Behavioral Responses to Changes in Enforcement Priorities. [download]

Abstract: This paper asks whether a criminal sanction makes crimes less violent. The motivation of this research is to shift from the literature’s common measure of criminal activity, whether one decides to commit a crime or not, to the concept of marginal deterrence, where a criminal act is a set of continuous actions along the spectrum of harmfulness. I test whether weakening criminal sanctions based on the use of violence reduces the severity of a criminal act. In 2013, the United States’ Department of Justice instructed federal prosecutors to decline mandatory minimum sentencing charges for drug criminals who do not possess weapons. This policy lowered the expected punishment of a drug offense, provided the offender did not carry a weapon, and made it more likely that a rational criminal would choose not to carry a weapon. Regression Discontinuity Design estimates of the probability of weapon possession of drug offenders show an 11-percentage point reduction in Florida’s border counties. The results are consistent with marginal deterrence effectiveness: a harsher sanction on more harmful activities lowers the degree of harmful activity that criminals choose.

Refugees and crime in Germany. [download]

Abstract: This study examines the impact of refugee arrivals on the aggregate crime rate in Germany. When the country received over 476,000 refugees in 2015, the country recorded an additional 45,000 crimes committed by refugees. However, refugee arrival did not increase the aggregate crime rate of the country; the overall crime rate dropped by 0.1% in 2015. This study disentangles the effect of refugee arrival on the aggregate crime rate through perpetrator-level analysis. I estimate the impact of the refugee influx on recorded crimes by the immigration status of accused parties. To achieve causal inference, this study uses the presence of vacant military facilities, which are converted to emergency shelters for refugees as the number of incoming refugees exceeds the capacity of the existing accommodation, as an instrument for refugee populations. I find one additional recorded refugee crime for every ten incoming refugees. However, the refugee arrival did not increase the overall crime rate in Germany, because the increase in recorded crimes committed by refugees was offset by a statistically and economically significant reduction in the number of recorded crimes committed by natives.

The Effects of Child Support on Unmarried Teenage Mothers.

Abstract: This study estimates the effect of child support payments on unwed single parents. U.S. child support (CS) payments reached $29 billion in 2019, 1.8 times the volume of TANF benefits in the same year. The economics literature has little to say about the effect of payments on outcomes such as education and income of the custodial parents. Women who gave birth as a teen are the most vulnerable cohort among single parents; only 38% finish high school and 66% are living in poverty. I estimate the effect of child support on educational achievement and income of women who gave birth as a teenager. Using Current Population Survey (CPS) child supplement data, I find that mothers with CS agreements are 10% more likely to finish high school degrees by their mid-20s and 18% more likely by their late 20s. However, child support payments do not appear to increase single mothers’ incomes.