Innovation and Size at the Firm Level

Abstract

Empirical studies examining the relationship between firm size and innovative activity have produced what superficially might appear to be contradictory results.' While some studies have found a positive relationship between firm size and technological change, still others have identified no relationship or even a negative one. There are two main reasons for these seemingly inconsistent findings. The first is that different measures have been used to quantify technical change. These measures have typically involved either some measure of inputs into the innovative process, such as expenditures, or else the number of patented inventions. While neither of these are direct measures of innovative output, they clearly represent different aspects of the innovative process. Thus, it is perhaps not too surprising that different results have tended to emerge when the input measures are used than for the patent measures. The second reason is that virtually every study examining the relationship between firm size and technical change has had to use a truncated distribution of firm sizes where either no or only a few small firms were included. For example, Scherer's [14, 234-35] conclusion that the empirical results, . .. tilt on the side of supporting the Schumpeterian Hypothesis that size is conducive to vigorous conduct of R&D was based on the responses of 443 large corporations participating in the Federal Trade Commission's Line of Business Survey. Similarly, Scherer [15] used the Fortune annual survey of the 500 largest U.S. industrial corporations and found that the number of patented inventions increases less than proportionately along with firm size. Soete [16] found that expenditures tend to increase more than proportionately along with firm size using a sample from Business Week, consisting of the most intensive firms. Bound et al. [6] were able to include a considerably wider spectrum of firm sizes in their sample of 1,492 firms from the 1976 COMPUSTAT data. They found that increases more than proportionately along with firm size for the smaller firms in their sample, but that a fairly linear relationship exists for the largest firms. Inferences about the relationship between firm size and technical change based on a severely *We thank F. M. Scherer and an anonymous referee for helpful suggestions, as well as Sigrid Raasch and Jianping