2. The geographic concentration of advanced technology adopters
Archived Content
Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.
If some kind of knowledge spillovers across technology adopters are indeed present, then the technology adopters should exhibit a higher geographical concentration than plants overall. To investigate whether or not the above hypothesis is supported by the data, the degree of geographic concentrations of advanced-technology-user plants versus all plants in the manufacturing sector in Canada is examined. As a measure of the degree of agglomeration, we employ the Ellison-Glaeser index of concentration. 4 This index measures the excess concentration beyond that which would be expected to occur randomly. It takes on a value of zero when an industry is as concentrated as one would expect to result from a random location process, and assumes a positive value when an industry is concentrated more than what one would expect to occur randomly.
Figure 1 exhibits the Ellison-Glaeser index of concentration for all manufacturing plants and advanced-technology adopting plants in 1993 for 2-digit Standard Industrial Classification (SIC) manufacturing industries at the economic region level. 5 The grey bar represents the concentration by all plants in each industry. All 2-digit manufacturing industries have a positive value of the index, indicating excess geographic concentration. This excess concentration is not surprising, since the geographic concentration of economic activity is a well-documented fact in the literature (Krugman 1991b, Ellison and Glaeser 1997). What is more interesting, however, is the degree of concentration among adopters of advanced technologies, which is indicated by the black bars. It shows that technology adopters not only exhibit excessive concentration in every industry, but are substantially more concentrated than the overall number of plants for most industries. This fact has never been documented in the literature. 6 While a positive value of the index is not sufficient evidence for the presence of knowledge spillovers among technology adopters, it is a necessary evidence of them. What then would be the explanation for the higher degree of agglomeration among technology adopters? Is it localized-knowledge spillovers across technology-adopting plants, or alternatively, is it other agglomeration economies or 'common environment' effects that attract plants to certain regions that also facilitate technology adoption?
4 . The Ellison-Glaeser index is defined as
is the spatial
Gini coefficient, where xi is location
i 's share of employment in a particular
industry. 
is
the Herfindahl index of the j plants in the industry,
with zj representing the
employment share of the j th plant. Let s1, s2,…, sM be the shares of an
industry's employment in each of M geographic
areas, and x1, x2,…., xM be the shares of total employment in each
of M areas. It corrects for the random concentration
arising from industrial structure that the spatial Gini does
not control for.
5 . See Section 5.2 for a more detailed explanation on economic regions.
6 . Audretch and Feldman (1996) show that innovative activity is substantially more concentrated than overall production, and industries that emphasize research and development tend to be more spatially concentrated. A related result is obtained by Jaffe, Trajtenberg and Henderson (1993), who show that patent citations are highly spatially concentrated.
- Date modified: