Research

Here are my topics of research:

Asymptotic Full Revelation in Cheap Talk with Many Senders [PDF]

Strategic information transmission, commonly referred to as "cheap talk", was first introduced by Crawford and Sobel (1982) who showed that only a limited amount of information could be transmitted from an informed sender to an uninformed receiver. In this paper, I show that it is possible to achieve high precision in information transmission with multiple senders by constructing an equilibrium which converges to full revelation exponentially fast with the number of senders. This equilibrium is robust to small perturbations. It can also be constructed for multidimensional state spaces and it achieves arbitrarily high precision even if each sender is significantly biased and is allowed to send only two types of messages.

Profit Sharing and Incentives with Emre Ozdenoren [PDF], International Journal of Industrial Organization, Volume 83, July 2022, 102857

We model a firm as a team production process subject to moral hazard and derive the optimal profit sharing scheme between productive workers and outside investors together with incentive contracts based on noisy performance signals. More productive agents with noisier performance signals are more likely to receive shares which can explain why managers are motivated by shares, and law or consulting firms form partnerships. A firm that grows by opening branches is held almost entirely by outside investors when its output noise grows faster than the number of branches. Otherwise, insiders hold substantial amount of a large firm's shares.

Rank Based Contracts with Emre Ozdenoren

We present a moral hazard problem with many principal-agent pairs where each agent takes an unobservable costly action which affects the payoff of his principal. Each principal receives a rating of her agent which is a discrete signal of the agent's effort. The ratings can also depend on the other agents' efforts. A notable special case of ratings is a rank-order tournament.  In equilibrium, each principal offers her agent an optimal contract which determines a transfer from the principal to the agent given the agent’s rating and the efforts of other agents. We show that when the principals and agents are risk neutral, rank-order based contracts can be sufficient to achieve efficient levels of effort for all principals. We provide necessary and sufficient conditions for the implementability of different levels of efforts and the existence of symmetric equilibria. We also provide an example where rank-order based contracts generate externalities that are welfare reducing and a planner can improve on the equilibrium contracts.

Two-Sided Cheap Talk 

I study the situation of two-sided cheap talk when two agents both have unique information and can affect the final action. Therefore, both agents act as both senders and receivers in the classical cheap talk game. I find that typically in this setting the information transmission between the agents is typically impossible.

Pre-Election Opinion Polls with Miguel Espinosa

Do pre-election opinion polls affect the outcome of the election? We construct a model where the release of an opinion poll improves the information about the voters' preferences over parties, and therefore changes their individual votes, as well as the outcome of the election. This paper studies how pollsters are able to influence the public opinion. We analyze how either the reputation or the political bias of a pollster can affect the opinion poll and discuss how the results of the opinion poll published by a media outlet can be altered by its political slant.