The Impact of Food Assistance on Food Insecure Populations during Conflict in Mali
Publisher: International Initiative for Impact Evaluation
Author(s): Aulo Gelli, Jean-Pierre Tranchant, Lilia Bliznashka, Amadou Sekou Diallo, Moussa Sacko, Emily Siegel, Elisabetta Aurino, and Edoardo Masset
Date: 2018
Topics: Conflict Causes, Conflict Prevention, Gender, Governance, Humanitarian Assistance, Livelihoods, Programming
Countries: Mali
Mali, a vast landlocked country at the heart of West Africa in the Sahel region, is one of the least developed and most food insecure countries in the world. Mali suffered from a series of political, constitutional and military crises since January 2012, including the loss of government control of northern territories from April 2012 until January 2013. A range of humanitarian aid interventions were scaled up in response to these complex crises. In this study, we build on data from a unique pre-crisis baseline to evaluate the impact of food assistance on food security-, nutrition- and education-related outcomes in rural populations. We design a longitudinal, quasi-experimental study based on two survey rounds, five years apart, in the Mopti region in Northern Mali. Data was collected from 66 communities randomly selected from within food-insecure districts. Study outcomes include household expenditures and food consumption (measured through 7-d recall), child nutritional status, and school enrollment, attendance and attainment.
We estimate program impact by combining propensity score matching and difference-in-difference. In terms of food security and nutrition, food assistance was found to increase household non-food and food expenditures and micronutrient availability. Disaggregating by degree of conflict exposure showed that effects on expenditures and children’s height were mostly concentrated in areas not in the immediate vicinity of the conflict: The effects on height were also concentrated on households in these areas who received at least two forms of food assistance. In villages where armed groups were present, food assistance improved household micronutrient availability from food consumption, including zinc and vitamin A. In terms of education outcomes, school feeding led to increases in enrollment by 11 percentage points, and to about an additional half a year of schooling. Attendance among boys residing in households receiving GFD, however, decreased by about 20% compared to the comparison group. Disaggregating by conflict intensity showed that receipt of any program led to increases in enrollment mostly in higher conflict intensity areas and that the negative effects of GFD on attendance were also concentrated in most affected areas. Conversely, school feeding mostly increased attainment of children residing in areas not in the immediate vicinity of the conflict. Programme receipt triggered adjustments in child labour: school feeding led to lower girls’ participation and time spent in work, while GFD increased children’s labour, particularly for boys.
Evidence from this study suggests that there is scope to improve the design and scale-up of food assistance interventions during conflict. However, humanitarian operations during conflict face important trade-offs involving on the one hand programme scale and effectiveness, and on the other the practicalities of operating in areas under the control of armed groups, including security, governance and transparency.