Why the Feldstein-Horioka “Puzzle” Remains Unsolved

This paper argues that the 40-year-old Feldstein-Horioka “puzzle” (i.e., that in a regression of the domestic investment rate on the domestic saving rate, the estimated coefficient is significantly larger than what would be expected in a world characterized by high capital mobility) should have never been labeled as such. First, we show that the investment and saving series typically used in empirical exercises to test the Feldstein-Horioka thesis are not appropriate for testing capital mobility. Second, and complementary to the first point, we show that the Feldstein-Horioka regression is not a model in the econometric sense, i.e., an equation with a proper error term (a random variable). The reason is that by adding the capital account to their regression, one gets the accounting identity that relates the capital account, domestic investment, and domestic saving. This implies that the estimate of the coefficient of the saving rate in the Feldstein-Horioka regression can be thought of as a biased estimate of the same coefficient in the accounting identity, where it has a value of one. Since the omitted variable is known, we call it “pseudo bias.” Given that this (pseudo) bias is known to be negative and less than one in absolute terms, it should come as no surprise that the Feldstein-Horioka regression yields a coefficient between zero and one