Publications

We study rainmaking as an instrumental religious belief. We present a model in which a religious leader tries to persuade people to believe. Praying for rain can persuade only where the hazard of rainfall during a dry spell is increasing over time, so that prayer is most likely to succeed when people most want rain. We present evidence from prayers for rain in Murcia, Spain, where the hazard rate is increasing, that the church’s prayers for rain predict rainfall over two centuries. To generalize this finding, we gather an original data set of whether ethnic groups around the world traditionally prayed for rain. We find that ethnic groups facing an increasing rainfall hazard are 47% more likely to pray for rain, consistent with our model’s prediction that societies are more likely to pray for rain where prayer is persuasive.

We study the problem of designing a dynamic reassignment mechanism when agents' preferences over objects change over time. In the context of Norway's system for (re)assigning patients to general practitioners (GPs), we provide direct evidence of misallocation under the current mechanism—patients waiting for each others' GPs, but who cannot trade—and estimate a structural model of GP-switching behavior to evaluate alternatives. Introducing top trading cycles (TTC) would, on average, reduce waiting times and increase patient welfare. However, patients endowed with less desirable GPs would be harmed. Prioritizing these patients can avoid these harms while preserving most of the gains from TTC.

This paper studies the implications of central bank credibility for long-run inflation and inflation dynamics. We introduce central bank lack of commitment into a standard nonlinear New Keynesian economy with sticky-price monopolistically competitive firms. Inflation is driven by the interaction of lack of commitment and the economic environment. We show that long-run inflation increases following an unanticipated permanent increase in the labor wedge or decrease in the elasticity of substitution across varieties. In the transition, inflation overshoots and then gradually declines. Quantitatively, inflation overshooting is persistent, and the welfare loss from lack of commitment relative to inflation targeting is large.

We state conditions under which choice data suffice to identify preferences when consumers may not be fully informed about attributes of goods. Our approach can be used to test for full information, forecast how consumers will respond to information, and conduct welfare analysis when consumers are imperfectly informed. In a lab experiment, we successfully forecast the average response to new information when consumers engage in costly search. In data from Expedia, our method identifies which attribute was not immediately visible to consumers in search results and allows us to compute the value of additional information.

Customs data reveal the heterogeneity and granularity of relationships among buyers and sellers, showing how more exports to a destination break down into more firms selling there and more buyers per exporter. We develop a quantitative general equilibrium model of firm-to-firm matching that builds on this insight to separate the roles of iceberg costs and matching frictions in gravity. In the cross section, we find matching frictions as important as iceberg costs in impeding trade, and more sensitive to distance. Because domestic and imported intermediates compete directly with labor in performing production tasks, our model also fits the heterogeneity of labor shares across French producers. Applying the framework to the 2004 expansion of the European Union, reduced iceberg costs and reduced matching frictions contributed equally to the increase in French exports to the new members. While workers benefited overall, those competing most directly with imports gained less, even losing in some countries entering the EU.

Do elite colleges help talented students join the social elite or help incumbent elites retain their positions? We combine intergenerationally linked data from Chile with a regression discontinuity design to show that, looking across generations, elite colleges do both. Lower-status individuals who gain admission to elite college programs transform their children's social environment. Children become more likely to attend high-status private schools and colleges and to live near and befriend high-status peers. In contrast, academic achievement is unaffected. Simulations combining descriptive and quasi-experimental findings show that elite colleges tighten the link between social and human capital while decreasing intergenerational social mobility.

Language skill gaps between advantaged and disadvantaged children emerge well before formal schooling, in part due to disparities in parent–child shared reading practices. This paper reports results from an 11-month randomized controlled trial. The study evaluates the impact of providing low-income families with an electronic tablet loaded with a digital library on the language skills of children aged three to five. The digital library included over 200 children’s books and, for a randomly assigned subset of families, incorporated behaviorally informed goal-setting and reminder messages. All tablet functions other than the reading application were disabled. Access to the digital library alone led to a significant improvement of 0.29 standard deviations in children’s language skills relative to families who did not receive it. We find no statistically significant evidence that behavioral messages improve outcomes beyond access to the digital library on average. Subgroup analysis suggests that the benefits of digital library access are concentrated among children with lower baseline language skills: children below the median improved by approximately 0.6 standard deviations, significantly more than their higher-skilled peers, closing about 15 percent of the baseline language gap. Overall, these findings demonstrate that educational technology designed for home use can meaningfully support early language development and may help reduce disparities in children’s learning outcomes.

This paper studies how the risk of hold-up affects procurement. I use data on the universe of solar power auctions in India. The Indian context allows clean estimates of counterparty risk, because solar plants set up in the same states, by the same firms, are procured in auctions intermediated by either risky states themselves or the trusted central government. I find that the counterparty risk of an average state increases solar prices by 10%. This risk premium sharply reduces investment, because demand for green energy is elastic. Contract intermediation by the central government eliminates the counterparty risk premium.

This article studies the effects of automation in a task-based economy in which some jobs pay workers rents—wages above workers' outside options. We show that automation targets high-rent tasks, dissipating rents, amplifying wage losses, and reducing within-group wage dispersion in exposed groups. This form of rent dissipation is inefficient and offsets the productivity gains from automation. Using U.S. data from 1980 to 2016, we find evidence of sizable rent dissipation and reduced within-group wage dispersion due to automation. Automation accounts for 52% of the increase in between-group inequality since 1980, with rent dissipation explaining one-fifth of this total. Our estimates imply that inefficient rent dissipation has offset 60%–90% of the productivity gains from automation over this period.

We study the possibility of identifying time-inconsistent preferences in empirical designs where preferences are elicited in advance at time 0 and then again at time 1, after the agent receives additional information. For single-peaked preferences, time consistency is rejected only when the time-1 ranking between a pair of alternatives is always the reverse of the time-0 ranking. We establish variations and generalizations of this result. Since such stark reversals are rarely observed, choice-revision designs require stronger identification assumptions than perhaps previously appreciated. But we show that time inconsistency is identifiable in environments where preferences over alternatives can be "priced out."

The US labor share has declined, especially in manufacturing and retail. Yet the labor share of a typical firm in these sectors has risen. We introduce a model where firms incur fixed costs to automate tasks. A decline in the price of capital goods used for automation reproduces the observed patterns: large firms automate tasks, reducing the aggregate labor share, while the median firm continues to operate a labor-intensive technology. When calibrating the automation fixed costs to match the observed adoption heterogeneity, the model generates the aggregate and firm-level facts quantitatively in response to lower capital prices, especially in manufacturing.

We develop a model of multi-dimensional misspecified learning in which an overconfident agent learns about groups in society from observations of his and others’ successes. We show that the average person sees his group relative to other groups too positively, and this in-group bias exhibits systematic comparative-statics patterns. First, a person is most likely to have negative opinions about other groups he competes with. Second, while information about another group’s achievements does not lower a person’s prejudice, information about economic or social forces affecting the group can, and personal contact with group members has a beneficial effect that is larger than in classical settings. Third, the agent’s beliefs are subject to “bias substitution”, whereby forces that decrease his bias regarding one group tend to increase his biases regarding unrelated other groups.

We examine the effects of international trade in the presence of a set of domestic distortions giving rise to informality, a prevalent phenomenon in developing countries. In our quantitative model, the informal sector arises from burdensome taxes and regulations that are imperfectly enforced by the government. In equilibrium, smaller, less productive firms face fewer distortions than larger, more productive ones, potentially leading to substantial misallocation. We show that in settings with a large informal sector, the gains from trade are significantly amplified, as reductions in trade barriers imply a reallocation of resources from initially less distorted to more distorted firms. We confirm findings from earlier reduced-form studies that the informal sector mitigates the impact of negative labor demand shocks on unemployment. Nonetheless, the informal sector can exacerbate the adverse real income effects of economic downturns, amplifying misallocation. Last, our research sheds light on the relationship between trade openness and cross-firm wage inequality.

Doctors often treat similar patients differently, which affects health outcomes and medical spending. We assess the recent literature on doctor decision-making through the lens of a model that incorporates diagnostic and procedural skills, beliefs, incentives, and differences in patient pools. Decision-making is affected by beliefs, training, experience, peer effects, financial incentives, and time constraints. Interventions to improve decision-making include providing information, guidelines, and technologies like electronic medical records and algorithmic decision tools. Economists have made progress in understanding doctor decision-making, but applications of that knowledge to improving health care are still limited.