The BCS as Path Dependency

I am ordinarily not much of a fan of the market-failure argument, whereby someone purports to show that the market has achieved an inferior outcome because of what economists call externality or public-goods problems, and that a particular government proposal would improve things.

But one often interesting assertion of market failure involves something known as path dependence. In a path-dependency problem, a market standard is chosen for historical reasons, and despite its inferiority persists because the costs of switching to a new standard would be too high. Americans persist in foolishly avoiding the metric system, for example, or Microsoft Windows maintains its dominance only because so many people are already Windows-fluent that the cheapest path to developing new software lies in designing it to be Windows-compatible, even though the operating system itself is lousy.

Many of these assertions are disputed. The Dvorak typewriter was said to allow typists to type faster, but the QWERTY keyboard, which according to some was actually designed to slow typists down so that the keys wouldn’t clog up, has persisted despite its inefficiency. Typists all train on QWERTY, and so keyboard makers keep building QWERTY keyboards. And no one bothers learning an alternative to QWERTY, because that’s the layout on all the keyboards. But an article some years ago called “The Fable of the Keys” in The Journal of Law and Economics found that most of the studies purporting to show that typing was faster on Dvorak were conducted by people with a financial interest in that layout’s success, and that well-trained typists could type as well on either layout.

And so that brings us, in a roundabout way, to LSU vs. Ohio St. It seems clear to me that if we were to design a college football postseason competition from scratch, it would bear no resemblance either to the bowl system in general or the BCS system for deciding a “champion” in particular. The only reason there is a BCS system is because the old lineup of big-time bowls with conference ties – Big Ten and Pac-10 in the Rose, Big 8 in the Orange, SEC in the Sugar – failed to produce a definitive 1 vs. 2 matchup, and fans demanded that the system be altered. But not by eliminating the bowls in favor of a playoff. Instead, the New Year’s system was woven into the championship framework, which has evolved into the separate BCS national-championship game we have now.

But as everyone who follows college football knows, there are a lot of teams with plausible claims on that game – the conference champions USC and West Virginia in particular, and arguably even Hawaii. A sensible system would look like every other NCAA sport, with a playoff elimination bracket culminating in a final game matching the survivors. But that has been successfully opposed by the people who make so much money off the BCS bowls (and by those who benefit from the absurdly large number of bowls overall, which only require six wins over a Division IA team as a qualification). Yet clearly the NCAA is content with the current system, in which computers and voters determine who one of their champions is.

And so we are locked into a system that most fans dislike, because the costs of switching are too high. As I indicated these kinds of arguments ordinarily do not impress me, and this is obviously not a question for government intervention of any sort, but the college-bowl system as a path-dependent outcome clearly inferior to what would occur if we could start over seems clear to me. I am hard-pressed to come up with any argument for the system the way it is from the point of view of total welfare, including the welfare of the fans who watch the games. (Unless arguing over the title game itself generates a lot of fan utility, which seems like a cheap way out.)