Abstract
Historically, geographers have been keen to adopt bodies of theory from other disciplines in a somewhat uncritical manner. This practice has surfaced again in the recent attempts to apply notions from catastrophe theory to the study of regional development. This paper is an examination of one such application to the modeling of interregional capital flows and critically evaluates it from a number of perspectives. On empirical grounds, there is little evidence that ‘catastrophic’ shifts in the geography of US manufacturing capital have indeed occurred. The presumption of a catastrophic shift is shown to be based on poor empirical indicators of capital mobility or on an incomplete understanding of the process of interregional investment decisions. The paper offers an alternative conception of this process in which the Kaldorian principles of scale economies, specialization, and division of labour are shown to be in continuing operation, albeit now in a spatially discontinuous manner.

This publication has 24 references indexed in Scilit: