Who Benefits From Job Corps? A Distributional Analysis of an Active Labor Market Program

Abstract

Using recently developed econometric techniques to estimate quantile treatment effects (QTE) and experimental data, we examine the impact of Job Corps on earnings distribution. Our results indicate a great deal of heterogeneity in the effects of Job Corps. The QTEs show an increasing pattern along the earnings distribution, with much more pronounced differences at the upper quantiles for males, whites, and ages 20-24. Moreover, we find the QTEs to be very small at quantiles below the median for males, ages 16-17 and 18-19, and non-resident students. We propose strong economic conditions and skill hypotheses to explain the heterogeneity observed over the earnings distribution.

Bounds on Average and Quantile Treatment Effects on Duration Outcomes Under Censoring, Selection, and Noncompliance

“…Point estimates based on the IV QT E (for the compliers), presented in Table 2, are inside the estimated bounds for SQT E α cEEU U on the 25 th and 50 th quantiles, but on the 75 th quantile the estimated lower bound is larger in magnitude than the estimated IV QT E. As shown in Figure I2 in the Internet Appendix, containing the estimates for different demographic groups, we only find positive and statistically significant effects on lower quantiles analyzed for the sample of Females. This finding stands in contrast with prior literature that has documented smaller effects for females on earnings and employment relative to other groups (e.g., Flores et al, 2012;Blanco et al, 2013a;Eren and Ozbeklik, 2014).…”

“…Commonly, to address noncompliance the treatment assignment indicator is employed as an instrumental variable for actual participation, which identifies actual training effects for individuals who comply with their treatment assignment, or compliers (Imbens and Angrist, 1994;Angrist et al, 1996). Studies analyzing the effects of actual JC participation on earnings, employment, and wages include Schochet et al (2001), Schochet et al (2008), Frumento et al (2012, Eren and Ozbeklik (2014), Chen and Flores (2015), and Chen et al (2018). However, to the best of our knowledge, despite the importance of analyzing the effects of training programs on duration outcomes (as previously discussed), no study has assessed the effects of this major program on such outcomes.…”

Vocational training for unemployed youth in Latvia

“…7 For the USA, Job Corps stands out as one of the few large-scale programmes of (residential) vocational education and training with a positive impact. Launched in 1964 to target disadvantaged youths in the 16–24 age group, it showed positive post-programme earning gains for older youths (Schochet et al 2008 ), and especially for white and black individuals but not Hispanics (Flores-Lagunes et al 2010 ), with much more pronounced differences for males, whites, and those aged 20–24 at the upper quantiles of the earnings distribution (Eren and Ozbeklik 2014 ). 8 The empirical literature also provides various examples of evaluations of youth-targeted labour market policies with clear eligibility criteria based on age, exploited using regression discontinuity designs (RDDs).…”

Quantile treatment effects in difference in differences models under dependence restrictions and with only two time periods

“…In fact, estimating distributional treatment effect parameters is becoming more common in applied work. Recently, distributional treatment effects have been estimated in the context of welfare reform (Bitler et al, 2006(Bitler et al, , 2008, conditional cash transfer programs in developing countries (Djebbari and Smith, 2008), head start (Bitler et al, 2014), and the effect of Job Corps (Eren and Ozbeklik, 2014). One thing that each of the above empirical papers have in common is that each uses experimental data.…”