Crack cocaine markets were associated with substantial increases in violence in the U.S. during the 1980s and 1990s. Using cross-city variation in the emergence of these markets, we show that the resulting violence has important long-term implications for understanding current levels of murder...

We propose the rise of crack cocaine markets as an explanation for the end to the convergence in black-white educational outcomes beginning in the mid-1980s. After constructing a measure to date the arrival of crack markets in cities and states, we show large increases in murder and...

Social Security eligibility begins at age 62, and approximately one third of Americans immediately claim at that age. We examine whether age 62 is associated with a discontinuous change in aggregate mortality, a key measure of population health. Using mortality data that covers the entire U.S....

While time out of work normally decreases subsequent employment, Social Security Disability Insurance (DI) may improve the health of disabled individuals and increase their ability to work. In this paper, I examine the employment of individuals who lost DI eligibility after the 1996 removal of...

Teens are risky drivers and often subject to extra restrictions. We examine the effects of an Australian intervention banning first-year drivers from carrying multiple passengers between 11:00pm and 4:59am, which had represented 3% of their accidents and 18% of their fatalities. Using daytime...

A crucial issue in studying social insurance programs is whether they affect work decisions through income or substitution effects. We examine this in the context of U.S. Social Security Disability Insurance (DI), one of the largest social insurance programs in the U.S. The formula linking DI...

Victor Fuchs has suggested that the persistent positive correlation between education and health habits can be explained by interpersonal differences in the discount rate. If Fuchs is correct, some health habits can be used as instruments for education in standard wage equations. We use whether...

Using data from the 2010-2014 American Community Survey, we use a procedure suggested by Capps et al. (2015) to identify refugees from the larger group of immigrants to examine the outcomes of refugees relocated to the U.S. Among young adults, we show that refugees that enter the U.S. before age...

As health care consumes a growing share of national income in the U.S., the demand for better estimates regarding both the benefits and the costs of new health care treatments is likely to increase. Estimating these effects with observational data is difficult given the endogeneity of treatment...

In recent years there has been a heightened public concern over the potentially harmful effects of environmental tobacco smoke (ETS). In response, smoking has been banned on many jobs. Using data from the 1991 and 1993 National Health Interview Survey and smoking supplements to the September...