Evidence on Saudi women’s take-up of unemployment assistance

Despite the narrowing of gender gaps in education, gender inequality in labor markets remains a global challenge, particularly in the Middle East and North Africa. In Saudi Arabia, for example, half of women are college-educated or hold advanced degrees, yet their labor force participation was just 22 percent in 2018, well below the global average of 50 percent. Moreover, many women who are eligible for Saudi Arabia’s unemployment assistance program, Hafiz, have not enrolled. 

Does this low take-up suggest that Saudi women are uninterested in seeking employment or that they lack information about the program? Rohini Pande and coauthors worked in partnership with AlNahda, a nonprofit dedicated to Saudi women’s social and economic empowerment, to experimentally test the impacts of different forms of messaging on Hafiz take-up. They found that take-up was higher when women learned that the program gave them full autonomy over registering for benefits and receiving financial benefits.
 

In recent years, the Saudi government has implemented a range of active labor market policies that differentially target women – including Hafiz, an unemployment assistance program launched in 2011 to reduce barriers for labor force entry or reentry. While anyone can join Hafiz, women comprised nearly 85 percent of its beneficiaries through 2017.

Hafiz provides active job seekers between the ages of 20 and 35 with a monthly stipend of up to 2000 Saudi riyal (about $530 USD), adjusted to beneficiaries’ incomes, for up to a year. Eligibility is limited to people who have been unemployed for at least three months and out of formal education for at least six months, and participants are required to take trainings and actively search for jobs. A 2014 expansion, Hafiz 2, targets jobseekers aged 36 to 60, with payments tapered over the twelve-month period.

Despite the fact that most Hafiz beneficiaries are women, many eligible women do not apply. To analyze what might motivate Saudi women to sign up, Rohini Pande – the Henry J. Heinz II Professor of Economics and Director of the Economic Growth Center – and co-authors Yazan Al-Karablieh of the International Monetary Fund and Rema Hanna of Harvard Kennedy School tested the role of different messaging approaches.

“Existing research on social protection programs,” notes Pande, “suggests that low participation often results from factors such as inadequate eligibility information, difficult application processes, stigma, and the perception that any benefits or wages would be captured by other household members.”

Using a database from the Saudi non-profit AlNahda, the researchers identified 746 women as potentially eligible for Hafiz or Hafiz 2. After a phone survey to collect baseline data, participants were randomly assigned to one of four groups that received differentiated follow-up text messages: a control group, an “eligibility” group, a “privacy” group, and an “agency” group.

All groups received a simple “thank you” message. The eligibility group also received a prompt to check whether they qualified for Hafiz. In addition to the eligibility message, the privacy group received information about the confidentiality of Hafiz registration and the fact that three million women had benefited from the program. Finally, the agency group received the eligibility message combined with a statement that registration required no other individual’s approval and that benefits would be deposited directly into a bank account that they alone could access.

Receiving any of the three follow-up messages more than doubled the Hafiz application rate, from 3.2 percent in the control group (which only received the thank you message) to 7.9 percent.

Across treatment groups, the agency message boosted Hafiz applications to 10.3 percent. This effect was even stronger on lower-income women, whose application rate increased to 14.0 percent – suggesting that control over benefits matters more in households that may have more competition for resources. The eligibility message also had a statistically significant effect, increasing the application rate to 7.2 percent, without any significant differences due to income or marital status.

While the privacy message had no significant effect on take-up on average and did not vary with incomes, among married women it increased the application rate to 12.8 percent – indicating that the confidentiality of Hafiz benefits is especially important to married women. Women living in households without men did not respond differently to each treatment, suggesting that the observed effects have a strong gendered component.

As Pande notes, these findings are directly policy-relevant.

“Take-up could be raised at minimal additional cost by simply ensuring potential applicants understand the program design,” she said. “Overall, the results suggest that the Saudi government could significantly increase take-up of unemployment assistance by distributing messaging that states the financial benefits of this assistance, underlines the sole agency of any applicant, and highlights the confidentiality of their application.”

Research Summary by Jonathan Bower.