Saving-investment associations and capital mobility - On the evidence from Japanese regional data

The paper examines the magnitude of the 'saving retention coefficient,' b, in a setting of known near perfect capital mobility, Japanese regions. First, we find that on total regional saving and investment rate data, inclusive of regional government saving and investment, the estimate of b is negative. This negative relationship in the total rates across Japanese regions appears to arise from the strong negative association in the government saving and. Investment rates. Second, on private regional investment and saving rate data, b is insignificantly different from zero. This is evidence consistent with the hypothesis of Feldstein and Horioka. (Economic Journal, 1980) that in a financially integrated economy b would be close to zero.