. Morck et al. (2000) provide evidence of the detrimental effects of this particular type of ownership (dubbed the ‘Canadian Disease’) on firm performance. However, Chirinko et al (2004) find that the Canadian concentrated share ownership is associated with fewer governance problems. Thus we argue that agency theory, even defined in terms of PPA, is unable to satisfactorily explain some of the fundamental issues involved in the adoption of dual class share structure by firms, and perhaps even with the mechanisms of corporate governance in other firms with concentrated or large ownership structures per se. Recently, some management researchers have also similarly moved beyond the one dimensional agency theory lens for examining the issues of corporate governance (Aguilera et al., 2003; Sundaramurthy et al., 2003; Lee and O’Neill, 2003; Daily, Dalton and Canella, 2003; Daily, Dalton and Rajagopalan, 2003; Morrisson, 2004; Wang, 2004). Recent Alternative Theoretical Lenses Aguilera and Jackson (2003) argue that agency theory fails to sufficiently explore how corporate governance is shaped by its institutional embeddedness. They propose a firm level governance model consisting of the domains of capital, labor and management, and define corporate governance as the relationship between these stakeholder groups in the process of decision making and control over firm resources. National institutional structures are seen as influencing the range of effects but not determining outcomes within organizations. Aguilera et al. (2003) thus link agency and institutional theory to assert that multiple institutions exert interdependent effects on firm-level governance outcomes. We argue that this view overlooks the important aspects of individual and industry level effects on the governance mechanism. Also, in the context of dual class capital structure specifically, this model considers the stakeholder group of ‘capital’ to be homogeneous within the domain, and does not differentiate between minority and majority shareholders. Thus this model, though considered to be more robust and predictive of corporate governance mechanisms than one based solely on agency theory, is still considered to be unsatisfactory in explaining the relevant issues of governance, especially in case of firms with dual class equity capital structure. Sundaramurthy et al. (2003) proposes an alternative paradox framework, using agency and stewardship theories to explain governance mechanisms at the individual level. They thus assert governance as a ‘control and collaborate’ issue. Lee et al. (2003) similarly look at agency – stewardship theories to explain their findings of differences in governance mechanisms between US and Japan. They hypothesize that agency theory explains the relationship between ownership structure and long-term orientation in United States firms, while stewardship theory best explains the Japanese context. They go on to assert that ownership and 231 strategic decisions are embedded in cultural and institutional processes and path dependencies. Such a paradox framework integrated with institutional theory perspective might well become central in future studies of corporate governance mechanism, but we argue that the industry level institutional pressures on a firm are not catered for in such a viewpoint. As Coles et al. (2001) assert, industry performance is a strong and significant driver in the governance and performance relationship. Also, especially in respect of dual class share firms, the governance issue of control between owners cannot be specifically addressed by the paradox framework alone. Thus even these new models do not suffice for explaining the governance mechanism in case of dual class firms. This paper proposes a contingent governance model to address this gap in literature and explores the dual class capital structure, building on the above mentioned multi-theoretical lenses at individual, industry and national levels, in an attempt to understand the underlying mechanism that might better explain the contradictory findings and to seek a rationale for adoption of dual class equity structure in firms worldwide. The Contingent Governance Model We start with summarizing the common findings in the studies on dual class share firms. The firms that generally adopt dual class structure are found to be mostly: • • • Family or founder owned Concentrated in media related and other industries that avail benefit of control Have larger book value to size ratios and are highly leveraged (high debt to asset ratios) • Differ significantly across countries in the premia (discount) between the valuation of the different classes of shares We build our theoretical development around these findings and attempt to explain these trends by asserting that dual class share structure is a contingent governance mechanism adopted by firms to safeguard against ‘outside’ influences, and this mechanism is better understood by utilizing multi-theoretical lenses instead of the mono-theoretic agency lens. Explanations for dual class share structure (and their governance implications) are manifested at several levels, thus we argue that the dual class structure adoption by a firm is best explained by: (i) Psychological ownership and paradox framework at the individual level, (ii) Resource dependency and institutional theory at the industry level, and (iii) Institutional and resource based view (RBV) at the national level. The effects on a specific level do not preclude similar effects at the other levels, but are merely indicative of the greater salience of that effect on that level, e.g. RBV is manifested at the firm level as well as at the national level, however, its effect is most salient at the national level. Likewise, control-collaborate at individual level is opined to be closely related to individualism-collectivism at national level; however, within a given country, control-collaborate behavior will better explain variation between individual firms. Our model is thus proposed as a refinement of Lee et al (2003) assertion that agency theory reflects U.S. firms adequately, while stewardship theory represents Japanese firms adequately – our model enables differentiating between individual firm behavior within a specific national culture. Also, in keeping with contingency theory (Galbraith, 1973), we further assert that the interaction between these levels will determine the adoption, composition of the capital structure (percentage of voting to 232 non-voting shares, premium/discount between these shares, extant of voting power) and utilization of control by a dual class firm, and thus is unique to each firm. Such an undertaking is also in keeping with Daily et al’s (2003a) call for a “multi-theoretic approach to corporate governance as essential for recognizing the many mechanisms and structures that might reasonably enhance organizational functioning”. The premises for this model are developed further in the following exploratory propositions section. Propositions Individual level: We argue that the tenets of psychological ownership (Pierce et al., 2001) better explain the governance mechanism of dual class firms, in view of the fact that most dual class firms are predominantly family or founder owned and/or managed. This theory asserts that the degree of control engenders feeling of ownership towards the firm. Such owners do not treat ‘their’ firm as an economically rational ‘homoeconomicus’ would do as per the tenets of agency theory. This might be the underlying cause for Becht et al. (2003) assertion that family-owned firms go public only if they could retain control through mechanisms like dual-class share structures, as they might be reluctant to go public if they risk losing control in the process. Hart (1988) argues that this would benefit the firm and the exchange, and is also unlikely to hurt minority shareholders. This is supported by the findings of Taylor & Whittred (1998) and Holmén & Högfeldt (2004) studying the IPO valuation of firms with different classes of shares. Smart and Zutter (2003) also find that dual class IPOs experience less underpricing as compared to single class firms. Psychological ownership might also explain the reason why dual class firms are uniformly so anti-takeover oriented (despite there being not many takeover attempts) that such firms are predominantly represented in takeover defense literature (Rosenbaum, 1990 - 2002). Morck, Shleifer, and Vishny (1988) also offer evidence that firms in which one of the top two officers is a member of the founding family are less likely to be acquired in a hostile tender offer than firms in which the top two officers are unrelated to the founder. This also explains Stulz (1988) finding that high equity holding management's preference for control and consequent refusal to tender its shares, forces acquirers to pay higher premiums to gain control when management's stake is higher, and sometimes leads to an increase in the target firm's ex ante value. Thus we propose: Proposition 1a: Dual class share structure will predominantly be adopted by founder or family controlled/managed firms Proposition 1b: Dual class firms will have higher representation of family members in the management team and on their board of directors Thus control is to maintain ownership, not with explicit intentions of either expropriating minority shareholders rights, or reducing monitoring or agency costs, or even as a safeguard against takeover attempts; but more as indicative of the owner’s identification and attachment with the firm. Shum et al. (1995) found that 40% of the US firms (in their sample of 69 stock recapitalizations to dual class) publicly stated that they created the dual class structure to retain voting control even when none of these firms were targets of takeover attempts. Anecdotal evidence also shows that most family firms are named after the founding family (e.g. Bombardier, Molson, Coors, Reitmans) and thus the family reputations are closely associated with the firm’s image. Donoher’s (2004) study also indirectly supports this assertion, as he found that contrary to agency theoretical predictions, firms with high levels of inside equity ownership and secured indebtedness file for bankruptcy in poorer financial condition than peer firms with low levels of these variables; and in contrast, firms with high levels of outside equity ownership and short-term indebtedness file when in relatively better financial condition. Aghion and Bolton (1992) consider a situation where ownership is concentrated and argue that family-owned firms want to limit control by outside 233 investors because they value the option of being able to pursue actions in the future which may not be profit maximizing. They may value family control so much that they may want to turn down acquisition bids even if they are worth more than the net present value of the current business. Or, they may prefer to keep the business small and under family control even if it is more profitable to expand the business. It will be logical to suppose that in such cases, the owning family will play an active part in managing and governing the firm and will attempt to minimize outside influence. Nelson (2003) finds support for this argument in her empirical study of founder firms at IPO stage. Thus we propose: Proposition 1c: Dual class firms will have smaller boards consisting of more insiders Proposition 1d: Management in dual class firms will exhibit higher commitment to the firm and thus will display lower turnover of management Such a ‘psychological ownership’ of the firm will also explain why dual class firms exhibit non-linear relationship between voting rights and performance (Morck et al., 2000; Gompers et al., 2004). Too little ‘psychological ownership’ might lead to “not my baby” sentiment and safeguarding of income/employment by the insiders and thus high risk aversion strategies (Ansoff, 1965); while too high an ‘ownership’ will lead to choosing a certain option over a risky one in keeping with prospect theory (Tversky and Kahneman, 1992). Higher ownership by insiders, the desire to maintain control and the consequently increasing lack of diversification by the insiders (unlike outside investors) will also lead to risk adverse strategies (Gompers et al., 2004). Such a firm behavior also explains Smart et al. (2003) finding of lower valuation of dual class firms, while there are no systematic differences in earnings vis-à-vis single class firms. Bhide (1993), Becht et al. (2003) and Neumann (2003) also assert that the national stock market liquidity effect the firm’s stock valuation. More liquid stocks are generally valued higher and are more frequently traded. Thus, in agreement with Gompers et al. (2004), we restate their assertion to propose: Proposition 1e: Dual class firms will be less likely to raise capital so as to avoid diluting control, invest less, grow slower, trade lesser volumes of shares, and thus will be valued lower. We further assert that due to the high degree of control in such firms, the values, norms and attitude of the owner will better indicate the outcome of the governing mechanisms and its effect on firm performance. Hence, the control-collaborate paradox framework might best explain the outcome of the utilization of control - agency theory for malfeasance and stewardship theory for benevolence, and we opine, somewhere between the two for most firms1. Concentrated control will result in a tendency towards self-reinforcing cycles. Due to such self reinforcing cycles, any change will have a more significant impact on the firm. We thus agree with Sundaramurthy et al. (2003) assertions regarding the control-collaborate paradox framework and merge their original proposals to state: Proposition 2a: Dual class firms with a history of high performance and a predominant emphasis on control or collaboration will experience reinforcing cycles that foster strategic persistence. 1 An illustrative example of the ‘psychological ownership’ agency-stewardship paradox is Conrad Black’s 2002 email to Hollinger executive Peter Atkinson: "There has not been an occasion for many months when I got on our plane without wondering whether it was really affordable. But I am not prepared to re-enact the French Revolutionary renunciation of the rights of nobility. We have to find a balance between an unfair taxation on the company and a reasonable treatment of the founder-builders-managers. We are proprietors, after all, beleaguered though we may be." (The Montreal Gazette, Friday December 31, 2004, page B 02). 234 Proposition 2b: In a low-performance context, dual class firms with a predominant emphasis on control or collaboration will experience reinforcing cycles that foster organizational decline. The degree of psychological ownership and its effect on the paradox framework might also suggest some possible outcomes in the context of dual class firms. Such a perspective will suggest that dual class firms will significantly alter in valuation and governance during or immediately after succession. This is argued as because a founder identifies more closely with the firm as compared to her descendents (unless the successor has previously been involved with the firm). This is supported by the findings of Smith et al. (1999) and Villalonga et al. (2004) that find higher valuation for founder led firms, and lower for descendent led ones. We can also expect that later generations will be groomed to run the family business and thus the descendants will undergo business education and be exposed to the mainstream corporate norms and values, and thus will act more like other ‘professional’ managers. In case the descendents show no interest in managing the firm, outsiders will be hired to manage the firm, who might lack the same degree of psychological ownership that the founder, or even the descendants, might exhibit. This is supported by Morck et al. (1988) finding that in older dual class firms, firm value is higher when the firm is run by a member of the founding family than when it is ran by an officer unrelated to the founder. Also, in cases of outsider management, change in management structure will be caused due to divergence from controlling shareholder’s desires, and thus there will be a drastic shift in direction by the firm at turnover, and the subsequent valuation based on the market’s reaction to the change. This assertion is in congruence with the DeAngelo & De Angelo (2000) study of the Times Mirror Company and the mechanism of governance in that case. Thus we propose: Proposition 3a: Dual class firms will drastically change in valuation as well as strategic direction with change in ownership/management as compared to single class firms. Propositions 3b: Older dual class firms will be more alike single class firms Industry level: As per the resource dependency perspective, firms do not merely respond to external constraints and control through compliance to their environmental demands, rather they undertake a variety of strategies to alter the situation confronting them to make compliance less necessary (Pfeffer & Salancik, 1978). We assert that firms adopt dual class equity structure in accordance to this perspective, as a means to minimize outside influence on its workings. We argue that firms that started up in capital intensive industries with relatively low operating cash flow /liquidity requirements (like most of the firms that have adopted dual class shares), were mostly well entrenched and established family firms. Once entrenched, these firms have lower needs for capital as most of the assets of these industries are lienable, these firms opted for the ‘cheaper’ capital of debt. This argument is supported by Anderson et al. (2003) finding that firms controlled by founding families pay lower interest rates on bank loans. Also, Gompers et al. (2004) also find dual class firms to be older and highly levered (though they interpret the leverage as a lack of equity raising capability of dual class firms), and have very high book value to size ratio. Smart and Zutter (2003) also find that dual class firms have high family involvement, high leverage and lower reliance on external equity. Pajuste (2003) also finds dual class firms to have low equity dependence. Thus we assert that equity dependence (or lack thereof) is the main reason for adopting or recapitalizing to dual class structure by firms. Thus we propose: Proposition4a: In industries with low equity dependence, i.e. high capital outlay (fixed assets) and low liquidity requirements, firms will tend to adopt dual class equity structure 235 Also, Demsetz and Lehn (1985) speculate that “amenity potential” (the non-financial benefits, such as fame and influence, obtained by controlling a newspaper or a television station) influence ownership structures in industries such as media, sports, and entertainment. For example, owners can enjoy ‘amenity potential’ from controlling the editorial privilege of a media company or affiliating with sports and entertainment celebrities. This term was thereafter referred to as non-pecuniary ‘private benefit of control’ (Grossman and Hart, 1988). DeAngelo and DeAngelo (1985) argue that the non-pecuniary private benefits of consumption may be high in media-related firms and hence may lead founders to establish a dual-class structure in order to preserve control. Gompers et al. (2004) also find support for such non-pecuniary ‘benefit of control’ amongst media and communication firms. Djankov et al. (2003) find that dispersed ownership firms are uncommon in these industries as widely held firms will have no controllers who would enjoy the amenity potential, and thus these will be ripe for takeover by others desiring this benefit. Smart and Zutter (2003) also find dual class firms are more prevalent in industries in which ‘private benefits’ are significant, and particularly non-pecuniary private benefit of control in most dual class media firms. Thus we state: Proposition 4b: In industries (like media-related) that have ‘amenity potential’, firms will tend to adopt dual class share structure to preserve control Zhang (2003) also finds that within media industries, dual class firms do not have a lower Tobin’s Q than single class firm while non- media industries dual class firms have a lower Tobin’s Q than single class firms. He also finds that in the media industries, dual class firms pay lesser dividends than single class firms, the dividend payouts increases as insider ownership of cash flow rights rises and declines as insider ownership of voting rights rises – while there is no such relation in the non-media industries. Thus we state: Proposition 4c: In industries (like media-related) that significantly benefit from control, dual class firms will not differ in valuation from single class firms We also argue that once adopted at the IPO stage, path dependency (Bebchuk and Roe, 2003) will cause firms to continue with their dual class equity structure, unless their liquidity requirements changes drastically. This is supported by Pajuste (2003) in a related study, where she finds that dual class firms with higher equity dependence also have higher probability of unification (restructuring into single class firm). Thus we propose: Proposition 4d: Once adopted, firms will maintain dual class equity structure unless there is a drastic change in their equity dependence National level: Numerous studies have linked national and cultural effects on institutional structures and governance mechanisms (Nenova, 2000; Coffee, 2001; Gedajlovic & Shapiro, 2002; Aguilera et al., 2003; Lee et al., 2003; Chang, 2003; Miguel et al., 2004) proposing how national culture, norms and values are reflected in the governance mechanism of the firms. The societal (media), institutional (stock market) and regulatory (trading and taxation laws) pressures also ensure that most firms within the context of a nation will adopt similar governance mechanisms, thus leading to convergence in the mechanism of governance being adopted by firms within that nation. Dual class firms also will perforce conform to these institutional pressures. However, in the case of dual class firms, we assert that the concentrated control (within its close knit ownership structure) acts like an organizational resource adding a unique value that is difficult for competitors to imitate (Barney, 1991), and this will enable dual class firms more commitment (or flexibility, depending in the concentrated owner’s strategic choice) in its strategy and thus might be a source of potential competitive advantage. Concentrated control is an asymmetric resource available only to 236 dual class firms, and thus is a rare, inimitable and non-substitutable. Support for this assertion can be Miller’s (2003) finding that many firms are able to turn asymmetries into sustainable capabilities, leveraging them across appropriate market opportunities (Miller assets that asymmetries are rare, inimitable and non-substitutable, although not connected to any engine of value creation, and, in fact, might often act as liabilities too). Thus we propose: Proposition 5a: Dual class firms will differ significantly from single class firms in their strategic orientation and/or commitment This strategic divergence and the information asymmetries might also better explain the national level variation in the premia (discount) in the value of the different class of shares, which essentially is the market’s expectation of the firm’s future performance. Support for such an assertion can be found in Amit et al. (1993) study that use resource based view to develop the notion of strategic assets of firms to obtain competitive advantage and organizational rent. They assert that resource market imperfections and asymmetry in discretionary managerial decisions about resource development and deployment can be a source of sustainable economic rent. This argument is also supported by Nenova (2000) finding that legal environment (and thus degree of market imperfections) explains most of the variation in premia. Dyck and Zingales (2004) finding of media (influencing information asymmetries) and regulations (influencing market imperfections) as the only significant governance mechanism also supports this argument. Further support can be Doidge (2003) finding that non US dual class firms, when cross list on the US exchange, experience a reduction in the voting premia. Building on these arguments, we state: Proposition 5b: In countries with high market imperfection and information uncertainties, dual class firms will exhibit greater variation in the premia (discount) between the types of shares Proposition 5c: Within a given country, dual class firms with higher information uncertainties will exhibit higher variation in the premia (discount) between the types of share Whether dual class firms will be able to consistently reap the benefits of this strategic asset of control is of course contingent on the fact if the chosen strategy was appropriate or not - the outcomes are ex post - we argue for the ex ante strategic asset of commitment (or flexibility) and the value of the perceived strategic advantage of control - and the corresponding market reaction based on information asymmetry and market imperfection, or lack thereof. Research and Managerial Implications This paper focused on the underlying issues of firms adopting a dual class equity capital structure; however our model and propositions are opined to be generalizable to other concentrated ownership structure. If such concentrated ownership is the norm, as La Porta et al. (1999) and Faccio et al. (2002) assert, this contingent governance model is considered to have implications beyond dual class firms. We also did not analyze the firm level effect of variation in stakeholder salience (Jawahar et al., 2001) and thus its effect on governance mechanisms. We speculate RBV and stakeholder salience at the firm level might better explain governance and performance relationship than agency theory. Thus, understanding the mechanism of governance in terms of multi-theoretic lens and at multiple levels has important research implications. Such an approach might be able to explain the inconsistencies found in previous studies that utilized the mono-theoretic agency lens only. This paper is considered to have important managerial implications too. Firstly, free of the constraining view of agency theory, managers might not be seen only as ‘agents’ who need to be ‘monitored’, ‘disciplined’ or ‘bribed’ with equity or other such incentives(which McGuire and Matta [2003] show anyways doesn’t translate into greater firm 237 performance). And secondly, understanding the importance of psychological ownership at the individual level will enable owners to devise better governance instruments and mechanisms designed to engender sense of ownership in the managers. Limitations This paper has viewed prior corporate governance studies purely from the agency perspective, which is in keeping with the commonly accepted view (Eisenhart, 1989). However, some researchers have approached agency costs in terms of transaction cost economics (TCE) also. We have not addressed the issues from the TCE perspective. This is accepted as a limitation of this paper, however examining corporate governance issues of dual class firms or other kinds of concentrated ownership, in terms of TCE might be a worthwhile focus of future studies. Another limitation is that we have not included the stultifying effect of embeddedness on the choice of strategy by dual class firms. What we term as ‘commitment to strategy’, might well be effect of embeddedness, i.e. the firm continues to ‘do what it has always done’ without suffering the disciplinary role of outside market or shareholder pressure to alter its strategy. We assert that the controlling owner-manager of dual class firms intends to work in the best interests of the firm because of the psychological ownership effects, and argue for Hendry (2002) assertion that managerial action that might eventually lead to lowering of firm value will be a case of ‘honest incompetence’- but definitely not a case of ‘agency’ problem against itself. References Anderson, Ronald C., Sattar A. Mansi, and David M. Reeb. 2003. Founding Family Ownership and the Agency Cost of Debt. Journal of Financial Economics.68, 2, 263-285 Aghion, P., Bolton P. 1987. Contracts as a Barrier to Entry. American Economic Review, 77. Aguilera Ruth V., Jackson Gregory. 2003. The Cross-National Diversity of Corporate Governance: Dimensions and Determinants. Academy Of Management Review, 28, 3, 447–465. Alchian A., Demsetz H. 1972. Production, information cost, and economic organization. American Economic Review, 62: 777-795 Amit Raphael, Schoemaker. 1993. Strategic Assets and Organizational Rent. Strategic Management Journal, 14, 1: 33-46 Amoako-Adu, Ben, Smith Brian F., 2001. Dual-Class Firms: Capitalization, Ownership Structure And Recapitalization Back Into Single Class, Journal Of Banking and Finance 25, 1083-1111. Bebchuk Lucian A., Roe Mark J. 1999. A Theory of Path Dependence in Corporate Governance and Ownership. Working paper. Columbia Law School. http://papers.ssrn.com/paper.taf?abstract_id=192414 Bebchuk Lucian Arye, Kraakman Reinier, Triantis George. 2000. Stock Pyramids, Cross-Ownership, and the Dual Class Equity: The Creation and Agency Costs of Separating Control from Cash Flow Rights. Concentrated Corporate Ownership, R. Morck, Ed., 445-460. Becht, M., Bolton, P., & Roell, A. 2003. Corporate Governance and Control. Handbook of the Economics of Finance, Edited by G.M. Constantinides, M. Harris and R. Stulz. Elsevier B.V. Bergstrom Clas, Rydqvist Kristian. 1990. Ownership of Equity in Dual-Class Firms. Journal Of Banking And Finance, 14, 255-269. Bhide, A.1993. The hidden costs of stock market liquidity. Journal of Financial Economics, 34: 31−51 Black, B.S. 1990. Shareholder passivity reexamined, Michigan Law Review 89:520. Boyd, B. K. 1994. Board Control And Ceo Compensation. Strategic Management Journal, 15 Brown L, Caylor M. 2004. Corporate Governance And Company Performance. Research Study Commissioned by the Institutional Shareholder Services 238 Burkart M., Gromb D., Panunzi F. 1998. Why Higher Takeover Premia Protect Minority Shareholders? Journal of Political Economy, 106; 172-204 Chang Sea Jin. 2003. Ownership Structure, Expropriation, and Performance of Group-Affiliated Companies in Korea. Academy Of Management Journal, 46, 2, 238–253. Chirinkoa Robert S., Schaller Huntley. 2004. A revealed preference approach to understanding corporate governance problems: Evidence from Canada. Journal of Financial Economics, 74 Claessens, Stijn, Djankov Simeon, Fan Joseph P.H., Lang Larry H.P. 2002. Disentangling the Incentive and Entrenchment Effects Of Large Shareholdings, Journal of Finance 57, 2741-2772. Coffee, John C. Jr. 2001. Do Norms Matter?: A Cross-Country Examination of the Private Benefits of Control. Working Paper No. 183. Columbia Law School Coles Jerilyn W, McWilliams Victoria B, Sen Ninaljan. 2001. An examination of the relationship of governance mechanisms to performance. Journal of Management, 27 (2001) 23–50 Cornett Marcia Millon, Vetsuypens Michael R. 1989. Voting Rights and Shareholder Wealth. Managerial and Decision Economics, 10, 3; 175 Dacin, M. T., Ventresca, M., Beal, B. D. 1999. The Embeddedness of Organizations: Dialogue And Directions. Journal of Management, 25: 317–356. Daily Catherine M., Dalton Dan R., Rajagopalan Nandini, 2003a. Governance through Ownership: Centuries of Practice, Decades Of Research. Academy Of Management Journal, 46, 2, 151–158. Daily Catherine M., Dalton Dan R., Cannella, Albert A. Jr. 2003b. Corporate Governance: Decades of Dialogue and Data. Academy of Management Review, 28, 3, 371-382 Daines R, Klausner M. 2001. Do IPO charters maximize firm value? Antitakeover protection in IPOs. Journal of Law Economics & Organization, 17, 1: 83 Dalton, D. R., Daily, C. M., Ellstrand, A. E., & Johnson, J. L. 1998. Meta-Analytic Reviews of Board Composition, Leadership Structure, and Financial Performance. Strategic Management Journal, 19 Dalton Dan R., Certo Trevis S., Roengpitya Rungpen. 2003. Meta-analysis of financial performance and equity: Fusion or confusion. Academy of Management Journal, 46, 1, 13-26 Davis J. H., Schoorman F.D., Donaldson L. 1997. Towards a stewardship theory of management. Academy of Management Review, 22, 20-47 DeAngelo H., Rice E. 1983. Anti-takeover Charter Amendments and Stockholder Wealth. Journal of Financial Economics, 11: 326-359 DeAngelo Harry, Deangelo Linda. 1985. Managerial Ownership of Voting Rights. A Study of Public Corporations with Dual Classes of Common Stock. Journal of Financial Economics 14, 33-69 DeAngelo Harry, DeAngelo Linda. 2000. Controlling Stockholders and the Disciplinary Role of Corporate Payout Policy: A Study of the Times Mirror Company. Journal of Financial Economics, 56 Djankov Simeon, McLeish Caralee, Nenova Tatiana, Shleifer Andrei. 2003. Who Owns the Media? Journal of Law and Economics, 46, 2, 341 Demsetz, H., Lehn, K., 1985. The structure of corporate ownership: causes and consequences. Journal of Political Economy 93, 1155–1177. Diamond, D.W.1989. Reputation Acquisition in Debt Markets. Journal of Political Economy, 97, 828 Diamond, D.W. 1991. Monitoring and Reputation: The Choice Between Bank Loans and Directly Placed Debt. Journal of Political Economy, 99, 689−721. Dimitrov, V., and Jain, Prem C. 2004. Recapitalization of One Class of Common Stock into Dual-class: Growth and Long-run Stock Returns. SSRN Working Paper series. http://ssrn.com/abstract=422080 Doidge Craig. 2003. US Cross-Listings and the Private Benefits of Control: Evidence from Dual-Class Firms. Journal of Financial Economics, 72, 3, 519-553 Donoher William J. 2004. To File or not to File? Systemic Incentives, Corporate Control, and the Bankruptcy Decision. Journal of Management. 30, 2, 239-262 239 Dyck, A., and L. Zingales. 2004. Private benefits of control: an international comparison. Journal of Finance 59, 2: 537-600 Easterbrook F., Fischel D. 1983. Voting in Corporate Law. Journal of Law and Economics, 26 Eisenhardt, K. M. 1989. Agency Theory: An Assessment and Review. Academy Of Management Review, 14: 57–74. Faccio M, Lang L., Young L. 2001. Dividends and Expropriation. American Economic Review, Faccio Mara, Lang Larry H.P. 2002. The Separation of Ownership and Control. An Analysis of Ultimate Ownership in Western European Corporations. Journal Of Financial Economics, 2002 Falkenstein, Eric G. 1996. Preferences for Stock Characteristics As revealed by Mutual Fund Portfolio Holdings, Journal of Finance 51, 111-135. Galbraith J.R. 1973. Designing Complex Organizations. Addison-Wesley, Reading, Ma. Gedajlovic Eric, Shapiro Daniel M.2002. Ownership Structure and Firm Profitability in Japan. Academy of Management Journal, 45, 2: 565-575 Gilson Ronald J., Gordon Jeffrey N. 2003. Controlling Controlling Shareholders. Social Science Research Network Electronic Paper Collection at: http://Papers.SSRN.com/Abstract=417181 Gompers P, Ishii J, Metrick A. 2003. Corporate Governance and Equity Prices. Quarterly Journal Of Economics Gompers Paul A., Ishii Joy, Metrick Andrew. 2004. Incentives vs. Control: An Analysis Of U.S. Dual-Class Companies. NBER Working Paper 10240, National Bureau Of Economic Research, Granovetter, M. 1985. Economic Action and Social Structure: The Problem Of Embeddedness. American Journal Of Sociology, 91: 481–510. Grossman S.J., Hart O.D. 1980. Takeover Bids, the Free Rider Problem, and the Theory of the Corporation. Bell Journal of Economics, 11, 42-69 Grossman, Sanford, Oliver Hart. 1988. One Share-One Vote and the Market for Corporate Control. Journal of Financial Economics, 20, 175-202. Harris Milton, Raviv Arthur. 1985. Corporate Control Contests and Capital Structure. Journal of Financial Economics, 20: 55-86 Hendry, J. 2002. The Principal's other Problems: Honest Incompetence and the Specification of Objectives. Academy Of Management Review, 27(1): 96-112 Holmén Martin, Högfeldt Peter. 2004. A Law and Finance Analysis of Initial Public Offerings. Journal of Financial Intermediation, 13, 324–358 Huberman, Gur. 2001. Familiarity breeds investment. Review of Financial Studies, 14, 659-680. Jawahar, I M; McLaughlin, Gary L. 2001. Toward a descriptive stakeholder theory: An organizational life cycle approach. Academy of Management Review; 26, 3 Jarrell Gregg A., Poulsen Annette B.1988. Dual-Class Recapitalizations as Antitakeover Mechanisms. The Recent Evidence. Journal Of Financial Economics, 20 129-152. Jensen M. C., Meckling W. H. 1976. Theory of the firm: Managerial agency costs and ownership structure. Journal of Financial Economics, 3, 305-360 Kabir Rezaul, Cantrijn Dolph, Jeunink Andreas. Takeover Defenses, Ownership Structure and Stock Returns In The Netherlands: An Empirical Analysis. Strategic Management Journal, 18:2 Kreps, D.M. 1990. Corporate culture and economic theory in J.E. Alt and K.A. Shepsle, eds., Perspectives on Positive Political Economy (Cambridge University Press) pp. 90–143. La Porta, R., F. Lopez-de-Silanes and A. Shleifer. 1999. Corporate ownership around the world. Journal of Finance 54:471−517. Lauterbach Beni, Malitz Ileen B., Vu Joseph. Takeover Threats, Antitakeover Amendments and Stock Price Reaction. Managerial and Decision Economics, 12, 6: 499-510Lee Peggy M., O’neill Hugh M. 2003. Ownership Structures and R&D Investments of U.S. and Japanese Firms: Agency and Stewardship Perspectives. Academy Of Management Journal, 46, 2, 212–225. 240 Lease, R.C., J.J. McConnell, W.H. Mikkelson. 1984. The market value of differential voting rights in closely held corporations. Journal of Business, 57: 443−467. Levy, H. 1983. Economic Evaluation of Voting Power of Common Stock. Journal of Finance, 38,1, 79-93 Lins, Karl V. 2003. Equity Ownership and Firm Value in Emerging Markets. Journal Of Financial And Quantitative Analysis, 38, 159-184. Lins, Karl V., 2003, Equity Ownership and Firm Value in Emerging Markets, Journal of Financial and Quantitative Analysis 38, 159-184. McGuire Jean, Matta Elie. 2003. CEO Stock Options: The Silent Dimension Of Ownership. Academy Of Management Journal, 46, 2 Morck Randall, Shleifer Andrei, Vishny Robert W. 1988. Management Ownership and Market Valuation. Journal of Financial Economics, 20: 293-315 Morck R, Stangeland DA, Yeung B, Page MP. 2000. Inherited Wealth, Corporate Control, and Economic Growth: The Canadian Disease. Concentrated Corporate Ownership, NBER, Univ. of Chicago Press Miguel Alberto De, Pindado Julio, La Torre Chabela De. Ownership Structure And Firm Value: New Evidence from Spain. Strategic Management Journal. 25: 1199–1207 Miller Danny. 2003. An asymmetry-based view of advantage: Towards an attainable sustainability. Strategic Management Journal; 24, 10; 961 Myers, S.C. 1977. Determinants of corporate borrowing. Journal of Financial Economics, 5 Myers, S.C., 2001. Capital structure. Journal of Economic Perspectives,15, 81–102. Nelson Teresa. 2003. The Persistence of Founder Influence: Management, Ownership, and Performance Effects at Initial Public Offering. Strategic Management Journal, 24 Nenova Tatiana. 2000. The Value of Corporate Votes and Control Benefits: A Cross-Country Analysis. Journal of Financial Economics, 68, 3, 325-351 Neumann Robert. 2003. Price Differentials Between Dual-class Stocks: Voting Premium or Liquidity Discount? European Financial Management, 9, 3, 315-332 Nicodano, G. 1998. Corporate groups, dual-class shares and the value of voting rights. Journal of Banking and Finance 22:1117−1137. Pajuste Anete. 2003. What drives the unification of multiple class shares? Working paper, Stockholm School of Economics, www.scholar.google.com Pfeffer J.; Salancik G. 1978. The External Control of Organizations: A Resource Dependence Perspective. New York. Harper & Row. Pierce Jon L., Kostova Tatiana, Dirks Kurt T., 2001. Towards a Theory of Psychological Ownership in Organizations. Academy of Management Review, 26, 2, 298-310 Pound John. Proxy Contests and the Efficiency of Shareholder Oversight. Journal of Financial Economics, 20 237-265. Rosenbaum, Virginia, Corporate Takeover Defenses, 1990, 1993, 1995, 1998, 2000, 2002. Investor Responsibility Research Center Inc, Washington, D.C. Seligman, J. 1986. Equal protection in shareholder voting rights : the one-share–one-vote controversy. George Washington Law Review 54:687. Shleifer, A., Vishny, R. W. 1997. A survey of corporate governance. Journal of Finance, 52 Shum, C., Davidson, W., Glascock, J., 1995. Voting rights and market reaction to dual-class common stock issues. Financial Review, 30, 275–287. Smart S.B., Zutter C.J. 2003. Control as a motivation for underpricing: A comparison of dual and single-class IPOs. Journal of Financial Economics, 69, 85–110 Smith, Brian F., and Ben Amoako-Adu. 1994. A comparative analysis of takeovers of single and dual class firms. The Financial Review 29(1). 241 Smith Brian F. Amoako-Adu Ben. 1999. Management succession and financial performance of family controlled firms. Journal of Corporate Finance, 5, 341–368 Stulz Rene M. 1988. Managerial Control of Voting Rights. Journal of Financial Economics, 20 Stulz, R. 1990. Managerial discretion and optimal financing policies. Journal of Financial Economics, 26, Sundaramurthy Chamu, Lewis Marianne. 2003. Control and Collaboration: Paradoxes Of Governance. Academy Of Management Review, 28, 3, 397-415. Taylor Stephen , Whittred Greg. 1998. Security design and the allocation of voting rights: Evidence from the Australian IPO market. Journal of Corporate Finance, 4, 107–131 Tversky A., Kahneman D. 1992. Advances in Prospect Theory: Cumulative Representation of Uncertainty. Journal of Risk and Uncertainty, 5, 297-323 Villalonga Belén, Amit Raphael. 2004. How Do Family Ownership, Management, and Control Affect Firm Value? NBER working paper Weber, M. 1978. Economy and Society. Berkeley: University Of California Press. Zingales, L. 1995. What determines the value of corporate votes? Quarterly Journal of Economics, 110 242 ASAC 2005 Toronto, Ontario Carol-Ann Tetrault Sirsly (student) John Molson School of Business Concordia University CORPORATE SOCIAL RESPONSIBILITY INITIATIVES: A FIRST-MOVER THEORY In assessing a strategic corporate social responsibility (CSR) initiative, I ask the question, does it provide the firm with a first-mover advantage (FMA)? Using the resource-based view, I construct a framework for predicting a FMA based on the CSR initiative possessing three strategic attributes: centrality, specificity and visibility. While marketers have long debated the first-mover advantage (FMA) in garnering market share, the translation into competitive advantage and superior profit performance has fuelled the debate in strategic management. As Lieberman and Montgomery argue the independent evolution of FMA and resource-based view when "taken separately, each suffers from serious deficiencies. We see a strong potential for synergy" (1998: 1112). While some might accept short-term costs as inevitable in meeting the firm's responsibility to society, corporate social responsibility (CSR) advocates assert it is possible for firm strategies to be aligned with societal objectives to generate benefits for the firm (Burke & Logsdon, 1996; Epstein & Roy, 2003; Griffin & Mahon, 1997; Waddock & Graves, 1997). This then brings us to the research questions this paper answers. Is there a strategic advantage to be a first-mover in adopting a CSR initiative? On the other hand, do early adopter or follower roles provide better strategic positions? This paper will explore these questions and propose a theory to identify which CSR initiatives are likely to provide a FMA. The three attributes used to determine the salience of a CSR initiative to a FMA are: (1) the centrality of the CSR initiative to the firm's mission, (2) the specificity of the benefits of the CSR initiative to the firm as opposed to the general good, and (3) the visibility of the CSR initiative to the firm's stakeholders. This theory's logic is inspired by Burke & Logsdon's (1996) examination of the pay off of CSR and contributes to CSR research by providing a framework for predicting a FMA for a strategic CSR initiative. The theoretical underpinnings of the argument lie with a resource-based view of FMA interpreted within stakeholder theory. I will briefly define and position each of the main elements of the argument, providing a review of the literature. I will then elaborate a theory of FMA of a CSR initiative, building a model based on the three key attributes of centrality, specificity and visibility. I will illustrate applications for management and strategy development. The implications for empirical and further research will also be discussed. Literature Review and Theory Development Resource-Based View The resource-based view's answer to why performance varies between firms contends that it is a firm's unique bundle of valuable, rare and costly to imitate resources and capabilities that provides competitive advantage (Barney, 1991, 2001a, 2001b; Hoopes, Madsen & Walker, 2003; Peteraf, 1993; Peteraf & Barney, 2003; Peteraf & Bergen, 2003; Wernerfelt, 1995). "A capability is firm-specific since it is embedded in the organization and its processes, while an ordinary resource is not" (Makadok, 2001: 388), with a further distinction lying in the ability to value and trade a resource, but not a capability (Hoopes et al., 2003; Teece, Pisano & Shuen, 1997). This unique bundle of resources and capabilities can also be constituted of the business processes, routines or activities of the firm (Porter, 1991), while "the ability of firms to pursue certain activities, routines or business processes may be limited by the resources and capabilities they control" (Ray, Barney & Muhanna, 2004: 35). The threat of imitation can be reduced by patent right protection and by either very expensive replication costs or when there is sufficient ambiguity as to what to imitate (Hoopes et al., 2003; Peteraf & Barney, 2003). The criteria of costly to replicate must also not ignore substitution as a tangible competitive threat (Peteraf & Bergen, 2003). Peteraf and Barney (2003) argue the resource-based theory addresses sustainable competitive advantage by creating greater net benefits and is a complement, not a substitute, to Porter's 5-forces (Porter, 1991) and game theory. They go on to suggest that resource-based theory can also be "useful for non-profit organizations and those with a stakeholder orientation" (Peteraf & Barney, 2003: 321), since value creation is segregated from its distribution. Russo and Fouts (1997: 536) welcome the resource-based view for CSR research, highlighting it "addresses the fit between what a firm has the ability to do and what it has the opportunity to do". Hoopes and colleagues (2003) relate the resource-based view to the growth, shakeout and maturity cycle of industry growth, where early entrants rely on innovation with a high degree of variability in the cost of resources. As the industry develops, early entrants face challenges from new entrants. Through further innovation and imitation survival is determined (Porter & van der Linde, 1995): "The combination of competitive advantage in the short term and dynamic capability in the long term determines whether a first-mover can sustain its superior position as an industry evolves" (Hoopes et al., 2003: 894). First-Mover Advantage Lieberman and Montgomery (1988: 41) define "first-mover advantages in terms of the ability of pioneering firms to earn positive economic profits (i.e. profits in excess of the cost of capital)" going on to attribute this pioneering opportunity to "some unique resources or foresight, or simply because of luck". In their reflection ten years later, they explicitly linked the FMA to the resource-based view of the firm (Lieberman & Montgomery, 1998), where a firm specific resource or capability which thrusts a firm into being a first-mover can thus afford it a competitive advantage (Hoopes et al., 2003; Lieberman & Montgomery, 1998; Teece et al., 1997). A first-mover is also most likely to have the capability to reduce costs as experience is acquired (Mueller, 1997), reinforcing this competitive advantage. Using game theory, a sequential-game structure where the slope of reaction functions slopes downwards, affords a FMA to a pre-emptive move by a leader (Gal-Or, 1985). However, not all first-movers achieve an advantage, as a cheaper cost to imitate, suggested to represent 65% of the cost to innovate, may favour free riding of followers over the pioneers (Kerin, Varadarajan & Peterson, 1992; Lieberman & Montgomery, 1988), particularly when the followers have alignable differences with the pioneers (Zang & Markman, 1998). Furthermore, later entrants may learn from and improve upon the first-mover positioning to gain a competitive advantage, thereby providing an alternative argument against first-movers (Kerin et al., 1992). One inherent premise in this argument is the assumption that all information is public, permitting rivals, as spectators, to assess first-mover outcomes and, "Each firm has an incentive to wait and see because adoption by a rival provides the same information as its own adoption" (Jensen, 2003: 244 98). Kerin and colleagues (1992: 48) argue, "Unless a firm has the expertise, resources and creativity necessary to exploit these opportunities, and forestalls or neutralizes the efforts of later entrants, being first to market will produce neither sustainable competitive advantages nor desired performance outcomes", echoing the resource-based view of capabilities. A strategy proposed for late entrants to redefine the market to benefit themselves at the expense of the pioneer (Shankar, Carpenter & Krishnamurthi, 1998) is also routed in a resource-based view of the innovative capabilities of the late-mover. Fundamental to FMA is the yardstick against which performance is compared. Utilizing a ten-year survival rate of industrial goods businesses, Robinson and Min (2002) compared 167 first-entrant market pioneers with 267 early followers to suggest the endurance of first-mover advantages, including the role in setting industry standards. While most first-mover studies have concentrated on products and markets, Nehrt (1996) used the timing and intensity of environmental investments in 50 chemical bleached paper pulp manufacturers to show a FMA on profitability. Stakeholder Theory Freeman's (1984) seminal "Strategic Management: A Stakeholder Approach" defined stakeholders to include "any group or individual who can affect or is affected by the achievement of the organization's objectives" (Freeman, 1984: 46). A narrower lens has been taken in suggesting legitimacy be required to be considered a stakeholder (Rowley, 1997), in segregating primary from secondary stakeholders based on their necessity to firm survival (Clarkson, 1995), and in a typology of stakeholder salience utilizing attributes of urgency, power and legitimacy (Mitchell, Agle & Wood, 1997). Stakeholders have also been identified as the involved stakeholder, who is known to the firm and who may influence organizational behaviour and the affected stakeholder, who does not have such an opportunity, other than by representation made by a witness, and whose claims can rarely be bounded (Vos, 2003). Stakeholder influence strategies have been categorized in terms of direct and indirect use or withholding of resources, building on resource dependence theory (Frooman, 1999), while firm responses to stakeholder pressures have been categorized as "commander, compromiser, subordinate, and solitarian" (Rowley, 1997: 888), using social network analysis. A stakeholder view of the firm integrates Porter's structure-conduct-performance (Porter, 1991) and the resource-based view (Barney, 1991 & 2001a) to provide a dynamic framework (Post, Preston & Sachs, 2002), where interaction between the various stakeholders constitutes the firm's access to resources. Both external and internal stakeholders' perceptions of these past interactions and expectations of future prospects are reflected in corporate reputation, which represents a strategic resource to the firm's competitive advantage (Mahon, 2002; Logsdon & Wood, 2002), consistent with the resource-based view. While it would be altruistic for firms to be intrinsically committed to stakeholder interests because it is the right thing to do, a more instrumental approach views stakeholder management as a strategic tool to improve financial performance and provide a sustainable competitive advantage over the long term (Donaldson & Preston, 1995; Harrison & Freeman, 1999; Wood, 1991). The institutional view of the corporation's legitimacy is developed in relationship to how a corporation manages its stakeholders (Wood, 1991), with any loss of legitimacy seen as a withdrawal of stakeholder support. While "legitimacy is a perception or assumption in that it represents a reaction of observers to the organization as they see it; thus legitimacy is possessed objectively, yet created subjectively" (Suchman, 1995: 574), making it both the cornerstone and 245 Achilles' heel of the stakeholder relationship. Corporate Social Responsibility Lacking a single, universal definition, CSR is often linked to sustainability, stakeholders and ethics. The World Business Council for Sustainable Development definition exemplifies the underlying partnership of CSR as "the commitment of business to contribute to sustainable economic development, working with employees, their families, the local community and society at large to improve their quality of life" (WBCSD, 2002). The breadth and diversity of the elements included in CSR range from the environmental aspects of air, water and habitat pollution and degradation to employee and human rights abuses in unfair labour practices and dangerous products sold to unsuspecting consumers (Harrison & Freeman, 1999). The relationship between CSR and financial performance has not yet established a consistent pattern (Griffin & Mahon, 1997; Waddock & Graves, 1997), however major corporations do recognize the advantage of social and environmental initiatives on long-term profitability (Epstein & Roy, 2003). The need to make the business case to ensure the endurance of otherwise vulnerable CSR initiatives (Epstein & Roy, 2003) is no different from any other strategic choice. Porter and van der Linde (1995: 98) argue, "Properly designed environmental standards can trigger innovation that may partially or more than fully offset the costs of complying with them". For example, studies into corporate environmental performance have concluded that "it pays to be green and that this relationship strengthens with industry growth" (Russo & Fouts, 1997: 534), leading to firm specific benefits. Defining the Strategic Attributes of CSR initiatives Three key dimensions identifying whether a CSR initiative is strategic are centrality, specificity and visibility (Burke & Logsdon, 1996). These may be compared to the three key attributes of power, legitimacy and urgency used to determine stakeholder salience (Mitchell et al., 1997), in that they distinguish a combination of characteristics necessary to attain a FMA.. Centrality. An intuitive yardstick of whether any firm initiative is truly strategic is the proximity of that initiative to the core mission or strategic objectives of the firm (Burke & Logsdon, 1996). Using the resource-based view, the resource needs of strategic objectives will have first claim to firm-specific resources, justified by expectations that fulfilling these objectives will maximize the benefits reaped on the resources invested (Burke & Logsdon, 1996; Peteraf & Barney, 2003). For example, funds spent by a professional sports team to sign star athletes are directly aligned with strategic objectives and would thus denote high centrality. However, funding of the junior league feeder team, while developing future talent, would be viewed as being more distant, or of lower centrality. In the context of a clear CSR initiative, the generic corporate giving programs would constitute low centrality, since firms do not define charitable donations as their mission. However, a CSR initiative congruent with corporate objectives, such as a publishing house's community program to increase literacy and encourage book ownership would have greater centrality. Similarly, government lobbying to exempt books from retail sales taxes, making them more affordable to the consumer, would have high centrality to the publishing house. However, these same initiatives might be undertaken for ideological reasons by another firm, such as the sports team, to have but very low centrality. Specificity. "The firm's ability to capture or internalize the benefits of a CSR programme, 246 rather than simply creating collective goods which can be shared by others in the industry, community or society at large" is defined as specificity by Burke and Logsdon (1996: 497). The strategic value of a CSR initiative has greatest significance when some of the benefits are exclusive to the firm, as opposed to just the general populace. The capturing of benefits from a CSR initiative does not detract from the ethical anchor (Woods, 1991), but moreover reinforces the firm's commitment and capacity to make a compelling business case (Epstein & Roy, 2003). The numerous ecological initiatives that provide societal benefits such as clean air and water, reduced reliance on non-renewable energy sources, reduced waste and more recycling do not have to trade off against firm-specific benefits (Porter & van der Linde, 97). Lovins, Lovins and Hawken (1999) cite examples of Johnson & Johnson's redesign of packaging saving 330 acres of forest and $2.8 million annually and Dupont's recuperation and recycling of its polyester industrial film, costing less and attracting a higher price from customers pleased with its greater strength for a significant net benefit to Dupont. Porter and van der Linde (1995) show how Dow Chemical's production process redesign, costing $250,000 to implement, generated an annual savings of $2.4 million, reducing caustic waste by 6,000 tons and hydrochloric acid waste by 80 tons. Russo & Fouts (1997: 537) argue the profitability of firms "going beyond compliance to focus on prevention". While many CSR initiatives may advantage the general community, they may not procure any monetary or tangible advantage for the firm. It is however, important to carefully evaluate the broader implications of a CSR initiative to recognize indirect benefits, such as enhanced reputation, in order to determine the specificity of benefits (Epstein & Roy, 2003). Visibility. A strategic CSR initiative is one that is visible and recognized by internal and external stakeholders (Burke & Logsdon, 1996) and by extension also a contributor to the creation of a firm's reputation (Logsdon & Wood, 2002). While positive visibility is largely controlled or even promoted by the firm, negative visibility of corporate social irresponsibility is generally beyond the control of the firm. With the telecommunications revolution and the scope of global media reporting, information rapidly reaches the public domain, facilitating stakeholder mobilization (Dawkins & Lewis, 2003). A firm whose stakeholders attribute a very high reputation for CSR risks a major backlash should there be any breach of CSR (Dawkins & Lewis, 2003; Lewellyn, 2002). Ironically a firm with a less strong CSR reputation would suffer less damage for the same behaviour (Lewellyn, 2002). Mahan's (2002: 415) quote of Shakespeare's Othello, "Reputation is an idle and most false imposition; oft got without merit, and lost without deserving", reinforces the notion that while costing nothing, reputation is priceless. Viewed as a firm resource (Barney, 1991, 2001a, 2001b; Oliver, 1997; Russo & Fouts, 1997; Teece et al., 1997), corporate reputation is built over time and is a perception of how well the firm meets stakeholders' expectations (Rowley, 1997), which also change over time (Lewellyn, 2002; Mahon, 2002). The fact that different stakeholders have different perceptions and biases (Lewellyn, 2002; Mahon, 2002) places added onus on what visibility a CSR initiative will receive and how it will be interpreted. First-Mover Advantage of a Corporate Social Responsibility Initiative The combination of strategic characteristics of a CSR initiative of (1) centrality, (2) specificity and (3) visibility are central to whether a pioneer can claim a FMA. The seven possibilities include one combination of all three characteristics, three possibilities for two characteristics and three single attributes, as depicted in the following model: 247 Figure 1: Strategic attributes of a CSR initiative 1-Centrality 4 2-Specificity FMA 5 7 6 4=Add visibility for FMA 5=Late-mover advantage 6=Early-mover advantage 7=First-mover advantage 3-Visibility When a firm institutes a visible new CSR initiative congruent with corporate objectives, yielding direct benefits to the firm, a FMA is highly likely to accrue to the firm by virtue of this CSR initiative constituting a valuable, rare and inimitable resource. A competitor would unlikely be able to replicate or surpass the firm-specific benefits of the CSR initiative, and any challenge would be able to be defended by the firm to maintain its FMA. Some of the possible outcomes of achieving FMA include being able to establish the firm as the "prototype against which all later entrants are judged" (Kerin et al., 1992: 35), setting the industry standards (Robinson & Min, 2002) and influencing the direction of environmental regulations (Porter & van der Linde, 1995). For example, The Body Shop International's (BSI, 2004) campaign to stop violence against women in Canada is a CSR initiative central to corporate objectives as documented in their mission statement entitled "Our Reason for Being". The nine-year history of this highly visible "STOP Violence Against Women" awareness and education program has given The Body Shop International (BSI) a very positive public image to reflect favourably on the corporate reputation. The sensitization and fund-raising initiatives within the retail outlets generate additional store traffic and attract consumers that might be more interested by the cause and BSI's implication, than in a specific BSI product. Much of the implication in supporting violence recovery and prevention programs also comes from staff volunteering their time (Hartman & Beck-Dudley, 1999), thereby providing this high visibility at a minimal cost while generating additional sales volumes to the benefit of BSI. The impact of this high visibility on employee advocacy (Dawkins & Lewis, 2003) and on employee morale are also seen as contributors to attracting and retaining high calibre employees (Burke & Logsdon, 1996). Therefore: Proposition 1. Where all three strategic attributes of centrality, specificity and visibility are high, a CSR initiative is likely to provide a FMA. Where there are only two of the strategic attributes, the pairings will determine the possibility of a FMA. As visibility is generally within the firm's control, a strategic CSR initiative meeting the other two criteria could easily be the subject of low-cost press releases or other media interviews. 248 With the three attributes this CSR initiative would then likely lead to a FMA. Hence: Proposition 2. Where the strategic attributes of centrality and specificity are high, a CSR initiative can easily obtain high visibility to likely provide a FMA. The visibility of the CSR initiative would provide information to competitors who could also have similar objectives encouraging them to replicate the initiative at a lower cost than innovation. The lack of any firm-specific benefit would not carry an opportunity cost to any delay in adoption. The possibility to free ride another firm's initiative and possibly create a firm-specific benefit would suggest a late-mover advantage situation. Returning to the earlier example of the publishing house and their lobbying efforts to eliminate retail sales taxes on books; letting another publisher spend money on such an initiative and only jumping in when the tax reversal is imminent would provide a late-mover advantage. Accordingly: Proposition 3. Where the strategic attributes of centrality and visibility are high, a CSR initiative provides no first-mover advantage but would support a late-mover advantage. While not centered on the firm's strategic objectives, a visible CSR initiative that provides tangible benefits will benefit from early adoption, since any delay would incur the opportunity cost of foregoing the benefits. A FMA would be less likely since further corporate investment to maintain the distinctiveness of a non-central CSR initiative would probably not be forthcoming. In describing Germany's leading recycling standards Porter and van der Linde (1995: 105) note these "gave German firms an early-mover advantage in developing less packaging-intensive products, which have been warmly received in the marketplace", attracting premium prices. Given that packaging is rarely a central corporate objective, these firms would readily be characterized as having specificity and visibility, pointing to the likelihood of having early-mover advantage. While competitors might replicate these CSR initiatives to eradicate any FMA, further investment would be unlikely and therefore a FMA could not be defended. Therefore: Proposition 4. Where the strategic attributes of specificity and visibility are high, a CSR initiative would not provide a FMA but could provide an early-mover advantage. The underlying rational applied to each single attribute follows. Centrality provides legitimacy for the CSR initiative, beyond just doing the right thing, however its vulnerability to changing priorities without any supporting firm-specific benefits or visibility (Epstein & Roy, 2003) would not provide a sustainable FMA. Specificity provides a financial justification for the CSR initiative, however this does not preclude imitation or being surpassed by competitors. It is likely that this type of CSR initiative will survive only as long as the firm-specific benefit is convincingly positive (Epstein & Roy, 2003), although it is highly unlikely to ever constitute a FMA. For example, extrapolating from the earlier example of Dupont's recycling initiative, should the manufacture of polyester industrial film not fall within core objectives and there be no particular visibility, the recycling operation would be vulnerable and unable to provide a FMA. It would be unable to attract corporate investment to maintain its technological advance if it is not part of core objectives, leaving the way open for a competitor to gain an advantage by free-loading and building a more cost-efficient process. Visibility alone cannot provide a FMA since the CSR initiative can easily be copied or bettered, while the firm would have no incentive to make further investments on an initiative that generates no firm-specific benefits, to procure a FMA. Although a CSR initiative having only the 249 attribute of being visible to stakeholders could be linked to the valuable resource of the firm's reputation, this would not be sufficient to ensure a FMA. Accordingly, Proposition 5. Where only one of the strategic attributes of centrality, specificity and visibility is high, a CSR initiative will not generate a FMA Discussion I have proposed that for a CSR initiative to provide a FMA, it must have high centrality, specificity and visibility. This strategic CSR initiative in the resource-based view of the firm constitutes a rare, valuable and inimitable resource whose distinctiveness provides the firm with a FMA. This FMA theory of CSR is presented as descriptive propositions. No prescription is offered; the model proposes the three strategic criteria of centrality, specificity and visibility as considerations for assessing the strategic advantage of a CSR initiative to provide a FMA. This new perspective has implications for both research and management. Empirical testing of whether the theory's explanation of the strategic criteria of centrality, specificity and visibility for CSR initiatives describes the conditions necessary for a FMA would be a starting point. The applicability to the various social and environmental CSR initiatives should also be tested. Case studies could also be informative and provide an insight to causality. While the CSR initiative's degree of centrality to corporate objectives is proposed, is it the CSR initiative that moves corporate objectives closer to the CSR initiative to then create a FMA? Further avenues of research could also include integrating the role of the stakeholders with the strategic CSR initiative to refine the model. Managers can also use this model when evaluating CSR initiatives to determine the degree of commitment and how to solidify the CSR initiative to create a FMA. The consideration of the three strategic criteria of centrality, specificity and visibility can help management frame a CSR initiative within their overall business strategy and build a business case justification. Conclusion More than just doing the right thing, strategic CSR initiatives can also be viewed as rare, valuable and costly to copy resources that under certain conditions can provide the firm with a FMA. These can translate to sustainable competitive advantages where visible CSR initiatives are central to corporate objectives and yield firm-specific benefits. This CSR theory opens a new avenue of theory which complements stakeholder theory and recognizes the strategic opportunities afforded by CSR initiatives. In aligning CSR with the sustainability of competitive advantage, future research may open a window to how CSR pays off for the firm. References Barney, Jay B., "Firm resources and sustained competitive advantage", Journal of Management, 17, (1991), 99-120. Barney, Jay B., "Is the resource-based "view" a useful perspective for strategic management research? Yes," Academy of Management Review, 26 (1), (2001a), 41-56. Barney, Jay B., "Resource-based theories of competitive advantage: A ten-year retrospective on the resource-based view," Journal of Management, 27, (2001b), 643-650. "Body Shop International" (BSI), 2004, http://www.thebodyshop.ca. [Accessed 2005-01-01]. 250 Burke, Lee & Logsdon, Jeanne M., "How Corporate Social Responsibility Pays Off," Long Range Planning, 29 (4), (1996), 495-502. Clarkson, Max B.E., "A Stakeholder Framework for Analyzing and Evaluating Corporate Social Performance," Academy of Management Review, 20 (1), (1995), 92-117. Dawkins, Jenny & Lewis, Stewart, "CSR in Stakeholder Expectations: And Their Implication for Company Strategy," Journal of Business Ethics, 44 (2/3), (2003), 185-193. Donaldson, Thomas & Preston, Lee E., "The Stakeholder Theory of the Corporation: Concepts, Evidence and Implications," Academy of Management Review, 20 (1), (1995), 65-91. Epstein, Marc J. & Roy, Marie-Josée, "Making the Business Case for Sustainability. Journal of Corporate Citizenship, 9, (2003), 79-96. Freeman, R. Edward, Strategic management: A stakeholder approach, Boston, Pitman Publishing, 1984. Frooman, Jeff, "Stakeholder influence strategies," Academy of Management Review, 4 (2), (1999), 191-205. Gal-Or, Ester, "First Mover and Second Mover Advantages," International Economic Review, 26 (3), (1985), 649-653. Griffin, Jennifer J. & Mahon, John F., "The Corporate Social Performance and Corporate Financial Performance Debate: Twenty-Five Years of Incomparable Research," Business & Society, 36 (1), (1997), 5-31. Harrison, Jeffrey S. & Freeman, R. Edward, "Stakeholders, social responsibility, and performance: Empirical evidence and theoretical perspectives," Academy of Management Journal, 42 (5), (1999), 479-485. Hartman, Cathy L. & Beck-Dudley, Caryn L., "Marketing Strategies and the Search for Virtue: A Case Analysis of The Body Shop, International," Journal of Business Ethics, 20 (3), (1999), 249-263. Hoopes, David G.; Madsen, Tammy L. & Walker, Gordon, "Guest editors' introduction to the special issue: Why is there a resource-based view? Toward a theory of competitive heterogeneity," Strategic Management Journal, 24 (10), (2003), 889-902. Jensen, Richard, "Innovative leadership: First-mover advantages in new product adoption", Economic Theory, 21 (1), (2003), 97-116. Kerin, Roger A., Varadarajan, P. Rajan & Peterson, Robert A., "First-Mover Advantage: A Synthesis, Conceptual Framework, and Research Propositions," Journal of Marketing, 56 (4), (1992), 33-51. Lewellyn, Patsy G., "Corporate Reputation: Focusing the Zeitgeist," Business and Society, 41 (4), (2002), 446-455. Lieberman, Marvin B. & Montgomery, David B., "First-mover advantages," Strategic Management Journal, 9 (1), (1988), 41-58. Lieberman, Marvin B. & Montgomery, David B., "First-mover (dis)advantages: Retrospective and link with the resource-based view," Strategic Management Journal, 19 (12), (1998), 1111-1125. Logsdon, Jeanne M. & Wood, Donna J., "Reputation as an Emerging Construct in the Business and Society Field: An Introduction," Business and Society, 41 (4), (2002), 365-370. Lovins, Amory B.; Lovins, L. Hunter & Hawken, Paul., "A Road Map for Natural Capitalism," Harvard Business Review, 77 (3), (1999), 145-158. Mahon, John F., "Corporate reputation: A Research Agenda Using Strategy and Stakeholder Literature," Business and Society, 41 (1), (2002), 415-445. Makadok, Richard, "Toward a synthesis of the resource-based and dynamic-capability views of rent creation," Strategic Management Journal, 22 (5), (2001), 387-401. Mitchell, Ronald K.; Agle, Bradley R. & Wood, Donna J., "Toward a theory of stakeholder identification and salience: Defining the principle of who and what really counts," Academy 251 of Management Review, 22 (4), (1997), 853-886. Mueller, Dennis C., "First-mover advantages and path dependence," International Journal of Industrial Organization, 15, (1997), 827-850. Nehrt, Chad, "Timing and intensity effects of environmental investments," Strategic Management Journal, 17 (7), (1996), 535-547. Peteraf, Margaret, "The cornerstones of competitive advantage: a resource-based view," Strategic Management Journal, 14 (2), (1993), 179-191. Peteraf, Margaret A. & Barney, Jay B., "Unravelling The Resource-Based Tangle," Managerial and Decision Economics, 24 (10), (2003), 309-323. Peteraf, Margaret A. & Bergen, Mark E., "Scanning dynamic competitive landscapes: A market-based and resource-based framework," Strategic Management Journal, 24, (2003), 1027-1041. Porter, Michael E., "Towards a Dynamic Theory of Strategy," Strategic Management Journal, 12, (Winter, 1991), 95-117. Porter, Michael E. & van der Linde, Claas, "Toward a New Conception of the Environment-Competitiveness Relationship," Journal of Economic Perspectives, 9 (4), (1995), 97-118. Post, James E.; Preston, Lee E. & Sachs, Sybille, Redefining the Corporation – Stakeholder Management and Organizational Wealth, Stanford, CA: Stanford University Press, 2002. Ray, Gautam; Barney, Jay B. & Muhanna, Waleed A., "Capabilities, business processes, and competitive advantage: Choosing the dependent variable in empirical tests of the resource-based view," Strategic Management Journal, 25 (1), (2004), 23-37. Robinson, William T. & Min, Sungwook, "Is the First to Market the First to Fail? Empirical Evidence for Industrial Goods Businesses," Journal of Marketing Research, 39 (1), (2002), 120-128. Rowley, Timothy J, "Moving beyond Dyadic Ties: A Network Theory of Stakeholder Influences," Academy of Management Review, 22 (4), (1997), 887-910. Russo, Michael V. & Fous, Paul A., "A resource-based perspective on corporate environmental performance and profitability," Academy of Management Journal, 40 (3), (1997), 534-559. Shankar, Venkatesh; Carpenter, Gregory S. & Krishnamurthi, Lakshman, "Late Mover Advantage: How Innovative Late Entrants Outsell Pioneers," Journal of Marketing Research, 35 (1), (1998), 54-70. Suchman, Marc C., "Managing Legitimacy: Strategic and Institutional Approaches," Academy of Management Review, 20 (3), (1995), 571-610. Teece, David J.; Pisano, Gary & Shuen, Amy, "Dynamic Capabilities and Strategic Management," Strategic Management Journal, 18 (7), (1997), 509-533. Vos, Janita F.J., "Corporate social responsibility and the identification of stakeholders," Corporate Social Responsibility and Environmental Management, 10, (2003), 141-152. Waddock, Sandra A. & Graves, Samuel B., " The corporate social performance-financial performance link," Strategic Management Journal, 18 (4), (1997), 303-319. Wernerfelt, Birger, "The resource-based view of the firm: Ten years after," Strategic Management Journal, 16 (3), (1995), 171-174. Wood, Donna J., "Corporate social performance revisited," Academy of Management Review, 16 (4), (1991), 691-718. World Business Council for Sustainable Development (WBCSD), "Corporate Social Responsibility: The WBCSD's Journey," (2002), http://www.wbcsd.org. [Accessed 2005-01-01]. Zang, Shi & Markman, Arthur B., "Overcoming the Early Entrant Advantage: The Role of Alignable and Nonalignable Differences." Journal of Marketing Research, 35 (4), (1998), 413-426. 252 ASAC 2005 Toronto, Ontario Melissa Toffanin (student) Martin L. Martens John Molson School of Business Concordia University EXAMINING PRE-IPO ACTIVITY: FIRM GROWTH BY ACQUISITION In this article, we examine the effects of engaging in acquisition activity before issuing an initial public offering. We find that pre-IPO acquisitions allow these firms to match the valuations of similar firms that do not engage such acquisitions. Additionally, firms that engage in pre-IPO acquisitions are more likely to fail three years after the IPO. Introduction Obtaining capital to finance firm growth through acquisitions is one of the primary reasons for going public (Pagano, Panetta, and Zingales, 1998; Shultz and Zaman, 2001; Wiggenhorn and Madura, 2004). Considering that financing acquisitions is a common reason listed in the use of proceeds, one might think that acquisitions in the period prior to issuing the IPO would be a rare or uncommon activity. Yet an examination of several IPO prospectuses quickly identifies firms issuing pro forma financial statements and discussions about acquisitions in period in the year to two before the IPO. While numerous researchers have investigated the acquisition activity of public firms and, more recently, the acquisition of newly public firms (Wiggenhorn and Madura, 2004), the motives for and implications of acquisitions conducted prior to firms going public have been ignored in both the IPO literature and the merger and acquisition literature. In this paper we begin to fill this gap by examining conceptual reasons for pre-IPO acquisitions by firms who go public soon after. We offer a series of hypotheses regarding potentially significant impacts of pre-IPO acquisition activity on post-IPO performance, both short and long term, on underpricing, and whether any noteworthy patterns might be inferred from our sample of firms. If differences between the performance of non-acquiring IPO and acquiring IPO firms exist, for example, then inferences may be made regarding the reasoning behind pre-IPO acquisitions. More specifically, if we find that IPO firms that engage in acquisition activity significantly outperform firms that do not engage in pre-IPO acquisitions, then we may reasonably infer that the acquisitions are value-increasing and were conducted to improve the overall performance of the firms prior to their offerings. Conversely, if non-acquiring IPO firms outperform acquiring IPO firms, then we may reasonably infer that the acquisitions may not have had a value-increasing motive, but were conducted for other reasons, such as managerial incentives, managerial hubris, and sending signals to the institutional investors to improve the legitimacy of the firm prior to its IPO. In this research, we hope to gain insight into the initial public offering process, merger and acquisition activities by young firms, and corporate strategies of firms preparing to issue initial public offerings. 253 Literature Overview In this section, we examine existing literature regarding motives for acquisitions, their performance, the effect of method of payment, and roll-up IPO transactions. In general, we examine three main theories regarding the motives for acquisitions: the synergy hypothesis, the hubris hypothesis, and the managerialism hypothesis. The Synergy Hypothesis The synergy hypothesis suggests that acquisitions may occur when the combined value of two firms is greater than the sum of the individual firm values (i.e.: Bradley, Desai, and Kim, 1988). Synergy potential is the only value-creating acquisition motive (Seth, Song and Pettit, 2000). In this conceptual framework, the management team is believed to maximize shareholder value by taking on only positive NPV projects that result in statistically significant wealth increases. Synergies can arise in three major forms: financial, operational, and managerial (Trautwein, 1990). He argues that financial synergies can lower a firm’s cost of capital through increased access to cheaper capital, the establishment of an internal capital market, and/or reduced systematic risk. Secondly, operational synergies can occur when economies of scale and scope are achieved by combining previously-separate operating units such as production, marketing, R&D, and administration. Thirdly, managerial synergies can be realized when the acquiring firm’s management has superior skills, abilities, and knowledge than incumbent s. Larsson and Finkelstein (1999) note that “the various sources of synergy define a combination’s potential, which in turn is expected to affect the extent to which synergies will be realized in an acquisition” (p. 5). Thus, mergers with high combination potential are likely to realize significant efficiencies, while those with low combination potential are unlikely to extract any sizeable synergies. Larsson and Finkelstein (1999) further note that the traditional view of enhanced synergy creation through combinations of firms with similar or related operations is incomplete as it ignores the potential for considerable synergy creation through “complementary synergy sources that may be present throughout the value chain” (p.6) According to Seth, Song and Pettit (2000), the synergy hypothesis assumes that the synergies created by combining two firms cannot costlessly be replicated by other firms, as many of the benefits realized can be intangible, and that market frictions exist to prevent the facilitation of replication. The Hubris Hypothesis The hubris hypothesis was first advanced by Roll (1986). Observing that acquisition premiums are sometimes greater than the total gain that the market assigns to the transaction, he concluded that at least part of the price paid is just a value transfer from the bidder to the target. Because it may be unreasonable to assume that the managers make frequent valuation mistakes and do not learn from them, Roll argues that takeovers produce no gains – the managers of the acquiring firms make errors in estimating the potential gains, but proceed with the merger assuming their valuations to be correct. Although errors made by managers can either overestimate or underestimate the value, Roll argues that the error we observe is usually in the same direction, and that the left tail of the 254 valuation distribution is truncated at the current market price, as no offers would be accepted for anything less than what can be achieved by selling the shares in the market. Roll notes that his hubris hypothesis is consistent with market efficiency of the strong-form, where all information about a firm is reflected in the current market price. Thus, in this case, managers are irrational if they make a bid at a price that is anything greater than the current market price, given that there are no expected synergistic gains. If the hubris hypothesis is taken to the extreme, takeovers result in no gains overall and are mere transfers of wealth from acquirer shareholders to target shareholders. Empirical evidence from Roll’s study indicates consistency with the extreme form of the hubris hypothesis, while Berkovitch and Narayanan (1993) and Firth (1991) find evidence that supports the hubris hypothesis, but not the extreme form. Hayward and Hambrick (1997) argue that it is important to examine managerial incentives in order to understand the motives for acquisition because the individual or group that makes the acquisition decision (the CEO or top management team) can be clearly identified. Malmendier and Tate (2004), concentrating on overconfident CEOs, found that they are more likely to conduct a merger in the first place, because they identify undervalued companies more often when compared to other managers. Malmendier and Tate (2004) also argue that the market reacts less intensely to the acquisitions announced by overconfident CEOs. In the case of firms issuing IPOs, however, many decisions regarding the reshaping and restructuring of the firm are not made or driven by the top management team but by the lead underwriters (Martens 2004). If the pre-IPO acquisition activity is driven by demands made by the underwriters or venture capitalists, and the CEO or top management team acquiesces to these demands, then the source of the hubris may be the underwriters or venture capitalists rather than the management team. If the acquisitions are made in an effort to signal the firm’s fitness to successfully operate as a public firm, or simply its fitness to issue an IPO, then the decisions may be an effort to legitimize the firm in the eyes of potential institutional investors. These acquisitions are, in effect, polishing the firm to get it ready for the IPO. The managerialism hypothesis, also known as the agency motive, was first advanced by Marris (1964), and suggests that it is not the best interests of the shareholders that guide management actions but managements’ own self-interests (Jensen and Meckling, 1976; Jensen, 1986). Managers act to maximize their own utility at the expense of the firm shareholders and knowingly overpay for the target firm. Marris argues that managerial compensation structures give managers the incentive to increase the value of assets under their control, rather than firm profits. In their study over the period 1963 to 1988, Berkovitch and Narayanan (1993) examine negative returns to acquiring firms and find significant evidence that managerialism influences acquisition decisions. Slusky and Caves (1991) focus their research on the premiums paid in acquisitions in order to find evidence of synergies, agency, and/or managerial hubris. While they do find evidence of financial (as opposed to real) synergies in their study, the overwhelming motive in their sample is agency-based. In terms of acquisition performance, the empirical results have been mixed – some researchers provide evidence that merger activity adds value, others find it destroys value, and several find a combination of the two. For example, Bradley, Desai, and Kim (1988) find positive total gains in their sample and thus purport that mergers are value-enhancing activities. Conversely, though target returns were found to be positive, returns to the acquirer were found to be negative at least half of the time. Mulherin and Boone (2000) also find large and significantly positive returns to target firm shareholders, but negative and insignificant returns to acquirers. Other research consistent with these results include Asquith (1983), Bradley, Desai, and Kim 255 (1983), Jensen and Ruback (1983), Malatesta (1983), Servaes (1991), Andrade, Mitchell, and Stafford (2001), and Fuller, Netter, and Stegemoller (2002). In a recent study, Wiggenhorn and Madura (2004) examine the acquisition activity and performance of newly-public firms, arguing that the transactions are more likely to be value-increasing as potential differences in life cycles, transparency in disclosure, and increased corporate governance for the newly-public firms can allow for greater firm value enhancement than found in acquisitions by established public firms. Wiggenhorn and Madura (2004) further note that the managerialism motive behind acquisitions of newly-public firms is reduced, given that much of the IPO proceeds have already been allotted and are unavailable to the management team for self-serving interests. They find that, contrary to the general findings for acquisitions by established public firms, there is a positive and significant acquisition announcement effect for newly-public firms. Thus, their argument of greater value enhancement in acquisitions by newly-public firms is supported by their findings. The effect of the method of payment in acquisitions has also been well-studied in the literature, where it has been found that there are significant differences in the return and operating performance of acquisitions financed by cash and those financed by stock. In general, it is found that cash payments are associated with positive but insignificant returns to acquirers, while acquisitions financed with stock result in significantly negative acquirer returns (e.g. Wansley, Lane, and Yang, 1987; Asquith, Brunner, and Mullins, 1987; Travlos, 1987). A related consideration in this literature is that of acquisitions and free cash flow. Free cash flow is generally defined as being the cash available after all positive NPV projects have been undertaken (Jensen, 1986). Ideally, since only negative NPV projects remain, any excess cash flow should be distributed to the firm’s shareholders. However, in many cases, this distribution is not made and the cash flow is used to finance other endeavors, including acquisitions. Jensen (1986) argues that managers act not in the interests of the shareholders but act to maximize their own utility, and thus they have the incentive to use the free cash flows of the firm for value-destroying investments that suit their own needs. Since issuing equity increases agency costs and the potential for misuse of the cash, the market reaction upon equity offerings is usually negative. Roll-up IPOs are a specific type of acquisition that is related to our research. Brown, Dittmar, and Servaes (2000) note that roll-up transactions occur where “a shell company goes public while simultaneously merging with a number of other firms that operate in the same industry” (p.1). The main motive of this approach is to consolidate firms in highly-fragmented industries in order to achieve revenue and cost synergies. The merger occurs just prior to or in conjunction with the IPO. In their study of 47 roll-up transactions over the period 1994 to 1998, Brown et al. (2000) find that while the initial performance of this type of IPO is similar to that of traditional IPOs, Roll-up IPOs significantly underperform several benchmarks and traditional IPOs over the long term. In a related paper, Brown, Dittmar, and Servaes (2004) found that Roll-up IPOs are more successful in the long run if the founder remains involved. Brown et al. (2003) argue that industry consolidation success is significantly determined by initial structure. In this article, we offer several hypotheses regarding the motives for pre-IPO acquisitions. The first come from the synergy hypothesis literature. In many IPOs, firms have little or no revenue inflows, especially those in the high-tech and pharmaceutical industries; in order to be able to report a steady income stream in the prospectus and thereby encourage risk-averse investors to participate in the offering, the IPO firm purchases firms with relatively stable incomes. In doing so, synergies (both tangible and intangible) can be achieved. At the very least, financial synergies 256 can be achieved as the acquisition can establish an internal capital market and can increase the firm’s access to outside financing. Yi (2001) examined the pre-IPO earnings of firms going public over the period from 1987 to 1991. For firms with positive earnings at the time of the IPO, he did not find any significant benchmark-adjusted abnormal returns, while firms with negative earnings had significant negative benchmark-adjusted abnormal returns. As well, Yi (2001) argued that IPO firms with positive earnings outperform those reporting negative earnings. Thus, as there is evidence that earnings can significantly impact long-term IPO performance, we argue that firms attempt to improve their reported earnings through acquisitions prior to their offerings and offer the following hypotheses: H1a: Compared to similar IPO firms that do not engage in acquisition activity prior to their IPOs, firms that engage in pre-IPO acquisition activity will have lower levels of ex-ante risk level for the IPO. H1b: Compared to similar IPO firms that do not engage in acquisition activity prior to their IPOs, firms that engage in pre-IPO acquisition activity will have better long-term survival performance. The next hypothesis comes from the hubris hypothesis literature, in which authors argue that firms will engage in pre-IPO acquisitions as a method for increasing firm size, but that managers may make errors in valuing the target firm and overestimate expected synergies. Thus, the acquiring firm may overpay for the target and the transaction will be value-decreasing overall. Klein (1996) studies the relationship between the price per share and several variables for 193 IPOs, finding a positive relationship between the IPO share price and both pre-IPO earnings per share (EPS) and pre-IPO book value of equity per share. Therefore, firms wishing to go public may acquire to increase their book values of equity in an attempt to increase their IPO valuations. Thus, we offer the following hypothesis: H2: Compared to firms that do not engage in acquisition activity prior to their IPOs, firms that engage in pre-IPO acquisition activity will have higher levels of IPO proceeds. The third set of hypotheses comes from the managerialism (or agency) hypothesis literature. Under this perspective, managers of the pre-IPO firm will make acquisitions that are in their own self-interests, and not in the interests of value-maximization. Mann and Sicherman (1991) note that: “…since managers may have incentives to expand firm size and purchase unrelated assets, any unencumbered (or nonbonded) cash requests by management are likely to be perceived negatively by shareholders. If managers hoard free cash flow rather than distributing it to shareholders, they can finance these projects internally and avoid seeking financing in the capital markets. When firms are seeking funds in well-functioning capital markets, it is easier to monitor management behavior.” (p. 215) Since a pre-IPO firm is, by definition, not yet public, investors cannot bid down the price of the firm’s shares when they believe that the firm is engaging in acquisitions that are only in the interests of management. Thus, the manager who may have built up cash reserves can time his/her non-value-maximizing acquisitions to occur while the firm is still private in order escape market discipline. While in efficient markets investors would take this “bad” transaction into account at the time of the IPO and price the firm’s shares accordingly, if the acquisitions are made close 257 enough to the IPO, it may be difficult to observe any negative impact on the firm’s value, and the manager’s non-value maximizing behavior goes relatively unpunished. The finance literature contains a plethora of documentation of the fact that, in general, IPOs are underpriced (e.g. Ritter, 1991; Ritter and Welch, 2002; and Loughran and Ritter, 2004). Many theories and hypotheses are devoted to attempting to explain this phenomenon (Tinic, 1988). One of the main theories relevant to our research is that of the relationship between ex-ante uncertainty (in terms of variables such as sales, amount and use of proceeds, industry, and underwriter factors) and underpricing. Beatty and Ritter (1986) and Clarkson and Merkley (1994) look at the relationship between ex-ante uncertainty and underpricing, arguing that the greater is this uncertainty, the greater will be the underpricing. They base their argument on the winner’s curse problem - to induce the uninformed investors to remain in the IPO market, issues need to be underpriced. Beatty and Ritter (1986) note that smaller issues, in general, are more risky in nature. Overall, the authors conclude that IPOs having greater ex-ante uncertainty do exhibit greater underpricing. Melnik and Thomas (2003) found that firms with larger revenues are less underpriced. Thus, in cases where the acquisitions are made for securing revenue streams, we might expect lower underpricing. Conversely, in cases where the acquisitions are conducted in managements’ interests and are not value-increasing transactions, we would expect greater underpricing. However, we can also make the prediction that the pre-IPO acquisitions will have no discernible effect on underpricing as the results of the transactions will have already been incorporated into the value of the firm. Considering this literature, we offer the following hypotheses: H3: Compared to firms that do not engage in acquisition activity prior to their IPOs, firms that engage in pre-IPO acquisition activity will have lower levels of underpricing. As we previously discussed, the method of payment used to finance an acquisition has been shown to impact the performance of the combined firm. Acquisitions financed with cash perform better than acquisitions financed with stock, with returns to acquirers being positive (or zero) and negative, respectively (i.e. Ritter, 1991; Ritter and Welch, 2002; Loughran and Ritter, 2004). In terms of the choice of financing method, Fishman (1989) purports that acquirers will finance deals with cash in order to deter competing bids when they have favorable private information that indicates a high value for the target, possibly stemming from expected synergies. Chang and Mais (2000) assert that “in mergers, bidding firms are predicted to finance with stock when they have high growth opportunities” (p. 143). Conversely, Travlos (1987) and Wansley, Lane, and Yang (1987) find that the use of stock financing for acquisitions is a signal that the acquiring firm is overvalued, and cash is used in cases where the firm is undervalued. In light of these findings, we offer the following hypotheses: H4: Compared to firms that finance their acquisitions with stock, firms financing the transactions with cash will have higher levels of long-term survival performance. H5a: Synergy will be the primary motive for those acquisitions financed with cash. H5b: Agency or managerial hubris will be the primary motive for acquisitions financed with stock. Data and Methods 258 Our dataset is composed of 435 firms in the computer software industry that issued an IPO on a US stock exchange between 1996 and 2000. We chose the computer software industry as it contained the largest number of IPOs for any single industry in this period and the single industry sample eliminates the potential for cross-industry accounting problems. All firms listed themselves in the SIC 2-digit classification 73 (Business Services). We collected the data using the prospectuses filed with the Securities Exchange Commission and available in the SEC’s EDGAR database. Of the 435 computer software firms that issued IPOs between 1996 and 2000 on US stock exchanges, we were able to identify 87 companies that made one or more acquisitions in the two years prior to their IPO. For these 87 firms, we found an average of 2.7 acquisitions in the pre-IPO period, with an average total consideration cost (cash, stock, and options) of US$24 million. The average age of these 87 firms is 7.21 years; the average size was 470 employees, with an average IPO valuation of US$81.6 million. Fifty-one of these firms made only a single acquisition before their IPO; however, one 3-year old firm (Verio, Inc.) made 20 acquisitions in the two years before its IPO while a 2-year old firm (Via Net Networks Inc.) made 19 acquisitions in the two years before its IPO. The pre-IPO acquisition consideration cost averaged 33.9% of the firm’s IPO valuation. One firm made two pre-IPO acquisitions that cost 3.35 times the eventual proceeds of its IPO. The firms that made pre-IPO acquisitions were somewhat larger, on average, compared to our entire dataset but the same age. The averages for all 435 IPO firms in our dataset are 7.19 years, 348 employees, and a valuation of US$69.1 million. Dependent variables. We examine four dependent variables in this study. The first is a measure of ex-ante risk, using the BROAD measure of ex-ante risk (MacCrimmon and Martens, 2001). This measure is created by examining the risk factor section of the prospectus for ten specific risk factors. This result is a value from zero to ten where zero indicates that none of the risk factors were listed and ten indicates that all ten were listed. This measure is more specific than a simple count of all risk factors listed (Beatty and Ritter, 1986; Clarkson and Merkley, 1994) and is similar to ex-ante risk factor measures (Welbourne and Andrews, 1996). The second dependent variable (value) is the amount of capital raised in the offering (# of shares * offering price), the valuation of the issue. As the data was left-skewed, we transformed it using a natural log. The third dependent variable (uprice) is first day underpricing (stock price close of first day minus the offering price). The fourth dependent variable (delist) is a dichotomous variable where 1 means that the firm was delisted, for a negative reason, within the first three years on the market. We include the condition ‘for a negative reason’ because some firms are delisted after being acquired by other firms at a price higher than their IPO issue price. We define a delisting failure as one where, within the first three years, the firm is delisted for failing to meet SEC or exchange requirements or if they were acquired by another firm at a stock market-adjusted value that was twenty-five percent below its IPO valuation or lower. We test our hypotheses on IPO valuation and years to delisting using OLS regression analysis. We use logistic regression for the 3-year delisting analysis. Key independent variables. In our study, we employ four acquisition-related variables. The acquisition indicator variable (ACQ) is a dummy where 1 equals a firm engaging in a pre-IPO acquisition and 0 if no acquisitions were made. Acquisition consideration (CONSID) is the total dollar value of the acquisition, in millions. The method of payment variables, CASH and STOCK, are the percentages of total consideration of each respectively. Thus, the cash variable is the total amount of cash paid as a percent of total consideration, while the stock variable is the total value of stock paid as a percent of total consideration. 259 Control variables. We include a number of control variables. Since firms on the NYSE tend to be substantially larger, we include AMEX and NASDAQ as market-level controls. IPO performance may also be influenced by momentum effects and be subject to non-rational enthusiasm over the possible future return of IPOs (Jenkinson and Ljungqvist 2001). IPO Date is the Excel-coded date of the IPO issue (for example 35227 is the equivalent to June 11, 1996). We also include a “Hot IPO” period variable that measures the number of IPOs during the week the firm went public. The measure “BUBBLE” is a dichotomous variable indicating if the firm was part of the 1998-2000 IPO bubble. The NASDAQ prior 30 days return variable controls for the NASDAQ market performance in the 30 days prior to the IPO. The variable is measured as the percentage return on the NASDAQ index during the 30 calendar days prior to the IPO date. The NASDAQ is chosen for this control because the large majority of the issues in our sample trade on this exchange. We include a number of firm-level controls; age at the time of the IPO; Firm size (LOGEES) is measured by taking the natural log of the number of employees at the time of the IPO. We also measure firm financial size by including the total revenue in the year prior to the IPO (TOTREV). Auditor reputation is measured as a dichotomous variable differentiating the involved auditing firms in terms of size (Big 6 vs. non-big 6). Loughran and Ritter’s (2004) underwriter prestige measures are used to control for underwriter reputation (UWRANK). We include a dichotomous measure for profitability (NETINC) where 1 represents a company that reports a net profit in the year prior to the IPO. Fraud is a variable used to take into account firms that are involved in a class-action laddering lawsuit, identified on the website IPOfraud.com. This class action lawsuit was filed in 2003 against 309 IPO firms and the underwriters involved in those firms. 129 of the 309 firms (42%) named in the lawsuit are present in our sample. We include a control for firms where the Founder is still the CEO after the IPO (FCEO) and we control for firm ownership by the CEO (CEORE) that measures the percentage of the outstanding shares owned by the CEO. Results For the first set of models, we examine the effect of acquisitions on our four dependent variables. The results are presented in Table I. In testing the hypothesis that firms make acquisitions pre-IPO to reduce ex-ante uncertainty, we find the acquisition dummy variable to be significant and negative. This supports Hypotheses 1a. Pre-IPO acquisition activity appears to be associated with lower levels of ex-ante risk. When IPO valuation is the dependent variable, we find the acquisition dummy variable is insignificant and thus Hypothesis 2 is not supported. Pre-IPO acquisition activity does not appear to have an effect on the IPO valuation, compared to firms that do not engage in acquisitions. This result suggests that firm growth through internal measures and growth via acquisition are valued essentially the same for IPO firms. Almost all of the other variables are significant in this model, save CEO retained earnings, total revenue, net income, and the hot IPO period dummy. In terms of pre-IPO acquisitions and underpricing, we find an insignificant relationship. Thus, although the coefficient on the acquisition dummy variable has the expected sign, Hypothesis #4b is not supported in this model. Pre-IPO acquisitions do not seem to have an effect on subsequent IPO underpricing. Other significant explanatory variables include the two stock exchange dummies, underwriter rank, IPO fraud, net income, NASDAQ 30-day return, and the hot IPO dummy variable. 260 Interestingly, when the 3-year delisting dummy is the dependent variable, the acquisition dummy variable is positive and significant. Firms conducting pre-IPO acquisitions are more likely to be delisted in the third year than non-acquiring IPO firms. Thus, Proposition #1b is not supported by our data as long-term performance appears to be negatively impacted by pre-IPO acquisition activity. The Effect of Acquisition Consideration and Payment Type For the next set of models, we reduce the sample size to the 87 acquiring firms and consider the effect of acquisition consideration and payment type on our four dependent variables. The results are presented in Table II. In our analyses, we find no support for Hypotheses 4, 5a, or 5b. In terms of ex-ante risk level, we find no significant relation for the acquisition consideration, cash, or stock variables. Thus neither acquisition size nor payment method appear to impact ex-ante risk levels. Auditor reputation, IPO fraud, CEO retained earnings, IPO date, and log of employees, conversely, are found to be significantly related to the BROAD measure. When considering IPO valuation, we find that acquisition consideration and method of payment are not significantly related to IPO proceeds. Larger acquisitions do not necessarily result in higher proceeds, and the effect of payment method is weak at best. Variables found to be significantly related to IPO valuation include the NASDAQ dummy, underwriter rank, total revenue, and the BROAD ex-ante risk measure. In terms of underpricing, we find a negative but insignificant relation for each of acquisition consideration, cash, and stock. Underwriter rank, IPO fraud, and IPO date are significantly positively related to IPO underpricing for our sample of acquiring firms. The model including the 3-year delisting dummy as the dependent variable reveals that acquisition size and cash payment are not statistically significantly related to later delisting. Significant relationships are found between the delisting dummy and each of net income, NASDAQ 30-day return, log of employees, and the hot IPO dummy variable. 261 Table I - The Effect of Acquisitions on the Dependent Variable List Dependent Variables N=435 2 BROAD (Adj. R = 0.3283) 2 VALUE (Adj. R = 0.4712) UPRICE (Adj. R2 = 0.4049) DELIST (Adj. R2 = 0.1250) b s.e. p-value b s.e. p-value B s.e. p-value b s.e. p-value INTERCEPT ACQ AMEX NASDAQ AUDREP UWRANK FRAUD CEORE FIRMAGE FCEO TOTREV NETINC N30DRET IPODATE LOGEES HOTIPO BUBBLE BROAD VALUE 6.4109 -0.3567 -0.1165 1.1158 0.005 0.069 0.771 -0.018 -0.080 -0.088 -0.001 0.000 -0.011 0.000 -1.209 ----------------------------------------- 1.027 0.194 1.129 0.424 0.646 0.075 0.179 0.008 0.015 0.156 0.001 0.000 0.009 0.000 0.248 ----------------------------------------- <0.0001 0.067 0.918 0.009 0.994 0.360 <0.0001 0.022 <0.0001 0.574 0.190 0.398 0.227 0.013 <0.0001 ----------------------------------------- 1.094 -0.008 -0.659 -0.343 -0.141 0.064 0.053 0.001 -0.004 -0.070 0.000 0.000 0.004 0.000 0.201 0.001 0.098 0.021 ----------- 0.141 0.025 0.142 0.054 0.082 0.009 0.024 0.001 0.002 0.020 0.000 0.000 0.001 0.000 0.032 0.001 0.026 0.006 ----------- <0.0001 0.752 <0.0001 <0.0001 0.085 <0.0001 0.027 0.417 0.036 0.001 0.607 0.963 0.000 0.002 <0.0001 0.102 0.000 0.001 ----------- -0.561 -0.036 0.309 0.166 -0.078 0.037 0.297 0.000 -0.003 -0.027 0.000 0.000 0.003 0.000 -0.045 -0.003 0.037 -0.009 0.167 0.168 0.028 0.163 0.063 0.092 0.011 0.027 0.001 0.002 0.022 0.000 0.000 0.001 0.000 0.038 0.001 0.029 0.007 0.055 0.001 0.190 0.060 0.009 0.396 0.001 <0.0001 0.929 0.134 0.227 0.976 0.084 0.019 0.540 0.238 0.001 0.213 0.191 0.003 1.519 0.525 -1.599 -0.228 -1.064 -0.186 -0.015 0.021 -0.032 -0.454 -0.001 -0.027 0.033 0.000 -0.379 0.016 -0.279 0.055 0.267 1.949 0.281 1.709 0.816 0.963 0.116 0.317 0.011 0.025 0.234 0.002 0.009 0.014 0.000 0.420 0.009 0.322 0.076 0.597 0.436 0.061 0.350 0.780 0.269 0.106 0.962 0.058 0.196 0.052 0.647 0.002 0.019 0.971 0.367 0.094 0.388 0.470 0.655 UPRICE ----------- ----------- ----------- ----------- ----------- ----------- ----------- ----------- ----------- -0.790 0.506 0.118 Table II - The Effect of Consideration and Payment Method on Dependent Variable List Dependent Variables N=87 BROAD (Adj. R2 = 0.4100) VALUE (Adj. R2 = 0.4041) UPRICE (Adj. R2 = 0.3465) DELIST (Adj. R2 = 0.2927) b s.e. p-value b s.e. p-value B s.e. p-value b s.e. p-value INTERCEPT CONSID CASH STOCK NASDAQ AUDREP UWRANK FRAUD CEORE FIRMAGE FCEO TOTREV NETINC N30DRET IPODATE LOGEES HOTIPO BUBBLE BROAD VALUE 9.370 0.001 0.337 0.997 0.428 -3.539 0.161 0.815 -0.037 -0.061 -0.284 -0.004 -0.006 -0.028 0.000 -1.312 ----------------------------------------- 3.109 0.005 0.681 0.679 1.002 1.701 0.170 0.396 0.017 0.037 0.352 0.005 0.007 0.018 0.000 0.520 ----------------------------------------- 0.004 0.895 0.622 0.146 0.671 0.041 0.346 0.043 0.027 0.107 0.422 0.493 0.427 0.117 0.089 0.014 ----------------------------------------- 1.268 0.001 0.031 -0.038 -0.289 -0.063 0.065 0.024 0.001 -0.005 0.012 0.002 -0.002 0.005 0.000 0.023 0.001 0.071 0.034 ----------- 0.546 0.001 0.111 0.111 0.162 0.290 0.028 0.068 0.003 0.006 0.058 0.001 0.001 0.003 0.000 0.089 0.002 0.076 0.020 ----------- 0.023 0.420 0.782 0.736 0.080 0.831 0.022 0.721 0.710 0.425 0.835 0.082 0.179 0.108 0.112 0.798 0.664 0.353 0.092 ----------- -1.621 -0.001 -0.117 -0.068 0.269 0.180 0.076 0.231 -0.001 -0.005 -0.010 0.000 0.001 0.004 0.000 0.037 0.000 0.087 0.017 0.134 0.638 0.001 0.125 0.125 0.187 0.326 0.032 0.076 0.003 0.007 0.065 0.001 0.001 0.004 0.000 0.101 0.002 0.086 0.023 0.136 0.013 0.360 0.353 0.590 0.155 0.584 0.021 0.004 0.734 0.462 0.883 0.664 0.507 0.248 0.061 0.717 0.896 0.315 0.461 0.330 -2.465 0.005 1.023 2.092 -1.469 8.726 -0.124 0.352 -0.021 -0.013 -0.598 0.001 -0.045 0.066 0.000 -1.892 0.046 -0.724 -0.024 -1.576 762.3 0.009 1.320 1.360 1.922 762.2 0.314 0.775 0.034 0.074 0.625 0.010 0.021 0.037 0.000 1.084 0.024 0.841 0.216 1.371 0.996 0.578 0.438 0.124 0.445 0.991 0.694 0.650 0.528 0.864 0.352 0.892 0.033 0.071 0.686 0.081 0.054 0.390 0.913 0.250 UPRICE ----------- ----------- ----------- ----------- ----------- ----------- ----------- ----------- ----------- -1.790 1.210 0.139 263 Conclusion Individually, acquisitions and initial public offerings are momentous transactions for a firm. Each may significantly improve or impair the prospects of a firm, and a plethora of research is devoted to documenting their effects. However, the combination of the two, as far as we can find, has been ignored in the literature. This paper begins to fill that gap by examining the acquisition activity by pre-IPO firms in order to examine the motives behind and the impact of such transactions, both pre- and post-IPO. Following previous literature in the M&A framework, we examine a few perspectives in our study: the synergy hypothesis, the managerial hubris hypothesis, and the agency hypothesis. The first argues that acquisitions are conducted pre-IPO in order to increase the value of the firm; while the latter two purport that the transactions are value-decreasing, either through managerial error or self-serving behavior, respectively. We further proposed that the method of payment is indicative of acquisition motives and that it will have an impact on post-acquisition performance. Competing theories makes forming a single proposition regarding the effect of the acquisitions on later IPO underpricing difficult, and thus it is left to empirical analysis. The results of our study suggest that engaging in pre-IPO acquisitions may be explained by both the synergy and managerial hubris hypotheses. We found that engaging in pre-IPO acquisitions allows firms to match the IPO valuations that other firms obtain through internal growth. This appears to support the synergy hypothesis. However, our analysis suggests that firms that engaged in pre-IPO acquisitions were then more likely to be delisted, for negative reasons, three years after the IPO. This appears to support the managerial hubris hypothesis. In combination, engaging in pre-IPO acquisition activity helps firms obtain more money in the IPO but then it appears that they are more likely to fail in three years. We also offer a new possibility for pre-IPO acquisition decisions. If the decisions are driven by the underwriters or VCs, then perhaps we should consider that it is the underwriters or VCs who are overconfident and thus it is underwriter or VC hubris. If the acquisitions are primarily done as a way to signal the firm’s fitness or legitimize the firm in the eyes of potential investors, which then leads to higher IPO valuations but not better long-term performance, then we should question the wisdom of this type of IPO polishing. In addition to testing the hypotheses we proposed in this article, investigating the effects of pre-IPO acquisition activity offers many other avenues worth researching. For example, in the pre-IPO period, why do firms make acquisitions? Our research suggests that some acquisitions are made to show rapid revenue growth while others appear to be efforts at product diversification. The two firms mentioned above appear to use acquisitions as a method of assembling a firm for an IPO, as they both appear to be a rapidly assembled collection of firms with similar or related products. Some pre-IPO acquisition activity significantly shifts the firm’s strategic direction. How do these different motivations affect the firm’s performance at the time of the IPO and its long-term performance? Additionally, how does pre-IPO acquisition activity affect post-IPO acquisition activity and subsequent firm performance? Firms do use the IPO proceeds for acquisitions. Are firms that make pre-IPO acquisitions better able to successfully conduct post-IPO acquisitions? If the Top Management Team is able to learn from their pre-IPO acquisitions, they may be more capable at the post-IPO acquisition activity, as suggested by Wiggenhorn and Madura (2004). We expect that our research will lead to such new and important avenues of investigation. 264 References Andrade, G., Mitchell, M., and Stafford, E., “New Evidence and Perspectives on Mergers,” Journal of Economic Perspectives, 15, (2001), 103-120. Asquith, P., “Merger Bids, Uncertainty, and Stockholder Returns,” Journal of Financial Economics, 11, (1983), 51-83. Asquith, P., Brunner, R., and Mullins, F., “Merger Returns and the Form of Financing,” Proceedings of the Seminar on the Analysis of Security Prices, 34, (1987), 115-146. Beatty, R.P., and Ritter, J.R., “Investment Banking, Reputation, and the Underpricing of Initial Public Offerings,” Journal of Financial Economics, 15, (1986), 213-32. Berkovitch, E., and Narayanan, M.P., “Motives for Takeovers: An Empirical Investigation,” Journal of Financial and Quantitative Analysis, 28, (1993), 347-362. Bradley, M., Desai, A., and Kim, E.H., “The Rationale Behind Interfirm Offers: Information or Synergy?" Journal of Financial Economics, 11, (1983), 183-206. Brown, K.C., Dittmar, A., and Servaes, H., “Roll-ups: Performance and Incentives for Industry-Consolidating IPOs.” American Finance Association Conference, New Orleans, July 2000. Brown, K.C., Dittmar, A., and Servaes, H., “Corporate Governance, Incentives and Industry Consolidations,” Review of Financial Studies, forthcoming, RFS Advance Access published on April 2, 2004. Chang, S., and Mais, E., “Managerial Motives and Merger Financing,” The Financial Review, 35, (2000), 139-152. Clarkson, P., and Merkley, J., “Ex Ante Uncertainty and the Pricing of Initial Public Offerings: Further Canadian Evidence,” Canadian Journal of Administrative Sciences, 11, (1994), 54-67. Firth, M., “Corporate Takeovers, Stockholder Returns and Executive Rewards,” Managerial and Decision Economics, 12, (1991), 421-428. Fishman, M.J., “Preemptive Bidding and the Role of the Medium of Exchange in Acquisitions,” Journal of Finance, 44, (1989), 41-58. Fuller, K., Netter, J., and Stegemoller, M., "What Do Returns to Acquiring Firms Tell Us? Evidence from Firms That Make Many Acquisitions," Journal of Finance, 57, (2002), 1763-1793. Hayward, M., and Hambrick, D., “Explaining the Premium Paid for Large Acquisitions: Evidence of CEO Hubris,” Administrative Science Quarterly, 42, (1997), 103-127. Jenkinson, T., Ljungqvist, A., Going Public: The Theory and Evidence on How Companies Raise Equity Finance, Oxford University Press: New York, 2001. Jensen, M.C., “Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers,” American Economic Review, 76, (1986), 323-329. Jensen, M. C., and Meckling, W., “Theory of the Firm: Managerial Behavior, Agency Costs, and Ownership Structure,” Journal of Financial Economics, 3, (1976), 305-350. Jensen, M.C., and Ruback, R.S., "The Market for Corporate Control: The Scientific Evidence," Journal of Financial Economics, 11, (1983), 5-50. Klein, A., “Can Investors Use the Prospectus to Price Initial Public Offerings?” Journal of Financial Statement Analysis, 2, (1996), 23-40. Larsson, R. and Finkelstein, S., “Integrating Strategic, Organizational, and Human Resource Perspectives on Mergers and Acquisitions: A Case Survey of Synergy Realization,” Organization Science, 10, (1999), 1-27. Linn, S.C., and Switzer, J.A., “Are Cash Acquisitions Associated with Better Post-Combination Operating Performance than Stock Acquisitions?” Journal of Banking and Finance, 25, (2001), 1113-1138. 265 Loughran, T., and Ritter, J.R., “Why Has IPO Underpricing Changed Over Time?” SSRN Working Paper, posted February 2004. Loughran, T., and Vijh, A.M., “Do Long-Term Shareholders Benefit from Corporate Acquisitions?” Journal of Finance, 52, (1997), 1765-1790. MacCrimmon, K., and Martens, M., “Predicting Post-IPO Performance From Ex-Ante Risk Measures,” In the proceedings of the Administrative Sciences Association of Canada Conference, London Ontario, 2001. Malatesta, P.H., “The Wealth Effect of Merger Activity and the Objective Functions of Merging Firms,” Journal of Financial Economics, 11, (1983), 155-182. Malmendier, U., and Tate, G.A., “CEO Overconfidence and Corporate Investment,” NBER Working Paper No. W10807, 2004. Mann, S.V., and Sicherman, N.W., “The Agency Costs of Free Cash Flow: Acquisition Activity and Equity Issues,” Journal of Business, 64, (1991), 213-227. Marris, R.L., The Economic Theory of Managerial Capitalism, London, UK: Macmillan, 1964. Martens, M.L., “IPO Effects: Corporate Restructuring When a Firm Goes Public,” Journal of Public Affairs, 4, (2004), 125S-139S. Melnik, A., and Thomas, D.C., “Value-Relevance of Accounting Information and the Predictability of IPO Underpricing,” Academy of Entrepreneurial Finance, Fourteenth Annual Conference, 2003. Mulherin, J.H., and Boone, A., “Comparing Acquisitions and Divestitures,” Journal of Corporate Finance, 6, (2000), 117-129. Pagano, M., Panetta, F., and Zingales, L., “Why Do Companies Go Public? An Empirical Analysis,” Journal of Finance, 53, (1998), 27-64. Ritter, J., “The Long-Run Performance of Initial Public Offerings,” Journal of Finance, 46, (1991), 3-27. Ritter, J., and Welch, I., “A Review of IPO Activity, Pricing, and Allocations,” Journal of Finance, 57, (2002), 1795-1828. Roll, R., “The Hubris Hypothesis of Corporate Takeovers,” Journal of Business, 59, (1986), 197-216. Schultz, P., and Zaman, M., “Do the Individuals Closest to Internet Firms Believe They are Overvalued?” Journal of Financial Economics, 59, (2001), 347-381. Servaes, H., "Tobin's q and the Gains from Takeovers." Journal of Finance, 46, (1991), 409-419. Seth A., Song K.P., and Pettit R., “Synergy, Managerialism or Hubris? An Empirical Examination of Motives for Foreign Acquisition of US Firms,” Journal of International Business Studies, 31, (2000), 387-405. Slusky, A.R., and Caves, R.E., “Synergy, Agency, and the Determinants of Premia Paid in Mergers,” Journal of Industrial Economics, 39, (1991), 277-296. Tinic, S.M., “Anatomy of Initial Public Offerings of Common Stock,” Journal of Finance, 43, (1988), 789-823. Trautwein, F., “Merger Motives and Merger Prescriptions,” Strategic Management Journal, 11, (1990), 283-295. Travlos, N.G. 1987. “Corporate Takeover Bids, Method of Payment, and Bidding Firms’ Stock Returns.” Journal of Finance, 42: 943-963. Wansley, J.W., Lane, W.R., and Yang, H.C., “Gains to Bidder Firms in Cash and Securities Transactions,” Financial Review, 22, (1987), 403-414. Welbourne, T. and Andrews, A., “Predicting the Performance of Initial Public Offerings: Should Human Resource Management be in the Equation,” Academy of Management Journal, 39, (1996), 891-919. Wiggenhorn, J., and Madura, J., “Performance of Acquisitions by Newly Public Firms,” Paper presented at the Financial Management Association Conference, New Orleans, LA, 2004. 266 Yi, J.H., “Pre-Offering Earnings and the Long-Run Performance of IPOs.” International Review of Financial Analysis, 10, (2001), 53-67. 267 ASAC 2005 Toronto, Ontario Taiyuan Wang (student) Richard Ivey School of Business The University of Western Ontario GENERIC STRATEGY AND CAPITAL STRUCTURE: THE IMPACTS OF PRODUCT DIFFERENTIATION ON FINANCIAL LEVERAGE This study examines the relationship between product differentiation and financial leverage. It argues that product differentiation is associated with investments in firm-specific assets that have lower liquidation value than general assets. Debt investors need to charge a higher interest rate to cover the risk of default, making the firm resort to equity financing and resulting in low financial leverage. Modiglian and Miller (1958) argued that firm value was independent of capital structure. This means that ways of financing (debt or equity) have no relationship to firm performance. On the contrary, Myers and Majluf (1984) pointed out that different means of financing reflected managers’ different decisions, which are influential in any firm’s performance. Other theory is needed to explain the inconsistence. Williamson (1988) pointed out that debt and equity could be considered as different means of governance. Equity investors can access the management of a firm by means of the board of directors. Debt investors have no right of access to the management, but they have priorities to claim against the firm’s assets when it goes bankrupt. Since special assets are more difficult to be redeployed, they have lower liquidation value than do general assets. As a result, debt investors are less likely to provide funds for firms that have more special assets. Prior studies found that some specific strategies would require investments in firm-specific assets and consequently made the firm difficult in borrowing (Balakrishnan and Fox, 1993; Jordan, Lowe et al., 1998; Kochhar and Hitt, 1998; Riahi-Belkaoui and Bannister, 1994; Vicente-Lorente, 2001. etc.). However, the impacts of generic strategies on capital structure have been ignored. According to the Resource-Based View, competitive advantages are built on firm-specific, rare, and inimitable resources (Barney, 1991). A successful differentiation strategy must involve firm-specific assets to enable the firm to achieve sustainable competitive advantage. At the same time, the specialty of firm-specific assets has a negative influence on debt borrowing capacity (Williamson, 1988). Therefore, the strategy of differentiation should influence capital structure, but there is no study concentrating on this research question. This paper fulfills this gap. It theoretically examines the relationship between the strategy of product differentiation and capital structure, and argues that product differentiation has negative impacts on financial leverage. The contribution of this study is twofold. Firstly, the proposition that product differentiation has negative impacts on financial leverage would be important to researchers, financial managers, and bankers. For researchers, it broadens the domain of research by linking capital structure with the strategy of differentiation; for financial managers, taking the account of strategy as a factor to determine capital structure is practically meaningful; and for bankers, examining a firm’s strategy may mean making better investment decisions. Secondly, it argues for an improvement in the measurement of strategic position by examining accumulated investments in particular assets. Prior studies used yearly investments in 268 specific assets to measure strategy (Balakrishnan and Fox, 1993; Jordan, Lowe et al., 1998; Kochhar and Hitt, 1998; Riahi-Belkaoui and Bannister, 1994), but yearly investments cannot reflect the stock of particular assets (O'Brien, 2003). At the same time, strategic position is less likely to change frequently. The accumulated investments, which can reflect the stock of particular assets, are more appropriate to represent strategic position. Theoretical Background Agency Theory and Debt Financing The interests of shareholders and creditors may be incongruent. Shareholders have the right to harvest the firm’s residual value, while creditors’ profit lies in the pre-determined interest rate. After obtaining funds from creditors, shareholders are inclined to invest in more profitable projects, which are riskier. If the riskier projects succeed, shareholders will harvest more, but creditors’ earnings will not change (Balakrishnan and Fox, 1993). Put differently, creditors bear risks with shareholders for a riskier project but, even taking more risks, they cannot share the potential profit with shareholders. The interests of shareholders and managers are not always consistent. The return to shareholders is determined by profitability, while the reward to managers is positively associated with firm size. Therefore, managers may try to invest as much as possible, probably resulting in some inferior investments (Jensen, 1986). Agency Theory (AT) can theoretically explain the conflict between shareholders and creditors as well as the inconsistency between shareholders and managers. Williamson (1988) argued that people were subject to bounded rationality and opportunistic, which can interpret why shareholders have the propensity to take riskier projects. Creditors recognize that shareholders could steal their wealth. They will increase the interest rate to cover the risk of default (Balakrishnan and Fox, 1993). Similarly, managers may concern their own benefits more than those of shareholders: they over-invest the surplus capital and make the firm as large as possible, seeking for the increase in salary, power, and social status (Jensen, 1986). Agency costs between shareholders and managers can be reduced by debt financing (Jensen, 1986). Debt generates interest expense and therefore decreases the slack of cash flow, making it difficult for managers to waste shareholders’ wealth. Further, a high debt ratio means high financial distress - which can increase the risk of default, but can force managers to work hard. The costs of bankruptcy to managers are dual: they will not only lose their current positions, but also lose the opportunity to find new jobs because their reputation will be impaired by the experience of bankruptcy (Balakrishnan and Fox, 1993). The mixture of strengths and weaknesses of debt financing implies that there should be an optimal debt ratio, but empirical investigations showed that this optimal debt ratio was not easy to determine (Shyam-Sunder and Myers, 1999). Some researchers argued that there was a ‘pecking order’ when firms make financing decisions: they are likely to use funds generated from operation; if the retained earnings from operation are insufficient, they will choose debt financing; and if the debt loan is still not enough, they will resort to equity financing (Myers and Majluf, 1984; Shyam-Sunder and Myers, 1999). 269 TCE, Asset Specificity, and Capital Structure Williamson (1988) theoretically linked capital structure and strategy. He argued that debt and equity could be considered alternative means of governance. Equity investors can access management by means of the board of directors; monitor managers’ behavior; and, if necessary, control managers’ strategic choices. From this perspective, equity is a kind of hierarchical control (Williamson, 1988). Debt investors cannot access management, but they have the preemptive right to claim against a firm’s assets if the firm goes bankrupt. Before providing a firm with capital, debt investors will evaluate the condition of the firm’s assets so that they can make appropriate investment decisions. Compared with equity, debt is a kind of marketing control (Williamson, 1988). A firm with more valuable firm-specific assets that are rare and inimitable may outperform its competitors (Barney, 1991), but firm-specific assets have lower liquidation value than do general assets that can be redeployed and reused. Because debt investors cannot access the management, they are unable to discern that the firm can perform better than its competitors. Consequently, debt investors are likely to increase interest rate to cover the risk of default. As a result, firms with more specific assets have to resort to equity financing, resulting in lower financial leverage. Theoretically, the specialty of firms’ assets and the difference of debt governance (market control) and equity governance (hierarchy control) establish a bridge between capital structure and strategy (Balakrishnan and Fox, 1993; Williamson, 1988). Prior Research on Capital Structure and Strategy Barton and Gordon (1987) ignited the studies of capital structure from a strategy perspective. They argued that capital structure was influenced by top managers’ priorities and behavioral characteristics. In their subsequent study (1988), they empirically concluded that both strategy and top managers’ behavioral characteristics could explain capital structure. In the past decade, more researchers joined in the studies of capital structure with a strategy paradigm (Balakrishnan and Fox, 1993; Jordan, Lowe et al., 1998; Kochhar and Hitt, 1998; O'Brien, 2003; Riahi-Belkaoui and Bannister, 1994; Vicente-Lorente, 2001. etc.). Most studies are based on the theory that strategic investments in firm-specific assets impair firms’ debt borrowing capacity. Investments in R&D are expected to influence firms’ borrowing capacity (Balakrishnan and Fox, 1993; Vicente-Lorente, 2001). Empirical investigations found a “negative and significant” relationship between R&D intensity and financial leverage (Balakrishnan and Fox, 1993, p. 12). Vicente-Lorente distinguished between firm-specific R&D and external R&D (out-sourcing) and found that firm-specific R&D intensity had negative influence on financial leverage, while external R&D had “no significant effect” (Vicente-Lorente, 2001, p. 170). Diversification and integration are considered to be associated with the change of capital structure because such investments need substantive external funds (Kochhar and Hitt, 1998). A firm that pursues related diversification or vertical integration is likely to look forward to the synergies between the current business and the potential business. This means that investments in related diversification or vertical integration bring more firm-specific assets than do unrelated diversification investments. Therefore, a firm pursuing related diversification or vertical integration will have lower financial leverage than if it had pursued unrelated diversification (Kochhar and Hitt, 1998; Riahi-Belkaoui and Bannister, 1994). 270 Differentiation as a Generic Strategy Generic strategy has been widely studied since the introduction of Porter (1980)’s typology: cost leadership, differentiation, and focus. A firm pursuing cost leadership makes every effort to reduce costs, so as to achieve a cost advantage over competitors. On the contrary, a firm taking a strategy of differentiation tries to create something special on product and/or service, making it unique in the industry so that the firm can charge a higher price than its competitors. Focus refers to the concentration on a niche market segment, regardless of pursuing cost leadership or differentiation. Commitment to one of the three as the primary target will result in better performance than if the firm had become stuck in the middle (Porter, 1980). Cost leadership cannot neglect the quality of products and services, and differentiation cannot ignore low cost operation (Porter, 1985). Differentiation brings more transaction costs than does cost leadership (Jones and Butler, 1988). Whichever strategy they take, firms need to make potential customers familiar with the characteristics of their products and services. Through differentiation, a firm makes its products and/or services more special and complex than if the firm had merely pursued cost leadership. Therefore, the firm needs to invest more to inform and train its customers about its differentiated products and/or services. To cover the transaction cost generated from differentiation, the firm can charge a higher price to achieve a higher margin, or increase the production volume to reduce production costs (Jones and Butler, 1988). Product Differentiation Differentiation can be achieved by employing some strategies that can be perceived by customers as benefits; cannot be imitated by competitors; and are valuable for business partners such as suppliers (Calorir and Ardisson, 1988). New product development, brand identification, and innovation in marketing and technology are main differentiation strategies (Dess and Davis, 1984). Priem (1992) noted that differentiation could be achieved by improving product features, product design, technology, customer service, and distribution network. To summarize, differentiation can be realized through product differentiation and marketing management (Miller, 1986). Product differentiation can be realized through product innovation, quality and reliability improvement, and customer service enhancement (Murray, 1988). Product innovation is viewed as a main differentiation strategy (Dess and Davis, 1984; Kotha and Vadlamani, 1995; Miller, 1988; White, 1986. etc.). In mature industries, customers are already familiar with product features. They are likely to choose standard products at a low price. Competitors have similar technology and production capability: they can easily imitate the innovated products (Murray, 1988). Under such a condition, the improvement on product quality, liability, and customer services becomes the source of sustainable advantage (Murray, 1988; Porter, 1985). Product differentiation and product innovation are different. Product innovation is widely considered as a way to improve product features (Miller, 1986), while product differentiation should include brand building, improvement in quality and reliability, and the strengthening of customer service (Murray, 1988). Put differently, a strategy of product differentiation is associated with product improvement, product advertisement, and customer service enhancement. Product improvement refers to the improvement in product features, quality, and reliability; product advertisement refers to those advertisements particularly designed for the brand building of products and/or services; and customer service enhancement refers to the improvement on customer service facilities, such as call-center, after sales service network, etc. 271 Murray (1988) argued that the total costs to a customer included price, the expense of configuration and repair, the inconvenience of poor performance, and the unhappiness due to product or service problems. High quality and reliable products reduce repair expenses. Good customer service can reduce customers’ unhappiness when they have problems. By means of improving product quality and customer service, a firm can lower the total costs to its customers, resulting in the achievement of customer loyalty, which is a competitive advantage. Further, the process of improvement on product quality and customer service is sufficiently complex to make imitation difficult (Murray, 1988). Therefore, such differentiation strategy can help the firm achieve sustainable advantages. Product Differentiation and Financial Leverage Measurements of Financial Leverage Capital structure is generally measured by financial leverage, which is defined as the ratio of debt over the sum of debt and equity (Jordan, Lowe et al., 1998). Although some studies used gearing (the ratio of debt over equity) (Kochhar and Hitt, 1998) to reflect capital structure, leverage were more widely accepted in prior studies (Balakrishnan and Fox, 1993; Jordan, Lowe et al., 1998; Vicente-Lorente, 2001. etc.). Albeit different in forms, leverage and gearing reflect similar information. The issue related to the measurement of financial leverage lies in the selection of debt (long-term or short-term) and equity (book value or market value). Because long-term debt can avoid the fluctuations of short-term financing (Barton and Gordon, 1988), and because strategic positions are not likely to change frequently, compared with short-term debt, long-term debt is more appropriate for reflecting strategic investments. Both book and market value equity are important factors in evaluating financial leverage. Book value can reflect the information of historical investments, while market value can reflect the economic rent generated from the firm-specific assets that have been invested before. Several measurements of financial leverage have been used in prior studies. Titman and Wessels (1988) used the ratio of long-term, short-term, and convertible debt over book and market value equity. Barton and Gordon (1988) used the ratio of owners’ equity over the difference of total assets and current liabilities. Balakrishnan et al (1993) defined financial leverage as the ratio of total debt over the sum of total debt and market value equity. Vicente-Lorente (2001) used two financial leverage variables, one employing book value equity and the other using market value equity. In this study, I define financial leverage as the ratio of long-term debt over the sum of long-term debt and equity, considering both book value and market value equity. Product Differentiation and Investments in Firm-Specific Assets Product differentiation requires investments in firm-specific assets. To improve product features, firms need to invest in R&D; to improve product quality and reliability, firms need to invest in quality management system such as Total Quality Management; to improve brand identification, firms need to invest in product advertisement; and to increase customer loyalty, firms need to invest in customer service facilities. These investments can be divided into three categories: the investments in product improvement on features and quality & reliability; the investments in product advertisement; and the investments in customer service. 272 To achieve product differentiation, a firm needs to continuously invest in the related assets. Temporary or occasional investments cannot help the firm achieve sustainable advantage (O'Brien, 2003). Historical or accumulated investments in the related assets are more appropriate for the reflection of a firm’s strategic position than temporary investments. The related assets generated from these investments are firm-specific. If a firm goes bankrupt, the liquidation value of these assets will have very low liquidation value. Consequently, the specificity of such assets will influence the firm’s financial leverage. The relationships between product differentiation and accumulated investments in product improvement, product advertisement, and customer service, and the relationships between such investments and financial leverage are shown in Figure 1. Figure 1 Product Differentiation and Financial Leverage Accumulated investments in product Improvement + + Product Differentiation Accumulated investments in product Advertisement _ _ Financial Leverage _ + Accumulated investments in Customer Service Proposition 1a: Product differentiation is positively associated with accumulated investments in product improvement. Proposition 1b: Product differentiation is positively associated with accumulated investments in product advertisement. Proposition 1c: Product differentiation is positively associated with accumulated investments in customer service. Product improvement requires the investments in firm-specific assets such as R&D (Jordan, Lowe et al., 1998). A firm pursuing product differentiation by means of continuous investments in product improvement will have a high ratio of firm-specific assets. The firm-specific assets will impair the firm’s debt borrowing capacity (Williamson, 1988). Consequently, the firm will seek for equity financing, resulting in lower financial leverage. Proposition 2a: Financial leverage is negatively associated with accumulated investments in product improvement. 273 On the one hand, the investments in product advertisement can be considered firm-specific. If a firm that has invested much capital in product advertisement goes bankrupt, debt investors will collect nothing from the invested advertisement: it is completely firm-specific. Therefore, debt investors will increase their interest rate to cover the risk of default (Williamson, 1988). On the other hand, continuous advertisements can help a firm build customer loyalty, which, in turn, improves debt investors’ safety factor of investment (Balakrishnan and Fox, 1993). Further, advertisement can be viewed as a signal showing that the firm will continue to operate in the market, resulting in the improvement of debt investors’ confidence in investing (Balakrishnan and Fox, 1993; Vicente-Lorente, 2001). The competing effects of the investments in product advertisement on debt loan capacity make it difficult to make a clear discernment. According to TCE (Williamson, 1988), it is the liquidation value that makes debt investors prefer general assets. From this perspective, liquidation value is negatively related to the difficulty of debt loan. The liquidation value of the investments in advertisement is likely to be extremely low, if not zero, if the firm goes bankrupt. Therefore, we can tentatively expect that the effects of the investments in product advertisement on financial leverage are more negative than positive. Proposition 2b: Financial leverage is negatively associated with accumulated investments in product advertisement. Customer service facilities such as call-centers, door-to-door service, international warranties, and customer clubs have been widely established in recent years. The investments in such customer service facilities are firm-specific. If a firm that has invested much capital in customer service facilities goes bankrupt, its debt investors will recover less than if the firm had had more general assets. Therefore, debt investors will request a high interest rate, making it difficult for the firm to borrow and resulting in low financial leverage. Proposition 2c: Financial leverage is negatively associated with accumulated investments in customer service. Conclusions and Implications Prior research that links strategy and capital structure mainly focused on specific strategies (Balakrishnan and Fox, 1993; Jordan, Lowe et al., 1998; Kochhar and Hitt, 1998; Riahi-Belkaoui and Bannister, 1994; Vicente-Lorente, 2001. etc.). Generic strategy has been widely studied in the field of strategy, but no study concentrates on the relationship between generic strategy and capital structure. This study fulfills the gap. It examines the impacts of product differentiation on capital structure and argues that product differentiation is related to product improvement, product advertisement, and customer service. Each aspect is necessary but not sufficient. Product improvement on product features and quality & reliability is the essence of differentiation. Product advertisement improves brand identification, which is a key element of product differentiation (Dess and Davis, 1984; Murray, 1988). It can also reduce transaction costs by means of providing customers with more information (Jones and Butler, 1988). Good customer service reduces total costs to customers (Murray, 1988), resulting in the improvement in customer loyalty. At the same time, the investments in product improvement, product advertisement, and customer service are firm-specific. Such investments will impair firms’ debt borrowing capacity, resulting in low financial leverage (Williamson, 1988). Therefore, a firm that pursues product 274 differentiation is likely to have lower financial leverage than does a similar firm pursuing cost leadership. The propositions of this study are useful for researchers, financial managers, and bankers. For researchers, earlier theorists (Modiglian and Miller, 1958) argued that firm value had no relationship with capital structure. This study articulates that product differentiation does have impacts on capital structure. It broadens the domain of the research for both strategy and finance fields. For financial managers, it suggests that strategy should be considered when they adjust capital structure. For bankers, understanding the influence of strategy on capital structure can help them make better investment decisions. The discrimination against firm-specific assets might make bankers lose good investment opportunities. An advantage of this study is the recommendation of using accumulated investments in product improvement, product advertisement, and customer service to reflect product differentiation. Prior studies measured specific strategies by examining yearly investments in particular assets (Balakrishnan and Fox, 1993; Jordan, Lowe et al., 1998; Kochhar and Hitt, 1998; Riahi-Belkaoui and Bannister, 1994), but yearly investments cannot reflect the stock of a firm’s particular assets (O'Brien, 2003). Therefore, accumulated investments in particular assets are more closely to reflect specific strategic position. A limitation of this study might be the measurements of product differentiation. Product advertisement is defined as an item of product differentiation, but it was viewed as marketing differentiation in prior research (Miller, 1986). Empirical investigation is needed to test the representation of product advertisement. Another potential limitation is the weak argument that product advertisement negatively influences financial leverage. Prior research, by arguing that the reputation effect is more significant, found the opposite relationship (Balakrishnan and Fox, 1993; Vicente-Lorente, 2001). Further theoretical explanation and empirical investigations are needed to dig out the real impacts of product advertisement on capital structure. It is not easy to “substantiate” the relationship between differentiation and financial leverage (Jordan, Lowe et al., 1998, p. 7). As a generic strategy, differentiation has been widely studied in the field of strategy. Capital structure is also a main topic in the field of finance. Any further study able to pinpoint a more profound relationship between the two fields would constitute a significant contribution. References Balakrishnan, Srinivasan & Fox, Isaac, "Asset Specificity, Firm Heterogeneity and Capital Structure," Strategic Management Journal, 14(1), (Jan 1993), 3-16. Barney, Jay B., "Firm Resources and Sustained Competitive Advantage," Journal of management, 17(1), (Mar 1991), 99-120. Barton, Sidney L. & Gordon, Paul J., "Corporate Strategy: Useful Perspective for the Study of Capital Structure?," Academy of Management. The Academy of Management Review, 12(1), (Jan 1987), 67-75. Barton, Sidney L. & Gordon, Paul J., "Corporate Strategy And Capital Structure," Strategic Management Journal, 9(6), (Nov/Dec 1988), 623-632. Calorir, R. & Ardisson, J. M., "Differentiation Strategies in 'Stalemate Industries'," Strategic Management Journal, 9(3), (May/Jun 1988), 255-269. 275 Dess, Gregory G. & Davis, Peter S., "Porter's (1980) Generic Strategies as Determinants of Strategic Group Membership and Organizational Performance," Academy of Management Journal, 27(3), (Sep 1984), 467-488. Jensen, Michael C., "Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers," The American Economic Review, 76(2), (May 1986), 323-329. Jones, Gareth R. & Butler, John E., "Costs, Revenue, and Business-Level Strategy," Academy of Management. The Academy of Management Review, 13(2), (Apr 1988), 202-213. Jordan, Judith, Lowe, Julian & Taylor, Peter, "Strategy and financial policy in UK small firms," Journal of Business Finance & Accounting, 25(1/2), (Jan-Mar 1998), 1-27. Kochhar, Rahul & Hitt, Michael A, "Research notes and communications linking corporate strategy to capital structure: Diversification strategy, type and source of financing," Strategic Management Journal, 19(6), (Jun 1998), 601-610. Kotha, Suresh & Vadlamani, Bhatt L, "Assessing generic strategies: An empirical investigation of two competing typologies in discrete manufacturing industries," Strategic Management Journal, 16(1), (Jan 1995), 75-83. Miller, Danny, "Configurations of Strategy and Structure: Towards a Synthesis," Strategic Management Journal, 7(3), (May/Jun 1986), 233-249. Miller, Danny, "Relating Porter's Business Strategies To Environment And Structure: Analysis and Perrormance Implications," Academy of Management Journal, 31(2), (Jun 1988), 280-308. Modiglian, F & Miller, M., "The cost of capital, corporation finance and the theory of investment.," The American Economic Review, 48(3), (Jun 1958), 261-297. Murray, Alan I., "A Contingency View Of Porter's 'Generic Strategies'," Academy of Management. The Academy of Management Review, 13(3), (Jul 1988), 390-400. Myers, Stewart C. & Majluf, Nicholas S., "Corporate Financing and Investment Decisions When Firms Have Information That Investors Do Not Have," Journal of Financial Economics, 13(2), (Jun 1984), 187-221. O'Brien, Jonathan P, "The capital structure implications of pursuing a strategy of innovation," Strategic Management Journal, 24(5), (May 2003), 415-431. Porter, Michael E., Competitive Advantages, New York: 1980. Porter, Michael E., Competitive Advantages: Creating and sustaining superior performance, New York: Free Press, 1985. Priem, Richard L., "An Application of Metric Conjoint Analysis for the Evaluation of Top Managers' Individual Strategic Decision Making Processes: A Research Note," Strategic Management Journal, 13(8), (Summer 1992), 143-151. Riahi-Belkaoui, Ahmed & Bannister, James W, "Multidivisional Structure and Capital Structure: The Contingency of Diversification Strategy," Managerial and Decision Economics, 15(3), (May/Jun 1994), 267-276. Shyam-Sunder, Lakshmi & Myers, Stewart C, "Testing static tradeoff against pecking order models of capital structure," Journal of Financial Economics, 51(2), (Feb 1999), 219-244. Titman, Sheridan & Wessels, Roberto, "The Determinants of Capital Structure Choice," The Journal of Finance, 43(1), (Mar 1988), 1-20. Vicente-Lorente, Jose David, "Specificity and opacity as resource-based determinants of capital structure: Evidence for Spanish manufacturing firms," Strategic Management Journal, 22(2), (Feb 2001), 157-177. White, Roderick E., "Generic Business Strategies, Organizational Context and Performance: An Empirical Investigation," Strategic Management Journal, 7(3), (May/Jun 1986), 217-231. Williamson, Oliver E., "Corporate Finance And Corporate Governance," The Journal of Finance, 43(3), (Jul 1988), 567-591. 276 ASAC 2005 Toronto, Ontario Jaana Woiceshyn Haskayne School of Business University of Calgary HOW ‘GOOD MINDS’ THINK: A STUDY OF CEO DECISION MAKING This study of 16 oil company CEOs, recognized as effective thinkers, found that they combined reason and intuition, primarily by relying on both explicit and ‘automatized’ principles. The CEOs also shared three thinking-related traits: self-awareness, certain motivation, and an active mind. Suggestions for effective thinking are derived from the findings. “I have a strongly held belief that intuitive or gut feelings are just pattern recognition, almost instantaneous pattern recognition, whereas logic is the more painstaking process of making a pattern emerge. I think that one is just as important as the other.” --A CEO participating in the study To probe how ‘good minds’ think, I asked 16 CEOs who ran (or had been running until recently) successful oil and gas companies to read a realistic decision scenario which presented three strategic alternatives: to invest in a new technology, to explore in the Arctic in a joint venture, or to acquire another oil company. The chief executives were then asked to think out loud how they would decide in the scenario. See Table I for the research methodology. This paper 1) explains how effective thinkers combined reason and intuition, 2) shows what principles they relied on, 3) describes the three common characteristics of effective thinkers, and 4) discusses implications of the study’s findings for those wanting to improve their thinking. Effective thinking Thinking is a process of asking and answering questions. Its effectiveness is tested when one is trying to solve a problem or make a decision. For example, how to design a new product? Who to hire to manage a new project? Where to drill for oil? All these decisions require thinking—and preferably thinking that is quick rather than slow, and leads to ‘right’ rather than ‘wrong’ answers. If thinking takes too long, the product might be obsolete when it is ready for a launch, or a competitor might have hired the best project manager, or bought the property with significant oil reserves. If the product design is ‘wrong’, there may be no buyers. If the ‘wrong’ manager is hired, the project may not succeed. If one drills a well in a ‘wrong’ area, no oil will be found. This suggests there are at least two dimensions of effective thinking: speed and finding a ‘right’ answer (cf. McGregor, 2002). A right answer is one that is consistent with facts (e.g., product qualities match demand, the project manager’s skills match the requirements of the job, oil is drilled where seismic analysis indicates oil). An answer consistent with facts is objective. 277 Table 1. How the study was done I asked nine experts knowledgeable of the oil industry (oil company CEOs, investment bankers, a financial analyst and a business reporter) to name effective thinkers among successful oil company Chief Executive Officers in Calgary, home of the second largest concentration of oil company headquarters in North-America. This led to a list of 72 names, 32 of which had been nominated twice or more. I requested participation of the 32 CEOs; 16 of them agreed. The companies these CEOs ran (or had run) ranged in size from about 1,000 to 300,000 boe/d (barrels of oil equivalent per day), with the median of 7,000 boe/d. Three of them were private. The Chief Executives had mainly technical degrees (engineering or geology); six also held MBAs. Two were women. The median oil industry experience was 24 years. To tap into the CEOs’ decision processes, I wrote a short, hypothetical decision-making scenario (available upon request). It was revised based on the comments of three knowledgeable industry observers, and ‘piloted’ with three oil company presidents (who were not part of the group of 16) to make it as realistic as possible (cf. Thomas, Clark & Gioia, 1993). The scenario involved an oil company CEO about to choose between strategic options for his firm and was used in a ‘think-out-aloud’ procedure in the interviews that averaged 90 minutes and were all taped. I analyzed the interview transcripts in a number of rounds, first grouping comments together by the question, then by similarity in their content. From these groups of comments, patterns about using intuition and reason and active minds started to emerge. For example, the use of various mid-range principles was directly observable in the interviews. I further integrated them into a handful of ‘grand principles’ discussed below (Eisenhardt, 1989; Miles & Huberman, 1984). However, thinking fast and objectively is not sufficient for effectiveness—one might be focusing on a wrong problem altogether. Effective thinking also requires asking and answering the right questions. Maybe a new product design is not what is needed for the business to succeed, but improvement of the efficiency of the production process instead. Maybe a new project and a manager are not needed. Maybe it would be better to acquire another oil company than to explore for more oil. Effective thinkers need to ask the right questions, and answer them fast and objectively. How do they do it? How the CEOs made decisions: reason and intuition Cognitive research has shown that experienced decision-makers in various fields--e.g., fire fighters, chess masters, pilots, business executives--do not rely solely on rational analysis but also utilize intuition in solving problems (Burke & Miller, 1999; Isenberg, 1984; Klein, 2001; Simon, 1987). How they combine the two is less well understood. The next sections describe how the CEOs used reason and intuition and combined them in the decision-making process (Figure 1). Reason. Reason is the faculty for thinking: it observes reality and integrates those observations into abstract conceptual knowledge, which it then applies to various concrete problems. Use of reason involves conscious cognitive processes. Three such processes manifested themselves in this study: essentialization and logic, judgment, and explicit use of principles. When dealing with the decision-making scenario, these executives focused on essentials: those factors 278 that fundamentally affected the long-term performance and survival of the company in the scenario. For example, the CEOs determined that the company needed to take significant action to sustain value creation for its shareholders. This enabled them to put aside less fundamental issues such as internal champions of and resistance to a new technology and post-acquisition integration—issues they knew they could control after the essential concerns had been addressed. Judgment is often considered an inexplicable process associated with intuition. However, here it refers to the conscious decision as to what problem to focus on, and as to when there is sufficient information and analysis to make a decision. Many of the CEOs related how they are able to make such judgments quickly, often faster than their peers. Many of them indeed attributed this to intuition or experience. However, I suggest that quick judgment follows from the decision-maker’s ability to see fundamental causes (essentialization) and to project consequences of decisions. When one has grasped the fundamental cause(s) of problems and projected consequences, the decision to act can be made quickly and confidently. The third conscious process was the explicit use of principles. For example, the rationality principle guided the CEOs to base their decisions on facts as opposed to emotions, the objectivity principle guided them to seek outside advice and prevented them from rushing to conclusions, and the productivity principle told them to focus their company’s activities on its strengths in order to create value. These and other principles were held explicitly—some CEOs told how they even kept check lists of principles—and consulted them in making decisions. Intuition. Intuition refers to subconscious cognitive processes which can be hard to observe. I was able to observe some of them, thanks mostly to the CEOs’ introspection. The CEOs relied on intuitive processes of pattern recognition, visualization, and what I call ‘automatized’ use of principles. These processes contributed primarily to the speed of decision-making. Pattern recognition refers to the instantaneous process of identifying events or situations as familiar. But for the subconscious mind being able to quickly locate such experiences in one’s ‘memory files’, they have to be filed by essentials. For example, ‘technology adoptions over which management team was divided’ is an essential categorization, whereas ‘technology adoptions during sunny weather’, is not. (Understanding why the former happened will help with further technology adoptions, whereas the latter does not provide any useful knowledge.) If the ‘filing’ has been done properly, the recognition of a new situation as familiar is instantaneous, and problem-solving can proceed faster. All the CEOs in this study commented how the decision scenario was familiar, and were able to identify feasible alternatives quickly. Visualization is another intuitive process. Its use varied among the CEOs, with some being strong visualizers and others less so. Those who used visualization did it to project different decision alternatives and solutions. This was a quick way for them to concretize alternatives and assess their feasibility, again enhancing the speed in which they could handle the decision. I label automatized use of principles as a subconscious, intuitive process, although one could argue that it really lies in between conscious and subconscious activity. ‘Automatized’ here refers to principles that have been internalized to such an extent that a conscious checklist is not needed and their use is almost routine, not unlike driving a car. Many of the principles guiding the thinking or decision-making process itself (as opposed to the substance of the decision) were like this. For example, the principles of questioning the quality of information, of not using emotions as decision tools, and of not latching onto the first solution that looks feasible, were held almost as automatized (but recognized by the CEOs who were introspective of their thinking processes). 279 Figure 1. Framework for effective thinking Effectiveness: • speed • objectivity in asking and answering the right questions INTUITION • pattern recognition • visualization • ‘automatized’ use of principles* REASON • essentialization and logic • judgment • explicit use of principles* ACTIVE MIND • reality focus: facts vs. emotions • question asking • varied interests MOTIVATION SELF-AWARENESS *rationality, productivity, first-handedness, justice, self-interest, honesty Combining reason and intuition. How did the chief executives rely on both reason and intuition? The participants of the study exhibited a certain common pattern in approaching the decision scenario: • Instead of methodical analysis of the three proposed alternatives, they zoomed in quickly on what they considered the most feasible ones. They were able to do this as they recognized the pattern of the problem as familiar and then defined the problem by essentialization (identification of fundamental causes). For example, most of the CEOs rejected the option of exploring in the Arctic as something they had contemplated before and as too risky with too distant a payoff, given the company in the scenario was publicly traded with a short reserve life index and limited resources. • The CEOs then focused on the feasible alternative(s) and rejected the non-feasible ones by using principles such as value creation explicitly to project an alternative’s consequences, but also by visualizing. For example, the value creation principle was used as a criterion for the new technology option (if the technology could not increase production beyond initial projection, i.e., create meaningful value, it was rejected). Also, the alternatives visualized as feasible were adopted. • The decision-makers analyzed the selected alternatives with the help of automatized and explicit principles, judgment, and visualization. For example, if the quality of information about an alternative was not sufficient, the alternative was put on hold. • Finally, they developed additional alternatives, or combinations of alternatives using essentialization as well as visualization. For example, once the CEOs had identified fundamental causes for the company’s problems, they were able to develop a hierarchy of alternatives, addressing both short-term and long-term performance of the firm. Visualization of the alternatives also helped to put them in a hierarchy or a sequence. Using principles in decision-making. Principles are important decision tools used by strategic thinkers to cope with complexity (Keelin & Arnold, 2002), yet they have not been 280 addressed much by those who study thinking. As discussed above, principles were used by the chief executives in this study, both explicitly and ‘automatically’, to guide decision-making. Here I will discuss the role of principles in cognition in general, as well as the specific principles used by the executives in this study. Role of principles in cognition. Principles are generalizations reached by induction and used as guidelines in decision-making (Peikoff, 1989). The need for them arises as people do not have automatic knowledge as to what goals to choose nor how to achieve them, yet many human choices are complex and can have far-reaching consequences that are not immediately visible. Principles guide decision-makers through complex choices so that long-term goals can be reached, by helping to project the future and to choose between concrete alternatives in any given situation. For example, the principles of nutrition help project the long-term consequences of various nutritional choices as well as choose what to buy at the grocery store for supper. The principles of competitive strategy help project the consequences of various strategic options, and to choose which particular strategies to implement in the face of competition. Some principles can be almost ‘automatic’: they are so internalized that one does not need to consciously apply them. For example, the ethical principle of honesty is so ingrained in many people that they don’t even have to think whether to hand back the excess change that a cashier mistakenly gives them or whether to return, unread, the confidential competitive information that accidentally ends in their mailbox; they do it ‘automatically’. Examples of such automatized principles in this study were the mental ‘standing orders’ about the decision-making process that the CEOs followed, e.g. “Question the quality of the information” and “Do not to latch onto the first feasible sounding alternative.” Other principles are more explicit and abstract, and thus warrant to be placed on one’s checklist. An example in this study is the value creation principle which screens out alternatives that dilute or do not create value for the shareholders. Principles used by the CEOs. Many principles the CEOs in this study used were immediately observable or were described by them. Such principles tend to be quite concrete or narrow in their application. I call them ‘mid-range’. For the purposes of integration, I grouped them into more abstract categories which I call ‘grand’ principles: rationality, productivity, first-handedness, justice, self-interest and honesty. They are described here, showing how they were integrated from the various mid-range principles (see Table 2). This is a descending order, starting with the principles most frequently mentioned (all were mentioned in most interviews). Rationality guides one to use reason (i.e., to adhere to facts by the means of observation and logic), as opposed to emotions, mystic insights, or majority opinion. The decision-makers in the study strongly subscribed to this principle. They emphasized the focus on facts and keeping emotions in check when making decisions. In the words of one of the CEOs: “This business has the highest highs and the lowest lows and if you are not excited about what you do and a little bit emotional, it is a tougher business. But at the same time you have just got to pull back and say: ‘What are the facts, Jack?’” They also said their religious beliefs, if they had them, did not influence their decisions. The CEOs kept asking questions to get to facts and to establish connections between them. Finally, they emphasized ‘having done your homework’ before leaping, i.e., having a good grasp of the relevant facts before making a decision: The principle of rationality also manifested itself in the CEOs’ strong desire for objectivity. They sought outside opinion (from the board members, consultants, investment bankers) to ensure they did not get blind-sided by their own or their management team’s possible bias. For the same reason, they wanted to have teams with diverse backgrounds and opinions, and ‘embraced the skeptics’ within their organization. These executives also emphasized disciplined thinking and 281 continuously assessed the quality of information they used. Although they made decisions quickly, they insisted considering multiple options and not latching onto the first one too soon, without substantiation by evidence. One CEO commented: “I would take that step back to make sure that these are the only options that I should be considering. In that process you will probably come up with another option.” Table 2. ‘Grand’ Rationality Principles Mid-range - Focus on facts principles - Homework/analysis before deciding - Objectivity: ● Seek outside opinion ● Use diverse teams ● Embrace skeptics ● Disciplined thinking ● Check quality of information ● No rushing ‘Grand’ Justice Principles Mid-range - Hire best people and reward them principles - Honest criticism - Terminate nonperformers - The Golden Rule - Fairness - Respect - Justice to oneself: Accountability Principles used by the CEOs Productivity First-Handedness - Value creation - Building/creating something meaningful - Focus on the company’s goals and strengths - Commitment - Manage/reduce risk (risk threatens value creation) - Work hard - Consult others – but decide on your own - Control your destiny (majority stakes in projects) - Hire and seek advice from independent thinkers - Confidence in your own vision and ability to solve problems Honesty Self-Interest - No sacrifice - Passion/having fun - Making money - Long-term perspective - Balance in life (work vs. other values) - Not faking reality to gain values - Not only towards other people but towards oneself The principle of productivity guides one to create material values; this stood out as important in every CEO’s interview. They assessed each alternative with the question: Does this create value for the shareholders? And if the answer was negative, they rejected the option. The principles of productivity also manifested itself in the CEOs’ goal of building or creating a successful business. While they were motivated by the financial returns, virtually all of them said that was not enough to keep them working. They derived a sense of purpose from building their businesses. “I just want to build something that you are proud to run and build and create some value and good for your shareholders and for your own family,” said one CEO. This goal of ‘creating something meaningful’ was supported by the principle of focus, or the idea that the company needs to select its goals carefully, and match those with its resources and strengths. The CEOs also emphasized commitment to these long-term goals through hard work, and protecting the value created by managing risk. 282 The principle of first-handedness guides one to do one’s own thinking, as opposed to following others. The 16 CEOs demonstrated this in many ways. They wanted to consult others but process what they had learned themselves, and decide on their own. To quote one of the CEOs: “You have a discussion, it is open and frank, and let everybody say what they have in their mind. But ultimately it is not a democracy; I will make the decision at that point in time.” They also sought to ‘control their destiny’ by retaining majority stakes in projects or by being the sole owners and operators. They resisted market pressures for maximizing quarterly profits, even at public companies, being confident in the long-term vision and focus they had set for their firms. These CEOs did not feel threatened by others’ independent thinking, either. Quite the contrary, they wanted independent thinkers as employees and advisors, and felt confident that they were capable of assessing others’ conclusions. The principle of justice guides one to evaluate other people objectively—critically important, given that so much of business success is based on social interaction. As the first-handedness principle indicated, the CEOs in this study did not seek recognition from others. Instead, they sought ability, talent and hard work. They recognized that productivity requires the combined efforts of many people, and strove to hire and retain the best people they could. The justice principle manifested itself primarily in two ways: how the CEOs evaluated others, and how they treated them accordingly. They sought the best people (i.e., those who were talented, hard-working, independent, accountable) and then rewarded them with autonomy and generous compensation. The tougher aspect of justice was treatment of those who did not live up to expectations. Some offered honest criticism first, and when it did not lead to an improvement, terminated those employees who did not contribute to the company in a positive way. Most of the CEOs found these dismissal decisions the toughest ones they had to make but yet necessary, and just to both the employee and the company. There was no ‘egalitarianism’ in these CEOs’ applying justice: everybody was not treated equally regardless of conduct. Those who helped create value were rewarded, and those who did not, were reprimanded or dismissed. Neither was there room for playing favorites: people were assessed on merit, not on connections or relationships. Many CEOs also subscribed to the Golden Rule (treating others as you would like to be treated), and the principles of fairness and respect. Justice extended into the CEOs’ assessment and treatment of themselves. They held themselves accountable for their decisions, admitted errors, and took and granted credit appropriately. All the CEOs were driven by the principle of self-interest. They did not come to work and run their companies from a sense of duty or sacrifice. Rather, they emphasized the enjoyment of or even passion for their work. One said, reflecting the sentiment of many: “I love the business. You want to make a bunch of money through the process, so obviously that is a motivating factor but more so for myself, I love the business, I love coming to work every day. So I’ll call that fun.” Many cited the intellectual challenge, and the satisfaction from solving the puzzles and problems they encountered. Making money was considered an important motivator but not sufficient by itself; work also needed to offer challenge, enjoyment and purpose. The pursuit of self-interest did not consist of single-minded focus on work although it was a central value. Even if the CEOs “lived to work”, they had lives outside of it. They all had families and spent time with them. Those with no young children had other interests ranging from carpentry to community work, but even in these activities they selected those that they enjoyed or believed in. Some explicitly advocated balance in life, even if that balance came in shifts: periods of more 283 intense focus on work when that was required, alternating with periods where there was more emphasis on outside things. The principle of honesty guides one not to fake reality in order to gain a value. The 16 CEOs considered it a given; when asked about ethical principles, it was usually raised first. By honesty, they meant remaining true to reality, not just in communicating with others but in your own thinking. This view of honesty contrast with those who suggest that a little dishonesty is necessary to get ahead in business and is to be expected (Carr, 1968). Common characteristics of effective thinkers The CEOs participating shared three inter-related characteristics that seem to facilitate effective thinking: a certain motivation, self-awareness and an active mind—all of which can be cultivated with some effort and introspection. All the CEOs showed a certain motivation. They all loved their business; they were not CEOs of companies because their parents had wanted them to go into the oil business but because they found the industry fascinating. They started and ran oil companies because they wanted to, and they wanted their companies to succeed. This motivated their thinking: they wanted to solve problems and find ever better solutions. They were motivated to find out what were the right questions to ask; they were motivated to find out what the right (objective) answers were; and they were motivated to think fast, to find solutions before their competitors did. Being motivated to succeed made the CEOs introspect: how do I do it? Many of the CEOs were also very aware of their own decision-making approach. Nobody had any trouble thinking out loud how they would handle the decision-making scenario. Many recited the steps of their generic decision-making approach and referred to principles they used. The CEOs were able to analyze why they were effective, although this ability varied somewhat. For example, some referred to either experience or intuition as their source of effective thinking, whereas others elaborated how they bolstered their thinking by defining problems correctly, visualizing alternatives, or recognizing familiar patterns to aid decision-making. The motivation to succeed and self-awareness of the CEOs helped cultivate the third characteristic facilitating effective thinking: an active mind. An active mind is curious; it’s engaged with facts, and constantly processes them in search for the ‘right answers.’ The CEOs immediately engaged with the scenario (i.e., ‘facts’ for the purposes of the exercise), and they also emphasized the importance of focusing on reality and not deciding on emotion. As they were thinking out aloud, they were asking a lot of questions, for more information but also as a habitual thinking process, to check their assumptions and the quality of the information available. Besides the reality focus and continuous question asking, the CEOs’ active minds were evidenced by their varied interests. For example, most were avid readers, of fiction, history, biographies, of technical and business-related material. One executive read a number of newspapers daily. Those who did not read a lot (i.e., beyond the required technical and business material) cited lack of time and importance of other priorities, such a family. Many were keen travelers, and participated in outdoor activities and sports. 284 Lessons for effective thinking This study’s findings suggest a number of ways in which executives, managers or other business people can cultivate ‘good minds’, or at least improve the effectiveness of their thinking: • • • • • Cultivate an active mind. Ask a lot of questions, being focused on reality and seeking the facts behind problems and issues. Having interests outside of work can also help. Introspect. Being aware of you own thinking process facilitates its effectiveness. You can catch yourself when emotions try to take over, or when you are tempted to evade a problem. By introspecting, you can also guide yourself away from dead ends. Find and follow your passion. Thinking is difficult when you are not motivated, and the opposite is true when you are doing it in relation to something you enjoy. Being motivated to solve problems is the first step in effective thinking. Use intuition to speed up decision-making. In order to do this, you need to file things properly, i.e., by essentials, in your subconscious. Your subconscious will not be able to offer you patterns, if you have filed “technology adoptions during sunny weather” as opposed to “successful technology adoptions”. Some people are able to use intuition to visualize decisions and their consequences; and most people are able to internalize certain decision principles so that their use becomes a second nature, or almost automatic. Identify and apply principles. When you encounter a problem, ask yourself what principles are involved. Value creation? Objectivity? Honesty? Justice? Or something else? (You can induce principles from observation by generalizing). And how does the principle apply to the problem? The more one identifies and applies principles, the easier it becomes—and the more effective one’s decision making. References Burke, L.A. & Miller, M.K. “Taking the mystery out of intuitive decision making,” Academy of Management Executive, 13:4, (November 1999), 91-99. Carr, A. “Is business bluffing ethical?” Harvard Business Review, (January/February1968). Eisenhardt, K.M. Building theories from case study research. Academy of Management Review, 14:4, (1989) 532-550. Isenberg, D.J. “How senior managers think,” Harvard Business Review, (November-December 1984), 81-90. Keelin, T. & Arnold, R. “Five habits of highly strategic thinkers,” Journal of Business Strategy, 23:5, (2002), 38-42. Klein, G., Sources of power: How people make decisions, Cambridge, MA: MIT Press, 2001. McGregor, L. “Improving the quality and speed of decision making,” Journal of Change Management, 2:4, (2002), 344-356. Miles, M.B. & Huberman, M.A. 1984. Qualitative data analysis: A sourcebook of new methods. Newbury Park, CA: Sage. Peikoff, L. “Why should one act on principle?” Intellectual Activist, 4:20, (February 1989), 2-6. Simon, H. A. “Making management decisions: The role of intuition and emotion,” The Academy of Management Executive, 1:1, (February 1987), 57-64. Thomas, J.B., Clark, S.M.& Gioia, D.A. “Strategic sensemaking and organizational performance: Linkages among scanning, interpretation, action, and outcomes,” Academy of Management Journal, 36:2, (1993), 239-270. 285 ASAC 2005 Toronto, Ontario Samia Belaounia Professeur Rouen Graduate Business School D’UNE APPROCHE DETERMINISTE A UNE APPROCHE CONTINGENTE DU LIEN « DIVERSIFICATION –PERFORMANCE » : QUEL IMPACT SUR LA VALIDITE DE CONTENU DES INDICATEURS « CLASSIQUES » DE LA DIVERSIFICATION ? SUMMARY Une large part du management stratégique concerne l’impact de la diversification sur la performance. La plupart des travaux empiriques (notamment ceux des années 80) visent à montrer, dans la perspective de Rumelt (1974), que la stratégie de diversification liée, basée sur la recherche de synergies opérationnelles, génère une performance économique supérieure à celle du conglomérat. Ces études comparent les stratégies de diversification sur la base des principaux ratios de performance, en contrôlant l’effet de la différenciation et de la structure des industries concernées sur les écarts de prix et de rentabilité. Postulant l’avantage concurrentiel de la diversification liée, l’approche de Rumelt est donc déterministe. Toutefois, les résultats empiriques contradictoires obtenus amènent à s’interroger sur sa pertinence, en particulier face à une perspective contingente. Celle-ci prend en effet en compte les coûts administratifs liés à la recherche des synergies. La diversification liée suscitant les synergies mais aussi les coûts les plus élevés, elle a un impact indéterminé a priori sur la rentabilité. En s’appuyant sur une approche systémique de l’entreprise diversifiée, la perspective contingente éclaire de plus sur les modalités permettant à la diversification liée de créer de la valeur. Toutefois, les résultats contradictoires des études empiriques peuvent renvoyer également aux faiblesses des indicateurs de la stratégie de diversification ou à leur divergence qui empêche de comparer les résultats obtenus par des travaux n’exploitant pas les mêmes mesures. Dans un premier temps, sont présentés les coûts administratifs de la diversification liée, dans une perspective contingente du lien « diversification – performance ». Nous partons pour cela de la conception systémique de l’entreprise multi-produit propre à Thompson (1967) Puis, après un passage en revue des indicateurs de la diversification, nous montrons l’incompatibilité de ces derniers avec une approche contingente combinée à une perspective du portefeuille d’activités en termes de ressources, dans la lignée de Penrose (1957). Cela amène à définir la stratégie de diversification à partir des liens réellement mis en place entre les activités au sein de l’organisation. Nous dégageons enfin l’intérêt d’un recours à la structure formelle, à partir des huit études de cas réalisées. Contrairement à Wiersema [2003], le construit des indicateurs n’est pas au centre de la discussion sur la validité de contenu. Ce travail vise à montrer le décalage existant entre données externes et internes pour caractériser la stratégie de diversification, sur le mode de Nayyar [1992] mais en prenant en compte les paramètres structurels. Celui-ci procède par le biais de questionnaires. Or, le recours à la structure organisationnelle gagne en objectivité et en finesse dans la caractérisation des liens inter-métiers. De plus, les entretiens semi-directifs font accéder à des dimensions qui auraient pu être ignorées par le regard de l’observateur extérieur. 286 ASAC 2005 Denise Ghanam Toronto, Ontario Bristol Business School, UK (student) Odette School of Business, University of Windsor DYNAMIC CAPABILITIES: THE STRATEGY-HRM INTEREST SUMMARY Much is written on current trends in strategic HRM, focusing on the new role of HR, and its relationship to strategy. Issues like globalization, increased competition, and technology have spawned a growing emphasis on the role of the individual employee and teams. Effective use of human capital has become a valued source for competitive advantage, as well as a unique and proprietary organizational asset. As firms realize they need innovation and information are for success, and these attributes come most often from their people, focus moves to the connection between the firm’s strategy and its HR practices and policies. Regardless of how one views strategy, or what school of thought guides planning and formulation, it is people that are critical to those processes. People develop strategy and vision, people create opportunities for success, and people implement the strategic plans. While the meaning of HRM is debated, typically a more strategic approach to managing people encompasses what follows. Employees are an asset for achieving competitive advantage; effective planning is utilized; and employment policies and practices are aligned in a coherent fashion (internal integration) with business strategy (external integration) to be proactive. This strategic HRM view ties together several schools of strategy, in that learning processes, planning processes, design processes and cultural processes are connected to a configurational view of strategy and its fit inside and outside the firm. As well, there are overtones of the positioning and environmental schools as well. Michael Porter, even as he embodies positioning, reinforces the importance of unique processes and intangible relationships for transfer of knowledge between business units to provide competitive advantage through value chains. It is this multi-dimensional cogency which must guide future work in the strategy area, in order for it to be truly useful. Although this paper functions within a theoretical stance of strategic management, its focal point remains the human factor. The more integrated HRM is with overall business strategy, the more it reflects a unique way of managing employees that embodies the essence of the dynamic capabilities approach. The overlap of the literature of HRM with that of strategy most often incorporates a resource-based view (RBV) of human capital as value added, unique, difficult for competitors to imitate, and non-substitutable. This dynamic capabilities view is the vital point at which strategy and human resources intersect to ensure innovation and change. Human capital theory emphasises the importance of individuals, considering them as investors in their own education and training to achieve future benefits. Thus, intellectual capital is an asset, similar to physical or financial assets. Management of intangibles should stress that the total intellectual capital of the firm (its dynamic capabilities) can be increased either by leveraging the level of human, structural, and relational capital or by promoting the interactions between these interlinked groups. To be viable, existing managerial processes must align with an intellectual capital system through effective HRM to become the firm’s dynamic capabilities. This is a useful paradigm through which to view the strategy-HRM connection and has implications on a wide range of strategic firm decisions. 287 ASAC 2005 Toronto, Ontario Jean D. Kabongo Virginia State University THE DYNAMICS OF RESIDUAL REVALORIZATION: A CASE STUDY IN THE CANADIAN ENTERPRISES SUMMARY Dynamics of residual revalorization are a critical issue in the development of industrial ecology as a field of research. Studies relating to these considerations have been so far directed more toward the description of the overall conditions and the implementation of technology innovations. Few empirical researches have attempted to identify and to analyze the underlying causes of change and growth of the recovery and the transformation of discarded waste industry. To address this issue, fifty-six in-depth interviews with Canadian managers in charge of revalorization in twelve industrial firms were carried out. Findings of the study show that faster growth and opportunities for the firms in the particular revalorization industry rely basically on six types of dynamics perceived as supply irregularity, waste materials asymmetry, process enrichment and enlargement, green partnership, indefiniteness, and ecological managerial values. These identified dynamics that likely impact industrial businesses go beyond the institutionalized technology implementation and modeling. The results of the study are presented and discussed; perspectives for further researches are open. 288 ASAC 2005 Toronto, Ontario John R. Phillips (Student) Richard Ivey School of Business University of Western Ontario CEO MORAL CAPITAL AND STRATEGIC LEADERSHIP IN TURBULENT TIMES SUMMARY I build on the existing view that strategic leadership involves the effective management of human and social capital (Hitt & Ireland, 2002) by incorporating the concept of moral capital: CEO moral capital is the expectation that the CEO will justly* balance the disparate interests of individual and group stakeholders; applied as material and symbolic investments to achieve positive returns that benefit the firm, its stakeholders and the CEO. CEO moral capital is a particularly valuable resource in a developing socio-culturally turbulent environment. It generates cooperation in social networks and contributes to competitive advantage and superior firm performance. I build on the sociology of Pierre Bourdieu to introduce the concept of moral spheres of influence and I present five propositions: Proposition 1: CEO moral capital is positively related to firm performance. Proposition 2: To the extent that CEO moral capital is aligned with the communal morality of the firm moral sphere it has a positive effect on firm performance. Proposition 3: To the extent that CEO moral capital is aligned with the communal morality of the moral sphere of economic production it has a positive effect on firm performance. Proposition 4: To the extent that CEO moral capital is aligned with the communal morality of the moral sphere of wider society it has a positive effect on firm performance. Proposition 5: The effect of CEO moral capital on firm performance is negatively related to the social distance between the CEO moral sphere and the aligned moral sphere. I emphasize that CEOs have a large job ahead of them to achieve competitive advantage and superior firm performance in the developing socio-culturally turbulent environment. To create firm value, it appears they need to become more proficient at distributing the value created by the firm. Materially and symbolically, CEO’s need to demonstrate that they justly balance the interests of all individual and group stakeholders. This requires that they develop moral capital. * Justly = lawfully and equitably distributes, corrects and reciprocates 289 ASAC 2005 Toronto, Ontario Huanglin Wang (student) Ivey School of Business University of Western Ontario AN ANALYSIS OF THE RELATIONSHIP BETWEEN ALLIANCE NETWORKS AND FIRM PERFORMANCE SUMMARY From a dyad level of analysis, the relational view provides a new perspective in strategy; that is, the links that a firm has with other firms and its embeddedness in the networks that it has with these firms, affects firm performance and competitive advantage. This approach suggests that the source of competitive advantage may lie beyond a firm’s boundaries and in interfirm collaboration. This is contrary to what is suggested by the resource based view which is focused on intrafirm factors. Therefore, firms making relational specific investments may gain competitive advantage. Research has supported the conclusion that firms benefit from alliance networks on survival, profitability, and so on. But there has not been systematic study to explore the mediating factors in this relationship. This paper fills this gap and identifies interorganizational learning and interoganizational trust between the partners as two of the mediators in this relationship. Both the number and the properties of network ties can facilitate interorganizational learning. A central location in a network brings more linkages and external learning resources for a firm, direct ties that a firms has in alliance networks are efficient and effective in facilitating learning; therefore, alliance networks promote interorganizational learning. On the other hand, risk sharing, interdependence, asset specificity and reputation that are developed among the firms in alliance networks may all produce interorganizational trust. Both interorganizational learning and interoganizational trust are positively related to firm performance. Firms must not overemphasize the benefits of alliance networks and neglect to develop intrafirm competencies. Interfirm relationships, on the one hand, are sources of competitive advantage because competitors cannot replicate these relationships, but on the other hand, they constrain firms from adapting to a new environment. Firms should keep both independent capabilities and develop interfirm relationships to create effective dual sourcing orientations at the same time to consistently perform well. Future research areas include the other mediators between alliance networks and firm performance, the moderators that affect this relationship, the effect of structures of alliance networks and the firms’ positions in alliance networks on firm performance, and the management of the alliance networks that a firm is involved. These questions are important for current strategy research. The key reason why they are important is there appears to be a trend toward alliance networks as the basis for achieving sustainable competitive performance. Research regarding these alliance networks is needed to gain an understanding of this new form of competition. 290