41st Congress of the European Regional Science Association: "European Regional Development Issues in the New Millennium and their Impact on Economic Policy", 29 August - 1 September 2001, Zagreb, Croatia
The concept of agglomeration economies, first considered in a systematic (though rather restrictive) manner by Weber, has proven to be an important feature in the analysis of industrial location, whether this is of a theoretical or empirical nature. In either case considerable reliance has been placed on the categories of agglomeration economy proposed by Ohlin. These were termed by Hoover scale economies, localisation economies and urbanisation economies, and were later discussed in some detail by Isard. Leaving aside the reasonable concerns of McCann, who argued that attention should be concentrated on the cost issues underlying agglomeration economies, such as tripartite classification is incomplete in several respects, and therefore represents at best a partial summary. It has recently been argued that the agglomeration economies enjoyed by a firm can be divided into those based on internal economies and those based on external economies, and also that each kind of economy can be considered in terms of scale, scope or complexity. A classification organised around these distinctions subsumes the Ohlin-Hoover classification, and also permits a sharpening of his categories. The concern here is to explore certain implications of this classification, and to examine a number of issues that have probably not received adequate attention. These issues include: the residual nature of agglomeration economies; the possibility of agglomeration without agglomeration economies; and the spatial context of agglomeration economies. It will be argued that such issues need to be addressed if the concept of agglomeration is to be employed effectively in the analysis of industrial location.