We study the effects of Secure Communities, an immigration enforcement program that dramatically increased interior removals of Hispanic noncitizens from the United States, on participation in means-tested social insurance programs among co-ethnic citizens. Exploiting county-level variation in the roll-out of enforcement together with its ethnic specificity, we find that Hispanic-headed citizen households significantly reduced their participation in two large federal safety net programs. Our results are most consistent with network effects that propagate fear through minority communities rather than stigma or lack of benefit information.

We examine the effect of forest loss around the time of birth on infant mortality and the early childhood health of children in rural Nigeria. We find that forest loss leads to an increase in neonatal mortality—one standard deviation of forest loss is associated with a 9%–15% increase in the likelihood of death within the first month of life. The mechanism linking forest loss to infant death is maternal exposure to malaria when the child is in utero. Such exposure also results in worse birth outcomes, proxied by lower weight-for-age and height-for-age of surviving infants.

This paper studies the executive-legislative exchange of centrally allocated benefits (jam) for legislative support in Colombia using data from road building projects, legislative roll-call votes, and a leaked database which uncovered the assignment of road contracts to individual legislators. We draw hypotheses from a model in which an executive spreads jam to sway legislators. We document that assigned projects had excess costs, legislators targeted were more likely to be swing voters in congress, and legislators increased their support for the executive after their contracts were signed. The results are driven by legislators representing remote regions and constituencies with weaker political institutions.

This paper draws on employer-employee and longitudinal plant data from Mexico to investigate the impact of exports on wage premiums, defined as wages above what workers would receive elsewhere in the labor market. We decompose plant-level average wages into a component reflecting skill composition and a component reflecting wage premiums. Using the late-1994 peso devaluation interacted with initial export propensity as a source of exogenous changes in exports, we find that exports have a significant positive effect on wage premiums and that the effect on wage premiums accounts for essentially all of the medium-term effect of exporting on plant-average wages.

We develop a framework for measuring under- and overreaction in expectation formation. The basic insight is that under- and overreaction to new information is identified (up to sign) by the impulse response function of forecast errors. Our measurement procedure yields estimates of under- and overreaction to different shocks at various horizons. In an application to inflation expectations, we find that forecasters underreact to aggregate shocks but overreact to idiosyncratic shocks. We illustrate how our approach can be used to (i) quantify the importance of different biases, (ii) estimate theoretical models, and (iii) shed light on existing empirical approaches and puzzles.

Workplace safety policies are designed to ensure that employers internalize the costs of injuries, but employers can undermine these policies with threats of dismissal. We show that states’ adoption of the public policy exception to at-will employment—an exception forbidding employers from firing workers for filing workers’ compensation claims or for whistleblowing—led to a substantial reduction in injuries. The widespread adoption of the public policy exception explains 14% of the decline in fatal injury rates between 1979 and 1994. Statutory protections from retaliatory firing also improved safety, but only when employers faced sufficiently strong penalties for violating them.

We report the results of a cost-effective intervention to improve workers' performance in small cassava processing firms in Ghana. We train workers to track their daily output and then randomly assign a sub-sample to set daily production goals. Achieving or missing a goal does not carry monetary consequences. Goal setting increases workers' output by 16%, their productivity by 8% and the average product of labor in firms by 13%. Goal setting is particularly effective for piece-rate workers, increasing their output by 32% and productivity by 24%. While not conclusive, evidence suggests that goals serve as a self-regulation device.