Knowledge that Transforms

Recent studies have linked shale gas development (SGD) to ground water contamination. The extent of these environmental externalities, to date, remains uncertain. To address this gap, we examine whether shale gas development systematically affects drinking water quality by creating a novel dataset that relates SGD to public drinking water samples in Pennsylvania. Our difference-in-differences strategy finds evidence that additional well pads drilled within 1 kilometer of a community water system intake increases shale gas-related contaminants in drinking water. These results are striking considering that our data are based on water sampling measurements taken after municipal treatment.

We employ a triple difference-in-difference approach, using censuses and georeferenced temperature data, to quantify heat effects on internal migration in Central America and the Caribbean. A 1-standard deviation increase in heat would affect the lives of 7,314 and 1,578 unskilled young women and men. The effect is smaller than observed in response to droughts and hurricanes but could increase with climate change. Interestingly, youth facing heat waves are more likely to move to urban centers than when exposed to disasters endemic to the region. Research identifying the implications of these choices and interventions available to minimize distress migration is warranted.

Under the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) of 1996, individuals convicted of drug-related felonies were permanently banned from receiving welfare and food stamps. Since then, over 30 states have opted out of the federal ban. In this paper, I estimate the impact of public assistance eligibility on recidivism by exploiting both the adoption of the federal ban and subsequent passage of state laws that lifted the ban. Using administrative prison records on five million offenders and a triple-differences research design, I find that public assistance eligibility for drug offenders reduces one-year recidivism rates by 10 percent.

The magnitude of and heterogeneity in systematic earnings risk has important implications for various theories in macro, labor, and financial economics. Using administrative data, we document how the aggregate risk exposure of individual earnings to GDP and stock returns varies across gender, age, the worker's earnings level, and industry. Aggregate risk exposure is U-shaped with respect to the earnings level. In the middle of the earnings distribution, aggregate risk exposure is higher for males, younger workers, and construction and durable manufacturing. At the top of the earnings distribution, aggregate risk exposure is higher for older workers and finance.

Abduction is the process of generating and choosing models, hypotheses and data analyzed in response to surprising findings. All good empirical economists abduct. Explanations usually evolve as studies evolve. The abductive approach challenges economists to step outside the framework of received notions about the "identification problem" that rigidly separates the act of model and hypothesis creation from the act of inference from data. It asks the analyst to engage models and data in an iterative dynamic process, using multiple models and sources of data in a back and forth where both models and data are augmented as learning evolves.

We report on two experiments investigating whether there is a gender difference in the willingness to compete against oneself (self-competition), similar to what is found when competing against others (other-competition). In one laboratory and one online market experiment, involving a total of 1,200 participants, we replicate the gender-gap in willingness to other-compete but find no evidence of a gender difference in the willingness to self-compete. We explore the roles of risk and confidence and suggest that these factors can account for the different findings. Finally, we document that self-competition does no worse than other-competition in terms of performance boosting.

This paper adds to the empirical evidence that criminal records are a barrier to employment. Using data from 2,655 online applications sent on behalf of fictitious male applicants, we show that employers are 60 percent more likely to call applicants that do not have a felony conviction. We further investigate whether this effect varies based on applicant race (black versus white), crime type (drug versus property crime), industry (restaurants versus retail), jurisdiction (New Jersey versus New York City), local crime rate, and local racial composition. Although magnitudes vary somewhat, in every subsample the conviction effect is large, significant, and negative.

The share of U.S. workers in alternative work arrangements has increased substantially in recent decades. Micro longitudinal analyses show that unemployed workers are much more likely to transition into alternative work arrangements than other workers. Macro time-series evidence shows that weak labor market conditions lead to an increase in non-traditional work. But the estimated magnitudes imply that the Great Recession and high unemployment in the 2000s can account for only a modest part of the rise in alternative work. Secular factors associated with rising inequality and technological changes making it easier to contract out work appear to be the driving forces.

The secular decline in safe interest rates since the early 1980s has been the subject of considerable attention. In this short paper, we argue that it is important to consider the evolution of safe real rates in conjunction with three other first-order macroeconomic stylized facts: the relative constancy of the real return to productive capital, the decline in the labor share, and the decline and subsequent stabilization of the earnings yield. Through the lens of a simple accounting framework, these four facts offer suggestive insights into the economic forces that might be at work.

This paper discusses recent trends in the use of replications in economics. We include the results of recent replication studies that have attempted to identify replication rates within the discipline. These studies generally find that replication rates are relatively low. We then consider obstacles to undertaking replication studies and highlight replication initiatives in psychology and political science, behind which economics appears to lag.