Weak and strong cross-sectional dependence: a panel data analysis of international technology diffusion

This paper provides an econometric examination of geographic R&D spillovers among countries by focusing on the issue of cross-sectional dependence. By applying several unit root tests, we first show that when the number of lags of the autoregressive component of augmented Dickey Fuller test-type specifications or the number of common factors is estimated in a model selection framework, the variables (total factor productivity and R&D capital stocks) appear to be stationary. Then, we estimate the model using two complementary approaches, focusing on spatial autoregressive errors and unobserved common correlated factors. These approaches account for different types of cross-sectional dependence and are related to the concepts of weak and strong cross-sectional dependence recently developed in the literature.