Publications

Do elections aggregate the private information of office-motivated candidates? Our answer stems from a general result for two-player constant-sum Bayesian games with type-independent payoffs. Under a “completeness” statistical condition, every “identifiable” equilibrium is an ex-post equilibrium. Applied to Downsian elections, the ex-post property implies a sharp bound on information aggregation: equilibrium voter welfare is at best equal to the efficient use of a single candidate’s information. In canonical specifications, politicians may “anti-pander” (overreact to their information), whereas some degree of pandering would be socially beneficial. We discuss other applications of the ex-post result.

The Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) provides infant formula to low-income households through state-level exclusive supply contracts that cover roughly half of U.S. infants. Although these contracts directly govern purchases by WIC participants, they also affect households that receive no WIC benefits. We study these spillovers using household-level in-store and online transaction data matched to WIC contract changes across states from 2018 to 2025. When a manufacturer becomes the WIC supplier, its market share among non-WIC households rises by roughly 30 percentage points in the regular-size formula segment. The response is concentrated in physical stores, where the new supplier’s market share rises by roughly 50 percentage points, while we find little systematic response online or for bulk-size products. This contrast points to the physical retail environment as a central mechanism. Additional evidence from WIC-authorized retailers and hospital formula-practice data suggests that direct exposure to WIC labels and hospital-provided samples account for relatively little of the spillover. The results show that public procurement can substantially reshape demand among untreated consumers by changing the retail environments in which choices are made.

Economic development is often conceived as structural transformation and migration out of rural agriculture (Lewis, 1954; Fei & Ranis, 1964). Understanding development therefore requires us to identify how rural economies transform in response to emigration. We study how international migration reshapes domestic economic activity in origin households using a randomized visa lottery that gave Bangladeshi men job opportunities in Malaysia. Winning the lottery raises total household income through remittances, but income earned in Bangladesh declines due to contractions in nonfarm business activity. In contrast, crop income remains stable despite large reductions in agricultural input expenditures, because farming gets delegated through land rental markets. We identify a “supervision constraint” that explains why households divest from nonfarm business activity. The key input that is lost through male migration is not land, labor, or physical capital, but “management capacity”.

This paper proposes a semi-endogenous growth theory that incorporates technology vintages and the endogenous evolution of multiple technological paradigms through innovation. It provides a characterization of both balanced growth equilibrium and transitional dynamics in an environment where new technologies continuously emerge. From a positive perspective, the model rationalizes two distinct empirical patterns. Using two centuries of US patent data, I first document that the age profile of patents has a pronounced hump shape: most contemporary patents build upon technologies that are between 50 and 100 years old. Second, this age profile has remained stable throughout the past century. From a normative standpoint, the theory underscores a misallocation of research effort induced by the tendency among profit-maximizing firms to overinvest in further developing mature technologies. This yields a suboptimally slow development of emerging technologies. According to a calibrated version of the model, correcting such misallocation could generate welfare gains of 7%.

We study the effects of H-1B immigration on U.S. industries that employ H-1B workers and their trading partners. Using a novel cross-industry design and the 1999–2003 expansion of the H-1B visa cap for identification, we find that H-1B exposure raised incomes for natives and pre-existing immigrants, with gains concentrated in non-STEM occupations. Income gains propagate forward through supply chains to downstream industries but not backward to upstream industries, consistent with a productivity shock rather than a labor supply shock. We find no direct effect on patenting, suggesting that productivity gains arise from better task execution rather than patentable invention.
 

Health insurance lowers the out-of-pocket price of healthcare, and it is well-established that this leads to higher utilization of care. This manifestation of “moral hazard” is typ ically viewed as a social cost of insurance. Within a standard model, this paper shows that a consumer’s ability to change her behavior in response to insurance can also play a central role in the ability of insurance to protect her from risk. We provide a theoretical characterization of this channel and quantify its importance empirically. Under stan dard parameterizations and estimates in the literature, we find that insurance-induced healthcare utilization can account for more than half of the total value of risk protection derived from insurance. Preventing consumers from changing their behavior would lower healthcare spending, but also result in a major loss of risk protection, on-net reducing social welfare in some cases. Our results suggest that under-utilization of healthcare may thus be an equally important threat to welfare as over-utilization.

We provide evidence that domestic outsourcing increases young-worker entry into the formal sector. Leveraging a pair of 1993–1994 Brazilian reforms that reduced the relative cost of outsourcing security guards, a triple-differences design shows that the reforms increased formal guard employment by 4% and hiring from unemployment or informality by 7%, while reallocating formal employment from older to younger workers and leaving demographic-adjusted wages unchanged. Census data corroborate the rise in formality, driven by the youngest cohorts. The compositional shift mirrors a general pattern in Brazil’s matched employer–employee records: conditional on total firm size, employers with greater occupation-specific scale hire younger workers, paid less at entry and more likely to be entering formal employment for the first time. The evidence is most consistent with contract firms supplying at scale the capabilities needed to hire productive workers from outside the formal sector—a demand-side channel for increasing formal sector employment.

Evidence that domestic outsourcing lowers pay comes largely from on-site transfers, in which workers move to a contractor but keep the same jobs. Displacement is rarely observed: whether workers lose their jobs, where they go, how earnings evolve. In Brazil’s 1993–1994 pro-outsourcing reforms, which differentially affected security guards, such transfers were rare; firms instead used occupational layoffs, shedding their guards while keeping other workers. Displaced guards’ employment recovered within five years, but many changed occupations and wages stayed about 12% lower. Lifetime losses average 1.2 to 1.5 years of pre-layoff earnings, concentrated among workers from high-wage firms, reflecting lost premia.

We consider the problem of explaining a data generating process (DGP) to a decision maker (DM) who cannot understand it. Explanations are information, not approximations: they are useful because they rule out possible DGPs. If the DM maximizes her average payoff, explanations using OLS are robustly useful, and augmenting them with summary statistics makes them more so. Sampling error can destroy these guarantees, but theoretical assumptions linking that error to the DGP can restore them. If the DM is sufficiently ambiguity averse, explanations that are affine in outcomes are not robustly useful, but certificates reporting lower bounds on outcomes are.

By age 17, a quarter of U.S. students have experienced a peer suicide. Using linked administrative data from South Carolina and a matched difference-in-differences design, we find that exposure to a peer’s self-harm death increases the probability of a self-harm diagnosis by nearly 50% and both the incidence and frequency of mental health visits. Effects on care use and criminal behavior are concentrated among white boys, and responses diverge by prior mental health history: students without a prior diagnosis increase felony offending rather than care-seeking. Deaths from assault and transportation accidents produce no comparable rise in self-harm, consistent with contagion.

This paper develops a framework for fast online inference on semiparametric models with large sample sizes and possibly many covariates. The computational algorithm itself is the object of statistical study: after a globally consistent warm start in the first phase, the path of averaged online iterates generated in the second phase automatically delivers estimators with optimal convergence rates and valid confidence sets. Both phases require only a single pass over the data stream and are well suited to streaming data or to settings with storage/privacy constraints. For semiparametric monotone index models, the averaged trajectory of the second phase lead to estimators that are automatically orthogonalized and satisfy the laws of the iterated logarithms, and policy functionals are updated along the same trajectory at negligible additional cost. The averaged trajectories satisfy functional central limit theorems, which yield fast online inference via random scaling and bypass the explicit variance estimation that complicates inference for semiparametric models. Applied to a fixed large sample, our online algorithm achieves substantial computational gains over corresponding offline procedures without sacrificing statistical performance. Monte Carlo experiments show adequate behavior. Our methods are applied to 19 million traffic-stop online records from the North Carolina State Patrol (Pierson et al. 2020) and to the international trade data of Helpman et al. (2008) with over 300 regressors. We also compare our estimator to its parametric benchmark in both empirical illustrations.

Empirical models of multi-product demand rely on low-dimensional product representations to capture substitution patterns, increasingly using proxies built from unstructured data. When proxies are imperfect, standard workflows yield biased counterfactuals and invalid inference. We develop a practical toolkit to address these issues. Our methods apply to market-level and/or individual data, require minimal additional computation, provide simple standard-error formulas, and accommodate proxies from fine-tuned models. Further, we propose diagnostics to assess proxy quality. Our methods yield meaningful improvements in predicting substitution in empirically calibrated simulations and in an application where we assess counterfactual prediction performance against a ground truth.

The optimal mechanism for selling a divisible good under convex production costs can be arbitrarily complex, yet firms overwhelmingly use two-part tariffs: a fixed fee plus a constant markup over production cost. We quantify the profit this simplicity sacrifices. For regular value distributions, a two-part tariff guarantees a fraction of the optimal profit that depends only on a lower bound m on the elasticity of the marginal cost, independent of the value distribution. The guarantee approaches 1 as m grows, and is impossible without such a bound. Beyond regularity, no two-part tariff guarantees a constant fraction, but a menu of K + 1 tariffs with a common markup does when the ironed virtual value has K ironing intervals; moreover, the menu size must scale with K. With multiple product lines, the better of separate sales and a single access fee granting purchases at production cost achieves a constant fraction of the optimal profit. These results provide a theoretical foundation for the ubiquity of simple cost-based pricing.

Divorce reshapes family life, yet little is known about one of its most consequential features: the allocation of child custody. We study the impact of joint versus sole custody on both parents and children using rich administrative data from Sweden linked to over 25 years of newly-collected court custody rulings. To address selection concerns, we exploit random assignment of custody disputes to judges who differ sharply in their propensity to grant joint custody. For fathers, joint custody substantially raises earnings and improves mental health, consistent with sustained paternal involvement enhancing labor market attachment and psychological well-being. In contrast, there are no measurable labor market or mental health effects for mothers. Turning to children, joint custody increases standardized test scores and school quality without affecting mental health outcomes. Joint custody increases fathers’ chances of remarriage, keeps separated parents in closer geographic proximity, and has no effect on intimate partner violence allegations against either partner. These findings inform longstanding debates over the role of child custody in shaping post-divorce family life.

With uncertainty about persistence, we show that forecasts necessarily become more persistent and over-react at long horizons. For these reasons, correctly specified and Bayesian forecasts may under-react at short horizons and over-react at long horizons. These results provide a unified explanation for several asset pricing and forecasting puzzles, including: (i) the excess responsiveness of long-horizon rates to short rates, (ii) the dominance of apparent term premia for long-term rates, (iii) the ex post predictability of bond yields, (iv) the excess volatility of long-horizon forward prices, (v) the excess persistence of long-horizon forecasts, and (vi) the over-reaction of long-horizon forecasts.