Publications 

"Leader-Follower Dynamics in Shareholder Activism" (with Gonzalo Cisternas, Aaron Kolb, and S. "Vish" Viswanathan) Journal of Finance (2026) (Online Appendix)

"Exit Dilemma: The Role of Private Learning on Firm Survival" (with Chiara Margaria) AEJ:Microeconomics (2024) (Online Appendix)

"Collective Progress: Dynamics of Exit Waves" (with Can Urgun and Leeat Yariv) Journal of Political Economy (2023) (Online Appendix) 

"Renegotiation and Dynamic Inconsistency: Contracting with Non-Exponential Discounting" (with Felix Feng and Can Urgun) Journal of Economic Theory (2023) 

"Uncertainty-driven Cooperation" (with Ilwoo (Iru) Hwang and Ayca Kaya) Theoretical Economics (2020) (Best TE Paper Prize 2022)

"Implementing Equal Division with an Ultimatum Threat" (with Emin Karagozoglu) Theory and Decision (2014)

Working Papers

"Flexible Testing under Strategic Manipulation" (with Francesc Dilmé) (2026)

 We study dynamic monitoring when a principal chooses what evidence to seek, and how intensively to seek it, about an agent whose persistent type is private. The principal wants to hire the good type and fire the bad type, whereas both types want to be hired. The good type always exerts high effort, while the bad type can imitate him at a flow cost. Incentive compatibility turns experiment choice into an endogenous information budget that limits how intensively the principal can test while keeping the bad type from imitating the good type. We show that monitoring is one-sided away from stationary beliefs: the principal seeks either pass news leading to hiring or fail news leading to firing, but not both. Depending on probation and effort costs, equilibrium features no probation, pass-only probation, fail-only probation, or a fail–pass sequence. Probation achieves accurate classification with delay, whereas immediate stopping is faster but error-prone.

"Midterm Review" (with Yonggyun Kim, Fei Li and Curtis Taylor) (2026)

We study why organizations conduct interim performance reviews when monetary rewards are limited. Review creates incentive capacity by allowing future work and career opportunities to serve as rewards for past performance. Optimal review policies map a continuum of performance outcomes into a simple incentive ladder: termination, tough or easy continuation, and, for exceptional performance, an early maximal reward with no further work. Review can even sustain high effort when terminal compensation alone cannot. Its timing balances two forces: waiting improves the information revealed by performance, but leaves less future work available to motivate the agent.

"AI and the Market for Signals" (with Emre Ozdenoren) (2026)

Early-career work signals a worker's ability and builds long-run skills. We introduce an AI industry selling tokens that workers convert into output indistinguishable from what effort produces, and compare three regimes: no token trade, purchases only, and open trade with side sales. With purchases only, the market still learns types, and the distortion is ordered by type: lower buyers work less than before AI, some falling below the full-information level, while higher buyers, and the top types who buy nothing, can work harder than before. With sales and abundant AI, the token price equals the output value and every type weakly under-invests. Token trade can raise welfare while workers in aggregate lose: in education, where output has no value, active trade always hurts them, but with valuable output and cheap tokens they gain alongside welfare. Token taxes and watermarking can help or hurt, depending on output value and market structure.

"Dynamic Signaling in Wald Options" (with Chiara Margaria) (2025) Revise & Resubmit Review of Economic Studies 

A sender engages in costly signaling to influence a decision maker, who observes a biased noisy signal and decides when to irreversibly take an action to match the binary state. We characterize Markov equilibria in terms of a two-dimensional boundary value problem for fixed discount rates and analyze equilibrium behavior as players become arbitrary patient. The leading example is a dynamic limit pricing game between an incumbent and a potential entrant who uses price to infer the industry conditions. A sufficiently patient incumbent always produces at capacity, and consumers can be hurt because the potential entrant strategically delays its entry.

"Efficiency in Repeated Partnerships" (2024) (Online Appendix) Revise & Resubmit Review of Economic Studies 

Two partners contribute to a common project over time. The value of the project is determined by the aggregate effort of the partners and by a common productivity parameter that each partner is privately informed about. At each instant, the two partners observe a noisy public signal of total effort. An equilibrium of this game is Markov if effort choices of agents depend only on the beliefs about the value of the project and on calendar time. I characterize the linear Markov equilibrium as the solution to a nonlinear boundary value problem. Equilibrium is unique if agents are symmetric. The equilibrium features a mutual encouragement effect, as agents exaggerate their effort in order to signal their private information, which counteracts free-riding incentives. Indeed, if the project lasts sufficiently long, the diffused information structure approximates the first-best in terms of welfare. If, instead of distributed private information, one agent has all the information about the productivity parameter, the excessive signalling effect is accentuated. As a result, the centralized information structure can yield output levels above the first best.

"Optimal Project Management" (with Alessandro Bonatti and Juuso Toikka) (2026)

We analyze optimal dynamic contracts for long-term projects where a risk-averse agent exerts hidden effort to increase the drift of a Brownian process toward a completion threshold. Relative to the first best, the optimal contract slows progress, makes success less likely, induces earlier termination, and narrows project scope. Allowing for costly project resets generates further distortions: resets may be excessive or insufficient depending on the residual effort required after a reset. With endogenous risk-taking, optimal incentives introduce penalties for failure and reduce rewards for effort above a critical progress level, resulting in discontinuous and non-monotonic effort in progress.

"Cutting Back to Move Forward" (with Chiara Margaria, Tatiana Mayskaya and Arina Nikandrova) (2026)

We study a dynamic moral hazard model without monetary transfers, in which a principal can gradually and costlessly transfer knowledge to raise an agent's productivity. Although transferring knowledge is efficient, the principal may deliberately limit it to deter the agent's procrastination, and this inefficiency persists even with infinite patience. Small differences in effort cost or learning rate can generate starkly different outcomes: one agent is made minimally productive, while the other receives maximal training. Commitment has no value for the principal, who can achieve her commitment-optimal payoff via a three-phase training with a mid-career dip, consistent with empirical evidence.

"Competitive Search and Moral Hazard in Dynamic Project Management" (with Mayur Choudhary and Emre Ozdenoren) (2025)

This paper introduces moral hazard in dynamic project management into a competitive search framework. In this model, agents’ unobservable actions influence the likelihood of a breakthrough. Search frictions determine agents’ outside options endogenously, affecting contract terms like tenure length and bonuses. Equilibrium tenure exceeds the benchmark with exogenously set outside options. Comparative statics reveal how search costs, degree of moral hazard, the benefits and costs of the project influence contract tenure and bonuses, offering insights into trends like declining CEO tenure and rising bonuses. We also allow the agents to have heterogeneous private valuations for success and show that separating equilibrium does not exist. In the pooling equilibrium, firms offer a menu of contracts in a single market.

"Mediation in Dynamic Contracting" (with Dino Gerardi, Lucas Maestri, and Ignacio Monzón) (2024)

We study communication in dynamic contracting with limited commitment. A firm and a worker interact over time. The worker is privately informed about his productivity. In each period, the firm offers a menu of short-term contracts. The interaction between the firm and the worker continues provided that the worker accepts a contract and finishes otherwise. We let the parties communicate through a trustworthy mediator. We allow for a class of natural and simple communication protocols (good news, bad news or both). We characterize the optimal communication protocol for the firm. We find that it is optimal to only use bad news and to reveal information gradually .

Work in Progress

"Regulating Pre-Election Spending: A Dynamic Analysis" (with Leandro Gorno, Helios Herrera and Lucas Maestri) (2026)

"When Financing Reveals Information" (with Luigi Paciello) (2026)