Abstract
The objective of this chapter is to analyze the impact of potential agglo- merative economies, which arise from reducing transportation costs through spatial juxtaposition of manufacturing industries, and other factors as identified in Chapter 3, on the geographic locations of firms in selected manufacturing industries, as described in Chapter 4. Results from Chapters 3 and 4 are first summarized, integrated and compared with previous results. Particular attention is paid to the relationship of linkages to geographic associations. Regression analysis for a sample of industries is then used to expand the analysis of the impact of linkages and other factors on geographic associations. In the final section of the chapter the identification of industrial complexes is considered.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Notes
All three of these values are significantly greater than 0 at the.01 level of significance and significantly different from each other at the.01 level. Streit thought that the value of.13 was not statistically significant, and so stated. He apparently used the number of industries rather than the number of different industry pairs as the sample size in computing his test statistics [47: 179].
Another way of analyzing the importance of processing stage for individual industries is to consider the relative importance of input sources and output purchases for the average industry. The average number of input source industries for the 199 manufacturing industries in this study is 94.77 and the average number of industries purchasing an industry’s output is 69.52. This large difference alone suggests that any single output purchaser might be potentially more important locationally than any single input source to an industry. Table 5-4 shows that there are significantly more (at the.05 level) supply links than demand links per industry, however, indicating that more input sources are important to the average industry. This is because the ratio of number of supply links to number of input sources is not significantly different from the ratio of number of demand links to number of output purchasers so that the larger number of input sources is transformed into a large number of supply links. The only factor supporting the importance of output purchasers is that the number of strong demand links is significantly larger than the number of strong supply links.
Klaassen [25b] and Spiegelman [44] also employed regression analysis in their studies of locational behavior but their objectives and approaches were somewhat different and thus their results are not strictly comparable to those of the present study. While the objective of the present study is explicitly to explain geographic associations, Spiegel-man’s was to explain the level of output using characteristics of the region and Klaassen’s was to explain the level of output using characteristics of the industry, including its supply and demand relationships with other industries. Use of these relationships makes Klaassen’s analysis more closely related to that of this study. It is interesting that he, too, finds that interindustry links are important locational variables. Because he examined only three-digit SIC industries for larger regions, direct comparisons of the magnitude of effects is not possible.
Note that only seven of the ten industries are the same in both cases. For industry 134 the sign of L is negative but, because the value of the coefficient is not significantly less than zero at the.05 level of significance and because the theoretical model cannot explain a negative influence of a linkage, it might be regarded as zero.
In the actual calculating of the regression equations the reciprocal of the measure described in Chapter 3 was used.
While the general agglomeration variable is related to the measures of market and material orientation used in this study these variables are not highly colinear. G is based on the correlation between regional employment for a specific industry and total regional employment while M and R utilize the correlation between regional employment in an industry and total regional population. Total regional population is highly correlated with total regional employment. G is not highly correlated with M or R, however, because the latter two variables use additional information that is not correlated with total regional employment.
The inclusion of only variables with coefficients significantly different from zero means that the estimated equations for industries 67, 101 and 134 are identical to those reported in Table 5-9 for Model 2.
Leontieff has suggested a means of identifying the hierarchal structure of industry by ‘triangulation’ of an input-output table [27: 4]. In this process, material oriented industries are listed first, then the industries linked to them, and so on until the market oriented industries might be omitted from the complex. Links to industries in the study but not in with specifying industry orientations uniquely.
This statement applies to the 199 industries used in the study. Reciprocal relationships with industries not in the study might exist; but the only error possible is that some industries might be omitted from the complex. Links to industries in the study but not in the complex could be important for some not shown. However, it remains true that among those industries in the complex, each might be interested in the location of another in it which might reciprocate this interest. For example, a firm producing electric instruments would be interested in being close to a firm producing electronic components and the latter would like to locate close to the former. The presence of both of these would make a location highly attractive to a firm producing communications equipment.
Note that other practitioners have allowed the same industry to be part of more than one complex so that it is not necessary to allocate each industry to only one complex.
Stevens, Douglas and Neighbor report that they attempted a factor analysis approach to the identification of complexes several years prior to 1969. Unfortunately they do not specify what the results of the attempt were [46: 21].
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1976 H. E. Stenfert Kroese B.V., Leiden
About this chapter
Cite this chapter
Latham, W.R. (1976). The impact of linkages on industrial location. In: Locational behavior in manufacturing industries. Studies in applied regional science, vol 4. Springer, Boston, MA. https://doi.org/10.1007/978-1-4613-4369-1_5
Download citation
DOI: https://doi.org/10.1007/978-1-4613-4369-1_5
Publisher Name: Springer, Boston, MA
Print ISBN: 978-90-207-0638-3
Online ISBN: 978-1-4613-4369-1
eBook Packages: Springer Book Archive