The analysis of lab data entails a joint test of the underlying theory and of subjects' conjectures regarding the experimental design itself, how subjects frame the experiment. We provide a theoretical framework for analyzing such conjectures. We use experiments of decision making under uncertainty as a case study. Absent restrictions on subjects' framing of the experiment, we show that any behavior is consistent with standard updating (“anything goes”), including those suggestive of anomalies such as overconfidence, excess belief stickiness, etc. When the experimental protocol restricts subjects' conjectures (plausibly, by generating information during the experiment), standard updating has nontrivial testable implications. (JEL C91, D11, D81, D83)

Leading equilibrium concepts require agents’ beliefs to coincide with the model’s true probabilities and thus be free of systematic errors. This implicitly assumes a criterion that tests beliefs against the observed outcomes generated by the model. We formalize this requirement in stationary environments. We show that there is a tension between requiring that beliefs can be tested against systematic errors and allowing agents to disagree or be uncertain about the long-run fundamentals. We discuss the application of our analysis to asset pricing, Markov perfect equilibria, and dynamic games.

An agent observes the set of available projects and proposes some, but not necessarily all, of them. A principal chooses one or none from the proposed set. We solve for a mechanism that minimizes the principal's worst‐case regret. We compare the single‐project environment in which the agent can propose only one project with the multiproject environment in which he can propose many. In both environments, if the agent proposes one project, it is chosen for sure if the principal's payoff is sufficiently high; otherwise, the probability that it is chosen decreases in the agent's payoff. In the multiproject environment, the agent's payoff from proposing multiple projects equals his maximal payoff from proposing each project alone. The multiproject environment outperforms the single‐project one by providing better fallback options than rejection and by delivering this payoff to the agent more efficiently.

A sender persuades a receiver to accept a project by disclosing information about a payoff‐relevant quality. The receiver has private information about the quality, referred to as his type. We show that the sender‐optimal mechanism takes the form of nested intervals: each type accepts on an interval of qualities and a more optimistic type's interval contains a less optimistic type's interval. This nested‐interval structure offers a simple algorithm to solve for the optimal disclosure and connects our problem to the monopoly screening problem. The mechanism is optimal even if the sender conditions the disclosure mechanism on the receiver's reported type.

We study how to regulate a monopolistic firm using a robust-design, non-Bayesian approach. We derive a policy that minimizes the regulator’s worst-case regret, where regret is the difference between the regulator’s complete-information payoff and his realized payoff. When the regulator’s payoff is consumers’ surplus, he caps the firm’s average revenue. When his payoff is the total surplus of both consumers and the firm, he offers a piece rate subsidy to the firm while capping the total subsidy. For intermediate cases, the regulator combines these three policy instruments to balance three goals: protecting consumers’ surplus, mitigating underproduction, and limiting potential overproduction. (JEL D21, D42, D83, H25, L51)

We define the empirical content of an economic theory as the least restrictive observationally equivalent theory. We show that the empirical content of a theory is captured by a certain kind of axiomatization, with axioms that are universal negations of conjunctions of atomic formulae. (JEL B41, D01, D11)