Do Borders Matter? Evidence from Japanese Regional Net Capital Flows

Abstract

Are international borders barriers to capital flows? We use evidence on net capital flows among regions within a country as a benchmark. For this purpose we develop a data set of saving and investment rates of Japanese prefectures. We find that the correlation between saving and investment rates is higher for OECD countries than for Japanese regions in both time-series and cross-sectional data. After controlling for factors that are expected to contribute to a positive correlation in the absence of barriers to capital flows, we conclude that primarily long-term capital flows are hindered by national borders, as reflected in the cross-sectional evidence.

The Global Side of the Investment‐Saving Puzzle

“…Attempts to uncover common shocks to saving and investment have usually centered on Feldstein–Horioka regressions augmented by domestic variables considered as jointly impacting saving and investment 4 Taylor (1994). and Iwamoto and van Wincoop (2000) propose a different approach and estimate a factor‐free FH coefficient by performing the FH regression on the unexplained residuals from conditional saving and investment regressions.…”

Between Mao and markets: new evidence on segmentation of the bank loan market in China

“…National-and provincespecific shocks that can affect the magnitude of correlation coefficients between deposit and loan rates, other than those owing to the integration of the bank loan market, should be controlled. I, therefore, apply the conditional FH test developed by Iwamoto and van Wincoop (2000) to the Chinese provinces to control for these factors. 4 The steps are as follows.…”

“…Iwamoto and van Wincoop (2000) themselves implement this methodology to examine financial integration across Japanese prefectures.…”

¿Se está rompiendo el mercado español? Una aplicación del enfoque de Feldstein-Horioka

“…Yamori (1995) y Dekle (1996) encontraron unos resultados muy similares para Japón, en los que destacan unos valores de ␤ negativos y signifi cativos que estos autores atribuyen a las políticas redistributivas. Iwamoto y Van Wincoop (2000), y Van Wincoop (2000 encuentran para Japón unos resultados muy diferentes, al mostrar ␤ un valor positivo y signifi cativo. Sin embargo, la interpretación es similar debido a la diferente defi nición de las tasas de ahorro e inversión que hacen estos autores.…”