This article systematically compares 12 distinct refugee subpopulations living in seven refugee camps and three capital cities across Kenya, Uganda, and Ethiopia. The analysis addresses three key questions: What variation exists in income, assets, and consumption among refugees? What strategies and resources, including work, remittances, and aid, do refugees rely on? And how are refugees’ opportunities and constraints shaped by host country institutions and their individual characteristics?
The analysis is based on both quantitative and qualitative data collected between 2016 and 2018. Quantitative data comes from household surveys conducted in seven refugee camps (Kakuma in Kenya, Nakivale in Uganda, and the Dollo Ado camps in Ethiopia) and three capital cities (Nairobi, Kampala, and Addis Ababa). The total sample size comprises 8,996 refugee respondents from 3,673 households. The surveys covered themes such as demographics, economic activities, income, expenditures, assets, networks, mobility, food security, and health and well-being. Qualitative research included over 500 interviews and focus group discussions with refugees and non-refugee stakeholders.
Main findings:
- There is significant variation in income, assets, and food security between host countries and urban/camp contexts. Refugees in capital cities generally have higher incomes, greater asset ownership, more diverse diets, and lower rates of severe food insecurity compared to those in camps. However, certain subpopulations in each capital city also experience high levels of food insecurity.
- Income, assets, and food security vary notably across nationality groups. For example, South Sudanese refugees in Kakuma have significantly lower incomes, fewer assets, and less dietary diversity, along with higher rates of food insecurity compared to other nationalities. Somali refugees in Nairobi, Kampala, and Nakivale generally fare better than Congolese refugees, with higher household incomes and lower prevalence of food insecurity. While in Addis Ababa, Somali households are worse off than Eritrean households, with higher rates of food insecurity and fewer assets.
- There is significant variation in income, assets, and food security across households. Both the fixed effects and the Gini coefficient for per capita monthly income vary across the 12 refugee subpopulations, indicating significant differences within these groups at the household level.
- Refugees’ economic strategies vary significantly. Those in camps largely depend on humanitarian aid, while urban refugees rely more on work and remittances. Somali households tend to have higher incomes due to better access to remittances. Overall, about one in three refugees earn an income from work, but this proportion is much lower in Ethiopia, where refugees’ labor opportunities are more restricted. In Kenya and Uganda, Congolese refugees are more likely to earn an income compared to Somali refugees. Employment rates are particularly low among South Sudanese refugees in Kakuma and Somali refugees in Addis Ababa. Additionally, there is a significant gender gap in the likelihood of earning an income, with men being 13 percentage points more likely to be employed than women.
- Average incomes vary substantially. On average, working refugees earn USD 86 per month, but urban workers earn significantly more than those in camps (USD 127 vs. USD 53). This higher urban income is primarily driven by Somali refugees in Kampala and Nairobi, who earn two to three times more than other refugees. There is also a substantial gender income gap, with men earning 32 percent more than women on average.
- The substantial differences in income among refugees can be partly explained by the types of activities they engage in, which vary by nationality and context. In the Kakuma refugee camp, nearly half of working refugees are incentive workers, receiving payments averaging about USD 74 per month. In the Dollo Ado camps, nearly 60 percent of working refugees receive incentive payments from international organizations and NGOs, averaging about USD 56 per month. However, in Nakivale, few refugees work as incentive workers. When not employed by international organizations or NGOs, refugees tend to specialize in certain activities based on their nationality. For example, Somali refugees are more likely to be involved in trade, while Congolese refugees often engage in manual labor or farming.
- There are wide differences in access to remittances between cities and camps and between nationalities. Urban refugee households are more likely to receive remittances than those in camps, largely due to better networks. Somali households are 32 percentage points more likely to receive remittances than households of other nationalities, and the monthly amount per capita received is USD 44 higher on average. However, Somali refugees in Dollo Ado are an exception, with only 13 percent of households receiving remittances.
- Humanitarian aid also plays an important social protection role, especially for refugees living in camps or settlements who are entitled to food assistance from the World Food Programme (WFP). Urban refugees have limited access to material assistance, with some exceptions like the Urban Assistance Programme (UAP) for Somali refugees in Addis Ababa.
- The regulatory environment in each host country significantly shapes refugees’ economic opportunities. Uganda has relatively progressive policies on socio-economic rights, granting refugees the right to work and freedom of movement. Ethiopia is transitioning from restrictive to more inclusive policies but has yet to fully implement progressive legislation in practice. Additionally, the locality of refugee camps, often situated in remote and economically marginalized areas, further constrains economic opportunities. Informal regulations also limit refugees’ livelihood activities.
- Refugees’ economic strategies and constraints depend on their individual characteristics, such as nationalities, ethnicities, education, and skills. These characteristics strongly affect choice of location, with education being the most critical factor. The location where refugees reside predicts about half of the explained variation in labor market outcomes. Refugees’ demographic characteristics also explain how they fare in the labor market. Beyond labor market outcomes, refugees’ resources mostly depend on where they live and where their relatives live.
The authors conclude that although the economic lives of refugees have some distinguishing and common features, they are also heterogeneous by host country, urban/camp context, nationality, and household/individual-level characteristics. One-size-fits-all approaches to refugee assistance are likely to be ineffective. Instead, assistance programs should be designed with subpopulations in mind, considering differences in living standards, strategies, and how institutions and identity shape economic lives in each context.