Issued by the Director General of Telecommunications Contents

Chapter 1 Introduction
Points to bear in mind
Outline of the report

Chapter 2 Internet access – basic
Introduction
Market developments
Methodology
Results for residential baskets
Results for business baskets
Sensitivity analysis
Detailed results
Conclusions

Chapter 3 Internet access – broadband
Introduction
Market developments
Methodology
Results for residential users
Results for business users
Sensitivity analysis
Comparison of dial-up and broadband pricing
Service availability and takeup
Conclusions

Annex A Exchange rates and VAT rates

Annex B Detailed results for Internet access baskets (basic)

Annex C Detailed results for DSL and cable modem services

Summary

Scope

S.1 This report is an International Benchmarking study of Internet access covering both basic dial-up access and broadband services (DSL and cable modem) in France, Germany, Sweden, United Kingdom and the US (the States of Ohio and California are used for dial-up services; the study covers a number of the largest providers of broadband who offer services across a range of States).

S.2 For dial-up Internet access this study is a follow up to the study published by Oftel in June 2001 (however, changes to the methodology used in this report mean the results are not strictly comparable with those published in the June 2001 report). For broadband Internet access, this study is a follow up to the study published by Oftel in January 2001. The analysis and drafting of this report has been carried out by Oftel. The models developed by Oftel used to generate the results contained in this report have been audited by an independent consultant, Analysys Ltd. The tariff data used in this report have been collected by consultants Total Romtec.

S.3 The details of the approach used, and the points which should be borne in mind in interpreting the results, are contained in the main chapters of this report.

S.4 The results presented are based on a snapshot of prices as at 9 August 2001.

Results for ‘basic’ Internet access

S.5 Tables 1, 2 and 3 show the results for the price index for basic Internet access (ie Internet access for consumers who have no particular requirement for access speed), expressed relative to the UK price index, for residential off-peak usage, residential peak usage and for business users respectively.

S.6 These results show that, based on the sample of ISPs/operators selected:

• for residential consumers, UK prices compare favourably with the other European countries for both peak and off-peak access and for both limited and unmetered services. Prices in the US are generally up to 10% cheaper; and
• for business consumers, UK prices for limited usage services are above prices seen elsewhere. UK prices for unmetered services are above those in Germany, close to those in the US and below those in France and Sweden.

S.7 In the UK and US, unmetered dial-up services are widely available. In the other European countries this is not the case, and broadband services are more cost effective than dial-up for consumers with high usage. The results are based on the cheapest available option, and hence include some broadband packages.

Table 1: Comparison of residential off-peak Internet access prices

Table 2: Comparison of residential peak Internet access prices

Table 3: Comparison of business Internet access prices

* only one package used for this, due to the limited number of packages in some countries

S.8 The results show some movement compared to those in the June 2001 report. For residential services, prices in the UK are now above those in the US, and generally the range of prices has narrowed. The inclusion of broadband services has improved the position for some countries.

Results for broadband Internet access

S.9 Table 4 shows the results for the price level, price index (relative to the UK) and (downstream) bandwidth provided for broadband residential services. The results are based on the average of the two cheapest deals available. Results are presented both with and without cable modem offers included.

S.10 From Table 4 it is seen that:

• When cable modem services are included, prices in the UK are lower than in all countries except Sweden; and
• When cable modem services are excluded (ie a comparison of DSL services only) UK prices are higher than in all countries.

Table 4: Results for price level, price index and (downstream) bandwidth for residential broadband services

S.11 The UK's relative position for residential services has improved relative to the January 2001 report. This is due to price falls for cable modem services.

S.12 Table 5 shows the results for the price level, price index (relative to the UK) and bandwidth provided for broadband business services, for a variety of consumer types. The results are based on the average of the two cheapest deals available.

S.13 From Table 5 it is seen that:

• for the most basic services available to business (ie where the business has no minimum bandwidth requirement):

• prices in Sweden and the US are around 20% cheaper than in the UK;
• prices in Germany are similar to prices in the UK; and
• prices in France are more expensive than in the other countries – but provide a higher bandwidth;

• for business services where a minimum geometric mean bandwidth of 500kbit/s is required:

• Prices in Sweden and the US are very much lower than elsewhere;
• UK prices are between prices in France (10% cheaper) and those in Germany (16% higher); and
• The bandwidth in the UK is about the same as in France and Sweden but less than in Germany and the US; and

• for business services where a minimum geometric mean bandwidth of 1000kbit/s is required:

• Services are only available in Germany, Sweden and the US; and
• Prices in Germany are much more expensive than Sweden or the US.

Table 5: Results for price level, price index and bandwidth for business broadband services (cable modem services included)

Note: DS = downstream, GM = geometric mean

S.14 Prices for UK business services are similar to those in the January 2001 report. There have been some price falls elsewhere, but also some price rises.

Takeup and availability of broadband services

S.15 Data on takeup are presented as a % of population (using OECD population figures for 1999) in Figure 1. Broadband takeup rates in each country are shown as at end 1999, end 2000 and mid 2001.

S.16 While current levels of broadband takeup are low (below 5% of population) in all of the countries benchmarked, takeup in the US and Sweden is markedly higher than in the other European countries. However, if the data for DSL takeup is replotted to show takeup in relation to time since service launch, it is seen the current low takeup in France and the UK may be largely explained by their later start date. The UK's current takeup level is similar to the level seen in other countries a year or so after service launch.

Figure 1: Broadband takeup per 100 population by country for end 1999, end 2000 and mid 2001.

Source: OECD

Chapter 1

Introduction

1.1 Oftel’s aim is to provide the best possible deal for telecommunications customers in terms of quality, choice and value for money through effective competition. International comparisons of telecoms services are one way of assessing the achievement of this aim.

1.2 Oftel carries out regular reviews of the major telecoms markets to determine the appropriate level of regulation for the future. Regular benchmarking exercises allow a time trend to be developed, which provides greater information than a single snap shot. The results in this report will form part of the data input for future reviews.

1.3 This study focuses on comparing the cost for representative users for Internet access in each country using a ‘basket’ methodology. Oftel has carried out its own benchmarking work (as opposed to using alternative published sources) to ensure that the methodology is rigorous and robust and that it addresses the following issues:

• Usage baskets need to reflect a range of patterns of use representative of different groups of consumers; and
• Consumers have a potentially wide range of choices for their telecoms services:

• Consumers have a choice of competing operators all of whom will offer different packages; and
• Each operator may offer a variety of different packages and discount schemes targeted at different segments of the market;

1.4 The following countries are included in the analysis:

• France;
• Germany;
• Sweden;
• UK; and
• US (the States of Ohio and California are considered for dial-up Internet access; the study covers a number of the largest providers of broadband who offer services across a range of States).

1.5 For dial-up Internet access services, this study is a follow up to the study published by Oftel in June 2001 (the ‘June 2001 study’), which was based on tariff data valid as at February 2001. For broadband Internet access, this study is a follow up to the study published by Oftel in Jan 2001 (the ‘January 2001 study’), which was based on tariff data valid as at October 2000. A list of the ‘tariffs valid at’ date and date of publication of previous studies is given in Table 1.1.

Table 1.1: List of previous studies

Reports can be found at: www.oftel.gov.uk/publications/research/index.htm

1.6 The analysis and drafting of this report has been carried out by Oftel. The models developed by Oftel used to generate the results contained in the report have been audited by an independent consultant, Analysys Ltd, who confirm the following:

• The models are logically sound (ie they correctly implement the methodology described in this report); and
• The tariff data has been checked and has been correctly entered and interpreted.

1.7 Data have been collected from the vast range of tariff packages available to residential and small/medium sized businesses, for major operators and service providers in each country as at August 2001. Data collection has been carried out for Oftel by consultants Total Romtec.

Points to bear in mind

1.8 There are a huge range of operators and service providers offering services to consumers and it is not possible to cover all packages on offer. In addition, it is not possible to fully capture all aspects of innovative tariff products. While it is believed that the sample chosen is fully representative and covers the range of offers available by country, the possibility that consumers have access to additional offers outside the range calculated for the sample used in this report cannot be excluded. Not all tariffs included will be available to all consumers in a particular country.

1.9 The objective of this report is to provide a comparison between countries for a range of consumer usage profiles. A range of profiles is chosen to be illustrative of the most frequent types of customer calling patterns, but is not comprehensive. Prices are based on a snap shot as at 9 August 2001. Although providing a sound overall picture, the comparison of companies within a country should not be taken as necessarily being representative for an individual consumer.

1.10 The analysis is based on a basket methodology, which allows prices to be compared across different operators in different countries. It is not, however, intended to be a buyer’s guide to services as:

• it provides only a snapshot of the price levels for a continually changing set of price packages; and
• the residential or business consumer will have to use more specific usage information in order to determine the best option for that person or business.

The objective is to illustrate the scale and level of prices available to consumers in different countries.

1.11 There are separate industry funded initiatives and reports which seek to help UK consumers be better informed purchasers of telecommunications services (examples are: individual enquiry based price comparisons such as – www.phonebills.org.uk; fixed and mobile quality of service indicators – www.cpi.org.uk; and information for small businesses about using telecoms and the Internet – www.TelecomsAdvice.org.uk). This report is not part of that body of work as it seeks to compare positions between, not within, countries and it is not, therefore, intended in any way to be a buyer's guide to services in the UK.

Outline of the report

1.12 There are two subsequent chapters which present the results for:

• Internet access – basic (Chapter 2);
• Internet access – broadband (Chapter 3).

1.13 In general, the methodology used is very similar to that used in previous reports. However, this is the first time that results for dial-up Internet access and broadband Internet access have been presented in the same report. Some changes in methodology have been incorporated in order to recognise the overlap in provision of Internet via different access mechanisms. In particular:

• comparisons of ‘basic’ Internet access (where the consumer is unconcerned about speed or ‘always-on’ service) are no longer restricted to dial-up packages – where DSL/cable modem services are cheaper than dial-up (as is the case for heavy users in some countries), these prices are chosen as the cheapest available package; and
• the comparisons for broadband (or ‘higher bandwidth’ services) now include a definition on minimum bandwidth (ie they are based on service capability, rather than technology).

1.14 Details are provided in chapters 2 and 3. This approach also allows a comparison to be made of the gap between unmetered dial-up and broadband products in each country.

Chapter 2

Internet access – basic

Introduction

2.1 This chapter covers ‘basic’ Internet access, ie access to the Internet by consumers who have no requirement for high speed access. Mainly this will mean PSTN dial-up access, as this is usually the cheapest option. However, there are circumstances where other access mechanisms may be cheaper. In countries where unmetered PSTN dial-up access is not available, broadband access may be a cost effective alternative for consumers with high usage. In this chapter a range of access mechanisms are considered (ie PSTN and ISDN dial-up, DSL and cable modem) and the results then focus on the cheapest available options. Price comparisons for consumers requiring the greater access speed, and ‘always-on’ facility, provided by DSL and cable modem, are presented in Chapter 3.

2.2 Oftel has recently published a consultation document for dial-up narrow band access and has proposed that the retail market for dial-up Internet access is effectively competitive. The final statement will be published towards the end of this year. This study will form part of the data input to the final statement. It also forms part of the ongoing monitoring of the market for dial-up Internet access and the resulting time series will be used in subsequent reviews.

2.3 The results provide an update of the results presented in the June 2001 report (based on prices as at February 2001). This study is the fourth in a series that started with a study carried out by Teligen for Oftel and published in May 2000 (based on prices as at February 2000).

2.4 This chapter follows broadly the same format as the previous reports and shows:

• a summary of developments in the UK market over the last six months;
• a summary of the methodology used;
• results for business and residential consumers showing:

• price indices for a range of baskets;
• price movements over the last six months;
• charts of the spread of offers by country;

• sensitivity analysis;
• detailed results; and
• conclusions.

2.5 Detailed results for all the packages modelled are shown in Annex B.

Market developments

2.6 Results from Oftel’s latest consumer research (conducted in August 2001) show that 39% of homes are now connected to the Internet (compared to 30% of homes connected as at Nov 2000) and that, of these, over 80% use PSTN dial-up access. Of those homes with Internet access, 28% use a fully unmetered package and a further 12% a partially unmetered package (compared to 20% using fully unmetered and 11% partially unmetered in Nov 2000). Average time spent on line has risen from 30 hours a month (seven hours a week) to 35 hours a month (eight hours a week) over the same period, though average usage for those on fully unmetered packages is 56 hours a month (13 hours a week).

2.7 Around 61% or so of small and medium sized businesses are connected to the Internet, compared with 55% in May 2001. The majority of small businesses used ordinary phone line/dial-up access to connect to the Internet, whilst medium-sized businesses were more likely to use ISDN or leased lines. Overall, 73% of businesses used ordinary phone line/dial-up access to connect to the Internet, 28% used ISDN and 1% used leased lines. 37% of Internet users were using some form of unmetered package in August 2001 – broadly similar to the figures for May 2001.

Methodology

2.8 In previous reports, price comparisons for Internet access via PSTN dial-up services and via DSL/cable modem services have been presented separately. In countries like the UK, where unmetered dial-up access is available, it makes sense to distinguish price comparisons for ‘basic’ (low speed) Internet access from those for higher speed access. However, in countries where unmetered dial-up access is not available, broadband services are likely to be a cost-effective alternative to metered dial-up access for heavy users. In previous benchmarking exercises, it has been found that unmetered dial-up products are not available in some countries, leading to an incomplete comparison. Including broadband offers in comparisons of basic Internet access does not affect the level of ‘cheapest available’ offers in the UK, but does have an impact in some other countries.

2.9 Inclusion of broadband services raises the question of what costs should be included in the ‘basket’ to ensure dial-up and broadband services are compared on a like for like basis. In particular, there is the question of the treatment of the fixed costs of the PSTN line. In previous studies, all but the lowest usage baskets have included the fixed cost of the PSTN line. This is on the basis that:

• a high Internet user may have a second line installed for Internet use and/or they may see the primary reason for having a fixed line as Internet access (rather than voice services);
• not including the fixed cost would distort the results in cases where there is a trade off between the line rental and the cost of a local call. Eg in the US a user will pay a higher line rental in order to have unmetered local calls (including Internet calls) – not including the line rental would ignore this cost element; and
• including the fixed PSTN costs is consistent with the OECD methodology.

2.9 For broadband services, the issues are different. While the user needs a PSTN subscription, Internet use over DSL or cable modem does not affect voice telephony (which can occur at the same time). In the following analysis, the fixed PSTN line costs are not included. This produces a consistent comparison for unmetered dial-up access and broadband access: in both cases there is a dedicated line for Internet access. A small number of ISDN packages have also been included. In this case, the incremental cost of ISDN over PSTN is included. The results are also run removing the fixed PSTN line costs as a sensitivity (as has been done in previous studies).

2.10 Some DSL and cable modem packages contain a usage element, often expressed as a cost per GigaByte (GB), with some level of usage already included in the monthly price. For these packages the basic cost only has been modelled – it is assumed that where a broadband service is used as a substitute for dial-up, the levels of usage are unlikely to exceed the included allowances.

2.11 The change in methodology has implications for the trend in prices and the price spread charts. To make these meaningful, results have been run on a ‘dial-up only’ basis.

2.12 This approach does not explicitly address the question of availability, ie while most dial-up packages will be available throughout the country, the availability of broadband services may be limited. The availability of broadband services is now in excess of 50% of households in the countries studied, so can be considered to be fairly widely available. Previous benchmarking studies have also not explicitly addressed limits to availability on dial-up offers (eg offers from cable companies will not be country wide, some unmetered services may be offered to a limited number of consumers). Limits to availability should be borne in mind in interpreting the results.

2.13 Apart from the above points, the methodology remains as in the June 2001 report, and the details are not repeated here.

Results for residential baskets

Price index for residential off-peak services

2.14 The price index (based on the average of the two cheapest packages from different operators in each country) is presented in Table 2.1 for residential off-peak baskets.

Table 2.1: Comparison of residential off-peak package prices

2.15 From Table 2.1 it is seen that:

• UK prices are the cheapest of the European countries included in the study for unlimited off-peak access;
• UK prices for limited use are close to the cheapest overall (prices in Germany are 2% cheaper; and
• UK prices for limited use compare more favourably with the other European countries studied the higher the usage level.

2.16 In several countries, cable modem or DSL services appear as the cheapest package for some baskets. In France, Germany and Sweden, there are no dial-up unmetered packages (with the exception of one package in Germany – but this is found to be expensive and does not feature in the index), and prices for the unmetered basket are based on DSL/cable modem offers. In Germany and Sweden, cable modem or DSL packages start to become cost effective at 20 hours usage per month.

2.17 The results present a positive picture for the UK, however, there has been some movement in the UK's relative position – in the previous study the UK was the cheapest overall by a fairly large margin. This change is due to a combination of: increased prices in the UK; price falls for PSTN dial-up in some countries; and the change in methodology used in this report – broadband services are now included and these are cost effective in some countries at relatively low levels of usage.

Price trend

2.18 A summary of price changes since February 2001 study is provided in Table 2.2. The average price change shown is the mean of the percentage changes across each of the original five baskets. To ensure the price trend is meaningful, only dial-up packages have been included in this analysis.

Table 2.2: Percentage change in price levels for residential off-peak baskets February 2001 – August 2001

2.19 From Table 2.2 it is seen that:

• prices in the UK have risen particularly for the low usage metered baskets (given the relatively low monthly spend on Internet access for these consumers, a relatively small change in £ per month gives rise to a large percentage change);
• overall prices for metered tariffs have fallen in the US and Germany, risen in France, and remain at similar levels in Sweden; and
• prices for unmetered dial-up Internet access are only available in the UK and US – they have risen slightly in the UK and fallen in the US.

2.20 However, price changes need to be considered together with price levels. The following chart (Figure 2.1) shows price changes for metered baskets relative to UK prices as at February 2000. Over the period February 2000 to August 2001 there has been some convergence of prices across the countries benchmarked.

2.21 Relative prices as at August 2001 on the basis of the new methodology (ie including broadband products) are also shown for comparison. The overall spread of prices is similar to that for dial up only, though there is some effect on individual countries ranking.

Figure 2.1: Price trends by country from February 2000 to August 2001

Price spread

2.22 The spread of prices was also considered. To make this meaningful, only dial-up packages have been included (otherwise the spread for the low usage baskets would be very wide, as broadband offers are not appropriate for these usage levels). The charts in Figure 2.2 show the price for the cheapest package offered by each operator as a percentage above the cheapest price across all countries, for each basket. Amongst other things, the price spread provides an indication of the level of choice for customers at, or around, the best deal. Figure 2.2 shows that the spread of prices varies by basket. The spread of unmetered prices is not presented due to the relatively small number of packages currently on offer, ie only available in the UK and US.

2.23 The price spread for low usage baskets is particularly large for the UK, US and France. In these countries there are some operators who offer only unmetered/partly unmetered packages and these are not cost-effective at low levels of usage. For higher usage the spread is generally narrower, with the highest spreads seen in the UK and the lowest in the US.

Figure 2.2: Price spread charts for residential off-peak baskets (dial-up only)

Baskets are shown in order of usage level

Price index for residential peak services

2.24 The price index (based on the average of the two cheapest packages from different operators in each country) is presented in Table 2.3 for residential peak baskets.

Table 2.3: Comparison of residential peak package prices

2.25 From Table 2.3 it is seen that:

• overall, UK prices for limited use are the cheapest of the European countries studied – only prices in the US are lower; and
• for unmetered access, UK prices are also the lowest of the European countries studied, with prices in the US 4-8% cheaper.

2.26 As for the off-peak baskets, in several countries, cable modem or DSL services appear as the cheapest package for some baskets. In France, Germany and Sweden there are no dial-up unmetered packages (with the exception of one package in Germany – but this is found to be expensive and does not feature in the index), and prices for the unmetered basket are based on DSL/cable modem offers. In Germany and Sweden, cable modem or DSL packages start to become cost effective at 20 hours usage per month.

2.27 The results present a positive picture for the UK, however, there has been some movement in the UK's relative position – in the previous study the UK was the cheapest overall by a fairly large margin. This is due to the factors listed in paragraph 2.18.

Price trend

2.28 A summary of price changes since the February 2001 study is provided in Table 2.4. The average price change shown is the mean of the percentage changes across each of the original four baskets. To ensure the price trend is meaningful, only dial-up packages have been included in this analysis.

Table 2.4: Percentage change in price levels for residential peak baskets February 2001 – August 2001

2.29 From Table 2.4 it is seen that:

• there has been a rise in the cost of metered baskets in the UK and France, particularly for low usage baskets (given the relatively low monthly spend on Internet access for these consumers, a relatively small change in £ per month gives rise to a large percentage change);
• prices for metered Internet access have fallen elsewhere, particularly in Germany;
• prices for unmetered dial-up access (available only the UK and US) have fallen (but only slightly in the UK).

2.30 However, price changes need to be considered together with price levels. The following chart (Figure 2.3) shows price changes for metered baskets relative to UK prices as at February 2000. Compared to off-peak prices, there is less convergence in pricing.

2.31 Relative prices as at August 2001 on the basis of the new methodology (ie including broadband products) are also shown for comparison. The spread of prices is much narrower when broadband services are also included.

Figure 2.3: Price trends by country from February 2000 to August 2001

Price spread

2.32 The spread of prices was also considered. To make this meaningful, only dial-up packages have been included (otherwise there would be a very wide spread for the low usage baskets, as this would include broadband offers not appropriate to these usage levels). The charts in Figure 2.4 show the price for the cheapest package offered by each operator as a percentage above the cheapest price across all countries, for each basket.

2.33 The price spread in the UK is wide – to some extent this reflects the fact that some operators may only offer unmetered packages (expensive for low users) or pay as you go (expensive for high users).

Figure 2.4: Price spread charts for residential peak baskets (dial-up only)