Supporting disadvantaged NEETs: Is a combination of active measures and cash transfers the answer?

What is the benefit of offering active and passive labor market policies in combination, rather than separately? The French “Garantie Jeunes” offers a year of cash transfers and intensive activation measures to youths not in employment, education or training (NEETs) in disadvantaged economic conditions. An impact evaluation found positive effects of the program on employment, albeit mostly in precarious jobs. The results provide important lessons about the role of active and passive policies.

Guest post by Francesco Filippucci | March 2022

Large numbers of young people not in employment, education or training (NEETs) are a significant policy concern in several European countries. Over the last decade, NEET rates for youths aged 15-24 ranged between 12% and 22% in countries such as Spain or Italy, and were persistently above 10% in others, such as France. High levels of NEETs also represent a harsh equity problem. Not only are disadvantaged youths more likely to become NEETs, but NEET spells can have long term “scarring” consequences on employability, demotivating youths and weakening their soft skills, potentially leading to a poverty trap.

Governments have tried to support young NEETs by offering active and passive labor market policies. However, offering active or passive policies in isolation might have ambiguous effects for disadvantaged NEETs. Passive policies in Europe are mostly made of social security cash transfers. Offering cash transfers without increasing activation requirements can disincentivize job search, for example if youths are afraid of losing the benefit once they start working, or if they reduce search effort once cash transfers satisfy their basic needs. In turn, activation measures consist of training, search assistance, or subsidized employment. When these activities are offered without a cash incentive, youths might not feel motivated to attend or put in the necessary effort. For these reasons, international institutions, such as the OECD and the ILO, have advocated for offering active and passive policies jointly. Yet, rigorous evaluations of the effect of combined active and passive labor market policies are scarce..

Evaluation of “Garantie Jeunes”

My recent study fills this gap in the literature by evaluating an innovative French program, Garantie Jeunes, that targets disadvantaged NEETs between 16 and 25 years old. The program combines a year of cash transfers equivalent to the French minimum income (about 500 Euros per month) with intensive activation measures, including:

6 weeks of soft-skills training, covering for example time-management, job-search strategies, preparation of CVs and interviews. The courses are conducted in small groups, with the aim of fostering motivation and information sharing, increasing the effectiveness of the sessions.
Personalized counseling throughout all the program, with individual inverviews held at least once every two weeks and frequent email or phone interactions.
Short-term job experiences, during which youth familiarize themselves with the professional environment and carry out some small tasks. These one-the-job experiences last up to 15 days each and are concentrated in the middle part of the program.

The timeline of Garantie Jeunes is reported in Figure 1 below.

Figure 1: Timeline of Garantie Jeunes

To enroll in the program, youths have to be registered with French Youth Employment Centers (YECs), earn less than the minimum income, and undergo an informal and formal selection process. The informal selection is operated by YECs officials, who tend to propose the program only to youths expected to benefit from it. The formal selection is conducted by local independent commissions, who assess applications from eligible youths based on their motivation and economic difficulties.

A counterfactual impact evaluation of the program was possible thanks to the fact that the program started out from only a few areas in late 2013, and got progressively extended until early 2017. The French Labor Ministry allowed me to create a new dataset on employment paths of youths who registered with YECs. These youths represent more than a third of the French population aged 16-25, as youths register to YECs for many other reasons than participating to Garantie Jeunes, such as receiving information on job offers or participating in other programs.

In my analysis, I exploit that some YECs have already adopted the program (treatment YECs) while others will adopt it only later (control YECs), resulting in a situation where youths are exposed to the program at different points in time and for different durations. To analyze the impact of the program on young people’s employment outcomes, I conduct two comparisons: (i) Within the treatment YECs, I assess the outcomes of youth while Garantie Jeunes is offered compared to youth when the program did not yet exist at the YEC; and (ii) I compare the overall employment outcomes of youth in the treatment YECs to those youth in the contol YEC where the program was not (yet) available at all. This method is called difference-in-differences and has undergone several methodological innovations in recent years. To ensure the comparability of youths across time, I checked that the baseline characteristics of youths registering with YECs don’t change when Garantie Jeunes is introduced.

Promising impacts on employment, but the quality of jobs remains a challenge

The estimated effect of being exposed to the program (i.e. when YECs start offering the program) is reported in Figure 2.

Enrollment: Prior to Garantie Jeunes being implemented in YECs (before the first vertical blue line), by definition nobody enrolls in the program. When youths start having Garantie Jeunes available in their YECs, an increasing share of youths enrolls in the program. Yet, the share of youths enrolling among all those registered at YECs remains quite small. This is not surprising, given the selection process operated by YECs and local independent commissions.
Employment rates: Before the introduction of Garantie Jeunes, there is no significant difference in employment rates of youths from different YECs, reassuring us on the soundness of our methodology. There is also no significant difference during the first four quarters while youth were exposed to the program. A significant positive effect arises from the second year of program implementation (after the second vertical blue line), driven by youths enrolled during the first year and completing the program. Because the effect is driven by a small share of youths in YECs actually enrolled in the program, the magnitude of the underlying effect is quite large – roughly +50% compared to untreated youths — suggesting that the program is very successful in increasing participants’ employability.
Hours worked: The dynamic of the effect of hours worked is similar to the one on employment, with no effect during the first year of program implementation and a positive effect emerging from the second year onwards, driven by youths who have completed the program. The magnitude corresponds to 24 additional hours worked per month.
Hourly wages: Wages seem to be completely unaffected by the program. Because youths tended to work mostly at the minimum wage at baseline, the result suggests that the extra chunk of employment created by the program are still minimum-wage jobs.

Figure 2: Main results obtained through difference-in-difference comparison of exposed vs. not yet exposed youths in YECs. Blue line represents estimated effects, gray lines represent 95% confidence intervals upper and lower bounds

It must be noted that the large effects on employment and hours worked do not translate into an improvement in the quality of jobs. Further analysis estimates the effect of the program on employment in specific kinds of contracts, highlighting that the new jobs found by treated youths are mostly temporary or by the hour, similarly to untreated youths.

How can we interpret this result? A closer look at the behavior of youths while enrolled in the program offers some insights on the mechanisms. First, program beneficiaries tend to reduce job search when they are busy in time-consuming training and subsidized job immersions. Second, they are less likely to work in part-time jobs with a pay above 300 Euros, where the cash transfer gets steeply reduced. This suggests that disadvantaged NEETs are potentially quite time-constrained and very sensitive to the phase-out of the cash transfer with job earnings, therefore not seeking more and better employment while enrolled in the program.

To sum up, a year of cash transfers jointly offered with intensive activation measures seems to strongly increase participants’ employment, although in precarious and minimum-wage jobs. The effect materializes only after completion of the program. In fact, during the program youths’ employment is hampered by the fact that they are busy with activation measures and that they fear losing the cash transfer when finding better paid jobs.

Implications for policy and future research

There are multiple policy lessons arising from this study. First, active and passive policies combined can have large positive effects on disadvantaged NEETs’ employability. In comparison with previous studies which found mild positive effects of activation alone and zero effect of cash transfers alone in similar contexts, this suggests large complementarities between active and passive policies. Second, youth may reduec job search when they are busy in activation measures, when they are subject to phase-out of cash transfers with job earnings and as long as they have the right to the cash transfer. This suggests that policy makers might think about reducing program duration, concentrating activation measures in a short period of time and allowing cumulation of benefits with job earnings.

Of course, the results are obtained in a very specific context. Participants in Garantie Jeunes are very disadvantaged but motivated youths, and the positive effect of the program can decrease if the selection process becomes less stringent, as it was proposed in France. With lower overall impacts, the program might lose its positive benefit-cost ratio, as the estimated impacts come with large costs (only 15% smaller than the current benefits). In addition, it is not guaranteed that the results remain valid in other countries, where (Youth) Employment Centers may not be well developed and the labor market differs in terms of minimum wage, demand and supply of low-skill jobs, and reasons for difficulties of disadvantaged NEETs. For example, some studies found insignificant results in developing countries.

To conclude, the encouraging results provide empirical support to current policy dialogue that advocates for linking active and passive policies, but more research will be needed to extend external validity. Further research could also allow for a better understanding of the mechanisms, for example on the sources of complementarities between active and passive policies. If complementarities arise from activation measures providing some sort of monitoring, then frequent interviews, tight conditionality, and sufficiently long lasting cash transfers seem reasonable. If instead other aspects are key – such as contact with enterprises in job immersions or soft-skills – then the funding should be concentrated on these aspects of the program. As more and more programs of this kind are introduced, policy makers should not refrain from evaluating them, learning important lessons aboout how to maximize their effectiveness.

About the author:

Francesco Filippucci is a PhD Candidate in Labor and Public Economics at the Paris School of Economics. His research revolves around policy evaluation, both in the field of human capital accumulation and its interactions with employment and productivity. His PhD thesis covered employment policies, professional training and place-based policies, in close collaboration with the French Labor Ministry. He has experience as a policy consultant both in the private sector and for the Italian Government, and is active in several policy and data-collection projects.