Abstract
The paper questions the ‘classical’ concept of distance on which interregional interaction models of the linear programming and the gravitation type are based. In contrast to the conclusions drawn from several econometric studies by Gordon inter alia the following hypotheses are suggested: There is no spatial stability in the relationship between flow levels and geographic distance with respect to particular commodity groups because of historical and geographical particularities in connection with product heterogeneity and oligopolistic market structure. If at all, a distance variable may explain the patterns of production and trade on an aggregate level. This factor of distance is neither equivalent to the transportation costs of today, nor does it fully represent the communication costs of today. It rather reflects the influence of the past on the contemporary spatial pattern of production and trade. With respect to international trade the impact of distance is of minor importance compared to the influence of cultural and linguistic affinity of spatial association and political relationship. These hypotheses are supported by quoting the results of several studies of the Institut für Regionalforschung der Universität Kiel arrived at in the past few years. All this research work centred on the question of how the spatial pattern of production and trade was influenced by national borders and geographic distance in the course of the European economic integration process. The final conclusion of the paper is the need for a theoretical reorientation.