Cross-country growth regressions: problems of heterogeneity, stability and interpretation

Abstract

The paper discusses the issues of heterogeneity and stability of cross-country growth regressions that have been used to study the problem of convergence. Almost all studies use pooled regressions. The paper considers the issue of pooling under heterogeneity using a hierarchical Bayesian method and estimates growth regressions for different panels studied in earlier papers, and different regimes. The conclusion is that the convergence rates are higher than those obtained from pooled regressions under the assumption of homogeneity and that there is instability over time in the relationships.