Social Welfare: Reed Asks Yoo

DPRPhotoSmall.jpg David P. Reed asks a question and Christopher S. Yoo responds on Farber’s Interesting People list. I’m posting both in full here, with my thoughts at the end; basically, law isn’t a science, and anecdotes can turn into legal cases; some have already regarding net neutrality.
From: David P. Reed []
Sent: Saturday, May 10, 2008 11:50 AM
To: David Farber
Cc: ip
Subject: Re: [IP] re-distribution of op-ed on Net Neutrality — a reaction and a reply from one of the authors

I read through the long comment by Chris Yoo below, and as a non-lawyer interested in policy, I ask the following simple question:

Is there a well-regarded (one might ask for scientifically reasoned) argument that antitrust law as currently interpreted and practiced has a substantial impact measured in some currency like $ on social welfare?

Otherwise this entire argument is about nothing more than vaporware proceeding from a faith that competition (however loosely defined) creates social welfare best. AFAIK, this is largely an article of faith, just as the “End of History” was a grand article of faith posited by many of the same people as “truth”.

It is just not fair to imply that the core of “today’s settled antitrust law” carries even the level of weight as Darwin’s Theory of Evolution. There have been no replicable studies of its practice.

Law professors and lawyers who don’t challenge its truthiness squarely are merely behaving as dogmatic mandarins always do – asserting authority of professional status, rather than rigor of reasoning, experiment, or argument.

I say this not as FOX News or Hillary Clinton would call an elitist, but as a person who genuinely is unconvinced by magical faith in authorities.

That’s Reed’s question. Yoo’s response, and my thoughts, after the jump.

From: Christopher S. Yoo [] I think David Reed’s question is right on. The areas of antitrust law most closely related to the network neutrality debate are the law of vertical integration and vertical restraints (preference to proprietary inputs, exclusivity contracts, tying to implement price discrimination, etc.). Unsurprisingly, the theoretical literature is all over the map, with studies coming down on both sides.

Perhaps the most important development in antitrust law is the improvement of empirical tools. These have given us real insight into whether the current focus of antitrust law is actually benefiting consumers. There have been two recent surveys of the empirical literature that are quite telling.

The first was conducted by four members of the FTC’s staff and was published in the International Journal of Industrial Organization. It examined 22 existing empirical studies on vertical restraints (including a number on issues like exclusivity in cable television that are quite analogous to network neutrality) and found “a paucity of support for the proposition that vertical restraints/vertical integration are likely to harm consumers.” Indeed, only one study unambiguously found that vertical integration harmed consumers, and in that case, “the losses are miniscule ($0.60 per cable subscriber per year)”. On the other hand, “a far greater number of studies found that the use of vertical restraints in the particular context studied improved welfare unambiguously.” The survey thus concluded that “[m]ost studies find evidence that vertical restraints/vertical integration are pro-competitive.” The weight of the evidence thus “suggests that vertical restraints are likely to be benign or welfare enhancing,”

Another survey was conducted by two respected academics (Francine Lafontaine and Margaret Slade) and was just published last month by the MIT Press in the Handbook of Antitrust Economics. It surveyed twenty-three published empirical studies of vertical restraints. Despite the relatively small sample size, the authors found the empirical evidence to be “quite striking,” “surprisingly consistent,” “consistent and convincing,” and even “compelling.” As a general matter, “privately imposed vertical restraints benefit consumers or at least do not harm them,” while government mandated vertical restraints or restrictions on vertical restraints “systematically reduce consumer welfare or at least do not improve it.” Together “[t]he evidence … supports the conclusion that in these markets, manufacturer and consumer interests are apt to be aligned, while interference in the market [by the government] is accomplished at the expense of consumers (and of course manufacturers).” The authors conclude that “the empirical evidence suggests that in fact a relaxed antitrust attitude towards [vertical] restraints may well be warranted.”

So there does seem to be some scientific support for claim the current thrust of antitrust law as it relates to the network neutrality debate (as well as other areas of competition policy) is promoting economic welfare. It is, of course, always possible to quibble with empirical studies. But I would think that this growing corpus of work published in peer reviewed journals should be sufficient to put the burden on the other side of the debate to adduce at least some evidence to the contrary.

Which brings me to one of my pet peeves in this debate. It seems to me that the question David asks should apply with equal force to both sides of this debate. In other words, we should also inquire whether network neutrality proponents have “a well-regarded (one might ask for scientifically reasoned) argument” that mandating network neutrality would have “a substantial impact measured in some currency like $ on social welfare.”

The problem is that when asked the types of questions that that David Reed poses here, the other side of the debate has generally either invoked anecdotes or pointed to straight theory papers. Anecdotes and theory can be helpful, but cannot ultimately resolve how likely it is that the practice will enhance social welfare and seem inadequate when weighed against the sizable peer-reviewed empirical literature I have described above. At this point, the onus would seem to lie on the other side of the debate to prove that their arguments are not mere “vaporware” that are “largely an article of faith.”

I take it as high praise that the other side of this debate has always complimented me for basing my arguments on facts. I would encourage anyone interested to look through the transcripts of the FCC hearings, the House Judiciary hearings, the House Commerce hearings, or any of the many deliberations in which I have participated and make their own assessment of which side is trying to get by simply by invoking a “magical faith in authorities.”

In other words, I think that David Reed is asking absolutely the right questions. I just would like to see both sides of the debate held to the same, appropriately high standards.

I see Dr. Yoo has still not directly addressed competition in Internet provision, such as the massive decrease in numbers of ISPs as the FCC stopped enforcing UNE. As for anecdotes, the recent incidents of AT&T censoring Pearl Jam and other bands, Comcast and Cox stifling BitTorrent (and sometimes Lotus Notes), etc., are the sorts of anecdotes that lead to lawsuits, which would exercise the antitrust law that Dr. Yoo advocates as a solution. At least one of them already led to complaints to the FCC.

As for customer harm, why is it that Japan, which has some real ISP competition, has Internet access speeds ten times that of the U.S.? Why is it that countries that have four or more ISP competitors tend to have faster customer speeds?

Note I did say “customer”, not “consumer”. “Consumers” passively consume what broadcasters send them, as in cable television. Internet participants can just as easily produce and share content, if ISPs don’t get in their way. This is what net neutrality is about.

Framing the discussion as about “consumers” assumes centralized control. The Internet is about participation.