Does the impact of refugees on host communities pass down through generations?

Following decades of violent conflicts in Burundi, Rwanda, South Sudan and the Democratic Republic of Congo (DRC), Tanzania became home to thousands of refugees seeking shelter.  By the end of 1994, Tanzania— known for its hospitality and open-door policy— was hosting close to 1.3 million refugees in its northwestern region , making it one of the top four refugee-receiving countries in Sub Saharan Africa. The majority of these refugees settled in 13 main camps in the northwestern districts of Karagwe, Ngara, Kasulu, Kigoma and Kibondo.  In some of these districts, refugees outnumbered Tanzanians five to one— making it perhaps the most pronounced forced displacement crisis. By the end of May 1994, the Benaco refugee camp in Ngara district had become the largest in the world.

The sudden influx of refugees triggered profound short-and long-term effects on the livelihoods and economic activities of Tanzanian hosts.  We recently published a paper, Intergenerational Impact of Population Shocks on Children’s Health: Evidence from the 1993-2001 Refugee Crisis in Tanzania, that looks at the long-term impacts of this sudden inflow of refugees on parents and their children in host communities. It looks at whether the documented health impacts that major parental early life shocks like natural disasters, famine, war or pollution have on their children, also holds true for temporary population shocks like a refugee crisis.

More specifically it asked, do the children born approximately 15 years later to parents that were living in refugee-hosting areas and exposed to a large influx of forcibly displaced persons in early childhood, experience intergenerational health outcomes? 

To determine the health outcomes of these children, we looked at the children’s Height- for -Age Z score (HAZ) — a key health indicator used to determine the probability of stunting in children, which effectively captures the long-term health effects and socio-economic development prospects of a child over his/her lifetime.

We used data from the Tanzania Demographic and Health Survey collected between 2015 and 2016 containing the migration history of mothers and fathers and the exact years they moved to their current area of residence.  The data contained unique GPS information of each household cluster, which accurately assessed the intensity of refugee influx by calculating the distance between host community households and each of the 13 refugee camps. In total, the analysis relied on a sample of 13,266 women between ages 15 and 49 and 10,223 children under age five living in 608 clusters across the entire country.

The results were stark. Findings from preliminary descriptive statistics suggest that on average, the probability of stunting for children under age five born to mothers who were living closer to the refugee camps was 10.1 percentage points higher (45.1 percent) as compared to their peers in low refugee- receiving areas (35 percent). Similarly, on average the probability of stunting for children born to mothers who lived their first five years in high refugee-receiving areas (41.7 percent) is 7.4 percentage points higher as compared to children born to mothers who were either in their late childhood and/or were living in low refugee -receiving areas. (34.2%).

Our main econometric analysis confirms these findings. Children whose mothers lived closer to the 13 camps during their early childhood were found to have a lower height for their age as compared to their peers.

How can these findings be explained? Previous studies in Tanzania (see here and here) have found that while socio-economic opportunities increased for households living closer to refugee camps, health and service provision deteriorated due to the refugee crisis. The sudden arrival of refugees created temporary secondary employment opportunities for mothers and fathers which may have impacted the intra-household allocation of labor and child care. At the same time, the inflow of aid for refugees generated labor market opportunities for skilled labor-- but that led to lower quality and fewer health and education services in host communities. In other words, the positive growth in labor market opportunities for the host population during the refugee crisis may have been detrimental for the well-being of children in host communities including child care, nutrition, morbidity and education.

While results are still very preliminary, this paper finds some support for this hypothesis. Mothers who had spent their first five years of life alongside refugees were more likely to participate in the labor market than their older peers. But mothers who were 5 or younger at the time of the refugee influx were also more likely to have fewer years of education and were less likely to own key assets such as land and a house. In short, their underachievement and lower assets had a negative impact on their children’s health, which is reflected in high stunting rates for those children. Fathers who spent their first five years with refugees were also less likely to own land and a house.  But unlike mothers, early life exposure to refugees had no long-term impacts on fathers’ labor market participation or their attainment of a post-secondary education. This difference between the long-term effects between mothers and fathers could be explained by understanding that often, in times of household shocks, the girls tend to be more neglected than the boys.  Therefore, any early life shocks happening to the mothers are more likely to affect their children rather than shocks happening to fathers.

These early findings are potentially important for policy. While increasing labor market opportunities for hosts is key to maintaining stability during a refugee crisis, possible indirect effects of aid policies such as reduced child care and disruptions in the provision of essential services like health have to be addressed to reduce the long-term consequences of a refugee crisis on host communities.

This research is part of the program “Building the Evidence on Protracted Forced Displacement: A Multi-Stakeholder Partnership" funded by UK aid from the United Kingdom's Department for International Development (DFID).