TheoreticalEconomics

Walrasian equilibrium in large, quasilinear markets: Correction econometricsociety.org/publi…

The result of Theorem 6 in Foster and Hart (2023) is correct, but the proof of its more detailed version, Theorem 12 in the appendix, contains an error. We provide a slightly refined statement of Theorem 12, together with a more extensive proof econometricsociety.org/publi…

An expert typically knows much more than a decision maker. Aybas & Callander show that the complexity of the environment matters to cheap talk: an expert can hold power and communicate precisely, in contrast to simple environments econometricsociety.org/publi…

How should society reconcile efficiency with fairness under uncertainty? Focusing on ex-ante fairness, we characterize new aggregation rules within a framework where individual utilities are not comparable a priori econometricsociety.org/publi…

Gaming is costly for everyone. Perez-Richet and Skreta design (non-market) allocation rules immune to gaming, and provide conditions that make honesty optimal for the planner econometricsociety.org/publi…

When asking questions is costly, optimal decision-making can generate apparent biases: it prioritizes more likely choices and makes them even more likely, which we call an endogneous confirmation effect econometricsociety.org/publi…

Why do employers like exploding offers? Should universities discourage their graduating students from reneging on accepted offers? This paper studies when firms make offers with recall and the impact of policies discouraging reneging econometricsociety.org/publi…

Every ton of CO2 pushes us toward a tipping point nobody can locate. In "Acting in the Darkness" (forthcoming, TE), Louise Guillouet and I show optimal policy mixes caution (irreversibility), hope (surviving means the threshold may) econometricsociety.org/publi…

How should Machina's Mom allocate an indivisible good? A variation of Herstein and Milnor's mixture set theorem suggests Mom should use an entropy function econometricsociety.org/publi…

How does a profit-maximizing certifier design and price tests? Two information channels shape the optimal menu: what the test results reveal and what the buyer's choice of test reveals econometricsociety.org/publi…

How do contracts shape assortative matching when workers’ histories are private? This paper shows that screening contracts deliver geometric convergence, while adverse selection without them slows adjustment or stops it altogether econometricsociety.org/publi…

How can directions of preference improvement be characterized without addition, subtraction, multiplication, division, limits, or epsilons and deltas? No algebra, no topology, just orders. This paper proposes such a framework econometricsociety.org/publi…

In our dynamic model, users decide whether to share randomly drawn stories, and spotting false ones takes costly attention. Endogenous attention means early draws affect the outcome; less credible falsehoods can spread more. @tuvaldan @DrewFudenberg econometricsociety.org/publi…

We show, under certain conditions, that if the players start reasoning about more information sets in a dynamic game, then the associated rationalizability procedure will lead to a smaller, or equal, set of outcomes econometricsociety.org/publi…

Is efficiency preserved for stable matchings under two-sided incomplete information? The paper reveals a fundamental asymmetry: it holds for submodular but not supermodular premuneration values, hinging on worst-case misassignment deduction econometricsociety.org/publi…

We study optimal mechanisms for allocation problems without transfers where agents have correlated types. Using graph-theoretic methods, we show that exactly optimal mechanisms are impractically complex. We propose approximately optimal alternatives econometricsociety.org/publi…

We provide a foundation for stochastic choice as the long-run frequency of repeated choices over time econometricsociety.org/publi…

When is ruling out beneficial two-agent swaps enough for full efficiency? The paper shows that for monotonic object-allocation rules, pairwise efficiency and non-wastefulness together imply Pareto efficiency econometricsociety.org/publi…

We prove that any stable matching mechanism suffers from systematic inefficiency of striking magnitude: in large random markets, any stable allocation is Pareto-inefficient with high probability, and almost all students can simultaneously improve econometricsociety.org/publi…

An agent privately observes a Markov chain online and reports it to a designer. We show that, in general, keeping track of the empirical frequency of transition counts in the agent’s reports is insufficient, despite the true state being Markovian econometricsociety.org/publi…