OECD Broadband Report Questioned - US Broadband Penetration Grows to 81.8% Among Active Internet Users - May 2007 Bandwidth Report
In our April bandwidth report we reported that the US dropped to 15th out of 30 countries surveyed by the OECD. The Information Technology and Innovation Foundation (ITIF) says that the broadband penetration data reported by the OECD is not an adequate measure of a nation's relative position in broadband technology. The ITIF suggests that adding speed and price to the equation would show a more complete picture of a nation's broadband rankings. Calculated in this way the US ranks 12th in global broadband penetration, two spots behind Canada at 10th overall (see Table 1). Turkey paid the highest price for high-speed broadband at $115.76 per month while Japan paid the least for high-speed broadband service at $0.27 per month.
The OECD data was also questioned by Market Clarity recently, a "telecommunications intelligence" firm based in Australia. Not surprisingly Australia came out ahead in the Market Clarity report. The OECD responded with questions of their own about Market Clarity's methodology. In our last annual US versus Canada broadband survey we found the US fast approaching Canada for broadband penetration among home users. As of March 2007, the US passed Canada in home broadband penetration (see Figure 1). Note that this data is from two different sources, who measure broadband penetration differently. Our US data is for active Internet users, which inflates the figures somewhat. However, the trends show that Canada appears to have reached a plateau in broadband penetration growth while US broadband uptake is still growing. Figure 1: US versus Canadian Broadband Penetration Growth (home users) US broadband penetration grew to 81.76% among active Internet users in April 2007. Narrowband users connecting at 56Kbps or less now make up 18.24% of active Internet users, down 0.95 percentage points from 19.19% March 2007 (see Figure 2). Figure 2: Web Connection Speed Trends - Home Users (US) In April 2007, broadband penetration in US homes grew 0.95 percentage points to 81.76% up from 80.81% in March. This increase of 0.95 points is above the average increase in broadband of 0.75 points per month over the last six months (see Figure 3). Figure 3. Broadband Adoption Growth Trend - Home Users (US) As of April 2007, 93.46% of US workers connected to the Internet with broadband, up 0.43 percentage points from the 93.03% share in March. At work 6.54% connect at 56Kbps or less (see Figure 4). Figure 4: Web Connection Speed Trends - Work Users (US)
"Market Clarity's omission of current, official data sources and subsequent estimation of broadband totals undermines the statistical validity of the report's findings. In addition, Market Clarity has chosen to adopt a different methodology for counting broadband (e.g. including 3G) than the OECD but then applies it inconsistently. Finally, Market Clarity is also inconsistent across countries in how it attempts to remove connections slower than 256 kbit/s." - Source: OECD
Table 1: ITIF Broadband Rankings - Top 20 OECD Countries
Rank
Nation
Penetration
Speed
Price
Overall Score
Subscribers per Household
Average Speed (mbps)
Price per Month for 1 mbps, Fastest Technology (USD PPP)
1
Korea
0.90
45.6
0.45
15.73
2
Japan
0.52
61.0
0.27
14.99
3
Iceland
0.83
6.0
4.99
12.14
4
Finland
0.57
21.7
2.77
12.11
5
Netherlands
0.73
8.8
4.31
11.87
6
Sweden
0.49
18.2
0.63
11.54
7
France
0.49
17.6
1.64
11.41
8
Denmark
0.70
4.6
4.92
11.37
9
Norway
0.64
7.4
4.04
11.29
10
Canada
0.62
7.6
6.50
11.11
11
Belgium
0.54
6.2
6.69
10.60
12
US
0.51
4.8
3.33
10.47
13
Switzerland
0.68
2.3
21.71
10.40
14
Australia
0.50
1.7
2.39
10.23
15
Austria
0.42
7.3
5.99
10.08
16
Portugal
0.42
8.1
10.99
9.92
17
UK
0.50
2.6
11.02
9.92
18
Germany
0.38
6.0
5.20
9.81
19
Italy
0.38
4.2
3.36
9.78
20
Luxembourg
0.51
3.1
18.48
9.71
US Passes Canada in Broadband Penetration
US versus Canadian Broadband Penetration Over Time
Source: Nielsen//NetRatings and Ipsos Reid
Home Connectivity in the US
Web Connection Speed Trends - Home Users (US)
Source: Nielsen//NetRatings
Broadband Growth Trends in the US
Broadband Adoption Growth Trend - Home Users (US)
Extrapolated from Nielsen//NetRatings data
Work Connectivity
Web Connection Speed Trends - Work Users (US)
Source: Nielsen//NetRatings
Further Reading
By website optimization on 21 May 2007 AM
Comments
Dear friends:
Could you send E-mail and inform me what are the keypoints made in OECD report (released April 2007) as it applies to New Zealand's Environmenttal Performance?
Appreciate and all the best!!
White
By: White at June 1, 2007 7:57 PM
I operate a small, local ISP in the US, and the price shown above for our country ($3.33 per Mpbs per month) is misleadingly low. The wholesale cost of Internet backbone bandwidth in our area of the US is $50 to $100 per Mbps per month -- sometimes more. And ISPs need to stay solvent, so obviously the bandwidth can't cost less than a tenth of that at retail. In the US, the quoted speed is usually often the speed of the modem -- not the device's actual maximum throughput, which is typically 10-15% less. And it is certainly NOT the amount of backbone bandwidth that the provider will actually let you use. (If you use the full capacity of an 8 Mbps cable modem 24 hours a day, the provider will cut you off in a hurry because it is losing large amounts of money on you.) Our ISP is different. We quote to users the actual amount of bandwidth that users can get all the time, and so the figure for us is much higher -- about $100 per Mbps per month (in line with our wholesale costs). But believe it or not, we sometimes suffer for this honesty. Prospective customers too often go for the false and misleading inflated claim than for ours, which can be empirically verified. Different countries have different constraints on deceptive advertising and marketing practices, so it is possible that this entire chart may be skewed. The compilers of this chart should investigate the true speeds of these services -- not just the advertised speeds -- and also determine whether that amount of bandwidth can really be used 24x7 without the provider "firing the customer" or throttling the connection.
By: Brett Glass at July 24, 2007 1:44 PM
I operate a small, local ISP in the US, and the price shown above for our country ($3.33 per Mpbs per month) is misleadingly low. The wholesale cost of Internet backbone bandwidth in our area of the US is $50 to $100 per Mbps per month -- sometimes more. And ISPs need to stay solvent, so obviously the bandwidth can't cost less than a tenth of that at retail. In the US, the quoted speed is usually the "raw" speed of the modem -- not the device's actual maximum throughput, which is typically 10-15% less. And it is certainly NOT the amount of backbone bandwidth that the provider will actually let you use. (If you use the full capacity of an 8 Mbps cable modem 24 hours a day -- it's really something like 5 or 6 Mbps after overhead -- the provider will cut you off in a hurry because it is losing large amounts of money on you.) Our ISP is different. We quote to users the actual amount of bandwidth that users can get all the time, and so the figure for us is much higher -- about $100 per Mbps per month (in line with our wholesale costs). But believe it or not, we sometimes suffer for this honesty. Prospective customers too often go for the false and misleading inflated claim than for ours, which can be empirically verified. Different countries have different constraints on deceptive advertising and marketing practices, so it is possible that this entire chart may be skewed. The compilers of this chart should investigate the true speeds of these services -- not just the advertised speeds -- and also determine whether that amount of bandwidth can really be used 24x7 without the provider "firing the customer" or throttling the connection.
By: Brett Glass at July 24, 2007 2:54 PM