Two essays on the economics of strategic information transmission

Abstract

In Chapter 1 we compare the relative efficiency between two modes of certification. Third party, expert certification may help mitigate the asymmetric information problem between a seller and abuyer. The certication may be initiated by the seller or by the buyer and hence the terms of "sellercertification" and "buyer certification" for the two modes respectively. We extend the work by Stahl and Strausz (2017) (SS in short) in two aspects: seller cost types are continuously distributed (versus binary in SS) and the certication technology is imperfect (versus perfect technology inSS). We find that the main result in SS that, given the assumption of monopoly certifier, seller certification being more efficient continues to hold true in our setting. However, we also find that when the certification fee approaches zero (because of competition between certiers or because of price regulation, for example), buyer certification will be more efficient. Both low certification feeand noisy certification technology are needed for our new result to hold.In Chapter 2, I study two models of competition for persuasion. In each of them, two experts with different action preference compete to reveal verifiable information to an uninformed decision maker (DM) who has private preference for state-contingent action. The DM then chooses from available alternatives as specfied by the experts. The two models differ in the cost associated with concealment: in the first, experts bear an exogenous concealment cost if they censor observedinformation; in the second, a reputation cost of concealment arises endogenously in the agenda-pushing experts who want to appear honest in the presence of an objective counterpart. There arethree main findings: (1) the two models lead to different equilibrium outcomes and information is more effectively revealed under exogenous concealment cost; (2) increased agenda divergence canlead to more disclosure in both environments but more effectively so under exogenous concealment cost; and (3) increased reputational concern in one strategic type may backre and cause him todisclose less, but unambiguously leads to more disclosure by his strategic counterpart, while increased exogenous concealment cost in one expert unambiguously leads to more disclosure by both.