Population growth, the dependency rate, and the pace of economic development

Abstract

The analysis of population's impact on the economy has frequently been developed in the context of the dependency-rate argument. The dependency rate, typically measured as the proportion of the total population outside the labour force, is a summary statistic which is intended to capture the influence of a population's age structure on the process of economic growth. Unfortunately, there has been substantial confusion surrounding the economic interpretation of the dependency rate. This confusion derives from the fact that the dependency rate has been used as a proxy for several age-specific influences of population on economic growth. Additionally, for anyone of these influences of population, the impact on the economy will be determined by the particular economic model within which the dependency rate is being analysed. In other words, depending on which age-specific economic aspect of the dependency rate is being examined, and depending on which economic model forms the basis of the analysis, it is possible that an increase in the dependency rate may be associated with either an increase or a decrease in the economy-wide growth of output per head. As a result, the widely used dependency-rate statistic may not be a particularly useful predictor of economic - demographic - growth-rate interrelationships unless the analyst makes explicit his underlying economic framework, and unless the particular economic influences for which the dependency rate is taken as a proxy are delineated.

Additional information

Notes on contributors

Allen C. Kelley

A portion of this paper was presented at a National Bureau of Economic Research Seminar for the Commission on Population Growth and the American Future. I am grateful for the comments on an earlier version of this paper by Professors Richard Billsborrow, Glen G. Cain, Richard A. Easterlin, Charlie Hirsham, Nathaniel H. Leff, Peter H. Lindert, Soren T. Nielsen, Nathan Rosenberg, Julian L. Simon and Jeffrey G. Williamson. A grant from the Population Council facilitated the computational work of this study.