from EVIDENCE to POLICY Learning what works, from the Human Development Network December, 2012 74527 Do wage subsidies help young women get jobs? Unemployment among young adults is a problem throughout Development Goals. In order to build evidence of what the world, and it’s of particular concern in the Middle East, works, the World Bank funded the Jordan NOW pilot where half the population is under the age of 25 and more than program, which was designed to encourage employment of a quarter of those aged 15-24 are out of work. Young women female college graduates in Jordan through wage subsidy fare worse than men when it comes to finding jobs. Cultural vouchers and soft skills training. Built into the project was norms can discourage them from working or traveling on their an evaluation to measure the impact. Researchers found that own, meaning that some young women never even make the vouchers did boost employment—but only for as long as the transition into the workforce. How to reduce youth unemploy- vouchers were valid. After that, the new hires were let go or ment in general and give women a boost in particular is of key left their jobs. The training didn’t show any significant ef- concern to policymakers and development groups trying to fect on employment. The evaluation’s results, similar to other make a difference. But it’s not yet clear what steps can reverse studies that have found limited effect of wage subsidies in de- the problem. veloped countries*, show that raising employment rates may The World Bank understands that skills development not only be a matter of helping graduates find their first job, Skills and Jobs and jobs creation is necessary to improving people’s lives but also tackling the structural frameworks that can hamper and helping countries meet the United Nations Millennium job creation, hiring, and retention. Context Jordan, like many countries in the Middle East, has difficulty agers or in their early twenties are less likely to ever work; meeting the demand for jobs even as the economy continues those who marry later are more likely to have held at least to grow. The national unemployment rate is 9 percent for one job, and more likely to continue working (but even then, men and 22 percent for women, but for those under the age labor market participation for married women is less than 20 of 30, the unemployment rate jumps to 19 percent for men percent, among the lowest in the region). Going to college and almost 50 percent for women. With some 70 percent of doesn’t seem to give young women much of an advantage. the country’s population under the age of 30, jobs creation While 70 percent of unemployed young men don’t have any and encouraging new hires is a primary concern. schooling beyond high school, the composition of female Barriers to entering the labor force and getting a job are unemployment is much different: 75 percent of unemployed especially acute for young women. Those who marry as teen- young women have a community college or university degree. Helping women enter the labor force is a priority for the Did you know: Jordanian government, and government officials requested Woman in the Middle East and North Africa participate in the labor World Bank support for developing a pilot program for fe- force at lower rates than any other region in the world. male community college graduates. This group of women The rate for the Middle East and North Africa region is 21 percent, was seen as particularly vulnerable because they compete with compared with a high of 71 percent in the East Asia & Pacific region. university graduates for jobs. The program, called the Jordan Source: World Bank data figures, 2010 New Work Opportunities for Women (NOW) pilot and eval- uation, was launched in August 2010 and ran for 11 months. * Betcherman, Gordon, Karina Olivas, and Amit Dar (2004) “Impacts of Active Labor Market Programs: New Evidence from Evaluations with Particular Attention to Develop- ing and Transition Countries,� World Bank Social Protection Discussion Paper no. 402 Evaluation All women who passed the August 2010 graduating exams JD ($210) per month. The training, which was five hours from eight of the country’s 14 public community colleges a day for nine days, dealt with effective communication, were eligible for the program. In total, 1350 out of the 1395 time-management, and business writing skills. A baseline female graduates were randomly chosen to participate. They survey was conducted one month before graduation, with were randomly assigned to receive either a voucher, soft follow up surveys eight months after graduation, and 16 skills training, both, or neither. The vouchers reimbursed months after graduation. There was a 93 percent participa- employers for up to six months (between October 2010 and tion rate in all the surveys. August 2011) of salary, equal to the minimum wage of 150 Findings Skills and Jobs Vouchers boosted young women’s employment—at Nonetheless, in April 2011, about nine months after least while the vouchers were active. graduating from college (and at a time when the vouchers were still valid), close to 80 percent of the Young women with the vouchers were 39 percent more likely young women, regardless of whether they were in to work than those in the control group. At the midline sur- an intervention or control group, either had a job or vey, when the vouchers were still active, 57 percent of those were actively looking for one. with vouchers were employed compared with 18 percent of the control group. The high labor force participation rate is reflected in the base- line survey, when more than 90 percent of the young women But the skills training program had negligible effect said they wanted to look for work after graduation and more on employment. than 80 percent preferred the public sector. They also had a very positive outlook, with 82 percent saying they expected to Young women who received just the training had an em- have a job within 6 months (the reality is that 40 percent of ployment rate of about 21 percent during this same pe- community college graduates find at least one job within the riod, compared with the control’s group 18 percent em- first year and a half after entering the labor market). More than ployment rate. The difference between the control and 90 percent said they would like to work outside the home after treatment groups wasn’t statistically significant. Likewise, marriage and they didn’t believe they would face any obstacles those who received both the voucher and training had vir- (even as 50 percent said they couldn’t travel alone to the market tually the same employment rate as those who received or a health center; and 44 percent said a girl should obey her only the voucher. brother, even if he was younger). This policy note is based on boyh the World Bank’s Policy Research Working Paper 6141, “Soft Skills or Hard Cash? The Impact of Training and Wage Subsidy Programs on Female Youth Employment in Jordan,� Matthew Groh, Nandini Krishnan, David McKenzie, Tara Vishwanath, July 2012, and “Soft Skills or Hard Cash? What Works for Female Employment in Jordan?� (www.worldbank.org) percent of young women who had received either a voucher or a voucher and training were dropped from their jobs. Among those in the control and training groups, some seven to eight percent of young women lost their jobs. By the last survey, four months after the voucher period expired, the employment rate for all groups hovered around 25 percent. Most of the young women—66 percent—in the voucher group said that they were hired because of the voucher, and that their job ended because the voucher was no longer sub- sidizing their wages. Eighteen percent said they weren’t satis- fied with the job, nine percent quit for family reasons, four The positive outlook was reflected in voucher take-up percent due to salary dispute and two percent said they had to and training program attendance. About half of those of- leave for other reasons. fered the voucher used it, and 17 percent of those who didn’t Similarly, 62 percent of employers said they wouldn’t have redeem it found a job on their own. Among women who made the hire without the voucher. Of these firms, close to 60 qualified for the training-only intervention, some 60 per- cent completed the course (new graduates who were married were 20 percentage points less likely to attend the training). Barriers to employment? For those offered both the training and the voucher, take-up 63% of firms say men are overall was around 75 percent: 40 percent used the voucher more productive and work and took the training, 10 percent used only the voucher, and longer hours than women 25 percent only attended the training. 30% say they prefer to hire men over women Young women with vouchers were able to find 24% say women are bad hires because they aren’t ambitious employment—but employers didn’t register them as Evaluation survey formally employed. Using administrative data, researchers discovered that many percent said the reason was that they had no real need for the of the women employed through the vouchers were not for- worker, indicating the jobs were always viewed as temporary. mally registered in the Social Security System. When employ- ers register workers as formally employed, they have to pay Nonetheless, women who received vouchers did gain social security taxes (12.25 percent of wages) and payroll taxes important workforce experience and were more likely (seven percent of wages), adding 20 percent to the cost of to still be in the labor force—even if this meant they employing a worker. were unemployed but looking for work—than others. Eighty-five percent of women who used the voucher said they were paid exactly the amount covered by the voucher By the endline survey, about a year-and-a-half after the (equal to the minimum wage), meaning most employers young women had graduated college, those who received the didn’t contribute anything to the women’s salaries. job vouchers were 30 percentage points more likely to have worked than those in the control group. They were also 10 The fact that employers didn’t register their new percentage points more likely to be looking for work than hires may have been a sign that they weren’t plan- those in the training or control group. Job vouchers also im- ning to keep them on. Indeed, four months after the proved young women’s subjective well-being (training alone voucher program ended, there was little difference didn’t appear to affect this) and their optimism on the future, in employment rates among young women who had as measured on the Cantril scale. Training alone did appear vouchers and those in the control group. to help build confidence, according to participants’ feedback after completing the course, and also reduced the likelihood Between the midline survey, when the vouchers were valid, and of major depression by almost five percentage points, accord- the endline survey, when the vouchers were no longer valid, 38 ing to the Mental Health Inventory scale. Conclusion Using vouchers is something that policymakers often consider minimal effect when the job-market itself is limited both in job as a way to encourage hiring and give job-seekers on-the-job creation and social outlooks on women’s employment. experience. But as the Jordanian pilot shows, wage-substitution Although the vouchers and training didn’t boost long-term employment, an argument can be made that even short-term jobs are important for empowering women. These jobs can help them gain the self-confidence and work experience they will need later on if they pursue employment in the limited private sector job market. As such, vouchers may be a useful policy op- tion for governments seeking to craft programs with an impact. The average participant was 21 years old and unmarried. Almost half came from one of Jordan’s major cities – Amman, Salt or Zarqa —and had a car or computer at home, but limited internet access. More than 90 percent reported that their mothers didn’t work. Half of the young women said they were not allowed to go to the market unaccompanied. Their studies reflected the traditional areas for working women: teaching, medical assistance, and administration and finance. Eight vouchers don’t necessarily overcome labor market rigidities, percent had jobs lined up upon graduation. Overall, the young wom- en, like new college graduates across the Middle East, expressed a government policies, and overall economic constraints that preference for joining the public sector over the private sector. The limit job creation and hiring. And while training can give job- main reason given was because of the stability of the job. seekers a new sense of confidence and a better outlook, it has The Human Development Network, part of the World Bank Group, supports and disseminates research evaluating the impact of development projects to help alleviate poverty. The goal is to collect and build empirical evidence that can help governments and development organizations design and implement the most appropriate and effective policies for better educational, health and job opportunities for people in developing countries. For more information about who we are and what we do, go to: http://www.worldbank.org/sief. This Evidence to Policy note series is produced with the generous support of SIEF. THE WORLD BANK, HUMAN DEVELOPMENT NETWORK 1818 H STREET, NW WASHINGTON, DC 20433 Produced by Office of the Chief Economist, Human Development Network, Communications/Aliza Marcus amarcus@worldbank.org For further information on the evaluation, contact Tara Vishwanath at TVishwanath@worldbank.org