Abstract
M.Com. (Development Economics)
This paper investigates the relationship between the age dependency ratio and the savings rate in South Africa for the period 1970-2014. In the process it explores the extent to which the Life-Cycle Hypothesis holds. It attempts to make advances in the field by testing the traditional Life-Cycle Hypothesis using Autoregressive Distributed Lag bounds testing approach and Toda and Yamamoto’s (1995) Granger Causality Test. Diagnostic tests are also conducted, whilst functional stability is tested by applying the Cumulative Recursive Residuals and the Cumulative Recursive Residuals of Square.
The study finds a cointegration relationship between the rate of savings, the age dependency ratio, the growth rate of income per capita, and the real interest rate. The significance of the error correction term establishes that the long-run relationship exists in an equilibrating manner. Confirmation of the relationship between savings rate, the age dependency ratio, and the real interest rate is enhanced by the Granger Causality Test. In particular, a unidirectional relationship between the age dependency ratio, the real interest rate, and the savings rate exists. Cumulative Recursive Residuals tests confirm the long-run relationship between the variables, and reveals the stability of the model.
For the period under study, the Life-Cycle hypothesis does not hold in the case of South Africa, as the age dependency ratio contrasts with the expected negative sign, however statistically significant. The sign of coefficient of the real interest rate also deviated from expectations in this regard, suggesting a negative impact on savings rates. The gross domestic product per capita growth coefficient is not significant in the long run, suggesting the irrelevance of income in explaining savings rate. Therefore, policies that aim to prop up savings through population age structure management will be futile.