Abstract
Many young people enter the labor market with few qualifications, limited information about available opportunities, and little experience navigating employment services. Governments often respond by offering career guidance, training, job search assistance, internships, or subsidized employment. Yet participation in these programs is frequently low. Young people may not enroll, may drop out early, or may attend meetings without taking up the investments that could improve their employment prospects. This paper studies whether financial incentives can address this problem. We evaluate a randomized conditional cash transfer linked to participation in the French national career guidance program for young, low-skilled jobseekers. The program was administered by local Job Youth Centers and offered individualized counseling, information on job offers and training opportunities, and support with career planning. In the experiment, 5,498 young people who had enrolled in the guidance program were randomly assigned either to the standard program or to the same program plus a monthly cash transfer. Treated participants could receive up to €4,800 over two years, conditional on complying with the program. In practice, because the guidance program was highly individualized, the main enforceable condition was attendance at regular meetings with the counselor.
The study provides, to our knowledge, the first clean randomized test of a conditional cash transfer tied to participation in labor market activation. This differs from traditional conditional cash transfer programs, which usually condition payments on children’s school attendance or health clinic visits. It also differs from unemployment insurance or welfare systems, where benefits are often tied to job search requirements, but where it is difficult to isolate the effect of the transfer itself. In our setting, the cash transfer was added experimentally to an existing activation program that all participants could access. The results show that young jobseekers respond strongly to financial incentives. The cash transfer substantially increased program participation. Treated participants spent more time in the guidance program, attended more counselor meetings, and were much less likely to drop out. The average number of months spent in the program rose from 13.1 to 19.9, and the number of meetings with a counselor increased from 8.2 to 13.3 per participant. Because counselors saw participants more often, they proposed more opportunities: job offers, training options, and career-building activities increased by around 55 percent. However, these additional opportunities did not translate into greater employability investment. Participants did not take up more training, apprenticeships, internships, job search activities, or other career-building investments. The estimated effects on a broad range of investment and search outcomes are close to zero and precisely estimated. Employment also did not improve. If anything, full-time employment declined during the first six months of the program, consistent with standard work disincentives created by cash transfers and income tapering. The findings point to a sharp distinction between increasing participation in an activation program and increasing the behaviors that such programs ultimately aim to promote. The transfer successfully changed the behavior it directly incentivized: attending meetings. But it did not change downstream decisions, such as enrolling in training, increasing job search effort, or taking steps that would improve employment prospects. The study therefore does not show that financial incentives are ineffective. On the contrary, it shows that they can be highly effective. The key issue is that incentives work on the margin to which they are attached. This has an important policy implication. If the objective is to increase employability investment or employment, incentives should be tied as directly as possible to those outcomes. Conditioning payments on meetings may increase contact with counselors, but meetings are only an intermediate step. In personalized activation programs, the relevant investments are harder to define, verify, and enforce. This creates a practical challenge for policy design: governments need contractible conditions that are observable, fair, and sufficiently under the participant’s control, while avoiding incentives for low-quality training, mechanical job applications, or unstable job matches. A second implication concerns the size of transfers. If cash transfers are intended to relax financial constraints, they may need to be large enough to allow young people to meet subsistence needs while investing in their human capital. If a young person relies on low-skilled or informal work to cover basic expenses, and cannot combine this work with training because both require time, a transfer that covers only part of subsistence needs may ease liquidity without enabling a switch from short-term work to longer-term investment. The evidence is consistent with this interpretation. In our experiment, realized income increased by much less than the theoretical transfer amount, partly because transfers were tapered against earnings and partly because they substituted for other income sources. Measures of financial hardship did not improve. Treated participants increased savings, but did not increase consumption or employability investment. Overall, the study highlights both the promise and the limits of conditional cash transfers in labor market activation. Financial incentives can bring young jobseekers into closer contact with employment services. But if payments are conditioned on intermediate or weakly contractible activities, they may buy attendance without changing the investments that matter for employment. More effective designs may require a combination of larger transfers and conditions linked more closely to employability investments.