Reconciling Observed Tariffs and the Median Voter Model

We investigate the empirical validity of the median voter theory of trade policy. Mayer’s classic framework predicts that the optimal tariff is positive for a capital-abundant country and negative for a labor-abundant country (Mayer, 1984). However, import subsidies are rare so the tariff level prediction seems unrealistic. In this paper we examine two approaches to reconcile the theory with the lack of “pro-trade” bias across countries. First, we test the variation inequality-tariff prediction proposed by Dutt and Mitra (2002). We find support for it using human capital during the 1980s and 1990s. Using physical capital, the prediction is validated for the 1990s but not for the 1980s. Second, we extend the Mayer framework to a large country. The tariff level in a large country is the sum of the median voter component and a positive terms of trade component. We provide empirical evidence for this level prediction and a positive terms of trade component in the 1990s. Using human capital, we find that the median voter component is positive in capital-abundant countries and negative in labor-abundant countries. Consequently, positive terms of trade effects can overcome the median voter component in labor-abundant countries, reconciling theory with observed protectionism.