As a growing number of refugees remain stuck in protracted displacement, policy-makers are looking to economists and other researchers to better understand the economic impact of refugees on local communities.
However, researchers often portray the people living around a refugee camp as a homogeneous group. In particular, the term “host population” reduces an incredible diversity of livelihoods, education levels and interests and can lead to misguided strategies and policies.
One example is a recent article in Refugees Deeply, which summarizes an innovative study using a range of data sources – including nighttime luminosity around a refugee camp – to measure the economic effects of Kenya’s Kakuma camps on the local population. The full study, which draws on research by the World Bank, was published in the Journal of Development Economics (JDE).
The authors tentatively conclude that refugees in Kakuma have a positive impact on economic activity and consumption among locals. This host population is predominantly made up of Turkana people, whom the authors of this study describe as “a community of semi-nomadic pastoralists.”
However, the population of Turkana County is not a homogeneous community but a highly stratified society, divided along lines of livelihood, education, material wealth and identity.
A majority of people are Turkana herders relying primarily on livestock sustained through seasonal migrations; many have never attended school. By contrast, primary and secondary school graduates generally leave the pastoral sector and join an urban class in the more ethnically diverse towns, where they seek business opportunities and employment.
There is also a growing class of former pastoralists whose herds were decimated by drought or raiding. Locally referred to as “those without animals” (ng’ikebootok), they must seek alternative means of survival in urban areas like Kakuma, but are often excluded from formal employment due to their lack of education.
Recognizing the diversity within the host population opens alternative conclusions to those reached in the economics study. The authors use satellite data on luminosity as a proxy for economic activity and consumption, which another study suggests can improve GDP studies in data poor regions.
However, nighttime luminosity also correlates with population density. Rather than increasing consumption, the observed increases in luminosity from 1993 to 2013 could be due to an influx of people, including destitute herders seeking alternative means of survival in town.
If the camp serves as a source of refuge for impoverished herders, this is still an important benefit. But it should not be mistaken for improvements in wealth and household consumption.
I also question the proxy indicators of consumption used in the study, including the presence of “non-grass walls and roofs,” which are not necessarily indicative of economic wealth. Certainly, local businesspeople have benefited materially from the camp economy. But most successful pastoralists remain mobile and invest in their herds, not in permanent shelters. Those who lose their herds are the ones who construct sedentary houses in town.
Finally, the study treats job opportunities as a universal good for the local population. The authors write: “If there is labor market competition from refugees, its overall impacts are swamped by the job opportunities provided by the camp and the increased demand generated by refugees.”
Yet these job opportunities are not the same for everyone. Educated urban residents are often the first to access employment through the government, NGOs and local businesses, including as drivers for motorbike taxi owners.
However, the uneducated poor, including former pastoralists who have lost their herds, tend to work for refugee households, carrying out menial tasks like carrying water or washing clothes for small fees. So long as the labor market is divided into these formal and informal sectors, it may widen the rift between these two classes of the host population.
Meanwhile, most Turkana herders rely less on either formal or informal employment generated by refugees. Many reside in the pasturelands farther from the camps, and their primary interaction with refugees is via trade in charcoal, firewood and basic construction materials procured from the local environs.
But herders are also concerned about losing access to pasture and water sources for their herds, especially after further land was allocated for refugee shelters at the Kalobeyei Settlement. This arrangement is based on a 2015 Terms of Engagement agreement between the national government and the “host community.” However, this label obscures the differing priorities and land use practices of wholesalers, public officials, herders and others.
In Turkana, as in most places, economic change will create winners and losers. Different groups have interdependent and sometimes conflicting interests. For example, urban residents want jobs, while those with herds are more concerned with access to land for grazing.
To its credit, the JDE paper notes: “our results show only average effects. This average is likely to mask important heterogeneity.” However, such limitations are often footnoted and then excluded from abstracts, conclusions and journalistic summaries.
Even if refugees’ presence increases overall economic activity for the host population, it’s important to understand these internal divisions. If the benefits and risks of hosting refugees diverge greatly for different groups, the local population may become increasingly divided by inequality, the consolidation of economic power by an educated commercial elite and pauperization of pastoralists.
Cory Rodgers has worked in Kakuma as a research consultant for the Refugee Economies project at Oxford’s Refugee Studies Centre. The views expressed are his alone and do not necessarily reflect the editorial policy of Refugees Deeply.