I spent Thursday evening and Friday at UCLA’s event in memory of one of my main mentors, the late Harold Demsetz. The event ended late yesterday afternoon with a bang: the keynote speech by MIT economist Paul Joskow (pictured above.) It was first-rate.

Various people through the day talked about the importance of various of Harold’s articles. Joskow focused on his 1968 article in the Journal of Law and Economics, “Why Regulate Utilities?” Joskow pointed out that the basic insight in that article is that even if an industry is a natural monopoly, that doesn’t imply that getting to a competitive price requires government regulation of the natural monopoly’s price and output. Instead, Demsetz had argued, you could have competition for the market rather than competition within the market. How so? Various firms could compete on price for the right to be the lone producer.

Joskow pointed out that that insight seems obvious but also that when Demsetz came up with it, it wasn’t obvious at all. He mentioned his MIT colleague Jean Tirole, who, in his textbook with Jean-Jaques Laffont, devotes 70 pages to laying out the insight. Joskow had said to Tirole that he used 70 pages to say what Harold Demsetz had said in 11 pages. Tirole had responded, “Yes, but we have more equations.”

A large part of Joskow’s talk was on laying out how extensive the application of Demsetz’s insight has been, both in local U.S. governments and in poorer countries.

He led with a story about his city of Brookline, Massachusetts. He pointed out just how left-wing Brookline is. It was one of the last cities in Massachusetts to get rid of rent control. In the last presidential election, 88% of the voters voted for Biden. Etc.

But some years ago, there was a choice to be made between having an ambulance service run by the city government or putting it out for competitive bidding. The government chose the latter. The results were so good that later the Brookline government put garbage collection out for bidding.

Joskow’s point, of course, is that even some left-wing governments understand the power of competition in what might be natural monopolies.

He then turned to the record in poorer, developing countries. In many of those countries, governments are doing what Demsetz recommended: having competitive bidding to be the one producer in the natural monopoly industry.

Joskow also pointed out that economists who are critical of Demsetz’s article will point out, correctly, that there are potentially big issues in such contracting having to do with uncertainty, sunk costs when it comes time to rebid, and a number of other things. But, he pointed out, many of these critics don’t then point to tough issues that arise when governments do it the old-fashioned way, either with government production or with regulation of a natural monopoly. In comparing the reality of competitive bidding with standard government regulation, they leave the latter unexamined. In doing this, they are guilty of what Demsetz called, in another path-breaking article, the Nirvana approach. (I’ve forgotten whether Joskow used this term. A number of economists through the day did refer to that insight. To a person, they called it the “Nirvana fallacy.” I had pointed out in the lunch roundtable that Harry DeAngelo of USC emceed and in which Joe Kalt of Harvard, Bob Topel of the University of Chicago, and I shared our reminiscences, that Demsetz had actually called it the “Nirvana approach.” The Nirvana approach consisted of three fallacies: the “free lunch” fallacy, the “grass is greener” fallacy, and the “people could be different” fallacy. So far, I don’t seem to have persuaded anyone to correctly refer to Harold’s term.)

Joskow then went on to talk about how within various industries thought to be natural monopolies, there are components that are not. He gave the example, which many people have given, of electric utilities. Generation of electricity can be provided competitively.

He ended with two points. First, he noted, a natural monopoly is not necessarily a natural monopoly forever. So it’s important to pay attention to how technology evolves so that we can pare down the parts that governments need to put to a bid and for the rest rely on regular competition within markets.

The final point Joskow made is that the awesome citation count that Harold’s article has received is a major understatement of its influence. Citations measure references in other academic articles. But when governments in other countries make decisions based explicitly on Demsetz’s insight, that doesn’t show up in citations. But, he said, in these countries, “the name that’s always mentioned is Harold Demsetz.”



I deal with natural monopoly in The Concise Encyclopedia of Economics entry, Monopoly, written by George Stigler. It’s at the end of his article. I don’t discuss, but should have discussed, Demsetz’s insight.

Steve Globerman and I, in our recent book, The Essential UCLA School of Economics, do discuss the Nirvana approach at some length. You can download the book here.

This is my bio of Harold in The Concise Encyclopedia of Economics.