R&D and Knowledge

Analysis of data shows an uncertain relationship between knowledge capital and research and development.

These days, "knowledge" is a popular term for justifying increased spending on technology tools. But beyond anecdotal explanations, I have seen no hard evidence to demonstrate that spending more money on KM programs, devices and solutions would improve a company's financial results.

When pressed for proof of profitability, proponents of knowledge development programs respond that hard data is difficult to obtain. In the absence of generally accepted metrics or even a theoretical model, some of these experts resort to techniques of the sort used in food testing, beauty contests and opinion polls. They produce conceptual presentations, logic diagrams, pie charts and bar graphs, but nothing that a skeptical investor with limited funds would consider to be verifiable justification.

In the presence of such an analytical vacuum, I am amazed that so little attention is paid to a knowledge acquisition activity for which data is widely available, whose results can be traced for more than a decade and whose individual data points possess as much reliability as our corporate institutions are capable of producing. I am referring to annual data for research and development expenditures, which appear regularly as line items on audited financial statements.

The accounting practices for identifying these expenditures are well established and regulated by the Financial Accounting Standards Board. Reporting of R&D expenditures is scrutinized by government agencies, including the Securities and Exchange Commission and the Internal Revenue Service, that have power to enforce credibility in what is reported. The history of granting investment tax credits for R&D also contributes to the consistency of such data over an extended time.

To demonstrate the reliability of this source of data, I will use the 1999 R&D expenditures for 2,004 U.S. corporations, as reported by Standard & Poor's, to test the proposition that more R&D increases knowledge capital (KC). The companies' annual financial statements offer insights into the rewards of organized knowledge gathering in R&D departments that are usually under seasoned professional management.

The Right Ratio
For purposes of this article, I have chosen to examine the relationship between the ratio of R&D spending to cost of goods (R&D/COG) and the ratio of knowledge capital to shareholder equity (KC/EQ). The R&D/COG ratio should tell us how many cents of deliberate spending on enhancing a corporation's knowledge are embedded in every dollar's worth of goods produced. The KC/EQ ratio should tell us whether the worth of knowledge capital exceeds the nominal worth of financial capital (also called a company's book value).

Over the past year in this column, I have described how to measure knowledge capital; you can find these articles at strassmann.com/pubs/km or www.destinationKM.com. However, for this particular analysis I have applied a more rigorous approach to calculating the effective cost of capital by using the capital asset pricing model, a method widely used in financial portfolio analysis to determine the expected rate of return on an investment at a specified level of risk. The capital asset pricing estimates are currently listed in Standard and Poor's Compustat tables.

I used this method in this case because the reported R&D numbers are sufficiently accurate to attribute company-specific internal risks to their R&D investments. For my usual knowledge capital calculations, that is not possible because there we deal largely with external and strategic risks.

A comparison between R&D/COG and KC/EQ ratios does not suggest that corporations that spend more on R&D will improve their ratio of knowledge capital, as the graph "R&D vs. Knowledge Capital" shows. In other words, knowledge accumulation through R&D expenditures is only a contributory influence on corporate performance as a creator of knowledge capital--not a decisive influence.

The graph demonstrates that as R&D/COG--the ratio of R&D spending compared with the cost of goods sold--increases from a negligible amount of 0.001 percent for an agricultural products company to a high of 100 percent for a biochemistry start-up, the gains in knowledge capital relative to shareholder equity do not show a favorable trend. Put simply, more spending on research and development (a knowledge acquisition investment) does not show up as gains in knowledge capital.

The random pattern in the scatter graph also appears in all of my prior studies dealing with knowledge enhancement efforts, such as spending on computer systems. Yet random patterns do not invalidate the need to spend money on knowledge management, on R&D or on computers. Rather we should understand the absence of correlation between spending and knowledge acquisition primarily as a warning that unless sound business practices, superior management and advantageous competitive positioning are also present, spending on even the most worthy of endeavors will not necessarily produce superior results.

Paul A. Strassmann originated the trademarked concepts "information productivity", "return-on-management" and "knowledge capital."