An Econometric View on the Estimation of Gravity Models and the Calculation of Trade Potentials

This paper presents an AR(1) model in the spirit of Hausman and Taylor zero differences between actual and in-sample predicted trade flows. Large systematic differences between observed and in-sample predicted trade flows only indicate model misspecification and econometric problems. Out-of-sample predictions make sense if countries of interest are in an early stage of the transformation process. The gravity model remains an interesting tool for the calculation of counterfactuals (e.g. the impact of catching up in GDP per capita in the CEEC or of the reduction of trade costs via infrastructure investments or tariffs on bilateral trade, etc.). For this line of research, the difference between short-term and long-term influences could provide interesting insights