Structural stability in a cross-country neoclassical growth model
This paper tests for structural stability of the Solow growth model, as recently extended to human capital and applied to a large section of countries by Mankiw, Romer and Weil. The evidence is obtained by ranking each explanatory variable of the model in ascending order and then running recursive regressions and by then splitting the original sample according to cluster analysis before running separate regressions. The evidence shows that the model exhibits overall structural breaks; the convergence coefficient, however, is very stable; the coefficients of the production factors become unstable where the factors become very abundant or very scarce; the coefficient of labour growth is negative and significant, as required by the theory, only if a small group of countries with scarce labour is considered; the same coefficient is instead positive and significant for a group of countries with abundant labour and favourable initial income and investments; the coefficient of investment in human capital is not significant if abundant labour countries or, simply, if influential countries in the regression are not considered.
No Reference information available - sign in for access.
No Citation information available - sign in for access.
No Supplementary Data.