Modeling Agglomeration and Dispersion in City and Country: Gunnar Myrdal, Francois Perroux, and the New Economic Geography - Critical Essay

American Journal of Economics and Sociology, The,  Jan, 2001  by Stephen J. Meardon

• E-mail
• Print
• Link

STEPHEN J. MEARDON [*]

ABSTRACT. The "new economic geography" is a recent body of literature that seeks to explain how resources and production come to be concentrated spatially for reasons other than the standard "geographic" ones. Unlike alternative explanations of the geographic distribution of industry, the literature is not interdisciplinary. The new economic geography lies well within economics proper: it is an offspring of international trade theory, with models characterized by increasing returns, factor mobility, and transportation costs. The models explain the distribution of industry in terms of the opposition of an agglomerating force, the interaction of transportation costs and increasing returns to scale, with a dispersing force, commonly the interaction of transportation costs and a partially fixed input or output market.

Some authors outside the new economic geography (e.g., Martin 1999) have criticized it as simplistic, irrelevant, or passe. They claim it employs overly abstract analysis, prioritizes mathematical technique over realistic explanation, and is reminiscent of the much earlier works of Gunnar Myrdal and Francois Perroux--in comparison to which, however, it falls short.

This paper investigates the similarities and differences between the new economic geography and the work of Myrdal and Perroux, who in the previous special issue of this journal were ranked by Zafirovsky (1999, pp. 596, 598) as among the leading twentieth century economic sociologists. I examine how the techniques of analysis and intuitive explanations of agglomeration compare between these economic sociologists and the new economic geographers. The paper highlights what has been gained and what has been lost by the new economic geographers, who generally eschew interdisciplinary study.

IN THE PAST DECADE "the new economic geography" has emerged as a means of thinking about and modeling the spatial agglomeration of economic activity. Questions emblematic of the literature are, "Why, as late as 1957, was 64% of U.S. manufacturing employment concentrated in the Northeast and Eastern Mid-West?" (Krugman 1991b, p. 12), or "Why, as late as 1980, was 40% of Mexico's manufacturing employment concentrated in Mexico City?" (Krugman and Livas 1996, p. 138). The questions are answered in terms of interactions between just a few variables and parameters: transportation costs, a production technology featuring increasing returns to scale, and productive factors that are partly fixed and partly footloose.

Questions about the spatial arrangement of markets were not, of course, alien to economics prior to the new economic geography. In the past fifty years (not to mention the past one hundred and fifty) spatial questions have been central to some subdisciplines of economics, among them urban economics and regional science, whose practitioners outnumber the new economic geographers and continue to disseminate their own work in widely circulating scholarly journals. Claims that the new economic geography makes new forays into uncharted territory--or, as more commonly argued (e.g. Krugman 1995), partly charted but subsequently lost territory--are overstated. [1]

Nevertheless the models of the new economic geography have brought greater attention to spatial questions than was previously afforded in general interest economics journals. The models have awakened interest particularly among international economists, for whom one might think spatial questions would come naturally, but who instead have a longstanding tradition of modeling "wonderlands of no spatial dimensions." [2] Indeed it is by way of international economics that the new economic geography has found its way into the general interest journals, not vice versa. The "new international trade theory" of the 1980s gestated the new economic geography of the 1990s, and the latter bears the former's distinguishing features. Both employ the metaphor of a system of simultaneous equations to represent the interdependence of quantities and relative prices; both use the same technical tricks to make increasing returns to scale compatible with equilibria in which firms enter freely and make zero profits. Undoubtedly it is the new economic geography's use of these tools, not only understandable but de rigueur to mainstream economists, that has allowed it to attract widespread attention.

Naturally, attention has incited criticism. Among the most vocal critics are social scientists outside the mainstream of economics, whose largely methodological critiques could just as well be applied to other classes of models in economics. They are applied with particular force to the new economic geography, however, because the fragmentation of geographical economics along methodological lines in the mid-twentieth century left numerous alternative ways of thinking about the subject. [3] Whatever methods the new economic geography brings to bear, it is said they obscure rather than illuminate the relevant determinants of geographical location; whatever insights the new economic geography offers, it is said they were expressed earlier and more profoundly. As Paul Krugman has paraphrased the complaints of his critics, "it's obvious, it's wrong, and anyway they said it years ago" (Gans and Shepherd 1994, p. 178).