Does participation in non-farm activities provide food security? Evidence from rural Ethiopia

Abstract

This study examines the impact of non-farm activities on the food security status of farm households in southern Ethiopia. Data are collected from 400 sample farm households. Food security was measured using the household calorie intake/day/AE. We combine the endogenous switching regression model with binary propensity score matching and inverse-probability weighting to correct bias in the sample selection from unobserved factors that affect the decision of households to engage in non-farm activities, and food security status that most previous studies did not consider. Age of the household head, family size, landholding, extension services, total income, and membership in agricultural groups are the major factors that determine the non-farm income-generating activities. Participants in non-farm works are more food secured compared to their counterfactual non-participants. Hence, policies encouraging non-farm activities, like small and micro businesses, and establishing and supporting businesses that can absorb extra labor from the farming activity can help farmers diversify income and improve their food security.