摘要：Both cross-sectional regression and panel data methods of growth convergence estimation are used to determine the impact of various forms of capital controls on economic growth and growth convergence. We suggest a method to control for measurement error in the cross-sectional regressions based on data quality estimates. The effect of capital controls is found to be strongly dependent on the specific type of control in place, with restrictions on current account payments having the most detrimental effect on growth. A key finding is the importance of interaction effects between capital controls and openness to trade, and capital controls and initial income. The interaction between capital controls and initial income suggests that not only do capital controls impair growth, they also reduce the rate of conditional growth convergence. The omission of these interactions may explain the failure of earlier cross-sectional studies to find effects of capital controls on growth.
