During the last century there has been a huge reduction in trade costs at almost every spatial scale impulsed not only by technological advances applied to transport and communication systems, but also by the spread of regional trade agreements and other related schemes. This estringent fall has provoked an explosion of physical, trade and investment integration and, hence, important and lasting effects on the economy at different dimensions. Among them, one that has received special attention from economic literature is the spatial dimension, which indeed is the focus of this article. Since our emphasis is put on countries or 'large spatial scales', where the type of externalities that more likely operates is neither localization nor urban economies but pecuniary external effects, the paper concentrates on NEG models. More specifically, it presents a very up-dated and complete survey of the NEG literature, both theoretical and empirical, on the spatial effects of trade costs on the distribution of economic activity within countries. The NEG framework has successfully evolved thanks to many fruitful contributions and extensions proposed enhancing first-generation models. As a result, there is by now an extensive and rich theoretical literature that examines the role of trade costs in determining the distribution of economic activity. Nonetheless, despite the considerable advances, to date theoretical research still has some limitations regarding relevant issues, such as the application of a general model of monopolistic competition, the discussion on the appropriate treatment of alternative cumulative causation processes at different spatial scales, and the proper inclusion of the transport sector, which is central for thereafter carrying out insightful applied work. With reference to empirical literature, this article finds that the number of papers studying intra-country spatial effects of trade costs has multiplied within the last ten years. The challenge now is, taking theory more seriously, to apply structural specifications, to exploit natural experiments for analysing spatial phenomena and to use innovative research tools - such as spatial econometric techniques and CGE simulations. Moreover, the invitation is to further advance in useful policy-oriented analyses.
