This paper proposes a new measure and test of international capital mobility. In order to investigate capital flows, we utilize a small open economy version of Campbell and Mankiw's [NBER Macroeconomic Annual (1989) 185-216] permanent income model. Our capital mobility measure focuses on the correlation between the country's consumption and net output instead of the correlation between domestic saving and domestic investment proposed by Feldstein and Horioka [Economic Journal 90 (1980) 314-329]. The model is estimated for 11 OECD countries. The results suggest that the null hypothesis of perfect capital mobility cannot be rejected for more than a half of these countries. On the other hand, some kind of capital mobility constraints are detected for several countries including the United States and Japan. (C) 1998 Elsevier Science Ltd. All rights reserved. JEL classifications: E21, E22, F21, F32, F41.