Publications
Too good to be true? Retention rules for noisy agents, with Debraj Ray, AEJ: Microeconomics, 15(2), pp. 493-535.
(Online Appendix)
An agent who privately knows his type seeks to be retained by a principal. Agents signal their type with some ambient noise, but can alter this noise, perhaps at some cost. Our main finding is that in equilibrium, the principal treats extreme signals in either direction with suspicion, and retains the agent if and only if the signal falls in some intermediate bounded set. In short, she follows the maxim: “if it seems too good to be true, it probably is.” We consider extensions and applications, including non-normal signal structures, dynamics with term limits, risky portfolio management, and political risk-taking.
A principal-agent relationship with no advantage to commitment, with Debraj Ray and Rajiv Vohra, Pure and Applied Functional Analysis, 6 (5), 1043-1064 (2021)
This paper explores conditions under which the ability to commit to a menu of contracts in a principal-agent relationship creates no additional benefit for the principal, over and above simultaneous interaction without commitment. A central assumption is that the principal's payoff depends only on the payoff to the agent and her type.
Working Papers
Local merger screens play a central role in the assessment of horizontal mergers with local competition, where agencies and parties must decide which local areas warrant closer scrutiny. Despite their practical importance, evidence on their reliability remains limited. This paper develops a "Local Markets Lab", a stylised model economy in which purchase patterns, catchments, and merger-induced price effects are observed, allowing screening methods and rules to be evaluated under controlled variation in demand and market structure. We show that the merger-induced delta share within a local market is highly informative of price effects within a fixed environment but does not translate robustly across environments, undermining one-size-fits-all threshold rules. We also find that overlap-based methods better approximate true local shares, yet presence-based methods can perform better under current rules by reducing systematic over-flagging from small peripheral overlaps. Overall, screen performance depends on the combined design of catchments, metrics, and thresholds.
In this article, we analyse how the existence of multi-market contact affects the increase in the incentives to collude resulting from a horizontal merger that occurs in one of two markets. We find that the effect of multi-market contact is ambiguous, and we pay particular attention to those cases in which multi-market contact reduces the effect of the merger. This is the case of the (within-market) symmetric versions of the Cournot and Bertrand models of competition, which we consider as illustrative examples. In conclusion, competition authorities should not automatically assume that multi-market contact aggravates the change in incentives to collude brought about by the merger. Quite the contrary.
I study how GDP shocks affect corruption and political turnover, both from a theoretical and an empirical perspective. The theoretical setting presumes politicians are heterogeneous and privately informed about their degree of corruptibility. They face exogenous shocks to GDP, which affect appropriable government revenues. The political incentives to appropriate these rents systematically vary with such shocks, which affects the endogenous re-election/replacement decisions of voters. I predict that transitory output shocks that lie above trend represent a good opportunity to grab rents today - current gains increase relative to expected continuation values. Therefore, corruption is predicted to be pro-cyclical, as is subsequent political turnover. This has the further implication that booms in current mandates decrease corruption in future mandates, because voters eliminate the more corrupt incumbents. I test these three predictions using an annual panel of countries over 1985-2011. I find evidence in support of the theory. Moreover, these cyclical properties are stronger for more democratic countries, suggesting electoral accountability is indeed the bridge that links corruption to business cycles.
A model-based framework for the analysis of anti-dumping investigations is proposed. By taking the determination of dumping as given, the paper focuses on the other two conditions needed for the imposition of anti-dumping duties on imports: the determination of injury to the domestic industry and causation between dumping and such injury. A formula for the computation of anti-dumping duties that follow the Lesser Duty Rule is derived from the model. This novel formula relaxes an implicit assumption of perfect substitution between the domestically produced and the imported products which is standard in current anti-dumping investigations. The theory indicates that quality differentials between domestic and imported products are irrelevant for optimal anti-dumping duties, and a central role is instead played by elasticities of substitution. I empirically evaluate the proposed methodology by applying it in two real-world antidumping cases. With the first case I show how to calibrate the parameters of the theoretical model in order to compute the proposed antidumping duties, and I illustrate how big the difference between current and proposed duties can be in practice: antidumping duties decrease from 97% to 37%. The second example is considered to show that antidumping duties might be insufficient to eliminate the injury to the domestic industry, even if they are infinitely large.