91227 Finance & PSD Impact SEPT 2014 The Lessons from DECFP Impact Evaluations ISSUE 30 Our latest note tests the extent to which search and matching frictions are the cause of high levels of unemployment for educated youth in Jordan, and whether a job matching service can reduce unemployment. Testing Job Matching Services for Unemployed Educated Youth in Jordan Matthew Groh, David McKenzie, Nour Shammout and Tara Vishwanath Unemployment rates for educated for the job. E.g. extroversion is likely to be youth are high through-out most of the important for sales positions, and Middle East. Yet many firms say they have organizational personality types for difficulty finding competent graduates. administrative positions, while some Search and matching theory offers positions may require analytic ability, and one potential explanation for this others depend more on English ability. phenomena. According to these theories, BDC then identified three to five persistent unemployment can arise when matches for each job opening. These high search costs prevent firms with candidates were called to explain the job, vacancies from linking with qualified and if interested, their details were passed graduates. These search frictions may be onto the firm, along with a description of particularly high in developing countries in why they would be a good fit for the job. which employers are less confident with the Firms could then interview candidates, and signal obtained from grades and university hire them. than in developed countries, and where youth are less likely to have done Testing this via a Randomized internships or had serious work experience Experiment while studying. As a result, many workers We worked with a sample of 1,354 and firms rely on a system of connections unemployed graduates – on average known by its name in Arabic, wasta, to fill individuals aged 22 who had graduated job openings. about 9 months ago from university. These were randomized into a treatment group who The Labor Market Matching Service we attempted to match to firms, and a We worked with a Jordanian control group which we didn’t. psychologist and Business Development On the firm side we had an Center (BDC), a leading training services experimental sample of 2,279 small and firm in Jordan to develop and test a labor medium firms who said they were interested market matching service for educated in hiring a worker in the next 6 months, graduates. again randomized into treatment and control, This matching service began by as well as a booster sample of 175 larger testing young job-seekers on a range of firms to provide more job openings. skills. They received four hours of tests that measured mental reasoning, English Results proficiency, Excel proficiency, soft skills  56 percent of the treatment group of such as the ability to work well in groups, job-seekers were randomized into at and personality traits. least one match, with many matched Firms with job openings were then to multiple jobs. This resulted in asked about their needs for workers, and a skill and personality profile was developed 1,143 matches being made. Do you have a project you want evaluated? DECRG-FP researchers are always looking for opportunities to work with colleagues in the Bank and IFC. If you would like to ask our experts for advice or to collaborate on an evaluation, contact us care of the Impact editor, David McKenzie (dmckenzie@worldbank.org)  However, in 28 percent of matches, to be economic costs of taking such jobs, so job-seekers said they weren’t the reason appears to be perceived negative interested in the job; and in 55 social costs and family prestige from taking on certain jobs. percent the firm didn’t offer the job- seeker an interview. Policy implications  Only 10 percent of the matches (115 Our results suggest the solution to matches) resulted in an interview. high educated youth unemployment is more  Job offers were extended by the firm complicated than would be the case if the in almost half the cases where an problem was just high minimum wages or high search costs. We see two avenues for interview took place (54 cases). future policy work:  But job seekers refused 30 out of the 1. Interventions on the firm side to spur a 54 job offers received, and quit private sector which creates more skilled within the first month in another 15 jobs - a combination of unequally applied cases. regulations and other barriers to entry limits  The end result was only 9 jobs that the entry of new firms and reduces the dynamism of the private sector. lasted at least a month out of over 2. Efforts on the labor supply side to 1,000 matches between job-seekers encourage graduates to consider a wider and firms. range of jobs. This could include public sector reforms to make these jobs more Why was this not more successful? comparable to the private sector, rather than On the labor demand side, despite the main thing graduates aspire to; efforts to firms saying that they had difficulty telling promote and celebrate work in the private good and bad candidates apart before our sector, including internships; and more intervention, and saying it is difficult to find career support within universities to help qualified employees, in practice firms seem graduates broaden their range of jobs they to be able to fill positions more easily than might consider. they say. We tracked open positions and It is less clear whether there are policy found 88 percent were filled within 4 weeks. actions that can help reduce the social costs of The question then remains as to why taking a less prestigious job, although this youth turn down so many job opportunities. process may occur slowly as a response to We consider several explanations, and find some of the other changes proposed. the evidence most supportive of the idea of We also don’t believe this one reservation prestige – educated youth are experiment should stop other attempts to unwilling to take jobs they consider beneath develop better job matching services for them, even for a temporary amount of time. graduates in other countries – the approach A second experiment (described in the used here could be more successful in other working paper listed below) provides more societies. evidence for this view. There do not appear For further reading see: M. Groh, D. McKenzie, N. Shammout and T. Vishwanath (2014) “Testing the Importance of Search Frictions, Matching, and Reservation Prestige Through Randomized Experiments in Jordan”, World Bank Policy Research Working Paper no. 7030. Recent impact notes are available on our website: http://econ.worldbank.org/programs/finance/impact