Human development index: Dynamic and political factors

Economists have long been interested in measuring the level of development observed among nations and in using such measures as the basis of inter-country comparisons and policy discussions.  Traditional measures of development based upon per capita income and output have long been recognized as quite narrow conceptually relative to broadly accepted development goals.  This perceived deficiency has led analysts to supplement national income statistics with a variety of social indicators measuring comparative achievement in education, nutrition, health and longevity.  In addition, composite indexes such as the physical quality of life index (PQLI) have been developed to combine several social indicators in a measure that can be used to compare levels of development across countries.

The most recent approach is the attempt by the United Nations Development Program (UNDP) in its 1990 and subsequent Human Development Report to combine per capita income with social indicators of education, literacy and longevity to calculate a composite indicator of development called the Human Development Index (HDI). This paper proposes extensions of the HDI that incorporate both political and dynamic growth factors into the basic UNDP framework.  It concludes with an analysis of the empirical performance of the proposed alternative human development measure relative to that of the index used by the UN.

We use data from the World Bank (World Development Indicators) to reproduce HDI as calculated by the UN. Our additions to HDI include the previous 10 year average growth rate of real per capita GDP (also from WDI) as a measure of expansion of opportunity. In addition we use the index of civil and political freedom produced yearly by Freedom House as our measure of political freedom.

We compare the original HDI index with three alternatives that include economic growth, political freedom, and both economic growth and political freedom. Using rank correlation coefficients we analyze the changes that result from these additions to HDI. We compare our empirical results to discussions in the literature of the “best” additions to HDI.