“At that time the United States was a leader in broadband,” Mr. Medin recalled. Today, he lamented, “I don’t see anybody arguing that the U.S. is anything but mediocre.”

These days, Mr. Medin leads Google’s effort to deploy superspeedy 1 gigabit-per-second networks — 100 times faster than the 10 Mbps plans @Home introduced long ago — in several cities around the country, starting in Kansas City last fall.

Most of the nation’s innovation today relies on a broadband connection. Yet broadband seems to be the one area of the information economy that has not followed Moore’s law, named after the proposition by Intel’s co-founder Gordon Moore that the power of digital devices would roughly double every couple of years, radically expanding their capability and driving down their cost.

“Internet access is constraining what people can do,” Mr. Medin said. “This puts American companies at a disadvantage. It puts Google in a place where we can’t innovate as well as we could.”

President Obama has made much of this deficit. In 2010 his administration introduced a National Broadband Plan that promised a path of rapid deployment of high-speed networks, offering 100 million households affordable access to connections of 100 Mbps or more.

“We will not succeed by standing still, or even moving at our current pace,” Julius Genachowski, Mr. Obama’s first chairman of the Federal Communications Commission, told Congress at the time. Yet most Americans are still stuck in the Internet slow lane, far from the frontier of our possibilities. And the main roadblock remains much the same as it has been for years: a lack of competition.

Last week, President Obama nominated Tom Wheeler, a veteran lobbyist for the telecommunications industry, to succeed Mr. Genachowski. He has his job cut out for him: achieving fast universal broadband requires figuring out how to shake up the oligopolies that run the nation’s high-speed Internet.

There has been progress lately. The FCC points out that more fiber-optic cable has been laid in the United States than in Europe in the last two years. According to Akamai, the nation’s average broadband download speed is about 7.4 Mbps per second, about twice as fast as it was two years ago. This puts the nation in eighth place in the world, up from 22nd in 2009.

Still, speeds in the United States remain behind those in the world’s most connected countries, like South Korea, Japan and Switzerland. Equally importantly, American broadband, at an average price of $6.14 per Mbps, is more expensive than in most other developed nations.

This has little to do with the actual cost of moving bits. The price of transporting data wholesale across the Internet has fallen to about $1.57 per Mbps, down from $1,200 when Mr. Medin was helping start @Home. And high prices discourage Americans from opting for higher speeds. Though 10 Mbps broadband is available in 90 percent of homes around the country, and four out of five homes have access to 100 Mbps service, last year only 28 percent of homes that had access to broadband at a speed above 6 Mbps actually bought it.

What’s most worrying is that the handful of companies offering high-speed broadband to American consumers may have little incentive to expand their networks, increase their speeds and lower their prices.

According to the F.C.C.’s latest calculation, under one-third of American homes are in areas where at least two wireline companies offer broadband speeds of 10 Mbps or higher. Even including the spottier service offered by wireless providers, which tends to come with strict data caps limiting use, the share is less than half.

That means that in most American neighborhoods, consumers are stuck with a broadband monopoly. And monopolies don’t strive to offer the best, cheapest service. Rather, they use speed as a tool to discriminate by price — coaxing consumers who are willing to pay for high-speed broadband into more costly and profitable tiers.

Blair Levin, who headed the F.C.C.’s broadband initiative until three years ago and is now at the Aspen Institute, traces the roots of broadband’s limits to telephone companies’ decision, back in the 1990s, not to match cable’s costly investments in fiber, trusting that their DSL service would be an adequate competitor.

But DSL couldn’t follow cable past 3 Mbps. Verizon did eventually get on the ball — investing in its FiOS fiber network, which is expected to reach 17 million homes when it is completed. But that’s the exception.

Article source: http://www.nytimes.com/2013/05/08/business/google-project-may-spur-broadband-competition.html?partner=rss&emc=rss