Trade and Cities

Abstract

Many developing countries display remarkably high degrees of urban concentration, incommensurate with their levels of urbanization. The cost of excessively high levels of urban concentration can be very high in terms of overpopulation, congestion, and productivity growth. One strand in the theoretical literature suggests that such high levels of concentration may be the result of restrictive trade policies that trigger forces of agglomeration. Another strand in the literature, however, points out that trade liberalization itself may exacerbate urban concentration by favoring the further growth of those large urban centers that have better access to international markets. The empirical basis for judging this question has so far been weak: in the existing literature, trade policies are poorly measured (or not measured as when trade volumes are used spuriously). Here, we use new disaggregated tariff measures to empirically test the hypothesis. We also employ a treatment-and-control analysis of pre- versus post-liberalization performance of the cities in liberalizing and non-liberalizing countries. We find evidence that, controlling for, among others, largest cities that have ports and, thus, have better access to external markets, liberalizing trade does lead to a reduction in urban concentration. Finally, by using a cross-country level of analysis we provide some external validity to the more careful empirical studies that rely on single country data.