Alvin Roth on Matching Markets

Episode Summary

Executive Summary: Russ Roberts and Alvin Roth discuss how many important markets are not cleared by price alone, but by matching institutions designed to handle thickness, congestion, and safety. They focus on kidney exchange and school choice, showing how market design can save lives, reveal true preferences, and improve allocation without relying on money.

Main Topics: Matching markets vs. commodity markets (Priority: 5/5): Roth explains that in labor, school, organ, and college markets, participants must be chosen by others, so prices alone cannot clear supply and demand. These markets require applications, interviews, priorities, and algorithms to coordinate exchanges. Kidney exchange and organ allocation (Priority: 5/5): The conversation details how kidney exchange evolved from simple paired swaps to larger cycles and long chains started by non-directed donors, transforming transplantation while avoiding illegal kidney sales. School choice and deferred acceptance (Priority: 5/5): Roth describes why early school-choice systems were unsafe and congested, and how deferred acceptance lets families state true preferences without strategic distortion while improving assignment efficiency. Repugnance, ethics, and regulation (Priority: 4/5): They explore why some transactions, especially kidney sales, are widely prohibited despite similar payments being accepted for surgeons and hospitals, and how concerns about coercion and exploitation shape policy. Market design, incentives, and information (Priority: 4/5): The discussion shows that properly designed rules can reveal hidden information, reduce strategic behavior, and even help identify failing schools or underused capacity in public systems. Broader applications and future markets (Priority: 3/5): Roth and Roberts consider other markets being redesigned, including taxi services, restaurant reservations, Wi-Fi sharing, and high-frequency financial trading, illustrating the expanding scope of market design. Hayek, engineering, and humility (Priority: 4/5): The interview ends with a reflection on economists as market designers: interventions should be humble, but thoughtful rules can make competition work more effectively rather than replacing markets.

Key Arguments: Many markets are matching markets, where buyers and sellers cannot simply transact by price; participants must be mutually selected, so institutions beyond prices are essential. Thickness and congestion are central design problems: markets need many participants, but also mechanisms that process many simultaneous choices without overwhelming participants or administrators. Kidney exchange shows that market design can materially improve outcomes, including by creating compatible swaps and long chains that increase the number of transplants without using money. Deferred acceptance in school choice makes it safe for parents to reveal true rankings, preventing schools from exploiting visible first-choice lists and reducing strategic misrepresentation. Repugnance toward kidney markets reflects not just moral discomfort with prices, but worries about coercion, exploitation, and the vulnerability of poor donors; thus regulation matters even when markets are allowed. The absence of prices does not mean the absence of market forces; it means incentives must be embedded in rules, priorities, and algorithms that channel competition productively. Market design can expose hidden demand and quality signals, as when truthful school preferences help reveal which schools families avoid and may justify closure or reform. Even in price-based markets like finance, design matters because timing rules can distort competition toward speed; call markets may restore price competition over millisecond speed races. Economists should act as engineers with humility: designed rules can improve outcomes, but institutions must be built carefully because unintended consequences are always possible.

Data Points: U.S. deceased donor kidney transplants per year: about 11,000 - Roth cites the number of viable kidneys recovered from deceased donors in the United States each year. People on deceased-donor kidney waiting list: 100,000 - He describes the size of the U.S. waiting list for deceased donor kidneys. Living donor kidney transplants per year: about 6,000 - Roth notes the number of additional living donor donations annually in the U.S. Length of kidney chains: 30 or more transplants - Long non-simultaneous chains started by a non-directed donor can extend to 30+ transplants. People in a 30-link kidney chain photo: 60 people - Roth explains that each transplant chain photo would include 30 donors and 30 recipients. New York City high school applicants: almost 90,000 - He cites the scale of NYC ninth-grade admissions each year. Students unmatched under old NYC system: about 30,000 - The old mail-based process left many students unmatched before school started. Students with multiple admissions in old NYC system: about 17,000 - The Department of Education mailed letters to students admitted to multiple high schools. Improvement after NYC algorithm: 30,000 unmatched down to 3,000 - The computerized deferred acceptance system sharply reduced the number of unassigned students. Millisecond speed in financial trading: about 8 milliseconds from New York to Chicago - Roth discusses how speed competition in markets can dominate price competition.

Pivotal Quotes: "in matching markets, you care who you're matched with" — Alvin Roth: Explaining why labor, school, and marriage-like markets differ from commodity markets. "it makes it safe to list what's your true first choice, what's your true second choice, what's your true third choice" — Alvin Roth: Describing the central benefit of deferred acceptance in school choice. "there is in particular all the difference between deliberately creating a system within which competition will work as beneficially as possible and passively accepting institutions as they are" — Alvin Roth / quoting Hayek: Closing reflection on the role of market design and the legitimacy of intervention.

Implications: Well-designed rules can improve markets without replacing them. Expect continued expansion of matching algorithms in public services and private platforms, but with careful attention to ethics, incentives, and unintended effects.