The influence of age composition on saving rates: a meta-analytical approach

Abstract

For a long time economists have held different views on the effect of age composition of population on saving rates. Those who support the idea that the proportion of people outside the labour market adversely affects the saving rate appeal to the life cycle hypothesis of saving. The life cycle hypothesis predicts that people will save when young and spend their savings when old at a constant rate. Therefore the lesser the proportions of the people in the working age group the lesser is the saving rate. Their opponents argue that households have a scope to internalise the effect of family size on consumption with little or no effect on saving and that the retired group in the population might not dissave at the rate predicted by the life cycle hypothesis. This has led to conflicting results when the proposition is tested empirically. This thesis is an attempt to reconcile these conflicting views. The thesis uses meta-analysis to determine whether age composition of the population, measured as dependency ratios, has adverse effect on saving rates. Meta-analysis is a quantitative methodology that uses statistical theories to cumulate quantitative studies. As such it is capable of providing definitive answers to whether the dependency ratio influences saving rates by synthesising all previous studies on the topic. The present study synthesises 34 studies on saving rates and dependency ratios covering over 30 years of research on the topic. The study finds that the dependency ratios adversely affect the saving rate. This validates the claims of the life cycle hypothesis that the proportion of young and old people in the economy is inversely related to the saving rate. However, it is not clear whether or not the magnitude of these effects should be of concern.