An Oddity in the Human Development Index

Abstract

The Human Development Index is a summary measure of a country’s achievement in key dimensions of human progress. It is estimated using three indicators: life expectancy at birth, average years of schooling, and national income per capita. The aggregate index is calculated, after certain transformations, using their geometric mean to decrease the level of substitutability among those three dimensions. However, the indicator of education is estimated by simply averaging the mean of years of schooling received by adults and the expected years of education for children entering school. This paper shows that the latter variable is often overestimated and fluctuates widely across countries that are similar otherwise. This may distort the HDI by unintendedly magnifying the index in numerous instances, and by making countries with comparable levels of development to be ranked away from each other due to different schooling expectations. To ameliorate that distortion, we propose to estimate the indicator for education using 20-year windows to build a weighted geometric mean that captures countries’ factual advances in schooling two decades later. This paper shows that such a methodological improvement enhances the HDI’s consistency in identifying the levels of human development across countries. JEL codes: I25, I28, I31, O15