Rethinking the Benefits of Youth Employment Programs: The Heterogeneous Effects of Summer Jobs

Abstract

This paper reports the results of two randomized field experiments, each offering different populations of youth a supported summer job in Chicago. In both experiments, the program dramatically reduces violent-crime arrests, even after the summer. It does so without improving employment, schooling, or other types of crime; if anything, property crime increases over 2-3 post-program years. To explore mechanisms, we implement a machine learning method that predicts treatment heterogeneity using observables. The method identifies a subgroup of youth with positive employment impacts, whose characteristics differ from the disconnected youth served in most employment programs. We find that employment benefiters commit more property crime than their control counterparts, and non-benefiters also show a decline in violent crime. These results do not seem consistent with typical theory about improved human capital and better labor market opportunities creating a higher opportunity cost of crime, or even with the idea that these programs just keep youth busy. We discuss several alternative mechanisms, concluding that brief youth employment programs can generate substantively important behavioral change, but for different outcomes, different youth, and different reasons than those most often considered in the literature.

How Do Summer Youth Employment Programs Improve Criminal Justice Outcomes, and for Whom?

“…Finally, this paper is also comparable to Davis and Heller (2017a), which investigates the heterogeneity in outcomes across subgroups as well as potential mechanisms. The Boston results differ in terms of the former but are consistent with the latter.…”

“…Two studies find that the New York City SYEP increases average earnings and the probability of employment during the program, but also that these effects subsequently faded (Gelber, Isen, & Kessler, 2014;Valentine et al, 2017). Another study using machine learning to identify subgroup impacts in Chicago finds that employment improved for only a subset of SYEP participants; this group was younger, more likely to be Hispanic, female, and enrolled in school, and less likely to have an arrest record (Davis & Heller, 2017a).…”

Improving Regulatory Effectiveness through Better Targeting: Evidence from OSHA

“…5 We extend this literature on machine learning, which has focused on predicting outcomes, to instead estimate variation in the causal effects of regulatory inspections, which Athey (2017) points out is the relevant criterion to judge an optimal resource allocation problem. Our approach to estimating heterogeneity in policy effectiveness is shared with a recent study in the context of youth employment programs (Davis and Heller 2017). Relatedly, by illustrating how machine learning can help identify if agents' behavior is best achieving their objectives, our paper is related to work diagnosing deviations from "optimal" behavior, such as pricing in the car rental market (Cho and Rust 2010) and by retail chains (DellaVigna and Gentzkow 2019).…”