Topics in Strategic Communication

National Science Foundation, SES-0965577
Principal Investigator: Navin Kartik


The strategic communication of private information is of fundamental interest in economic theory and various applications, including in political economy, bargaining, finance, and law and economics. This research consists of three projects in strategic communication, each of which employs the methodology of game theory.

The first project develops new models of communication that span the literature's standard analysis of "cheap talk" and "verifiable disclosure". Much of what economists know about strategic information transmission is based on models of these polar settings, where evidentiary private information is either perfectly manipulable or entirely non-manipulable, respectively. However, reality often lies in between: manipulation is possible, but may incur a cost. The research studies the structure of strategic communication in such cases, developing theoretical frameworks and results.

The second project studies why biased experts may be appointed to advise a decision-maker, often by the decision-maker herself. Most extant analysis typically takes as given that there is a conflict of interest between experts and decision-makers, but does not address why such a conflict exists. The project develops a model that combines strategic communication with endogenous information acquisition, showing that soliciting advice from an expert with a difference of opinion can be an optimal arrangement when the expert must be motivated to both acquire information and then reveal it.

The third project tackles information transmission in politics. There is a long-standing view that even if voters are uninformed about economic or political variables, electoral institutions encourage well-informed candidates for office to reveal their information to the public. The current research examines the validity of this notion through formal models of electoral competition, where privately informed candidates propose policy with the objective of winning office. The question is whether competition between candidates results in voters being able to extract the socially valuable information possessed by the candidates.

Broader Impacts

The first project makes a novel contribution to understanding the nature of strategic information transmission as a function of the technology of communication. As a consequence of the research, both cheap talk and verifiable disclosure can be viewed conceptually as extremes of a unified model, and likewise for their predictions. Moreover, the results may be useful for applied work in not only economics, but also fields such as political science, where partially manipulable information is strategically communicated. Project 2 provides a rationale for the existence of conflicts of interest and divergence of opinion between experts and decision-makers. It may be a first step towards a theory of rational diversity in organizations. The third project informs our understanding of the incentives for politicians and parties to disseminate socially valuable information during elections. It may have implications for which electoral institutions promote information revelation.


Working Papers and Publications

Che, Yeon-Koo & Dessein, Wouter & Kartik, Navin, 2010. "Pandering to Persuade," CEPR Discussion Papers 7970, C.E.P.R. Discussion Papers.
Full text: