Broadband access in the United States is even worse than you think.

The intersection of government and technology.
April 28 2010 7:03 AM

Denial of Service

Don't believe the telecoms. Broadband access in the United States is even worse than you think.

Given the dismal state of broadband connections in America, it was illuminating recently to hear a major telecom executive paint a rosy picture of where the country stands. When Wall Street Journal Deputy Managing Editor Alan Murray asked how the United States ranks in broadband, Verizon CEO Ivan Seidenberg didn't hesitate: "One. Not even close."

To support his statement, Seidenberg claimed that "in the U.S., there is greater household penetration of access to the Internet than any country in Europe." Compare that with what Federal Communications Commission Chairman Julius Genachowski recently told a Senate committee: "Our record shows roughly 65 percent adoption in the U.S. compared to significantly higher adoption percentages—up to 90 percent or more—for some countries in Asia and Western Europe."

How can two people arrive at such radically different assessments? Seidenberg is deliberately conflating "access" and "adoption"—the difference between who has the option of buying broadband service and who has actually done so. Using Seidenberg's logic, Americans also have universal access to health care, college, and employment. Dispelling this sort of misinformation, however, isn't always easy. One of the big problems in this debate is that the data about broadband are as spotty and unreliable as the connections themselves. And, taking a page from the playbook of big oil and tobacco, the telecom companies are spending millions to further confuse the issue, spending about $100 million in 2009 alone in lobbying fees. With all the bogus information out there, hucksters like Seidenberg can lie through their teeth and get away with it.

Here's what we do know: If you simply look at broadband "penetration"—a measure of broadband subscribers relative to the population—the U.S. is ranked 15th by the Organization for Economic Co-operation and Development, with 27 broadband subscribers per 100 people (check out Table 1d). And another key organization, the International Telecommunications Union, ranks the United States 16th. Just one decade ago, the United States was at the top of the list.

But penetration doesn't tell the whole story. To get an up-to-date picture of where we actually stand, the New America Foundation—where we both work—recently took a very close look at both speeds and prices in more than a dozen leading broadband countries. As it turns out, U.S. residents paid more for bandwidth than nearly every other country surveyed. Typically, the lowest price for broadband in the United States, not counting promotions and bundled deals, costs an average of $35 a month for a measly 1 megabit per second connection. Twice this speed is available in Denmark and Canada for lower prices; more strikingly, Hong Kong, Taiwan, and Sweden have broadband available for under $20 a month. Additionally, the fastest speeds in the United States are comparatively slow. The common top speed available for residential services in the Unites States is 50 Mbps (and costs $145 a month), while several nations have speeds available that are up to four times  faster, for less than $60 a month.

According to the FCC's National Broadband Plan, the no. 1 reason that those without broadband cite for not having broadband is cost. Given that broadband is more expensive here than abroad, it's no surprise the United States lags behind a growing list of other countries. Subscribers in the United States pay more per megabit of bandwidth than countries across both our oceans. To remedy this, the FCC has a plan that's the equivalent of the United States entering the Grand Prix with the goal of finishing last. The National Broadband Plan wants all Americans to have access to 4 Mbps download and 1 Mbps upload speeds by 2020. In that same time frame, the plan also proposes a neatly framed 100 Mbps download, 50 Mbps upload connection for 100 million homes.

By way of comparison, Taiwan already has near-universal access to 10 Mbps and South Korea achieved 1 Mbps universal access in 2008. By the end of 2010, Germany and Ireland both plan to reach universal 1 Mbps while Sweden, Denmark, and the U.K. are working to 2 Mbps to everyone by the end of the year. In essence, many nations expect to achieve goals by the end of 2010 that will rival what we hope to achieve in 2020.  Furthermore, the 100 million households that will get 100 Mbps speeds represent only 75 percent of the population. By comparison, South Korea plans to have 50 Mbps available for 95 percent of the population in 2013, Sweden's goal is 100 Mbps for 90 percent of the population by 2020, and Finland is striving for universal 100 Mbps availability by 2015.

Many commentators have pointed out that competition is sorely lacking among broadband providers. As the FCC noted in its national plan, 96 percent of all households are served by two or fewer providers. But even when some choice is present, precious little information is available for customers to make informed decisions about their broadband service offerings. Speeds are advertised as "up to"—even though systematic testing documents that customers usually receive only half this advertised speed. And advertised prices almost always exclude hidden fees and additional costs, often require bundling with additional services that customers neither want nor need, are usually only good for short promotional periods, and come with a mountain of caveats and other fine print allowing providers to sever connections, manipulate customers' Internet traffic, and even spy on your online activities.

The FCC's plan falls far short of providing the meaningful information customers need. In two detailed examples, the plan suggests that ISPs provide the average speed—a far more useful comparison than max speed—but not a guaranteed minimum speed. We'd never buy a package of "up to a dozen eggs" at the supermarket, so why are broadband providers allowed to systematically promise more than they deliver?

In the same way that it's useful to know the processor speed, screen size, amount of RAM, and hard drive space of a computer before you buy it, broadband measures such as latency, jitter, and uptime are key pieces of information needed to know whether you can run a growing number of online applications, even if they sound complicated at first. For example, if you use Skype, stream NetFlix movies, play World of Warcraft, or use any other of the countless real-time applications, metrics like latency and jitter affect your quality of service.

What can we do to fix these problems?  First, the FCC can mandate that all ISPs provide a "broadband nutrition label" that clearly lays out the details of what's actually being offered so that customers can make informed decisions about which service to buy.  Second, the FCC should systematically collect information on the speeds, pricing, and adoption of broadband across the country—especially in un- and underserved areas.  Third, the commission needs to address the duopoly in the United States and formulate a competition policy that will bring back a meaningfully competitive market. If we don't put these solutions in place,  the United States could end up on the wrong side of the international digital divide.

Like Slate on Facebook. Follow us on Twitter.

James Losey is a policy analyst with the New America Foundation's Open Technology Initiative. He writes about user freedoms and the digital divide have been published in Advances in Computing, Ars Technica, CommLaw Conspectus, IEEE Internet Computing, IEEE Spectrum, and Slate.