The contributions of agricultural growth to poverty reduction in Ethiopia

This study investigates agricultural growth-poverty relationships at the national and household levels. A rural household model is used to measure the impact of agricultural growth (or decline) on consumption first and the effect of consumption changes on poverty using regression analyses. Two approaches are used here to estimate the agricultural growth elasticity of poverty. First, the growth elasticity of poverty is determined and then using the agriculture elasticity of growth, the agriculture growth requirement for the change in poverty is estimated indirectly. In the second approach, a two-stage analysis is used to arrive at the growth requirement by estimating first the determinants of welfare (household consumption expenditure) and estimating the poverty impact of a certain growth using the relationship between expenditure and poverty incidence. Accordingly, a forty-year time-series national account data has indicated that a 1 percent growth in agriculture would lead to a 0.32 percent growth in GDP. Given the GDP growth elasticity of poverty, it also followed that a 1 percent increase in agricultural production would lead to a 0.24 percent decline in poverty incidence at the national level. At the Household level, the base simulation has provided a poverty incidence of 40.2 percent in rural areas with 11.2 and 24.9 poverty gap and severity respectively for the year 1995/96. Given the actual and base simulation poverty measures and assuming the percentage change in percapita value added of agricultural production reflects the same percentage change in consumption per capita (adult) and that income distribution remains the same, the application of the same poverty line in real terms4, has provided a poverty incidence of 63.6 percent and a poverty gap of .22 for the year 2000/01. Such a rapid increase in poverty level goes very well with the general perception of the public as expressed during the PRSP consultation and with other studies by non-governmental organizations. Here agricultural growth elasticity of poverty becomes -3.62, which is very high. When inflation is ignored the agricultural growth elasticity of poverty becomes -2.12. Based on another assumption that takes the long-term trend in agricultural growth, poverty incidence and gap for 2000/01 become 46.5 percent 13.7, implying agricultural growth poverty elasticity of - 0.98, or almost equal percentage change in opposite direction. So, it is likely that a one percentage increase in agricultural per capita value added will result into a one percent decline in poverty level of rural households.