Assessing the Direct and Spillover Effects of Shocks to Refugee Remittances

Jennifer Alix-Garcia, Sarah Walker, and Anne Bartlett

World Development, Volume 121 (2019), Pages 63-74


Over 60 percent of refugees in Kenya’s Kakuma refugee camp receive remittances, over half through informal money transfer services. On average, remittances comprise half of the income of families that receive them. In April 2015, in response to an Al Shabaab terrorist attack on Garissa University College, the Kenyan government shut down Somali-operated money transfer agencies, known as hawala. Hawala is most commonly used by Muslims, since its modalities are consistent with Islamic norms. Hawala operators in Kakuma camp were estimated to transfer US$3 million into the camp each year. This paper examines the impact of the remittance shut down on refugees in the Kakuma refugee camp in terms of access to financial resources and household consumption, as well as the impact on host communities living in proximity to the camp. The authors rely on qualitative interviews and household surveys in Kakuma refugee camp and surrounding host communities during or shortly after the shut down.

Key findings:

  • The remittance shutdown decreased remittance inflows into the camp, at least temporarily.
  • Although overall consumption of food items did not change for the average refugee household, households most likely to use Islamic transfer schemes experienced large decreases in the consumption of milk, meat and phone cards. Refugees who actively used hawala in the past or were members of ethnicities likely to be Muslim decreased their consumption of meat and milk by magnitudes ranging from 26 to 33 percent.
  • At the same time, households more likely to use other transfer systems increased their consumption of these same items, suggesting that prices may have decreased due to weakened demand from remittance-receiving households.
  • Less than 10 percent of host households living in close proximity to the camp receive transfers, compared to over 60 percent of refugees.
  • Prior to the shutdown, host households living in close proximity to the camp had higher purchases of some key items, suggesting that host households living round the camp are somewhat better off than those farther away.
  • The shut down negatively affected host households living in close proximity to the camp, which were less likely to purchase sugar and tea. These households experienced statistically significant decreases of 12 and 25 percent in the probability of purchasing sugar and tea, which are considered luxury items among this population. These effects were stronger for households selling animals near the camp. This is consistent with the theory that host households that are more engaged with markets are more likely to be affected by a decrease in cash flows to the camp.


These findings highlight the economic integration of refugee camps and surrounding host communities, even when refugees are limited in their ability to engage in the labor market, and demonstrate the multiplier effects of remittances. The authors conclude that policies that directly reduce the ability of refugees to purchase goods are likely to hurt the host populations living near the camps. Conversely, policies supporting refugee’s ability to purchase goods, such as cash transfers rather than in kind assistance, are likely to have positive repercussions for host populations.