Abstract

This paper experimentally dissects the preferences for honesty into two components: lying aversion and deception aversion. For a separate identification, we consider two reputation-building environments with a two-dimensional belief domain. In the unique equilibrium of one environment, a sender must tell the truth to conceal her misaligned preference with a receiver, while in the other, she must lie to reveal her aligned preference. These distinct settings enable us to isolate and identify individuals' aversion to making statements that are literally untrue (lying aversion) and their aversion to statements intended to manipulate others' beliefs (deception aversion).

Abstract

We study a two-period model of electoral accountability between an incumbent politician and a voter. The incumbent possesses two attributes: competency and experience. Competency determines the accuracy of the incumbent's private signal on policy outcomes. An experienced incumbent is aware of her competency, unlike an inexperienced one. The voter evaluates the incumbent's competency from the incumbent's experience and her policy choice, deciding whether to reelect her. If the incumbent is experienced, she exhibits "anti-pandering" behavior: to signal her competency, the incumbent chooses the risky policy even when its outlook is not promising. In contrast, if the incumbent is inexperienced, she may engage in "pandering" behavior because signaling through the risky policy choice no longer works. The voter's welfare is either nonmonotonic or monotonic in electoral incentives depending on the incumbent's experience because of its impacts on the distortion-selection trade-off. Notably, the value of leadership experience, in terms of the voter's welfare, can be negative.

Summary

We compare the value of reputation-building and commitment power when a sender with imperfect expertise attempts to persuade a receiver. A sender with high expertise observes the state more precisely than a sender with low expertise. Our theory predicts that a sender with high expertise benefits from reputation-building and commitment power, whereas a sender with low expertise only benefits from commitment power. However, our experimental results demonstrate that commitment performs worse than reputation-building in terms of the sender's payoff, contrary to the theoretical prediction. We discuss the main mechanisms underlying these findings.

Summary

One of the recent concerns in the science of science is the declining trend in disruptive innovations. Building upon Garfagnini and Strulovici (2016), we develop a model that explores the relationship between research evaluation systems and researchers' experimentation choices. Even though the evaluation system highlighting the citation index lowers the boldness of innovation at the frontier, such an evaluation system opens the possibility of incremental research that fills the gap in the current knowledge.

Summary

We study the dynamic competition between two media platforms, each deciding the news structure and vying to capture consumers' attention as long as possible. In each period, consumers decide which media platform to use and whether to take an irreversible political action. Media platforms face a trade-off between winning the competition and capturing attention: supplying news with a higher arrival rate wins the attention of more consumers but encourages them to take the action earlier.