This page collects my work related to financial market design. My best known conceptual ideas in this area are frequent batch auctions and sniping. Frequent batch auctions is a market design alternative to continuous trading on financial exchanges. Mathematically, instead of trading in continuous time as a serial process — which means that being a nanosecond faster is treated as economically meaningful by the market design — trade is in discrete time as a batch process, using auctions. Sniping refers to arbitrage that is primarily a contest in speed, also known as picking off stale quotes. Mathematically, sniping refers to arbitrage rents that arise from symmetric public information — information that is broadly disseminated to the whole market at the same time — as distinct from asymmetric private information.
This paper collects my work related to the Covid-19 crisis. My research argues that the Covid-19 crisis demanded a novel play in the epi playbook: vaccinate the world as fast as possible, and in the interim, aim to maximize societal and economic well-being subject to the constraint that the virus does not spread exponentially. In math, “Maximize Utility subject to R<1”.
This page collects my work on the market design problem of course allocation, also known as multi-unit or combinatorial assignment. My best known conceptual ideas in this area are the course allocation mechanism “Approximate Competitive Equilibrium from Equal Incomes” (A-CEEI) and the associated approximate fairness and incentives criteria: Envy Bounded by a Single Good (EF-1), Maximin Share (MMS), and Strategy-proof in the Large (SP-L).
This page collects my work on market design theory. Readers may also wish to look at the page specifically on the course allocation problem, which has significant overlap.
This page collects my work on cryptocurrencies and blockchains. My main conceptual idea in this area is in equations (1)-(3) of the June 2018 “Economic Limits” paper. These equations show that the anonymous, decentralized trust invented by Satoshi Nakamoto is very expensive: the ongoing, “flow” cost of maintaining the trust has to be large relative to the one-off, “stock” value of attacking the system.
This page collects my work related to patents and innovation.
This page collects my work on ticket markets. This market has been of interest to me for a long time, dating back to queues at Ticketmaster kiosks for concert tickets as a teenager, and going with my dad to see the Mets clinch the 1986 pennant with scalped tickets.