Accepted Paper 


Latest version: here

We study doctor–hospital matching markets with contracts, in which hospitals can offer a range of contract terms to doctors. We show that introducing new terms can reduce doctors’ welfare, whereas withdrawing terms may generate a Pareto improvement. We establish the preference domains under which welfare is preserved and show that only agent-lexicographic preferences for all agents ensure that no doctor is worse off when additional terms are introduced. Since this condition is rarely satisfied in practice, our results imply that most real-world markets are vulnerable to welfare losses when the set of contract terms expands.


Working Papers


Job Market Paper, latest version: here (OA)

Subsidies and taxes are widely used in labor markets to influence employment outcomes. This paper uses a Kelso-Crawford framework to assess how transfers affect the welfare of minority workers. We show that affirmative action policies, though well-intentioned, can unintentionally harm this group. We establish that only uniform transfers—where all workers in a group receive the same subsidy or tax across all firms—guarantee no welfare loss. Building on this insight, we explore how transfers can be designed to achieve representation goals for minority workers without reducing their welfare. 


Latest version: here

This paper examines a preference revelation game that utilizes the student-proposing deferred acceptance mechanism (DA) in the context of school choice. Our focus is on assignments that Pareto-dominate the student-optimal stable assignment, specifically investigating the structures of Nash equilibria that implement these assignments with DA (i.e., Nash equilibria that lead to these outcomes with DA). While some assignments cannot be implemented with DA, we identify a strategy profile that determines whether a given assignment is implementable when each school has a capacity of one. However, this profile cannot determine implementability when at least one school has more than one available seat. As a result, we focus on specific assignments that Pareto-dominate the student-optimal stable assignment and establish sufficient conditions on strategy profiles that can implement these assignments with DA. 


Latest version: here

This paper explores the impact of application fees on student strategies within the Deferred Acceptance (DA) mechanism. We show that application fees reduce the set of Nash equilibria under DA. While they can lead to Pareto-efficient assignments, application fees may also prevent Nash equilibria that result in assignments Pareto-dominating the student-optimal stable assignment. This occurs when application fees are positive for all students at a given school. 


Latest version: here

Affirmative action policies, by establishing representation thresholds for protected groups, seek to balance fairness and equity in various assignment problems. Fairness is maintained by prioritizing individuals based on merit scores, while equity is ensured through guaranteed group representation. We focus on overlapping reserves, where individuals can belong to multiple groups, and introduce the Maximal Score and Minimum Guarantee (MSMG) choice rule, which upholds representation requirements while preserving fairness. We define the score of an assignment as the sum of the merit scores of the selected individuals. We demonstrate that the assignment produced by the MSMG choice rule achieves the highest possible score among all fair assignments that satisfy the given representation thresholds. 


Joint with Resul Zoroglu, latest version: here

This paper examines rebate rules in the context of public goods provision. These rules aim to redistribute the surplus when total contributions exceed the cost of the project. Using an axiomatic approach, we establish impossibility results that highlight the inherent tensions between fairness, participation incentives, and contribution incentives. To address these limitations, we propose the Proportional Rebate with Threshold rule, which strikes a balance between these competing objectives.


Joint with Alice Mazzacurati, Best Junior Paper Award at Match-UP 2026, draft available upon request 

This paper studies the labor market within a Crawford–Knoer framework with unemployment benefits. We first characterize the worker-optimal stable allocation and show how a marginal increase in the benefit of a single unemployed worker propagates through the market. To capture these interdependencies, we construct a Directed Acyclic Graph that depicts how workers’ utilities interact.  We show that increasing the unemployment benefit of an unemployed worker raises the utility of all employed workers directly or indirectly influenced by that worker by the same amount.  We then extend the analysis to all stable allocations. In this more general analysis, we examine policies that target benefits to employed workers and design a policy that raises welfare while preserving efficiency at no cost.


Latest version: here

Regulations imposing hiring constraints are widely used to promote diversity in labor markets. Motivated by these interventions, this paper studies their effects on the welfare of minority workers in stable allocations. Building on a Crawford–Knoer framework, we show that such regulations can unintentionally reduce minority welfare. Using firm-optimal and worker-optimal stable allocations as benchmarks, we characterize the class of policies that, across all markets, never reduce minority welfare under either allocation. This class consists only of the null policy and policies that exclude non-minority workers from the market without imposing reserves. We then introduce policy rules, which allow policies to depend on the assignment observed prior to intervention. Restricting attention to environments with two worker groups, we characterize the set of policy rules that never reduce the welfare of minority workers.