Latent heterogeneity to evaluate the effect of human capital on world technology frontier

Abstract Although human capital has been recognized as playing important role in spurring productivity growth, its empirical effect remains ambiguous due to the possibility of latent heterogeneity. To reveal impact this driver economic we propose an alternative methodology, robust frontier non parametric location-scale models for accommodating simultaneously problem model specification uncertainty, heterogeneity and cross-section dependence modelling technical efficiency. We estimate a nonparametric define world technology 40 countries over period 1970–2007. Conditional versions enables us investigate external factors on production process. One these will allow measure openness economy country second be factor linked absorptive capability hence country. have adjust methodology take into account panel structure our data set, i.e., handling time dimension cross-sectional affecting Our findings prove that plays accelerating technological catch-up (increase efficiency) but not changes (shifts frontier). This result seems confirm theoretical hypothesis benefit from new (technological catch-up) only when they ability exploit it, high level capability.