Association for Public Policy Analysis & Management

Ethiopia is one of the countries with the lowest income per capita in the world and has been suffered from numerous famines that significantly hit the economy. Despite the continuous food assistance from the international community and the effort of the Ethiopian government, however, Ethiopia still remains as one of the most food-insecure countries.

            In order to tackle the food insecurity issues in Ethiopia, aid donor countries send in-kind food transfers to food-insecure regions often as emergency assistance. While such food assistance greatly helps the Ethiopian population in need, however, the influx of in-kind food transfers may result in negative consequences in local commodity markets. In particular, critics argue that the inflow of food aid imports into local markets brings down the commodity price, depressing local production. Called the “disincentive effect,” such impact may ultimately defeat the purpose of food aid, which is poverty reduction. However, due to confounding factors in market dynamics, policy choices, and seasonality, previous literature presents ambiguous empirical findings in measuring the effect.

            While the majority of the previous research has focused on empirically quantifying the disincentive effect on the national level, the focus of this study is on the sub-national level because the microeconomic environments of food recipient countries greatly vary. In doing so, this study aims to estimate the effect by using vector autoregression (VAR) measuring the dynamic relationship between sub-national level (woreda, districts) food production and food aid receipt of rural households in Ethiopia from 1994 to 2009. To that end, this study uses various data sources including the Ethiopian Rural Household Survey (ERHS) and sub-national food production statistics from the Central Statistical Agency of Ethiopia. Here, the food refers to the production and aid receipts in cereal as the primary crop production and food assistance are in cereal in Ethiopia.

            By using VAR, the expectation of this study is to estimate short- and long-term effects of food aid influx on food production in rural districts of Ethiopia. In the process, as a robustness check, Granger-causality tests are carried out in order to examine whether the lagged values of food aid receipts properly predict district level food productions.