Agglomeration and location of manufacturing activity: the theoretical framework

  • Gerald A. Carlino
Part of the Studies in applied regional science book series (SARS, volume 12)


The primary purpose of this chapter is to articulate the theoretical framework upon which the empirical analysis, in the next chapter, will be conducted. The view put forth is that indivisibilities in production are the main forces leading to scale economies in the widest sense, and agglomeration of manufacturing activity. Since production function analysis is a technique consistent with this theoretical scheme, a variant of the CES production function is proposed to measure scale economies. A two-fold approach will be employed: (1) time-series analysis will be utilized to estimate agglomeration or scale coefficients directly for each two-digit SIC industry in a given SMSA via production function techniques; (2) these scale coefficients will then be employed as a dependent variable in a cross-sectional industry-by-industry regression model designed to decompose the agglomeration measure into: (a) internal scale economies; (b) localization economies; (c) urbanization economies; and (d) urbanization diseconomies.


Production Function Capital Stock Demand Curve Scale Economy Urbanization Economy 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Actually, the shape of the demand curve depends upon a number of conditions as demonstrated by Greenhut and Ohta (1975). They show that the free spatial (regional) demand curve is convex to the origin regardless of the shape of the individual demands which make it up. The shape of the spatial competitive market demand curve depends upon the behavioral assumptions used in the competitive model. The linear demand curve drawn in fig. 3.3 is constructed as such for convenience.Google Scholar
  2. 2.
    The scenario described above is overly simplified to highlight the main features of the Lösian system. Questions regarding the entry of new firms when pure profits exists such as: how many, where will they locate, what impact will these new firms have on the location, size and shape of existing firms, and ultimately what will the equilibrium configuration of firms in space look like, have been ignored. For a more complete discussion of these and related questions see Segal (1977).Google Scholar
  3. 3.
    Robinson (1953–4:47) For an excellent discussion of the Cambridge Capital Controversy see Harcourt (1972), and Harcourt and Laing (1971).Google Scholar
  4. 4.
    Actually, Equation 3.8 above which generates Equation 3.9 is based upon the assumptions in 3.5 and 3.6 provided that 3.9d holds as well.Google Scholar
  5. 5.
    Dhrymes’ model has been employed by Shefer (1969; 1973) to investigate localization economies in two-digit manufacturing industries in the U.S. See the previous chapter for extensive criticisms of the Shefer analysis.Google Scholar
  6. 6.
    In much the same fashion a potential type variable for localization economies for the i-th industry in j can be defined. This variable would account for the intra industry agglomeration economies do to the presence of the same industry in nearby SMSA’s. An example here is the potential localization economies for establishments in primary metals in the Pittsburgh and Youngstown SMSA’s due to the large concentration of this industry in these areas. This variable, however, is subject to the same problems as the urbanization economies potential proxy described above.Google Scholar

Copyright information

© H. E. Stenfert Kroese B. V., Leiden 1978

Authors and Affiliations

  • Gerald A. Carlino
    • 1
  1. 1.University of MissouriKansas CityUSA

Personalised recommendations