A The Implication of Geospatial and Households Characters on Multidimensional Vulnerability in Rural Ethiopia

Abstract

This study investigates the influence of household and geospatial characteristics on vulnerability to multidimensional poverty in rural Ethiopia, utilizing pooled data from the Ethiopian Socioeconomic Survey and applying the Feasible Generalized Least Squares (FGLS) method. The findings reveal persistent poverty dynamics, with 95% of poor households in 2012 and 97% in 2019 remaining vulnerable, while around 6% of previously non-poor households face the risk of falling into poverty. Geographic isolation emerged as a significant driver of vulnerability: each additional kilometer from the nearest main road and market was associated with a 0.3% and 0.02% increase in poverty vulnerability, respectively. Conversely, favorable climatic conditions reduced risks, with a one-unit increase in average annual rainfall linked to a 0.5% decline in vulnerability. Demographic and livelihood factors also shaped outcomes—older household heads experienced a 0.4% reduction in poverty vulnerability per additional year of age, while engagement in non-agricultural activities decreased vulnerability by 2.4% per additional hour allocated. Religious affiliation further influenced household risk, with non-Orthodox and non-faith households exhibiting lower vulnerability levels. Regional disparities were evident, as households outside Harari were more vulnerable than those in Tigray, underscoring the spatially differentiated nature of multidimensional poverty. Overall, the results highlight the need for region-specific, socially sensitive, and infrastructure-focused policy interventions to mitigate poverty risks and enhance rural resilience in Ethiopia.