Archives

Measuring the Market for Legal Firearms

The Massachusetts Firearms Records Bureau recently published administrative data covering the universe of legal firearm transactions in the state. We use these data to validate state-level background checks as a proxy for firearm transactions and show that historical trends in transactions within Massachusetts align with the rest of the United States. Using auxiliary data from a national survey, we show that the Massachusetts dataset can detect patterns in the demographics of both gun ownership and type of firearm purchased. Our analysis suggests that this dataset is a promising source of information for studying interactions in the market for legal firearms.

Employment and Earnings of Men at High Risk of Gun Violence

Since Becker (1968), economists have modeled crime as resulting from higher returns to criminal activity than legal work. Yet contemporary employment data for people engaged in crime is scarce. We surveyed men at extreme risk of gun violence in Chicago about their work in the formal, informal, and criminal sectors. Noncriminal work is common. Two-thirds of respondents specialize solely in the criminal or noncriminal sectors, both earning about minimum wage at the median. Those who mix across sectors typically earn higher wages. We describe workers by type to demonstrate how better understanding sectoral specialization could inform program design.

Local Exposure to School Shootings and Youth Antidepressant Use

While over 240,000 American students experienced a school shooting in the last two decades, little is known about the impacts of these events on the mental health of surviving youth. Using large-scale prescription data from 2006 to 2015, we examine the effects of 44 school shootings on youth antidepressant use in a difference-in-difference framework. We find that local exposure to fatal school shootings increases youth antidepressant use by 21.4 percent in the following two years. These effects are smaller in areas with a higher density of mental health providers who focus on behavioral, rather than pharmacological, interventions.

Trauma at School: The Impacts of Shootings on Students’ Human Capital and Economic Outcomes

We examine how shootings at schools—an increasingly common form of gun violence in the United States—impact the educational and economic trajectories of students. Using linked schooling and labor market data in Texas from 1992 to 2018, we compare within-student and across-cohort changes in outcomes following a shooting to those experienced by students at matched control schools. We find that school shootings increase absenteeism and grade repetition; reduce high school graduation, college enrollment, and college completion; and reduce employment and earnings at ages 24–26. We further find school-level increases in the number of leadership staff and reductions in retention among teachers and teaching support staff in the years following a shooting. The adverse impacts of shootings span student characteristics, suggesting that the economic costs of school shootings are universal.

Is electronic monitoring contributing to gun violence?

Amid a historic rise in gun violence, debates about the underlying factors driving the surge in Chicago have grown to include electronic monitoring (EM), a condition of pretrial release that equips people who have been arrested and assigned to EM with a GPS-monitored ankle bracelet while awaiting trial from home instead of jail. However, the lack of easily accessible public data makes it difficult for all sides to agree on the basic facts about EM and its potential effects on public safety. Analysis of the data we’ve received from the Cook County Sheriff’s Office (CCSO) allows us to ground the debate about EM in a few essential takeaways: for one, it’s an oversimplification to say that EM is driving gun violence in Chicago, given the timing of the relevant changes in the justice system and the small number of people arrested for gun violence while out on EM compared to overall victimizations. Nonetheless, the makeup of who is on EM has changed over the past few years: while the largest increase in the use of EM was for gun possession offenses, the number of people on EM for a homicide or shooting is also much higher than it was in 2016. We also see in the data that people out on EM are themselves victims of violence at a much higher rate than others in the city, suggesting that there’s more to be done to connect people on EM to helpful social services. In any data analysis, there are numerous details about both the data and analytical methods that can be important for understanding the nuances of what the data can and can’t tell us, as well as reasonable alternative choices about how to analyze the data, interpret and present the results, and address some of the limitations of the data themselves. Because of the space constraints in the article, we have omitted many of those details and sensitivity analyses from the published version. We provide those additional details here.

Probable Causation, Ep. 88: Sara Heller & Max Kapustin

EPISODE DETAILS:

In this episode, we discuss their work on how to reduce gun violence:

“Predicting and Preventing Gun Violence: An Experimental Evaluation of READI Chicago” by Monica P. Bhatt, Sara B. Heller, Max Kapustin, Marianne Bertrand, and Christopher Blattman.

OTHER RESEARCH WE DISCUSS IN THIS EPISODE:

Predicting and Preventing Gun Violence: An Experimental Evaluation of READI Chicago

Gun violence is the most pressing public safety problem in U.S. cities. We report results from a randomized controlled trial (N = 2,456) of a community-researcher partnership called the Rapid Employment and Development Initiative (READI) Chicago. The program offered an 18-month job alongside cognitive behavioral therapy and other social support. Both algorithmic and human referral methods identified men with strikingly high scope for gun violence reduction: for every 100 people in the control group, there were 11 shooting and homicide victimizations during the 20-month outcome period. Fifty-five percent of the treatment group started programming, comparable to take-up rates in programs for people facing far lower mortality risk. After 20 months, there is no statistically significant change in an index combining three measures of serious violence, the study’s primary outcome. Yet there are signs that this program model has promise. One of the three measures, shooting and homicide arrests, declined 65% (p = .13 after multiple-testing adjustment). Because shootings are so costly, READI generated estimated social savings between $182,000 and $916,000 per participant (p = .03), implying a benefit-cost ratio between 4:1 and 18:1. Moreover, participants referred by outreach workers—a prespecified subgroup—saw enormous declines in arrests and victimizations for shootings and homicides (79% and 43%, respectively) which remain statistically significant even after multiple-testing adjustments. These declines are concentrated among outreach referrals with higher predicted risk, suggesting that human and algorithmic targeting may work better together.