This paper adopts a fixed-effect panel methodology that enables us to take into account both TFP and neoclassical convergence. We use a sample of 76 countries, 1960-2003 and estimate TFP values obtained by using different estimators such as LSDV, Kiviet-corrected LSDV, and GMM à la Arellano and Bond. In our estimates, cross-country TFP dynamics shows that most countries in the sample do not catch up with the USA. We also find conditional convergence in TFP levels and that human capital acts as a robust enhancing factor of technology adoption, as suggested by Nelson and Phelps in 1966. In contrast with previous evidence, in our results even very low level of human capital stocks allow a country to enter a “conditional TFP convergence club” – a result again consistent with the original version of the Nelson-Phelps hypothesis. Further, our results imply a plausible link between stages of development and returns to different levels of education. Finally, the positive influence of human capital on technology catch up is robust to the inclusion of controls for a country’s institutional quality.