An Alternative Interpretation of Conditional Convergence Results

A study examines 2 of the most widely used control variables in conditional convergence regressions, the investment-to-GDP ratio and the population growth rate, and finds that these variables do not appear to be exogenous with respect to growth. The investment-to-GDP ratio rises and the population growth rate declines with income growth. The endogeneity of the control variables causes a negative bias in the commonly used regression coefficient for the conditional convergence test. The bias appears sufficiently important that the conditional convergence result is not found when alternative regressions are designed to avoid the bias.