AEI-Brookings Joint Center Policy Matters 06-18

What U.S. Broadband Problem? Scott Wallsten, Seth Sacher. July 2006.

"Tenth is ten spots too low," President Bush declared in 2004, referring to the share of Americans with high-speed Internet connections compared to other countries. Today, the U.S. doesn't even make the top ten.  These comparisons have led to calls for government subsidies, investment, and regulations on how broadband providers can use and charge for their infrastructure. Policymakers and others hope that telecom legislation working its way through Congress will improve U.S. international broadband competitiveness.

International comparisons, however, must be considered carefully.  A closer look suggests that despite the hype, it's not clear that there is a problem.

The precise rankings are fairly meaningless.  International numbers are generally self-reported by governments or government agencies that do not collect data in comparable ways and may be interested in appearing to have a high rank.  To make matters worse, the OECD and International Telecommunications Union, which report the numbers, generally do not publish their own methodologies, making it even less clear how the rankings are calculated.  Still, even if the U.S. is not behind exactly eleven other countries in broadband penetration, the numbers are clean enough to know that the U.S. is not in the top tier.

Or is it?

The share of the Americans that are Internet users, as opposed to the share of the population with broadband lines, compares much more favorably to the rest of the world, and is higher than other countries often held up as models to be emulated, such as Japan.

What explains this difference between broadband penetration and Internet users? The answer is that about half of U.S. Internet users still connect via dial-up. Flat-rate unlimited local telephone calls, innovations like "web accelerators," and competition that has pushed prices below $10 per month have allowed dialup to remain an attractive option for many people. While the number of people with dialup connections will surely continue to fall, a recent survey by the Pew Internet and American Life Project reports that nearly 60 percent of dialup users claim to have no interest in broadband (Horrigan 2006).  Some critics may point to the high use of dialup as evidence of a problem in the U.S., but that misses the point.  Aside from people who would prefer to subscribe to broadband but cannot, these numbers show that some people have less demand for broadband than do others.  For many, dialup--often supplemented by broadband access at work--is sufficient for their current needs.

What about available bandwidth, or "speed" as it is more colloquially called?  In some countries, notably Japan and Korea, providers advertise Internet speeds many times faster than those available here. Advertised speeds, however, may not accurately represent actual speeds. For example, providers in Hong Kong advertise some of the highest connection speeds in the world, but the island's telephone authority recently censured the Hong Kong Broadband Network for misleading claims regarding the speed of its service. Numerous factors affect speeds actually available to customers including line lengths, the number of users sharing the bandwidth connection, and the location of the content accessed.

Other factors also affect differences across countries.  It is less costly, for example, to build out the necessary infrastructure in more densely populated areas.  Korea is about 16 times more densely populated than is the U.S. and more than half of all Koreans live in large apartment buildings.  This concentration of people makes infrastructure upgrades less costly.  While some countries with low population densities, like the current leader in penetration, Iceland, have higher penetration rates, econometric evidence shows a strong statistical correlation between population density and broadband penetration (Wallsten 2006).

Meanwhile, a closer look at the U.S. reveals a dynamic broadband market. Critics contend that the U.S. broadband market is a “cozy duopoly” of DSL and cable providers that leaves consumers poorly served. But the facts suggest otherwise. The vast majority of broadband connections are indeed DSL or cable, but they compete vigorously with each other and with new technologies like wireless broadband, whose share is growing steadily.

Companies are investing bundles in broadband and consumers are signing up for it at record rates. According to the Federal Communications Commission (2006), the number of high-speed lines in the U.S. increased by nearly one-third to almost 43 million between June 2004 and June 2005. And the Pew survey found that the number of Americans with broadband access at home increased by 40 percent between March 2005 and March 2006.  Companies like AT&T and Verizon are rolling out fiber optic lines that connect directly to residences, allowing for much faster connections than have been available so far.
 
Prices for broadband access, moreover, are coming down while available speeds are going up. The Pew survey found that the average price of residential DSL service decreased from $38 per month in February 2004 to $32 per month by December 2005. Some evidence suggests prices may have fallen further still, with Verizon, AT&T, Comcast, and others offering service plans at less than $20 per month. And as prices fall, providers are upgrading connections and investing in new infrastructure to provide higher speeds to consumers.

Nonetheless, investments in information infrastructure, including broadband, have brought huge economic benefits--possibly accounting for a third of our productivity growth over the past decade.  Policy makers should ensure that legislation, regulation, and other initiatives do not have adverse consequences and undermine investment.  For example, current policies keep valuable spectrum away from its most valuable uses.  Putting more spectrum into the market, allowing its use to be flexible, and making it tradable could yield substantial economic benefits, including additional broadband competition.

The real problem with the international rankings is not that the U.S. appears to be behind, but that they cause politicians and policymakers to focus on silly statistics and favor equally silly policies.

Seth Sacher is a partner at the economics consulting firm of Bates White, LLC, and Scott Wallsten is a senior fellow at the AEI-Brookings Joint Center for Regulatory Studies and a resident scholar at the American Enterprise Institute.

A version of this article appeared in CNET on July 3, 2006.