Locational Dynamics of Manufacturing in the Atlanta Metropolitan Region, 1968-1976

Southeastern Geographer Vol. XX, No. 2, November 1980, pp. 10(Vl 19 LOCATIONAL DYNAMICS OF MANUFACTURING IN THE ATLANTA METROPOLITAN REGION, 1968-1976 James S. Fisher and Sam Ock Park Locational patterns of manufacturing industries have long been recognized as having a dynamic character. At least three processes, decentralization , diffusion, and dispersen have been identified as integral to these locational shifts in manufacturing. (J) Cohen and Berry utilized a three-fold locational framework for an analysis of manufacturing location changes during the 1950s: (2) Y) Relative location within the city-suburb structure of metropolitan areas. Locational shifts of manufacturing outward within this framework are considered as decentralization. 2)Relative location within the urban hierarchy. Locational shifts of manufacturing within this framework are considered as diffusion and presumably occur downward in the urban hierarchy. 3)Relative location within the heartland-hinterland zones of national space. Locational shifts of manufacturing within this framework are considered as dispersion. The results of the Cohen and Berry study generally confirm the findings of Fuchs and Creamer, but greater attention is given to the intricacies of nationwide change. (3) Of crucial importance in the Cohen and Berry study was the recognition that manufacturing shifts could and do occur simultaneously within each of these locational frameworks. Stated otherwise, it is possible for a large metropolitan area to lose manufacturing firms to nearby centers at a lower level in the urban hierarchy (diffusion), experience suburbanization of manufacturing (decentralization ), and lose manufacturing to hinterlands outside the national manufacturing belt (dispersion). The Cohen and Berry analysis did not treat individual cases but rather manufacturing locational shifts among urban places as a group and by position in the urban hierarchy. Studies of locational change in manufacturing at the metropolitan Dr. Fisher is Associate Professor of Geography at the University of Georgia in Athens, GA 30602. Mr. Park is a doctoral student in the Department of Geography at the University of Georgia in Athens, GA 30602. Vol. XX, No. 2 101 scale also suggest that manufacturing relocation may not be a single process of decentralization common to all large metropolitan areas. For example, Cameron, in a study of new plant location in Clydeside, found that while comparative growth outward from the central city did occur, such growth was not necessarily consistent for all industries. (4) His findings suggest that industries tend to replicate the location pattern of related or like industries, whether centralizing for external economies or decentralizing for low rent peripheral locations. In a study of manufacturing location in the Vancouver area, Steed found the primary contributor to the changing location patterns in that city to be the entry (birth) of new plants; plant migration was secondary. (5) Steed also found that as late as the 1960s the high-cost inner core remained the location of nearly one-half of new industrial plants, despite high land costs. The incubation function and external economies partially explain this phenomenon. Intrametropolitan migration predominantly occurred over short distances, particularly within the inner core. He found relatively little absolute industrial decline in the number of plants in the inner core, but a steady relative decline in comparison with other parts of the metropolitan area stemming from three processes: 1) a decreasing number of sites for new or migrant plants, 2) high outmigration rates, and 3) a great deal of locational flux due to the interacting processes of plant entry, mortality, and migration. The composition of remaining industry was also found to change considerably because of the demise of selected industries, particularly food and beverage plants. Struyk and James, in a study of intrametropolitan industrial location, found enormous mobility rates for industry and increasing decentralization as general patterns, (6) and that the composition of industrial activities and source of net change varied considerably by Standard Metropolitan Statistical Area (SMSA). (7) It seems that while many urban areas have experienced decentralization for some time, there is considerable likelihood that several large cities are only currently beginning to experience this process, primarily because of their "youthfulness" or stage of development. (8) OBJECTIVE. The identification of three processes, decentralization, diffusion, and dispersion, and the seeming variable nature of intrametropolitan manufacturing shifts among different cities which have been the object of...