Economics Letters 59 (1998) 237–242 International capital mobility in OECD countries: The Feldstein–Horioka ‘puzzle’ revisited * Khaled A. Hussein Department of Economics, University of Kent at Canterbury, Canterbury, Kent CT27NP , UK Received 3 September 1997; accepted 20 January 1998 Abstract We utilise the most recent time series techniques of dynamic OLS and examine international capital mobility across 23 OECD countries. We adopt the saving–investment approach where the endogeneity problem is confronted. Our findings show that capital has been remarkably mobile in most countries over the last three decades. 1998 Elsevier Science S.A. Keywords: Cointegration; Feldstein–Horioka puzzle; Saving–investment JEL classification: F32 1. Introduction Over the past two decades, industrial countries have experienced a continuing process of financial market deregulation and capital controls have been liberalised in many economies. By the 1980s, capital movements among major industrialised countries was thought to have reached a very high level (Frankel and MacArthur, 1988). To test the extent of international capital mobility, Feldstein and Horioka (1980) (FH hereafter) examined the correlation between national savings and investment across 16 OECD countries. They argued that with perfect capital mobility there should be no relation between a country’s domestic savings and its domestic investment. An increase in saving in any one country should add funds to the world capital market. These funds would then be shared among countries with favourable investment opportunities. Using cross-sectional analysis FH’s findings showed that 85–95% of national savings is domesti- cally invested and the regression coefficient of savings on domestic investment is insignificantly different from unity. They concluded that the level of international capital mobility is very low and the high correlation between savings and investment has not weakened over time. In this paper, we provide empirical evidence against FH’s findings. Using the recently developed time series techniques of dynamic OLS, where the endogeneity of national savings and investment is taken into account, our findings show that capital is highly mobile in 18 OECD countries. We could only find support for FH’s findings in five cases out of 23. * Tel.: 144 1227 827655; fax: 144 1227 827850; e-mail: k.a.hussein@ukc.ac.uk 0165-1765 / 98 / $19.00 1998 Elsevier Science S.A. All rights reserved. PII: S0165-1765(98)00045-7