Estimating the extent of poverty is a preliminary task before implementing any anti-poverty project. This can be done by creating a holistic individual deprivation index for different life dimensions. Currently, there are a growing number of multidimensional poverty index (MPI) studies. However, there exist only few empirical papers for Ethiopia that used country-specific well-being indictors in a panel data framework. This paper estimated the MPI for Ethiopia using the Ethiopian Rural

more »

... Panel Data Survey (ERHS) for the years 2004 and 2009. The study used five dimensions and nineteen country specific well-being indicators to estimate MPI. The MPI estimation with the counting approach showed that the percentage of multidimensional poor people for the years 2004 and 2009 were 35% and 25% respectively. The paper decomposed MPI across regions, household sizes, and depth and breadth of poverty. The MPI decomposition across time estimate confirmed that the large decrease in the MPI was the result of a significant decrease in the number of poor households among the middle and the large families than a decrease in the number of deprivations. The panel MPI decomposition also showed that, once a household slips into poverty, the probability of exiting from it is very low. Child mortality contributed more for the decrease in the breadth of poverty. Moreover, the decrease in asset deprivation contributed the largest for the decrease in MPI across the two periods. The use of country-specific indicators such as land holding and crop stored for agrarian economy are more likely to identify poor people than indicators used in internationally comparable MPI estimation. This panel based MPI estimation clearly showed the progress or regresses of the household in achieving a particular indicator over time.