Evolution des prix Internet 2010-2011

Département fédéral de l’Environnement, des
Transports, de l’Énergie et de la Communication DETEC
Office fédéral de la communication OFCOM
Division services de télécommunication
Bienne, novembre 2011
Prix de détail des services à large bande (DSL
et câble) : comparatif et évolution des
produits les meilleur marché
2010-2011
Table des matières
Vue d’ensemble
3
Résultats indiciels par profil d’usager
5
Prix réels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5
Prix normalisés . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5
Résultats indiciels par profil d’usager et positionnement des fournisseurs de services
7
Prix réels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7
Prix normalisés . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9
A Produits considérés (2011)
11
B Produits les meilleur marché exprimés en CHF (2011)
12
B.1 Produits les meilleur marché – prix réels . . . . . . . . . . . . . . . . . . . . . . . . . . . .
12
B.2 Produits les meilleur marché – prix normalisés . . . . . . . . . . . . . . . . . . . . . . . .
13
C Méthodologie de l’OCDE
14
Liste des tableaux
1
Profils d’usager des services à large bande . . . . . . . . . . . . . . . . . . . . . . . . . .
4
2
Indices de prix selon les profils d’usagers – prix réels . . . . . . . . . . . . . . . . . . . . .
5
3
Indices de prix selon les profils d’usagers – prix normalisés . . . . . . . . . . . . . . . . .
5
4
Produits considérés pour la période 2011 – petit usager . . . . . . . . . . . . . . . . . . .
11
5
Produits considérés pour la période 2011 – usager moyen . . . . . . . . . . . . . . . . . .
11
6
Produits considérés pour la période 2011 – gros usager . . . . . . . . . . . . . . . . . . .
11
7
Produits les meilleur marché en CHF par mois, TVA incluse – petit usager (prix réels) . .
12
8
Produits les meilleur marché en CHF par mois, TVA incluse – usager moyen (prix réels) .
12
9
Produits les meilleur marché en CHF par mois, TVA incluse – gros usager (prix réels) . .
12
10
Produits les meilleur marché en CHF par mois, TVA incluse – petit usager (prix normalisés) 13
11
Produits les meilleur marché en CHF par mois, TVA incluse – usager moyen (prix normalisés) 13
12
Produits les meilleur marché en CHF par mois, TVA incluse – gros usager (prix normalisés) 13
Table des figures
1
Indices de prix selon les profils d’usagers – prix réels . . . . . . . . . . . . . . . . . . . . .
5
2
Indices de prix selon les profils d’usagers – prix normalisés . . . . . . . . . . . . . . . . .
6
3
Produits les meilleur marché par opérateur (petit usager, prix réels) . . . . . . . . . . . . .
7
4
Produits les meilleur marché par opérateur (usager moyen, prix réels) . . . . . . . . . . .
7
5
Produits les meilleur marché par opérateur (gros usager, prix réels) . . . . . . . . . . . .
8
6
Produits les meilleur marché par opérateur (petit usager moyen, prix normalisés) . . . . .
9
7
Produits les meilleur marché par opérateur (usager moyen, prix normalisés) . . . . . . . .
9
8
Produits les meilleur marché par opérateur (gros usager, prix normalisés) . . . . . . . . .
10
3
Vue d’ensemble
L’indice des services à large bande (DSL et câble) d’un usager moyen (produits ayant des débits
descendants se situant de 2.5 à 15 Mbit/s inclus) a diminué de 5.2% par rapport à l’année précédente.
Le débit moyen des produits sélectionnés pour un usage moyen a augmenté en 2011 de 57.3%. On
combinant cette baisse à l’augmentation du débit moyen, on constate que l’indice de prix du Mbit/s a,
sur la même période, fortement diminué de 36.9%.
Diminution des prix et élévation globale des débits
Entre 2010 et 2011, les coûts supportés par un usager moyen pour des services à large bande
ont diminué (-5.2%). La baisse constatée s’explique par l’apparition de charges non récurrentes liées à
certains produits. Globalement, les débits ont crû en 2011. Ceux considérés dans le panel des produits
de l’usager moyen ont augmenté de près de 3000 kbit/s (+57%) pour s’élever en moyenne pondérée à
8087 kbit/s (5141 en 2010). Cette tendance s’explique par la commercialisation de produits à débits plus
élevés par plusieurs opérateurs.
Différences entre les opérateurs
A l’instar des années précédentes, 2011 est caractérisée par des différences de prix entre les opérateurs. Pour un usage moyen, deux groupes d’opérateurs se distinguent : les offres de VTX,Sunrise et
Cablecom affichant une tarification agressive montrant que la libéralisation du dernier kilomètre tend à se
concrétiser dans les faits et les offres de Swisscom, Improware, Quickline et Orange pour lesquelles le
statu quo prévaut. En comparant la moyenne de ces deux groupes, on constate une économie potentielle
sur les prix de l’ordre de 30%, pour des débits souvent comparables.
Forte baisse de l’indice du prix de 1 Mbit/s
La conséquence directe de la forte hausse des débits et de la diminution des prix pour l’usager moyen
est une forte diminution de l’indice des prix normalisés à 1 Mbit/s. L’indice décroît en effet de 36.9% par
rapport à l’année précédente. Notons encore qu’une majorité d’opérateurs proposent depuis 2007 des
produits affichant des débits descendants de plus de 10 Mbit/s. Cablecom a même commercialisé en
2009, dans certaines villes, un produit affichant un débit descendant de 100 Mbit/s.
4
Méthodes
L’indice des prix des services à la large bande pour l’accès à Internet repose depuis 2010 sur une
méthodologie définie par l’OCDE. L’indice couvre une période de 11 ans, soit de 2001 à 2011. La
méthode se base sur l’analyse de l’offre la meilleur marché de chaque opérateur de manière à refléter
les habitudes de consommation de trois profils d’usagers (petit, moyen et gros). Un autre indice, qui
calcule le coût théorique de chaque service si le débit était toujours de 1 Mbit/s, est présenté en parallèle
afin de préciser l’analyse. Cette approche ne reflète cependant pas la réalité budgétaire des usagers et
s’écarte ainsi de ce qu’un indice de prix à la consommation est censé refléter. La méthodologie détaillée
peut être consultée dans l’annexe C.
En 2010, l’OCDE a défini plusieurs paniers de consommation pour les services à large bande qui se
différentient en fonction des débits commercialisés et de l’usage du client. L’OFCOM en a retenu trois
sur les dix proposés et leur composition est présentée dans la table 1.
Type d’usager
Petit
Moyen
Gros
Débits descendants
> 0.250 Mbit/s et ≤ 2.5 Mbits/s
> 2.5 Mbit/s et ≤ 15 Mbits/s
> 15 Mbit/sx
Usage mensuel
Gbit/s Heures
6
30
18
45
33
60
Tab. 1 : Profils d’usager des services à large bande
Limites de la méthode
– L’analyse porte uniquement sur le produit le meilleur marché par opérateur, c’est-à-dire sur la minimisation des dépenses de l’usager pour son profil de consommation ;
– Le prix du raccordement à la ligne d’abonnés RTPC ou éventuellement au câble n’est pas considéré ;
– Lorsqu’un un abonnement au service téléphonique est obligatoire pour s’abonner au service à
large bande, ce dernier est ajouter au prix du service à large bande ;
– La méthode des prix réels traduit la réalité budgétaire de l’usager, mais les prix ne sont pas toujours
directement comparables en raison des différences de débits qui peuvent exister entre les offres
retenues. Les services proposant les débits les plus bas sont favorisés lorsqu’il s’agit de retenir le
produit le meilleur marché ;
– La méthode des prix normalisés à 1 Mbit/s ne traduit pas la réalité budgétaire de l’usager. Les
produits retenus sont en général ceux offrant les débits les plus grands ;
– Notons que plusieurs opérateurs proposent des produits qui combinent différents types de services
(par exemple, le raccordement, la téléphonie fixe, l’accès à Internet, voire même la téléphonie
mobile). Nos paniers de consommation ne s’intéressent qu’aux services d’accès à Internet sans
tenir compte des éventuels gains en matière de services associés à l’accès large bande. Dans
les cas où le raccordement téléphonique est inclus dans une offre combinée, il est volontairement
exclu à hauteur de ce qui est facturé à un client ne profitant pas d’une offre combinée.
– Les produits qui ne s’adressent qu’aux usagers pour lesquels le raccordement est déjà dégroupé
sont considérés dans l’analyse, même si de telles offres ne couvrent pas encore l’ensemble du
réseau de l’opérateur.
5
Résultats indiciels par profil d’usager
Prix réels
Année
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
Petit usager
100.0
100.1
Usager moyen
179.4
156.3
156.6
109.0
109.0
109.2
109.7
105.6
99.9
100.0
94.8
Gros usager
100.0
98.5
Tab. 2 : Indices de prix selon les profils d’usagers – prix réels (100 = période 2010)
200
200
200
150
150
150
100
100
100
50
50
50
0
0
0
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Année
Année
Année
(a) Petit usager
(b) Usager moyen
(c) Gros usager
Fig. 1 : Indices de prix selon les profils d’usagers – prix réels (100 = période 2010)
Prix normalisés
Année
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
Petit usager
100.0
74.0
Usager moyen
5438.1
4035.5
3968.7
1192.0
1008.0
561.5
236.3
117.6
100.7
100.0
63.1
Gros usager
100.0
93.2
Tab. 3 : Indices de prix selon les profils d’usagers – prix normalisés (100 = période 2010)
6
200
6000
200
5000
150
150
4000
100
3000
100
2000
50
50
1000
0
0
0
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Année
Année
Année
(a) Petit usager
(b) Usager moyen
(c) Gros usager
Fig. 2 : Indices de prix selon les profils d’usagers – prix normalisés (100 = période 2010)
7
Résultats indiciels par profil d’usager et positionnement des fournisseurs de services
Prix réels
Prix réels - petit usager
Année
2010
2010
2010
2010
2010
2011
2011
2011
2011
2011
2011
Fournisseur
Cablecom
Indice
Swisscom
QuickLine
ImproWare
Cablecom
VTX
Indice
Swisscom
QuickLine
ImproWare
Indice
64.6
100.0
109.8
113.0
170.5
64.7
90.6
100.1
110.0
113.2
156.3
200
150
●
●
100
●
50
Indice
Swisscom
Cablecom
ImproWare
Orange
QuickLine
Sunrise
Tele2
VTX
0
2010
2011
Année
Fig. 3 : Produits les meilleur marché par opérateur exprimés par rapport à la valeur de l’indice pondéré du petit
usager (prix réels, 100 = période 2010). Les valeurs de la position des opérateurs doivent uniquement être
comparées année par année pra rapport à l’indice et jamais de manière temporelle
Prix réels - usager moyen
Année
2010
2010
2010
2010
2010
2010
2010
2010
2011
2011
2011
2011
2011
2011
2011
2011
Fournisseur
Sunrise
VTX
Indice
QuickLine
ImproWare
Tele2
Swisscom
Orange
Sunrise
Cablecom
VTX
Indice
Tele2
Orange
Swisscom
ImproWare
Indice
72.9
81.4
100.0
102.8
103.5
104.8
105.0
105.0
72.8
75.1
81.5
94.8
104.9
105.1
105.1
125.1
200
150
●
●
100
●
50
Indice
Swisscom
Cablecom
ImproWare
Orange
QuickLine
Sunrise
Tele2
VTX
0
2010
2011
Année
Fig. 4 : Produits les meilleur marché par opérateur exprimés par rapport à la valeur de l’indice pondéré de l’usager
moyen (prix réels, 100 = période 2010). Les valeurs de la position des opérateurs doivent uniquement être
comparées année par année pra rapport à l’indice et jamais de manière temporelle
8
Prix réels - gros usager
Année
2010
2010
2010
2010
2010
2010
2010
2011
2011
2011
2011
2011
2011
Fournisseur
Cablecom
VTX
Indice
QuickLine
ImproWare
Swisscom
Orange
QuickLine
ImproWare
Cablecom
VTX
Indice
Swisscom
Indice
77.3
94.4
100.0
100.5
105.6
106.7
114.5
74.3
74.8
77.4
94.5
98.5
106.8
200
150
●
●
100
●
50
Indice
Swisscom
Cablecom
ImproWare
Orange
QuickLine
Sunrise
Tele2
VTX
0
2010
2011
Année
Fig. 5 : Produits les meilleur marché par opérateur exprimés par rapport à la valeur de l’indice pondéré du gros
usager (prix réels, 100 = période 2010). Les valeurs de la position des opérateurs doivent uniquement être
comparées année par année pra rapport à l’indice et jamais de manière temporelle
9
Prix normalisés
Prix normalisés - petit usager
Année
2010
2010
2010
2010
2010
2011
2011
2011
2011
2011
2011
Fournisseur
Cablecom
QuickLine
Indice
Swisscom
ImproWare
VTX
Swisscom
QuickLine
Indice
Cablecom
ImproWare
Indice
57.1
58.8
100.0
114.2
169.7
47.2
57.3
59.0
74.0
123.6
155.7
200
150
●
100
●
50
Indice
Swisscom
Cablecom
ImproWare
Orange
QuickLine
Sunrise
Tele2
VTX
●
0
2010
2011
Année
Fig. 6 : Produits les meilleur marché par opérateur exprimés par rapport à la valeur de l’indice pondéré du petit
usager (prix normalisés, 100 = période 2010). Les valeurs de la position des opérateurs doivent uniquement
être comparées année par année pra rapport à l’indice et jamais de manière temporelle
Prix normalisés - usager moyen
Année
2010
2010
2010
2010
2010
2010
2010
2010
2011
2011
2011
2011
2011
2011
2011
2011
Fournisseur
Sunrise
QuickLine
ImproWare
VTX
Indice
Tele2
Swisscom
Orange
Sunrise
Swisscom
VTX
Indice
Cablecom
Tele2
Orange
ImproWare
Indice
42.8
55.4
55.7
76.1
100.0
112.8
113.0
113.0
42.9
56.6
59.0
63.1
85.6
113.1
113.3
134.8
200
150
●
100
●
50
Indice
Swisscom
Cablecom
ImproWare
Orange
QuickLine
Sunrise
Tele2
VTX
●
0
2010
2011
Année
Fig. 7 : Produits les meilleur marché par opérateur exprimés par rapport à la valeur de l’indice pondéré de l’usager
moyen (prix normalisés, 100 = période 2010). Les valeurs de la position des opérateurs doivent uniquement
être comparées année par année pra rapport à l’indice et jamais de manière temporelle
10
Prix normalisés - gros usager
Année
2010
2010
2010
2010
2010
2010
2010
2011
2011
2011
2011
2011
2011
Fournisseur
Cablecom
QuickLine
Orange
Indice
VTX
ImproWare
Swisscom
Cablecom
ImproWare
QuickLine
Indice
VTX
Swisscom
Indice
29.8
71.5
89.7
100.0
107.8
122.2
123.4
26.3
37.6
37.6
93.2
104.2
117.9
200
150
●
●
100
●
50
Indice
Swisscom
Cablecom
ImproWare
Orange
QuickLine
Sunrise
Tele2
VTX
0
2010
2011
Année
Fig. 8 : Produits les meilleur marché par opérateur exprimés par rapport à la valeur de l’indice pondéré du gros
usager (prix normalisés, 100 = période 2010). Les valeurs de la position des opérateurs doivent uniquement
être comparées année par année pra rapport à l’indice et jamais de manière temporelle
11
A Produits considérés (2011)
Accès
cable
cable
cable
adsl
adsl
adsl
adsl
cable
Fournisseur
Cablecom
ImproWare
QuickLine
Swisscom
Swisscom
VTX
VTX
ImproWare
Service
Basic Internet
Light
Internet 2
DSL Start
DSL Mini
ADSL Light
Surf and phone 2 megas
Light
Upload (kbit/s)
100
150
200
100
200
100
200
150
Download (kbit/s)
500
1000
2000
500
2000
500
2000
1000
Tab. 4 : Produits considérés pour la période 2011 – petit usager
Accès
cable
cable
adsl
adsl
adsl
adsl
adsl
adsl
adsl
adsl
adsl
adsl
cable
Fournisseur
Cablecom
ImproWare
Swisscom
Sunrise
Sunrise
Sunrise
Sunrise
VTX
VTX
VTX
Orange
Tele2
ImproWare
Service
Fiber Power Internet 5
Standard
DSL Standard
click+call 5000+
click+call 15000+
click+call relax+
click 10000+
Surf and phone 5 megas
Surf and phone 10 megas
VDSL Single 10000
ADSL 5000
Tele2 ADSL 5000
Standard
Upload (kbit/s)
200
500
1000
500
1000
1000
1000
500
1000
1000
500
500
500
Download (kbit/s)
5000
5000
10000
5000
15000
15000
10000
5000
10000
10000
5000
5000
5000
Tab. 5 : Produits considérés pour la période 2011 – usager moyen
Accès
cable
cable
cable
cable
cable
adsl
adsl
adsl
adsl
adsl
adsl
adsl
Fournisseur
Cablecom
Cablecom
Cablecom
ImproWare
ImproWare
QuickLine
QuickLine
QuickLine
Swisscom
VTX
VTX
VTX
Service
Fiber Power Internet 25
Fiber Power Internet 50
Fiber Power Internet 100
Economy
Premium
Internet 20
Internet 50
Internet 100
Infinity
Surf and phone 20 megas
VDSL Single 20000
VDSL Single 30000
Upload (kbit/s)
2500
5000
7000
2000
5000
2000
5000
10000
2000
2000
2000
3000
Download (kbit/s)
25000
50000
100000
20000
60000
20000
50000
100000
20000
20000
20000
30000
Tab. 6 : Produits considérés pour la période 2011 – gros usager
12
B Produits les meilleur marché exprimés en CHF (2011)
B.1 Produits les meilleur marché – prix réels
Année
2011
2011
2011
2011
2011
Accès
cable
adsl
adsl
cable
cable
Fournisseur
Cablecom
VTX
Swisscom
QuickLine
ImproWare
Service
Basic Internet
Surf and phone 2 megas
DSL Mini
Internet 2
Light
Upload (kbit/s)
100
200
200
200
150
Download (kbit/s)
500
2000
2000
2000
1000
CHF
20.00
28.00
34.00
35.00
48.30
Tab. 7 : Produits les meilleur marché en CHF par mois, TVA incluse – petit usager (prix réels)
Année
2011
2011
2011
2011
2011
2011
2011
Accès
adsl
cable
adsl
adsl
adsl
adsl
cable
Fournisseur
Sunrise
Cablecom
VTX
Tele2
Swisscom
Orange
ImproWare
Service
click+call 5000+
Fiber Power Internet 5
Surf and phone 5 megas
Tele2 ADSL 5000
DSL Standard
ADSL 5000
Standard
Upload (kbit/s)
500
200
500
500
1000
500
500
Download (kbit/s)
5000
5000
5000
5000
10000
5000
5000
CHF
33.92
35.00
38.00
48.90
49.00
49.00
58.30
Tab. 8 : Produits les meilleur marché en CHF par mois, TVA incluse – usager moyen (prix réels)
Année
2011
2011
2011
2011
2011
Accès
adsl
cable
cable
adsl
adsl
Fournisseur
QuickLine
ImproWare
Cablecom
VTX
Swisscom
Service
Internet 20
Economy
Fiber Power Internet 25
Surf and phone 20 megas
Infinity
Upload (kbit/s)
2000
2000
2500
2000
2000
Download (kbit/s)
20000
20000
25000
20000
20000
CHF
48.00
48.30
50.00
61.00
69.00
Tab. 9 : Produits les meilleur marché en CHF par mois, TVA incluse – gros usager (prix réels)
13
B.2 Produits les meilleur marché – prix normalisés
Année
2011
2011
2011
2011
2011
Accès
adsl
adsl
cable
cable
cable
Fournisseur
VTX
Swisscom
QuickLine
Cablecom
ImproWare
Service
Surf and phone 2 megas
DSL Mini
Internet 2
Basic Internet
Light
Upload (kbit/s)
200
200
200
100
150
Download (kbit/s)
2000
2000
2000
500
1000
CHF
13.03
15.83
16.29
34.13
43.01
Tab. 10 : Produits les meilleur marché en CHF par mois, TVA incluse – petit usager (prix normalisés)
Année
2011
2011
2011
2011
2011
2011
2011
Accès
adsl
adsl
adsl
cable
adsl
adsl
cable
Fournisseur
Sunrise
Swisscom
VTX
Cablecom
Tele2
Orange
ImproWare
Service
click+call 15000+
DSL Standard
Surf and phone 10 megas
Fiber Power Internet 5
Tele2 ADSL 5000
ADSL 5000
Standard
Upload (kbit/s)
1000
1000
1000
200
500
500
500
Download (kbit/s)
15000
10000
10000
5000
5000
5000
5000
CHF
3.45
4.56
4.75
6.89
9.10
9.12
10.85
Tab. 11 : Produits les meilleur marché en CHF par mois, TVA incluse – usager moyen (prix normalisés)
Année
2011
2011
2011
2011
2011
Accès
cable
cable
adsl
adsl
adsl
Fournisseur
Cablecom
ImproWare
QuickLine
VTX
Swisscom
Service
Fiber Power Internet 100
Premium
Internet 100
Surf and phone 20 megas
Infinity
Upload (kbit/s)
7000
5000
10000
2000
2000
Download (kbit/s)
100000
60000
100000
20000
20000
CHF
0.72
1.02
1.02
2.84
3.21
Tab. 12 : Produits les meilleur marché en CHF par mois, TVA incluse – gros usager (prix normalisés)
14
C Méthodologie de l’OCDE
Unclassified
DSTI/ICCP/CISP(2009)14/FINAL
Organisation de Coopération et de Développement Économiques
Organisation for Economic Co-operation and Development
18-Mar-2010
___________________________________________________________________________________________
English - Or. English
DIRECTORATE FOR SCIENCE, TECHNOLOGY AND INDUSTRY
COMMITTEE FOR INFORMATION, COMPUTER AND COMMUNICATIONS POLICY
DSTI/ICCP/CISP(2009)14/FINAL
Unclassified
Working Party on Communication Infrastructures and Services Policy
REVISION OF THE METHODOLOGY FOR CONSTRUCTING TELECOMMUNICATION PRICE
BASKETS
English - Or. English
JT03280342
Document complet disponible sur OLIS dans son format d'origine
Complete document available on OLIS in its original format
DSTI/ICCP/CISP(2009)14/FINAL
FOREWORD
This document contains the 2009 revisions to the OECD's basket methodologies. These revisions
reflect discussions at an expert meeting held in Barcelona on 25-26 June 2009 as well as subsequent
written comments from governments and telecommunication operators. Governments and operators also
provided data to assist in the process of making appropriate adjustments in the baskets. The Working Party
on Communication Infrastructures and Services Policy (CISP) recommended, at its meeting in December
2009, forwarding the document to the Committee for Information, Computer and Communications Policy
(ICCP) for declassification. The ICCP Committee agreed to make the document publicly available in
March 2010.
© OECD/OCDE 2010
2
16
DSTI/ICCP/CISP(2009)14/FINAL
INTRODUCTION
The OECD has developed a set of methodologies for comparing retail prices of telecommunication
services for the purpose of assessing the price levels experienced by consumers and businesses in the
member countries.
The methodologies are based on a “basket” approach where a consumption pattern is described for
different types of users, and the prices of corresponding services from each provider covered are used to
calculate the resulting cost for each type of user. These baskets are referred to as the “OECD Baskets”.
The baskets are reviewed and revised periodically as consumption patterns change. Previous
revisions of the baskets took place in 2000, 2002 (mobile only), 2005 and now in 2009 with an OECD /
Teligen workshop in Barcelona.
The baskets are based on actual traffic data and consumption data, collected from operators and
regulators in the OECD member countries. The OECD, together with Teligen, has developed an approach
to the collection of data and the development of basket methodologies that is based on the following
elements:
•
Input from the OECD / Teligen basket workshop in Barcelona June 2009.
•
Traffic data and views submitted by operators and regulators in the OECD member countries up
until the end of October 2009 and comments on the proposal presented in December 2009.
Much of the background data which underpins the basket methodologies is commercially sensitive
and operators have submitted the data to Teligen to be analysed on behalf of the OECD. The following
basket definitions take into account all data received from operators and regulators.
3
17
DSTI/ICCP/CISP(2009)14/FINAL
New in these basket definitions
Voice call calculations
The 2009 basket revision includes a new method of accounting for different billing durations (e.g. per
second, per minute). This issue concerns both the PSTN and the mobile baskets.
The former calculation of individual call costs was based on the standard (average) call duration
specified in the basket for each type of call (destination and time). The call cost was calculated using the
actual billing period, i.e. using the full number of billing units covering the call duration. Essentially, calls
of various durations were rounded up to the next billing increment.
Table 1. Former method for calculating call durations by rounding to the nearest billable unit
60 second billing:
20 second billing:
Per second billing:
(75/60) = 2 units = 120 seconds
(75/20) = 4 units = 80 seconds
(75/75) = 75 units = 75 seconds
This meant that a 75 second call was billed as 75 seconds with per-second billing, and as 120 seconds
with per minute billing. This approach took into account the billing system effect on the cost, but not the
distribution of call durations. The selection of call duration in the basket would have a large influence on
the level of overbilling of actual call time. For example, a call just over the billing increment would require
a larger rounding up to reach the next increment while a call approaching the actual billing increment
would have very little overbilling.
If the average duration used in the basket was just below a multiple of billing units there was a small
additional time added as an effect of the billing. If, by contrast, the duration selected for the basket
happened to be just above a multiple of billing units the additional time (and cost) added could be
significant. The effect was somewhat arbitrary.
This 2009 revision changes the calculation to use the average duration defined by the basket, and then
add an element related to the billing method and distribution. The rationale behind this can be explained as
follows:
The new system for capturing differences in billing systems essentially calculates the price of a call
based on the actual number of seconds stipulated by the basket and then adds an additional adjustment
reflecting the average “overbilling” of calls corresponding to the billing system. Figure 2 below shows how
the amount of overbilling varies for each call.
4
18
DSTI/ICCP/CISP(2009)14/FINAL
Figure 1. Implications of the new methodology on different time-based billing systems
A call can have any duration, defined as A (seconds), and it ends P seconds into the last unit. When
the billing unit duration is N (seconds) the actual billing will overstate the average duration with
S=0, 1, 2, 3, ……, N-1 seconds,
depending on the value of A. If we assume that there is an even likelihood of any of these
durations (which is a simplification), the average overstated duration will be S=(N-1)/2. By
adding this to the average duration before calculating the cost of the call, the effect of both the
billing unit and the call distribution will be taken into account. The call calculation will then
always be calculated as if the billing is based on average per minute charge, calculated to the
fractions of a second before an adjustment for the billing systems is made.
Per minute billing
Per 20 second billing
Per second billing
= 29.5 second adjustment
= 9.5 second adjustment
= 0 second adjustment
Given an example with a 75 second call, the “adjusted” call duration will be calculated as follows
under different billing systems:
Table 2. Revised method for calculating call durations based on the average overbilling for each method
60 second billing:
20 second billing:
Per second billing:
75 + (60-1)/2 = 75 + 29.5 = 104.5 seconds
75 + (20-1)/2 = 75 + 9.5 = 84.5 seconds
75 + (1-1)/2 = 75 + 0 = 75 seconds
With this “adjusted” call duration the cost per call is calculated on the basis of the charge per second
in all cases.
5
19
DSTI/ICCP/CISP(2009)14/FINAL
If the “adjusted” call duration is shorter than the billing unit the full price of one billing unit will be
considered as the cost of the call. Any call set up charge will be added to the cost per call.
Selective discounts
Selective discount plans allow users to typically specify 1, 2, 3 or up to 10 or more numbers
(depending on the plan) to which calls and/or messages will be free or discounted. The basket definitions
generically refer to these plans as “selective discounts” although they are also known under brand names
like “Friends and Family”, “Bestmates”, “Preferred numbers”, “Calling circle” etc.
It is also worth noting that there will normally be an overall traffic increase with the introduction of
selected discounts. Most of this increase will go to the nominated numbers, and will only have a limited
effect on the end user cost as such calls are free or heavily discounted. In the basket calculations this
increase is not considered.
The handling of the selective discount is based on the following elements and assumptions:
The total number of minutes for all calls in the basket is V
The discount applies to N nominated numbers
The discount D (%) applies to each of these calls
(D=100% is a free call)
The proportion of minutes A (%) receiving the discount is calculated based on the formula below,
using V and N as input data. The proportion A is adjusted according to the discount D
(A2 = A x D)
Mapping information will indicate which call types are affected by the discount.
The remaining proportion A2 is used to calculate the number of minutes to be deducted from the
basket minutes according to the call type mapping.
Cost of remaining minutes is calculated as usual.
Where discounted SMS is offered as part of the discount they will be deducted using the same
value A from the formula below, and the relevant discount applicable to the SMS.
The critical element is the calculation of A. This proportion is based on N (number of nominated
numbers) and V (total minutes in basket), and an empirically developed function can be used to calculate
the proportions as shown on the graph below. The function is:
A (%) = Log(10 x N1.5) / Log(10 x V)
This function provides a proportion that resembles the amount of calls going to nominated numbers in
the data received from operators in this basket review.
6
20
DSTI/ICCP/CISP(2009)14/FINAL
Figure 2. Selective call discount volumes
Principle of selective discount call volumes
100%
90%
Percentage of all minutes
going to nominated numbers
80%
70%
60%
50%
60 minutes/month
40%
200 minutes/month
30%
600 minutes/month
1660 minutes/month
20%
10%
0%
1
2
3
4
5
6
7
8
9
10
Nom inated num bers
The selective discount will be taken before any minute, message and value allowances included in the
package. The amount of minutes that will be deducted because of the selective discount is calculated as:
V(2) = V(1) x (Log(10 x N1.5) / Log(10 x V)) x D
where V(1) is the total number of minutes defined by the basket, and V(2) is the number of minutes
going to the nominated numbers
V(2) is then distributed to the specific call types according to the selective discount mapping. Each
call type will have between zero and V(2) minutes to be deducted. The remaining minutes for each
call type is used for the following distribution of allowances and calculation of call costs.
Local calling areas for the PSTN baskets
Previous versions of the baskets provided 14 discrete distances for national fixed line calls. This
enabled both a distance distribution and a method for incorporating the size of local calling areas. The
baskets now use only local and national areas to describe the national destinations for fixed line calls, due
to considerable simplifications in the pricing of such calls in recent years.
In order to allow for differences in the size of local calling areas the following adjustment of the local
and national call proportions will be used, based on the closest size of local calling area:
7
21
DSTI/ICCP/CISP(2009)14/FINAL
Table 3. Local and national call proportions
Average local call radius
Local adjustment
10 km
15 km
20 km
25 km
30 km
50 km
100 km
National adjustment
-8.9%
+8.9%
-4.2%
+4.2%
-1.4%
+1.4%
0.0%
0.0%
3.2%
-3.2%
6.4%
-6.4%
9.6%
-9.6%
In addition the percentages above must be adjusted with the proportion of fixed line calls in each
basket, as given in Table 4 below.
Table 4. Basket adjustment
Basket
adjustment
81%
75%
73%
90%
67%
66%
20 calls basket
60 calls basket
140 calls basket
420 calls basket
100 calls business basket
260 calls business basket
An example: The 60 calls basket will have the following adjustment factors:
Table 5.
Average
Local call
area radius
Example – 60 call basket
Local
adjustment
National
adjustment
10 km
15 km
20 km
25 km
30 km
50 km
-6.7%
-3.2%
-1.1%
0.0%
2.4%
4.8%
6.7%
3.2%
1.1%
0.0%
-2.4%
-4.8%
100 km
7.2%
-7.2%
If the operator, for example, uses an average local calling radius of 15 km, the fixed-to-fixed local
proportion will be 60% - 3.2% = 56.8%, and the fixed-to-fixed national portion will be 15% + 3.2% =
18.2%.
8
22
DSTI/ICCP/CISP(2009)14/FINAL
PSTN BASKETS
General basket rules: PSTN
Only incumbent operators are covered.
For each operator a relevant number of packages shall be included. Combinations of packages are
allowed. Discounts shall be incorporated in the calculations of prices for each basket. Only tariffs
presented clearly as current tariffs on the operator web pages will be considered.
Selective discounts are calculated with the algorithm described above.
There are 4 residential and 2 business baskets:
Table 6. Residential and business PSTN baskets
Type of basket
Residential
Residential
Residential
Residential
Business
Business
Basket
20 calls basket
60 calls basket
140 calls basket
420 calls basket
100 calls business basket, single user
260 calls business basket, single user
Nonrecurring charges are covered using the charge for a new installation of a service.
Nonrecurring charges are distributed over 5 years, except where the installation is a tradable asset
(Japan) where the charge is distributed over 20 years.
Call costs are calculated using the duration of D + (Unit(seconds)-1)/2, based on basket call duration D
converted to seconds and average per second charges. Unit is the billing unit in seconds.
National call charges to fixed networks are based on a local / national split. While this is adequate for
most prices, some operators may split their prices into local / regional / national. In such cases only the
prices for local and national areas will be considered.
The proportion of calls to the local calling area will be adjusted for local calling area radius as
described above in Table 3.
When call charges to mobile networks differ by network, the weighted average charge for calls to all
national mobile networks shall be used, based on available subscriber numbers.
International calls to other OECD countries are included, with call charges weighted according to
actual traffic volumes. This means that those destinations with the most traffic will carry the most weight.
Traffic volume weights can be provided by national regulators. Until such weights are provided the
existing weights are used.
For international calls the highest charge is used for peak time, and the lowest is used for off-peak
time.
9
23
DSTI/ICCP/CISP(2009)14/FINAL
Allowances are deducted in the following order: Selective discounts, minute allowance, value
allowance. Specific volume discounts will be deducted from the total cost at the end.
Results are presented in USD / PPP per month, excluding VAT for business baskets and including
VAT for residential baskets. Nominal exchange rates can be used.
PSTN basket parameters
Table 7.
Overall basket volumes and destination distribution (Fixed)
Call distribution
Calls per month
20 calls basket
60 calls basket
140 calls basket
420 calls basket
100 calls business basket
260 calls business basket
Total
calls
Fixed to
fixed
Local
Fixed to
fixed
National
Fixed to
mobile
International
20
60
140
420
100
260
61%
60%
58%
73%
48%
43%
20%
15%
15%
17%
19%
23%
17%
21%
23%
8%
30%
25%
2%
4%
4%
2%
3%
9%
Table 8. Time of day distribution: Fixed to fixed
Day
20 calls basket
60 calls basket
140 calls basket
420 calls basket
100 calls business basket
260 calls business basket
53%
60%
52%
52%
69%
75%
10
24
Fixed to Fixed
Evening
Weekend
25%
22%
26%
26%
17%
15%
22%
18%
22%
22%
14%
10%
DSTI/ICCP/CISP(2009)14/FINAL
Table 9. Time of day distribution: Fixed to mobile
Day
20 calls basket
60 calls basket
140 calls basket
420 calls basket
100 calls business basket
260 calls business basket
Fixed to Mobile
Evening
Weekend
45%
57%
46%
46%
69%
77%
28%
22%
27%
27%
18%
14%
27%
21%
27%
27%
13%
9%
Table 10. Time of day distribution: International (fixed)
International
Peak
Off peak
20 calls basket
60 calls basket
140 calls basket
420 calls basket
100 calls business basket
260 calls business basket
45%
44%
47%
47%
75%
87%
55%
56%
53%
53%
25%
13%
PSTN call durations
Table 11. PSTN call durations: Fixed to fixed local, minutes per call
Day
20 calls basket
60 calls basket
140 calls basket
420 calls basket
100 calls business basket
260 calls business basket
2.6
2.6
3.1
3.6
1.9
2.0
Fixed to fixed local
Evening
Weekend
4.0
3.8
4.8
5.4
2.3
2.8
2.6
2.9
3.7
5.4
2.1
3.1
Table 12. PSTN call durations: Fixed to fixed national, minutes per call
Fixed to fixed national
20 calls basket
60 calls basket
140 calls basket
420 calls basket
100 calls business basket
260 calls business basket
Day
Evening
Weekend
4.0
4.1
4.7
5.3
2.3
2.4
6.3
6.4
7.6
8.1
3.3
2.7
5.4
6.4
7.1
8.1
3.3
3.4
11
25
DSTI/ICCP/CISP(2009)14/FINAL
Table 13. PSTN call durations: Fixed to mobile, minutes per call
Fixed to mobile
20 calls basket
60 calls basket
140 calls basket
420 calls basket
100 calls business basket
260 calls business basket
Day
Evening
Weekend
1.5
1.9
1.7
1.8
1.6
1.7
2.1
2.4
2.3
2.3
1.9
2.2
1.3
1.9
2.1
2.3
1.5
1.9
Table 14. PSTN call durations: International
International
Peak
Off peak
20 calls basket
60 calls basket
140 calls basket
420 calls basket
100 calls business basket
260 calls business basket
4.6
4.7
4.7
5.0
3.2
3.7
12
26
6.2
6.8
6.8
8.1
5.4
4.1
DSTI/ICCP/CISP(2009)14/FINAL
MOBILE BASKETS
General basket rules
At least two largest network operators are covered for each country, based on subscriber numbers.
The operators covered should between them have at least 50% of the market share.
Discount brands offered by the network operators will only be included when clearly linked with the
network operator’s website and brand.
Offers shall be typical 2G and 3G services with the main focus on voice.
A range of relevant price packages shall be covered, allowing the lowest cost package to be selected
for each operator. Only price presented clearly as current on the operator web pages will be considered.
The basket methodologies can be used to compare different types of offers such as pre-paid, post-paid
or SIM-only plans.
Table 15. Mobile baskets
Basket
30 calls per month
100 calls per month
300 calls per month
900 calls per month
40 calls per month pre-paid basket
400 messages per month basket
Nonrecurring charges are distributed over 3 years.
Selective discounts are calculated with the algorithm described in the section on selective discounts
above.
The value of call and message allowances included in the package will be deducted from the usage
element of the basket, up to the value of actual usage.
Allowances are deducted in the following order: Selective discounts, most restricted minute
allowance, least restricted minute allowance, message allowance, value allowance. Specific volume
discounts will be deducted from the total cost at the end.
Off-net mobile-to-mobile charges are weighted according to subscriber numbers for each
country, where relevant for the pricing of calls.
Call costs are calculated using the duration of D + (Unit(seconds)-1)/2, based on basket call duration D
given below concerted to seconds and average per second charges. The Unit is the billing unit in seconds.
13
27
DSTI/ICCP/CISP(2009)14/FINAL
Results are presented in USD / PPP per month including VAT. Nominal exchange rates can be used.
Mobile basket parameters
Table 16. Mobile: Overall basket volumes and destination distribution
Call distribution
Total
calls per
month
30
100
300
900
40
8
Volume per month
30 calls basket
100 calls basket
300 calls basket
900 calls basket
40 calls prepaid basket
400 messages basket
Mobile to
fixed
16%
17%
14%
14%
14%
8%
On-net
55%
52%
46%
55%
64%
55%
Off-net
25%
28%
37%
28%
18%
25%
Voicemail
4%
3%
3%
3%
4%
12%
SMS
100
140
225
350
60
400
Table 17. Mobile time of day distribution
Voice call distribution
30 calls basket
100 calls basket
300 calls basket
900 calls basket
40 calls prepaid basket
400 messages basket
Day
46%
51%
49%
49%
46%
46%
Evening
29%
26%
32%
32%
29%
29%
Message distribution
Weekend
25%
23%
19%
19%
25%
25%
Peak
66%
66%
66%
66%
66%
66%
Off-peak
34%
34%
34%
34%
34%
34%
On-net
53%
51%
50%
50%
53%
50%
Table 18. Mobile voice call durations
30 calls basket
100 calls basket
300 calls basket
900 calls basket
40 calls prepaid basket
400 messages basket
Call duration (minutes / call)
Mobile to
On-net
Off-net
fixed
2.0
1.6
1.7
2.1
1.9
1.8
2.0
2.0
1.8
1.9
2.1
1.9
1.9
1.9
2.0
1.6
2.2
1.6
14
28
Voicemail
0.9
1.0
1.0
1.1
0.9
1.1
Off-net
47%
49%
50%
50%
47%
50%
DSTI/ICCP/CISP(2009)14/FINAL
LEASED LINE BASKETS
General basket rules
The basket shall include transparent end-to-end leased lines from the incumbent operators. Virtual
circuits can be included in cases where traditional leased circuits do not exist anymore; xDSL services,
however, fall outside the scope of the basket.
Non-recurring charges (installation) are excluded from the basket. Only monthly rental charges are
included.
Circuits above 2 km shall include two 2 km local tail circuits within the defined distance. This means
that, for example, a 50 km circuit will have 2 local tail circuits of 2 km, and a main circuit of 46 km. Some
operators include the local tail circuits in the total price, some do not.
Circuits are assumed to be within or out of the major city in the country. This means that the 2 km
circuit is a local circuit within the major city, and the rest of the distances will have one end in the major
city, and the other end outside.
Where the distance exceeds the possible distance for a country, the highest available price is used for
that distance. This means that even when a circuit length would go beyond the borders of a country, this
circuit is included in the basket, using the price of the longest possible circuit.
Leased line capacities vary slightly among countries. For example, a T1 line in the United States has a
capacity of 1.5 Mbit/s while an E1 line in Europe has a capacity of 2 Mbit/s. In order to compare the two
services, the price of the T1 line is adjusted upward proportionately to match bit rates. In this case, the
price of
T1 (1.5 Mbit/s) = P(T1)
E1 (2.0 Mbit/s) = P(E1)
T1 (adjusted to 2.0 Mbit/s) = P(T1 adjusted) = P(T1) * [2/1.5] = P(T1) * 1.33
For 34 Mbit/s circuits there is a situation where sometimes a 45 Mbit/s is used as a proxy when the
34 Mbit/s service is not offered. In these cases the price of the 45 Mbit/s is adjusted down to reflect the
implied per-Mbit/s price of 34 Mbit/s. As an example:
T3 (45 Mbit/s) = P(T3)
E3 (34 Mbit/s) = P(E3)
T3 (adjusted to 34 Mbit/s) = P(T3 adjusted) = P(T3) * [34/45] = P(T3) * 0.756
Results are presented in USD per month, excluding VAT.
15
29
DSTI/ICCP/CISP(2009)14/FINAL
Leased line basket parameters
Table 19. Overall basket volumes and destination distribution by distance (km) and capacity (Mbit/s) for
leased lines
2 Mbit/s
34 Mbit/s
2 km
20 km
50 km
100 km
200 km
500 km
50%
42%
18%
18%
6%
15%
8%
9%
10%
8%
8%
8%
16
30
DSTI/ICCP/CISP(2009)14/FINAL
FIXED BROADBAND BASKETS
General basket rules
The baskets shall include fixed broadband services provided over ADSL, cable and fibre networks.
The basket shall include the following charge elements: Connection (one off) charge, Monthly rental
charge, Usage allowance (time or volume), Overage charges.
Each basket will describe the minimum bitrates covered within the basket and the level of usage.
Basket results will contain fixed and usage cost for each package. Where an allowance is exceeded
and the service is stopped or speed reduced until the end of the billing period, the package will not be
included in the basket results.
Results are presented in USD/ PPP per month including VAT. Nominal exchange rates can be used.
Given the range of data allowances across OECD countries, the broadband basket will include two
profiles – one for lower and another for higher use (see Table 20).
Broadband basket parameters
Table 20. Fixed broadband basket profile, by advertised download capacity
Usage per month
Lower use
Higher use
Speed range
GB
Hours
GB
Hours
>0.25 Mbit/s
>2.5 Mbit/s
>15 Mbit/s
>30 Mbit/s
>45 Mbit/s
2
6
11
14
18
10
15
20
25
30
6
18
33
42
54
30
45
60
75
90
17
31