Essays on spatial international economics.

The essays that constitute this dissertation emphasize the importance of space in understanding the production and trade of goods in the international economy. The first essay, "Monopolistic Competition and International Trade: Reconsidering the Evidence", re-examines the empirical support for a theory of monopolistic competition and international trade advanced by Elhanan Helpman and Paul Krugman. Their theoretical work gave rise to two predictions about the volume and composition of trade that found support in previous empirical studies. I suggest that geography (simply, the proximity of trading countries) might be a critical determinant of the volume and composition of trade not accounted for either by Helpman and Krugman's theory or earlier empirical studies. Correcting for geography calls into question the early empirical support for this theory. In the second essay, "Global Income Patterns: Does Geography Play a Role", I begin by developing the stylized fact that the distribution of national income is spatially clustered: entire continents are relatively rich, others impoverished. I find that there is a strong geographic pattern to the distribution of income growth as well; nearly all of the thirty fastest growing countries over the last three decades are located in the European or Asian near periphery. Next, I perform empirical tests of the pure effect of geography on national income through the use of a Solow growth model augmented to include geographic effects. I find that, while the Solow model explains much income variation, there is an unexplained residual that is correlated with geography. This pure geography effect explains a significant portion of cross-sectional variation in per capita income, even after accounting for factor endowment differences. In the third essay, I develop a model of international trade to explain these patterns. Trade generates external economies due to increasing specialization of intermediate inputs. Introducing transport costs to this model reduces the value of these external economies, so that the scale of the external economies that the firm enjoys depends on its location. This results in a spatial distribution of wages that closely resembles the stylized facts introduced earlier: wages are a decreasing function of distance from economic mass.PhDEconomicsUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/104633/1/9542864.pdfDescription of 9542864.pdf : Restricted to UM users only