Local Versus Global Convergence Across National Economies

This paper reexamines the ability of the Solow-type growth models to explain the pattern of cross-country growth rates. Recent authors, most notably Mankiw, Romer and Weil [1990], have argued that differences in national growth rates are compatible with the view that each country has access to a common, neoclassical aggregate production function. Such models imply that, conditional on population growth and savings rates, disparate economies are converging over time to the same level of per capita output. We argue that cross-country growth is better explained by a model of local versus global convergence. Countries converge locally in the sense that economies with similar initial conditions tend to converge to one another. However, we find little evidence of convergence across economies with substantially different initial conditions as measured by per capita output or literacy rates. Further, the impact of capital formation on aggregate output increases with the level of economic development. These results are consistent with models of multiple equilibria in long run behavior. Our results suggest that the Solow growth model should be supplemented with a theory of aggregate production function differences in order to fully explain international growth patterns.