15.3. BASIC RESEARCH: WHO NEEDS IT?
The conduct of basic research that does not have immediate commercial benefit raises questions such as "Who needs it?" and "Why do it?"
The genius of Vannevar Bush in articulating the importance of basic research over half a century ago, when it was an untested concept, is quite remarkable. In commenting (Bush, 1945, pp. vi, 1) on the importance of investment in basic research, which is responsible for scientific progress, he stated:
Scientific progress is one essential key to our security as a nation, to our better health, to more jobs, to a higher standard of living, and to our cultural progress.... Without scientific progress no amount of achievement in other directions can insure our health, prosperity, and security as a nation in the modern world.
Benefits from basic research are hard to predict. Even the scientists are unable to see how their discoveries could someday address crucial human needs. J. Michael Bishop (Nobel Prize winner in physiology) describes his experience to illustrate this point (1995, p. 63):
In 1911, Peyton Rous at the Rockefeller Institute in New York City discovered a virus that causes cancer in chickens, a seemingly obscure observation. Yet 65 years later, that chicken virus was the vehicle by which Harold Varmus and I, and our colleagues, were able to uncover genes that are involved in the genesis of human cancer. The lesson of history is clear, the lines of inquiry that may prove most fruitful to science are generally unpredictable.
As discussed in Chapter 1, "R&D Organizations and Research Categories, a great majority (approximately 75 percent) of the research conducted at universities falls into the category of basic research. However, basic research conducted at U.S. industry and government laboratories is a much smaller proportion of their overall research activities: about 4 percent and over 20 percent respectively. Thus, investment in basic research by the nation is of special concern to the university community.
In the public sector, as in the industrial sector, decision-makers are often concerned with current problems and issues. Since basic research involves discovering fundamental mechanisms rather than achieving practical applications, and there is considerable uncertainty and risk, it is not difficult to understand why support for basic research is not always as strong as one might expect.
Personnel involved in basic research need resources to undertake such activities. In addition, scientists want freedom to investigate the topics they deem worthwhile. Basic research requires a great deal of time, but decision makers tend to be impatient; when faced with the problem of distributing scarce resources among competing requirements, decision makers may find it difficult to support basic research.
One could argue that if resources are not invested in basic research, the foundation necessary for technological innovation (infrastructure for training scientists and engineers at universities, trained personnel, and new inventions) would be missing. On the other hand, without technological innovation and investment in technology, the increase in productivity and the general economic well-being of society would be missing. After all, commercialization of technology is the engine that produces resources for use in basic research and for investment in the future. While the scientific and the university community may want the decision makers to understand the importance of basic research for the long-term viability of an industrialized society, it is equally important that scientists understand the importance of the innovation process, which turns outputs from research and development into useful commercial products so that the extra resources needed for basic research can become available.
Basic research focuses on the development of new knowledge, much of which is embodied in scientific information that cannot be turned into a marketable private property. As Merton (1973, p. 273) has suggested, the findings of science are a product of collaboration within the scientific community. Discoveries are the property of the commons, and the rights to these properties are assigned to the wider scientific community. This implies that in most cases the output from basic research is not directly marketable. This raises questions such as:
Who should fund basic research?
How should the resources devoted to basic research be determined?
How should efficiency in the use of these resources be achieved?
15.3.1. Funds for Basic Research
There seem to be three major possible funding sources: private enterprise through the free market economy, governmental agencies, and nonprofit foundations. It is useful to examine which sources are most likely to fund activities that will improve society.
Discussing the allocation of resources for invention, Arrow (1974, pp. 144–163) has treated this subject very elegantly and thoroughly. He suggests that the possible failure of perfect competition to achieve optimal resource allocation through the free market system is related to several factors. His discussion is too technical for review here, but the conclusion is clear: The free market system is not going to be able to allocate the necessary resources to basic research activities. In addition, no matter what the demand for the output of basic research might be, it would be less than optimal in a free market system, for two reasons: "(1) since the price is positive and not at its optimal value of zero, the demand is bound to be below the optimal; (2)...at any given price, the very nature of information will lead to a lower demand than would be optimal" (Arrow, 1974, p. 154). What this means is that even if the free market system could provide the necessary basic research funds, this would not be in the best interest of society since the basic research output would be utilized at a level that is less than optimal for society. Noncommercial sources have historically been the major funding sources of basic research. And this not only makes sense, but is desirable from a societal viewpoint.
It is important to note that sometimes the output from basic research is needed to conduct applied research and that science (the output from basic research) and technology interact with the market or human needs to foster the innovation process. Logically, the only basic research supported by industry then would be in areas where investment in basic research is essential to complete product developments that have present or potential commercial benefits.
In discussing this issue of basic research, Professor Andrew Schofield of Cambridge University pointed out that in pursuing new knowledge, in addition to the wages paid to a researcher and the capital invested in laboratory facilities, there is the fundamental issue of the individual scientist's motivation and drive to discover. Many scientists invest considerable personal time, effort, and emotional energy in pursuing new knowledge regardless of its marketability or immediate utility. This investment by the individual scientist may be far more important than the institutional investment in basic research. Management practices that recognize this and foster an innovative environment—via job design, organization development, and reward system—may result in increased effectiveness of institutional investment in basic research.
The realms of industry, as well as government, academia, and not for profit labs are clearly placing more emphasis on applying science and technology often to commercial ends. Where does basic research—research undertaken with no application in mind—fit? The quest for knowledge as a human endeavor has intrinsic value, but fundamental research may eventually lead to innovation. However, it typically requires long periods of time and its application can often only be appreciated retrospectively. Given its intentional lack of market orientation, is it possible the attention to innovation and economic rewards may eclipse the need for investment in basic research, and thus ultimately darken the nation's long term leadership prospects in technological innovation?
Three trends underlay the concern:
Industry dominates R&D funding and naturally tends to focus on applied research and development that will lead to profitability.
The federal government, the major supporter of basic research, has become a less prominent funding source of R&D in general.
Universities, the major performer of basic research, have turned to licensing of technology as a source of revenue, which may divert attention away from basic research.
Table 15.1 sheds light on the issue by contrasting funding for basic research overall and by sector in the years 1980 and 2006.
The key points from this analysis are as follows:
In 2006, basic research accounted for about 18 percent ($53 billion) of the total R&D funded in the United States, while in 1980, basic research accounted for 14 percent of total R&D funding.
Industry, which dominates R&D overall, represented about the same proportion of all basic research funding in 2006 (17 percent) as it did in 1980 (15 percent).
Between 1980 and 2006, industry funding of basic research increased 282 percent.
Of industry's funding for all R&D, it dedicated about the same percent of funds to basic research in 2006 (5 percent) as it did in 1980 (4 percent).
Between 1980 and 2006, federal funding of basic research increased 174 percent.
The federal government dedicated more funding for all R&D as a percent of funds to basic research in 2006 (38 percent) as it did in 1980 (20 percent).
While universities represent a small proportion of R&D funding overall, funding of their own basic research has more than quadrupled over the period from 1980 through 2006; universities in 2006 represented 10 percent of all basic research funding versus 6 percent previously.
Other sources of funding—foundations (10 percent) and state and local governments (3.5 percent)—are playing an increasingly important role in funding basic research at universities—up from 9 percent to 14 percent.
The comparison of these two snapshots in time implies that support for basic research in the United States has increased, and that the decreased proportion funded by the federal government has been compensated for by increases in funding by universities, as well as foundations and state and local governments. Industry as a proportion has remained stable. While this analysis is encouraging, a close look at the growth trends of basic research suggests that concern is warranted.
Figure 15.1 shows trends in total funding (federal, industry, and university sources) for basic research from 1980 through 2006. In the first 20 years, funding for basic research grew at about 5 percent per year. However, during the five-year period from 2001 through 2006, the average annual growth rate was 2.6 percent.
Figure 15.1. Trends in Basic Research Funding
Further, the NSF (Science and Engineering Indicators 2008) has reported, "in this century the industry share of support for basic research in universities and colleges, the primary performers of U.S. basic research, has also been declining. Likewise, Federal Government support for academic R&D began falling in 2005 for the first time in a quarter century, while Federal and industry support for their own basic research has stagnated over the last several years. These trends are especially alarming in light of the growing importance of knowledge-based industries in the global economy." Just as global competition in R&D and innovation is growing stronger, the United States may not be priming its basic research pump adequately.
Economic and political changes clearly influence the amounts and types of research funding. Even the meaning of basic research can drift, so that it implies less directed applied initiatives versus undirected fundamental research. In the commercial realm, a handful of companies with deep financial resources are actively pursuing basic scientific research. IBM, for example, recently announced basic research initiatives in physics, chemistry, and math, though these are focused on generating breakthroughs in the applications of semiconductor electronics, data processing, quantitative problem solving, and cloud computing. Companies need to maintain fundamental research initiatives in order to retain research scientists and engineers who can maintain external linkages to knowledge generated in universities. Pharmaceutical and scientific R&D services companies especially need to maintain capabilities and communication channels in basic research. However, industry sponsored R&D overall—and basic research in particular—is subject to the vagaries of larger economic trends. Historically large high-profile corporate research laboratories like those at RCA, Xerox, and AT&T's Bell Labs have been dismantled. More recently, Motorola has significantly downsized its basic research efforts and redirected R&D toward applications. Venture capital funding also mirrors economic trends. As was shown in Chapter 14, U.S. venture funding has not returned to the levels prevalent in 2000, and the distribution is away from high-risk seed and early-stage funding based on emerging technology and toward expansion of more proven ventures.
Pressure to maintain profitability will tend to drive industrial R&D toward shorter-term applications. Thus the federal government must remain vigilant in its support for basic research, especially at universities. There is good reason behind their dominant role in basic research. Universities are particularly well suited to higher-risk research. While the university setting does not offer the salary levels of industry, university scientists and engineers value the relative autonomy of academic institutions, and the ability to publish their research and thus promote dissemination of findings. And while, on the surface, it might appear that increased emphasis on commercial licensing of research performed at universities might dampen basic research efforts, this does not appear to be the case. In their study of university licensing, Thursby and Thursby (2007) concluded that, although researching the issue is difficult, the threat of eclipsing basic research efforts is not strong, given that a minority of faculty on U.S. campuses is involved in licensing. As long as a meaningful proportion of university licensing revenues is reinvested in fundamental research, a healthy balance can be achieved.
15.3.2. Level of Resource Allocation
Because of the reasons cited above, private enterprise is not able to allocate the necessary resources for basic research using the efficient competitive market processes, except where basic research is needed for commercial product development. Any time government or other nonprofit organizations find themselves engaging in such activities, the level of resource allocation becomes a problem. Other surrogate measures and social needs have to be used, recognizing that the level of resource allocation in such cases cannot be as efficient or as self-correcting as the market mechanism. Discussion of some of these measures and needs follows.
15.3.2.1. Economic Considerations
Presumably, resources should be allocated so that the expected marginal social benefit exceeds or equals the marginal social benefit of competing usages. Because of considerable uncertainty and other complexities, these computations may simply be based on a preference index, which might itself be determined by the expected economic benefits. As mentioned earlier, authoritative studies have shown that support for basic research is a prudent investment that is estimated to pay dividends to society exceeding 20 percent in real economic terms (Lane, 1996).
15.3.2.2. Its Contribution to Social Welfare
As Freeman and Soete (1997) have stated, "The advance of science and technology must find its support and its justification, not merely in the expectation of competitive advantage, whether national or private, military or civil, but far more in its contribution to social welfare, conceived in a wider sense." One of the colleagues at Cambridge University stated that basic research should be looked at as "an important cultural activity. If one is looking for direct returns, one is not likely to see a penny back." Therefore, decisions on allocating resources for basic research have to consider its enormous societal benefits.
15.3.2.3. Based on Comparative Needs
By looking at the historical record, one could gather percentages of gross national product devoted to basic research by other nations and compare these data to the impact of varying percentages of investment in basic research on economic productivity and the social welfare of society.
15.3.2.4. Other Considerations
In discussing this with some eminent R&D managers (identified in the preface), a number of other practical considerations emerged, as reflected in the following questions. Basic research investment should run 10–15 percent of total R&D and depends on:
"Do we have the people who are interested?"
"Do we have the people who have the time?"
"Do we have the people who are able to conduct basic research?
This really determines the level of basic research funding in most cases.
As discussed in Chapter 1, a scientist who engages in a mix of basic and applied research is likely to be more productive than one who does not. Therefore, even for a profit-oriented industrial organization, allowing the scientist to undertake some basic research, regardless of its profitability or immediate utility, is best for all concerned. This will keep the scientist at the cutting edge of the discipline, provide a higher level of motivation, and, inevitably, result in greater productivity in the applied aspects of the research as well.
15.3.3. Efficient Usage of Basic Research Investment
At the macro level, efficient usage is partially achieved by establishing an appropriate science policy. For basic research, by and large, payment is independent of the results achieved. So, the traditional market mechanism is irrelevant. There are, however, other mechanisms that ensure efficiency at the macro and micro levels:
The peer review process to determine the merits of proposed basic research projects
Awarding research contracts to those who have previously performed research successfully
The ethos of the scientific community (universalism, communalism, disinterestedness, and organized skepticism) that provides a vigorous review and analysis mechanism contributes to efficiency
Suggestions presented in this book, for creating a productive and efficient R&D organization, that might be helpful
Peer review and other processes mentioned here have historically worked well, but they tend to be a bit conservative and lack flexibility. Major breakthroughs are unlikely to be anticipated by a committee. In an R&D organization a portion of total R&D funding can be provided to managers at different levels as discretionary funds to be used for high risk research projects. Our experience indicates that this approach has proved very fruitful. It reduces the lead time required for the normal funding cycle for a project and provides the flexibility so necessary to undertaking high-risk, exploratory research requiring relatively low investment.