The saving-investment correlation puzzle is still a puzzle

A number of simulation studies claim to have solved the Feldstein-Horioka p uzzle by demonstrating that a high time-series correlation between saving a nd investment naturally arises from business cycle shocks, This paper uses panel data of saving and investment controlled for business cycle shocks to empirically test the significance of cyclical shocks - productivity, fisca l and the terms of trade shocks - in explaining a high saving-investment co rrelation. The estimation results reveal that conventional aggregate shocks only partially explain the high saving-investment correlation. Moreover, c ountry differences in the size of the GNP and the non-traded sector do not significantly affect the saving-investment correlation. The saving-investme nt correlation puzzle remains a puzzle after all. (C) 2001 Elsevier Science Ltd. All rights reserved.