Data Ports in a Storm

The debate over whether Internet providers can play favorites harks back to arguments over British wharves and Midwestern grain elevators

Related materials

Colloquy: Read the transcript of a live online discussion with Christopher S. Yoo, a law professor at Vanderbilt University, about whether broadband providers should be forbidden to offer different rates or levels of service to different customers.

Printer friendly	E-mail article	Subscribe	Order reprints
	Discuss any Chronicle article in our forums
Latest Headlines If Kent State Beats Goals, Professors Will Profit The university will offer cash bonuses to professors when institutional goals for fund raising, research dollars, and student retention are met. Hurricane Gustav's Toll Is Mostly Psychological McCain Presses Fight Against Earmarks in His Speech Students and Professors Help Convention Speakers Connect Video: 'I Support John McCain' Congress Demands More Accountability of Colleges Broad Institute's Founders Will Raise Their Gifts by $400-Million Update on Billion-Dollar Campaigns at 28 Universities Institution Does Its Research Roadside Commentary Race on the Occoquan: a President's Second Freshman Year

By DAVID GLENN

During the decade after the Civil War, farmers in the American Midwest produced millions of tons of grain, a large proportion of which eventually passed through the ports and rail yards of Chicago. The city was served by just 14 grain elevators — and it appeared unlikely that any more would be built because open land along the rivers and tracks was scarce.

So the Illinois legislature — citing a danger of profiteering from the bottlenecks — passed a law that set maximum rates for grain storage. Two operators challenged it, arguing that they were being unjustly deprived of their property rights. The case made its way to the U.S. Supreme Court, which upheld the price restrictions in an 1877 decision known as Munn v. Illinois.

Writing for the majority, Chief Justice Morrison R. Waite said the Illinois law fell under a longstanding principle in English common law, under which private operators of ports, wharves, and roads are "common carriers," meaning that they are obliged to serve all willing customers and must not charge arbitrary or excessive fees.

Chief Justice Waite acknowledged that grain elevators were a newfangled technology. But that made no difference, he wrote. The fact that the legislature had bothered to enact the law meant that "during the twenty years in which this peculiar business had been assuming its present 'immense proportions,' something had occurred which led the whole body of the people to suppose that remedies such as are usually employed to prevent abuses by virtual monopolies" were necessary.

Now economists and legal scholars are arguing about another peculiar business that has assumed immense proportions: the Internet. The proponents of "network neutrality" — a principle being furiously debated in Congress — argue that broadband services are dominated by a tiny number of cable and telephone companies. Because the industry is a near-monopoly, they say, the companies should not be allowed to play favorites. That is, they should not be allowed to charge different rates, or offer different transmission qualities, to different content providers and application developers. Like a port or road or grain elevator, neutrality's supporters say, the Internet should be open to all willing customers, and its prices should not be arbitrary.

Only if the Internet remains "neutral" in that sense, proponents say, will inventors have the incentive to create more blockbuster applications on the scale of Google or YouTube, a video-sharing site. If such inventors fear that their sites might not reach the full universe of Internet users, or that costly and elaborate deals with broadband providers would be necessary, they might not bother to develop the sites in the first place — or so the argument goes. Proponents also say network neutrality is crucial to maintain the Internet as an inexpensive forum for political speech and public conversation.

"The underlying debate here has to do with whether there truly is something special about certain types of infrastructure," says Timothy Wu, a professor of law at Columbia University and a leading advocate of net neutrality. "Whether there's something about plumbing, something about roads, something about bridges, something about broadband, that makes them likely to be sources of competitive abuse — but also likely to be of great public benefit if they're done right."

Skeptics of that view say there is no evidence that broadband providers would or could operate in monopolistic ways. The histories of the railroad and telephone industries, they observe, suggest that ostensibly pro-competition laws often have perverse effects. The companies might initially bristle, as the two Chicago grain operators did, but eventually they grow comfortable with a regulatory regime that allows them moderate but stable profits, and which makes it less likely that new competitors will emerge.

"What ends up happening is that the regulated firms and the regulators decide that it's in their mutual interest to make a nice, cozy life for themselves," says Bruce M. Owen, a professor of public policy at Stanford University. "You get this really pathological relationship between the government and the regulated firms."

The two sides agree on one point: The decisions Congress makes this year have the potential to shape the structure and the ecology of the Internet for decades. The stakes, they say, are high.

Bottleneck Blues

The debate over network neutrality has simmered for several years, but it came to a boil last summer. In National Cable and Telecommunications Association et al. v. Brand X Internet Services et al., the U.S. Supreme Court ruled that cable broadband providers are offering a "data service" and not a "telecommunications service," which freed them from certain regulations of the Federal Communications Commission. Under this new framework, cable broadband companies are theoretically permitted to block their competitors' offerings and to strike special deals with content providers. For example, the cable-television giant Comcast could guarantee high-quality downloading of, say, videos from Apple Corporation's iTunes service; in such a deal, iTunes downloads would be given priority on Comcast's network over whatever else subscribers were downloading.

Under pressure from content providers both large and small, some members of Congress have introduced bills that forbid such deals and guarantee network neutrality. Cable and other telecommunications companies are lobbying furiously against those proposals. Much like the Chicago grain-elevator owners of the 19th century, the companies say government regulations would prevent them from making decent returns on their investments. Action is expected on several competing bills this month.

The scholars who have weighed in on behalf of network neutrality offer arguments of two types. First, they are making "bottleneck" arguments that directly descend from the classical debates about English ports and Chicago granaries. "Should these large entities like AT&T have a public duty to refrain from discriminating?" asks Mr. Wu, of Columbia. "Many people would argue that you need to have a nondiscriminatory infrastructure to have a market in the first place."

Barbara van Schewick, a senior researcher in telecommunications at the Technical University Berlin, wrote a working paper last fall in which she described scenarios in which broadband providers might have incentives to squeeze content providers. For example, Verizon might naturally want to exclude Skype and other voice-over-Internet companies from its broadband service, since Verizon's core business is, after all, selling telephone time.

But Mr. Wu and his allies also make arguments that are unique to the Internet. Much of the Internet's value, they say, lies in its vast number of users. An oft-cited example of this concept is that a fax machine is more valuable to its owner if many people own fax machines. Similarly, the Internet is more valuable to its users if they know that they can send and receive data to and from the entire network. If the Internet becomes bifurcated in various ways — if, for example, the sports Web site ESPN becomes available only on the telephone giant Verizon's broadband service — the entire network would become less valuable, both to its end users and to people who are dreaming up new Web sites and Internet applications.

Lawrence Lessig, a professor of law at Stanford with whom Mr. Wu has sometimes collaborated, also emphasizes the value of the Internet as a "commons" for public exchanges of ideas. Without network neutrality, the two legal scholars fear, large-scale broadband providers would inexorably shape the Internet into something akin to television — a medium with content that is passively consumed by its audience, and not actively created by a diverse public.

Mr. Wu points out that most broadband providers today offer much more "downstream" than "upstream" bandwidth — that is, they assume that Internet customers receive a lot of data and do not broadcast much of their own content. "I don't understand why you'd design your network that way," he says. "It's a perfect example of why every time the existing network owners favor one type of business, they disfavor others implicitly."

The implicitly disfavored businesses in this case, Mr. Wu says, are the "many to many" Internet applications, like YouTube, in which customers upload their own, often homemade, videos. "The nightmare scenario," he says, "is a world where AT&T doesn't approve a start-up like YouTube. YouTube's idea of video and AT&T's are so different. ... YouTube has been incredibly successful, but it's exactly the kind of thing that might not get anywhere in a world without network neutrality."

Stifling a Growth Industry

Net neutrality's skeptics marshal several types of argument in reply. First, they say that even though most cities now have a monopoly or duopoly in broadband (typically, one cable company and one telephone company offer the service), that situation will quickly change. New technologies will allow households to receive Internet connections wirelessly, or even through their standard electric lines. Those new services, they say, will eliminate any scarcity of broadband suppliers and make the market much more competitive, quashing price-gouging and anticompetitive behavior. In effect there will be not 14 grain elevators but hundreds.

"Within five years, all of these technologies will be emerging," says Philip J. Weiser, an associate professor of law and communications at the University of Colorado at Boulder and a co-author of Digital Crossroads: American Telecommunications Policy in the Internet Age (MIT Press, 2005). "On top of that, you'll also see improvements in digital processing and who knows what."

Some skeptics also suggest that network-neutrality regulation would perversely make it more, not less, likely that a near-monopoly will persist in the industry. With regulation in place, they argue, new companies will have fewer financial incentives to enter the market. Mr. Weiser says small, upstart wireless-broadband services might be able to raise crucial capital only if they form exclusive partnerships with, say, television networks or telephone companies. Network neutrality would prevent such deals.

And even if there were a permanent monopoly or duopoly in broadband, net-neutrality skeptics contend, providers would have no incentive to distort the markets for content providers or application developers. (A venerable theorem in economics holds that monopolists can extract excess profits only from the industry they monopolize, and not from "complementary" industries.) "I believe that most firms that are rational strategists would actually welcome new applications to the network because they will increase the value of the network," says Mr. Weiser.

Skeptics also say that the most important Internet innovations might involve improvements in the structure of the network itself — and that network-neutrality regulations would stifle such improvements. "If we were to have standardized electric power at an early time, we would have standardized on DC power," says Christopher S. Yoo, a professor of law at Vanderbilt University. "And that would be a problem." (Alternating-current systems are cheaper and safer than the direct-current systems they replaced.)

"A lot of this debate has to do with whether you believe innovations happen at the edges of the network or at the core of the network," says Mr. Weiser. "Some people are almost religious about this issue." Supporters of the Internet's "end to end" architecture, like Mr. Wu and Mr. Lessig, believe that the most important innovations happen at the edges. They say the "core" of the network — its central backbone — should not be tinkered with in ways that would make it difficult for innovators to create and install new applications.

Finally, the skeptics say, enforcing network neutrality would inevitably require a cumbersome regulatory apparatus, no matter how simple the principle itself might sound. "If you have an unwilling network provider, they have basically an unlimited bag of dirty tricks they can play to frustrate the quality of the connection," says Mr. Yoo. "So you have to regulate each and every one of those things." It would be better, he suggests, simply to give the unwilling network what it wants: the right to block certain content providers from its network.

Stiff regulation would be especially inappropriate, Mr. Yoo adds, in an environment "where the technology is changing and costs are falling and prices are jumping all over the place. It's just very unworkable."

Commons Ground?

Columbia's Mr. Wu denies, however, that regulations would necessarily become complex. "Network neutrality basically says that you can't pick favorites," he says. "I don't think that's a complicated rule. On the margins, it may provoke some complications — but it's worth it. What we're talking about is protecting the flourishing market on top of the Internet. There's been a long history of very important and effective antidiscrimination rules, going back to common carriage in ports."

Whatever their disagreements on net neutrality, the two sides agree that the regulatory framework that is embraced this year will very likely persist for a long time.

In his new book, The Wealth of Networks: How Social Production Transforms Markets and Freedom (Yale University Press), Yochai Benkler, a professor of law at Yale, points out that "a series of choices made during a short period in the 1920s" gave the United States, Britain, and Continental Europe extremely different radio systems. The structure and culture of the American broadcasting system — or the Internet — eventually came to seem natural and inevitable, but all three systems are in fact deeply shaped by legal and political decisions. There is no reason to believe, Mr. Benkler writes, that "the most efficient institutions win out in the end."

Mr. Weiser — whose opposition to network neutrality is more moderate than Mr. Owen's or Mr. Yoo's — sees a feasible compromise.

First, he says, large, existing broadband providers should not be permitted to block any content providers outright. (Small, upstart firms should be given more leeway, he argues.) Second, the government should scrutinize "quality of service" guarantees (like the hypothetical Comcast-iTunes deal) to see if they have anticompetitive effects. Third, broadband providers should be required to provide a minimum amount of bandwidth on a "best effort" basis, so that all content providers can be confident that they will receive a baseline amount of transmission service. Finally, Mr. Weiser says, the Federal Trade Commission should ensure that broadband providers are accurately describing their networks to their customers.

"Unfortunately, that four-part suggestion is nowhere in the debate today," he says. "Instead you have a polarized debate in which one side says that we need to maintain the Internet exactly as it has always existed, which I think is a simplification of the truth, and the other side says there's no reason for any regulation at all, because we will be trustworthy. I think there's got to be an answer somewhere between those two."