Title: Information Spillover in Markets with Heterogeneous Traders
Abstract: This paper studies the welfare impact of information spillover in divisible-good markets with heterogeneous traders and interdependent values. In a setting in which two groups of traders trade two distinct but correlated assets, one within each group, the information content in the price of one asset spillovers to the other market. Some “informed” traders who submit demand schedules may condition their demands on the prices of both assets, while others not. We prove the existence of a linear equilibrium and examine how information spillover affects trading, information efficiency, and welfare, as the fraction of the informed traders varies. In the two symmetric benchmarks, full information spillover (all traders are informed) dominates no information spillover (all traders are uninformed) in terms of trading volume and welfare. However, in markets with heterogeneous traders, information spillover can hurt overall welfare, while still improving information efficiency and liquidity; we characterize the non-monotone impact of information spillover on aggregate welfare in large finite markets. Furthermore, information spillover can account for the empirical evidence of excessive price co-movement and volatility transmission in financial markets.
Title: Price Heterogeneity as a source of Heterogenous Demand
Abstract: We explore heterogenous prices as a source of heterogenous or stochastic demand. Heterogenous prices could arise either because there is actual price variation among consumers or because consumers (mis)perceive prices differently. Our main result says the following: if heterogenous prices have a distribution among consumers that is (in a sense) stable across observations, then a model where consumers have a common utility function but face heterogenous prices has precisely the same implications as a heterogenous preference/random utility model (with no price heterogeneity). Keywords. random utility, stochastic demand, augmented utility, misperceived prices, reference prices, equivalence scales, Afriat’s Theorem.
Title: Predicting Choices from Information Costs
Abstract: An agent acquires a costly flexible signal before making a decision. We explore the degree to which knowledge of the agent's information costs help predict her behavior. We establish an impossibility result: learning costs alone generate no testable restrictions on choice without also imposing constraints on actions' state-dependent utilities. By contrast, for most utility functions, knowing both the utility and information costs enables a unique behavioral prediction. When the utility function is known to belong to a given set, we provide an exact characterization of rationalizable behavior. Finally, we show that for smooth costs, most choices from a menu uniquely pin down the agent's decisions in all submenus.
Title: Deviation-Based Learning
Abstract: We propose deviation-based learning, a new approach to training recommender systems. In the beginning, the recommender and rational users have different pieces of knowledge, and the recommender needs to learn the users' knowledge to make better recommendations. The recommender learns users' knowledge by observing whether each user followed or deviated from her recommendations. We show that learning frequently stalls if the recommender always recommends a choice: users tend to follow the recommendation blindly, and their choices do not reflect their knowledge. Social welfare and the learning rate are improved drastically if the recommender abstains from recommending a choice when she predicts that multiple arms will produce a similar payoff.
Title: Background Risk and Small-Stakes Risk Aversion
Abstract: We show that under plausible levels of background risk, no theory of choice under risk can simultaneously satisfy the following three economic postulates: (i) Decision makers are risk-averse over small gambles, (ii) their preferences respect stochastic dominance, and (iii) they account for background risk. This impossibility result applies to expected utility theory, prospect theory, rank dependent utility and many other models.
Title: Background Risk and Small-Stakes Risk Aversion
Abstract: We show that under plausible levels of background risk, no theory of choice under risk can simultaneously satisfy the following three economic postulates: (i) Decision makers are risk-averse over small gambles, (ii) their preferences respect stochastic dominance, and (iii) they account for background risk. This impossibility result applies to expected utility theory, prospect theory, rank dependent utility and many other models.
Title: Disclosure of Bank-Specific Information and the Stability of Financial Systems
Abstract: We find that disclosing bank-specific information reallocates systemic risk, but whether it mitigates systemic bank runs depends on the information disclosed. Disclosure reveals banks’ resilience to adverse shocks, and shifts systemic risk from weak to strong banks. Yet, only disclosure of banks’ exposure to systemic risk can mitigate systemic bank runs because it shifts systemic risk from more vulnerable banks to those less vulnerable. Optimal disclosure thus maximally differentiates such exposure, provided that banks experience runs simultaneously, if inevitable. Disclosure of banks’ idiosyncratic factors does not differentiate such exposure, rendering the resulting reallocation of systemic risk ineffective in mitigating systemic runs. In the context of disclosing stress-test results, when the quality of the banking system deteriorates, the regulator may have to face a sudden run on a huge mass of banks rather than gradually abandoning weak banks.
Title: Magic Mirror on the Wall, Who Is the Smartest One of All?
Abstract: In the canonical model of bounded rationality each player best-responds to their belief that other players reason to some finite level. We propose a novel behavior that reflects the player’s belief that while other players may be rational, the player cannot model and hence predict the behavior of others. This encompasses a situation where a player believes that their opponent can reason to a higher level than they do. We propose an identification strategy for such behavior, and evaluate it experimentally
Title: Empirical Welfare Economics
Abstract: Given demand data for a group of agents, we seek to make counter-factual welfare statements. Our main result considers whether there are convex preferences for which some candidate allocation is Pareto optimal. We show that this candidate allocation is possibly efficient if and only if it is efficient for the incomplete relation derived from the revealed preference relations and convexity. Similar ideas are used to address related questions: when the Kaldor criterion may be used to make welfare comparisons, what prices can be Walrasian equilibrium prices, and the possibility of a representative consumer when the income distribution is endogenous.
Title: Learning by Consuming: Sequential Screening with Endogenous Information Provision
Abstract: We study the revenue-maximizing mechanism, in which a seller sells one unit of a divisible good to a buyer in two stages. In stage one, the buyer can rely on his private, rough valuation of the good to determine his first-stage consumption level. Consuming more leads to a more precise private valuation estimate of the good in stage two, after which he determines his second-stage consumption level. Since the single-crossing condition fails, the monotonicity in allocation plus the envelope condition are not sufficient for global incentive compatibility. The optimum is a menu of contracts, consisting of a first-stage price-quantity pair and a second-stage per-unit price for the remaining quantity. A larger first-stage quantity is paired with a higher first-stage price but a lower second-stage per-unit price. In equilibrium, a higher first-stage valuation buyer pays more to have higher first-stage consumption and enjoys a lower second-stage price.
Title: An Evolutionary Perspective on Updating Risk and Ambiguity Preferences
Abstract: Using an approach predicated on evolution and adaptation, we provide foundations for a model of choice under uncertainty based on adaptive preferences. We argue that our approach can be applied in most contexts involving ambiguity, and we show that adaptive preferences nest variants of several established models of ambiguity and risk preferences as special cases. Our model provides a tight connection between ambiguity attitudes and violations of expected utility, and it generates novel predictions about the role of random choice in hedging against ambiguity. We also find that updating of adaptive preferences in response to new information respects dynamic consistency even at the cost of violating consequentialism, addressing a prominent tension in the ambiguity and non-expected-utility literature.
Title: A Natural Adaptive Process for Collective Decision-Making
Abstract: Consider an urn filled with balls, each labeled with one of several possible collective decisions. Now, draw two balls from the urn, let a random voter pick her more preferred as the collective decision, relabel the losing ball with the collective decision, put both balls back into the urn, and repeat. In order to prevent the permanent disappearance of some types of balls, once in a while, a randomly drawn ball is relabeled with a random collective decision. We prove that the empirical distribution of collective decisions converges towards a maximal lottery, a celebrated probabilistic voting rule proposed by Peter C. Fishburn (Rev. Econ. Stud., 51(4), 1984). In fact, the probability that the collective decision in round n is made according to a maximal lottery increases exponentially in n. The proposed procedure is more flexible than traditional voting rules and bears strong similarities to natural processes studied in biology, physics, and chemistry as well as algorithms proposed in machine learning.