Abstract: We study adverse selection markets where consumers can choose to learn how much they value a product. Information is acquired after observing prices, so it is endogenous. This presents a trade-off: information increases the quality of consumers’ choices but worsens selection. We characterize how this trade-off translates into distortions of consumers’ demand and producers’ cost. Then, we show that information decisions produce a negative externality, may decrease welfare, and may lead to new forms of market breakdown. Moreover, efficiency is typically non-monotone in information costs. Two implications are that (1) standard measures underestimate the welfare costs of adverse selection; and (2) information policies can help correct its inefficiencies. Finally, we propose an empirical test to detect endogenous information in the data, and develop a framework for counterfactual policy analysis.

(Online Appendix)

Abstract: A monopolist offers a menu of quality-differentiated products. After observing the offer, a consumer can costly and flexibly learn which product is right for them. In the optimal menu, all types typically receive lower-than-efficient quality and distortions are, on average, more intense than under standard screening, even when no information is acquired. Profits are non-monotonic in the level of information costs, and the consumer may be better off when such costs are low than when information is free. We illustrate how an econometrician who ignores information acquisition might underestimate the level of inefficiency in the market.

Abstract: We study asymptotic learning when the decision maker is ambiguous about the precision of her information sources. She aims to estimate a state and evaluates outcomes according to the worst-case scenario. Under prior-by-prior updating, we characterize the set of asymptotic posteriors the decision maker entertains, which consists of a continuum of degenerate distributions over an interval. Moreover, her asymptotic estimate of the state is generically incorrect. We show that even a small amount of ambiguity may lead to large estimation errors and illustrate how an econometrician who learns from observing others’ actions may over- or underreact to information.

(Previous Title: Information Aggregation under Ambiguity)

Abstract:  A large empirical literature on the timing of binary choices documents that quicker decisions are often more accurate than slower ones. This evidence suggests individuals decrease the standards with which they choose over time, at odds with the classic sequential sampling model in which standards are time-independent. We show that incorporating risk aversion can account for time-dependent standards, and we find sufficient conditions for standards to be decreasing for a family of utility functions. Our approach sidetracks some of the difficulties in solving non-stationary optimal stopping problems and allows us to partially characterize the optimal boundaries.

Abstract:  We study the interaction between insurance and financial markets. Individuals who differ only in risk have access to insurance contracts offered by a monopolist and can also save through a competitive market. We show that an equilibrium always exists in that economy and identify an externality imposed on the insurer’s decision by the endogeneity of prices in the financial market. We argue that, because of that externality and in contrast to the case of pure contract theory, equilibrium always exhibits under-insurance even for the riskiest agents in the economy and may even exhibit pooling. Importantly, the externality does not disrupt the single crossing property of the economy.