What have we learned from Google Fiber?

Commentary: Blair Levin, chief author of the National Broadband Plan, offers insight on what Google Fiber has taught policymakers about how to get very high-speed broadband service to all Americans.

Blair Levin
Blair Levin is the former chief of staff for FCC Chairman Reed Hundt serving from 1993 to 1997. He has worked as a policy analyst for investment firm Legg Mason and then was managing director at Stifel Nicolaus. In 2008, he co-chaired the technology, innovation, and government reform transition team for President-elect Barack Obama and subsequently served as the executive director for the effort that produced the National Broadband Plan from 2009 to 2010. He currently heads up the Gig.U project, a consortium of 36 research university communities seeking to accelerate the deployment of next generation networks in the United States.
Blair Levin
5 min read

In February, 2010, Google announced its Fiber project, designed to offer a gigabit service with broadband speeds 100 times faster than what most Americans were receiving at that time. For several years, incumbent providers did not react with improved services, saying instead that such services were too expensive and consumers didn't care about getting a gig.

In the last year, however, as Google started to offer gigabit services at prices comparable to much slower offerings by cable and phone company providers and expanded Fiber's service territory, the incumbents and others have started to announce their own gigabit offerings at similar price points, setting the stage for a potential "game of gigs," in which tens of millions of Americans have the potential to receive faster, better and cheaper broadband.


There is no question that Google Fiber is a seminal development in the broadband market. The question is what is the lesson for policy?

Writer Karl Bode of Techdirt.com provides one answer: Google Fiber proves the National Broadband Plan was nothing more than a "political show pony." In 2009, Congress asked the Federal Communications Commission to write a National Broadband Plan to make recommendations to Congress, the FCC and other federal agencies with an eye toward improving broadband service, making it ubiquitous, improving adoption and using broadband to improve government services. The FCC delivered the National Broadband Plan in 2010.

Much of Mr. Bode's criticism of the National Broadband Plan, compares it unfavorably to recent government actions and private sector deployments. As neither was authorized in the plan's Congressional mandate, such criticism belongs in the same bucket one would put criticism of a baseball game due to the absence of touchdowns.

But Mr. Bode suggests a serious criticism of the plan at the core of one of the lessons Google Fiber teaches. He chastises the National Broadband Plan for not doing enough for competition. The core policy related to network competition, however, that he proposed was forcing incumbents to offer an open access network, in which the incumbents would offer all, or parts, or their networks to rival providers at wholesale rates. Under Mr. Bode's proposal, which others also supported, the government would define both what parts of the network would be "unbundled" and set the rates.

The National Broadband Plan team evaluated, and then rejected this option, believing it would not lead to investments to deploy the next generation networks necessary to deliver the affordable, abundant bandwidth America needed.

As Google Fiber, which emerged from discussions with the team developing the National Broadband Plan, started evaluating the economics of building a fiber network, it concluded that building a network that it shared with other providers would not make economic sense, even on an unregulated basis. Google, therefore, decided to proceed with a proprietary model. If the government had required open access that required Google to share its infrastructure with competitors, Google Fiber would have been stillborn.

This is not to suggest that open access cannot work in some places; indeed I have worked with some communities that have chosen that path. But Google Fiber would not have expanded under those rules.

The lesson here is competition does not arise from the desire for it; it comes from rules that attract investment into competitive networks.

But it is not enough to simply steer clear of the dangers of unbundling or forcing companies to share their infrastructure with competitors. If government is going to rely on competitive facilities to deliver abundance, it has an obligation to adopt policies that affirmatively make such networks economically viable. That's why last week's House hearing on a broadband deployment agenda was so welcome. The Energy and Commerce Committee heard from a number of witnesses, including an executive with Google Fiber, about the different ways specific government policies add unnecessary costs to deploying or upgrading networks.

Since Google has begun to deploy its network, it has pointed to a number of policies that government could adopt to facilitate such competition, such as improved pole access, competitive set-top box rules and federal agencies taking actions to make sure their rules did not raise the cost of deployment. The National Broadband Plan proposed actions in each of these areas, some of which have been adopted. Recently, Google has added another issue -- access to video programming -- which neither those who lobbied the author's of the National Broadband Plan, nor the plan itself, addressed.

Government's commitment to a deployment agenda cannot be time-limited. It must be open to new understandings about the barriers to competition.

Another critical lesson, highlighted at last week's hearing by the executive director of Next Century Cities, is the important role of local levers and leadership. While a federal deployment agenda is vital, what Google Fiber understood, and others such as my own organization, Gig.U, also exploited, was that the policies that most move the needle on lowering construction and operating costs are decided in city hall.

The most profound lesson, however, that I draw from Google Fiber, stems from how the effort has stayed true to its vision of bandwidth abundance but has constantly adjusted its strategy and tactics as markets adjust, competitors respond, and policy issues arise. Google Fiber is both an enterprise and an experiment. Its success at both depends on understanding new data as it emerges.

Policy also has an experimental component. Mr. Bode and I will disagree on the intent and impact of the National Broadband Plan, but I assume we both agree that due to Google Fiber, and other efforts, such as those of the 25 Gig.U communities taking affirmative steps to accelerate next generation deployments, the competitive outlook for affordable abundant bandwidth is better today than it was in 2009.

Markets, however, can shift, experiments can fail, and it is unclear how far the kind of gigabit competition we are seeing in some places will extend. As the chairman of last week's hearing Rep. Greg Walden (R-Oregon) wisely noted, "This is not the last hearing. We expect to continue this work going forward."

We should all hope Congress and others at all levels of government agree.

Blair Levin was executive director of the National Broadband Plan and, as executive director of Gig.U, recently co-authored "The Next Generation Network Connectivity Handbook." Gig. U is aimed at helping communities across the U.S. get wired with super-fast broadband.