Essays on Information and Uncertainty

Abstract

This dissertation investigates how limited attention and the lack of precise information shape individual behavior in markets and decision problems. Chapter 1 studies a novel form of the holdup problem that emerges when a consumer invests in learning about product quality set by a producer. Flexible information acquisition, modeled using rational inattention, mitigates this holdup problem by enabling positive quality provision: in equilibrium, the producer randomizes between zero and a higher-than-price quality. A trade-off arises. When attention costs decrease, high quality rises as the consumer learns about quality more precisely. However, by acquiring better information, the consumer trades less often to discourage producer deviations. When attention costs vanish, the holdup becomes inevitable: to deter deviations, the consumer never trades, causing the market to fail. Chapter 2 develops a tractable framework to incorporate ambiguity aversion into rational inattention. Our solution generalizes Matejka and McKay’s (2015) multinomial logit formula by adding an exponential multiplicative state fixed-effect that depends on ambiguity and the attitudes towards it. We provide an axiomatic characterization of this formulation, and also study when our solution follows Luce’s (1959) multinomial logit model. In Chapter 3, coauthored with Giulio Principi and Lorenzo Stanca, we represent preferences that exhibit absolute or relative attitudes towards ambiguity without assuming convexity of preferences. Decreasing absolute ambiguity aversion is characterized by constant superadditive certainty equivalents and admits an act-dependent variational representation (Maccheroni et al., 2006). Decreasing relative ambiguity aversion relates to positive superhomogeneity and admits an act-dependent confidence preference representation (Chateauneuf and Faro, 2009). We apply our characterizations to retrieve a risk sharing result on trade efficiency (Rigotti et al., 2008).