In this paper, we revisit the saving and investment nexus as postulated by Feldstein and Horioka (FH) [Econ. J. 90 (1980) 314]. We test for cointegration between saving and investment using the recently developed bounds testing approach to cointegration and derive the long-run elasticities using the autoregressive distributed lag modeling approach for Japan over the period 1960–1999. We establish the unit root properties of the data in the presence of structural break(s) using the Zivot and Andrews (ZA) [J. Business Econ. Stat. 10 (1992) 251] and the Lumsdaine and Papell (LP) [Rev. Econ. Stat. 79 (1997) 212] tests. Finally, we ascertain the direction of causation between saving and investment by using the bootstrap approach. Amongst our key results we find that saving and investment are cointegrated for Japan; investment causes saving and saving causes investment; shocks to saving and investment have a permanent effect; and the long-run coefficient on saving is 0.68, implying a moderate rate of correlation. From the latter finding, we believe that there is no puzzle between saving and investment in the case of Japan, a result contrary to FH (1980).
Field of Research
140203 Economic History 140302 Econometric and Statistical Methods