It is causal: revisiting the savings and investment nexus

Domestic savings and investment are positively correlated across countries and through time, as first suggested by Feldstein and Horioka (1980). However, whether this long-lasting correlation implies causation is still an open question. In this paper, we use instrumental variables in a panel setting and show that domestic savings do cause investment in developing economies. In contrast, for advanced countries the statistically significant correlation found in the data seems to result from endogeneity bias. Our identification strategy relies on the idea that age structure influences savings, but not total investment directly. Time-series data patterns for the United States and the theory of Direct Technological Change lend credence to our exclusion restriction.