July 6, 2007
TechKnowledge no. 106
by Timothy B. Lee
Timothy B. Lee is an adjunct scholar at the Cato Institute in Washington, DC. To subscribe, or see a list of all previous TechKnowledge articles, visit www.cato.org/tech/tk-index.html. A version of this article originally appeared in Ars Technica.
Published on July 6, 2007
Share with your friends:
Last month, the Washington Post reported that Google is swiftly expanding its lobbying presence inside the Beltway. As part of that effort, the company recently unveiled a new public policy blog which pledges to "do public policy advocacy in a Googley way."
The launch of Google's blog offers a good opportunity to review the debate over network neutrality regulation. In a recent post, Google noted the challenge of "defining what exactly the term means." Richard Whitt, Google's Washington telecom and media counsel, offered details about what would and wouldn't be permitted under the company's ideal regulatory scheme.
The requirements Whitt enumerates closely mirror the terms of the Snowe-Dorgan bill, the most prominent network neutrality legislation introduced last year. Google wants to bar ISPs from imposing "surcharges on content providers that are not their retail customers," discriminating among packets based on their source, or building a new "fast lane" limited to content providers that paid extra to use it.
Google worries that if cable and telephone incumbents are allowed to build such a tiered network, Google could be forced to pay tribute to companies like AT&T and Verizon, or be blocked from access to certain customers altogether. Whitt notes that 99.6 percent of broadband customers get their service from either their phone company or their cable company. He contends that two firms per market is insufficient to ensure a robust, competitive marketplace.
Many computer geeks are concerned that the Baby Bells -- companies with a historical hostility to open networks -- will not respect the Internet's open architecture. They rightly observe that the Internet's decentralized, non-proprietary design is essential to its entrepreneurial culture. And given Google's dependence on that open architecture, it's not too surprising that Google would have a keen interest in keeping it that way.
But Google's post also highlights the difficulties that are likely to arise if Congress attempts to codify the technical principle of network neutrality into a detailed regulatory scheme. If regulation is too complicated, it can tie ISPs up in red tape and make it more difficult for them to manage their networks effectively. On the other hand, overly vague regulations can open loopholes that undermine the entire purpose of the regulation. Finally, we can be sure that once a network neutrality rule is on the books, regulated firms will look for ways to turn the rules to their advantage.
One potential loophole can be found in Google's list of what behaviors are permitted under its proposed rules (and in the Snowe-Dorgan bill on which it is based). Google would allow ISPs to provide "managed IP services and proprietary content (like IPTV)" and to prioritize "all applications of a certain general type, such as streaming video." This suggests one strategy that an ISP could employ to evade the intent of the network neutrality rules: It could give video services a lower priority on its broadband service than other types of content (which would apparently be legal as long as all video services were treated the same), and then it could syndicate the video content of partner companies via its IPTV service. It's not clear how the law would distinguish between a prohibited "Internet fast lane" and permitted "managed IP services and proprietary content." This would appear to have precisely the effect Google fears -- allowing an incumbent to sell preferential delivery of video content -- without running afoul of the letter of the law.
There's also a real danger of what economists call "regulatory capture": when regulations designed to control industry incumbents are manipulated to serve those incumbents' interests. One of the first examples of this problem in the United States was the regulation of railroads in the 19th Century. The railroads were the high-tech industry of their day, and there was a lot of concern in the 1870s and 1880s that the railroads had become too monopolistic. Congress responded by creating the Interstate Commerce Commission in 1887, giving it the power to regulate the railroads.
As I wrote in the New York Times last year, the story of the ICC does not have a happy ending. President Grover Cleveland appointed Thomas M. Cooley, a railroad ally, as its first chairman. The Commission quickly fell under the control of the railroads, gradually transforming the American transportation industry into a cartel. By 1935, when it was given oversight of the trucking industry, the commission was restricting competition and enabling price increases throughout virtually the entire surface transportation industry. In 1970, a report released by a Ralph Nader group described the Commission as "primarily a forum at which transportation interests divide up the national transportation market."
Not all examples of regulatory capture are ancient history. The cable industry has used cable franchising rules -- which were originally intended to protect consumers -- to exclude the Baby Bells from the video market. This protects cable's ability to charge monopolistic rates for its services. And the Bells themselves are using a slew of telecom regulations -- Universal Service fees, E-911 requirements, CALEA, and others -- to harass up-and-coming VoIP providers.
Similar things could happen if network neutrality regulations are adopted. Once Congress passes a bill like Snowe-Dorgan, it would fall to the FCC to enact specific regulations that implement the bill's requirements. Google advocates imposing several new obligations on all Internet service providers: "requiring carriers to submit semiannual reports with broadband deployment data," interconnection and open access requirements, a "ban on most forms of packet discrimination," and "an effective enforcement regime." These requirements could lead FCC bureaucrats to second-guess the decisions ISPs make about their router configurations. At a minimum, it would require every ISP in the country to hire telecom lawyers to deal with the FCC's oversight process.
That's worrisome because new technologies are likely to introduce additional competition to the broadband market in the next decade or two. Some of them may not have the deep pockets of a Verizon or a Google. Overly burdensome regulations could become a barrier to entry for these smaller firms. Incumbents are likely to file a slew of complaints against these competitors. And even if the incumbents ultimately lose, such harassment could be sufficient to drain the upstarts' resources, slow them down, and drive them into bankruptcy.
A cynic might suggest that Google isn't concerned about regulatory capture because Google has sufficient resources to defend itself -- perhaps even because it wants to do some of the capturing itself. Regulatory capture tends to harm small firms the most, and Google is no longer small.
More likely, though, Google's leadership is simply naive about the regulatory process. Google's executives are amateurs at the regulatory game. Like Moby, Craig "List" Newmark, and generations of idealistic activists, they have an unreasonably optimistic view of the regulatory process. But if Google and their allies succeed in enacting new regulations, the battlefield will shift to the FCC, which will translate the general principles articulated by Congress into specific regulations. This arena is the Bells' home court, and Google could very easily find themselves outmatched.
Rather than seeking regulation, Google should focus its efforts on the area where it does have an advantage over telecom incumbents: developing great technology.
Share with your friends:
� 2010 The Cato Institute
Please send comments to webmaster