From Redding and Rossi-Hansberg:
The observed uneven distribution of economic activity across space is influenced by variation in exogenous geographical characteristics and endogenous interactions between agents in goods and factor markets. Until recently, the theoretical literature on economic geography had focused on styl- ized settings that could not easily be taken to the data. This paper reviews more recent research that has developed quantitative models of economic geography. These models are rich enough to speak to first-order features of the data, such as many heterogenous locations and gravity equation rela- tionships for trade and commuting. Yet at the same time these models are sufficiently tractable to undertake realistic counterfactuals exercises to study the effect of changes in amenities, productivity, and public policy interventions such as transport infrastructure investments. We provide an extensive taxonomy of the different building blocks of these quantitative spatial models and discuss their main properties and quantification.