CONCERNS
Overall U.S. Investment in Research and Development
The FS&T budget is one part of the broader measure of R&D in the federal government, as well as of the total public and private support of R&D. Total R&D spending has been rising as a share of GDP and is now nearing 3 percent—2.8 percent in 1999. While hard and precise targets for total R&D spending and for its composition are hard to establish, there is a wide consensus that U.S. economic growth and scientific preeminence depend on maintaining and possibly increasing the share of GDP devoted to R&D. The administration has suggested a target goal of total R&D as 3.0 percent of the U.S. gross domestic product (GDP).3 Movement in recent decades toward that goal has been achieved through growing private rather than public investment. As seen in Figure 3, there has been a significant divergence of federal and non-federal investment patterns in R&D as shares of GDP since 1987.4
Basic research receives its principal support from publicly supported R&D, whereas privately sponsored R&D emphasizes applied research and development. The continued effectiveness of industry expenditures on applied research and development depends on the continued flow of basic research findings and the associated training of scientists and engineers.5 Industry benefits from, and invests in, the development of products based on basic research conducted in prior decades. Thus, continued growth of basic research will help sustain continued high returns to private R&D outlays, and ensure a pipeline of new knowledge accessible to future generations.
The growth of industry spending on R&D should, therefore, not lull observers into thinking that the federal research budget can consequently be reduced. This growth does not reduce the need for a strong federal research budget.
Balancing the FS&T Portfolio
The differences in the growth rates of FS&T investments across fields are a concern.6 In essence, the life sciences budget has surged ahead while the FS&T
3 |
Science in the National Interest, The White House, 1994. |
4 |
Federal R&D as a percentage of total R&D in the United States reached a high point in 1964 at 66.8 percent, equaled 46.4 percent in 1987, and in 1999 was 26.7 percent. See NSF, National Patterns of Research and Development Resources 1999 Data Update (NSF 00-306). |
5 |
Capitalizing on Investments in Science and Technology, COSEPUP, National Academy Press, 1999. |
6 |
For an NRC review of this problem using data through FY1997, see Securing America's Industrial Strength, Board on Science, Technology, and Economic Policy, Appendix A. |
budgets for other fields have increased only slightly or have decreased with the cuts in the DOD budget. FS&T at NIH provided an increase to life sciences research between FY1998 and FY2000 that is greater than all of FS&T proposed for NSF for FY2001.
Research in the life sciences is motivated by a need to improve health. Yet many of the improvements seen in the past decades are due to advancement of knowledge that comes from other fields. Examples would include magnetic resonance imaging, positron emission, and miniaturization in athroscopic surgery. As Harold Varmus, former Director of NIH, has often explained, discoveries in biology and medicine depend on progress in physics, chemistry, engineering, and many allied fields. The FY2001 budget recognizes the need for balanced expansion of research with substantial increases proposed for the National Science Foundation in particular.
While reallocation of funds within a limited budget is inescapable, abrupt decreases can raise difficult problems. Among major programs, for instance, a cut of 14 percent is proposed for DOD FS&T. DOD has been and remains a major sponsor of academic research in the physical sciences and engineering. Much greater attention needs to be given to the impact of such reductions on fields, as available retrospectively in NSF data, where multi-year trends may sig-
nal the erosion of U.S. capability. Analysis could assess whether those trends affect U.S. global leadership in science and engineering.
Swings in FS&T levels pose difficulties for those planning careers in science and engineering. Federal research funding directly and indirectly supports the training of the next generation of scientists and engineers. Only rarely do budget decisions take into account those effects of the various agencies funding research. A recent National Research Council review of major fields with substantial declines in federal research support (chemical engineering, mechanical engineering, and electrical engineering) in the 1990s shows a strong correlation with reduced graduate enrollments in those fields.7
Mission agencies contribute substantially to fields not readily identifiable in the stated missions. The cuts in programs at NASA, DOE, and DOD are especially notable. These declines, accumulated over a period of years, can be damaging to a research infrastructure that takes years to build and to maintain at state-of-the-art condition.
The adequacy of physical infrastructure also requires close attention. 8 The rapid, recent increases in FS&T funding for the life sciences challenge the capacity of research institutions to respond to the demands for expanded programs. Construction lead time for buildings and laboratories can be long. Without such physical capital, ambitious research programs may be needlessly costly or simply unattainable. Recent FS&T proposals have inadequately reflected the long-term costs to research institutions of raising the funds and building the human and physical infrastructure to maintain an adequate research capacity.9
Rising levels of federal support for research programs increase tension between the government and universities over indirect cost recovery. Universities are presently contributing substantially toward making this investment effective with their own resources as a result of incomplete cost recovery and other forms of cost-sharing.10 Unless universities can find additional revenue, this cost burden will cause tradeoffs with other university functions. The successful completion of efforts to reform rules governing reimbursement for indirect costs deserves high priority.
7 |
Securing America's Industrial Strength, Board on Science, Technology, and Economic Policy, National Academy Press, 1999, pp. 89-93. |
8 |
An example of the analysis needed has been done for information technology: National Research Council, Funding A Revolution: Government Support for Computing Research Infrastructure, CPSMA, National Academy Press, 1999. |
9 |
The federal government directly paid for 9 percent of construction, renovation, and repair of academic research facilities in 1998, with the rest of the funds coming from state/local governments (about 30%) and internal university funds (about 60%). National Science Foundation, Scientific and Engineering Research Facilities at Colleges and Universities: 1998: An Overview (NSF 99-413), Arlington, VA: NSF, 1999, pp. xii-xiii. |
10 |
See National Science and Technology Council, Presidential Review Directive 4, Chapter 5, April 27, 1999. |