Robust multiperiod poverty comparisons

Building on the literature on multidimensional poverty, we propose stochastic dominance tests for multiperiod poverty that allow poverty orderings to be established over time and space that are robust to the chosen poverty measure, poverty line, and aggregation procedure. These tests imply the creation of dominance surfaces for different time spans and testing for significant differences. We elaborate the method first for the bi-dimensional case, using as the dimensions income observed over two periods: One at the beginning and another at the end of a time span. Subsequently, we extend it to the case, where incomes are observed over n periods. We illustrate our approach by performing poverty comparisons by using data for Indonesia and Peru. The discussion is embedded in the literature on chronic poverty.