17 17 17 17 17 RBEE Revista Brasileira de Economia de Empresas RBEE Revista Brasileira de Economia de Empresas RBEE Revista Brasileira de Economia de Empresas RBEE Revista Brasileira de Economia de Empresas RBEE Revista Brasileira de Economia de Empresas Brazilian Journal of Business Economics Brazilian Journal of Business Economics Brazilian Journal of Business Economics Brazilian Journal of Business Economics Brazilian Journal of Business Economics Vol. 3 - nº 1 Vol. 3 - nº 1 Vol. 3 - nº 1 Vol. 3 - nº 1 Vol. 3 - nº 1 Janeiro - Abril 2003 Janeiro - Abril 2003 Janeiro - Abril 2003 Janeiro - Abril 2003 Janeiro - Abril 2003 p. 17-22 p. 17-22 p. 17-22 p. 17-22 p. 17-22 * School of Management and Business, University of Wales, Aberystwyth. † Royal Holloway, University of London. LEFF WAS RIGHT AFTER ALL: NEW EVIDENCE ON DEPENDENCY RATES AND SAVINGS RATES G.Reza Arabsheibani* Merrony Cowling† Abstract: The purpose of this paper is to determine whether a negative relationship exists between dependency rates and savings rates as suggested by a number of empirical studies, including Leff (1969). We re-estimate the savings function as a function of, among other variables, dependency ratios and extend the analysis to the 1990s. Our major finding is that although this negative relationship is not confirmed in the 1980s, as was found by a number of empirical studies, the relationship is negative and significant for the Less Developed Countries in the 1990s. This confirms Leff’s hypothesis. Keywords: dependency ratios, savings, LDCs. Introduction The importance of the relationship between population growth and savings is highlighted by their prominence in models of growth and development, particularly in the Solow (1956) model. For over 30 years, economists have debated whether or not high dependency ratios are an important factor for explaining the large disparity in aggregate savings rates between Developed Countries (DCs) and Less Developed Countries (LDCs). The major reason is that empirical studies have provided mixed evidence on this relationship. Leff’s (1969) study, the first major analysis of the effect of demographic factors on aggregate savings rates, used cross section data for 74 DCs and LDCs. He hypothesised that there is an inverse relationship between dependency ratios and savings rates. The logic behind this hypothesis was that children constitute a heavy charge for expenditure and since they contribute to consumption but not production, a high ratio of dependants to the working age population might be expected to impose a constraint on a society’s potential for savings. His empirical results showed that demographic factors are important determinants of aggregate savings rates and that dependency ratios have a statistically distinct and quantitatively important influence on