New Working Papers
"The Design of Optimal Collateralized Contracts," with D. Cao, April, 2016.
This paper examines the nature of collateral in a two-period contracting problem. A risk averse borrower contracts to obtain a loan under asymmetric information. The borrower values a capital good and a composite non-capital good. He privately observes an income shock in the composite good in the second period. There is full commitment and no frictions other than asymmetric information. Very generally, collateralization of both goods emerges endogenously in the optimal contract, whereas it does not under full information.
In the optimal contract, the capital good is distorted toward over-consumption in the initial loan period and under-consumption in the repayment period. The contract is strictly separating for high income types, but requires forfeiture of assets and possibly pooling at the bottom. The relation between forfeiture and contractual distortion is summarized by a simple formula showing that higher distortions are associated with larger increases, or smaller decreases, in forfeited collateral. The formula shows that forfeiture is strictly decreasing in income at both tails of the income distribution, and collateral forfeited by low income types exceeds that of high income.
We obtain a closed form solution in a parameterized model. There, forfeited collateral is globally decreasing in income. Pooling occurs at low income levels when the borrower's initial assets are low enough. We give a straightforward algorithm for computing optimal contracts.
"Do Actions Speak Louder Than Words? Auditing, Disclosure, and Verification in Organizations" with L. Anderlini and D. Gerardi, revised February 2016.
Abstract: We study the relative performance of disclosure and auditing in organizations. We consider the information transmission problem between two decision makers who take actions in sequence at two decision dates. The first decision maker has private information about a state of nature that is relevant for both decisions, and sends a message to the second. The second decision maker can commit to exclusively observe the disclosed message, or can retain the option to observe the action of the first decision maker (auditing) or, at some cost, to verify the state. In equilibrium, state verification will never occur and the second decision maker effectively chooses between auditing and disclosure.
When the misalignment in preferences reflects an agency bias - a bias in a decision maker's own action relative to that of the other - then the second decision maker chooses to audit in equilibrium. Actions speak louder than words in this case. When the misalignment in preferences reflects an ideological bias - one decision maker prefers all actions to be biased relative to the other decision maker - then, for a large enough misalignment, the second decision maker chooses disclosure in equilibrium. In this case words speak louder than actions. Our results indicate that the ability to commit to forego the audit has value in the latter case.
"Tipping Points and Business-as-Usual in a Global Carbon Commons," with R. Harrison, January, 2016.
Abstract: This paper studies climate tipping in a dynamic strategic model of carbon consumption among nations. Each period, countries extract carbon from the global ecosystem. A country's output depends both of its carbon usage and on "stored carbon" in the ecosystem. Its desired mix of extracted versus stored carbon is determined by its output elasticity with respect to carbon input. These elasticities vary across countries and evolve stochastically over time.
We analyze the countries' strategic incentives in the absence of an effective international agreement. We characterize Business-as-usual (BAU) equilibria: continuous Markov Perfect equilibria in carbon extraction profiles. Under non-concave carbon dynamics, BAU equilibrium extraction profiles may deplete the carbon stock to the point that the global commons spirals downward toward a steady state of marginal sustainability. These thresholds, or tipping points, emerge endogenously in a BAU equilibrium. If the profiles of carbon factor elasticities become large enough, a tipping point is shown to be breached. We find that countries will, in fact, accelerate their rates of carbon usage the closer they are to tipping. Even so, there remains a time span (a "negotiation window") in which a collapse may be averted if the countries agree to implement the socially efficient profile of carbon usage.
"Revealed Political Power," with J. Bai, International Economic Review, 54: 1085–1115 (2013).
"Dynamic Mechanism Design for a Global Commons," with R. Harrison, forthcoming in International Economic Review (2016).
"Communication and Learning," with L. Anderlini and D. Gerardi, Review of Economic Studies, 79(2): 419-450 (2012).
"On the Faustian Dynamics of Policy and Political Power, with J. Bai, Review of Economic Studies, 78: 17-48 (2011).
"Social Memory, Evidence, and Conflict," with L. Anderlini and D. Gerardi, Review of Economic Dynamics, 13: 559-574 (2010).
"Markov Perfect Equilibria in Repeated Asynchronous Choice Games" with H. Haller, Journal of Mathematical Economics, 46: 1103-1114 (2010).
"The Dynamic Stability and Reform of Political Institutions , Games and Economic Behavior, 67: 569-583 (2009).