Economics

Agreeing to be fooled: Optimal ignorance about information sources (PDF)

Last updated: 10 January 2023

Should a decision maker learn whether a piece of information is from a reliable source? I show that the decision maker can obtain more useful information by strategically avoiding learning the reliability of the information source. Specifically, I study a persuasion game in which the sender uses a public signal about a payoff-relevant state to persuade the receiver to take an action. With some probability, the sender is unreliable and covertly manipulates the signal realisation. The receiver can investigate the sender's reliability. I show that it is optimal for the receiver to commit to investigations that do not fully reveal the reliability, and I characterise the receiver-optimal investigation strategy and payoff. I also study the extent to which the receiver can implement commitment outcomes by delegating investigations to a third party. In particular, I find that the receiver benefits from delegating investigations to the sender's adversary who also cares about the receiver's payoff. Moreover, I show that the receiver can use such an adversary to implement the optimal commitment outcome whenever the sender is sufficiently unreliable. My results shed light on the efficacy of cross-examination of witnesses in courts and the benefit of ad hominem arguments.

Knowing the Informed Player’s Payoffs and Simple Play in Repeated Games (SSRN), R&R at JET

with Elliot Lipnowski and Doron Ravid

We revisit the classic model of two-player repeated games with undiscounted utility, observable actions, and one-sided incomplete information, and further assume the informed player has state-independent preferences. We show the informed player can attain a payoff in equilibrium if and only if she can attain it in the simple class of equilibria first studied by Aumann, Maschler, and Stearns (1968), in which information is only revealed in the game’s initial stages. This sufficiency result does not extend to the uninformed player’s equilibrium payoff set.

Audience Design (PDF, Slides)

(Previously circulated with the title: The Power of Semi-Public Communication)

I study how diversity of opinions in an audience can increase a speaker's perceived sincerity. Optimal grouping, or partition, of the audience trades off the benefit from increased perceived sincerity through greater diversity in each audience against the cost from the inability to tailor communication to individuals in an audience. Public and private communication are restrictions on the partitions the speaker can adopt, and the speaker prefers semi-public communication whenever there is a possibility to benefit from differently diverse groups of audiences. I show that it is optimal for the sender to separate individuals that need to be persuaded from those that do not, and derive further properties of optimal partitions when receivers are "single minded" and when receivers' preferences lie on a spectrum.

Selection in Information Design (work in progress)

with Andy Choi

In many applications of Bayesian persuasion, the state space represents agent types so that the choice of experiment by the designer can be expected to have a selection effect. For example, the choice of grading policy for a class is likely to affect the type of students who choose to take the class. Similarly, a seller may not participate on a platform because the platform has a tough rating system. We study how accounting for selection effects in Bayesian persuasion problems affects the optimal payoffs and communication strategy by explicitly modelling the agent’s participation decision.

Optimal contracting with a Luce agent (PDF)

MPhil thesis

I study how introducing a boundedly rational agent to an otherwise standard optimal contracting problem alters the characteristics of the optimal contract. Specifically, I assume that the agent’s response function follows the Luce model such that he is more likely to undertake an action that gives him a higher (expected) utility. When there is moral hazard, optimal contracts exhibit the usual monotone contract property "in reverse"; i.e., the principal does not find it optimal to pay the agent always less when she observes a better outcome. This finding is similar to that from the standard model with a fully rational agent.

Money Illusion and its Implication on Unemployment (Link) Undergraduate Economic Review (2011)

Undergraduate dissertation

The paper discusses the implication of money illusion on persistent unemployment. A particular form of money illusion is assumed and this is modeled into the efficiency wage theory while separating the analysis into nominal and real frames. The model shows that the level of unemployment in the nominal and the real frame are likely to be different and that the government has an incentive to provide a signaling mechanism to the workers to reduce unemployment levels. Additionally, the government is shown to have an incentive to announce unemployment rates.

Takuma Habu

Job Market Paper

Other works in progress