A ‘Nobel’ Reward for Auction ‘Engineers’

Paul R. Milgrom and Robert B. Wilson, both of Stanford University, have been jointly awarded this year’s Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel for “improvements to auction theory and inventions of new auction formats”. 

The practice of selling valuable items through auctions to a highest bidder has been in vogue since ages. Two simple auction formats, namely the English auction, where ascending bids are made until one buyer is left willing to pay a certain amount and a higher bid is not received during a given time period; and the Dutch auction, where auctioneer sets a high opening price which is gradually reduced until a bidder is found, have dominated the scene for long. However, as their usage has expanded covering variety of assets, auctions have become more complex, besides acquiring far greater importance.  

In fact, it is William Vickrey, who first came out with auction theory in 1960s by narrowly focusing his study on each person’s private or subjective evaluation of the goods or services for sale. But in reality, each bidder’s valuation of the good being auctioned is not independent of the other bidder’s valuation as is noticed in English auction, where each bidder, after having started at a low price, moves on to higher prices depending on the price quoted by other bidders, which is, of course, again based on their own private information, and thus arrives at a ‘common value’. 

It thus shows that entirely private values are a rarity in auctions. Robert Wilson who became the first economist to research taking common value—a value that is uncertain at the beginning of the auction, but in the end it is the same for everyone—of the good into consideration, using game theory, showed in three of his classic papers published in 1960s and 70s how the best bidding strategies in common value auctions leads to low bids as participants in an auction try to avoid the ‘winner’s curse’—overestimating the common value and winning the auction at too high a price. He also showed that the problems caused by the winner’s curse are even greater when some bidders have better information than others. It means that bidders with information disadvantage will bid even lower or may even abstain from participating in the auction. 

Analyzing the bids in auctions with private and common values have turned out to be trickier than what Vickrey and Wilson thought of, for the technology/specialization owned by a bidder significantly differentiates the private values of a good from one bidder to another. For instance, private value of an oilfield not only depends on the estimate of the oil reserve, but also on the cost of extraction which depends on the technology used by a company and hence it varies from bidder to bidder. This riddle was finally cracked by Paul Milgrom—incidentally a doctoral student of Robert Wilson—in a couple of papers published by him in the 1980s. His research revealed that the auction structures that elicit more private information from bidders, such as English auctions, where every bidder observes who bids what and who drops out at what price, reduce the winner’s curse problem vis-à-vis with formats such as sealed bids that divulge very little private information. Looking at this link, it becomes imperative that a seller, in his own interest of maximizing his revenue, may have to provide bidders as much information as possible about the goods being auctioned. 

To their credit, Milgrom and Wilson have not limited themselves to developing fundamental auction theory. They have also put their theoretical knowledge to practical use: evolved new and better auction formats for complex situations where the existing formats were found inadequate. In the 1990s when the US Congress permitted the Federal Communications Commission to use auctions to sell radio spectrum to telecom companies, it posed a big challenge to the Commission, for no one knows how it works in the light of the value of a piece of spectrum in a specific region as it also depends on the other frequency bands owned by a specific bidder. Interestingly, to tackle this problem, Milgrom and Wilson, partly in association with Preston McAfee, came up with a new auction format, the Simultaneous Multiple Round Auction (SMRA) that offers all pieces of spectrum simultaneously for bidding. The SMRA model offered scope for participants to bid on all items in a number of rounds as a result of which some information about bids and prices is revealed to bidders whereby the winner’s curse is reduced. Since then many other countries, including India, have followed this model successfully. 

These two economists, who developed auction theory, turning engineers have also designed auction models, which have benefited buyers, sellers, and society as a whole, besides enabling them to win a Nobel Prize.