GiveDirectly Uganda Endline Report

Executive Summary

Many studies have shown that large, unconditional cash transfers (UCTs) reduce poverty and improve quality of life, at least in the short run. But will UCTs be as successful in refugee communities, where residents face larger barriers to income-generating opportunities? And will they have similar effects in the context of the dual shocks of COVID-19 and aid cuts?

The non-profit organization GiveDirectly is providing a 1,000 USD UCT to ~10,000 refugee households and ~5,000 host community households in Kiryandongo District, Uganda over three years. As the evaluation partner, IDinsight conducted an impact evaluation to examine the impact of cash transfers on households and to inform future decision-making about cash programs by donors and implementing organizations.

The impact evaluation included a randomized controlled trial and a longitudinal, qualitative study. Through a public lottery, IDinsight and GiveDirectly randomly assigned ~9,000 refugee households to one of 24 cohorts to receive the cash transfers sequentially. For our study, cohorts 1 and 2 are the “treatment” group, and a random sample of cohorts 17 to 20 is the “control” group. We measure outcomes ~19 months after most treatment households received the cash transfer.

The impact evaluation shows that – amidst the COVID-19 pandemic and cuts to the monthly World Food Programme cash and food aid – refugees in Kiryandongo effectively used the cash transfer to improve their economic and psychological well-being. By our endline survey, treatment households owned more assets, consumed more goods and services, and earned more business income compared to the control group. The observed effects are in line with the effect sizes observed in other large UCT studies conducted outside of refugee settlements.

The key findings include:

• Refugees in Kiryandongo used the 1,000 USD transfer to improve their houses, purchase land, and start businesses. Treatment households had 11% (32.3 USD above the control mean of 296.9 USD) higher monthly consumption at endline.
• Transfer recipients were 8.6 percentage points more likely to own household businesses, and total monthly business revenue was 64% higher compared to control households (14.3 USD above the control mean of 22.3 USD). Recipient households did not have higher agricultural production or revenue compared to non-recipients.
• Households used the transfer to invest in land and home improvements. At endline, recipient households had assets worth 60% (1,386 USD above the control mean of 2,286 USD) more than non-recipients.
• While most recipients planned to spend the transfer on education, the long, pandemic-induced school closures in Uganda led to no significant changes on education spending.
• The GiveDirectly transfer aided refugees in improving their psychological well-being: our composite psychological well-being index was 0.28 standard deviations higher for recipients compared to non-recipients. Despite the increase, both the treatment and control groups show signs of depression and moderate stress.
  We find no evidence that the transfer affected employment, migration, or household composition.
• We find no significant differences in transfer effects between men and women and no effect on female empowerment.
• Most respondents felt generally safe in their communities. However, theft was a commonly mentioned issue and concern, despite improvements over time. There is no indication that the GiveDirectly transfer has affected safety and security positively or negatively, though some respondents felt more worried about theft when receiving the transfer.
• Our findings on social cohesion are mixed, with many respondents mentioning positive relations and improvements, as well as tensions over resources such as water and land and violent conflicts triggered by inter-tribal romantic relationships. While the GiveDirectly transfer did not affect most dimensions of social cohesion (positively or negatively), it may have contributed to conflicts within families as well as exacerbated existing prejudices and tensions around price discrimination towards refugees and refugees getting “free money.”

Read the full report.