Getting Progress Right: Measuring Progress Towards the MDGs Against Historical Trends

Most numerical targets within the framework of the Millennium Development Goals (MDGs) are overly ambitious for the poorest countries when interpreted as country-specifi c goals. As a consequence, the current system undermines accountability and ownership and jeopardizes the public support the MDGs have drawn in the past. This paper proposes an alternative approach to evaluating progress towards non-income MDGs that allows a sensible appraisal of countries' progress. We fi rst estimate transition paths towards high levels of achievement for three MDG indicators (under- five mortality, primary completion, and gender equality in education). In line with previous empirical work, we fi nd that the sigmoid-shaped transition path captures several features of past transition episodes. Accounting only for initial levels and time elapsed, our models explain up to 80 percent of the within-country variation in the data depending on the indicator considered. Estimated transition paths are then used to project progress towards high levels of achievement since 1990. Comparing actual with projected progress allows us to identify over- and underachievers based on realistic expectations. For example, we find that while some countries in Sub-Sahara Africa have in fact shown considerable performance towards low levels of under-five mortality, the bulk of the the countries in that region is still lagging behind. Finally, we provide some preliminary regression results.