Which are really bringing new products to market, big companies or small ones? Or is that even the right question for policymakers to ask?

Innovation ranks high among the desirable characteristics of any country's economy. As economist Joseph Schumpeter pointed out more than 50 years ago, it isn't price competition but new technologies, new products, and new forms of organization that lead to decisive cost or quality advantages.

By the same token, innovation ranks high among policymakers' concerns. Too often, however, policy debates pit small companies against large ones in the scramble for resources. Often, too, the debates are based more on anecdotal evidence than on what's happening in the real world and how that might be changing.

The earliest academic studies -- in the 1950s and 1960s -- looked at the amount of money budgeted for research and development or the proportion of a company's work force involved in R&D, which weighted the scales in favor of large corporations. Critics quickly pointed out the limitations of those criteria. For one thing, they ignore informal R&D, which is more common in small companies. More important, the results in the marketplace are what should be measured, and they may not be captured in research expenditures.

After R&D studies, the next systematic attempt to discover who was doing the innovating was to document patents. That measure, too, had its limitations. On one hand, many patents are filed that never lead to a commercial product; on the other hand, many innovations come to market with nary a patent.

Not until well into the 1970s were systematic attempts made to measure innovation directly. More recently, new databases from the U.S. Small Business Administration and other sources have become available to researchers. In my own studies I have compared the intensity of R&D with the new data for different size companies. (Large companies are defined as those with 500 or more employees; small companies are those with fewer than 500 employees.) As is often the case, the data, while newly released, are not current. Still, they give us some insights into the source of new products. What follows are the conclusions we've reached using the newer databases and examining the most innovative companies in 34 U.S. industries in 1982.

Both large and small corporations contribute to innovation. Looking at manufacturers in 1982, we found small companies in the lead: the mean small-company innovation rate was 322 innovations per million employees, compared with 225 innovations per million employees for large companies.

There are considerable differences in innovation across industries. While large-company innovation rates were relatively high in the rubber industry, small-company rates were high in the production of nonelectrical machinery and electrical equipment. Large-company rates were relatively low in leather, textiles, and petroleum; small-company rates were relatively low in food, rubber, and paper. Not surprisingly, large companies hold the relative advantage in industries that are capital intensive, advertising intensive, concentrated, and highly unionized. Small companies hold the relative advantage in industries such as computers and process-control instruments.

Service companies understandably have fewer product innovations than manufacturers do. Still, we found 6.6 innovations per million employees in small service companies, compared with 12.9 in large service companies. In 1982 large corporations in the service sector appeared to be twice as innovative as small companies were.

While more recent data will not be available for another year, we can say with assurance that the role of small companies has become more, not less, important. From steel to computers, small companies continue to lead. In the all-important biotechnology industry small companies have a virtual monopoly on innovations.

Policymakers would do well to look at the characteristics both large and small companies bring to the table and at the particular types of industries in which each group excels.

* * *

Zoltan J. Acs is the Harry Y. Wright Professor at the University of Baltimore School of Business and editor of Small Business Economics . His most recent book, coauthored with David B. Audretsch, is Small Firms and Entrepreneurship: An East-West Perspective (Cambridge University Press, 1993).