Measuring human development: a stochastic dominance approach

We consider a weighting scheme that yields a best-case scenario for measured human development such as the official equally-weighted Human Development Index (HDI) using an approach that relies on consistent tests for stochastic dominance efficiency. We compare the official equally-weighted HDI to all possible indices constructed from a set of individual components to obtain the most optimistic scenario for development. In the best-case scenario index education is weighted considerably more than the other two components, per capita income and life expectancy, relative to the weight that it gets in the official equally-weighted index. It also turns out that the improvement in the official HDI is mainly driven by improvements over time in the education index, the component moving fastest relative to its targets, when compared with per capita income and life expectancy. We find that the best-case scenario hybrid index leads to a marked improvement of measured development over time when compared with the official equally-weighted HDI.