ZOLTÁN M. FERENCZI
Humboldt-Universität zu Berlin, Faculty of Life Sciences, Dept. of Agricultural Economics and Social Sciences, Germany
Existing theories about income dynamics of poor and vulnerable households are increasingly being called into question. Neoliberal growth theory cannot adequately explain why income gaps among the poor and the non-poor (be it households on a micro level or entire countries on a macro level) are increasing. Inequalities have also been on the rise not only within societies of developing countries but also in the OECD world, suggesting the existence of multiple equilibria in the process of economic growth and poverty traps: a self"=sustaining cycle of poverty keeping households and potentially entire societies from growth.
In terms of empirical evidence, results have been mixed. The majority of studies do find evidence for the existence of poverty traps; however, some do not. The reasons may be the varying empirical prevalence of poverty trap-like situations across different areas of life, suggesting that, for example, nutritional poverty traps are more significant than education"=based poverty traps. However, different study specifications and poverty measurements might also account for the non"=converging discourse about poverty traps. More traditional consumption and expenditure based poverty measures do not incorporate aspects of health, education, etc. that would be required to assess the more deeply embedded and long term effects of poverty. The Multidimensional Poverty Index (MPI) appears as a viable alternative but systematic studies within the poverty trap debate have been lacking.
The objective of this paper is to contribute to the above stated discourse by testing the observable implications of the poverty traps theory on an existing multidimensional, socioeconomic household panel data set, the Ethiopian Rural Household Surveys 1989-2009 and thereby adding to the existing empirical base of said debate. Traditional consumption based poverty indicators will also be included as controls so that comparisons between uni- and multidimensional poverty measures will be made.
Keywords: Ethiopia, multi-dimensional poverty, poverty measures, poverty traps