Economists have lately been called upon not only to analyze markets, but to design them. Market design involves a responsibility for detail, a need to deal with all of a market’s complications, not just its principle features. Designers therefore cannot work only with the simple conceptual models used for theoretical insights into the general working of markets. Instead, market design calls for an engineering approach. Drawing primarily on the design of the entry level labor market for American doctors (the National Resident Matching Program), and of the auctions of radio spectrum conducted by the Federal Communications Commission, this paper makes the case that experimental and computational economics are natural complements to game theory in the work of design. The paper also argues that some of the challenges facing both markets involve dealing with related kinds of complementarities, and that this suggests an agenda for future theoretical research.

Abstract not found

We report on the design of the new clearinghouse adopted by the National Resident Matching Program, which annually fills approximately 20,000 jobs for new physicians. Because the market has complementarities between applicants and between positions, the theory of simple matching markets does not apply directly. However, computational experiments show the theory provides good approximations. Furthermore, the set of stable matchings, and the opportunities for strategic manipulation, are surprisingly small. A new kind of “core convergence ” result explains this; that each applicant interviews only a small fraction of available positions is important. We also describe engineering aspects of the design process.

The entry-level market for American gastroenterologists was organized by a centralized clearinghouse from 1986 to 1996. Before, and since, it has been conducted via a decentralized market in which appointment dates have unraveled to well over a year before the start of employment. We find that, both before and after the years in which the centralized clearinghouse was used, gastroenterologists are less mobile and more likely to be employed at the same hospital in which they were internal medicine residents than when the clearinghouse was in use. This suggests that the clearinghouse not only coordinates the timing of appointments but also increases the scope of the market, compared to a decentralized market with early appointments.

The deferred acceptance algorithm proposed by Gale and Shapley (1962) has had a profound influence on market design, both directly, by being adapted into practical matching mechanisms, and, indirectly, by raising new theoretical questions. Deferred acceptance algorithms are at the basis of a number of labor market clearinghouses around the world, and have recently been implemented in school choice systems in Boston and New York City. In addition, the study of markets that have failed in ways that can be fixed with centralized mechanisms has led to a deeper understanding of some of the tasks a marketplace needs to accomplish to perform well. In particular, marketplaces work well when they provide thickness to the market, help it deal with the congestion that thickness can bring, and make it safe for participants to act effectively on their preferences. Centralized clearinghouses organized around the deferred acceptance algorithm can have these properties, and this has sometimes allowed failed markets to be reorganized.

harvard.edu). This work was partially supported by a grant from the National Science Foundation. We also thank David Brenner, M.D., Fred Gorelick, M.D., and Sean Sheehan, M.D., for agreeing to be interviewed about the demise of the gastroenterology match. Robert Beran, Elliott Peranson, and Cody Webb, M.D., helped orient us to the ongoing events in

The design of the New York City (NYC) High School match involved tradeoffs between incentives and efficiency, because some schools are strategic players that rank students in order of preference, while others order students based on large priority classes. Therefore it is desirable for a mechanism to produce stable matchings (to avoid giving the strategic players incentives to circumvent the match), but is also necessary to use tie-breaking for schools whose capacity is sufficient to accommodate some but not all students of a given priority class. We analyze a model that encompasses one-sided and two-sided matching models. We first observe that breaking indifferences the same way at every school is sufficient to produce the set of student optimal stable matchings. Our main theoretical result is that a student-proposing deferred acceptance mechanism that breaks indifferences the same way at every school is not dominated by any other mechanism that is strategy-proof for students. Finally, using data from the recent redesign of the NYC High School match, which places approximately 90,000 students per year, we document that the extent of potential efficiency loss is substantial. Over 6,800 student applicants in the main round of assignment could have improved their assignment in a (non strategy-proof) student optimal mechanism, if the same student preferences would have been revealed.

Abstract not found

Abstract: This essay discusses some things we have learned about markets, in the process of designing marketplaces to fix market failures. To work well, marketplaces have to provide thickness, i.e. they need to attract a large enough proportion of the potential participants in the market; they have to overcome the congestion that thickness can bring, by making it possible to consider enough alternative transactions to arrive at good ones; and they need to make it safe and sufficiently simple to participate in the market, as opposed to transacting outside of the market, or having to engage in costly and risky strategic behavior. I'll draw on recent examples of market design ranging from labor markets for doctors and new economists, to kidney exchange, and school choice in New York City and Boston. 1 This paper was prepared to accompany the Hahn Lecture I delivered at the Royal Economic Society

A great deal of late bidding has been observed on intemet auctions such as eBay, which employ a second price auction with a fixed deadline. Much less late bidding has been observed on intemet auctions such as those run by Amazon, which employ similar auction rules, but use an ending role that automatically extends the auction if necessary after the scheduled close until ten minutes have passed without a bid. This paper reports an experiment that allows us to examine the effect of the different ending rules under controlled conditions, without the other differences between intemet auction houses that prevent unambiguous interpretation of the field data. We find that the difference in auction ending rules is sufficient by itself to produce the differences in late bidding observed in the field data. The experimental data also allow us to examine how individuals bid in relation to their private values, and how this behavior is shaped by the different opportunities for learning provided in the auction conditions.