ISF 190, Project Map.
heterogeneity treatment effects of less educated subpopulations in randomized
and quasi-randomized experiments of youth job-training programs in low, middle,
and high-income countries.
How do active
labor-market youth employment interventions affect labor-market employment
outcomes among non-college educated youth in low, middle, and
Existing research presents mixed views about
the effectiveness of various active labor-market youth employment
interventions. Kluve et al. (2016) find that in a systematic review of 113
counterfactual impact evaluation studies of youth employment programs, only
slightly more than a third have positive impacts on labor market outcomes. The
rest have results that are either statistically insignificant or negatively
significant. Other authors (Grimm et al. 2014) argue that the study
design matters for impacts found; randomized control experiments produce
systematically smaller effects than quasi-experimental studies. Yet others
(Card et al. 2010) argue that program impacts are not affected by
most authors (Grimm et al. 2015, Cho et al. 2014, Card et al. 2010, and Card et
al. 2015) contend that program design matters; job search assistance programs
and entrepreneurship programs are found to be relatively favorable compared to
wage subsidy programs. There also seems to be a consensus on the effectiveness
of programs with regard to post completion time horizons with insignificant
short-term effects and increasingly positive effects for medium-term and
long-term time horizons.
prior studies are important to this study. Kluve et al. (2016) and Card et al.
(2015). In the former, the authors find that employment programs are more
successful in middle and low-income countries than in high-income countries.
They speculate that this maybe because program investments are especially
helpful for the most vulnerable populations – low skilled and low income. In
the later, the authors find systematic heterogeneity across gender and
long-term unemployed participant groups. This study builds on these two studies
by examining whether impact of employment training across different studies varies
within populations of non-college educated youth in low, middle and high-income
I aim to use 15+
counterfactual studies of youth targeting employment programs in countries
across different regions of the world.
Variables of interest: Employment status, self-employment,
wage/income, business start-up/expansion.
Variables: Job skills training, entrepreneurship and/or business
training, job-search assistance, employment services/wage subsidies.
of Studies: I
am using the following sources to identify potential studies to include in my
evaluation: Electronic databases like Harvard dataverse and EconLit.
Handsearching of journals like American Economic Review journals, Journal of Development
Economics, Econometrica, Quarterly Journal of Economics, Review of Economics
and Statistics, Journal of Political Economy, Review of Economics Studies, and Journal
of Labor Economics. I also plan to use Google Scholar to find grey literature, screen
youth employment websites, and literature snow bowl. Search terms: “job-training”, “youth-employment”, “job-search
assistance”, “entrepreneurship”, “employment” “youth” “low-income” and
period: I plan to include studies from 2000 – 2018 for 3 reasons:
First, this is the time when there was a world-wide concern with poverty
alleviation following the implementation of Millennium Development Goals (MDGs)
and Sustainable Development Goals (SDGs). Youth unemployment, though not
overtly stated in MDGs, was an implied policy concern and it is now an explicit
concern of the SDGs. Therefore, there are a lot more studies on youth
unemployment from 2000 than in previous times. Second, the use of RCTs and
quasi-experimental methodologies in impact evaluation studies became more
popular around this time. Third, journals and organizations gradually began to
adopt open data policies which would make it easier to get data for studies
without having to contact all individual authors.
of training programs: Job skills training, business and/or
entrepreneurship training, job-search assistance, and employment
assistance/wage subsidy. Methodology:
Experimental and quasi-experimental designs targeting youth (ages 15-35) and
have individual level data primarily with some measure of level of education
and other participant characteristics.
Extraction: For each of the studies that I end up
including in the study, I will extract information about the program and the
participants. That is, country location and income-level classification, the
type of program or program design, education levels, age, gender and other
variable characteristics of participants, dependent variables used,
methodology, time horizons, and size effect signs and estimates.
Card, D., Kluve, J., & Weber, A.
(2010). Active labor market policy evaluations: A meta?analysis. The economic
Card, D., Kluve, J., & Weber, A.
(2015). What works? A meta-analysis of recent active labor market program
evaluations (No. w21431). National Bureau of Economic Research.
Cho, Y., & Honorati, M. (2014).
Entrepreneurship programs in developing countries: A meta regression analysis.
Labour Economics, 28, 110-130.
Grimm, M., & Paffhausen, A. L. (2015).
Do interventions targeted at micro-entrepreneurs and small and medium-sized
firms create jobs? A systematic review of the evidence for low and
middle-income countries. Labor Economics, 32, 67-85.
Kluve, J., Puerto, S., Robalino, D. A., Romero,
J. M., Rother, F., Stöterau, J., … & Witte, M. (2016). Do Youth
Employment Programs Improve Labor Market Outcomes? A Systematic Review.