| Home |
 |
Student Resources |
|
Chapter 16: Patents and Technological Change |
|
Joseph A. Schumpeter (1950) started this debate by contending that large corporations with monopoly power were likely to advance industrial technology because of superior access to capital, ability to pool risks, and economies of scale in the maintenance of R&D laboratories. Critics argue that large bureaucracies stultify R&D.
According to Scherer (1984, ch. 11), Gelman Research Associates' large data base of major innovations indicates that small firms are relatively more likely to make major innovations. Companies with fewer than 1,000 employees are responsible for 47.3 percent of important innovations, a share greater than their 41.2 percent of total employment. Companies with over 10,000 employees are responsible for 34.5 percent of the innovations, but 36 percent of total employment. Thus, large firms innovate roughly in proportion to size, small firms innovate out of proportion, and medium-sized firms do relatively little innovating.
Also according to Scherer (1984, ch. 11), in another sample of major innovations reported in Industrial Research & Development, 72 percent of the major innovations came from U.S. profit-seeking firms, 12.9 percent from government laboratories, and 10.7 percent from other nonprofit organizations. Large firms in the Federal Trade Commission's (FTC's) line-of-business survey account for 73 percent of both company-financed and federal contract industrial R&D expenditures, 55 to 60 percent of manufacturing sales, and 55 percent of the major innovations. Again, there is no strong evidence that size leads to more innovation, at least in the United States. The large firms in the FTC data account for only 61 percent of all patented inventions, a smaller segment than their 73 percent share of the R&D expenditures.
Cohen, Levin, and Mowery (1987) find, using a statistical analysis of the FTC's line-of-business data, that business-unit size has no effect on the R&D intensity (R&D as a percent of sales) of firms conducting R&D but does affect the probability of engaging in R&D. Moreover, business-unit and firm size collectively explain less than 1 percent of the variance in R&D intensity, in contrast to industry effects, which explain nearly half the variance.
Bound, Cummins, Griliches, Hall, and Jaffe (1984) find, using sophisticated statistical techniques to control for the underreporting of R&D, that R&D expenditures increased with sales and gross plant size in 1976. At the sample mean, a 1 percent increase in sales causes a 0.7 percent increase in R&D. The impact of size is nonlinear, however, so that very small and very large firms are more R&D intensive than average-sized firms. Small firms that do research tend to patent more per R&D dollar than larger firms, and firms with large R&D programs tend to have a constant ratio of patenting to R&D. Jensen (1987) finds that there are no advantages or disadvantages to locating a given R&D program in a larger firm in the pharmaceutical industry. Holmes, Hutton, and Weber (1991) find that R&D intensity varies with firm size in some industries and not others, and that where it does vary, it may be negatively or positively related to size. Thus, these studies do not confirm the Schumpeter hypothesis for the United States. However, see Example 16.8.
|