Can Financial Frictions Account for the Cross-Section Feldstein-Horioka Puzzle?

This paper studies the famous Feldstein and Horioka finding, which is a high correla-tion between long period averages of savings rates and investment rates across countries. We first confirm the Feldstein-Horioka finding with a more recent data set, and then show that a calibrated complete markets model generates a cross-section savings-investment correlation close to zero. Thus, the cross-section Feldstein-Horioka finding is a puzzle to the complete markets model and further research is needed to account for this puzzle. We next explore roles of financial frictions in accounting for the cross-section Feld-stein-Horioka puzzle. The most popular incomplete markets model, the bond model with natural borrowing constraints, cannot account for the cross-section Feldstein-Horioka puzzle. We then propose the bond model with enforcement constraints in which uncon-tingent debt contracts are enforced by the threat of permanent exclusion from the mar-kets. This model generates endogenous borrowing constraints, which capture the incen-tives of countries to repay their debts instead of their abilities to repay under natural bor-rowing constraints, and accounts for the cross-section Feldstein-Horioka puzzle.