Spatial Equilibria: The Case of Two Regions

In this paper we characterize the set of equilibria in a generalized version of the canonical two-region economic geography model that nests the class of models in Allen and Arkolakis (2014) as well as Krugman (1991) and features an input-output loop. We provide sufficient conditions for uniqueness of equilibria that — in contrast to the well-know result in Allen and Arkolakis (2014) — allow for positive agglomeration externalities, which concentrate economic activity, even in the absence of congestion effects, which disperse it, and highlight the key role played by three additional parameters: the trade elasticity, which regulates the strength of the dispersion force associated with the decline in the terms of trade caused by migration into a region; trade costs, which weaken this dispersion force by limiting trade across regions; and the importance of the agricultural sector, which pushes against agglomeration forces in manufacturing.