Knowledge Spillovers in High Technology Agglomerations: Measurement and Modelling

  • Elsie Echeverri-Carroll
Part of the Advances in Spatial Science book series (ADVSPATIAL)


Malecki (1980) observed that one of the characteristics of innovative or high-tech firms is that they tend to cluster in relatively few places. In the United States, for instance, a large proportion of the high tech industry is concentrated in Silicon Valley. Indeed, if we use patents to measure innovative activity in the United States, we see that about 50 percent of U.S. patenting activity occurs in only six states: California [where Silicon Valley is located], New York, Texas, Illinois, Michigan, and New Jersey.


Knowledge Spillover Knowledge Externality Agglomeration Economy Efficiency Wage Cumulative Growth 
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.
    Dixit and Stiglitz (1977) derive demand functions from preferences [utilities] that exhibit a love for variety. Thus, for a given level of consumer spending in a product and a given price for the available varieties, welfare [consumer utility] rises as the number of varieties increases. The increase in utility is conditioned by the degree to which consumers desire variety.Google Scholar
  2. 2.
    The concept of iceberg transportation costs, introduced by Samuelson (1954) in the international trade literature, assumes that a fraction of any good shipped simply ‘melts away’ in transit, so transport costs are in effect incurred in the good shipped.Google Scholar
  3. 3.
    As Bona and Santos (1997, p. 243) pointed out, ‘While the arsenal of techniques applicable to the analysis of non-linear systems has recently grown enormously, it must be acknowledged that our collective hands are often tied when confronted with what appear to be simple non-linear problems’.Google Scholar
  4. 4.
    Where τ= 1/T < 1, with T[= quantity dispatched /quantity received] >1.Google Scholar
  5. 5.
    The DTM includes more than 90 percent of manufacturing establishments in Texas that have more than 10 employees, and more than 50 percent of those with fewer than 10. Because we expect to find most innovative high-tech firms in establishments with at least 10 employees, the DTM is an excellent database for our study.Google Scholar
  6. 6.
    The following industries [by 4-digit SIC code] were surveyed in these metropolitan areas. Austin: 3544, 3672, 3679, 3823, 3674, 2834, 3571, and 3842. Fort Worth: 3728, 3533, 3535, 3679, 3069, 3544, 2899, 3721, 2834, 3674. Dallas: 3544, 3661, 3721, 3728, 3674, 3679. Houston: 3533, 3823, 2899, 3561, 3569, 3511, 2821, 2869, 3571. San Antonio: 3728, 3544, 3679, 2899, 3674, 3537, 2842, 3721, 3531, 2834.Google Scholar
  7. 7.
    TDM involves a sequence of mailings and follow-ups designed to increase the response rate. Although TDM involves a fourth mailing of a letter and replacement questionnaire to non-respondents by certified mail [49 days after the initial mailing], this step was not performed due to time and funding restrictions.Google Scholar
  8. 8.
    The term maquiladora is derived from the Spanish word for the amount of corn paid by a farmer to the miller to grind the corn. Similarly, the maquiladora industry uses inputs provided by the client and returns the output to the same client.Google Scholar
  9. 9.
    Manfred Fischer highlighted this important point during my presentation of this chapter at the workshop on Knowledge, Complexity, and Innovation Systems, held in Vienna, July 1–3, 2000.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2001

Authors and Affiliations

  • Elsie Echeverri-Carroll
    • 1
  1. 1.The Red McCombs School of BusinessThe University of Texas at AustinUSA

Personalised recommendations