This paper considers panel growth regressions in the presence of model uncertainty and reverse causality concerns. For this purpose, my econometric framework combines Bayesian model averaging with a suitable likelihood function for dynamic panel models with weakly exogenous regressors and fixed effects. An application of this econometric methodology to a panel of countries over the 1960-2000 period highlights the difficulties in identifying the sources of economic growth by means of cross-country regressions. In particular, none of the nine candidate regressors considered can be labeled as a robust determinant of economic growth. Moreover, the estimated rate of conditional convergence is indistinguishable from zero.
