A Model of Expertise

Abstract

We study a model in which perfectly informed experts offer advice to a decision maker whose actions affect the welfare of all. Experts are biased and thus may wish to pull the decision maker in different directions and to different degrees. When the decision maker consults only a single expert, the expert withholds substantial information from the decision maker. We ask whether this situation is improved by having the decision maker sequentially consult two experts. We first show that there is no perfect Bayesian equilibrium in which full revelation occurs. When both experts are biased in the same direction, it is never beneficial to consult both. In contrast, when experts are biased in opposite directions, it is always beneficial to consult both. Indeed, in this case full revelation may be induced in an extended debate by introducing the possibility of rebuttal. Downloaded from1. Our results in the case of like biases concern monotonic equilibria where the action taken is a monotonic function of the state. While nonmonotonic equilibria exist, we provide sufficient conditions for all equilibria to be monotonic.2. Our results in the case of opposing biases hold generally for all equilibria, not just monotonic equilibria. 749 A MODEL OF EXPERTISE Downloaded from 3. The case where both b 1 , b 2 Ͻ 0 is qualitatively no different from the case where both b 1 , b 2 Ͼ 0. 752 QUARTERLY JOURNAL OF ECONOMICS at California State University, Fresno on July 24, 2015 http://qje.oxfordjournals.org/ Downloaded from 7. Although our analysis in the like bias case is confined to monotonic equilibria, we know of no instance where admitting a nonmonotonic equilibrium reverses the welfare comparisons. 759 A MODEL OF EXPERTISE

Learning under supervision: an experimental study

“…Our research also adds to past work on advice and its effects (Krishna and Morgan 2001;Ottaviani and Sorensen 2003). Both Krishna and Morgan (2001) and Ottaviani and Sorensen (2003) consider the issue of expert advice.…”

“…Both Krishna and Morgan (2001) and Ottaviani and Sorensen (2003) consider the issue of expert advice. Krishna and Morgan consider a situation where experts, who are perfectly informed, offer advice to decision makers.…”

Communication in dynastic repeated games: ?Whitewashes? and ?coverups?

“…Despite some similarities, the present model examines a very different type of communication than in typical sender-receiver, cheap talk models such as Crawford and Sobel (1982) and, more recently Krishna and Morgan (2001). The latter is representative of a more recent variety which, as in our model, features multiple senders of information.…”

Strategic political commentary

“…Djankov et al (2003) and Besley and Prat (2006) model and measure freedom of the press. 6 See Gilligan and Krehbiel (1987), Krishna and Morgan (2001), and Battaglini (2002).…”