Political preference functions,trade wars,and trade negotiations

The integration of elementary political considerations into computable general equilibrium models is considered, and an extended illustration to agricultural trade negotiations provided. The application involves an evaluation of the payoffs of alternative support levels to agricultural and non-agricultural interests in the EC and the US. A government political preference function for each region is calibrated as a CES aggregation of the payoffs to the two interest groups, with weights corresponding to their benchmark political influence. The political preference function is presumed to be employed by each government to determine the level of agricultural support. The analysis illustrates how sensitive such computable general equilibrium models can be to elementary political considerations, mainly due to the flatness of the implied Pareto frontier. It also illustrates how one can modify the traditional political preference function approach to accommodate possible convexities of the Pareto frontier in empirical models.