From mine to coast: transport infrastructure and the direction of trade in developing countries

Mine-related transport infrastructure specializes in connecting mines to the coast, and not so much to neighboring countries. This is most clearly seen in developing countries, whose transport infrastructure was originally designed to facilitate the export of natural resources in colonial times. We provide first econometric evidence that mine-to-coast transport infrastructure matters for the pattern of trade of developing countries, and can help explaining their low level of regional integration. The main idea is that, to the extent that it can be used not just to export natural resources but also to trade other commodities, this infrastructure may bias a country's structure of transport costs in favor of overseas trade, and to the detriment of regional trade. We investigate this potential bias in the context of a gravity model of trade. Our main findings are that coastal countries with more mines import less than average from their neighbors, and this effect is stronger when the mines are located in such a way that the related infrastructure has a stronger potential to affect trade costs. Consistently with the idea that this effect is due to mine-to-coast infrastructure, landlocked countries with more mines import less than average from their non-transit neighbors, but more then average from their transit neighbors. Furthermore, this effect is specific to mines and not to oil and gas fields, arguably because pipelines cannot possibly be used to trade other commodities. We discuss the potential welfare implications of our results, and relate these to the debate on the economic legacy of colonialism for developing countries