Printer Friendly

Spillover asymmetries and a comparative technological advantage.

This paper offers a framework for examining technology spillovers extant between research teams. A policy-relevant finding is that firms' research organizations must strive to absorb foreign technology. The result of a technology spillover imbalance or asymmetry is that from a country or policy perspective, even for countries sustaining a superior number of world-class research teams, a comparative technological advantage may be lost if the domestic R/D teams fail to absorb and exploit technology as aggressively as foreign rivals.

I. Introduction

Study of the effects of technology on international trade and firm competitiveness grew to prominence during the 1960's. Early in the decade Posner [1961] argued that new products and processes which afforded innovators monopoly rents helped explain international gaps in technology and the structure of trade. Shortly thereafter Vernon [1966] added skilled labor and mass production requirements to explain how a comparative advantage would shift to the less technological advanced, labor-inexpensive countries as products cycled toward maturity. Empirical support of the Posner and Vernon (product-cycle) model quickly followed [Gruber, Mehta and Vernon, 1967].

During the present decade much of our understanding of the international effects of technology has come directly from the economics of technology literature, where innovation, diffusion, patents and policy have been major themes. It has been often noted, for example, that high-technology firms differ widely in their ability to develop and commercialize technology quickly and efficiently. However, recent evidence suggests the difference may be more international than national, and hence a source of technological advantage.

In a recent survey of 200 firms, Mansfield [1988 a,b,c,] controlled for external and internal technology factors and found U.S. and Japanese firms generally comparable in innovative time and cost when exploiting internally developed technologies, but U.S. firms lagging in use of externally-based technologies. On average in the latter case, Japanese firms took 10% less time and 50% less money than their U.S. competitors. Thus, Mansfield's empirical findings suggest international spillover of technology are asymmetric, thereby corroborating numerous anecdotal accounts of U.S. and European firms suffering the NIH (not-invented-here) syndrome [cf. NAS, 1987; Brandin, 1987], and of Japanese firms pursuing foreign technology more aggressively and efficiently than their rivals. [e.g., Charles, 1989; Cutter, 1989; Herbert, 1989].

An asymmetry in spillovers complicates measurement of a nation's technological advantage. Conventional estimates of a nation's technological strength (for example in inventiveness or innovative potential) requires the use of proxy variables. Notwithstanding long-cited shortcomings in their use [Basberg, 1987; Gold, 1981; Narin, 1976; Cormanor and Scherer, 1969], the number of scientists/engineers, patent applications, citations, research centers, R/D funding are the oft-cited variables selected as proxies for technological measurements. This paper suggests numerical counts may be misleading; it determines the conditions under which a country, having superior strength in the number of research teams, has in fact relative technological inferiority owing to spillover asymmetry.

II. Spillover Framework

Steel's [1989] Communication Network (for analyzing the leakage of secrets) and Kacker's [1990] Interactive Graph (for planning fractional factorial experiments) suggested a framework for analyzing internal and external spillover characteristics. Technology spillovers between six research teams are depicted in Figure I as line segments. Each segment represents a potential bilateral flow of technology. Assume each R/D team, identical in size and quality, has the capacity to absorb (+ 1) units of external technology from each of its rivals for a total of (+ 5) technology-units.

The general case summarized in Table 1 is easily derived from Figure I. If country 1 is comprised of [U.sub.1] R/D teams, the total absorptive intracountry external benefit is [U.sub.1]([U.sub.1] - 1). Similarly the spillover benefit to country 2 from the aggregate of intracountry research teams is [U.sub.2]([U.sub.2] - 1), and from R/D teams within country 3 of [U.sub.3]([U.sub.3] - 1). All intracountry external benefits from the NW-SE diagonal of Table I.


Each R/D team in country 1 also receives an absorptive benefit of (+ 1) from each R/D team in country 2, hence [U.sub.1][U.sub.2] appears in the second row of the first column. All intracountry external benefits are similarly determined as the product of the number of research teams within the two countries.

The internal technological strength of a research team is given as a multiple (s) of external technology units of magnitude (+ 1). Hence country 1 has an internal technological strength of [U.sub.1]S, country 2 of [U.sub.2]S units, etc.

Let U = [U.sub.1] + [U.sub.2] + . . . . [U.sub.n]. Country 1's total technological strength from external and internal sources of [U.sub.1] (U - 1) + [U.sub.1] S can be derived logically as an independent calculation, or as the sum of terms in column 1 of Table I.

III. Spillover Asymmetry Attributable to

Country 1

Assume country 1 has the greatest number of identical R/D teams and country 2 the second greatest number such that [U.sub.1]>[U.sub.2]>[U.sub.3]. ... >[U.sub.n]. Under what conditions will country 1's failure to absorb technology from country 2 be sufficient for it to lose its technological superiority? Country 1 is technologically inferior to country 2 if:

[U.sub.1]([U - 1) + [U.sub.]S - [U.sub.1][U.sub.2]

<[U.sub.2](U - 1) + [U.sub.2]S or

([U.sub.1] - [U.sub.2])/[U.sub.2] < [U.sub.1]/([U + S - 1

The leftmost terms represents country 1's numerical superiority of research teams over country 2, expressed as a percentage. The righthand term is the technological strength of country 1 relative to the worldwide external benefit and s, the internal-external multiplier.

To illustrate, assume that out of 21 research teams worldwide (U = 21), countries 1 and 2 have 10 and 7 R/D teams ([U.sub.1] = 10; [U.sub.2] = 7) and that the internal-external technology strength factor is ten (S = 10). Country 1 is technologically superior since:

([U.sub.1] - [U.sub.2])/[U.sub.2] = (10 - 7)/7

However, if country 2 gains one more research unit, country 1, despite its 5:4 numerical advantage in R/D teams, becomes technologically inferior to country 2:

([U.sub.1] - [U.sub.2])/[U.sub.2] = (10 - 8)/8

>[U.sub.1]/(U + S - 1) = 10/30

The curvilinear line (quadratic) shown in Figure II shows technological parity for country 1 and 2 for various combinations of [U.sub.1] and [U.sub.2]. In the example case illustrated, both the internal-external technology factors (S = 10), and the number of R/D teams operating in countries outside of country 1 and 2 (U less [U.sub.1] and [U.sub.2] = 6) are assumed constant. Combinations of [U.sub.1] and [U.sub.2] for which country 1 is numerically superior to

country 2 in R/D teams, but technologically inferior owing to spillover asymmetry, is shown as a shaded area in Figure II.

The example plotted in Figure II reveals that twelve R/D teams absorbing no technology from country 2 (but absorbing technology from third countries) are needed in country 1 to match the technological strength of nine researching teams in country 2. A smaller internal-external technology factor (s) would ceteris paribus, aggravate the imbalance.

To the extent technology spillovers (e.g., published papers, conference lectures, cross-firm recruitment) imperfectly communicate a rival's R/D activities, research will be replicated. The greater the number of research teams, the greater the amount of replication. But replication means spillovers from four firms aren't likely twice that from two. And therefore for an individual country replication means that even if the number of spillover paths increases by the square of the number of researching units, the overall spillover benefit probably does not increase by [U.sub.n](u - 1) as we have assumed, but increases at a diminishing rate. Hence the curvilinear technology-parity line given in Figure II likely has less curvature than that illustrated in the example.

On the other hand, working to further increase the curvature is the geographical proximity of the researching units. Jaffe [1985, 1986] found, for example, that clusters of firms in close proximity to one another receive more patents per R/D dollar spent, have greater productivity growth, and more heavily invest in R/D than the average firm in their class.

Neither are spillovers binary variables, either present or not present, as has been assumed for the asymmetry condition. And as indicated, even a small amount of spillovers imply replication of research, and therefore curvature to the technology-parity line of Figure II. Despite these apparent qualifications, several useful policy-relevant observations can be drawn.

Policy Observation 1:

The more a developing technology resides within just two countries, the greater the numerical superiority in R/D teams the under-absorbing technological leader must have to maintain its leadership

Policy Observation 2:

The more a developing technology resides within just two countries, the easier it is for the under-absorbing technological leader to lose its technological superiority even while maintaining a superior number of R/D teams.

Rationale: the righthand term of the inequality condition for technological inferiority becomes larger, allowing the inequality to hold for even greater percentage differences in ([U.sub.1] - [U.sub.2])/[U.sub.2]

Policy Observation 3:

During the emergence of a new technology, as the number of R/D teams in all countries grows (we assume proportionally to the initial R/D team distribution), it becomes increasingly easy for the under-absorbing technological leader to lose its superiority.

Rationale: assume country one is (just barely) superior viz. ([U.sub.1] - [U.sub.2]) / [U.sub.2] > [U.sub.1] (U + S - 1). As [U.sub.1], [U.sub.2], U increase, only the righthand term increases (owing to S - 1), reversing at some point in time the inequality. Logically, the number of spillover linkages between R/D units increases by the square of the number of R/D units, hence as the number of R/D teams grows the effects of spillover asymmetry became more pronounced and detrimental to the under-absorber.

IV. Conclusion

The MIT commission on Industrial Productivity and The National Academy to Engineers have recently urged U.S. firms to forge closer and more cooperative ties to domestic competitors and to exploit more fully the technological breakthroughs of foreign rivals. [MIT, 1987; NAS, 1987] As the foregoing discussion suggests, these seemingly bland recommendations should be heeded, particularly if we are influenced by simple comparative counts of the number of researching teams or the number of scientists and engineers as measures of technological strength or superiority. Even if domestic firms command a numerical advantage, the national perspective of technology-linkages noted above reveals the ease by which a technological superiority may vanish, if domestic firms fail to absorb and exploit foreign technology to the same extent as foreign competitors. Hence even in research, domestic firms less aggressive than their foreign rivals do so at their own peril.


Basberg, B. J., "Patents and the Measurement of Technological Change: A Survey of the Literature," Research Policy, 16, pp. 131-141. Brandin, D. H. and Harrison, M. A., 1987, The Technology War: A Case for Competitiveness (John Wiley and Sons, New York) p. 24. Charles, D., 1989, "Congress Attacks Foreign Research Ties," in Perspectives, Research-Technology Management, Sept-Oct, p. 4. Cormanor, W. S. and Scherer, F. M., 1969, "Patents Statistics as a Measure of Technical Change," Journal of Political Economy, 77, pp. 392-398. Cutler, R. S., 1989, "A Survey of High-Technology Transfer Practices in Japan and in the United States," Interfaces, 19, pp. 67-77. Gold, B., 1981, "Technological Diffusion in Industry: Research Needs and Shortcomings," The Journal of Industrial Economics, 29, pp. 247-269. Gruber, W. Mehta D. and Vernon R., 1966, "The R/D Factor in International Trade and Investment of United States Industries," Journal of Political Economy. Herbert, E., 1989, "Japanese R and D in the United States," Research-Technology Management, 32, pp. 11-20. Jaffe, A., 1985, "Quantifying the Effects of Technological Opportunity and Research Spillovers in Industrial Innovation," Ph.D. Thesis, Harvard University. Jaffe, A., 1989, "Technological Opportunity and Spillovers of R/D: Evidence for Firms Patents, Profits and Market Value," American Economic Review, 76, pp. 984-1001. Kacker, R. N. and Tsui, K. L., 1990, "Interaction Graphs: Graphical Aids for Planning Experiments," Journal of Quality Technology, 22, pp. 1-14. Mansfield, E., 1988a, "Industrial R/D in Japan and the United States: A Comparative Study," American Economic Review, 78, pp. 223-228. Mansfield, E., 1988b, "The Speed and Cost of Industrial Innovations in Japan and the United States: External vs Internal Technology," Management Science, October, pp. 1157-1168. Mansfield, E., 1988c, "Industrial Innovation in Japan and the United Stats,' Science, 241, pp. 1769-1774. MIT Commission in Industrial Productivity, 1989, Made in America: Regaining the Productive Edge, (MIT Press, Cambridge, MA.) Narin, F., 1976, Evaluation Bibliometrics: The Use of Patent Statistics and Citation Analysis in the Evaluation of Scientific Activity (Contract NSF C-627, National Science Foundation, Washington, D.C.). National Academy of Engineers, 1987, Strengthening U.S. Engineering Through International Cooperation (U.S. Government Printing Office, Washington, D.C.). Posner, M. V., 1961, "International Trade and Technical Change," Oxford Economic Papers. Steele, M. J., 1989, "Models for Managing Secrets," Management Science, 35, pp. 240-248. Vernon, R., 1966, "International Investment and International Trade in the Product Cycle," Quarterly Journal of Economics, 80.
COPYRIGHT 1992 Omicron Delta Epsilon
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1992 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Carlisle, E.R.
Publication:American Economist
Date:Mar 22, 1992
Previous Article:Institutions and economic theory.
Next Article:Slave labor as labor hoarding - input use inside the economic region of production.

Related Articles
Spillovers, rivalry and R&D investment.
Two-stage R&D competition: an elasticity characterization.
Multinational firms and the diffusion of skills and technology.
Patents, patent citations, and the dynamics of technological change.
The contribution of science and technology to production.
Determinants of employment growth at MNEs: evidence from Egypt, India, South Africa and Vietnam.
Foreign Direct Investment. Six Country Case Studies.
Modeling sectoral and welfare impacts of information technology spillover: a tale of two Souths.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters