Gravity Model Estimation: Fixed Effects vs. Random Intercept Poisson Pseudo Maximum Likelihood

Since the work of Feenstra (2002), the standard Anderson & van Wincoop (2003) Gravity Model has been estimated using a fixed effects approach. However, a fixed effects approach has a major drawback: it does not allow for the estimation of exporter- and importer-invariant variables. Thus, economically relevant variables such as exporter and importer gross domestic product are disregarded. Here, we propose a random intercept model to address this gap. This approach not only provides identical estimates to a fixed effects approach, but also allows for the estimation of exporter- and importer-invariant variables