Competitive Enterprise Institute | 1899 L ST NW Floor 12, Washington, DC 20036 | Phone: 202-331-1010 | Fax: 202-331-0640
In response to my analysis of Google Fiber, Timothy B. Lee at Ars Technica says the Google Fiber deployment is “hardly an example of the free market in action.” Lee notes that Google received subsidized access to local government resources in the Kansas City area to support its fiber deployment. As I disclosed in my initial analysis, however, I agree that Google’s deployment did not occur in an unfettered free market. Our analyses don’t differ on the publicly available facts surrounding the Google Fiber deployment—they differ on the inferences and policy lessons that can be reasonably drawn from those facts.
Lee infers that local government subsidies were a significant factor in Google’s decision to deploy a fiber network in Kansas City. Though that is possible, there is no publicly available evidence indicating that subsidies were economically necessary for Google’s deployment. In its testimony before Congress, Google said it passed over other cities due to stringent regulatory requirements governing infrastructure deployment, not because they refused to offer subsidies. Most cities require network operators to serve all or substantially all neighborhoods irrespective of market demand, which is something Google was unwilling to do. Google’s statements and behavior indicate that Kansas City’s willingness to allow demand to drive Google’s deployment and avoid unnecessary regulatory costs and delay were the critical factors in Google’s decision to enter the market.
Google’s statements also indicate that it would have been willing to pay reasonable fees for access to public rights of way and to obtain power and office space at its own cost if Kansas City had not offered these benefits for free. Google has objected only to government regulation that “results in unreasonable fees, anti-investment terms and conditions, and long and unpredictable build-out timeframes.” Most franchise authorities impose fees on infrastructure providers that are designed to raise substantial revenue rather than recover the costs to the community of access to public rights of way. It infers too much to conclude that in-kind subsidies are required to build competitive fiber networks merely because Google objected to unreasonable fees and regulations and accepted the support offered by Kansas City.
Lee says “we should acknowledge the possibility that it simply doesn’t make economic sense for private firms to build new fiber networks without taxpayer subsidies.” That possibility has been considered a fact since the early 20th century. The United States has always subsidized networks in rural areas (e.g., areas that lack economies of density). Rural networks were traditionally subsidized indirectly by state-mandated monopoly telephone and cable networks that were subject to universal service requirements—i.e., in exchange for receiving a state monopoly, they were required by regulation to serve uneconomic neighborhoods and rural areas.
In my view, Kansas City’s willingness to abandon this traditional quid pro quo is what makes Google Fiber a win for the free market. If Google had been willing to build its network without the in-kind subsidies it received, as I believe the evidence suggests, its deployment belies the false choice presented by “natural monopoly” advocates who believe our only options are (1) mandating a monopoly and requiring it to provide universal service (a public utility model), or (2) allowing “wasteful” competition and abandoning the goal of universal service.
As Jerry Brito at the Mercatus Center noted in an article contrasting Google Fiber with Verizon FIOS, these advocates assume that all broadband network operators will profit solely from the subscription fees they collect from subscribers. Plain-old-telephone service uses a subscription model because telephone networks were historically limited to the transmission of temporary content generated solely by consumers. Networks that could offer complementary services, however, have not always relied on a subscription model to generate revenue. Over-the-air broadcast networks offer television service to consumers for free and generate profits through the sale of advertising to other businesses. Advertising is Google’s primary revenue source as well, which may explain why it is willing to offer Google Fiber for free to consumers who pay for installation. Google can also generate non-subscription revenue by bundling its hardware (Nexus 7), software (Android), and content (YouTube) with its premium network subscription, which serves to increase its share of these complementary markets. The profits generated by these complementary markets could sustain competition among networks without a government subsidy.
This is the possibility we should acknowledge—that there is an alternative to our current regulatory approach to networked industries, which assumes there are “natural boundaries” between networks, devices, and content. The free market aspects of the Google Fiber deployment suggest that these boundaries are not “natural,” they are the result of an outdated regulatory approach that has mandated market boundaries for nearly a century. If these regulatory barriers were removed, many areas would likely be able to support competition among multiple high-speed platforms that offer differentiated packages of services, which would likely increase consumer choice and lower prices.
Eliminating regulatory barriers would not provide assurance that competitive networks would be built everywhere without any government subsidies. There is significant variation in population density and geography in the United States, and it would likely remain uneconomic to serve remote areas. The relevant question, however, is how to best address this issue. The traditional answer, providing a state monopoly in exchange for privately subsidized service in uneconomic areas, throws the baby (market-based competition) out with the bath water (low density). Google Fiber indicates that we should be encouraging private firms to build competitive networks without government subsidy whenever possible and rely on explicit subsidies only when necessary.
Google Fiber also indicates that the marketplace is better than the government at determining the circumstances in which competitive entry is economically possible. If Google believed existing network providers had a “natural monopoly” in Kansas City, it would have been irrational for it to enter the market. Though that is a possibility, I’m willing to give Google the benefit of the doubt.
Finally, in circumstances in which government subsidies are required, economic efficiency and fundamental fairness require that those subsidies be available to everyone, i.e., that all interested parties have an opportunity to receive the same treatment. On that issue at least, Lee and I agree.