This study investigates the effects of a significant housing subsidy program in Jordan on the economic outcomes of Syrian refugee recipients and their social cohesion with Jordanian neighbors. Jordan hosts approximately 650,000 registered Syrian refugees, representing about 6 percent of its 11.1 million population. Notably, over 80 percent of these refugees reside outside of camps.
The study uses a Randomized Control Trial (RCT) involving 2,870 Syrian refugee households in Irbid and Mafraq governorates, with sufficient statistical power to detect moderate impacts on living standards and subjective well-being. The subsidy covered about a year of rent and included funding for landlords to improve housing quality. The design, which restricts subsidies to existing rental relationships, minimizes migration responses typically linked to rental subsidies.
Data was collected through in-person and phone surveys over three and a half years, with refugee households surveyed three times and Jordanian neighbors once. Surveys covered housing expenditures, housing quality, total household consumption, mental health, and child socio-emotional well-being (using the Strengths and Difficulties Questionnaire). Additionally, data from a detailed survey of Jordanian neighbors of both treatment and control households was analyzed to examine the effects of refugee-targeted transfers and housing improvements on host communities’ policy views, altruism, and interactions with refugees.
Main empirical findings:
- The program had largely null effects on living standards, including household consumption and respondent well-being, except for a reduction in housing expenditures.
- There was a significant decrease in the socio-emotional well-being of children in treated households, by 0.34 standard deviations on average. Contributing factors may include reduced food security (due to an unanticipated decrease in food aid), increased COVID-19 exposure, and changes in household composition (including the inflow of additional adolescent household members). This negative effect persisted 1.5 years after the program ended.
- The program led to a significant deterioration in relations between Syrian refugees and their Jordanian neighbors, with a 0.33 standard deviation decrease in an index of Jordanian social attitudes and perceptions of refugees. Increased visibility of aid to refugees may have fueled resentment among neighbors, persisting more than a year after the program ended.
- The negative impacts on child wellbeing and social cohesion may be linked, as decreased interactions between refugee and Jordanian households and children may have had adverse consequences on child wellbeing. Weaker social ties may have also reduced informal support from Jordanian neighbors.
- Any short-term benefits of the program dissipated after the program ended. While the program was ongoing, treated households saw short-term gains in housing quality and financial stability, including increased access to clean water and reduced reliance on loans, but experienced worsening food security and self-reported health, likely due to reduced food aid and increased household size. Shortly after the program ended, treated households reported reduced housing expenditures and increased savings but decreased subjective well-being. One to two years after the program ended, the only significant lasting effect was the substantial decrease in child socio-emotional well-being.
The housing assistance program provided limited short-term economic improvements that dissipated after the program ended, while negative psychological and social cohesion effects persisted. The significant housing subsidy did not lead to transformative positive changes for recipient households, highlighting the numerous constraints refugees face in accessing livelihood opportunities, credit, and quality housing. The deterioration in social cohesion suggests that assistance targeted exclusively at refugees can provoke host community backlash. Delivering benefits more discreetly, such as through cash or mobile money, may reduce the risk of host community backlash. Alternatively, pairing refugee assistance with enhanced host community support could mitigate negative social impacts and foster broader community improvements.