Friday, February 20, 2004

Everyone complains about highway congestion but no one does anything about it. Well, not quite. Lots is done but mostly the wrong things. Anyone with a passing acquaintance with economics knows that if price does not ration, something else will. In this case, it is crowding which results in substantial dead-weight losses.

Not long ago, time-of-day highway access pricing was dismissed as making sense only to economists . Yet, the feasibility of the approach is now clear because, not only have there been a mountain of studies but, there have also been numerous trials around the world that demonstrate that Econ 1 really works. Singapore since 1975 is an auspicious case as is London since last year. Now, even the socialist Mayor of London gets it.

Not here. The latest U.S. study of the problem, by the Highway Users Alliance, finds 233 crossings and interchanges in the U.S. that, together, account for one-half of all the congestion. Right on schedule, there is a multi-billion transportation bill pending in Congress that promises relief via highway construction and more public transit projects

Pricing is usually fended off as "inequitable" -- and it would undermine the porkfest, the perennial political favorite. We all know how "equitable" pork spending tends to be.

Milton Friedman has demonstrated that the costs of government are not simply measured by the taxes we pay (or by what they spend), but must include the costs of mandates that they routinely impose on the private sector. In this case, not only are wasteful highway and transit projects added as a putative solution but an uncountable number of land use regulations and construction barriers have been put in place to create settlement patterns that are supposed to limit auto traffic. As mentioned in yesterday's blog entry, these are expensive and ineffectual.

The costs of avoiding the road pricing option are, then, staggering. Most textbooks refer to congestion as an example of a market failure. It is really much more a policy failure.