IHS Global Insight Perspective | |
Significance | The OECD data provides a useful snapshot of global broadband uptake, with dimensions of analysis including technology, speed and price. |
Implications | The release always stirs up some controversy over both methodology and results, and is often used as ammunition to drive debate and policy. |
Outlook | Fibre deployment is ongoing and reaching new countries, with speeds increasing while price per bit is falling, but wide disparities exist. |
The Organisation for Economic Co-operation and Development (OECD) has released its semi-annual data on broadband subscription in member countries. Key data show that there are a total of 267 million broadband subscribers in the 30 countries surveyed, equal to 22.6% of the population and up from 22.1% a year earlier. The total number of subscribers (based on government reports or estimates) is up by 13% year-on-year (y/y), with no sign of a slowdown in the latter half of the year as the financial crisis and recession rolled out around the world, growing by 6.23% in the second half of the year and 6.16% in the first half of the year. The Slovak Republic was noted as showing the strongest growth, with Greece, New Zealand, Norway, Germany, France and the United States all adding more than three subscribers per 100 inhabitants over the year.
Denmark again led with 37.2 connections per 100 of the population (the more useful penetration per household is not provided), with fellow Nordic countries dominating the top rankings. Eastern Europe, the Aegeans (Greece and Turkey) and Mexico take up the bottom of the ranking table and there is a fairly large correlation of 0.65 between GDP and penetration rates. There was a very low correlation (0.22) with population density and total landmass (0.02), but given that this does not account for city clustering, it does not really show anything—although the advantages of Korea and Japan in deploying FTTx are visible in their leading population densities and an older analysis by the OECD of penetration against percentage of landmass used by 50% of the population showed a better correlation.


Technology Use
Broadband is counted to include DSL (13.6 subscribers per 100 of population), cable (6.4%), fibre (2.2%) and other (0.4%) including power-line, leased line and fixed-wireless technologies (satellite, LMDS, MMDS, fixed WiMAX and other technologies) with speeds over 256 Kbps, but not 3G mobile services unless sold as a fixed-wireless solution. The use of different technologies varied significantly between countries, with the United States, Canada and Korea the only countries with more cable users than DSL—in Korea this is mainly due to the high penetration of fibre. Denmark, Belgium, the Netherlands and Switzerland all had cable penetration of around 10% or higher.

Fibre-based services now account for more than 10% of all subscriptions across the OECD, up from 9% six months ago. Japan and Korea continue to lead by a wide margin in the amount of fibre-based (FTTx) connections, with Korea counting 13.8 fibre users per 100 inhabitants to Japan's 11.3. Deployment is now well under way in other countries, with the likes of Sweden, Denmark and Norway counting a fair number of fibre-based connections. Given the Slovak Republic's low penetration, it has a high rate of fibre connections, but this is likely to be the result of effectively no mandated pricing for local loop unbundling, thus forcing new competitors to build out new plants (see Slovakia: 26 March 2009: GTS Slovakia Signs LLU Agreement, Will Focus on Corporate Clients).

Speed
The increase in fibre-based connectivity is enhancing the service speeds available, with 12 countries able to provide speeds of 50 Mbps or higher and DSL offering an average of 9.6 Mbps, cable 14.9 Mbps and fibre 65.3 Mbps. Japan has the highest advertised speeds at an average of 93Mbps, with a top advertised speed of 160 Mbps. The average speed rankings show Korea in second place at 81 Mbps, France third at 51 Mbps and Finland fourth on 19 Mbps. The United Kingdom falls 15th with 10.7 Mbps, while the United States averages 9.6 Mbps. At the bottom, average advertised speeds in Turkey are 3.3 Mbps and in Mexico 1.5 Mbps. Advertised cable speeds have grown by 30% per year on average since 2005, while DSL speeds increased by 22%.
Prices


The data available for September/October 2008 show price per Mbps averaging US$12. Entry-level service plans were all under US$34 per month, with DSL subscribers paying an average of US$40 per month and cable subscribers US$45 per month. Japan was cheapest for DSL at US$26 per month, while the French average price of US$22 per month was the cheapest cable broadband. Prices for cable services fell by an average of 15% for cable and 14% for DSL each year since 2005. Bit caps, which have proved controversial in some countries such as the United States, are widely used, although not significantly for FTTx-based connections where, as a premium product, high-use expectations are likely to exist (see United States: 20 April 2009: Time Warner Drops Metered Pricing Plans).

Pricing and Bit Caps (US$) | ||||
% Offers with Bit Caps | Average Bit/Data Cap (MB) among Surveyed Firms with Caps | Number of Offers to Calculate Average | Average Price per Additional MB After Reaching Cap | |
DSL | 0.40 | 29,936 | 59 | 0.04 |
Cable | 0.31 | 27,730 | 29 | 0.04 |
Wireless | 0.63 | 5,012 | 15 | 2.82 |
FTTx | 0.08 | 6,800 | ||
Outlook and Implications
While this data provides a useful snapshot of the industry in the major countries around the world, the release of OECD data tends to stir up controversy because countries are ranked against each other, thus garnering plaudits or ammunition for interested parties and often driving policy. Accusations fly that non-comparable data are used, such as the use of 256 Kbps as a benchmark for broadband that is not collected by many nations; no new data actually published by the claimed sources; or that available data are inconsistent with local regulators. The data based on advertised speeds and prices are also notoriously prone to errors from effects such as local advertising regulations, offer construction and hidden additional costs. Other points are made that the data are plainly incorrectly recorded, while little reference is given to source material to facilitate validation.
Even so, the data show that a number of countries should be nearing the penetration ceiling, with one connection to every three people similar to the average household size, and the deployment of premium and fibre-based services has the potential for higher revenues, mainly from new services as prices continue to fall for basic connectivity.
