Here is my Research Statement. See below for abstracts of selected working papers.
The Transparency of Opportunity, with Garance Genicot.
Educational credentials reflect both productive ability and the opportunities available to acquire them. We develop a signaling model in which opportunity lowers education costs and may be hidden from firms. Hidden opportunity makes education a composite signal of ability and advantage, generating endogenous pooling and investment distortions. We introduce Meritocratic Efficiency, the share of ability variation recovered by labor-market beliefs, and decompose its loss into confounding and pooling. Transparency restores full revelation of ability, while opportunity expansion has distinct equilibrium effects and can reduce Meritocratic Efficiency. Hidden opportunity also creates incentives for tuition discrimination by profit-maximizing schools.
Keywords: educational signaling, opportunity, transparency, meritocractic efficiency, tuition discrimination, multi dimensional types.
JEL Codes: D82, I24, I26, J31, D31.
Stable Government and Coups in Political Networks, with Matthew O Jackson and Konstantin Sonin.
Leadership often changes hands not through elections but through a “coup” within a limited elite. In a military junta, parliamentary faction, or corporate board, the security of an incumbent leader’s position depends on both the number of supporters and their proximity. We study structural foundations of political stability within an elite connected by a network. Each member prefers leaders who are closer to them in the network, and a leader is coup-stable if no rival is preferred by a strict majority, so that a coup-stable leader is a Condorcet winner under network-based preferences. First, we extend the classic median voter theorem to a general class of networks that goes well beyond linear networks and trees. Second, we show that a stable leader need not be a network median: a leader sharing sufficiently many connections with potential challengers is protected from being replaced. Third, networks admitting coup-stable leaders differ in robustness: a leader in a hierarchical network can be overthrown by adding or deleting a single link, while a leader in a dense core-periphery network can withstand more substantial changes. Our results shed light on survival strategies of leaders who control elite network connections.
Keywords: political economy, networks, leaders, coups, stability, Condorcet winner.
JEL Codes: D74, D72, D85, P16, C71.
Inherited Opportunity: Social Mobility, Technical Change and Class Polarization, with David Austen-Smith.
We study a dynastic model in which individuals acquire costly education, and the cost of doing so depends on both productive ability and the opportunity that comes with one’s social class: financial resources, social networks, and other advantages that make education easier to acquire. A child’s class is inherited from parental income, so every education decision doubles as a decision about the next generation’s starting point. Building on this base model, we fix three classes — low, middle, and upper — and extend it with two occupations, a middle-skill and a high-skill one, whose schooling requirements depend on technology. We show that a technology shock which raises the schooling required to enter the middle-skill occupation while narrowing the gap to the high-skill occupation pushes low-opportunity families out of the middle-skill occupation and pulls high-opportunity families past it. Because opportunity is inherited, this is not a one-time reshuffling: the steady-state middle-class share strictly falls by more than any single generation’s transition would suggest, and more dynasties are pushed down than pulled up, so the effect is regressive on net.
Keywords: intergenerational mobility, educational signaling, technology shocks, occupational choice, middle-class hollowing, class polarization, social class dynamics.
JEL Codes: D82, J62, J24, I24, O33.
Who Gets Hired? The Strength of Weak Ties and the Signaling of Strong Ties, with Yves Zenou.
This paper presents a model of referrals in which strong ties serve as a signaling mechanism, allowing employees to convey private information about referred workers. Matching occurs through three channels: strong ties (partially informative), weak ties (uninformative but frictionless), and formal screening in the market (fully informative but frictional). Low-skill workers use weak ties to avoid revealing their type, helping reconcile two stylized facts: Granovetter (1973)’s “strength of weak ties” and the prevalence of referrals in low-skill jobs. The model also reveals a complementarity between weak and strong ties: conditional on having a strong-tie offer, a weak tie raises expected wages — mechanically, through the higher outside option it provides, and, increasingly for higher-productivity workers, through the improved credibility of the referral message itself.
Keywords: signaling, strong ties, strength of weak ties, referrals, formal screening, job search.
JEL Codes: D82, D83, D85, J31, J63, J64.