COMMUNICATIONS COMMITTEE

Transcription

1 EUROPEAN COMMISSION Directorate-General for Communications Networks, Content and Technology Coordination Knowledge Base Brussels, 25 March 2014 DG CNECT/F4 COCOM COMMUNICATIONS COMMITTEE Working Document Broadband access in the EU: situation at 1 July 2013 This is a Committee working document which does not necessarily reflect the official position of the Commission. No inferences should be drawn from this document as to the precise form or content of future measures to be submitted by the Commission. The Commission accepts no responsibility or liability whatsoever with regard to any information or data referred to in this document. Commission européenne, B-1049 Bruxelles/ Europese Commissie, B-1049 Brussel Belgium. Telephone: (32-2) Office: BU25 1/174. Telephone: direct line (32-2) Fax: (32-2) cnect-f4@ec.europa.eu 1

2 Explanatory note The Commission has been publishing data on the number of broadband subscriptions in the Member States gathered in the context of the Communications Committee (COCOM) since The COCOM broadband report has been expanded to keep up with the technological changes of this dynamic market, and now includes information also on mobile broadband and speeds, as well as on retail prices. The information on speeds is especially relevant now, since the broadband targets of the Digital Agenda for Europe are expressed in terms of speeds. The report consists of: A Word document with the analysis of the data collected from the relevant ministries and regulatory authorities and other broadband-related statistics published by the Commission (data on coverage, retail and wholesale prices and the broadband state aid cases), An Excel document presenting the data tables. 2

3 Broadband access in the EU: situation at 1 July Executive summary 6 2. Fixed broadband take-up 7 3. Fast and ultrafast broadband take-up Fixed broadband technologies Fixed broadband market shares Fixed broadband speeds Fixed broadband retail prices Mobile broadband take-up Annexes 37 ANNEX 1: Definitions 37 ANNEX 2: Methodology 40 ANNEX 3: Prices of the local loop 43 ANNEX 4: Broadband state aid developments 45 3

4 List of figures Figure 1 Fixed broadband penetration at EU level, January 2004 July Figure 2 Fixed broadband subscriptions growth per day at the EU level, January 2004 July Figure 3 Fixed broadband penetration (subscriptions as a % of population), July Figure 4 Coefficient of variation Fixed broadband penetration (subscriptions as a % of population), July Figure 5 Fixed broadband penetration and speed of progress, July July Figure 6 Fast and ultrafast broadband penetration (subscriptions as a % of population) at EU level, January July Figure 7 Fast broadband (at least 30Mbps) penetration (subscriptions as a % of population), July Figure 8 Total NGA broadband coverage and high-speed broadband penetration Figure 9 Ultrafast broadband (at least 100Mbps) penetration (subscriptions as a % of population), July Figure 10 Fixed broadband subscriptions technology market shares at EU level, January July Figure 11 Fixed broadband net adds by technology at EU level, January July Figure 12 Fixed broadband subscriptions, technology market shares, July Figure 13 DSL subscriptions and market growth in the EU, January July Figure 14 Cable broadband subscriptions and market growth at EU level, January July Figure 15 NGA subscriptions at EU level, January 2012 July Figure 16 NGA subscriptions as a % of total fixed broadband subscriptions, July Figure 17 Fixed broadband subscriptions by operators at EU level, January July Figure 18 Fixed broadband subscriptions operator market shares at EU level, January July Figure 19 New additions market share per day at EU level, January July Figure 20 Fixed broadband subscriptions growth per day by operator at EU level, January July Figure 21 Fixed broadband subscriptions operator market shares, July Figure 22 DSL subscriptions operators' market shares at EU level, January July Figure 23 DSL subscriptions operators' market shares, July Figure 24 New entrants' DSL subscriptions by type of access at EU level, July Figure 25 New entrants' DSL subscriptions by type of access at EU level, July July Figure 26 Cable broadband subscriptions operator market shares at EU level, July Figure 27 NGA subscriptions incumbent market share at EU level, January 2012 July Figure 28 Fixed broadband subscriptions by speed at EU level, January July Figure 29 Fixed broadband subscriptions by speed, July Figure 30 Fixed broadband subscriptions by speed at EU level (Digital Agenda categories), July Figure 31 Fixed broadband subscriptions by speed (Digital Agenda categories), July Figure 32 Broadband retail prices standalone offers 12 to 30 Mbps, Figure 33 Broadband retail prices - Internet Access 12 to 30 Mbps + Fixed Telephony, Figure 34 Broadband retail prices Standalone offers at least 30 Mbps,

5 Figure 35 Broadband retail prices Internet Access at least 30 Mbps + Fixed Telephony, Figure 36 Share of median cost of Internet access in real adjusted gross disposable income of households per capita and fixed broadband take-up Figure 37 Mobile broadband penetration at EU level, January July Figure 38 Mobile broadband penetration all active users, July Figure 39 LTE coverage and Mobile broadband take-up Figure 40 Fixed broadband take-up and mobile broadband take-up July Figure 41 LLU monthly average total cost at EU level, Figure 42 LLU monthly average total cost, October 2011 October

6 1. Executive summary Broadband markets in Europe are fragmented. Looking at any of the key indicators measuring coverage/availability and take-up of fixed, NGA and mobile broadband, there are significant differences across Member States. The same can be observed when considering prices, service quality and competition dynamics. Although convergence is increasing in overall fixed broadband take-up, less mature segments of the market (NGA and mobile) are naturally characterised by higher divergence. Fixed broadband take-up (lines as a percentage of population) reached 29.4% in July 2013, which corresponds to a household penetration of 76%. Denmark (40.4%), the Netherlands (40.4%), France (37.3%) and Germany (34.6%) have the highest penetration rates, while Romania (18.9%), Bulgaria (19.2%), Slovakia (20.1%) and Poland (20.8%) the lowest. Broadband subscriptions are getting faster. There were 5.4 subscriptions of at least 30Mbps per hundred people as opposed to 3.4 a year ago. Fast broadband is most widely used in Belgium (20.7), the Netherlands (15.6) and Lithuania (12.7). Superfast broadband of at least 100Mbps is still very rare in Europe (1.2 subscriptions per hundred people). Take-up of Next Generation Access technologies increased sharply by 42% during the last twelve months; one in four subscriptions were NGA in July NGA already represents the majority of subscriptions in Romania, Belgium, Bulgaria, the Netherlands, Sweden, Latvia, Lithuania and Portugal, while it remains marginal in Greece, Croatia, Cyprus and Italy. Cable Docsis 3.0 is still the most widely used NGA technology with more than 50% of NGA subscriptions, followed by VDSL. The market share of incumbent operators continued to decline slightly and stood at 41.8% in July New entrants recorded 80.1% of all net additions (subscription growth) in the first half of On the other hand, new entrants provided 77% of NGA subscriptions. The number of mobile broadband users increased by 20.7% in the last twelve months and penetration (subscriptions as a % of population) reached 58%. Mobile broadband is most widely used in Finland, Sweden and Denmark, where there are already more than 100 subscriptions per 100 people. 6

7 2. Fixed broadband take-up In July 2013, there were 29.4 fixed broadband subscriptions per 100 inhabitants in the EU, which corresponds to a household penetration of around 76% 1. Compared to January 2013, this is an increase of 0.7 percentage points. Over 17,000 new lines were connected per day in the first six months in 2013, as opposed to only 12,600 in the first half of Figure 1 Fixed broadband penetration at EU level, January 2004 July 2013 Figure 2 Fixed broadband subscriptions growth per day at the EU level, January 2004 July Source: Eurostat 7

8 Denmark (40.4%), The Netherlands (40.4%), France (37.3%) and Germany (34.6%) remained on the top of the list followed by Malta (33.8%), Belgium (33.8%) and the UK (33.3%). Sweden (32.7%), Luxembourg (32.5%) and Finland (30.6%) are also above the EU average; the latter two are especially advanced in mobile broadband along with Denmark. At the bottom of the list, there were only two Member States (Romania and Bulgaria) below 20% followed by Slovakia (20.1% and Poland 20.8%). All these four Member States managed to lower the gap in the last twelve months. Figure 3 Fixed broadband penetration (subscriptions as a % of population), July 2013 Although fixed broadband penetration still varies considerably across Member States, there is an increasing convergence between 2008 and Indeed, the coefficient of variation 2 of fixed broadband penetration, which is a measure of the dispersion between countries, decreased from 38% in 2008 to 22% in This measure is computed as the ratio of the standard deviation of the distribution of fixed broadband penetration across Member States to the average fixed broadband penetration in the EU. 8

9 Figure 4 Coefficient of variation Fixed broadband penetration (subscriptions as a % of population), July Figure 5 Fixed broadband penetration and speed of progress, July July 2013 Countries where standard fixed broadband 3 has higher coverage tend to have higher broadband take-up, but the correlation is not strong. Looking at Member States with very high coverage, there are still large differences in broadband take-up. In these countries, other variables (such as income, prices, education and skills) explain the fixed broadband take-up. At the same, low coverage is a strong barrier to take-up in Poland, Slovakia, Slovenia and Latvia. 3 Source on coverage figures: Point Topic. Standard fixed broadband coverage refers to the overall coverage of xdsl, cable broadband, FTTP and WiMax technologies. 9

10 3. Fast and ultrafast broadband take-up The Digital Agenda for Europe set two objectives related to fast and ultrafast broadband: Fast broadband (at least 30Mbps) should be available to all by 2020, and 50% of homes should subscribe to ultra-fast broadband (at least 100Mbps) by As of end 2012, fast broadband was available to 53.8% of homes, ranging from 14% in Italy to 99.9% in Malta. 4 The actual take-up remains well below the coverage: there were only 5.4 subscriptions per hundred inhabitants in the EU in July 2013, which corresponds to an estimated household penetration of 14%. Nevertheless, fast broadband subscriptions increased by 59% in the last twelve months. As of July 2013, close to one in five fixed broadband subscriptions was at least 30Mbps download. Ultrafast connections (at least 100Mbps) are still rare in the EU, penetration (subscriptions as a percentage of population) stood at 1.2% in July Figure 6 Fast and ultrafast broadband penetration (subscriptions as a % of population) at EU level, January July 2013 Fast broadband connections are most widely used in Belgium, the Netherlands and Lithuania. Cable Docsis 3.0 and VDSL play a major role in Belgium and the Netherlands, while in Lithuania FTTB and FTTH are the prevailing technologies. At the bottom of the least, three Member States (Greece, Italy and Cyprus) have less than one fast broadband subscription per hundred people. 4 Source: Point Topic, Broadband coverage in Europe in

11 Figure 7 Fast broadband (at least 30Mbps) penetration (subscriptions as a % of population), July 2013 The correlation between high-speed broadband penetration and NGA broadband coverage is +57%. Countries with higher NGA coverage/availability tend to have higher take-up of at least 30Mbps connections, but there are large differences across the Member States. In Belgium, the Netherlands, Luxembourg and Malta, NGA is available to more than 90% of households, but the take-up varies greatly: Belgium has over 20 subscriptions per 100 people, while Malta below 5. Italy, Greece, Cyprus and Slovenia show a particularly low high-speed broadband take-up despite the fact that coverage is above the average in Slovenia and Cyprus. Figure 8 Total NGA broadband coverage and high-speed broadband penetration 11

12 Ultrafast connections are marginal in most of the Member States. Sweden is a striking exception with nine ultrafast subscriptions per hundred inhabitants. Latvia, Belgium, Romania, Finland and Portugal follow Sweden with % penetration of at least 100Mbps subscriptions. Figure 9 Ultrafast broadband (at least 100Mbps) penetration (subscriptions as a % of population), July

13 4. Fixed broadband technologies Although the market share of xdsl continued to slightly decline in the first six months in 2013, it still represented close to three quarters of all fixed broadband subscriptions. The market share of xdsl is 8 percentage points lower than in January Cable is the second most widely used fixed internet access technology in the EU with a share of 17% in total subscriptions. Figure 10 Fixed broadband subscriptions technology market shares at EU level, January July

14 Net additions increased both in cable and in DSL, but over one third of net additions came from technologies other than cable or DSL. Figure 11 Fixed broadband net adds by technology at EU level, January July 2013 The market share of DSL varies between 20% in Bulgaria and 100% in Greece. In about half of the EU Member States, more than 50% of fixed broadband lines are xdsl. Cable is still the main challenger of xdsl, especially in urban areas. Cable broadband is available in all but two Member States (Italy and Greece). Cable is the most widely used in Belgium (50%), Malta (48%), Hungary (47%) and the Netherlands (45%). At the same time, technologies other than xdsl and cable represent the majority of subscriptions in Lithuania (72%), Bulgaria (66%), Romania (63%) and Latvia (63%). All these four Member States have a strong presence of fibre based technologies. 14

15 Figure 12 Fixed broadband subscriptions, technology market shares, July 2013 Figure 13 DSL subscriptions and market growth in the EU, January July

16 Figure 14 Cable broadband subscriptions and market growth at EU level, January July 2013 Next Generation Access (NGA) technologies providing fast and ultrafast broadband represent a growing share in subscriptions. In this report, Fibre to the Home (FTTH), Fibre to the Building (FTTB), VDSL, Cable Docsis 3.0. and other NGA (any other technology capable of at least 30 Mbps download) are considered as NGA technologies. In July 2013, there were 35.5 million NGA subscriptions in the EU, up from 25.2 million a year before. One in four fixed broadband subscriptions are already NGA. Cable Docsis 3.0 is by far the most popular NGA technology with over 50% of NGA subscriptions. By the end of 2012, more than 90% of European cable networks were upgraded to Docsis 3.0 providing fast broadband, and by July 2013, three out of four cable subscriptions were migrated to NGA cable. Much weaker progress can be observed regarding VDSL. Although VDSL subscriptions increased by 86% in the last twelve months, they represent still only 5% of all xdsl connections. It is also to be noted that VDSL availability (24.9% of homes) falls well below that of cable NGA (39.4%) 16

17 Figure 15 NGA subscriptions at EU level, January 2012 July 2013 Member States vary greatly based on the take-up of NGA. On the one hand, the majority of fixed broadband lines are NGA in eight Member States: Romania, Belgium, Bulgaria, the Netherlands, Sweden, Latvia, Lithuania and Portugal. On the other hand, Greece, Croatia, Cyprus and Italy feature figures below 3%. Figure 16 NGA subscriptions as a % of total fixed broadband subscriptions, July 2013 Cable technology has the largest NGA customer base; at EU level it represents 55% of NGA lines. VDSL is the second largest technology with 16% of NGA lines followed by FTTB (15%) and FTTH (9%). 17

18 Table 1 NGA subscriptions by technology, July 2013 Jul-13 vdsl Fibre to the building - Fibre to the Home Ethernet + LAN Cable Other NGA SUM (FTTB) BE Confidential Confidential BG CZ DK DE EE Confidential Confidential Confidential Confidential EL ES FR HR IE Confidential IT CY LV LT LU HU Confidential Confidential MT NL Confidential AT PL PT RO SI SK FI SE UK

19 5. Fixed broadband market shares Market shares are calculated at the national level for the incumbent operators and new entrants. However, broadband markets are geographically fragmented in many cases suggesting that a large number of homes may be served by only one provider despite the competition at national level. There was very little change in the average market share of incumbent operators in the EU in the last few years. Incumbents had 41.8% of all fixed broadband subscriptions in July 2013, which is a slight decline of 0.9p.p compared to the situation a year ago. Figure 17 Fixed broadband subscriptions by operators at EU level, January July 2013 Figure 18 Fixed broadband subscriptions operator market shares at EU level, January July

20 New entrants are slightly gaining market share year by year, which is reflected in the chart below showing the net adds (subscriptions growth) by operators. New entrants were responsible for 70-80% of the market growth in the recent years. Figure 19 New additions market share per day at EU level, January July Figure 20 Fixed broadband subscriptions growth per day by operator at EU level, January July The peak in July 2010 was due to a revision of the data series by some NRAs 20

21 Looking at the Member State level, the market share of incumbent operators was the highest in Luxembourg (71%) and Cyprus (67%) and the lowest in the Bulgaria (23%), Poland (26%) and Romania (28%). Incumbents managed to gain market share in seven Member States (Estonia, Portugal, Sweden, the Netherlands, Greece, Latvia and the UK), but the largest growth was only 1.3 p.p. At the same time, new entrants gained over 2 p.p in Croatia, Slovakia, Cyprus, Ireland, Poland, Slovenia and Austria. Figure 21 Fixed broadband subscriptions operator market shares, July 2013 In the xdsl market, incumbent operators still have a share of more than 50% of subscriptions, while new entrants progressed only marginally (0.5 p.p) in the last twelve months. Figure 22 DSL subscriptions operators' market shares at EU level, January July

22 In Bulgaria, Romania, Latvia, Malta, Estonia and Lithuania there is virtually no competition in the DSL retail market. In all these Member States, however, there is a strong platform competition, and the share of DSL subscriptions in total fixed broadband is much lower than the EU average. At the same time, in the UK, Greece and France, new entrants have the majority of xdsl subscriptions followed by Italy (48%) and Germany (47%). In all these Member States, the vast majority of new entrants' xdsl subscriptions are full LLU, but in Italy bitstream access is also significant (29% of all new entrant xdsl subscriptions). Figure 23 DSL subscriptions operators' market shares, July

23 At the European level, the share of unbundled lines (either full LLU or shared access) stood at 82.1% of new entrants xdsl subscriptions in July There has been a strong migration towards full LLU from the other access types. Figure 24 New entrants' DSL subscriptions by type of access at EU level, July 2013 Figure 25 New entrants' DSL subscriptions by type of access at EU level, July July

24 Cable broadband is almost exclusively provided on new entrants' networks. The only exceptions are Finland, Denmark and Hungary, where incumbent operators have a considerable presence on the cable market. Figure 26 Cable broadband subscriptions operator market shares at EU level, July 2013 As far as NGA technologies are concerned, incumbent operators are well behind new entrants, as over 3 out of four NGA subscriptions belong to alternative operators. In technologies other than VDSL, incumbents have fairly low market shares. In VDSL, incumbents hold 93% of the subscriptions, which is much higher than their overall xdsl market share (54%). Figure 27 NGA subscriptions incumbent market share at EU level, January 2012 July

25 Table 2 Incumbent market share in NGA Incumbent market share Fibre to the Date building - vdsl Fibre to the Home Ethernet + LAN Cable Other NGA Total (FTTB) Jan-12 95% 48% 11% 2% 3% 22% Jul-12 94% 52% 12% 1% 3% 19% Jan-13 94% 57% 12% 2% 1% 21% Jul-13 93% 59% 13% 2% 0% 23% 25

26 6. Fixed broadband speeds In terms of headline speeds, as of July 2013, close to two out of three fixed broadband subscriptions provided download speeds of 10 Mbps and above. This is a large progress, as less than one in ten of subscriptions were that fast in January Subscriptions with below 2 Mbps have become marginal in the EU. Figure 28 Fixed broadband subscriptions by speed at EU level, January July 2013 Figure 29 Fixed broadband subscriptions by speed, July 2013 However, the Digital Agenda calls for fast and ultrafast broadband, which is still rare in Europe. In July 2013, only one in five subscriptions were at least 30 Mbps and only 4.2% at least 100 Mbps. This also reveals that one in four NGA subscriptions have less than 30 Mbps headline download speed. 26

27 Figure 30 Fixed broadband subscriptions by speed at EU level (Digital Agenda categories), July 2013 There are only two Member States with more than 50% of subscriptions of at least 30 Mbps (Belgium and Romania), while the same ratio is less than 10% in Italy, Greece, Cyprus, Slovenia and France. In ultrafast broadband (at least 100Mbps), Sweden, Latvia and Romania are the most advanced with at least 20% of subscriptions. Figure 31 Fixed broadband subscriptions by speed (Digital Agenda categories), July

28 7. Fixed broadband retail prices 6 This section compares the prices of the most popular categories of fixed broadband products. Purchasing Power Parity is introduced to better determine the relative costs of broadband offerings. These comparisons must be interpreted taking into account the disparities in consumption patterns of different Member States. In particular, comparisons should be done looking at the speeds, types of offers and technologies that are most popular in each Member State. The charts below are boxplots presenting prices for standalone broadband products and bundles of broadband and fixed telephony, with download speeds of 12 to 30 Mbps and at least 30 Mbps. A boxplot is a graphic that helps the reader to have a quick overview of the distribution of the data and helps to summarize the data at the same time. We use several statistical indicators to summarize the price of the offers in a boxplot: Median value: 50% of the offers have a price below this value and 50% of the offers have a price above this value Mean value: the sum of the prices divided by the number of offers (plotted with a diamond on the following charts) Minimum: the minimal price of the offers, excluding outlier(s) Maximum: the maximal price of the offers, excluding outlier(s) P25 (25th percentile): one-quarter of the offers have a price below this value and three-quarters of the offers have a price above this value P75 (75th percentile): three-quarters of the offers have a price below this value and one-quarter of the offers have a price above this value Outliers: offers with a price lower or higher than 1.5 times the interquartile range (the difference between P75 and P25) 6 This section has been drafted based on the study "Broadband internet access cost 2013" by Van Dijk. 28

29 At Member State level, minimum prices of broadband standalone products with a download speed between 12 and 30 Mbps were the lowest in Lithuania ( 10.3), Romania ( 11.2) and Latvia ( 14.6) and the highest in Cyprus ( 46.2), Spain ( 38.7) and Ireland ( 31.4). Median prices were the lowest in Lithuania ( 14.8), Bulgaria ( 19.2), and the Netherlands ( 19.7) and the highest in Croatia ( 89.3), Cyprus ( 78.8) and Hungary ( 56.7). The interquartile range was particularly high in Hungary, Cyprus and Poland, indicating relatively high price dispersion. In Estonia, Malta, Cyprus and the Netherlands, less than 5 offers were available, suggesting that this may not be a typical type of offer in these countries. Figure 32 Broadband retail prices standalone offers 12 to 30 Mbps, 2013 Source: EC services based on Van Dijk 29

30 Minimum prices of bundles of fixed broadband access (with a download speed between 12 and 30 Mbps) and fixed telephony were the lowest in Romania ( 16.9), Italy ( 22.7) and Austria ( 24.6) and the highest in Cyprus ( 70.4), Croatia ( 52.1) and Hungary ( 49.7). Median prices were the lowest in the Slovakia ( 28.6), Sweden ( 29.2) and Denmark ( 32.4) and the highest in the Croatia ( 87), Cyprus ( 76) and Ireland ( 62). The interquartile range was the highest in the Cyprus, Croatia and Slovenia, indicating relatively high price dispersion. 7 countries had less than 5 offers in this category and there was no offer available in Estonia and Lithuania. Figure 33 Broadband retail prices - Internet Access 12 to 30 Mbps + Fixed Telephony, 2012 Source: EC services based on Van Dijk 30

31 Figure 37 shows the boxplot of offers including broadband internet access only, with a download speed of at least 30 Mbps. The minimum prices were the lowest in Lithuania ( 11.3), Sweden ( 11.8) and Romania ( 12) and the highest in Cyprus ( 78.9, Malta ( 69.4) and Spain ( 64.2). The median prices were the lowest in Romania ( 23.4), Lithuania ( 24) and Latvia ( 26.4) and the highest in Malta ( ), Cyprus ( 96.4) and Slovenia ( 80.4). The interquartile range was particularly high in Portugal, Hungary and Slovenia, indicating relatively high price dispersion. There was no offer available in Croatia and less than 5 offers were available in Italy, Ireland and Spain, suggesting that this may not be a typical type of offer in these countries. Figure 34 Broadband retail prices Standalone offers at least 30 Mbps, 2013 Source: EC services based on Van Dijk 31

32 Finally, looking at the prices of bundles of fixed broadband access (with a download speed of at least 30 Mbps) and fixed telephony, minimum prices were lowest in Sweden ( 19), Romania ( 19.4) and Austria ( 24.6), and highest in the Croatia ( 102.7), Cyprus ( 81.6) and Malta ( 75.2). Median prices were lowest in Romania ( 32.4), France ( 36.5) and Sweden ( 36.8) and highest in Malta ( 116.6), Croatia ( 102.7) and Cyprus ( 102.4). In 3 out of the 28 countries, no offer was available for this basket (Bulgaria, Finland and Lithuania) and there were less than 3 offers available in Croatia, Cyprus and Italy, suggesting that this may not be a typical type of offer in these countries. Figure 35 Broadband retail prices Internet Access at least 30 Mbps + Fixed Telephony, 2013 Source: EC services based on Van Dijk 32

33 The correlation between fixed broadband take-up and the relative price of Internet access is negative (-66%). In other words, broadband take-up tends to be lower in countries where the cost of Internet access accounts for a higher share of income. Because income is correlated with other explanatory variables (skills, education, digital awareness, computer ownership, etc.), the existence of this correlation is insufficient to demonstrate a causality between broadband unaffordability and broadband take-up. In Hungary and Cyprus, the cost of broadband access is higher than 5% of the disposable income and fixed broadband take-up is relatively low in these countries. In comparison, the cost of broadband access accounts for around 1% of the disposable income in Sweden and in the Netherlands, where fixed broadband take-up is relatively high. Figure 36 Share of median cost of Internet access in real adjusted gross disposable income of households per capita and fixed broadband take-up 2013 Source: EC services based on Van Dijk, Eurostat and COCOM 33

34 8. Mobile broadband take-up Mobile broadband is mainly provided on HSPA networks, which was available to 96.3% at the end of Europe is lagging behind in 4G LTE deployments with a coverage of only 26%. 7 Mobile broadband is still growing faster than fixed broadband in terms of subscriptions, but the growth slowed down in the last two year. The total number of active mobile broadband users increased by 20.7% in twelve months, reaching a penetration rate (subscriptions as a percentage of population) of 58%. Nordic countries remained the leaders in the use of this technology with penetration rate exceeding 100% in Sweden, Finland and Denmark. On the other hand, Hungary and Romania registered the lowest penetration with figures of 25% and 31% respectively. Figure 37 Mobile broadband penetration at EU level, January July Source: Point Topic 34

35 Figure 38 Mobile broadband penetration all active users, July 2013 The correlation between 4th generation (LTE or 4G) mobile broadband coverage and mobile broadband take-up is medium (+48%). Countries with the highest 4G mobile broadband coverage tend to have the highest mobile broadband take-up. Finland, Sweden and Denmark have high LTE coverage and high mobile broadband take-up, while Belgium, Bulgaria and Cyprus have low LTE coverage and low mobile broadband take-up. Despite a high LTE coverage, Portugal has one of the lowest mobile broadband take-up rates in the EU. Figure 39 LTE coverage and Mobile broadband take-up Source: Point Topic and COCOM 35

36 The correlation between fixed and mobile broadband take-up is also medium (+40%). Countries with relatively high fixed broadband take-up tend to have relatively high mobile broadband take-up. However, fixed broadband take-up tends to be lower in countries where both mobile broadband take-up and LTE coverage are high. This could mean that citizens in these countries start to see 4G mobile broadband as a substitute of fixed broadband. Denmark, the UK and Luxembourg have both a relatively high fixed broadband take-up and a relatively high mobile broadband take-up. On the contrary, the fixed and mobile take-up in Romania, Bulgaria, Hungary and Portugal are lower than the EU average. Figure 40 Fixed broadband take-up and mobile broadband take-up July 2013 Source: COCOM 36

37 9. Annexes ANNEX 1: Definitions Definitions Broadband connection: a connection enabling higher than 144 Kbit/s download speed. As of January 2010 it is estimated that 1-2 Mbps is the minimum download speed and that just a fraction of all retail broadband lines provide speeds of 144 Kbit/s. Broadband penetration: Broadband subscription lines as a percentage of population. Incumbents: Organisations having enjoyed special and exclusive rights or de facto monopoly for the provision of voice telephony services before liberalisation, regardless of the role played in the provision of access by means of technologies alternative to the PSTN. New entrant: Alternative telecommunications operators, as well as internet service providers (ISPs). DSL: Digital Subscriber Line. Cable broadband: Broadband connections by means of cable TV access. Satellite: Broadband connections via satellite. NGA: Next Generation Access Technologies including VDSL, FTTH, FTTB, Cable NGA and other NGA as defined below. VDSL: Very high bitrate digital subscriber line. It uses copper networks in the access. Typically the physical network interface at the delivery point at subscriber's home would be a RJ-11 type connector. Fibre to the Node + vdsl lines should be included in this category. VDSL is deployed over existing wiring used for analog telephone service and lower-speed DSL connections. FTTH: Fibre to the Home. A communications architecture in which the final connection to the subscriber s premises is Optical Fibre. The fibre optic communications path is terminated on or in the premise for the purpose of carrying communications to a single subscriber. In order to be classified as FTTH, the access fibre must cross the subscriber s premises boundary and terminate: inside the premises, or on an external wall of the subscriber s premises, or not more than 2m from an external wall of the subscriber s premises. 37

38 FTTH services may deliver just one application, but generally deliver several such as data, voice and video. This FTTH definition excludes architectures where the optical fibre terminates in public or private space before reaching the premises and where the access path continues to the subscriber over a physical medium other than optical fibre (for example copper loops, power cables, wireless and/or coax). 8 FTTB: Fibre to the Building: An optical fiber reaches the boundary of the building, such as the basement in an multidwelling unit, and the final connection to the subscriber s premises is a physical medium other than Optical Fiber. NGA cable: Lines transmitting very high-speed data transfer on an existing coaxial cable TV network. Typically the physical network interface at the delivery point at subscriber's home would be an F connector type. Other NGA: Technologies other than FTTH, FTTB, VDSL and Cable NGA, which are capable of at least 30 Mbps download (headline speed). Fully unbundled lines: Fully unbundled lines supplied by the incumbent operator to other operators (new entrants), excluding experimental lines. In the case of full unbundling, a copper pair is rented to a third party for its exclusive use. As fully unbundled lines (LLU) supplied by the incumbent operator to the new entrants could in principle be used for services other than broadband, the total number of LLU for access to internet will be lower than the total number of LLU. Shared access lines: Shared access lines supplied by the incumbent to other operators (new entrants), excluding experimental lines. In the case of shared access, the incumbent continues to provide telephony service, while the new entrant delivers high-speed data services over that same local loop. Bitstream access: It refers to the situation where the incumbent installs a high-speed access link to the customer premises, and makes this access link available to third parties (new entrants), to enable them to provide high-speed services to customers. Bitstream depends in part on the PSTN, and may include other networks such as the ATM network. Bitstream access is a wholesale product that consists of the provision of transmission capacity in such a way as to allow new entrants to offer their own, value-added services to their clients. The incumbent may also provide transmission services to its competitor, to carry traffic to a 'higher' level in the network hierarchy where new entrants may already have a broadband point of presence. Simple resale: In contrast to bitstream access, simple resale occurs when a new entrant receives and sells on to end users a product (with no possibility of value added features to the DSL part of the service) that is commercially similar to the DSL product provided by the incumbent to its own retail customers, irrespective of the ISP service that may be packaged with it. Resale offers are not a substitute for bitstream access because they do not allow new entrants to differentiate their services from those of the incumbent (i.e. when the 8 Source: FTTH Council 38

39 new entrant simply resells the end-to-end service provided to him by the incumbent on a wholesale basis). Mobile broadband: internet access on third generation technologies (3G) and higher speed mobile technologies (i.e. HSPA or LTE). In the case of UMTS the unit of reference is SIM/USIM cards (including modem/dongles). For the CDMA standard, the unit of measurement should be the number of User Equipments. Mobile broadband standard mobile subscriptions: Number of subscriptions which have made an Internet mobile connection in the last 90 days through a standard mobile subscription. Standard mobile subscriptions are typical voice subscriptions which also provide access to the Internet but are not purchased separately. Standard mobile subscription excludes dedicated Internet mobile subscriptions. An Internet mobile connection is a connection to the open Internet using Internet Protocol (IP). Hence, subscriptions which only offer walled garden or -only services (or SMS/MMS only) as well as those offering access to the open Internet but that only have made access to "walled garden" and -only services in the last three months will not be considered. Bundled offers (i.e., voice and data access) for a unique (flat rate) tariff are to be counted if a data connection has been made in last 3 months. Mobile broadband - Dedicated data subscriptions for stand-alone services via cards/modems/keys only: Number of subscriptions to dedicated data services over a mobile network which are purchased separately from voice services as a stand alone service (modem/dongle), i.e. excluding mobile handset users. All dedicated data subscriptions with a recurring subscription fee are included as "active data subscriptions", regardless of actual use. Pre-paid mobile broadband plans (i.e. all non- recurrent fee subscriptions) require active use in previous 3 months. Subscriptions which only offer walled garden or -only services (or SMS/MMS only) will not be considered. Bundled offers (i.e., voice and data access) are excluded. Dedicated data subscriptions for add-on data package to a voice service requiring an additional subscription: Number of subscriptions to dedicated data services over a mobile network which are purchased separately from voice services as an add-on data package to voice service which require an additional subscription (i.e. excluding datacards/dongles). Recurrent fee subscriptions (i.e., contract) are included automatically. Prepayment subscriptions (or any other type of non-recurrent subscription) need to pass the activity criterion (a usage occurred in the last 3 months). Subscriptions which only offer walled garden or -only services (or SMS/MMS only) will not be considered. Bundled offers (i.e., voice and data access) are excluded. 39

40 ANNEX 2: Methodology The data (except for broadband coverage and broadband retail prices) have been collected by the European Commission, Communications Networks, Content and Technology Directorate General, from national ministries and regulatory authorities except when noted. The definitions have been agreed in the Communications Committee (COCOM). In some cases information for some types of access is not available. In a number of countries certain figures are estimates, as the National Regulatory Authorities had not received consolidated data from operators. It should also be noted that in some cases information only refers to major broadband access providers and that broadband access lines provided by small operators are not included. This report includes information from all 28 Member States. Data should be interpreted taking the below comments of Member States into consideration: Bulgaria January 2013 figures have been updated. Retail lines: Figures are based on expert estimation based on data submitted to CRC by 90.3% of registered undertakings. Retail lines by speed: Figures are based on expert estimation based on data submitted to CRC by 90.3% of registered undertakings. Denmark Retail lines: New numbers for NGA (VDSL, Cable DOCSIS 3.0, FiberLAN). Germany Wholesale: Incumbent activated main lines include retail analogue, ISDN and All-IP access lines. Wholesale unbundled lines, shared lines, bitstream access and simple resale are also included. Number of agreements on simple resale: data as of 01/01/2013. Retail lines: New entrants' VDSL own network DSL figures are based on estimates. Full ULL includes wholesale DSL lines (bitstream access, DSL resale) supplied by alternative operators to other alternative operators on the basis of unbundled local loops provided by the incumbent. Bitstream access includes only bitstream access lines supplied by the incumbent to new entrants. For resale only DSL resale lines supplied by the incumbent to new entrants are included. For FWA number of subscribers is reported. In case of cable modem and cable NGA, lines on cable infrastructures based on FTTB/FTTH are excluded. In case of cable NGA, lines with a capacity equal to or higher than 30 Mbps are reported. Other wireline access supplied by the incumbent refers to the estimated number of leased lines. Other wireline access supplied by new entrants refers to the estimated number of leased lines and powerline access lines. Fibre lines (FTTB/FTTH) include fibre lines provided by cable operators.total number of new entrant operators (incl. municipalities): Broadband services are also offered by clubs in Germany. Broadband services of clubs are not included. 40

41 Operator switching: Maximum number of days (calendar days, 5 operators). Croatia Wholesale lines: Fully unbundled lines are as of 1 August Finland Wholesale and retail lines: There were 28 SMP operators in SMP operators are operating as the incumbent in their own operating area and as entrants in other incumbents' operating areas. FICORA only collects data at country level. Therefore, incumbent/entrant data is not available. Incumbent activated main lines includes all operators' activated lines from their own operating areas and from other incumbents' operating areas. NGA/Cable includes subscriptions which are offered with Docsis 3.0 technology (Docsis 3.0 is used in the network and in the end user premise). Nearly all the cable television networks are upgraded to Docsis 3.0, but not all end user have Docsis 3.0 equipment in use. FTTB includes fibre/ethernet, fibre/homepna and fibre to building subscriptions. Mobile broadband: FICORA does not have the data needed to provide exact figures for the required breakdown. All mobile data subscriptions in Finland can in general be used with all kinds of terminals, with more than one terminal, and can also be used for voice services. Therefore, the data subscriptions breakdown is reported based on estimates from a consumer survey. Ireland Mobile broadband: 2013 January figures were updated and are not comparable with earlier data. Lithuania Wholesale lines: The same 2 agreements can be used and for fully unbundled lines and for shared access lines. Retail lines: The incumbents LLU lines are used in retail by new entrants for the provision other data transmission services (e.g. VPN) and 98 lines are used for the provision of broadband Internet access services. Retail lines by speed: ensured downstream speed rates are reported. The Netherlands Mobile broadband: January 2013 and July 2013 figures are not comparable. Austria Retail lines: NGA is estimated based on speeds January figures are updated. Retail lines by speed: 2013 January figures are updated. Poland Retail lines: Figures on cable and other technologies have been updated. 41

42 Portugal Retail lines: Figures refer to internet access broadband lines only. Cable and cable NGA figures have been updated. Operator switching: Data as of Romania All data are provisional. Retail lines: Some of the NGA subscriptions were re-allocated. Mobile broadband: Active use criterion refers to the last 180 days. Sweden Retail lines: NGA data for January and July 2013 are not comparable. UK Retail lines: VDSL/FTTP split for new entrants is estimated. Mobile broadband: Data are as of January or March

43 ANNEX 3: Prices of the local loop This annex illustrates the cost of connection and monthly rental for both Fully Unbundled Access (full Local Loop Unbundling (LLU)) and Shared Access (SA) to the loop. Monthly rental and connection fees are presented as well as the total average monthly cost, which is calculated as the monthly fee + the connection fee distributed over three years. Unless otherwise stated, connection fees include the technical expertise to assess the speed that can be conveyed through and disconnection fees (where applicable). Furthermore, only the price for a single line is presented here (charges may be different in the case of subsequent access). It is assumed that the loop is active and it will be used to provide both telephony and DSL services. Unless otherwise stated, figures exclude a whole range of additional one-off costs that may exist in some Member States. The following charts illustrate the monthly total cost for the full Local Loop Unbundling and Shared Access (connection and monthly fees) based on the assumption that the loop is used for three years. The EU average since 2005 is also shown. Figure 41 LLU monthly average total cost at EU level,

44 Figure 42 LLU monthly average total cost, October 2011 October 2012 For methodological remarks please see: Definitions, methodology and footnotes on Member State data (pdf) 44

45 ANNEX 4: Broadband state aid developments In 2013, the European Commission took 15 decisions regarding broadband projects involving public funding. The total amount of the aid approved was approximately m. Regularly updated information on all Broadband State aid decisions can be found here: # Decision name MS Decision Date Aid amount (million ) SA (2012/N) Brandenburg 1 Glasfaser GER 17/01/ SA (2012/N) - Austrian federal 2 broadband scheme AT 18/01/ SA (2012/N) - Amendment of the State aid to broadband scheme within the framework of the rural development program (modification of N30/2010 and 3 SA.33221) SWE 24/01/ ,4 SA (2012/N) - Regional 4 broadband network of Rzeszow PL 16/04/2013 4,3 Decision type Type Article 4(3) NGA Article 4(3) NGA Article 4(3) NGA Article 4(3) Basic SA (2013/N) - Development of Rural Area Information Technology 5 Network-Amendment LT 13/05/ ,5 Article 4(3) Basic/ NGA SA.35027, SA & SA (2012/N) - Local broadband network 6 projects in Podlasie PL 06/06/ Article 4(3) Basic/ NGA SA (2012/N) - Extension of high 7 speed broadband in Spain ES 05/07/ Article 4(3) NGA SA (2012/N) - Broadband network 8 in Czestochowa region PL 29/08/2013 6,5 Article 4(3) NGA SA (2012/N) - Broadband network 9 in Cieszyn region PL 02/09/2013 4,8 Article 4(3) NGA SA (2013/N) - Entwicklungskonzept Brandenburg Article 4(3) NGA 10 Glasfaser 2020 II GER 12/09/ SA (2012/N) - Regional 11 broadband network of Lodz PL 30/09/ Article 4(3) NGA 12 SA (2012/NN) - Next Generation IRL 04/12/ Article 4(3) NGA (backhaul) Network (NGN) alongside a gas pipeline in Galway and Mayo 13 SA (2013/N) - Broadband network BG 09/12/ Article 4(3) Basic project in Bulgaria 14 SA (2013/N) - Ro-Net project RO 12/12/ Article 4(3) NGA 15 SA (2013/N) - NGA Sachsen- Anhalt GER 13/12/ Article 4(3) NGA 45