Research


with Aaron Chalfin


A large literature establishes that hiring police officers leads to reductions in crime and that investments in police are a relatively efficient means of crime control compared to investments in prisons. One concern, however, is that because police officers make arrests in the course of their duties, police hiring, while relatively efficient, is an inevitable driver of "mass incarceration." This research considers the dynamics through which police hiring affects downstream incarceration rates. Using state-level panel data as well county-level data from California, we uncover novel evidence in favor of a potentially unexpected and yet entirely intuitive result --- that investments in law enforcement are unlikely to markedly increase state prison populations and may even lead to a modest decrease in the number of state prisoners. As such, investments in police may, in fact, yield a "double dividend'' to society, by reducing incarceration rates as well as crime rates.


Click here to download paper


with Aaron Chalfin and Maria Cuellar


In his 2014 Sutherland address to the American Society of Criminology, David Weisburd demonstrated empirically that the share of crime that is accounted for by the most crime-ridden street segments is notably high and strikingly similar across cities, an empirical regularity that Weisburd refers to as the “law of crime concentration.” We build upon recent work in this area that points out that the concentration of a large share of crime amongst a small number of street segments can, in some cases, be a mechanical artifact of the degree of crime density in a city rather than the signature of an empirical law. Using data from three of the largest cities in the United States, we identify crime concentration by comparing observed crime concentration to a counterfactual distribution of crimes generated by randomizing crimes to street segments. We show that this method avoids a key pitfall that causes existing methods of measuring crime concentration to overstate the degree of crime concentration in a city. Most (but not all) crimes are, in fact, concentrated amongst a small number of hot spots but the precise relationship is weaker — sometimes considerably so — than has been documented in the empirical literature. We further show that, within a city, crime is least concentrated in the neighborhoods that experience the largest number of crimes. Accordingly, the law of crime concentration sometimes holds only tenuously in the communities in which crimes are most prevalent. We conclude that Weisburd’s original intuition remains largely intact though the law of crime concentration requires some qualification.


Click here to download paper



Door locks are an ubiquitous form of security to control access to a building with the goal of reducing crime there. However, research on door locks is often limited by methodological issues and primarily focuses on residential or commercial locations. This paper assesses the impact of card reader door locks on school buildings on an urban universitycampus. Using a difference-in-differences approach, this paper estimates the effect of cardreader locks on crime in buildings. The results indicate that the locks do not significantly affect crime within the buildings. Avenues for future research, including suggestions of specific universities at which to conduct further studies, are discussed.


Click here to download paper


with Li Sian Goh


While all forms of domestic violence can be uniquely traumatizing, incidents resulting in serious injury can lead to lasting physical, mental, and financial consequences for the victim. Hence, it is surprising that most literature on the effects of policy intervention on domestic violence treats such incidents as homogeneous rather than considering differing levels of victim injury. This study provides evidence that decriminalization of marijuana leads to substantial declines in victim injury. Among domestic violence assaults where the victim suffered a serious injury, there was a significant decline in incidents where the offender was under the influence of alcohol or used a weapon.


Click here to download paper



The use of outdoor lighting, particularly through street lights, is a common tool for policy makers attempting to reduce crime. Research on the effect of lights on crime, however, are limited as installing or improving street lighting may affect the community in ways beyond merely increasing outdoor lighting. Welsh and Farrington’s (2008) study suggested that improving street lighting may also improve informal social control in the area as it reflects improved street usage and investments in the community. This paper uses moonlight as a unique measure of outdoor ambient lighting that avoids the issue of community cohesion and examines the effect of lighting directly. The amount of actual moonlight a city receives each night is measured using the interaction between the percent of the moon illuminated and the proportion of the night without clouds. This interactioncreates significant variation in moonlight between cities and across nights in the same city. Contrary to past research on lighting, this study finds that brighter nights, those with a full moon and no clouds, have significantly more crime than nights without any moonlight. These results suggest that there are heterogeneous effects of outdoor lighting by dosage and that more research on possible criminogenic effects of low dosages of outdoor lights are needed.


Click here to download paper


with Shichun (Asminet) Ling and Maria Cuellar


Recent advances, especially the use of DNA technology, have revealed that faulty forensic analyses may have contributed to miscarriages of justice. In this study we build on recent research on the general public’s perceptions of the accuracy of 10 forensic science techniques and of each stage in the investigation process. We find that individuals in the United States hold a pessimistic view of the forensic science investigation process, believing that an error can occur about half of the time at each stage of the process. We find that respondents believe that forensics are far from perfect, with accuracy rates ranging from a low of 55% for voice analysis to a high of 83% for DNA analysis, with most techniques being considered between 65-75% accurate. Respondents still believe that forensic evidence is a key part of a criminal case with nearly 40% of respondents believing that the absence of forensic evidenceis sufficient for a prosecutor to drop the case and that the presence of forensic evidence, even if other forms of evidence suggest that the defendant is not guilty, is enough to convict thedefendant.


Click here to download paper