Sylvain Chassang

Economic Shocks and Civil War

This article revisits the relationship between income per capita and civil conflict. We begin by documenting that the empirical literature identifies two different patterns. First, poor countries have a higher propensity to suffer from civil war. Second, civil war occurs when countries suffer negative income shocks. In a formal model we examine an explanation often suggested in the informal literature: civil wars occur in poor countries because the opportunity cost of fighting is small. We show that while this explanation fails to make sense of the first empirical pattern, it provides a coherent theoretical basis for the second. We then enrich the model to allow for private imperfect information about the state of the economy and show that mutual fears exacerbate the problem caused by negative income shocks.

Selective Trials: A Principal-Agent Approach to Randomized Controlled Experiments

We study the design of randomized controlled experiments in environments where outcomes are significantly affected by unobserved effort decisions taken by the subjects (agents). While standard randomized controlled trials (RCTs) are internally consistent, the unobservability of effort provision compromises external validity. We approach trial design as a principal-agent problem and show that natural extensions of RCTs - which we call selective trials - can help improve the external validity of experiments. In particular, selective trials can disentangle the effects of treatment, effort, and the interaction of treatment and effort. Moreover, they can help experimenters identify when measured treatment effects are affected by erroneous beliefs and inappropriate effort provision.

Calibrated Incentive Contracts

This paper studies a dynamic agency problem which includes limited liability, moral hazard and adverse selection. The paper develops a robust approach to dynamic contracting based on calibrating the payoffs that would have been delivered by simple benchmark contracts that are attractive but infeasible, due to limited liability constraints. The resulting dynamic contracts are detail-free and satisfy robust performance bounds independently of the underlying process for returns, which need not be i.i.d. or even ergodic.

Accounting for Behavior in Treatment Effects: New Applications for Blind Trials

The double-blind randomized controlled trial (DBRCT) is the gold standard of medical research. We show that DBRCTs fail to fully account for the efficacy of treatment if there are interactions between treatment and behavior, for example, if a treatment is more effective when patients change their exercise or diet. Since behavioral or placebo effects depend on patients’ beliefs that they are receiving treatment, clinical trials with a single probability of treatment are poorly suited to estimate the additional treatment benefit that arises from such interactions. Here, we propose methods to identify interaction effects, and use those methods in a meta-analysis of data from blinded anti-depressant trials in which participant-level data was available. Out of six eligible studies, which included three for the selective serotonin re-uptake inhibitor paroxetine, and three for the tricyclic imipramine, three studies had a high (>65%) probability of treatment. We found strong evidence that treatment probability affected the behavior of trial participants, specifically the decision to drop out of a trial. In the case of paroxetine, but not imipramine, there was an interaction between treatment and behavioral changes that enhanced the effectiveness of the drug. These data show that standard blind trials can fail to account for the full value added when there are interactions between a treatment and behavior. We therefore suggest that a new trial design, two-by-two blind trials, will better account for treatment efficacy when interaction effects may be important.

Decision Theoretic Approaches to Experiment Design and External Validity

A modern, decision-theoretic framework can help clarify important practical questions of experimental design. Building on our recent work, this chapter begins by summarizing our framework for understanding the goals of experimenters, and applying this to re-randomization. We then use this framework to shed light on questions related to experimental registries, pre-analysis plans, and most importantly, external validity. Our framework implies that even when large samples can be collected, external decision-making remains inherently subjective. We embrace this conclusion, and argue that in order to improve external validity, experimental research needs to create a space for structured speculation.

Uniform selection in global games

This paper brings together results which are required in order to extend the global games approach to settings where the game structure is endogenous. More precisely, it shows that the selection argument of Carlsson and van Damme [Global games and equilibrium selection, Econometrica 61(5) (1993) 989-1018] holds uniformly over appropriately controlled families of games. Those results also give proper justification for the inversion of limits which is often implicit in applied work taking comparative statics on the selected risk-dominant equilibrium.

Batched bandit problems

Motivated by practical applications, chiefly clinical trials, we study the regret achievable for stochastic bandits under the constraint that the employed policy must split trials into a small number of batches. We propose a simple policy, and show that a very small number of batches gives close to minimax optimal regret bounds. As a byproduct, we derive optimal policies with low switching cost for stochastic bandits.

Batched Bandit Problems

Motivated by practical applications, chiefly clinical trials, we study the regret achievable for stochastic bandits under the constraint that the employed policy must split trials into a small number of batches. Our results show that a very small number of batches gives close to minimax optimal regret bounds. As a byproduct, we derive optimal policies with low switching cost for stochastic bandits.

Rewards and punishments: informal contracting through social preferences

This paper develops a positive model of informal contracting in which rewards and punishments are not determined by an ex ante optimal plan but instead express the ex post moral sentiments of the arbitrating party. We consider a subjective performance evaluation problem in which a principal can privately assess the contribution of an agent to the welfare of a broader group. In the absence of formal contingent contracts, the principal chooses ex post transfers that maximize her social preferences. We characterize the incentives induced by the principals preferences, contrast them with ex ante optimal contracts, and derive novel testable predictions about the way externalities are internalized in informal settings.

Corruption, Intimidation, and Whistle-Blowing: A Theory of Inference from Unverifiable Reports

We consider a game between a principal, an agent, and a monitor in which the principal would like to rely on messages by the monitor to target intervention against a misbehaving agent. The difficulty is that the agent can credibly threaten to retaliate against likely whistleblowers in the event of an intervention. In this setting intervention policies that are very responsive to the monitors message provide very informative signals to the agent, allowing him to shut down communication channels. Successful intervention policies must garble the information provided by monitors and cannot be fully responsive. We show that even if hard evidence is unavailable and monitors have heterogeneous incentives to (mis)report, it is possible to establish robust bounds on equilibrium corruption using only non-verifiable reports. Our analysis suggests a simple heuristic to calibrate intervention policies: first get monitors to complain, then scale up enforcement while keeping the information content of intervention constant.