Spatial Evolution of Population and Industry in the United States

What features characterize the evolution of the U.S. urban system in this past century? To what extent has the 500-percent increase in national urban population during 1900 – 1990 been accommodated by growth in numbers versus sizes of cities? Given the radical changes in the U.S. economic geography, has the relative size distribution of city sizes changed significantly with, say, strong increases in urban concentration and primacy, a popular notion? As urbanization slowed after World War II, did fluctuations in city sizes and upward/downward mobility of cities dampen, as an Arthur scale economy framework would predict (Brian Arthur, 1990)? Finally, what natural geographic features of a city promote growth? How important are climate, location, neighbor cities, and the like? An interrelated set of questions concerns the agglomeration and mobility of industrial activity within the set of cities. Are industries that are subject to greater scale economies both more agglomerated and less mobile? For industries that are not subject to scale economies, why does agglomeration occur? Central to answering these questions is measuring the nature and extent of scale economies. This paper reports results on these two sets of issues, using Population Census data for 1900–1990 and Manufacturers Census data for 1963–1992 for a selection of machinery and high-tech industries.