KEY POINTS IN THIS CHAPTER
- Federal funding plays an essential role in supporting research, particularly basic research. The figures presented in this chapter illustrate the mission-oriented allocation of research funds by federal agencies and for certain areas of science.
- Impediments to growth in the research and innovation systems include insufficient funding for proof-of-concept research and swings in federal research funding.
- The United States remains at the forefront of research and development (R&D) with the world’s largest investment in R&D, the largest share of scientific publications, more than one-third of scientific publications cited in patents, and world-class research universities. However, this position is increasingly challenged by competition from many other nations, including China.
As discussed in Chapter 1, the U.S. research enterprise is a complex, dynamic system. This system is embedded within and evolving in conjunction with industry and market structures that are adapting to the intense competition resulting from globalization. Currently the strongest in the world, it has had a great deal of freedom in its develop-
ment; researchers and investors have available numerous options that are not constrained by government. Perhaps for that reason, the complex of research universities, private industry, government, and nonprofit organizations and the rich web of interactions among them have evolved into a system that is difficult for other nations to emulate. This chapter presents an overview of the evolution of the U.S. research enterprise, addressing in turn the role of research in the national economy; the complementary roles of industry, government, and philanthropic funding; historical trends in research funding; the performers of scientific research; impediments to the research and innovation systems; and how the U.S. research enterprise compares with those of other nations.
RESEARCH AND THE NATIONAL ECONOMY
Over the course of its history, the United States has developed from a primarily agricultural to an industrial or manufacturing economy, and then to an economy in which services have grown in importance, and knowledge, information, and human capital play a larger role than ever before. The development of today’s knowledge economy was enabled by research both directly and indirectly through the emergence of new sectors and the growing role of technology in production processes.
As noted in Chapter 1, the importance of public and private investments in R&D to the national economy was recently acknowledged by the government’s decision to revise gross domestic product calculations. Previously, business expenses on R&D and on the intellectual property of some creative and artistic works were treated as intermediate expenses of production; they were not included with other investments, such as those in new plants and capital equipment, that are expected to contribute to the production and sale of future products and services. When R&D expenses are treated as investments rather than expenses, their contributions to the national economy are highly visible and significant. This shift represents an important step in measuring the value of research.
COMPLEMENTARY ROLES OF INDUSTRY, GOVERNMENT, AND PHILANTHROPY IN FUNDING RESEARCH
Recent plateaus in research spending by the government and private industry raise concern about the future of the U.S. research enterprise. Figure 2-1 indicates that industry’s share of funding for U.S. research in 2009 was roughly at the same level as in 1953, while the federal share has declined since the early 1960s, partly because of reductions in the size of
FIGURE 2-1 Federal, industry, and nonprofit shares of total funding for U.S. basic and applied research, 1953-2011. Excludes university and college (U&C)-funded research performed in U&Cs and research funded by other government entities (e.g., state governments) performed in U&Cs.
NOTE: The character-of-work estimates for business R&D were revised for 1998 and subsequent years; therefore, the data for 1997 and earlier years are not directly comparable. Furthermore, the character-of-work estimation procedures for higher-education R&D were revised in 1998 and again in 2010; thus, the data for 2010 and beyond are not directly comparable with those for 1997 and earlier years, or for 1998-2009.
SOURCE: Data from National Science Foundation (2014a), Appendix Tables 4-7 and 4-8.
federal investments in defense-related research.1 Of particular concern is the future funding of high-risk, long-term research and of proof-of-concept research (described further in Chapter 6), which helps bridge the gap between research and development through the development of innovations. What are the roles of industry and government in funding such research?
1The figures in this chapter are subject to the quality of existing data on research spending. For more information about R&D expenditures, see National Research Council (2005a, 2013c).
Industry funds a portion of the nation’s basic research but places more emphasis than the government on leveraging internal funds, government contracts, and private support to create the end products of R&D. These end products encompass the manufacturing of chemicals, computers, electronics, aerospace and defense components, and automobiles, as well as the performance of services related to software, computers, and R&D. With a growing emphasis on applied engineering research, industry-funded research has become increasingly global and is not necessarily conducted in the United States, although it is far more likely than industry-funded development to be based in the United States (National Science Foundation, 2012). This difference between industry-funded research and development may reflect the substantial research infrastructure and budget that are supported by federal funds. In addition, the publicly supported U.S. research infrastructure and budget may attract foreign investments in U.S.-based industrial research.
The federal government, on the other hand, makes key investments in basic research, recognizing that the generation, distribution, and application of knowledge will prove fruitful in years to come by ensuring an eventual supply of marketable innovations and other societal benefits. Federally funded research is performed to benefit society and is intended to produce social returns more than private gains. Because others can benefit from those returns without sharing in the cost, the private sector has fewer incentives to produce them. This is particularly the case for basic research, which, as defined in Chapter 1, is scientific research conducted to increase fundamental understanding and not necessarily for a technological goal or other application. Moreover, government generally is better able than industry to tolerate the long wait—sometimes decades—for the transformation of knowledge into useful applications. Finally, in some cases the government funds research because it needs the end products for its own use to fulfill agency missions (e.g., for the development of defense weapon systems or of vaccines).
Increasingly, government is called upon to fund high-risk, long-term research and some types of applied research, particularly proof-of-concept research, at least to the point where the risks of investment in such research are reduced to attract private-sector funding. Inevitably, some high-risk research will not achieve its stated goals. Nevertheless, even failed research projects expand pools of talent and research capacity, can result in valuable redirection of research trajectories, provide valuable learning experiences for new researchers, and sometimes yield important but unanticipated discoveries. Regardless, government has a unique responsibility to support high-risk research because of its potential long-term benefits to society. Research portfolios can be pooled to reduce
risk and allow a few transformative discoveries to emerge that more than justify the entire portfolio.
Today, a number of government programs provide direct funding for applied research, filling the gaps in which industry lacks incentives to do so. Much of this funding is focused on the handoff from research to development. Through programs such as the National Science Foundation’s (NSF) I-Corps (discussed in Chapter 6), the government may be able to assist industry in bridging the gap between research and development, although few such programs have been rigorously evaluated to determine how well they accomplish this aim. These programs attempt to increase the efficiency of the transition from discovery to innovation through creative practices, most of which involve partnerships with industry. For example, I-Corps requires that academic researchers team with an entrepreneur possessing the skill sets and knowledge necessary to bring an invention to the market successfully.
The role of government in filling the gaps left by industry is illustrated by funding for so-called infratechnologies—the technical tools that enable the development and use of technologies. Examples include methods of measurement and testing for conducting research, for managing production for quality and unit cost, and for enabling marketplace transactions. The fact that these tools are in common use (e.g., as publicly owned standards in the semiconductor industry) results in underinvestment by private industry because no one company can capture all of their ultimate benefits. Accordingly, government agencies such as the National Institute of Standards and Technology, U.S. Department of Defense (DoD), and U.S. Department of Energy (DoE) invest in infratechnology research. Because of their nature, which requires government as well as industry funding, infratechnology research and some proof-of concept research are considered “quasi-public” goods.
The figures presented below highlight the critical role of federal funding in supporting research, particularly basic research. They illustrate the mission-oriented allocation of research funds by federal agencies, and they show how federal funding is allocated to certain areas of science.
Although industry and federal funding accounts for the majority of total U.S. research, shares of philanthropic funding are swiftly rising. Philanthropic funding shares of total U.S. research increased from 0.06 in 1953 to 0.12 in 2011 (see Figure 2-1), and this trend is expected to continue. Private foundations and wealthy individuals play a particularly important role in supporting scientific, medical, and engineering research at U.S. universities and colleges, where philanthropic contributions amount to an estimated $4 billion each year and provide nearly 30 percent of some institutions’ annual research funds (Murray, 2012). Much of this funding supports research operations, facilities, and endowments.
Despite the growing significance of philanthropic funding, these data are not widely reported, and relatively little is known about the patterns of private contributions to various fields of science and types of research. A recent analysis of science philanthropy at America’s top 50 research universities (Murray, 2012) suggests that at the very least, one thing is clear: the role of funding from wealthy donors differs from the broadly understood roles of federal and industry funding. The most striking difference is that philanthropic funding often shapes research around the preferences of its patrons, rather than taking account of the needs of the research enterprise as a whole. Private donors support translational medical research far more than other types of research, and often neglect the critical need for basic research. As Murray (2012, p. 1) notes:
The documented extent of science philanthropy and its strong emphasis on translational medical research raises important questions for federal policy makers. In determining their own funding strategies, they must no longer assume that their funding is the only source in shaping some fields of research while recognizing that philanthropy may ignore other important fields.
Private philanthropy does play an important role both in providing university endowments and in directly funding research. In some areas, such as translational medical research and astronomy,2 this role is particularly important. Still, a recent article in The New York Times (Broad, 2014) questions whether the increasing support of research by wealthy individuals will ultimately privatize American science by shaping research programs around the preferences of philanthropists rather than around national priorities. It is worthwhile to note that, for now at least, philanthropic donations may be on the rise, but they continue to account for a small fraction of total U.S. research support.
HISTORICAL TRENDS IN FUNDING FOR RESEARCH
Assessing the trends in funding for research is difficult because of the variations in data reporting. Many of the data reported fail to distinguish between research and development, let alone between basic and applied research, and not all of the data are reported for a consistent fiscal year. Furthermore, calculations of research spending are sometimes inconsis-
2One prominent example is the Giant Magellan Telescope, which has received more than $1 billion in support, mostly from philanthropic donors (see http://www.gmto.org/pressrelease11.html [August 2014]). Another example is the Sloan Digital Sky Survey, described in Box 6-5 in http://www.sdss.org/ [August 2014]).
tent across federal agencies (National Research Council, 2005a, 2013c). These variations make it difficult to draw comparisons across fields of research or among nations.
As depicted in Figure 2-1, the majority of basic and applied research in the United States has been funded by the federal government. Substantial contributions also come from private industry and smaller contributions from nonprofit organizations that have grown significantly as a share of total research funding since the 1980s.
Within the federal budget, research spending is distributed among 20 agencies or departments, each with multiple programs that receive funds according to their missions and priorities. Spending on development is allocated in much the same way, but the final distribution of funds by each agency varies quite dramatically for research compared with development as a result of differences in the agencies’ missions. Figure 2-2 shows how the shares of basic research, applied research, and development vary within the R&D spending of the top six federal R&D funding agencies. For example, NSF allocates the largest share of its R&D budget to basic research, while DoD allocates the largest share of its R&D budget to development.
In fiscal year (FY) 2011, federal funding for basic and applied research was $60 billion. In FY 2009, the National Institutes of Health (NIH) received more than half of the federal research investment, while NSF accounted for about one-tenth. These two agencies invest heavily in basic research.
DoD, DoE, and the National Aeronautics and Space Administration (NASA) fund predominantly applied research that supports their agency missions. In addition, both DoD and NASA have been large-scale purchasers of products incorporating the results of their R&D investments.
Over the past 40 years, the NIH share of the federal research investment has increased dramatically, and the NSF share has increased to a lesser extent. Other federal agencies that invest heavily in research are shown in Figures 2-2a and 2-2b. Of the federal agencies and departments that invest heavily in basic research, NIH and NSF are most prominent (see Figure 2-2b). A portion of federal support for applied research is funneled through the defense budget, with the government acting as a customer. In this role, the government purchases research that will yield innovative technologies to help the nation prepare for the future.
Excluding funding from the American Recovery and Reinvestment Act between FY 1993 and FY 1999, federal support for research in the fields of physics, chemical engineering, geological sciences, and electrical and mechanical engineering declined by more than 20 percent in real terms, while funding for research in the biological and medical sciences increased by more than 20 percent in the same period. Funding
FIGURE 2-2 Allocations for research and development (a) and total applied and basic research (b) by the top six federal funding agencies, 2013 (budget authority in millions of constant 2014 dollars).
NOTE: DoD = U.S. Department of Defense; DoE = U.S. Department of Energy; NASA = National Aeronautics and Space Administration; NIH = National Institutes of Health; NSF = National Science Foundation; USDA = U.S. Department of Agriculture.
SOURCE: Adapted from American Association for the Advancement of Science (2013b). Reprinted with permission.
FIGURE 2-3 Federal research funding by discipline, 1970-2010—budget authority in billions of constant FY 2012 dollars. “Other” indicates research not classified (including basic and applied research, excluding development and R&D facilities).
NOTE: FY = fiscal year; NIH = National Institutes of Health.
SOURCE: Adapted from American Association for the Advancement of Science (2013b). Reprinted with permission.
for research in computer science was the major exception to this pattern of reductions in the share of physical sciences and engineering within federal research spending, rising by more than 60 percent (Merrill, 2013). Figure 2-3 shows trends in funding for individual fields of science, illustrating how federal funds for life sciences research have increased significantly since the 1980s such that this field now accounts for slightly more than half of all federal basic and applied research funding.
The various funders of U.S. research have each played key roles in the emergence of transformative innovations. An example is the Internet, presented as a case study in Box 2-1.
PERFORMERS OF SCIENTIFIC RESEARCH
The federal research investment flows to federal laboratories, universities, nonprofit institutes and hospitals, and contractors. Laboratories both produce agency mission-related outputs and conduct commercial activities, as described below. Universities produce new knowledge and an educated workforce trained in research, as well as other research outcomes, and likewise conduct commercial activities. Contractors, by and large, produce commercial products.
Case Study: The Internet
This case study illustrates the long time lag generally entailed in moving from an initial idea to a profitable product, the need for fluid movement across the public-private boundary, the critical roles played by supportive institutions, the benefits of sustained diverse government investment, and the unpredictability of the system of research.
The Internet’s evolution from a small experimental network connecting three U.S. research facilities at speeds of 56,000 bits per second to a global network with more than 100 million hosts and a backbone capacity in excess of 2 billion bits per second relied heavily on federally funded innovations originating in the early 1960s, just years after scientists unveiled the world’s first computer. Those early innovations were aimed not at connecting ordinary people on opposite ends of the earth, but at providing the U.S. Department of Defense (DoD) with a means of connecting and sharing the scarce computing resources available at the nation’s top research centers.
Federal funding for R&D—particularly defense funding—enabled the creation of semiconductors, computers, software, and other information technologies on which the modern-day Internet is built. The 1960s research on packet switching, the Advanced Research Projects Agency Network, and protocols such as TCP/IP relied on a 15-year DoD investment in hardware and software. The NSF also played a key role, particularly in the 1980s, by funding the Computer Science Network, linking university computer scientists, and the NSFNET, connecting supercomputer centers across the United States at more than 1.5 megabits per second—a remarkable feat by 1988 standards. By 1991, the NSFNET had become the first national network operating at 45 megabits per second.
Federal funds also supported generations of future talent by strengthening universities’ research capabilities in computer science; facilitating university “spinoffs” such as BBN and Sun Microsystems; and training the technical workers who helped develop, adopt, and commercialize the Internet.
Privately financed research played an important role as well, supporting basic
The differing priorities of industry and government are clearly reflected in the distribution of research performers supported by different federal agencies. Trends in the nation’s total research investment (i.e., R&D investments from private as well as public sources) by performer reveal that industry performs most of the nation’s research (see Figure 2-4). However, academic institutions have traditionally performed most of the nation’s federally funded research (see Figure 2-5). When basic and applied research are considered separately, it becomes clear that industry performs most of the nation’s federally funded applied research, while universities perform most of the nation’s basic research (see Table 2-1). However, industry’s share of applied research has declined
networking technologies including networking hardware, Unix, and the Ethernet protocol. Start-up firms were crucial to the commercialization of Internet-related innovations. Equally important were the heavy investments in information technology (IT) by U.S. industry during the 1980s that supported the rapid diffusion of the TCP/IP network. But in many respects, this private investment complemented and responded to the incentives created by public policies and larger market forces.
In addition to providing funds, the U.S. government influenced the development and diffusion of the Internet through regulatory, antitrust, and intellectual property rights policies, encouraging rapid commercialization of Internet infrastructure, services, and content by new firms. The U.S. Internet explosion of the 1990s also relied on close university-industry links and an abundant supply of venture capital. As the focus shifted from development to application, defense R&D spending was largely overshadowed by private-sector R&D investment. The U.S. venture capital industry assumed a larger role in the commercial exploitation of the Internet than had been the case during the formative years of other postwar U.S. high-technology industries. Defense-related procurement, which had played a prominent role during earlier stages of the Internet’s development, was not an important factor during the 1990s. Defense-related R&D investment in Internet-related fields, such as computer science, also declined modestly throughout the decade, although cutbacks in DoD R&D investments in computer science were more than offset by increased investments from other federal agencies, such as NSF and the Department of Energy (National Academy of Sciences, 1999, pp. 83-84).
The Internet has a history resembling that of other postwar IT innovations in that it was first commercialized primarily in the United States—the first country to deploy a large national research computing network; the first to standardize on TCP/ IP; and the first to develop a large, competitive market for individual access. The United States remains an international leader in overall network penetration, and its national network continues to grow rapidly. And this history all began with basic research supported by federal investments—in technology and in talented people.
SOURCE: Adapted from Mowery and Simcoe (2002). More information is available from National Science Foundation (2003).
significantly in the past decade.3 The share of federally funded research that is performed intramurally (i.e., within federal laboratories, whether operated by federal agencies or contractors) has declined significantly since the 1970s.
Among the performers of federally funded research, national laboratories, including 17 under the purview of DoE, invest heavily in scien-
3Industry-funded basic research, although beyond the scope of this report, was extremely important to the evolution of the U.S. research enterprise. See, for example, Gertner (2012) for the many impacts of basic research at Bell Laboratories.
FIGURE 2-4 Federal, industry, and university performance shares of total U.S. research (basic and applied), 1953-2010.
NOTE: The character-of-work estimation procedures for academic and business R&D were revised in 1998; hence, these data are not directly comparable with data reported for earlier years. Furthermore, the methods of collecting data were revised in 2008 for business R&D and in 2010 for academic R&D; thus, these data are not directly comparable with data reported for earlier years. The federal shares include amounts for federally funded research and development centers. The sum of the shares does not equal 1 because these figures do not include university and college and other government (i.e., state) funding.
SOURCE: Data from National Science Foundation (2014a), Appendix Tables 4-3 and 4-4.
tific infrastructure. They carry out long-term scientific, technological, and operational missions with a direct focus on national priorities.
The national laboratories received much attention in the years before and immediately following the Bayh-Dole Act of 1980.4 Along with universities, federal laboratories were explicit targets for initiatives, including Bayh-Dole, focused on developing and commercializing research discoveries. In the wake of the Bayh-Dole legislation, however, the commercial-
4The Bayh-Dole Act allowed universities, small businesses, and nonprofit institutions to pursue ownership of federally funded research inventions. Prior to the act, most inventions were owned by the federal government, although research performers could and did negotiate with funding agencies for the rights to patent and license federally funded inventions.
FIGURE 2-5 Federal, industry, and university performance shares of federally funded U.S. basic and applied research, 1953-2010.
NOTE: The character-of-work estimation procedures for academic and business R&D were revised in 1998; hence, these data are not directly comparable with data reported for earlier years. Furthermore, the methods of collecting data were revised in 2008 for business R&D and in 2010 for academic R&D; thus, these data are not directly comparable with data reported for earlier years. The federal shares include amounts for federally funded research and development centers. The sum of the shares does not equal 1 because these figures do not include university and college (U&C) and other government (i.e., state) funding.
SOURCE: Data from National Science Foundation (2014a), Appendix Tables 4-7 and 4-8.
ization of technology produced by federal laboratories has received only modest attention (see, for example, National Research Council, 2013b, 2013d), for several reasons. Some national laboratories perform confidential work that often is not shared with the broader research community, and compared with research universities, national laboratories have less flow of information through talent and fewer training opportunities.
National laboratories use a number of process indicators to track internally the performance of R&D. In particular, DoE manages these indicators via the Corporate Planning System (CPS), which tracks milestone achievements and financial performance each quarter at the project and contract levels. In addition, the Joule system tracks annual output measures, such as percent progress toward program goals and other sig-
TABLE 2-1 Total U.S. Research Expenditures by Performing Sectors, 1979-2011
|Type of Work and Sector||Percent Distribution|
|Universities and colleges||48.84||46.68||53.99||57.88||52.73||54.63|
|Other nonprofit organizations||8.79||7.90||10.79||11.79||12.60||12.70|
|Universities and colleges||11.72||12.97||11.08||13.19||18.43||20.17|
|Other nonprofit organizations||5.62||3.59||4.67||6.17||7.98||6.91|
NOTES: FFRDC = federally funded research and development center. “Other” includes research not elsewhere classified. The federal shares include amounts for FFRDCs. The character-of-work estimation procedures for academic and business R&D were revised in 1998; hence, these data are not directly comparable with data reported for earlier years. Furthermore, the methods of collecting data were revised in 2008 for business R&D and in 2010 for academic R&D; thus, these data are not directly comparable with data reported for earlier years. The federal shares include amounts for federally funded research and development centers.
SOURCE: Data from National Science Foundation (2014a), Appendix Tables 4-3 and 4-4.
nificant objectives. Finally, the Executive Information System (EIS) acts as a central repository that tracks project- and program-level financial, portfolio, scheduling, and performance information, as well as trends across the Office of Energy Efficiency and Renewable Energy (EERE) portfolio. The CPS, EIS, and EERE are described in detail in U.S. Department of Energy (2007).
IMPEDIMENTS TO THE RESEARCH AND INNOVATION SYSTEMS
Insufficient funding for proof-of-concept research can be an impediment to the translation of basic and applied research discoveries into innovations. In addition to the personal sacrifices of time and effort that academic researchers must make to translate their ideas into marketable innovations, entrepreneurs can find it difficult to advance their discoveries after federal research funding has ended and before sufficient capital can be attracted from private sources to support development and scaling up. Those early stages of spin-off development are the most risky. A lack of short-term gains, a low success rate, and steep competition from other companies discourage private investment (Ewing Marion Kauffman Foundation, 2012). Even a delay in funding can be devastating, as the success of an innovation may be limited by the speed with which it can be scaled up and brought to market.
A further impediment to growth in the research and innovation systems is the widely remarked absence of any executive or congressional entity overseeing the federal research portfolio and performing policy analysis for research.5 NSF produces valuable data (e.g., Science and Engineering Indicators) that could be used in policy analysis. However, the role of its statistics agency, the National Center for Science and Engineering Statistics, differs from that of federal policy analysis agencies, such as the Office of the Assistant Secretary of Planning and Evaluation in the Department of Health and Human Services, as well as from that of statistics agencies that conduct policy analyses, such as the Bureau of Economic Analysis in the U.S. Department of Commerce and the Economic Research Service in the U.S. Department of Agriculture. In 1976, NSF established a Division of Policy Research and Analysis, which funded research on the returns to private and public R&D, but it was disbanded in 1995 (Hall et al., 2014). The lack of such an entity makes it difficult for various agencies to cooperate on common goals and strategies, as well as to foster more
5In 2011, NIH established the Office of Portfolio Analysis to enhance the impact of its funded research by giving NIH administrators the means to evaluate and prioritize current and emerging research areas.
effective research pursuits (National Research Council, 2010b, 2012d; Government-University-Industry Research Roundtable, 2012).
Swings in federal research funding, such as those resulting from the American Recovery and Reinvestment Act and sequestration, also can be detrimental to the growth of the research and innovation systems (Freeman and van Reenen, 2008). This issue is discussed in more detail in Chapter 6.
HOW THE U.S. RESEARCH ENTERPRISE COMPARES WITH THOSE OF OTHER NATIONS
The United States differs from many other nations in that there is no central government administration exclusively in charge of research and innovation (Amsden, 1989; Chang, 2008; Mazzucato, 2011). China is attempting to transition from top-down control to a bottom-up “open door” policy focused on industry; nonetheless, a strong central government cohort largely sets strategic directions, objectives, and policy frameworks for research. In Canada, the prime minister and cabinet formulate overall science, technology, and innovation policy, which is implemented by Industry Canada and the Department of Finance. In East Asia, national governments have taken control of the transfer of research findings to development, targeting specific investments, creating barriers to foreign competition, and establishing industries in the international market.
OECD performed an international comparison of research funding and impacts (OECD, 2013), finding that:
The U.S. remains the world’s largest spender on R&D, accounts for a large share of scientific publications, and for over one third of scientific publications cited in patents. The U.S. still enjoys three distinct advantages: world class universities, a scale that is unmatched, being the central node in the global network of science, technology and innovation [STI] (OECD, 2013).
Compared with other nations, basic research in the United States is closely tied to research universities rather than private research institutes. Furthermore, a relatively high percentage of the U.S. adult population has a tertiary education, although the United States no longer leads other industrial economies in the share of college-age citizens enrolled in higher education. The United States leads in the overall production of scientific publications, producing more than 4 million publications from 2003 to 2011, double the output of second-place China (OECD, 2013). Nevertheless, challenges in U.S. K-12 education may impede the future growth of the U.S. research enterprise, and America’s graduate- and faculty-level
pools of STEM talent rely in part on the strength of K-12 education in other countries, such as China and India. As Gordon (2014) notes:
The United States currently ranks 11th among the developed nations in high school graduation rates and is the only country in which the graduation rates of those aged 25-34 is [sic] no higher than those aged 55-64. . . . A UNICEF report lists the U.S. 18th out of 24 countries in the percentage of secondary students that rank below a fixed international standard in reading and math. The international PISA tests in 2013, again referring to secondary education, rated the U.S. as ranked 21st in reading, 24th in science, and 31st in math. A recent evaluation by the ACT college entrance test organization showed that only 25 percent of high school students were prepared to attend college with adequate scores on reading, math, and science.
The OECD comparison also reveals that the United States remains at the forefront of cutting-edge innovation, with a large and integrated marketplace and efficient capital and equity markets (OECD, 2012b, 2013). The United States is considered above average in business R&D expenditures and venture capital funding, although it is below the international average with regard to the share of university R&D that is funded by industry. However, a number of other studies have noted that the American university system works more closely with industry than is the case in many other economies, perhaps because of an overlap in federal funding that supports university-industry partnerships. (See the discussion of the university-industry relationship in Chapter 3.)
With respect to international collaboration between institutions in different countries, OECD (2013) finds that the United States is the “central node in the global network of science, technology, and innovation.” At the same time, the OECD report singles out the rise of China:
As China invests in its institutions, expands its funding and becomes a more active participant in the global STI network, some of the inherent advantages the United States has enjoyed for decades may be reduced and a new node for STI will begin to form. This creates challenges for the U.S. system, especially given its dependence on highly skilled talent from abroad. To maintain its position, the United States needs to continue to invest and let its system evolve to encompass new developments ranging from sophisticated IT applications to greater recognition of the importance of nontechnological innovation (OECD, 2013, p. 6).
This observation suggests that as China invests in its research institutions, it will likely follow the pattern of development of such other Asian innovation powerhouses as Japan, Singapore, South Korea, and Taiwan by improving its performance in both research and innovation. Therefore, just as was the case during the 1980s and 1990s, the U.S. research
system will be challenged by (and potentially be able to benefit from) the enhanced research and technological prowess of new competitors. It may become increasingly important to encourage foreign students who receive a Ph.D. in the United States to stay in the country and establish careers here.
NSF’s Science and Engineering Indicators for 2014 notes that Asian countries—led by China—are performing an increased share of the world’s R&D, rising from 25 percent in 2001 to 34 percent in 2011, while the share of the world’s R&D performed in the United States and Europe has significantly decreased. An NSF press release for that report states (National Science Foundation, 2014c):
Recognition on the part of national leaders that science and technology (S&T) innovation contributes to national competitiveness, improves living standards, and furthers social welfare has driven the rapid growth in R&D in many countries. China and South Korea have catalyzed their domestic R&D by making significant investments in the S&T research enterprise and enhancing S&T training at universities. China tripled its number of researchers between 1995 and 2008, whereas South Korea doubled its number between 1995 and 2006. And there are indications that students from these nations may be finding more opportunities for advanced education in science and employment in their home countries.
These observations drive home the point that the United States still dominates global research, but Asian nations, including China, are not far behind.