背景
It is a strange world when economists try to become geographers and geographers try to be something else.

Economists have rediscovered agglomeration and geographical distance as a fundamental driver of the location of economic activity.
Geographers, at least since the contribution of Vidal de la Blache (1917), dominated by observing and integrating all human activity developed in close geographical proximity, seem to have become liberated from the yoke of analyzing interaction in geographically-constrained spaces and have put their sights into exploring relationships which need not be limited by physical contiguity.

These two contrasting approaches are also contributing to the emergence of different policy paradigms, with mainstream economists advocating spatially-blind policies and geographers and some non-mainstream economists vaunting the virtues of place-based approaches.

地理的进展
When the literature on the ‘death of geography’ emerged, geographers were predictably outraged. Any view that envisaged the ‘end of geography’ (O’Brien, 1992), the ‘death of distance’ (Cairncross, 1997) or the advent of a ‘flat world’ (Friedman, 2005) was perceived as a threat to what the discipline of geography is about. Numerous articles have been written criticizing these views and pointing out that economic activity cannot happen anywhere (Cambridge Journal of Regions, Economy and Society, 2008; McCann, 2008, Special Issue).

How has this disciplinary migration come about? Economic geography in the 1980s and 1990s was very much about ‘neo-Marshallian industrial districts’ (Becattini, 1987), ‘new industrial spaces’ (Scott, 1988), ‘innovative milieus’ (Aydalot, 1986), ‘learning regions’ (Morgan, 1997) or ‘regional innovation systems’ (Cooke et al., 1997).

NEG
Transport costs provide an advantage to local firms in imperfectly competitive markets, and increasing returns transform this advantage into profits and industry expansion.

In parallel to developments in NEG, urban economists have broadened their scope to incorporate a variety of localized interactions. The systems of cities literature, pioneered by Henderson (1974), deals with the endogenous formation of a system of cities and exchanges between them.
A first, empirical, strand in this literature has made substantial progress in quantifying the productive advantages from locating in close geographical proximity to other firms and workers.
A second strand, with both theoretical and empirical components, has tried to improve the understanding of the causes of the advantages of urban density (Duranton and Puga, 2004).

Similar geographies, different explanations and different policies
观点不同
For geographers, the dominance of large metropoli derives from the relational proximity among economic agents. These relations create their own multifarious space within their interactive processes (Garretsen and Martin, 2010, 141–42).
For economists, density and contiguity are the determinants of economic agglomeration.
Economists thus seem to be replacing geographers as those who can explain better why economic activity has a tendency to agglomerate in reduced geographical spaces. But geographers may well have taken the lead in explaining why interaction can also take place regardless of geographical distance. Economists increasingly excel at analysing what happens within the boundaries of any given region, while geographers at identifying what happens between areas and beyond the region. Geographers are probably forgetting what drives economic activity to agglomerate in urban areas, as they no longer assume that all forms of proximity come together somehow mysteriously in agglomerated areas, while economists are struggling to acknowledge that agglomeration is not just the result of backwash effects in the immediate neighbourhood of cities, but also of the interaction among agglomerated spaces, often in far away locations.

政策启示不同
For economists, the solution for development problems are basically ‘one-size-fits-all’ ‘spatially-blind’ approaches; they can be applied anywhere regardless of context (Garretsen and Martin, 2010, 136).
For geographers, space is relative and variable and this makes context king. The multiplicity of interactions occurring at diverse geographical scales and the variegated spatial forms they generate mean that not only ‘one-size-fits-all’ approaches are anathema, but that even isolated place-based approaches (e.g. Barca, 2009; OECD, 2009), may be close to meaningless. Under a relational conception of space, the concept of space is embedded in the process and multiple processes may happen simultaneously. Hence, only a combination of different place-based policies may work in promoting

具体还可以看这篇文献的讨论。https://bbs.pinggu.org/thread-3057711-1-1.html


两门学科怎么合作
三个方面Where should this cross-disciplinary interaction come from? There are three potential areas for collaboration between geographers and economists: methods, research questions and topics and policy.

There is a need to combine, from a theoretical and empirical perspective, the views of economists and geographers about whether these policies are complementary and under which circumstances they may work together. This will inevitably lead to testing whether (i) place-based and (ii) spatially-blind policies deliver and, ultimately, (iii) whether place-based policies can also be ‘people-sensitive’ (i.e. whether they create differential impacts on individuals according to specific characteristics, such as income, age, gender, health, race and ethnicity or religion) and (iv) whether people-based policies are ‘place-sensitive’ (whether they affect different places in different ways).

推荐大家阅读 本文的第5节。