The new economic geography and policy relevance

Essentially, there are two ways that formal abstract models like those in new economic geography (NEG) can be used for policy analysis. First, formal models can be manipulated to draw out potential 'policy implications'. Second, given these theoretically derived implications, such models can be used to analyse specific policy questions. In recent years, both approaches have attracted attention in NEG work. This article assesses this 'policy turn' in NEG. It argues that the usefulness of NEG models for policy analysis is constrained by the questionable plausibility and credibility of those models. But at the same time, although proper economic geography can claim to be based much more closely on the observation of real-world phenomena, its methods and explanatory accounts are difficult to use for the sort of counterfactual 'what if' type policy analyses found in NEG. Each version of economic geography has epistemological strengths and weaknesses when it comes to policy analysis.