The impact of the 2015 El Niño-induced drought on household consumption: evidence from rural Ethiopia

ABSTRACT

This paper evaluates the impact of the 2015 El Niño-induced drought on household consumption in Ethiopia. A difference-in-difference method was used to compare consumption changes over time in a group unaffected by the drought to the changes in a group affected by the drought. By using the ESS household-level consumption aggregate data, we find that the 2015 drought reduces affected household’s annual consumption by 8%, and the reduction was largely driven by changes in the lower tails of the consumption distribution. Overall, we find a significant consumption decline due to the 2015 drought, and much of the decline has been experienced among the consumption poor, indicating shock resilience inequality among rural households.

Keywords:

• Drought
• consumption
• difference-in-difference
• rural Ethiopia

Acknowledgements

We would like to thank World Bank and Ethiopian Central Statistical Authority for making the Ethiopian Socioeconomic Survey data available for research. We are also grateful to Bahir Dar University, Institute of Economic Research for providing the training on ‘Impact Evaluation Design & Modelling’ to the research team.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Tesfahun Asmamaw Kasie is an assistant professor of Institute of Disaster Risk Management and Food Security Studies, Bahir Dar University, Bahirdar, Ethiopia.

Birhan Sisay Demissie is an assistant professor of Institute of Disaster Risk Management and Food Security Studies, Bahir Dar University, Bahirdar, Ethiopia.

Mihret Jember Bahry is a lecturer of Institute of Disaster Risk Management and Food Security Studies, Bahir Dar University, Bahirdar, Ethiopia.

Gashaw Mulu Gessesse is an assistant professor of Department of Development and Environmental Management, University of Gondar, Gondar, Ethiopia.

Letenah Ejigu Wale is an assistant professor of Institute of Economic Research, Bahir Dar University, Bahirdar, Ethiopia.

Notes

1 The ESS data are available through the CSA website: http://www.csa.gov.et/ or through the LSMS website: http://www.worldbank.org/ls

2 Income diversification index is created through factor analysis. A list of variables assumes value 1 or 0 is used, depending on whether or not a household has been involved in farming activity; employment activities; self-employment activities; received transfers and earned income from rent.

3 Livestock diversification is also created through factor analysis. A list of variables assumes value 1 or 0 is used, depending on whether or not a household has been involved in rearing cattle, shoats, equine and camels.

4 Crop diversifications are created through factor analysis. A list of variables assumes value 1 or 0 is used, depending on whether or not a household has been involved in planting barely, beans, chat (local tree), coffee, enset (local crop type), maize, sorghum, teff (local crop type), and wheat in their agricultural field in the past cropping season.

5 TLU standardizes different types of livestock into a single unit of measurement. The conversion factor adopted is: 1 camel; 0.7 cattle; 0.55 donkeys /mules/horses; 0.1 shoats.

6 Wealth index- is created through factor analysis. A list of variables assumes value 1 or 0 is used, depending on whether or not a household has specific non-productive assets; such as a gabi (local cloth), bed, clock, phone, radio, mofer (traditional ploughing tool) machid (traditional crop harvesting tool) plough axe.