How to Think About Net Neutrality
Saturday, September 26, 2009
History suggests the best course is not to borrow trouble, as the old phrase goes, but to let the system 'develop itself in its own way,' and act only in response to actual problems.
In a speech at the Brookings Institution September 21, FCC Chairman Julius Genachowski announced a rulemaking to embrace the concept of “net neutrality,” and the telecom world is crackling with the energy of renewing a fight continued for some years now. It is more than just a new round, too, because the speech almost casually added wireless as well as wireline to the scope of FCC’s concerns, which raises new and serious sets of uncertainties.
A problem with “net neutrality” has always been that no one knows exactly what the term means. In extreme form, its proponents would treat all Internet service providers as the equivalent of digital public roads, with anyone allowed to use the road for anything, with everyone paying the same, or nothing, and with no surcharges based on the weight of the trucks, and no efforts to deal with congestion except to build more roads. No road provider would differentiate charges according to the value of the cargo, contrary to the old value-of-service pricing built into ancient railroad and truck tariff regulation, but never part of communications rules.
The FCC took a small, but only a small, step toward this view in its 2005 policy statement that consumers are entitled: (1) to access the lawful Internet content of their choice; (2) to run applications and services of their choice, subject to the needs of law enforcement; (3) to connect their choice of legal devices that do not harm the network; and (4) to competition among network providers, application and service providers, and content providers.
FCC Chairman Julius Genachowski announced a rulemaking to embrace the concept of 'net neutrality,' and the telecom world is crackling with the energy of renewing a fight that has gone on for some years now.
The trigger for the statement was the Madison River affair, in which a phone company was accused of blocking access to VoIP. It paid $15,000 to settle the matter, so the FCC’s power to act against the practice was never confirmed.
The statement was advisory only, but enough to warn ISPs that if they tried to leverage their control over customer access to the Internet into control over complementary services or content, then they would have a problem with the FCC, even if Internet providers are not subject to the common-carrier rules that govern the telephone system, and even if the exact authority is unclear. No telecom company has been willing to argue that it has a right to cut off VoIP because it interferes with the company’s telephone franchise, so the truce has been maintained.
At Brookings, Genachowski took another step, perhaps a leap, in the direction of net neutrality by expressing an intent to make the four principles into rules rather than guidelines, and to add two more:
(5) Non-discrimination: “Broadband providers cannot discriminate against particular Internet content or applications. This means they cannot block or degrade lawful traffic over their networks or pick winners by favoring some content or applications over others in the connection to subscriber’s homes, nor can they disfavor an Internet service just because it competes with a similar service offered by that broadband provider.” (Plain English version: content providers cannot pay you to provide preferential access, nor can you favor your own VoIP or content.)
Caveats: “This principal will not prevent broadband providers from reasonably managing their networks … [such as for congestion, nor will it] constrain efforts to ensure a safe, secure, and spam-free Internet experience or to enforce the law” (including copyright law). Genachowski also spoke vaguely of some leeway for “broadband providers offering managed services in limited circumstances.”
(6) Transparency: “Providers of Internet access must be transparent about their network management practices.”
And, as noted, the promised rule making will cover wireless as well as wireline Internet access.
A problem with 'net neutrality' has always been that no one knows exactly what the term means.
Genachowski justified the proposal by the values of openness: “The Internet's open architecture pushes decision making and intelligence to the edge of the network, to end users, to the cloud, to businesses of every size and every sector of the economy, to creators and speakers across the country and across the globe. In the words of Tim Berners-Lee, ‘the Internet is a blank canvas allowing anyone to contribute and to innovate without permission.’”
He dismissed the value of our usual guardian of openness, competition, with two arguments—that the choice of Internet providers has narrowed, and that ISPs are also providers of telephone and cable television services, which increasingly compete with content sent over the Internet.
One can concede the core concern at stake here. Assuming that an ISP has any monopoly power at all, it would be unwise to allow it to leverage that power into control over all activities that rely on it. One parallel here is that we do not allow a railroad to extract all the profits from shippers who lack alternatives, and there are good reasons for this.
That said, the speech still leaves one baffled as to the agency’s collective thought processes. Immediately:
1. The new non-discrimination principle says that there can be no such thing as integration between an ISP and a content creator. This might make sense for a monopoly, but not in a system in which new forms of access are growing rapidly.
This approach assumes that each customer should have only one connection to the Internet, which must provide access to everything.
Further, this approach assumes that each customer should have only one connection to the Internet (or perhaps two, one fixed and one mobile), and this connection must provide access to everything. Why should this be so? Perhaps a better model would be for every consumer to have several connections, and for each of these to carry unique and proprietary content. Some could carry their own content and some could be dumb pipes, awaiting the genius of the edge. Each could price according to the attractiveness of its offerings. Let a hundred flowers bloom. But if we do not allow ISPs to integrate with content providers, and thus discriminate against others, we will shut down this entire line of possible development.
2. Is the FCC seriously thinking of applying openness obligations to wireless, with its multitude of providers and exploding usage? If so, why? And if this article were a diagram, an arrow would now point back to point one.
3. If the agency wants to talk about the incentives created by telephone and cable TV, perhaps it should also consider the regulatory structures governing these services, with their residue of inefficiency and cross-subsidy. Let’s look at universal service and rural subsidies, for example; perhaps hearings can be held in one of the beneficiary cities, such as Palm Springs or Hilton Head.
4. The academic mafia aside, these systems are not dumb pipes. They are highly sophisticated computers running on millions of lines of code, and managing traffic in incredibly complex ways. Why should their providers not share in the value created by the increased efficiency with which they move information?
5. Running through considerations of every kind of network since the canal and the railroad are the same fundamental dilemmas of pricing. Three parties are involved: consumers, network operators, and content providers (or shippers). All networks involve high fixed costs and low marginal costs. The question is, how do you price the service so as to allocate the fixed costs across the different interests, and how do you provide incentives for the proper level of investment? (Andrew Odlyzko has some interesting history on these issues.) The answers are non-obvious. For example, as a shipper of high-value manufactured goods, I should want the railroad to charge me more than it charges the coal company.
In many ways the Internet is more like a transportation network than a telephone, and the value of service pricing has always been part of transportation ratemaking and regulation.
There are no clear right answers, from either an economic or equitable point of view, but there are some wrong ones. And the most wrong lets players in the game use the oldest business model in creation, which is “find some sunk capital and steal it.” There may be genius at the edge, but without the network, “full many a flower is born to blush unseen,” and genius should be damn happy to share its fruits with those who make the seeing possible.
In any case, in many ways the Internet is more like a transportation network than a telephone, and, as Odlyzko notes, value-of-service pricing has always been part of transportation ratemaking and regulation.
6. What are the problems with which the FCC is concerned? The speech has three tales of woe: Madison River, an affair last year where Comcast was sued for traffic management relating to P2P, and “one service provider den[ied] users access to political content.”
Even in government by anecdote, one would like to see better stories, and perhaps even some new ones from time to time. (Years ago, reviewing the record for a proposed FTC rule, I wrote, “The agency is about to enact a provision of federal law because in a small town in Ohio ten years ago, a pair of red pajamas ran in the wash.”)
If these are the best horror stories the agency can come up with, then there is something to be said for the genius of the common law, which focuses on a specific fact situation, and then reasons from the specific to the general, rather than starting with an abstract assertion that there is cause for concern.
The history of regulation that starts from a priori assumptions is not encouraging:
In 1872, a committee on railroad amalgamation was appointed ... After taking a vast amount of evidence, they proceeded to review the forty years of experience ... They showed with grim precision how, during that period, the English railroad legislation had never accomplished anything which it sought to bring about, nor prevented anything that it sought to hinder. The cost to the companies of this useless mass of enactments had been enormous, amounting to some £80,000,000; for these were 3,300 in number and filled whole volumes.
The decision was made not to borrow trouble, as the old phrase goes, but to let the system "develop itself in its own way," and act only in response to actual problems. Adams says this worked quite well. You can read about it on the Internet.
FURTHER READING: In other articles for THE AMERICAN, DeLong wrote “Preparing the Obituary” on the decline of the newspaper industry and “Avoiding a Tech Train Wreck” on the intersection of technology and politics. His other pieces include “The Coming of the Fourth American Republic” on how the Special Interest State that has shaped American life for 70 years is dying and “Behind the Green Dam” on how the controversies over Internet filtering are only just beginning.
Image by Darren Wamboldt/Bergman Group.