Do Youth Employment Programs Improve Labour Market Outcomes? A Systematic Multi-National Review

The youth unemployment rate worldwide remains high at 13% compared to pre-recession levels, despite some improvement in the overall economy and the implementation of active labour market measures to connect youth to jobs. While previous studies have synthesized the evaluation evidence for such measures, very few have focused specifically on programs and outcomes for youth.

In response to this gap in knowledge, this study reviews evidence on the impact of youth employment programs on youth employment and earnings, as well as the role such factors as program design play in these outcomes. This information has the potential to inform improvements in the design and delivery of employment services for youth.

In their systematic review of youth employment programs, the authors – Jochen Kluve, Susana Puerto, David Robalino, Jose Manuel Romero, Friederike Rother, Jonathan Stöterau, Felix Weidenkaff, and Marc Witte – identified 113 “counterfactual” impact evaluations of 87 programs located in 31 countries, covering a wide range of methodologies and interventions involving both the private and public sectors. Counterfactual evaluations are considered the “gold standard,” in which the program effect is computed as the difference in outcomes (e.g., wage rate) between program participants (the treatment group) and a similar group of non-participants (the counterfactual or the comparison group) to control for the contributions of non-program factors (e.g., more education) to the effect in order to obtain an estimate of the “true” impact of the program. Using “meta-analysis” methods, whereby the results are statistically analysed as a whole in order to improve estimates of treatment effects, the researchers synthesized the evidence based on 2,259 effect sizes and the statistical significance of 3,105 treatment effect estimates.

The analysis also looked at the factors (“covariates”) that influenced program performance, most notably the type of intervention and program design/ implementation (see below). Other covariates considered included the country context (high or low income); characteristics of participants (e.g., age, gender, income level); and features of the evaluation (e.g., peer-reviewed, random control trial).

The types of youth employment interventions reviewed comprised the full range of assistance: skills training (technical and soft skills training); entrepreneurship promotion; employment services (job counselling, job-search assistance, mentoring services); and subsidized employment (wage subsidies, direct subsidy payments to individuals through vouchers etc., reductions in employer social assistance contributions, and labour-intensive infrastructure development projects).

Program design/implementation features that were considered as potentially affecting program impacts included whether or not the intervention had the following features:

  • Standalone or accompanied by other services,
  • Participant profiling to match the services provided with participant needs,
  • A participant engagement mechanism to monitor/follow participants and/or provide incentives to encourage completion or success, and
  • Results/performance-based funding to incentivize practitioners to improve outcomes.

Overall, the researchers found that, while most youth employment programs achieved positive effects on employment rates or earnings, only about a third showed a statistically significant positive impact. This implies that youth employment programs, in most cases, do not appear to work. Much of the difference in performance may be due to country context and to design and implementation factors, suggesting that a focus on design of the program increases the likelihood of its success.

In terms of country context, programs have been more successful in middle/low-income countries. The researchers speculate that this this may be because these countries’ investments are especially helpful for the most vulnerable (low-skilled, low-income) population groups that they tend to target. Also, programs in middle/low-income countries have been implemented more recently than in higher income countries and therefore might have benefited from design and implementation innovations.

As for the type of program, in middle and low income countries, skills training and entrepreneurship programs seem to have had a bigger impact than the other types of interventions. In contrast, in high-income countries, the role of intervention type is less definitive and it is context and how services are chosen and delivered that are greater contributing factors.

Indeed, the review pointed to the important role that design plays in program effectiveness:

  • Programs that integrate multiple interventions are more likely to succeed because they are better able to respond to the different needs of clients.
  • More effective programs are those with participant profiling, which better target assistance on clients’ needs and weaknesses, and those with individualised incentive/follow-up systems that encourage participant retention and success, implying the need for robust monitoring and evaluation systems.
  • Results/performance-based incentive systems for service providers (practitioners) to encourage them to meet their objectives had a positive influence on outcomes, though role of different types of incentive could not be discerned, a topic that requires more research.

This review would be of particular interest to both practitioners and program/policy designers who wish to enhance labour market outcomes for youth by making employment programming more effective.