Thomas W. Hazlett: Neutering the net
Richard A. Epstein: What we need is regulatory bed rest
The legendary Vint Cerf, co-creator of the Internet Protocol (IP) standard in the 1970s, is pleading for “network neutrality.” Cerf, now Google’s chief internet evangelist, argues for government regulation to ensure that broadband subscribers can use any network application or device, without extra fees.
The fear is that cable modem and digital subscriber line (DSL) networks will forge their own content deals, and then “click block” rival websites. In 2005, a small phone company prevented its DSL customers from using independent voice-over-internet (VoIP) services, reversing course when confronted with possible regulatory sanctions. Broadband network executives, however, say that they may start charging content suppliers – say, sticking Google with a 10 cents per search fee – to reach their customers.
Hence, the horror. “The internet was designed with no gatekeepers over new content or services,” Cerf writes. “By placing intelligence at the edges rather than control in the middle of the network, the internet has created a platform for innovation.”
Net neutrality advocates embrace flat rate, all-you-can-eat internet pricing, and are willing to mandate it. Broadband networks that charge content providers undermine the “open architecture” that Stanford law professor Lawrence Lessig claims as the internet’s “basic design.” Both Cerf and Lessig argue that government regulation is needed to save the internet as we know it.
Wrong prognosis, faulty diagnosis. The network economy has evolved through unregulated market transactions. Internet backbone carriers, for instance, exercise complete pricing freedom. Highly preferential treatment results, as the top backbones “peer” – charging themselves, the largest carriers, nothing to send traffic, while imposing hefty fees on smaller networks. This outcome yields incentives to construct larger and better facilities, improving bargaining power to obtain superior rates.
The internet is built, and grows, on the back of private property rights. Market structures that nurture innovative entrants have not been imposed by “design,” but have spontaneously emerged from “invisible hand” of self-interest.
Net neutrality is the current iteration of “open access,” the argument that cable TV operators be mandated to allow independent internet service providers (ISPs) to use their conduits, at regulated rates, to provide retail broadband service. Cerf is disappointed that regulators under both Presidents Clinton and Bush failed to enact open access. When rejecting a 1999 proposal, Federal Communications Commission Chair William Kennard said, “we don’t have a duopoly, we don’t have a monopoly, we have a no-opoly.”
That laissez faire cable modem policy, upheld in last June’s Brand X decision by the US Supreme Court, focused on encouraging new network creation. But the government’s policy on telephone company DSL service has differed. Extensive access rules were put in place in the 1990s, then relaxed following a February 2003 ruling. By September 2005, the FCC had substantially eliminated them, bringing DSL regulations into rough parity with cable modems.
The object of “open” rules is to provide consumers greater network value. The market test now reveals how consumers judge the results. Unregulated cable modems sprinted to a commanding lead among broadband subscribers, dominating regulated DSL networks nearly two-to-one, 1999 through year-end 2002. When DSL network access obligations were reduced in early 2003, however, the trend quickly switched. By 2004, new DSL subscribers pulled even with new cable modem customers. By 2005, DSL subscriber additions surged ahead. Overall, broadband penetration in the US increased from trend. The empirical evidence demonstrates that regulating open access failed to improve broadband networks.
While Google hawks “neutrality” for ISPs, its business is selling preferential access to advertisers, and its remarkable success owes much to its privately-managed IP network. This yields users lightning-fast searches by linking Google servers with conduits excluding competitors’ traffic. An upstart innovator would have to pay extra to duplicate this performance advantage, exactly the advantage that whetted Google’s creative juices.
The process whereby content owners and networks tailor terms and conditions is highly productive. Take Hughes DirecWay, a satellite broadband service that blocks video streaming. The policy is designed to rationally allocate capacity among customers, as attested to by the fact that DirecWay lacks market power. Rules forcing the network “open” its network would pre-empt economies and thwart competition.
Policies to invigorate such rivalry are the answer. There are crucial steps to take, such as unplugging local government franchise barriers and the asphyxiating constraints applied by the spectrum allocation regime. A recent statement on US broadband policy by 25 policy economists (including myself) states the pro-competitive case.
Cerf is aware that the alternative path, crafting regulations, is a tall order. He takes it as a challenge to find “language that would give us an objective measure of neutrality.” But no single policy mandate, articulate or clumsy, would capture the efficiencies that emerge from the trials and errors of the market. An irresistible irony is that cable TV systems are being hammered for flat, all-you-can-eat video pricing, with regulators threatening to impose a la carte menus, customers paying for each network separately. But flat rate pricing, politically popular in some instances and controversial in others, has thus far proven highly efficient, and has been widely adopted for both basic cable television and broadband access.
Allowing continued market-based evolution will not end the internet as we know it. Commencing to impose regulated solutions just might.
The writer is professor of law economics at George Mason University, where he is director of the Information Economy Project of the National Center for Technology and Law
Richard A. Epstein: What we need is regulatory bed rest
In the four or so years that I have written for FT.com on tech issues, I cannot recall having disagreed with Tom Hazlett even one occasion. Relax, for the shoe is not about to drop now either. The issue of network neutrality, or open access, is one that has long been an issue in network industries because of the simple observation that pure competitive solutions do not allow for the creation of a single network to which all persons have equal access. Some deviation from the usual rules of exclusive property rights and voluntary association is needed to allow any one person on the network to communicate with any other person.
So far it looks as though the case against pure competitive communications systems counts as a point in favour of the network neutrality proposal championed by such notables as Google’s Vint Cerf and Stanford’s Larry Lessig. But in reality that hasty conclusion begs the prior question: what is the smallest fix that allow for network integration. Answer: a rule that allows all persons to interconnect to the basic internet. There is nothing, however, about the proposal of various broadband operators to deny, limit, or charge access to their proprietary systems that serves to undo the interconnectivity of the Internet as a whole. The simple point here is that so long as consumers have a choice among broadband providers, all of who run over the internet, then the law should step aside. The situation here is no different from one in which multiple railroads are allowed to run separate trains on a common track. Once we know that there is competition up and down the line, then no single railroad should be under a duty to take all comers under a common carrier model that limits them to reasonable and nondiscriminatory prices.
Why? Hazlett mentioned that consumers are quite comfortable with switching between various broadband providers who operate either on cable modem or DSL. So long as these parties are in competition with each other, the dangers of any system of mandated access is clear. No regulator cannot mandate access without controlling price. No regulator knows how to set prices that will match in rapidity or subtlety the endless shift in prices and terms of competitive broadband providers. The great risk of regulation is that the effort to create parity in the ex post world will create massive distortions in the ex ante world that will reduce investment. We have already had bitter experience under the 1996 Telecommunications Act of how mispriced unbundled network elements could throw the entire industry into chaos. What we need now is not a new generation of price and access regulations. What we need is regulatory bed rest.
This writer is the James Parker Hall Distinguished Service Professor of Law at the University of Chicago, and the Peter and Kirsten Bedford Senior Fellow at the Hoover Institution.