Sensing History: Yoo Re Cherry

Dave Farber posted a response by Chris Yoo to Barb Cherry's post about myths and historical errors. Here's Chris's reponse in full. To me, it seems that he is conceding that she's right about the history, that antitrust says nothing about ISP competition, and that a few ISPs control most of the Internet in the U.S. But read it for yourself:

From: Christopher S. Yoo [mailto:[email protected]]

I don't pretend to be an expert on the history of common carriage regulation. Barbara has spent far more time thinking about this than I have, so I always appreciate hearing her reactions and learn from reading her work. That said, here are a few thoughts.

It is true that common carriage long predates both the Granger Movement and the Interstate Commerce Act of 1887. That said, one of the central problems is that the historic justifications for common carriage have not aged very well. Often times the common carriage obligations were regarded as a quid pro quo for a government grant of some economic privilege. Other times they were justified because the industry was "affected with a public interest," a concept that is usually traced to the landmark Supreme Court case Munn v. Illinois (1876). The Supreme Court struggled to imbue that standard with content (along with a number of early treatises trying to make sense of the concept) and would ultimately abandon it as analytically empty in Nebbia v. New York (1934). Legal scholars, such as Thomas Nachbar and James Speta in addition to Barbara, have attempted to recover lessons from this era. I have never spoken to Barbara about this in particular, but both Tom and Jim have noted the difficulty in extracting any useful lessons from the history.

The rest after the jump.

Put another way, even though it is true that common carriage was not originally based on statutes or the theory of natural monopoly, it is not clear what to make of that fact. One problem is historical. A review of the extensive early 20th century literature examining common carriage reveals that the original theory of common carriage was not that well articulated and failed to incorporate the history that Barbara has identified. The subsequent development in different directions over the past century or so is as much part of the history of common carriage as the history Barbara identifies. And as I indicated in my previous post, there is a counter history that dates back to the earliest days of the cable industry. The Supreme Court consistently rejected attempts to turn cable into a common carrier. So even from a historical perspective, extending common carriage obligations to cable modems would be just as radical a change as lifting common carriage obligations from DSL. The debate should thus consider the potential adverse impact that increasing the regulation of cable would have on competition, investment, innovation, and speech at the same time it considers the potential adverse effects of decreasing the regulation of telco-based broadband.

More importantly, we have learned a lot over the past 150 years. While history is often illumining, it would seem strange not to hold the old justifications up to modern standards to see if they still make sense. We now understand that price discrimination and product differentiation can provide benefits that we failed to appreciate a century ago. Thus, even if tort law once imposed a duty of nondiscrimination and just and reasonable prices, we should only adhere to that if it continues to make sense to do so in light of what we know today. We should also draw lessons from previous attempts to implement nondiscrimination regimes, both by the FCC (leased access to cable systems, UNE access under the 1996 Act) and by antitrust courts (essential facilities doctrine, Robinson-Patman Act). This history is not encouraging, particularly where the product and interfaces are complex and when technology is changing rapidly.

That's Chris's response to Barb.

He continued with a response to me, including to a remark I made to him (that Farber did not post) about being disappointed that he did not address ISP competition.

Two last thoughts regarding concerns that John Quarterman mentioned: He raises skepticism about the competitiveness of the first mile. I agree that the world would be better if the market for last mile service was more competitive and there were at least five independent providers in each market. That said, the trend seems to be going in the right direction. DSL and cable modem providers seem to have built out most of the country. The key is generating a third pipe to the home. In this regard, wireless seems to be the most promising. As of the beginning of 2005, there were no wireless broadband. By last September, they had 45 million. I realize that as of now wireless broadband can only provided limited bandwidth, but they are working on it. I find it interesting that the wireline network providers have repeatedly increased bandwidth without increasing the cost of their basic service one penny. I also find it interesting that small rural wireless broadband providers have submitted filings to the FCC indicating that they need to able to discriminate against high bandwidth applications (particularly video) if they are going to be able to provide adequate service. That said, the stakes are about to get much higher as the market approaches saturation. Instead of retooling existing technologies to race for new customers, cable and telephone companies are increasingly having to make significant investment in new technologies to try to provide greater value to existing customers. As the days of millions of new customers and the fiber glut recede into the past, network providers will inevitably find new ways to compete. Because change always creates winners and losers, I suspect that the evolution of competition will be a source of considerable friction with both business partners and consumer groups.

Um, let them eat wireless? I'm not surprised the wireline providers have increased bandwidth without increasing end-user pricing, considering that their pricing is already higher than in many countries. Change has created competition and universal access to broadband in other countries, but not in the U.S. The U.S. hare has been outrun by the Japanese (and Korean, and French, and ...) tortoise. More on that in another post.

Back to Chris:

John also notes that I do not discuss ISP competition. I may have missed it, but I didn't see any mention of ISP competition in any of the e-mail traffic or on the webpage he posted. That said, it is not clear that the history of common carriage has anything to say about ISPs. Common carriage has traditionally applied only to customers, not to other carriers, providers, or competitors. Moreover, the FCC has repeatedly found the market for ISPs to be quite competitive, except in the WorldCom-MCI merger when it required the spinoff of some backbone assets before it would provide regulatory clearance. What also strikes me is how much more heterogeneous the network has become, both in terms of structure and business relationships. Secondary peering and multihoming have reached the point where 70% of the nodes can interconnect without passing through the public backbone. The advent of new business relationships such as paid peering and partial transit has made the economic environment richer and more complicated as well. The effect has been to weaken the hold of the Tier-1 ISPs. The world is more complicated, but more open to creative solutions than ever before.

Chris's previous post was largely about competition as related to antitrust, so I was indeed disappointed that he mentioned competition in other arenas, such as telephone, but not in ISPs. His latest response mentions common carriage but not antitrust. This makes me wonder if antitrust has anything to say about ISP competition, which was the crux of the original issue. Also, while there was ISP competition back in the day of Worldcom, there isn't now. Plus, if "70% of the nodes can interconnect without passing through the public backbone" that means that most of the Internet is now at the mercy of a few ISPs. That is the problem. That is why we need net neutrality legislation.

-jsq