Research Universities and the Future of America: Ten Breakthrough Actions Vital to Our Nation's Prosperity and Security (2012)

5

Action

REPOSITIONING OUR RESEARCH UNIVERSITIES IN A CHANGED WORLD

The emergence of our nation’s research universities and the establishment of a strong federal-university partnership driving research and doctoral education has been a success story for the American people, contributing to our economic prosperity and national goals. Through education and research, American research universities have produced the talent and knowledge that generates innovation critical to economic growth and a high American standard of living.

The Great Recession and the “flattening of the world,” though, have made it clear that there is an urgent need to develop a compelling and effective national strategy for sustaining our world-class research universities that reinforces the partnership of research universities with federal and state governments and expands it to include a larger role for business. It is time to act. In the midst of the Civil War, one of our nation’s deepest crises, Congress passed, and President Abraham Lincoln signed, the Morrill Act, thereby laying the foundation for the land-grant universities that generated a productive agricultural and industrial society. So too, the nation now needs to act in the context of present circumstances to assure the vitality of its research universities in a global knowledge economy. For our research universities to continue to fulfill their obligations to the nation, they must have sufficient resources and a robust infrastructure, sound organizational and administrative structures, a vibrant intellectual community, and the ability to translate research discoveries into societal

benefits. Without these, states and regions that are currently sustained by their research universities may lose their competitive edge, and our nation may fall short in both meeting its national goals and continuing its strong global leadership.

PRINCIPLES

For the past half-century, the research and graduate programs of America’s research universities have been essential contributors to the nation’s prosperity, health, and security. Today, our nation faces new challenges, a time of rapid and profound economic, social, and political transformation driven by the growth in knowledge and innovation. Educated people, the knowledge they produce, and the innovation and entrepreneurial skills they possess, particularly in the fields of science and engineering, have become the keys to America’s future. We have taken stock of the organizational, financial, and intellectual health of our nation’s research universities today and have envisioned the role we would like them to play in our nation’s life 10 to 20 years from now. We can say without reservation that our research universities are, today, the best in the world and an important resource for our nation, yet, at the same time, in grave danger of not only losing their place of global leadership but of serious erosion in quality due to critical trends in public support.

Our vision for strengthening these institutions so that they may remain dynamic assets over the coming decades involves both increasing their productivity and ensuring their strong support for education and research. Therefore, it is essential that the unique partnership that has long existed among the nation’s research universities, the federal government, the states, and business and industry be reaffirmed and strengthened. This will require

•  A balanced set of commitments by each of the partners—federal government, state governments, research universities, and business and industry—to provide leadership for the nation in a knowledge-intensive world and to develop and implement enlightened policies, efficient operating practices, and necessary investments.

•  Use of matching requirements among these commitments that provide strong incentives for participation at comparable levels by each partner.

•  Sufficient flexibility to accommodate differences among research universities and the diversity of their various stakeholders. While merit, impact, and need should continue to be the primary criteria for awarding research grants and contracts by federal agencies, investment in infrastructure should consider additional criteria such as regional and cross-

institutional partnerships, program focus, and opportunities for building significant research capacity.

•  A commitment to a decade-long effort that seeks to both address challenges and take advantage of opportunities as they emerge.

•  A recognition of the importance of supporting the comprehensive nature of the research university, spanning the full spectrum of academic and professional disciplines, including the physical, life, and social and behavioral sciences; engineering; the arts and humanities; and the professions, that enable it to provide the broad research and education programs required by a knowledge- and innovation-driven global economy.

Within this partnership, our research universities—with a historical commitment to excellence, academic freedom, and service to society—must pledge themselves to a new level of partnership with government and business, recommit to being the places where the best minds in the world want to work, think, educate, and create new ideas, and commit to delivering better outcomes for each dollar spent. As articulated in the Millennium Declaration of 2001 on the future of research universities:

For a thousand years the university has benefited our civilization as a learning community where both the young and the experienced could acquire not only knowledge and skills, but the values and discipline of the educated mind. It has defended and propagated our cultural and intellectual heritage, while challenging our norms and beliefs. It has produced the leaders of our governments, commerce, and professions. It has both created and applied new knowledge to serve our society. And it has done so while preserving those values and principles so essential to academic learning: the freedom of inquiry, an openness to new ideas, a commitment to rigorous study, and a love of learning. There seems little doubt that these roles will continue to be needed by our civilization. There is little doubt as well that the university, in some form, will be needed to provide them. The university of the 21st century may be as different from today’s institutions as the research university is from the colonial college. But its form and its continued evolution will be a consequence of transformations necessary to provide its ancient values and contributions to a changing world.¹

RECOMMENDATIONS

With these principles in mind, the committee provides 10 recommendations that the federal government, the states, research universities,

¹ Declaration summarized in James J. Duderstadt, A University for the 21st Century. Ann Arbor, MI: The University of Michigan Press, 2003, p. 324. Original text of the declaration is available at: http://www.glion.org/pub_1999_millennium.aspx (accessed March 23, 2012).

and business and industry can act on to maintain the level of world-class excellence in research and graduate education necessary for the United States to compete, prosper, and achieve national goals for health, energy, the environment, and security in the global community of the twenty-first century. The first four recommendations reaffirm the commitments of each major partner, and the following six enable these commitments. It is important that these recommendations must be implemented together, as they reinforce each other in critical ways.

Universities are today among the most complex institutions in modern society. As James Duderstadt has noted, research universities are comprised of many activities, some nonprofit, some publicly regulated, and some operating in intensely competitive marketplaces. They teach students, conduct research for various clients, provide health care, engage in economic development, stimulate social change, and provide mass entertainment (e.g., athletics). In systems terminology, the modern university is a “loosely coupled, adaptive system,” with a growing complexity, as its various components respond to changes in its environment.²

As the major focus of the charge to the committee was graduate education and research, we preface our recommendations by reinforcing the importance of undergraduate education, both in the research universities that we are examining and in other important institutions, from liberal arts colleges to state universities that also provide undergraduate education. The strength of undergraduate teaching and learning to our nation’s workforce and prosperity and to preparing students who go on to graduate study cannot be overstated.

Similarly, the unusually broad intellectual needs of the nation and the increasing interdependence of the academic disciplines provide compelling reasons why such federal support should encompass all areas of scholarship, including the natural sciences, the social sciences, the humanities, the arts, and professional disciplines such as engineering, education, law, and medicine. Our report and its recommendations are designed to encourage support across all of these areas.³

² James J. Duderstadt, A University for the 21st Century, p. 50.

³ We look forward to a Congressionally requested report on the role of the humanities and social sciences in our nation, due in 2012 from the American Academy of Arts and Sciences.

Actors and Actions—Implementing Recommendation 1:

•  Federal government: The federal government should review and modify those research policies and practices governing university research and graduate education that have become burdensome and inefficient, such as research cost reimbursement, unnecessary regulation, and awkward variation and coordination among federal agencies. (See Recommendations 6 and 7.)

•  Federal government—Congress, Administration, federal science and technology (S&T) agencies: Over the next decade as the economy improves, Congress and the administration should invest in basic research and graduate education at a level sufficient to produce the new knowledge and educated citizens necessary to achieve national goals. As a core component of a national plan to raise total national R&D to 3 percent of gross domestic product (GDP), Congress and the Administration should provide full funding of the amount authorized by the America COMPETES Act that would double the level of basic research conducted by the National Science Foundation (NSF), National Institute of Standards and Technology (NIST), and Department of Energy (DOE) Office of Science as well as sustain our nation’s investment in other key areas of basic research, including biomedical research. Within this investment, as recommend by Rising Above the Gathering Storm,⁴ a portion of the increase should be directed to high-risk, innovative, and unconventional research.

•  Federal government—White House Office of Science and Technology Policy (OSTP), President’s Council of Advisors on Science and Technology (PCAST), U.S. Office of Management and Budget (OMB), National Economic Council (NEC), and Council of Economic Advisors (CEA): On an annual basis in the President’s annual budget request, OMB should develop and present, in coordination with OSTP, a federal science and technology budget that addresses priorities for sustaining a world-class U.S. science and technology enterprise. On a quadrennial basis, OSTP, in conjunction with PCAST, and OMB, in conjunction with the NEC and CEA, should review federal science and technology spending and outcomes, internationally benchmarked, to ensure that federal S&T spending is adequate in size to support our economy and appropriately targeted to meet national goals. We recommend that this process consider

⁴ National Academy of Sciences, National Academy of Engineering, and Institute of Medicine, Rising Above the Gathering Storm: Energizing and Employing America for a Brighter Economic Future, Washington, DC: National Academies Press, 2007.

U.S. global leadership, a focus on developing new knowledge, balance in the science and technology portfolio, reliable and predictable streams of funding, and a commitment to merit review.

Budget Implications

This recommendation calls for stable and effective federal research policies and practices, the budget implications of which are outlined under several recommendations below. The recommendation also aims to ensure robust financial support for critical federal basic research programs. It supports funding increases that Congress has already authorized through the America COMPETES Act for the doubling of funding for the NSF, NIST, and DOE Office of Science. These increases target stronger investment in physical sciences and engineering research, but do not imply any disinvestment in critical fields such as the life sciences and social, behavioral, and economic sciences. Indeed, we recommend Congressional action to at least maintain current levels of funding for basic research across other federal agencies, including the National Institutes of Health (NIH), as adjusted for inflation. Research universities, along with other research performers (national laboratories, nonprofit research and development organizations, and industry), will only benefit from these actions through their success in competing for federal grants and contracts from these agencies.

Expected Outcomes

Supportive federal research policies would ensure stable funding and cost-efficient regulation sufficient to enable corresponding university investment in research facilities and graduate programs. By completing the funding of the America COMPETES Act, the nation would achieve a balanced research portfolio capable of driving innovation necessary for economic prosperity. As research and education are deliberately intertwined in our American research universities, such funding will also ensure that we continue to produce the scientists, engineers, physicians, teachers, scholars, and other knowledge professionals essential to the nation’s security, health, and prosperity.

Discussion

Context

Nations around the world have recognized the importance of investment in research and doctoral education, both of which build their

nation’s research universities, contribute to economic growth, and improve global competitiveness. In most instances, they have developed comprehensive national strategies designed to strengthen their research base and their institutions to compete for students and faculty, resources, and reputation (see Box 4-1).

The United States has begun to lag on a key, internationally recognized indicator of national investment in the development of new knowledge: national (public and private) R&D expenditures as a percentage of GDP. As shown in Figure 5-1.1, public and private R&D expenditures in the United States have hovered between 2.5 and 2.8 percent of GDP over the last three decades. It stood at 2.79 percent in 2008. By comparison, as shown in Figures 5-1.1 and 5-1.2, Japan has increased its national R&D funding from about 2.8 percent of GDP in 1996 to 3.4 percent in 2007, while South Korea has increased its spending even further, reaching 3.5 percent of GDP in that year. While R&D in Germany as a percent of GDP is slightly lower than that of the United States, its nondefense R&D as a percentage of GDP is higher than that of the United States and the gap between the two countries is growing. The annual rate of growth in national R&D expenditures was 5-6 percent for the United States and the European Union (EU-27), while rates of growth for many Asian countries were far higher. China’s annual growth in national R&D expenditures was 20 percent for the period 1996 to 2007.⁵

Targets

Embedded in a broader federal innovation strategy that addresses national research and development priorities, the nation must develop a framework of national funding goals and supportive policies that sustain the nation’s research universities at world-class levels. The current Administration developed and issued the National Innovation Strategy in September 2009 and presented an updated version drafted by the National Economic Council, the Council of Economic Advisors, and the Office of Science and Technology Policy in February 2011. This strategy provides a broad policy context and includes a short section focused on strengthening and broadening “American leadership in fundamental research.” This provides an excellent foundation from which to craft a more detailed strategy for sustaining the nation’s R&D enterprise, fundamental research, and U.S. research universities. For example, it sets a national goal “for America to invest more than three percent of our GDP in public and private research and development” noting that “this investment rate

⁵ National Science Board, Science and Engineering Indicators, 2010, (NSB 10-01), Arlington, VA: National Science Foundation, 2010, Figures 4-13 and 4-16, pages 4-35 and 4-36.

FIGURE 5-1.1 Gross expenditures on R&D as share of gross domestic product, for selected countries: 1981-2007.

Source: National Science Board, Science and Engineering Indicators 2010. (NSB 10-01) Arlington, VA: National Science Foundation, 2010, Figure 4-16, page 4-36.

FIGURE 5-1.2 Gross domestic expenditures on R&D by United States, EU-27, OECD, and selected other countries: 1981-2007.

Source: National Science Board, Science and Engineering Indicators 2010. (NSB 10-01) Arlington, VA: National Science Foundation, 2010, Figure 4-13, page 4-35.

will surpass the level achieved at the height of the space race, and can be achieved through policies that support basic and applied research, create new incentives for private innovation, promote breakthroughs in national priority areas, and improve [science, technology, engineering, and mathematics] STEM education.”⁶ Table 5-1.1 displays 2008 U.S. R&D in current dollars and as a percentage of GDP and shows total national R&D spending in that year at 2.79 percent. An increase to 3 percent of GDP would potentially lift all components of R&D, including federally funded, university-performed research.

Indeed, the committee recommends federal R&D appropriations levels that would sustain and enhance university-based research. We strongly support the goals articulated by Rising Above the Gathering Storm and au-

⁶ The White House, A Strategy for American Innovation: Securing Our Economic Growth and Prosperity, February 2011.

TABLE 5-1.1 U.S. R&D, 2008 Expenditures

Sources: NSF/NCSES, Academic R&D Expenditures, Fiscal Year 2009, Tables 1, 2, and 3. Available at: http://www.nsf.gov/statistics/nsf11313/content.cfm?pub_id=4065&id=2 (accessed September 4, 2011). NSF/NCSES, national patterns of R&D Resources, Tables 6 and 13. Available at: http://www.nsf.gov/statistics/natlpatterns/ (accessed September 4, 2011).

thorized in the America COMPETES Act of 2010 that would increase the support of basic research key to sustaining the nation’s innovation necessary for prosperity and national security by doubling the budgets of the NSF, DOE Office of Science, and NIST. We also strongly urge that federal appropriations for basic research in support of other key national goals such as health (NIH), defense (Department of Defense [DOD]), space (National Aeronautics and Space Administration), and agriculture (U.S. Department of Agriculture) be sustained at least at the rate of inflation. Figure 5-1.3 shows that federally funded, university-performed R&D as a percentage of GDP increased from 1998 to 2005, while the NIH budget doubled, from about 0.17 percent to about 0.24 percent, and has since decreased to 0.22 percent. This mirrors the flattening of federally funded university R&D and the decline of federally funded research generally (in constant dollars) seen in Figures 5-1.4 and 5-1.5. Providing the appropriations we recommend can help reverse these declines and ensure that we are strongly investing for the future.

Principles

The following are important principles for federal R&D funding, many articulated in previous National Academies reports:

FIGURE 5-1.3 Federally funded, university-performed research and development as a percentage of GDP, 1990-2008.

Source: NSF/NCSES, Academic R&D Expenditures, Fiscal Year 2009, Table 1.

Available at: http://www.nsf.gov/statistics/nsf11313/content.cfm?pub_id=4065&id=2 (accessed September 4, 2011). NSF/NCSES, National Patterns of R&D Resources, Table 13. Available at: http://www.nsf.gov/statistics/natlpatterns/ (accessed September 4, 2011).

•  Focus on global leadership: At the end of the Cold War, the U.S. Congress asked the National Academies to identify priority areas for future federal investment and to provide a foundation upon which federal science and technology (FS&T) budgetary policy can be built and analyzed. In Science, Technology, and the Federal Government: National Goals for a New Era (1993), the National Academies recommended two goals to guide federal investment in science and technology:

o  First, the United States should be among the world leaders in all major areas of science. Achieving this goal would allow this nation to quickly apply and extend advances in science wherever they occur.

o  Second, the United States should maintain clear leadership in some areas of science. The decision to select a field for leadership would

FIGURE 5-1.4 University-performed research and development and federally funded, university-performed research and development, 1990-2008 (in millions of constant 2000 dollars).

Sources: NSF/NCSES, Academic R&D Expenditures, Fiscal Year 2009, Table 1. Available at: http://www.nsf.gov/statistics/nsf11313/content.cfm?pub_id=4065&id=2 (accessed September 4, 2011).

Data adjusted for inflation using implicit price deflator in NSF/NCSES, national patterns of R&D Resources, Table 13. Available at: http://www.nsf.gov/statistics/natlpatterns/ (accessed September 4, 2011).

be based on national objectives and other criteria external to the field of research.⁷

These remain critical national goals for federal R&D investment.

•  Focus on new knowledge: The National Academies’ Allocating Federal Funds for Science and Technology (1995) urged the Executive Office of the President and the U.S. Congress to develop a more coherent budget process for determining federal investment in programs that create new knowledge and technologies (i.e., the federal science and technology budget). It recommended that the President present annually a Federal

⁷ National Academy of Sciences, National Academy of Engineering, Institute of Medicine, Science, Technology, and the Federal Government: National Goals for a New Era. Washington, DC: National Academy Press, 1993.

FIGURE 5-1.5 Trends in and characteristics of national, industrial, and federal R&D, 1954-present.

Note: Fiscal 2012 is Administration Budget Proposal, not yet enacted at time of original figure composition.

Source for AAAS Figures: Patrick Clemins, Research and Development in the Federal Budget Presentation, AAAS S&T Policy Forum, May 5, 2011. Available at: http://www.aaas.org/spp/rd/forum2011/presentations/PatrickClemins_AAASForum2011.pdf (accessed September 4, 2011).

Science and Technology budget proposal that addresses priorities for sustaining a world-class U.S. science and technology enterprise.⁸ Within broader R&D appropriations, FS&T more narrowly focuses on the production of new knowledge and is roughly analogous to basic and applied research. This FS&T presentation was adopted in the late 1990s by the U.S. Office of Management and Budget and was continued into the early 2000s.⁹ The presentation has since been dropped from the President’s annual federal budget proposal. We recommend that it be restored.

•  Balance in the science and technology portfolio: Rising Above the Gathering Storm recommended to “increase the federal investment in long-term basic research by 10% each year over the next 7 years through reallocation of existing funds or, if necessary, through the investment of new funds.” The report also recommended “special attention should go to the physical sciences, engineering, mathematics, and information sciences and to Department of Defense (DOD) basic-research funding. This special attention does not mean that there should be a disinvestment in such important fields as the life sciences or the social sciences. A balanced research portfolio in all fields of science and engineering research is critical to U.S. prosperity.”¹⁰

•  Focus on accelerating scientific and technological advances: Rising Above the Gathering Storm also argued that “increasingly, the most significant new scientific and engineering advances are formed to cut across several disciplines” and that federal research agencies should “allocate at least 8% of the budgets of federal research agencies to discretionary funding…focused on catalyzing high-risk, high-payoff research of the type that often suffers in today’s increasingly risk-averse environment.”¹¹

•  Reliable and predictable streams of funding: The nation will increase the performance of its research enterprise by providing steady, predictable streams of funding for research over time. The last decade has seen damaging fluctuations in research appropriations. Instead, the federal government should provide steady, sustainable, predictable support for university research over the longer term. This would enable universities to plan their own investments in research, and it would make federal research expenditures more effective and efficient.

•  Commitment to merit review: The nation’s investments in university research should continue to emphasize the characteristics that have

⁸ National Research Council, Allocating Federal Funds for Science and Technology. Washington, DC: National Academy Press, 1995, p. v.

⁹ National Academy of Sciences, National Academy of Engineering, Institute of Medicine, Observations on the President’s Fiscal Year 2003 Federal Science and Technology Budget. Washington, DC: National Academies Press, 2002.

¹⁰ National Academy of Sciences et al., Rising Above the Gathering Storm.

¹¹ Ibid.

made it the most effective research investment in the world a research agenda driven by science and scientific opportunity and a commitment to peer-reviewed and competitively awarded research grants. In particular, the committee strongly encourages both federal sponsors and universities to avoid the use of earmarks or other political mechanisms for determining grant awards that both increase costs and erode research quality.

•  Importance of evaluation: Rising Above the Gathering Storm also recommended that federal investments “should be evaluated regularly to realign the research portfolio to satisfy emerging needs and promises—unsuccessful projects and venues of research should be replaced with research projects and venues that have greater potential.”¹²

Actors and Actions—Implementing Recommendation 2:

•  State governments: States should move rapidly to provide their public research universities with sufficient autonomy and agility to navigate an extended period with limited state support. (See also regulatory environment, below.)

•  State governments: For states to compete for the prosperity and welfare of their citizens in a knowledge- and innovation-driven global economy, the advanced education, research, and innovation programs provided by their research universities are absolutely essential. Hence, as state budgets recover from the current recession, states should strive to restore and maintain per-student funding for higher education, including public research universities, to the mean level for the 15-year period 1987-2002, as adjusted for inflation.¹³

•  Federal government: To provide further incentives for state actions to protect the quality of public research universities as both a state

¹² Ibid.

¹³ A 15-year period was used so as to ensure the funding recommendation was not unduly influenced by year-to-year fluctuations in state appropriations. The year 2002 was used as the endpoint of the period, as that year represents the beginning of a period of significant decline in appropriations.

and a national asset, federal programs designed to stimulate innovation and workforce development at the state level, including those recommended in this report, should be accompanied by strong incentives to stimulate and sustain state support for their public universities.

Budget Implications

This recommendation addresses the alarming erosion in state support of higher education over the past decade that has put the quality and capacity of public research universities at great risk. While the committee urges the states to strive to restore over time appropriation cuts to public research universities estimated to average 25 percent (and ranging as high as 50 percent for some universities),¹⁴ it acknowledges that current state budget challenges and shifting state priorities may make this very difficult in the near term. Hence, the committee views as equally important a strong recommendation that the states provide their public research universities with sufficient autonomy and ability to navigate what could be an extended period with inadequate state funding. The committee strongly believes that such recommendations are in the long-term interests of both the states and the nation.

Expected Outcomes

State appropriations per enrolled student have declined by 25 percent or more over the past two decades, resulting in the need for universities to increase tuition or reduce activities, or quality. As states strive to compete in a knowledge- and innovation-driven global economy, restoring state appropriations to levels sufficient to maintain advanced education, research, and innovation programs provided by research universities is absolutely essential for the prosperity and welfare of their citizens. Increasing the autonomy and agility of public research universities should increase their efficiency and productivity as well as their ability to re-

¹⁴ The National Science Board reports, “Over the decade [2002 to 2010], per-student state support to major research universities dropped by an average of 20 percent in inflation-adjusted dollars. In 10 states, the decline ranged from 30 percent to 48 percent.” National Science Board, Science and Engineering Indicators 2012, p. 8-68. Available at: http://www.nsf.gov/statistics/seind12/pdf/c08.pdf (accessed March 8, 2012). The states have enacted further and deeper cuts in 2011 and 2012, which suggests an overall decline for 2002-2012 of at least 25 percent. For example, the State Higher Education Executive Officers Association recently reported, “FY 2012 state appropriations [for higher education] (including a small residual of ARRA funding) were $72.5 billion, a decrease of 7.6 percent from $78.5 billion in FY 2011.” See SHEEO, “Commentary on FY 2012 state appropriations for higher education,” press release, January 23, 2012. Available at: http://grapevine.illinoisstate.edu/tables/FY12/SHEEO%20Commentary%20(2).pdf (accessed March 8, 2012).

spond to changing state and regional needs during an extended period when states may not be able to restore adequate support.

Discussion

Support for public research universities is a national challenge of immense importance, since these institutions produce the majority of advanced-degree recipients and basic research for the United States. Any loss of world-class quality for America’s public research institutions seriously damages national prosperity, security, and quality of life. In fact, for many state research universities, the national importance of these institutions is underscored by the fact that their federal support, through student financial aid and research grants, now exceeds state appropriations. But states still have a critically important role to play—one that supports these institutions and meets the local and regional needs of states and their residents.

The nation’s public research universities face great risk as the states that support them not only face serious financial challenges due to the recent recession, they also often no longer give priority to the support of graduate education and research. With increasing national and even international mobility of campus-generated knowledge and doctorates, states may support undergraduate education and the goal of broadening access at world-class levels, but they are less inclined to invest in research and graduate education at their public research universities given the uncertainty in their ability to capture the returns on their investments. However, state leaders should realize that a restoration of an adequate level of support for public postsecondary education generally—and their research universities more specifically—remains very much in their long-term interest. These institutions provide both the talent and ideas necessary for regional economic growth and for other local needs, including health, public safety, transportation planning, cultural enrichment, new elementary and secondary teachers, and more. The importance of highly educated citizens and universities with the ability to discover new knowledge, develop innovative applications of research, and transfer them into the marketplace is critical to the prosperity and welfare of the states just as it is to the nation. Yet the benefits of graduate education and university research are public goods whose high mobility extends far beyond state boundaries. Hence, with budget constraints and the shifting priorities of aging populations, many states have concluded that they can no longer justify giving high priority to sustaining their public research universities at world-class levels. Yet such actions represent not only a marked decline in regional advantage at the state level but also seriously harm the national interest. To be sure, not all states have the capacity to build and

FIGURE 5-2.1 Public FTE enrollment and state educational appropriations per FTE student, U.S., fiscal 1985-2010 (constant dollars).

Note: Net tuition revenue used for capital debt service is included in the above figures.

Source: State Higher Education Officers (SHEEO), State Higher Education Finance 2010, Figure 3, page 20. Available at: http://www.sheeo.org/finance/shef_fy10.pdf.

maintain large, comprehensive research universities at world-class levels, but all states have the capacity to focus resources to build high-quality graduate and research programs in select areas of high local priority.

Indeed, the states vary significantly in both the levels of, and trends in, support they provide for public higher education.¹⁵ However, for the nation as a whole, state postsecondary educational appropriations per full-time equivalent (FTE) student decreased 3.1 percent in constant dollars from FY 2005 to FY 2010. A 7.2 percent decrease in the past year due to the impact of the recession on the states wiped out interim gains from 2005 to 2008. In fact, as shown in Figures 5-2.1 and 5-2.2, state educational appropriations per FTE student in constant dollars have ebbed and flowed over time, but there has been a long-term downward trend since the late 1980s, and they were at their lowest levels in constant dollars in FY 2010

¹⁵ State educational appropriations per FTE at public higher education institutions in FY 2010 varied from a low of $3,781 in Colorado to $13,090 in Wyoming. The 5-year change in educational appropriations per FTE varied from–27.4 percent in Rhode Island to +26.6 percent in North Dakota. State Higher Education Executive Officers (SHEEO), State Higher Education Finance, FY 2010, pp. 10, 29, available at: http://www.sheeo.org/finance/shef_fy10.pdf.

FIGURE 5-2.2 Real state and local appropriations per student (FTE) in public research universities, by very high research and high research institutions, fiscal 1987-2007 (2007 constant dollars).

Source: Peter M. McPherson, President, Association of Public and Land-grant Universities, Presentation to the NRC Committee on Research Universities, November 2010.

than at any time in the past 25 years.¹⁶ This trend has resulted not just from the recession. State Higher Education Executive Officers (SHEEO) reports, “The proportion of state and local tax revenue allocated to higher education declined from 6.9 percent in 1998 to 6.6 percent in 2008.”¹⁷

There are important consequences for public research universities and their students that flow from these cuts in state appropriations. As shown in Figure 5-2.3, per FTE student expenditures at public institutions are lower than those at private institutions and have been growing more slowly. As shown in Figure 5-2.4, the median salaries of assistant, associate, and full professors at public institutions have decreased over time relative to their peers at private institutions. Consequently, the private institutions have the upper hand in hiring and have the additional ability to lure away “star” professors from public research universities. As shown in Figure 5-2.5, the ratio of students to full-time faculty has been lower in private institutions, and the gap between public and private institutions

¹⁶ Ibid., pp. 7, 20, 29.

¹⁷ Ibid., p. 10.

FIGURE 5-2.3 Total expenditures per FTE student at private and public nonprofit institutions, by institution category and type of expenses, 1999, 2004, 2008, and 2009 (2009 constant dollars).

Source: Donna M. Desrochers and Jane V. Wellman, Trends in College Spending, 1999-2009, Where does the money come from? Where does it go? What does it buy? A report of the Delta Cost Project. Available at: http://www.deltacostproject.org/resources/pdf/Trends2011_Final_090711.pdf (accessed September 16, 2011).

for this indicator is growing. This presumably affects the educational experience of students.

While the current budget difficulties faced by the states call for choices to be made, high-quality public research universities remain essential to providing America’s citizens with the advanced education and research necessary to compete in a knowledge- and innovation-driven global economy. As budgets revive, states should give high priority to restoring and maintaining funding sufficient to keep their research universities at world-class levels. The actions that each state may take will vary according to their particular needs and circumstances; yet the national aggregate of state postsecondary education appropriations per FTE student must be restored to at least the inflation-adjusted levels that existed in 1988 as

FIGURE 5-2.4 Ratio of salaries of full, associate, and assistant professors at private institutions to those at public institutions, 1976, 1986, 1999, and 2007.

Source: Peter M. McPherson, President, Association of Public and Land-grant Universities, Presentation to NRC Committee on Research Universities, November 2010.

FIGURE 5-2.5 Ratio of students to full-time faculty, for public and private research universities, 1989, 1997, and 2006.

Source: Peter M. McPherson, President, Association of Public and Land-grant Universities, Presentation to Committee on Research Universities, November 2010.

rapidly as possible. To be sure, there are other competing claims on state funding, but the recommended average increase per state is reasonable, attainable, and a wise investment in the future.

Increased state funding, moreover, should be targeted to their best use in the local and regional economic environment. For example, increased state funding could be used to

•  Fund expansion of undergraduate and graduate education at public research universities, including disciplines linked to the competitiveness of the state for retaining traditional business and attracting new business.

•  Provide funding for research work undertaken by public research universities, including but not limited to research supportive of state and regional business activity.

There will be other reasonable targets that reflect local needs and conditions.

Critically important, the committee agrees with the Association of American Universities (AAU) that federal funding should be used to leverage, not substitute for, state funding. “The allocation of federal funds in support of public research universities cannot be a substitute for state funds; maintenance of efforts by states should be committed and audited. Where possible, federal funds could be employed as an incentive for state funding, for example with support for scientific infrastructure such as new research facilities, facility modernization and research instrumentation.”¹⁸ Indeed, federal and state governments could enter into complementary matching opportunities to advance fields that are critical to societal needs and incentivize collaboration to a greater degree on large national and state projects.

In addition to restoring appropriations, states should provide public research universities with the autonomy and agility to restructure their operations to enable them to survive current public underfunding and to position them to capitalize on future opportunities as the economy improves. Greater autonomy and agility can be accomplished in several ways, including

•  Restructuring university governance so that boards better represent the broader “public” beyond just the states that now provide such small portions of overall university budgets.

¹⁸ Association of American Universities, Recommendations to the National Research Council Committee on Research Universities, February 2, 2011.

•  Allowing public research universities to set tuition and fees for their campuses.

•  Allowing public research universities to do their own procurement independent of the state government.

•  Allowing public research universities to obtain the bonds they need without state approval in order to move more quickly in the construction of dormitories, research facilities, and other buildings necessary to maintain high-quality education, research, and service. (In addition, moving construction projects ahead quickly creates jobs that are badly needed in this time of high unemployment.)

•  Providing incentives for public universities to form regional compacts with other universities for the purpose of ensuring that programs that might not be of scale in a single university continue to be collaboratively offered or otherwise made available within the region in a cost-effective manner.

•  Reducing state regulations that have attempted to take the place of university administrators and of university governing bodies that are already in place to effectively oversee the strategy and performance of the university as a whole, consistent with the particular mission and distinctive characteristics of the institution.

•  Conducting complete reviews of state compliance requirements and regulations that affect research with a focus on identifying their costs, making judgments about their efficacy, and producing recommendations about modifying or eliminating compliance requirements and regulations as appropriate.

This list is illustrative. There are certainly more, perhaps even better, steps that may be taken. As states undertake steps in this area, they should examine experiments already under way, such as the Restructured Higher Education Financial and Administrative Operations Act (Restructuring Act) that provides a framework for transforming public higher education in Virginia. As the University of Virginia reports, “The Restructuring Act grants Virginia’s public institutions of higher education including the University of Virginia greater financial and administrative autonomy allowing them to more effectively and efficiently manage day-to-day operations. In exchange for increased autonomy, each institution must commit to meet specific statewide goals. The Act provides for three levels of autonomy; the University, along with Virginia Tech, the College of William & Mary, and Virginia Commonwealth University, is currently operating at the highest degree of autonomy.”¹⁹

¹⁹ See http://www.virginia.edu/restructuring/ (accessed December 13, 2011).

Actors and Actions—Implementing Recommendation 3:

•  Federal government: Continue to fund and expand research support mechanisms that promote collaboration and innovation.

•  Federal government: Within the context of also making the R&D tax credit permanent, implement new tax policies that incentivize business to develop partnerships with universities (and others as warranted) for research that results in new U.S.-located economic activities.

•  Business, universities: The relationship between business and higher education should evolve into more of a peer-to-peer nature, stressing collaboration in areas of joint interest rather than the traditional customer-supplier relationship in which business procures graduates and intellectual property from universities.

•  Business, universities: Business and universities should work closely together to develop new graduate degree programs that address strategic workforce gaps for science-based employers.

•  National laboratories, business, universities: Collaboration among research by the nation’s national laboratories, business, and universities should also be encouraged, since the latter’s capacity for large-scale, sustained research projects both supports and depends critically on both the participation of university faculty and graduate students and the marketplace.

•  Universities: Improve management of intellectual property to improve technology transfer.

Budget Implications

Tax policies that create incentives for new university-industry research and development partnerships will have a cost to the federal budget as a “tax expenditure.” Although we are not in a position to estimate what that cost would be, it would be a relatively minor component of the cost of current proposals to make permanent the R&D tax credit.

Expected Outcomes

Effective use of research support mechanisms that promote collaboration will lead to the creation and efficient use of knowledge to achieve national goals.

The outcomes from the new tax policies would be new research partnerships; new knowledge and ideas; new products, processes, and industries located in the United States; economic growth; and new jobs. The outcomes from these efforts would be the creation of new partnerships, new knowledge and ideas, achieving national goals in key policy areas, and the economic growth and jobs that result from new activity.

Improvements in university management of intellectual property will result in more effective dissemination of research results, generating economic activity and jobs.

Discussion

Intellectual property—the ideas and knowledge that are key to innovation—moves out of universities to business through a variety of important paths. Chief among these is the education of students who leave to work in businesses, bringing their new knowledge with them. In some cases, they are able to innovate within their new environments based on what they have learned; in others, they create start-ups that represent new business and, in some important cases, new industries. Other means by which new ideas are disseminated include publication of scholarly papers, faculty consulting, creation of spin-off companies by faculty, patenting and licensing, and business-university research partnerships. This recommendation is primarily focused on the last two of these.

Research outcomes can be deepened and multiplied through improvements in the dissemination, translation, and commercialization of research results. Society’s problems are ever-more complex and need to be addressed by new learning and discovery. Industry-university-government engagement will take both prodding and resources to greatly expand. The most productive models of such engagement are interdisciplinary, flexible, and interconnected or networked—just like the problems that they need to address. Universities are not presently organized for this reality, and a deeper collaboration with industry and governments will provide further external pressure to make our institutions better able to address multidisciplinary societal problems. There is an equal responsibility for business to explore how new partnerships can benefit research, commercial innovation, and society.

To accomplish this, governments, industry, philanthropy, and academia should collaborate to enhance innovation and the dissemination

of research results. Improvements in university management of intellectual property and technology transfer from university laboratories to industry and the marketplace and stronger university partnerships with industry, national laboratories, and philanthropy are key ingredients for ensuring that research leads to innovation and jobs. This is the point at which the strong investments in Recommendations 1 and 2 lead to the breakthroughs needed to power the economy, enhance our culture and society, and help us achieve national goals.

Over the last two decades, long-term changes in the structure of the national research enterprise that have affected the roles and partnerships that all actors play are as follows:

•  Corporate practices for the funding and performance of basic research have shifted. Financial pressures have led to the disappearance of large industrial laboratories in many industries and, therefore, to new

FIGURE 5-3.1 Industry-funded basic research by perfomer, 1953-2008 (millions of constant 2000 dollars).

Source: National Science Foundation, National Center for Science and Engineering Statistics, National Patterns of Research and Development, 2008 Data Update, Table 6. Available at: http://www.nsf.gov/statistics/nsf10314/content.cfm?pub_id=4000&id=2 (accessed April 22, 2012).

FIGURE 5-3.2 U.S. basic research by performing sector, 1980-2008 (millions of constant 2000 dollars).

Source: National Science Foundation, National Center for Science and Engineering Statistics, National Patterns of Research and Development, 2008 Data Update, Table 2, Available at: http://www.nsf.gov/statistics/nsf10314/content.cfm?pub_id=4000&id=2 (accessed April 22, 2012).

strategies for obtaining the productive knowledge that allows for innovation and the development of new processes and products. As seen in Figures 5-3.1 and 5-3.2, industry funding and performance of basic research has recently been wildly erratic, but over the long term, essentially flat. Meanwhile university-performed research and industry-funded university research have grown. Corporate funding for basic research has increased on campuses, creating both new opportunities for research and commercialization and challenges such as the management of conflict of interest.

•  The missions of national laboratories—particularly those that have historically been focused on nuclear weapons research—have also shifted since the end of the Cold War. Changes in mission may be accompanied by new partnerships between research universities and these laboratories, many of which are university-managed, to increase the productivity of both.

•  Philanthropy has played a strong role in the history of the American research university. Philanthropic funding continues to play a powerful role, both through charitable giving that strengthens institutions and

through grants and gifts for facilities and lines of research. There was a significant decline in gifts during the recession, which added to the substantial challenge associated with declines in endowment value. The economy, endowments, and giving are all rebounding now, providing an opportunity to examine the best opportunities for philanthropic funding going forward.

Of particular note in these trends is the insufficiency of truly transformative R&D of the type that is usually attributed to the Advanced Research Projects Agency (ARPA) and Bell Laboratories in earlier decades and effective translational R&D capable of coupling fundamental scientific discovery with technological innovation and the marketplace, once characteristic of Bell Labs in industry or agricultural experiment stations in higher education.

Three areas provide opportunities for new approaches that may revitalize the U.S. research enterprise and enhance the contributions of academic research to meeting important national goals such as innovation and economic growth, national security, health, and energy. These areas are as follows:

1.   Technology transfer from universities

2.   Research-support mechanisms that promote collaboration and innovation

3.   Incentives for industry participation in partnerships with universities

These practices can be broadly implemented, though the character of their application may diverge depending on differences among industries and fields (e.g., information technology, advanced materials, biomedical) in their traditional patterns of university-industry roles and collaboration.

First, patent reform can increase the effectiveness of technology transfer from universities to the marketplace. The enactment of patent reform in September 2011 through the America Invents Act is a start in overhauling the general patent system. A recent National Research Council (NRC) report evaluates more specifically the university technology transfer system established more than 30 years ago through the Bayh-Dole Act (P.L. 96-517, the Patent and Trademark Act Amendments of 1980). This system, in which universities control intellectual property (IP) that results from research supported by federal agencies, has been much more effective than the previous practice of government control. Faculty invention disclosures, patenting, licensing, and other metrics of commercialization have all increased, without serious interference in other university missions. But the NRC report identifies several areas where universities

and the federal government could make significant improvements. For example, universities have uneven capabilities in the area of technology transfer, and there is a general need for universities to introduce more stakeholder involvement and accountability into the technology transfer system. The report suggests good practices for universities to follow in patenting, licensing, material transfer, and launching start-up companies based on university-developed technologies. In addition, the report recommends that universities take steps to streamline licensing negotiations with industry. Finally, the report calls for establishing an effective federal oversight framework with clear responsibilities and a supportive data collection system. During the course of this study, Congress passed and the President signed into law, on September 16, 2011, the America Invents Act, which overhauls the patent system, implementing many of the reforms recommended in the NRC report. This is a great step forward toward improving the flow of technological innovation.²⁰

Second, the federal government can continue and expand support for collaborative research mechanisms that promote knowledge generation and transfer and innovation. Regarding research support mechanisms that promote collaboration and innovation, the committee reaffirms that universities remain the primary source of fundamental science and engineering discoveries and that federal research investments are necessary to sustain this knowledge generation. In addition to merit-reviewed grants to individual investigators, though, federal agencies support university research in a number of ways. NSF has several long-standing “centers” programs, such as the Science and Technology Centers, Materials Research Science and Engineering Centers, Engineering Research Centers, and Industry University Cooperative Research Centers. In order to provide the American research enterprise and industry with better balance and cohesive linkages among sectors, the federal government should continue to develop new research paradigms that address current shortcomings, applying these paradigms judiciously across fields and industries. Several federal agencies (DOE, Department of Commerce, NSF, and DOD) have already launched such programs that merit strong support from the federal government and more can be developed. Examples include innovation hubs focused on translational research (proposed by DOE and Commerce) and the ARPA-Energy and ARPA-Education organizations for transformational R&D (proposed by Rising Above the Gathering Storm

²⁰ See http://www.whitehouse.gov/the-press-office/2011/09/16/president-obamasigns-america-invents-act-overhauling-patent-system-stim (accessed September 19, 2011). National Research Council. Managing University Intellectual Property in the Public Interest. Washington, DC: National Academies Press, 2011. National Research Council. A Patent System for the 2lst Century. Washington, DC: National Academies Press, 2006.

and PCAST). Clusters of these initiatives should be launched at scale and adequately funded from multiple sources (federal, state, industry, universities). Box 5-3.1 outlines additional federal initiatives under way to move ideas from the laboratory to the market.

Third, the committee looked at incentives for industry participation in partnerships with universities viewing this as the critical area that requires additional action. Support for university research is covered in the current

federal R&D tax credit, but several factors prevent it from being utilized optimally. For example, the year-to-year renewal of the tax credit prevents companies from making longer commitments to university partnerships. To overcome this and other barriers, we recommend the federal government institute new tax policies that create incentives for multiyear university-industry research and development collaboration. These new policies, which may be created through tax credits, will be provided to businesses that invest in

university-performed basic research. Research results that flow from these investments must be used in the development of U.S.-located economic activities so that the returns of the taxpayer investments are captured here to create economic growth and new jobs.

The National Academies report Rising Above the Gathering Storm strongly recommended strengthening and making permanent the research and experimentation tax credit, one of the most effective government tax incentives—clearly working as effectively as government spending on R&D in promoting research and development. Business accounts for about two-thirds of all R&D spending in the United States, providing laboratories and jobs for many of our scientists and STEM graduates. Although most business R&D is applied, the distinction between applied and basic is increasingly an artificial one, and business is the channel through which basic ideas developed in research universities reach the marketplace, often with the support of the R&D tax credit. Strengthening the R&D tax credit generally, particularly at a time when our incentives are declining relative to those of other countries, and creating specific opportunities within the credit to create productive research relationships between businesses and universities, will provide important additional tools for enhancing national innovation.

Finally, university management of intellectual property must be improved. The two main problems in this area are the amount of money that universities put into their technology-licensing offices and the institutional goals of the offices. First, much of the uneven approach to technology licensing results from the high cost of maintaining patents (about $20,000 per U.S. patent; more than $100,000 to protect foreign markets) until licensees are found, creating wide variations in approach depending on how much a university is willing to subsidize its licensing operations. University intellectual property policies should not only be simplified and streamlined but also better standardized across higher education, so that each negotiation between industry and university can move forward according to commonly accepted procedures. Second, a recent discussion of the trade-offs involved in institutional goals points out the near impossibility of satisfying the many competing goals of different stakeholders in the licensing process.²¹ The focus of technology transfer should first be on economic stimulus and only secondarily on revenue return and support of faculty.

²¹ Michael Sharer and Timothy L. Faley, The strategic management of the technology transfer function—Aligning goals with strategies and tactics, les Nouvelles, September 2008, p. 170.

Actors and Actions—Implementing Recommendation 4:

•  Universities: The nation’s research universities should set and achieve bold goals in cost-containment, efficiency, and productivity in business operations and academic programs. Universities should strive to constrain the cost escalation of all ongoing activities—academic and auxiliary—to the inflation rate or lower through improved efficiency and productivity. Beyond the implementation of efficient business practices, universities should review existing academic programs from the perspectives of centrality, quality, and cost-effectiveness, adopting modern instructional methods such as cyberlearning, and encouraging greater collaboration among research investigators and institutions, particularly in the acquisition and utilization of expensive research equipment and facilities.

•  University associations: University associations should develop and implement more powerful and strategic tools for financial management and cost accounting that better enable universities to determine the most effective methods for containing costs and increasing productivity and efficiency. As part of this effort, they should develop metrics that allow universities to communicate their cost-effectiveness to the general public.

•  Universities, working together with key stakeholders: Universities and key stakeholders should intensify efforts to educate key audiences about the unique character of U.S. research universities and their importance to state, regional, and national goals, including economic prosperity, public health, and national security.

Budget Implications

There may be an initial cost to institutions as they examine their operations in order to identify actions that will increase efficiency and as they invest in new infrastructure. In the long term, however, research universities will reap the rewards of these investments through greater productivity. Many institutions have already demonstrated that significant cost efficiencies are attainable. If research universities can take action, states and the nation will realize greater returns on their investments, and the savings associated with cost containment and greater productivity can

then be deployed to other priorities such as constraining tuition increases (a major national concern), increasing student financial aid, or launching new programs.

Expected Outcomes

By increasing cost-effectiveness and productivity, institutions will realize significant cost savings in their operations that may be used to improve performance by shifting resources strategically and/or to reduce growth in their need for resources (e.g., tuition). There are many ways to do this, but one of the easiest is to implement a “priority fund” in which the base funding of ongoing activities is reduced by 1 percent or so each year (with the “savings” reallocated to new university priorities).

Discussion

Without compromising the quality of their core programs and activities, the nation’s research universities should increase efficiency in their business operations, increase productivity and innovation in their academic programs, and report annually on their performance.

Revenue sources for U.S. research universities—and in particular for graduate education and research—have been constrained by the recent economic downturn and remain vulnerable for the near term: Federal appropriators have cut research funding as they negotiate how to reduce the federal debt, state appropriators have further cut already reduced support for higher education as their revenues have tightened and their policy priorities have shifted, corporate support has declined for both research and employee education, tuition increases are increasingly contested, gifts declined and endowments suffered during the financial collapse, and clinical income is threatened by new health legislation.

Consequently, most research universities have, along with other higher education institutions, faced budget crises that have necessitated emergency measures to balance their books through spending cuts and revenue enhancements, many of which work in the short term but are nonsustainable long-term strategies. Many institutions have increased tuition and fees, shifted enrollments toward higher-income or out-of-state students who pay higher tuitions, or increased enrollments while holding academic resources constant. Many have placed a freeze on hiring and salary increases, reduced benefits, instituted furloughs or laid off staff, or shifted to lower-cost, part-time instructors. Many have looked for one-time savings through cuts in administrative operations, academic programs, or student services. Table 5-4.1 displays the responses of presidents at public and private doctoral institutions to a question posed

TABLE 5-4.1 Strategies Deployed by Public and Private Doctoral Institutions to Address the Financial Consequences of the Economic Downturn (percentage that reported employing the strategy, Winter 2011)

Source: Kenneth C. Green with Scott Jaschik and Doug Lederman, Presidential Perspectives: The 2011 Inside Higher Ed Survey of College and University Presidents, Inside Higher Ed, 2011.

FIGURE 5-4.1 Cornell University, administrative streamlining program, projected savings by initiative, overall and by fiscal year, 2011-2015.

Source: http://asp.dpb.cornell.edu (accessed June 1, 2011).

by Inside Higher Ed about actions taken to address the current financial challenges.

Some institutions, however, have found that a thorough review of business operations can increase their overall efficiency and productivity by rationalizing their operations. For example, Cornell University; the University of California (UC), Berkeley; and the University of North Carolina (UNC) at Chapel Hill have each engaged Bain & Company, a management consulting firm, to examine ways to increase the efficiency of their administrative operations.²² As shown in Figure 5-4.1, Cornell expects to

²² Josh Keller, Universities can save millions by cutting administrative waste, panelists say, The Chronicle of Higher Education, July 25, 2010. Joe Wilensky, Update on budget cri-

save $75 million to $85 million over 5 years (FY 2011 to FY 2015) on its Ithaca campus, primarily by centralizing or improving negotiations in procurement or both ($30 million), reducing administrative layers and increasing direct reports per supervisor ($17.3 million), improving the use of facilities ($15.9 million), achieving efficiencies in information technology ($15 million), and achieving additional savings in student and academic services, finance, human resources, and communications. This amounts to annual savings of about 6 percent of Cornell’s Ithaca campus base budget, excluding external research funding.²³ Meanwhile, UC Berkeley envisions savings of about $75 million annually and UNC Chapel Hill, about $66 million per year.²⁴

All U.S. research universities should implement similar measures to ensure leaner and more productive academic and administrative operations in the long term. There is no single “magic bullet” for increasing efficiency and productivity, which will include both enhancing quality and improving cost control. Examples of efforts that are yielding results include the following:²⁵

Strategic Planning

•  Develop a strategic planning framework that lays out a university vision, supportive values, strategic goals, strategic initiatives, and progress indicators (University of Illinois).

•  Conduct administrative unit reviews and high-level external assessments of critical campus-wide functions to optimize the alignment of strategy and service with the campus’s mission and service stakeholders (University of California Davis).

•  Review university spending to ensure that academic programs are fully aligned with institutional mission and core competencies. For some institutions, this may mean paring back the number of graduate

sis and “Reimagining Cornell” stresses that all options are on the table, Cornell University Chronicle Online, July 23, 2009. Available at: http://www.news.cornell.edu/stories/July09/ReimaginingUpdate.html (accessed April 18, 2012).

²³ Cornell University, Division of Planning and Budget, Administrative Streamlining Program (accessed February 22, 2011). Available at: http://asp.dpb.cornell.edu (accessed June 1, 2011).

²⁴ Kevin Kiley, Where universities can be cut, Inside Higher Ed, September 16, 2011.

²⁵ Cornell University, http://asp.dpb.cornell.edu (accessed June 1, 2011); University of Illinois, http://strategicplan.illinois.edu/planning_process.html (accessed September 11, 2011); Pennsylvania State University, http://strategicplan.psu.edu/StrategicPlancomplete.pdf (accessed September 11, 2011); University of Minnesota, http://www1.umn.edu/systemwide/strategic_positioning/initiatives_ttu/admin_background.html (accessed September 11, 2011); University of California, Davis, http://vision.ucdavis.edu/local_resources/docs/vision_of_excellence.pdf (accessed September 11, 2011). Kiley, Where universities can be cut.

programs to focus on strengthening those in which the institution meets critical local, regional, or national needs.

Culture

•  Define and foster a culture that propels and reflects the university’s aspiration through excellence, service, and continuous improvement (University of Minnesota).

People

•  Implement voluntary separation programs for staff and faculty and rehire selectively and strategically (University of Illinois).

•  Improve the strategic hiring of faculty by instituting review of all faculty hiring plans by the Office of the Provost (University of Illinois).

•  Increase the number of staff in professional development and certification programs provided by the university, particularly in the areas of supervision, management, and leadership (University of California, Davis).

•  Improve performance management systems so every employee understands the expectation to continuously upgrade skills and knowledge and regularly receives feedback on performance (University of Minnesota).

•  Increase the use of part-time and part-year staff both to accommodate growing staff interest and to improve the efficiency of academic and administrative operations (Pennsylvania State University).

•  Increase the number of reports per supervisor, eliminating about 300 supervisory positions (University of California, Berkeley).

Administration and Finance

•  Recognize the university, its campuses, colleges, departments, and units as a single enterprise, establishing uniform standards and systems to reduce duplication of administrative processes and their associated support structures (University of Minnesota).

•  Implement a new financial system that not only replaces aged technology but also overhauls university financial processes and reporting mechanisms, resulting in improvements in processes, quality and quantity of information, and cost-effectiveness (University of Minnesota).

•  Consolidate administrative functions, such as payroll and accounts receivable, into a single university service center (University of Georgia System).

Information Technology

•  Upgrade information technology (IT) systems to capitalize on the latest advances, eliminate redundancies, and achieve interoperability.

•  Consolidate IT servers and establish centralized IT service centers, reducing IT staff by up to a third while maintaining service levels (University of Illinois).

•  Improve the capacity, costs, accessibility, efficiencies, and cyber-safety of campus computing systems, as a resource for both academic and administrative excellence (University of California, Davis).

Health Care

•  Encourage a shift from low (40 percent) to high (80 percent) use of generic medications to fill prescriptions in employee health care plans (University of Kentucky).

•  Increase the creativity of health care programs through a variety of measures, including additional wellness education and incentives, differential rates for employees who continue to engage in higher health-risk behaviors and for those who utilize in-system providers as opposed to other health care professionals (Pennsylvania State University).

Energy Efficiency and Conservation

•  Model energy efficiency and conservation in construction and maintenance operations and utilities consumption through pervasive and innovative application of green technologies (University of California, Davis).

•  Improve environmental stewardship through programs such as use of biofuels in service fleet, adopt a new Leadership in Energy and Environmental Design (LEED) policy for all new buildings,²⁶ reduce greenhouse gas emissions, encourage bicycle use, and increase recycling (Pennsylvania State University).

Facilities

•  Increase the effectiveness and efficiency of new construction, renovation, and maintenance projects as measured by timeliness, cost adherence, safety practices, and environmental certifications (University of California, Davis).

²⁶ Note: LEED certification costs about $100,000 and is only in part an economic model. Universities certainly should construct buildings to get the energy savings that would be paid back in energy cost savings over a reasonable period of time.

•  Improve the use of campus facilities during the day by investing in a state-of-the-art master scheduling system that will better spread classes across the day between 9:00 a.m. and 6:00 p.m. and by exploring the offering of additional evening classes (Pennsylvania State University).

•  Improve the use of campus facilities during the year by increasing the use of many university facilities that are now significantly underutilized during the summer months when students are away. By offering more courses during the summer, creating new summer programs, or allowing other organizations to use buildings during those months, universities may increase revenues without substantially increasing costs (Pennsylvania State University).

Instructional Productivity and Learning

•  Use tuition policies to encourage increasing 4-year graduation rates (University of Texas System).

•  Eliminate low-priority programs and reinvest the budget in higher-priority programs (Howard University, State University of New York (SUNY)–Albany, Washington State University).

•  Implement transparent workload policies, while accounting for legitimate differences among disciplines, articulating a clear expectation that all faculty supported on general funds will participate fully in the instructional programs of their respective units and that those not engaged in highly productive programs of research will have higher instructional workloads (Pennsylvania State University).

•  Achieve greater productivity in the classroom, particularly through the use of instructional technology, including blended and online learning. (The Open Learning Initiative at Carnegie Mellon University and the National Center for Academic Transformation provide important examples of how the use of technology and combinations of online and live instruction can dramatically reduce costs and improve learning in a variety of introductory and remedial classes. Stanford University recently ran an experiment that provided three online courses that enrolled 160,000 students, showing that there is a need, workable technology, and an opportunity for enhanced productivity.²⁷)

Accounting, Information, and Evaluation

•  Develop metrics to assess faculty and department productivity (University of Texas System).

²⁷ See http://www.nytimes.com/2011/08/16/science/16stanford.html?_r=1 (accessed January 18, 2012).

•  Develop the information and toll necessary to identify significant but currently unknown cost reductions.

On this latter point, a basic problem is that universities generally do not have the information and tools to identify significant reductions that are consistent with their agreed-upon outcomes. The core of what needs to be accomplished is to drive together the cost reductions with measurable outcome and quality improvements. To that end, there needs to be much better cost accounting; information data-informed decision making needs to be regularly used (i.e., the data must be collected and analyzed); and outcome measurements need to be agreed upon. Many of the outcome measurements must be developed by the individual institution because they measure what that institution seeks to achieve. This process causes fundamental questions to be raised and will make it possible to take steps otherwise impossible to understand and defend.

U.S. research universities should report annually to the public on their performance across their campuses to indicate how they have increased efficiency and productivity and met educational and research goals. They must show that they are good stewards of their resources, continuing to be fully transparent and accountable to the public and to policy makers through understandable performance measures. For example, the Association of Public and Land-grant Universities (APLU) and the American Association of State Colleges and Universities (AASCU) have been successful in developing the Voluntary System of Accountability (VSA) to enable universities to present clear, accessible, and comparable information on the undergraduate student experience to important constituencies through a common web report the College Portrait—so those constituents could better understand performance across institutions. Today, there are more than 300 institutions participating.²⁸ The development of similar or additional metrics on cost-effectiveness through VSA or another portal would enable institutions to provide the kind of information needed by policy makers, the public, families, and students to understand how institutions are utilizing their resources in a cost-effective manner.

Additional management and productivity issues are specific to academic health centers (AHCs), which have a rich history of an intricately entwined tripartite mission—education, research, and clinical care. These interactive missions have been a unique strength of the AHCs, as they have provided the underpinning for significant advances in biomedical discoveries and health care delivery. However, over the last four decades, the AHCs have relied on revenues from the clinical mission, with hos-

²⁸ See http://www.voluntarysystem.org/index.cfm?page=homePage (accessed January 17, 2012).

pital margins and physician practice plans, to sustain and expand their academic and research missions. It is estimated that the true cost of each medical student (i.e., the cost of training minus tuition recovery) requires subsidization of about $100,000, and as a whole, research grants require an investment of 30-40 percent by the university to cover the actual costs of conducting research.

With cuts in state appropriations to public universities, flattened federal budgets with lost spending power, increasing unfunded regulatory mandates, and mounting administrative burden, sponsored research funding at AHCs fails to recover the associated overhead costs incurred by universities. Meanwhile, clinical margins are in jeopardy. Managed care, changing payer mixes with increased Medicare and Medicaid patients, declining federal support for teaching hospitals, and complexity of care have contributed to strains on clinical revenues. Furthermore, the full impact of health care reform has yet to be realized by the AHCs. The net impact is unclear but, in general, the convergence of the current fiscal constraints puts the tripartite mission at risk. Erosion of the dollars typically used to fill funding gaps will require AHCs to work even more diligently to drive efficiencies, maximize productivity, diversify revenue sources, and enact a strategic approach to investing discretionary dollars across all three missions.

Actors and Actions—Implementing Recommendation 5:

•  Federal government: The federal government should create a new “Strategic Investment Program” supporting initiatives that advance education and research at the nation’s research universities. The program is designed to be a “living” program that responds to changing needs and opportunities. As such, it will be composed of term-limited initiatives requiring matching grants in critical areas that will change over time. The committee recommends the program begin with two 10-year initiatives: (1) an endowed faculty chairs program to facilitate the careers of young investigators and (2) a research infrastructure program initially focused on advancement of campus cyberinfrastructure, but perhaps evolving later to address as well emerging needs for physical research infrastructure as they arise. The federal investments in human capital and research infrastructure are intended for both public and private research universi-

ties. They require matching funds that different types of institutions may obtain from different sources. For example, public research universities may secure their matching funds from states sources, while private research universities may obtain their matches from private sources. However, the source that a particular institution taps for matching funds is not prescribed, so public and private institutions may draw from state support, philanthropy, business, or other sources for matching funds. While merit, impact, and need should continue to be important criteria for the awarding of grants, consideration should also be given to regional and/or cross-institutional partnerships, program focus, and opportunities for building significant research capacity, subject, of course, to the matching requirements for the federal grants.

•  Universities in partnership with state governments, business, philanthropy, and others: Universities should compete for funding under these initiatives, bringing in partners—states, business, philanthropy, others—that will support projects by providing required matching funds.

Budget Implications

In addition to increases in federal funding for basic research (in Recommendation 1), the committee recommends federal support for these first two initiatives in the program that will cost $7 billion per year over the next decade. These funds will leverage an additional $9 billion per year through matching grants from other partners.

Expected Outcomes

This program develops and leverages the human-, physical-, and cyberinfrastructures necessary for cutting-edge research and advanced education. Of particular importance is the investment in rapidly evolving cyberinfrastructure that will increase productivity and collaboration in research, but may also provide opportunities to increase productivity in administration and education. Also of critical importance is the endowment of chairs, particularly for promising young faculty, during a time of serious financial stress and limited faculty retirements. This will ensure that we are building our research faculty for the future, as we can reap the rewards of their work over the long term.

Discussion

Research universities operate today in a mixed environment that presents both fiscal challenges and innovation opportunities. To keep these institutions at the leading edge of research and education, the fed-

eral government should create a new “Strategic Investment Program” to support initiatives that advance innovations in key areas that, as with NSF’s Science and Technology Centers Program, eventually become self-sustaining. (See University of Illinois, http://strategicplan.illinois.edu/planning_process.html [accessed September 11, 2011] for an example of a strategic plan.) The first two initiatives should focus on (1) ensuring enhanced academic career opportunities for young faculty and (2) supporting investment in cyberinfrastructure that can increase the power of research and the productivity of institutional operations. These two initiatives are discussed below. The initiatives in the program should be term limited and change over time to reflect new needs and opportunities as they arise. This program should require matching funds from other stakeholders—states, business, philanthropic foundations, donors, and others—in order to ensure that the key parties that will benefit from these investments are contributing. Although the committee does not believe it appropriate to recommend a specific goal for the number of larger research universities that the nation should seek to maintain at world-class levels, the importance of advanced education and research to regional economic development suggests the need for a broader geographical distribution of research capacity than currently exists. Hence, we suggest that in addition to the usual criteria of merit, impact, and need used to determine research awards, some consideration be given to criteria such as program focus and opportunities for building significant research capacity and excellence. In particular, we would note that many universities that currently do not possess the resources or scale to become one of the larger research universities have demonstrated the capacity to mount high-quality research and graduate programs in more narrowly defined areas. By focusing resources, these institutions have managed to create peaks of excellence that can make significant contributions in particular areas of research and scholarship and have and can provide leadership to state and regional economic development. Such efforts should be supported as otherwise appropriate.

Faculty Chairs

To rebuild and sustain the faculties of research universities in key strategic areas during a period of serious financial stress, the federal government should launch an initiative under the Strategic Investment Program that provides matching grants to establish endowments for research faculty positions. Each faculty chair would be supported by a $3 million endowment, consisting of a $1 million grant from the federal government distributed through a competitive process based on research excellence and graduate student productivity, and a required $2 million matching

grant from private, state, or institutional resources. A total federal program of $2 billion per year would establish 2,000 new chairs each year, contributing significantly to the research and graduate education capacity of America’s research universities.

Replenishing the faculties of the nation’s research universities will bring new perspectives, capabilities, and energy. It would support young and mid-career scholars and investigators engaged in teaching and research who are often at the most creative point in their careers.²⁹ Yet many research universities, particularly public institutions now experiencing serious reductions in state appropriations, are limited in their ability to add faculty members at this time by serious financial constraints. Furthermore, the recent recession has shaken the confidence of senior faculty enrolled in defined contribution retirement programs, delaying their decisions to retire; consequently, our institutions have a rapidly aging and heavily tenured faculty cadre without the turnover necessary to open up positions for new junior faculty hires. Consequently, as shown in Figure 5-5.1, the age distribution of faculty is skewed toward the older end of the distribution, particularly in the humanities, social and behavioral sciences, and biological sciences. In the short term, this creates a bottleneck to refreshing the faculty. In the biomedical sciences, as shown in Figure 5-5.2, it has increased the average age at which an investigator receives a first NIH research grant to over 43 years old.

To address this current challenge, a federal program of matching grants to establish endowments for the support of junior and mid-career faculty positions would open faculty lines that will help build our nation’s faculties for the long term. There are other important reforms that may also be employed, but chief among them is the creation or strengthening of large, multiyear awards for early- and mid-career faculty (see Box 5-5.1). The use of chaired professorships has been very successful in building academic programs at many institutions and may be adapted specifically to the development of early-career faculty.

Examples of successful faculty development programs can be seen in a variety of places and times. In the late 1960s, for example, New York State established a program to award five Albert Einstein Chairs in Science and five Albert Schweitzer Chairs in the Humanities to institutions within the state. The awards were made competitively and available to both private and public institutions. While not explicitly stated, the

²⁹ Recent research provides hard evidence for this position. Pierre Azoulay, Joshua S. Graff Zivin, and Gustavo Manso, Incentives and creativity: evidence from the academic life sciences, The RAND Journal of Economics 42:3 (September 1, 2011): 527-554. This article showed that promising biomedical researchers who were appointed as HHMIs (Howard Hughes Medical Institute Investigators) did more innovative research than similar researchers who did not have such long-term support.

FIGURE 5-5.1 Age distribution of faculty in doctoral programs, by control (public, private), 2006.

Source: National Research Council, A Data-Based Assessment of Research-Doctorate Programs (Washington, DC: National Academies Press, 2011). Available at: http://www.nap.edu/rdp/ (accessed April 22, 2012).

program expected that new faculty hired with these awards would be recruited from institutions outside of New York State, since one of the goals was to build New York institutions. The 10 SUNY institutions receiving the awards benefited by hiring scholars who then served a focal point for building academic programs, attracting permanent and visiting faculty, and organizing international conferences. Similarly, in the 1980s, private foundations provided funding for regents chairs at the University of Texas at Austin, which used the program to create prestigious faculty positions, attract outstanding scholars to the university from around the world, and grow academic programs at the institution. In 1998, as a final example,

FIGURE 5-5.2 Average age of first-time R01-equivalent principal investigators, National Institutes of Health, by degree, 1980-2007.

Source: Sally Rockey, Deputy Director for Extramural Research, National Institutes of Health, Presentation to NRC Committee on Research Universities, November 22, 2010.

the University of Kentucky created its Endowment Match Program, more popularly known as the Bucks for Brains, or B4B, Program, in order to attract top researchers to Kentucky universities. B4B has required the universities to match the state funding through donations from philanthropists, corporations, foundations, and others, and the public and private funds have been invested to produce earnings that fund faculty positions, programs, or scholarships. During the program, the number of endowed chairs and professorships in Kentucky increased from 108 in 1997 to 523 in 2006. At the same time, extramural research and development grants at the Universities of Kentucky and Louisville more than tripled.³⁰

A national program that has built faculty, programs, and institutions in the global competition for talent is the Canada Research Chairs Program created by the Canadian government in 2000. The program

³⁰ Kentucky’s Bucks for Brains Initiative: The Vision, the Investment, the Future, 1997-2007. Available at: http://cpe.ky.gov/NR/rdonlyres/CA48D119-0E78-41BB-9D05-1FFBBA0CF7C5/0/BucksForBrains10YearReport.pdf (accessed September 13, 2011).

provided two types of awards: Tier 1 Chairs for experienced researchers and Tier 2 Chairs for junior faculty with acknowledged research potential. The program allocated $900 million between 2000 and 2005 to create 2,000 university chairs with an additional $250 million in infrastructure funding from the Canada Foundation for Innovation. Under the program, institutions were given 3 years to fill a chair, and the goal was to have all 2,000 chairs awarded by the 2007-2008 academic year. An evaluation of the program at the end of 2004 showed that, as of August of that year, funding had been used to create 1,282 chairs for researchers at 64 institutions.

Given the perceived success of the program, the government announced in 2010 the availability of $275.6 million to fund 310 additional chairs. The 2004 evaluation found the program to be successful in leveraging additional research funding and increasing research productivity. Further, while most of the awards under the program were made to faculty within the nominating institution, 364, or 29 percent, came from outside Canada. Almost all (84 percent) of the non-Canadian awardees indicated that the chairs program was influential in making their decision to join a Canadian institution. Tier 1 Chairs were awarded for 7 years and were renewable, but Tier 2 Chairs are for 5 years and are renewable once. The latter restriction was found to be a problem, since funds for the holding of these chairs may not be available after 10 years.

Creating a similar program in the United States, focused on the creation of endowed positions awarded through a national competition, would address the lack of research career opportunities for early- and mid-career scientists and ensure the existence of a research workforce for the future. There are successful young faculty programs such as the Presidential Early Career Award for Scientists and Engineers, NSF Faculty Early Career Development Program, and the NIH Pathway to Independence Award (see Box 5-5.2). Each of these programs awards support to an individual, and while that individual enhances the research program at the institution where they are employed or choose to be employed, this is not a position that the institution can use to build its programs. One additional advantage to endowed junior faculty positions over the above existing programs is the availability to support faculty in the humanities as well as in science and engineering. We believe it is the institutions’ responsibility to identify how best to allocate such positions across fields. These chairs would be handled like other university-endowed chairs, designed to support tenure-track faculty with both teaching and research obligations, and held in perpetuity. Whether faculty members would have fixed-term or lifetime appointments to the chairs would be addressed in the implementation phase through the proposals submitted by the institution.

Endowed junior faculty positions would be institution based and

once funded would remain with the institution and could be offered to any faculty members with the field specified for the position. A faculty member would hold the position for an initial 5-year term and would be renewable for an additional 5 years. At the end of the initial appointment or the renewal period, the faculty member would transition into a regular faculty slot. The allocation of federal grants to create these chairs should be merit driven, based on quality of proposals as well as the impact of the grant (e.g., the size of the graduate program or importance of the research area). But we also believe that the need of the institution for such endowed chairs should be a criterion, since well-endowed universities already have ample numbers of such positions, while less-wealthy institutions have serious needs.

Cyberinfrastructure

The National Academies report Rising Above the Gathering Storm recommended a $500 million federal investment in research infrastructure. Through the American Recovery and Reinvestment Act of 2009, the National Science Foundation and the National Institutes of Health made significant investments in research infrastructure during fiscal years 2009 and 2010. For example, NSF’s Academic Research Infrastructure Program: Recovery and Reinvestment and NIH’s Research and Research Infrastructure “Grand Opportunities” programs each provided $200 million in grants. These kinds of investments in research infrastructure are important. An area of investment that has the potential for significantly increasing productivity is cyberinfrastructure.

The federal government, philanthropy, and industry should focus one of its first two initiatives under the Strategic Investment Program on cyberinfrastructure necessary for cutting-edge research and advanced education. Rapidly evolving cyberinfrastructure (hardware, software, networks, and technical staff) will energize the conduct of research, collaboration and facilities sharing among elements of the national research enterprise (e.g., research universities, national agencies and laboratories, and industrial R&D activities), productivity enhancement through emerging IT-based paradigms (e.g., data centers, cloud computing, simulation), and progressive and innovative education (e.g., technology-enabled learning, cyberlearning).

It is increasingly clear that research in nearly all fields has entered a new era in which the discovery, application, and sharing of new knowledge rely fundamentally on advances in information technology.³¹ There are numerous examples of how IT-intensive research is transforming many disciplines. Some individual experiments, such as CERN’s Large Hadron Collider in physics and the Panoramic Survey Telescope and Rapid Response System in astronomy, generate enormous amounts of data that must be managed and analyzed. Climate research involves collecting and integrating disparate streams of data from observing systems in space, on land, and in the ocean. Advanced analytics and modeling increase the effectiveness and efficiency of enterprises in a service economy by mining large datasets and modeling effective practices. In biomedical research, advances are increasingly driven by patterns in data.³² In the

³¹ Tony Hey, Stewart Tansley, and Kristin Tolle, The Fourth Paradigm: Data-Intensive Scientific Discovery. Redmond, WA: Microsoft Research, 2009, available at: http://research.microsoft.com/en-us/collaboration/fourthparadigm/.

³² National Research Council, Steps Toward Large-Scale Data Integration in the Sciences: Summary of a Workshop. Washington, DC: National Academies Press, 2010.

humanities, arts, and social sciences, digitization and a variety of new IT-enabled tools are opening new possibilities for research.³³

A recent NRC report, Transforming Combustion Research Through Cyberinfrastructure, concluded, “The trends in continued use of fossil fuels and likely use of alternative combustion fuels call for more rapid development of improved combustion systems. New engines that are based on more predictive understanding of combustion processes must be designed for new fuel streams. A cyberinfrastructure (CI) that facilitates the timely dissemination of research results, experimental and simulated data, and simulation tools throughout the combustion community and extends into the engineering design process is necessary for shortening the time lines for combustion research (CR), development, and engineering. The current pace is rate-limited—by isolation, replication, and the reliance on experimentation, which is inherently slower than computer simulation.… A combustion CI will speed up the process of generating and testing designs and predictions preceding full-scale experimentation.”³⁴

In recent years, research universities, federal agencies, private foundations, and industry have launched a variety of programs to build a modern cyberinfrastructure—the research environments required for this new era of IT-intensive discovery. Federal agencies have made investments in domain-specific capabilities, such as the Bioinformatics Research Network supported by the National Institutes of Health. To coordinate and support the development of cyberinfrastructure resources broadly across the sciences and engineering, the National Science Foundation established the Office of Cyberinfrastructure in 2006. Within the higher education community itself, the Internet2 consortium leverages the capabilities of its members (primarily universities but also including industry and others) to develop and deploy advanced networking capabilities, which are a key component of cyberinfrastructure.

Although these and many other efforts have been valuable in enabling researchers to make crucial progress in ushering in twenty-first-century eScience, a significantly larger-scale national effort is urgently needed. In April 2011 the NSF’s Advisory Committee for Cyberinfrastructure (ACCI) released six task force reports to explore long-term cyberinfrastructure needs in campus bridging, cyberlearning and workforce development, data and visualization, grand challenges, high performance computing, and software for science and engineering. The ACCI task force reports

³³ Burton, Orville Vernon, and Simon Appleford. 2009. Cyberinfrastructure for the Humanities, Arts, and Social Sciences. Research Bulletin, Issue 1. Boulder, CO: EDUCAUSE Center for Applied Research.

³⁴ National Research Council, Transforming Combustion Research Through Cyberinfrastructure, Washington, DC: National Academies Press, 2011.

identify many pressing long-term needs in cyberinfrastructure and contain a number of specific recommendations.³⁵

The ACCI Task Force on Grand Challenges found that cyberscience and engineering (CS&E) has emerged as a distinct field and has become so important to advances across science and engineering that traditional mechanisms for NSF support, characterized by ad hoc, cross-directorate initiatives, should give way to a more permanent structure.³⁶ The task force also concluded that greater cooperation between agencies is required to ensure efficient progress in developing particular areas of CS&E—such as high-performance computing—and building a framework for multiagency support for specific grand challenges. The report also emphasizes the importance of training, education, and knowledge of how diverse communities can work together to build and upgrade cyberinfrastructure.

A comprehensive, sustained, and evolving cyberinfrastructure will support the fundamental requirements for the key activities of simulation and prediction, data mining, data management, online instruments and facilities, and interdisciplinary and interinstitutional collaboration. This capability is essential to the conduct of transformative research and associated education, and critical to the future well-being of the nation. To greatly enhance intellectual collaboration, productivity, and efficiency, much of this infrastructure should be shared across the national research enterprise (research universities, national laboratories, and industry R&D) and take advantage of rapidly evolving architectures, such as massive data centers, cloud services, and ultra-high-speed connectivity.

To realize this vision, the committee recommends that the federal government launch a national program to provide grants (with incentives for matching investments from institutions, industry, and the states) necessary to bring the cyberinfrastructure characterizing American research universities—and the broader national research enterprise—to the levels required for the conduct of world-class research, education, and collaboration. This necessary investment in cyberinfrastructure on campuses will also enable significant innovation gains and cost reductions in research and education, as it will expand that which has occurred through applications in business and industry. The program may also allow or even encourage joint efforts of institutions to spread the cost, build economies of scale, and share innovations.

³⁵ National Science Foundation, Advisory Committee on Cyberinfrastructure Task Force Reports. March 2011. Available at: http://www.nsf.gov/od/oci/taskforces/index.jsp.

³⁶ The ACCI Task Force defines grand challenges as science and engineering problems requiring breakthroughs in some combination of key areas, such as high-performance computing, computational models and algorithms, data management and visualization, software, and collaboration among diverse fields. National Science Foundation, Advisory Committee on Cyberinfrastructure Task Force Reports.

Actors and Actions—Implementing Recommendation 6:

•  Federal government—research sponsors: The federal government and other research sponsors should strive to support the full cost, direct and indirect, of research and other activities they procure from research universities so that it is no longer necessary to subsidize these sponsored grants by allocating resources (e.g., undergraduate tuition and patient fees for clinical care) away from other important university missions. Both sponsored research policies and cost recovery negotiations should be developed and applied in a consistent fashion across all federal agencies and academic institutions, public and private.

Budget Implications

Federal coverage of a higher portion of indirect costs would, at the margins, shift part of federal research funding from direct to indirect costs, so there will be no net change in cost to the federal government.

Expected Outcomes

This change will allow our research universities to hold steady or reduce the amount of their funding from other sources, such as tuition revenue or patient clinical fees, that they have had to provide for research procured by the federal government, amounts that have increased over the past two decades. Consequently, they will be able to use the flexibility this provides to allocate their resources from other sources more strategically for their intended purpose.

Discussion

In addition to a stable overall budget environment, a commitment from federal and other research sponsors to pay a fair share of the costs of research performed at universities would be an important step in strengthening U.S. research universities for leadership in the 21st century. Specifying the full costs of research and the appropriate contribution of sponsors has been a topic of debate, and occasional contention, par-

ticularly between research universities and the federal government, for several decades.

One perennial focus has been the ability of universities to recover the full and significant indirect costs of research from sponsors. The direct costs of research are those that can be attributed to a specific project, such as researcher salaries, travel, and the costs of laboratory materials. Indirect costs include outlays for facilities and administration (F&A), library costs, and other elements that support multiple projects or an institution’s entire research program.

There is ample support for the proposition that appropriately defined indirect costs of research incurred by universities should be fully recoverable. Indirect costs are real costs of research. For example, to launch a research project, laboratories and other facilities need to be built and maintained. In addition, existing evidence indicates that the indirect cost rates of research universities are comparable to those of government and industry labs, and perhaps slightly lower.³⁷ Sponsors are getting good value for their investments.

To be sure, there are countervailing views. Sponsors might argue that universities should make some contribution to research efforts because they reap rewards, too, including ownership of intellectual property, prestige for programs and schools, and enhanced recruitment of students and faculty. Parts of the philanthropic community, in particular, might see themselves as providing risk capital for the research enterprise. It is particularly difficult for these donors to understand indirect costs and the differences among different institutions in the levels of these costs. Furthermore, public concerns about rising tuition levels and medical costs, driven in part by declining support from state and federal government, will place many research universities in very awkward positions if they continue to be forced by research sponsors to utilize these sources to also subsidize the costs of sponsored research grants from public or private sources.

Naturally, federal agencies have an interest in maximizing the impact of taxpayer investments in research and have established mechanisms aimed at ensuring that indirect costs are justified, well-defined, and limited. Partly responding to a rise in F&A charges as a percentage of total award amounts during the 1970s and 1980s, the Office of Management and Budget established a cap of 26 percent on administrative costs in

³⁷ Arthur Anderson, The Costs of Research: A Report to the National Academies’ Government-University-Research Roundtable, 1996.

1991.³⁸ The Department of Defense and the Department of Health and Human Services (DHHS) are charged with negotiating F&A rates with research universities and other recipients. Institutions are regularly audited to make sure they are adhering to sound financial management practices and appropriately recovering indirect costs; universities found to have recovered costs inappropriately are required to reimburse the government.³⁹

Current policies and practices related to indirect cost recovery are causing significant problems for universities that are becoming more serious over time. First, the effective indirect cost recovery rates of many universities fall below the rates negotiated with DOD or DHHS.⁴⁰ For example, under some programs, such as NIH career awards and training grant programs, U.S. Department of Agriculture project grants, and U.S. Agency for International Development programs, F&A cost recovery is capped at lower rates by statute or by agency policy (OSTP, 2000). Similarly, agencies may formally request or favorably consider proposals that specify an indirect cost recovery rate below the institution’s negotiated rate. Second, during the two decades that OMB’s 26 percent cap on administrative cost recovery has been in effect, an increasing load of federal regulations and requirements have been added to research grants, raising the real costs of compliance borne by the institutions. These requirements in the context of the cap represent a significant unfunded mandate.

University contributions to research procured by the federal government and other sponsors are significant and growing. A report released in 2000 estimated that universities themselves were providing between $700 million and $1.5 billion to cover F&A costs in addition to those that the federal government covered.⁴¹ Since that time, institutions have reported that the costs of regulatory compliance in research programs have

³⁸ U.S. Office of Management and Budget, Circular A-21: Cost Principles for Educational Institutions, May 20, 2004. Available at: http://www.whitehouse.gov/omb/circulars_a021_2004 .

³⁹ Government Accountability Office. National Institutes of Health Extramural Research Grants: Oversight of Cost Reimbursements to Universities (GAO-07-294R). January, 2007. Washington, DC. Available at: http://www.gao.gov/new.items/d07294r.pdf.

⁴⁰ Government Accountability Office. University Research: Policies for the Reimbursement of Indirect Costs Need to Be Updated (GAO-10-937). September 2010. Washington, DC. Available at: http://www.gao.gov/new.items/d10937.pdf.

⁴¹ Charles A. Goldman, Traci Williams, David M. Adamson, and Kathy Rosenblatt. Paying for University Research Facilities and Administration. Santa Monica, CA: RAND Corporation, 2000. Available at: http://www.rand.org/pubs/monograph_reports/MR1135-1.html

increased dramatically, in one case quadrupling between 2000 and 2010.⁴² As shown in Table 5-6.1, institutional spending on university-performed R&D increased by 44 percent from 2004 to 2009, while R&D expenditures overall increased 27 percent and federally funded R&D expenditures increased about 18 percent. The pattern can also be seen in Figure 5-6.1 and the conclusion is the same: The institutional contribution to research has been growing faster than federal funding. To be sure, some of this may result from aggressive construction in anticipation of larger grant revenue, but a substantial portion of it stems from the increasing institutional subsidy for sponsored research. The institutional contribution must be funded through other sources, chiefly tuition, state appropriations, or private donations: As state appropriations and donations have declined in the recent recession and its aftermath, it is hardly fair to ask students and families to shoulder an ever-increasing share of these costs with tuition and fees already rising steeply.

While the basic principle that sponsors should pay for the full costs of research is straightforward, identifying and adopting the specific steps needed to realize it is more complex. There are several options that the committee discussed and should be carefully considered by OMB and others.

First, OMB could adjust its 26 percent cap on indirect cost recovery to account for the increasing costs of grant administration and regulatory compliance. This approach would be opposed by some out of concern that it would, in a flat-budget environment, represent a shift of federal expenditures from direct to indirect costs.⁴³ For this approach to be effective, the necessary additional funding would need to be appropriated, implementation would need to be gradual, and adjustment of the cap would need to be based on a rigorous analysis of actual costs.

A second approach would be the adoption of a flat F&A rate for all federal agencies and universities that would eliminate the negotiation of individual rates and the associated auditing. This approach would carry some advantages. It would increase certainty and reduce costs. Negotiating a standard rate across all federal agencies and institutions would create a level playing field for public-private competition for research grants. In the current economic climate, most universities cannot afford the current subsidy they are required to make for federal research grants

⁴² Association of American Universities, Association of Public and Land-grant Universities, and Committee on Government Relations. Regulatory and Financial Reform of Federal Research Policy Recommendations to the NRC Committee on Research Universities. January 21, 2011. Available at: http://www.aau.edu/policy/reports_presentations.aspx.

⁴³ Talman, William T. Letter from the President of the Federation of American Societies for Experimental Biology to Charles O. Holliday, Jr., Chair of the National Research Council Committee on Research Universities. March 14, 2011.

TABLE 5-6.1 Science and Engineering Research and Development Expenditures at Universities and Colleges: FY 2004-2009 (Millions of current dollars)

S&E = science and engineering.

NOTE: Because of rounding, detail may not add to total.

SOURCE: National Science Foundation, National Center for Science and Engineering Statistics, Academic Research and Development Expenditures, Fiscal Year 2009, Tables 1 and 2, available at: http://www.nsf.gov/statistics/nsf11313/content.cfm?pub_id=4065&id=2 (accessed April 22, 2012).

FIGURE 5-6.1 Federal and university funding for university-performed basic research, 1990-2008 (millions of 2000 constant dollars).

Source: National Science Foundation, National Center for Science and Engineering Statistics, National Patterns of R&D Resources, 2008 Data Update, Table 6. Available at: http://www.nsf.gov/statistics/nsf10314/content.cfm?pub_id=4000&id=2 (accessed September 4, 2011).

because of inadequate indirect cost recovery and excessive cost-sharing requirements. In reality, for all but the wealthiest private institutions, this subsidy today must come from the tuition dollars paid by students or the clinical fees paid by patients. It is no longer politically tolerable either at the institutional or federal level for undergraduates and patients to pay for federal research. The flat-rate approach would also involve some complications in implementation. For example, the facilities costs of different types of institutions vary widely. A variation on the flat-rate approach that could address variations in facilities costs is to limit the flat rate to the nonfacilities administrative costs.

A third approach would be to allow researchers to account for some portion of the time clerical and administrative staff spend on administrative and compliance tasks as direct costs. This approach would provide faculty with more flexibility, and allow any cost savings realized from the rationalization of regulatory requirements discussed below to flow

directly to research support. The disadvantage is that the true magnitude of indirect costs might grow in an opaque way over time, leading to lower productivity in research activity.

Notwithstanding the difficulty in reaching consensus on a specific approach to ensure that federal agencies and other research sponsors pay the full costs of research, there are areas where positive steps can clearly be made. Therefore, the committee recommends that

•  OMB should fully enforce existing cost-reimbursement rules and prohibit federal agencies from practices and policies inconsistent with federal cost principles. Some agencies establish rates for specific programs that are significantly lower than the 26 percent cap.

•  OMB should ensure that rate-setting practices by government negotiators are consistent and fair across all institutions. When different agencies negotiate rates, there can be inconsistent outcomes that are not fair to institutions.⁴⁴

The committee realizes that such harmonization among agencies may require statutory change, but also notes that a great deal can be achieved in the meantime by funding agencies working with OMB and with research institutions.

Another relevant aspect of policy toward research funding is the use of voluntary or mandatory cost-sharing provisions in federal grants, in which the grantee is to assume some of the direct costs of a project. The National Science Foundation has adopted a new policy based on a National Science Board report of several years ago.⁴⁵ The new policy eliminates voluntary cost sharing, and limits mandatory cost sharing to a set of programs where a financial commitment from the institution is seen as necessary for the project to succeed, or those involving partnerships with industry or state governments. Adopting this approach across all agencies, and adding a provision that exempts research universities from the mandatory cost-sharing requirements imposed on industry, would deliver significant benefits to institutions.⁴⁶

The steps recommended here would not, by themselves, ensure that private foundations, companies, and other research sponsors pay the full costs of research. The committee expects that the adoption of this principle and its implementation on the part of the federal government will

⁴⁴ AAU, APLU, and COGR, Recommendations to the NRC Committee.

⁴⁵ National Science Board. Investing in the Future: NSF Cost Sharing Policies for a Robust Federal Research Enterprise. Arlington, VA: National Science Foundation, 2009. Available at: http://www.nsf.gov/pubs/2009/nsb0920/nsb0920.pdf.

⁴⁶ AAU, APLU, and COGR, Recommendations to the NRC Committee.

have a positive impact on the willingness of nonfederal sponsors to do likewise.

Actors and Actions—Implementing Recommendation 7:

•  Federal government (OMB, Congress, agencies), state governments: Federal and state policy makers and regulators should review the costs and benefits of federal and state regulations, eliminating those that are redundant, ineffective, inappropriately applied to the higher education sector, or impose costs that outweigh the benefits to society.

•  Federal government: The federal government should also harmonize regulations and reporting requirements across federal agencies so universities can maintain one system for all federal requirements rather than several, thereby reducing costs.

Budget Implications

While the staff time to review regulatory and reporting requirements has a small, short-term cost, the savings to universities and federal and state governments over the long term will be substantial. Quantifying the burdens is difficult, so it is not feasible to estimate the savings in advance of a review, but we believe they could run into the billions of dollars over the next decade.

Expected Outcomes

Reducing or eliminating regulations can reduce administrative costs, enhance productivity, and increase the agility of institutions. We agree with the conclusion of the AAU, APLU, and Council on Governmental Relations (COGR) that “minimizing administrative and compliance costs ultimately will also provide a cost benefit to the federal government and to university administrators, faculty, and students by freeing up resources and time to directly support educational and research efforts.”⁴⁷ With greater resources and freedom, they will be better positioned to

⁴⁷ AAU, APLU, and COGR, Recommendations to the NRC Committee.

respond to the needs of their constituents in an increasingly competitive environment.

Discussion

The federal government—OMB, in conjunction with other federal agencies—should review the regulatory and reporting requirements it imposes on U.S. higher education institutions with the aim of eliminating those that are redundant, ineffective, onerous, or inappropriately applied to the higher education sector. Additions to the reporting or regulatory obligations of universities should be implemented only in light of an OMB cost-benefit analysis and should be accompanied by additional funding to support the higher resulting indirect and administrative costs.

As academic research activities have grown and become more complex, they have become subject to a broad array of regulations. Although state and local governments, as well as universities themselves, promulgate regulations affecting research, federal regulations are the main focus here because they constitute the predominant source of the research-related regulatory burden of universities. (State regulations and requirements were addressed in Recommendation 2.)

The vast majority of federal regulations are aimed at addressing legitimate issues and risks, and compliance and regulatory oversight are essential to the conduct of federally supported research. AAU, APLU, and COGR affirm that “research universities strongly support the objectives of accountability, transparency, and implementation of important policy and regulatory requirements.”⁴⁸

However, the sheer growth of requirements from many federal agencies, a substantial percentage of which were created with other types of organizations (e.g., industry) in mind, has raised the effort and costs necessary for compliance to a significant, unreasonable degree. AAU, APLU, and COGR argue that “in this environment, universities are often forced to institute one agency’s compliance requirements across an entire campus, even where they don’t make sense, and to sift through each agency’s specific rules and develop different compliance mechanisms all aimed at the same ultimate purpose.” They continue, noting that the uneven and unsynchronized implementation of regulations and reporting across many federal agencies create “a compliance miasma.”⁴⁹

AAU, APLU, and COGR note, “It is a growing fiscal challenge for universities to manage unfunded mandates as institutional budgets are being reduced, administrative cost reimbursements are being suppressed, and

⁴⁸ Ibid.

⁴⁹ Ibid.

cost-sharing requirements are increasing.”⁵⁰ While observing that compliance is difficult to measure, they provided the following as examples of the increasing costs of regulatory compliance:

•  One public university in the Northeast noted that the costs of managing its Sponsored Project Administration cost pool increased from $3.5 million in FY 2005 to nearly $6 million in FY 2010. Another, a private institution in the Midwest, estimated that its costs had increased from $4.2 million in 2002 to $7.3 million in 2008. A prominent medical school in the Southeast saw its compliance and quality assurance costs increase from approximately $3 million in 2000 to $12.5 million in 2010.

•  For that same prominent Southeastern medical school, compliance and quality assurance costs exhibited a cumulative growth rate of more than 300 percent between 2001 and 2010, while sponsored expenditures increased by only 125 percent during that same time.

Reviewing federal regulatory and reporting requirements will ensure both that important regulations are effectively enforced and that universities can use federal research funding efficiently and productively. In addition, efforts should be made to shift, where possible, from compliance-driven requirements to incentives for best practices. Most of the cost in compliance (for example, human subjects or animal treatment) is not the actual compliance. Rather, it is in maintaining, checking, and double-checking the bullet-proof audit records required. This is because it is an entirely compliance-driven regime, where the penalties of even a single infraction can be severe. By contrast, in a best-practices regime, an institution would be allowed a (limited) set of trade-offs between the cost of actual compliance and the cost of audit-proof documentation. An example is the system by which ISO-9000 certification is awarded. Firms are scored by whether their processes are up to best practices (with a percentage score allowing some variance), not audited at the single-incident level.

The current efforts on the part of the Obama administration to address the broad issue of regulatory reform are encouraging.⁵¹ The process put in place by Executive Order 13563, Improving Regulation and Regulatory Review, will hopefully lead to a lowering of regulatory burdens in areas relevant to universities. The ultimate results of this process and impacts on research universities should be evaluated at the appropriate time. A special effort focused on the regulatory burdens on research uni-

⁵⁰ Ibid.

⁵¹ Barack Obama. Executive Order 13563: Improving Regulation and Regulatory Review. January 18, 2011. Available at: http://www.gpo.gov/fdsys/pkg/FR-2011-01-21/pdf/2011-1385.pdf.

versities might still ultimately be needed. Fortunately, organizations and institutions that can help facilitate the necessary dialog among research universities and federal sponsors, such as the Federal Demonstration Partnership, are already in existence.

The problem of excessive regulatory burdens is itself an issue that puts a drag on the efficiency of all university research. The committee received testimony on many specific regulations and issues, several of which will be mentioned here by way of example. The full list of recommendations suggested by AAU, APLU, and COGR are provided in Box 5-7.1.⁵² The committee endorses this list as a basis for discussions moving forward.

In some cases, experts have identified regulations that do not add value or help ensure accountability, and have proposed alternative ap-

⁵² AAU, APLU, and COGR, Recommendations to the NRC Committee.

proaches. For example, effort reporting is the current mechanism used to support salary, wage, and related charges to federal contracts and grants.⁵³ Because effort is difficult to measure, the reporting mechanism is of little value as an internal financial control for the institution, while compliance is expensive and the reports are untimely from the standpoint of agency oversight. (Box 5-7.2 provides data collected by AAU, APLU, and COGR on the costs of effort reporting.) The current requirement puts a considerable burden on universities, with very little, if any, value to the federal sponsors or to the performing institutions. The committee

⁵³ Federal Demonstration Partnership. Payroll Certifications: A Proposed Alternative to Effort Reporting. January 3, 2011. Available at: http://sites.nationalacademies.org/PGA/fdp/PGA_055834

therefore recommends that effort reporting be eliminated or significantly modified.

In other areas, such as human subjects protection, animal welfare requirements, export controls, management and use of select agents, responsible conduct of research, and financial conflicts of interest, differing implementations and interpretations across agencies can cause inefficiencies in ensuring compliance and raise costs.⁵⁴ Standardized approaches to these across agencies would ease compliance burdens on universities significantly. (See further detailed suggestions in Table 5-7.1.)

⁵⁴ AAU, APLU, and COGR, “Recommendations to the NRC Committee TEXT MISSING

Further measures aimed at lowering and eliminating regulatory burdens on universities on a continuing basis should be considered. These measures would include the designation of a high-level ombudsman in the OMB’s Office of Information and Regulatory Affairs who would be charged with overseeing and regularly reviewing regulations affecting research universities and institutions, perhaps as part of an interagency effort under the National Science and Technology Council. Institutions could apply to the ombudsman to fix or eliminate inefficient regulations that do not add value.⁵⁵

During the course of this study, the committee received substantial testimony concerning the increasingly burdensome administrative and regulatory requirements associated with federally sponsored research imposed upon both institutions and investigators (including the statement that the majority of primary investigator time on NIH grants is now spent on project administration). Clearly this not only drives up university administrative costs, it also erodes research effort.

Actors and Actions—Implementing Recommendation 8:

•  Research universities: Research universities should restructure doctoral education to enhance pathways for talented undergraduates, improve completion rates, shorten time-to-degree, and strengthen the preparation of graduates for careers both in and beyond the academy.

•  Research universities, federal agencies: Research universities and federal agencies should ensure, as they implement the above measures, that they improve education across the full spectrum of research university graduate programs, because of the increasing breadth of academic and professional disciplines necessary to address the challenges facing our changing world, including the physical, life, social, and behavioral sciences; engineering; the arts and humanities; and the professions.

•  Federal government: The federal government should significantly increase its support for graduate education through balanced programs of fellowships, traineeships, and research assistantships provided by all science agencies dependent upon individuals with advanced training.

⁵⁵ Ibid.

TABLE 5-7.1 AAU-APLU-COGR Suggestions for Easing Compliance Burden on Research Universities

Source: AAU, APLU, and COGR, Recommendations to the NRC Committee on Research Universities

•  Employers: Business, government agencies, and non-profits that hire master’s- and doctorate-level graduates should more deeply engage programs in research universities to provide internships, student projects, advice on curriculum design, and real-time information on employment opportunities.

Budget Implications

Increasing the number of federal fellowships and traineeships to support 5,000 new graduate students per year in science and engineering would amount to $325 million in year one, climbing to a steady state expenditure of $1.625 billion per year. This funding is not designed to increase the overall numbers of doctoral students per se, but to provide incentives for students to pursue areas of national need and to shift support from the research assistantship to mechanisms that strengthen doctoral training. At the same time that the committee recommends increased federal funding for graduate education, the implementation of other aspects of our recommendation will also save money for the federal government, universities, and students. Reducing attrition and time-to-degree in doctoral programs, for example, will increase the cost-effectiveness of federal and other investments in this area.

Expected Outcomes

Improving pathways will ensure that we draw strongly from among the “best and brightest” for our nation’s future doctorates in science and engineering fields that are critical to our nation’s future.

Improving completion rates and shortening time-to-degree to an optimal length is the right thing to do for students and also increases cost-effectiveness, ensuring good stewardship of resources from the federal government and other sources.

Strengthening preparation of doctorates for a broad range of careers, not just those in academia, assists the students in their careers, and also assists employers who need their staff to be productive in the short term. This benefits new doctorates, employers, and society.

Discussion

Doctoral education in the United States represents the world’s leading effort for producing the next generation of faculty and researchers. By uniquely combining graduate education and research in the same place and at the same time, our universities have created a research and training system that is one of the nation’s greatest strengths—and the envy of

the rest of the world. Many countries globally are now moving to the U.S. model as they reform their own programs. Yet two sets of challenges now pressure us to reform key aspects of the process and substance of doctoral education to ensure that it remains vital, productive, and world class. The first are financial pressures on universities that call simultaneously for process improvements and additional financial support. The second are challenges and opportunities for realigning graduate education with labor markets to ensure students are trained for the careers they will have in academia, industry, government, or nonprofits. Many young people who imagine research careers also imagine careers as teachers in colleges and universities. Academic careers have become less attractive as salaries decline and permanent faculty positions become rarer. This affects the quality of those who choose research careers. The modern research enterprise requires a different mix of training levels and personnel capabilities than in previous generations. Then, the model was that every graduate student should be capable of becoming a Ph.D. and postdoc; every postdoc capable of becoming a junior faculty member; every junior faculty member, tenurable somewhere; every tenured faculty member capable of being an independent principal investigator. That is just not true anymore.

Response to Financial Pressures

Significant financial pressures on research universities impinge on their ability to provide support to doctoral students. Private institutions experienced falling endowment values during the recent recession. As we move out of that recession, these resources are rebounding, but political pressure on private institutions has forced them to increase spending rates from endowments for need-based financial aid for undergraduates, which is critically important but reduces resources for graduate education.⁵⁶ Meanwhile, public research universities have seen even greater pressure with deep cuts in state support for higher education compounded by substantial political pressure to use remaining resources to accommodate increases in undergraduate enrollments. For public and private institutions, these pressures result in reallocation of budgets that make support for graduate education extremely vulnerable, especially given the high mobility of doctorate recipients that creates incentives for states to under-invest in graduate education. Universities and governments must work together to place graduate education on a more solid financial foundation by improving the resource base for doctoral education, increasing the efficiency of doctoral education, and ensuring that doctoral programs are effectively meeting goals.

⁵⁶ The average discount rate for freshman in Fall 2008 was 42 percent, the highest ever.

Institutional Response

Given constraints on resources, institutions must become more efficient in educating graduate students. Two measures of this efficiency are completion rates and time-to-degree. By encouraging talented students to complete and to do so within a reasonable amount of time, institutions can save resources and propel graduates into early productive careers. The Council of Graduate Schools (CGS) has collected data on degree completion and attrition at 29 institutions in 23 fields across engineering, the sciences, and the humanities for students who began study in the early 1990s. As shown in Figure 5-8.1, the findings from this study show that after 10 years the overall completion rate for these programs was 57 percent, with a high of 64 percent in engineering and a low of 49 percent in the humanities. The flatness of the curves in this figure after about 8 years demonstrates that few students are likely to complete the degree after this point in time, at which about 50 percent have completed, the majority of those who will. While completion is not a measure of attrition, since students may eventually complete after 15 or 20 years, CGS data also show that the 10-year attrition rate averages about 31 percent in STEM fields and 30 percent overall. The highest attrition occurs in the physical sciences and mathematics at 37 percent, a level largely attributable to the attrition rate of more than 50 percent in computer science.⁵⁷

To be sure, attrition rates owe at least in part to the difference in the required talents and preferences for success in graduate versus undergraduate education and the fact that most university departments and students cannot easily determine suitability before admission. Unless advance training for research occurs early in a student’s undergraduate experience to allow sorting before attending a graduate program, there is bound to be a high rate of attrition; this situation has been true for many decades. Furthermore, completion statistics mask some clearly good outcomes for some noncompleters. Conservatively, half of the noncompleters studied in the CGS Ph.D. Completion Project left with a master’s degree awarded before they reached candidacy, and for many that was their original intended outcome. Still, given financial realities, it is clear that there are opportunities for improvement in completion statistics at the doctoral level, and it is incumbent upon doctoral programs to examine ways to reduce attrition, particularly for students who are prepared, talented, and otherwise eager to continue.

Hand in hand with the low completion rates are long times-to-degree, shown over time in Figure 5-8.2. In 2008, median duration between

⁵⁷ Council of Graduate Schools. Ph.D. Completion and Attrition: Analysis of Baseline Program Data from the Ph.D. Completion Project. Washington, DC: Council of Graduate Schools, 2008.

FIGURE 5-8.1 Average cumulative 10-year completion rates for cohorts entering doctoral study from 1992-1993 through 1994-1995, by broad field and year.

Source: Council of Graduate Schools. Ph.D. Completion and Attrition: Analysis of Baseline Program Data from the Ph.D. Completion Project. Washington, DC: Council of Graduate Schools, 2008.

starting graduate school and completing a doctorate was 7.7 years. This varied by field: 6.7 years in physical sciences, 6.7 years in engineering, 6.9 years in life sciences, 7.7 years in social sciences, and 9.3 years in the humanities.⁵⁸ These times-to-degree seem to match the completions data, as about 50 percent of the students complete in about 7 years. It also suggests that there is very little completion after 10 years and that attrition is near 40 percent for most fields.

Excessive attrition rates and time-to-degree represent inefficiencies in the current model of graduate education and incur considerable waste of both human capital and financial resources. Timely completion may be supported through a variety of means, including improved academic advising and mentoring, increased information about career opportunities, closer tracking of student progress, and activities to promote social integration within a department. Another aspect to the CGS study was

⁵⁸ National Science Foundation, Doctorate Recipients from U.S. Universities: Summary Report 2007-08. Arlington, VA: National Science Foundation, December 2009 (NSF 10-309), Table 18.

FIGURE 5-8.2 Average time-to-degree and age-at-degree for science and engineering Ph.D. recipients: 1978-2003.

Source: Mark C. Regets, Senior Analyst, National Science Foundation, National Center for Science and Engineering Statistics, Presentation to Committee on Research Universities, September 22, 2010. (Data from NSF/NCSES, Survey of Earned Doctorates.)

the development of interventions or best practices that could increase the completion rate. These included the following:

•  Selection and admissions policies to create a better “fit or match” between a prospective student and a specific program

•  Mentoring and advising, from student orientation to career guidance

•  Financial support in a form to optimize completion and enhance academic and social integration

•  Program environment that supports networks and support services

•  Research experience at the pregraduate level and exposure to different research options in the program

•  Curricular and administrative processes and procedures that provide support at different stages in graduate study and at the critical dissertation phase

These practices are particularly important for underrepresented minori-

ties in doctoral programs, since they represent a growing segment of the graduate application pool and, by CGS data, their completion rates are as much as 10 percent lower than white students.

Instituting these and other interventions to increase completion rates and reduce time-to-degree will require a different structure for graduate education, one that focuses on a programmatic commitment to student success and on preparing doctoral graduates for 21st-century careers. While the research and educational mission of universities is somewhat blurred at the doctoral level, it is important to students that they have clear objectives; only in this way will students be able to reach their full potential as researchers and contributors to the nation’s wealth.

A final note on doctoral education: The recent National Research Council Assessment of Research Doctorate Programs collected and analyzed data that provide a starting point for thorough institutional review of doctoral programs. These reviews should focus on how to strengthen programs that are viable and well targeted; right-size or redirect programs that are viable but need to be reoriented to meet current needs; or even eliminate programs that are not viable, as they do not meet goals or do not serve a current need.

Concerns about the length of time in training apply as well to postdoctoral study. Many postdoctoral fellows working in larger laboratories are engaged in interesting and productive work contributing to the science of their fields. This training allows them to mature as investigators and, eventually, move on to research positions in industry, faculty positions in research universities, or faculty positions in other higher education institutions. However, the uncertainty of and long time-to-career outcomes creates a strong disincentive to American college graduates to enroll in doctoral programs. Doctorates, mainly in the biomedical sciences, are experiencing long periods in training with little expectation of finding an academic research position that utilizes the training they received as a graduate student and a postdoctorate fellow. To shorten the postdoctoral period, many institutions have imposed time limits of usually 5 years, and once reached, these researchers either move on to another position or they find employment outside of research. Long postdoctoral appointments and poor prospects for a research career also deter newly awarded doctorates who are not electing the additional training; they select alternate career paths, possibly outside of their field of study. This may be viewed as an inefficient use of talent in the educational system.

Efforts are being made to address these concerns, such as the NIH Pathway to Independence Award Program. This program provides 5 years of funding for transition from a postdoctoral appointment to a research position at an institution or organization. The program will keep the career path open for the most promising researchers; however, it is limited

to a few hundred individuals, and a portion of the several thousand other postdoctorates will not find the research positions they trained for. Aside from the long periods in postdoctoral positions, these positions do not pay well, benefits given to regular institutional employees are not available, and the positions could be terminated at any time. A mechanism is needed for postdoctorates to continue their work in a research position that carries some job security and a reasonable salary level. Such positions as research faculty exist in educational institutions, but it is typically more economical for principal investigators to use lower-paid postdoctorates for the research. The NRC Research Associateship Programs provide an alternative career track for postdoctorates who work in national laboratories and often continue in these as permanent employees. These programs may serve as an example of innovation in this area.

To address these concerns, we await the final report of the National Academies’ committee that is currently undertaking, under the auspices of the Committee on Science, Engineering, and Public Policy (COSEPUP), an update of the report Enhancing the Postdoctoral Experience for Scientists and Engineers. That report played a key role in elevating the visibility of issues in postdoctoral training and, as a consequence, many institutions created postdoctoral offices and undertook reforms. The update will provide new recommendations based on current data.

Federal Response

While institutions are increasing efficiency, the federal government must also increase its support of graduate education, particularly for students in doctoral programs. Since the current financial climate facing American research universities makes it increasingly difficult for institutions to reallocate funds for this purpose (e.g., from undergraduate tuition revenues or endowment income), maintaining graduate enrollments and program quality in critical areas will require a significant increase in federal support for graduate education. As shown in Table 5-8.1, the number of federally supported, full-time graduate students in science and engineering peaked at almost 84,000 and has since declined to just above 78,000. It is critical at this time that the federal government compensate for this decline by committing to 5,000 new fellowships or traineeships. By providing multiple-year support, the federal government can signal to prospective students that they will have sufficient support to pursue advanced degrees, thereby enhancing the ability of graduate programs to attract the most outstanding undergraduates.

A program to increase federal support for doctoral students would also benefit from a review of the proper “package” of support for doctoral students during their time in graduate school. (See Box 5-8.1 for defini-

TABLE 5-8.1 Percentage of Full-Time Science, Engineering, and Health Graduate Students by Source of Support, Federal Agencies in 1988, 1998, and 2008

Source: National Science Foundation, National Center for Science and Engineering Statistics, Graduate Students and Postdoctorates in Science and Engineering, Fall 2008, Table 38. Available at: http://www.nsf.gov/statistics/nsf11311/content.cfm?pub_id=4072&id=2 (accessed, September 9, 2011).

tions of support mechanisms.) This review would examine the ways in which different mechanisms support both progress to the degree and experiences within the program. It may conclude that a shift in the support of graduate education away from research and teaching assistantships to multiple-year fellowships and traineeships is warranted, returning to a more balanced system of graduate student support similar to that of the 1960s. As seen in Figure 5-8.3, the shift from traineeships and fellowships to research assistantships began in the mid-1980s and increased rapidly in the early part of this century with the doubling of the NIH budget. In contrast to today’s graduate student support dominated by teaching and research assistantships that have as primary objectives providing low-cost support for the teaching and research enterprise, fellowships and traineeships have a primary objective of graduate student education.

Arguments for maintaining the proportion of support provided by different mechanisms focus on research efficiency and student eligibility. The shift of graduate support from traineeships and fellowship in the 1960s to research grants was driven by the dramatic increase in research

FIGURE 5-8.3 NIH graduate support, by mechanism, 1980 to 2008.

Source: National Research Council, Research Training in the Biomedical, Behavioral, and Clinical Research Sciences, Washington, DC: National Academies Press, 2011, Figure 3-16. Available at: http://books.nap.edu/catalog.php?record_id=12983 (accessed April 22, 2012).

grants and the need for individuals to do the research. The current system is very efficient at producing the research, but possibly at the expense of students who might seek research projects that suit their career goals. The shift to research assistantships also correlated with increases in non-U.S. citizens who are not eligible for traineeships that NIH restricts to U.S. citizens.

However, the downside to the current reliance on research assistantships in the natural sciences is that students on research grants are not necessarily provided with the kinds of programmatic commitment to success, alignment with 21st-century careers, and professional development activities (such as Responsible Conduct of Research) that are components of training grants. The NIH recognizes there may be problems with the current structure of support for graduate students and postdoctorates and has established a task force that will provide analysis of “the current composition and size of the workforce to understand the consequences of current funding policies on the research framework” to the Advisory Committee to the Director.

Alignment with Careers

The scientific workforce needs of our nation’s employers have evolved over the last several decades with changes in the work of science-based industries, government agencies, and non-profits. This highly trained, scientific and technical workforce has matured to include both more doctoral-level researchers and more staff who have both deep training in science at the graduate level and critically important skills in project or process management, sales, regulation, and similar areas. Consequently, as shown in Figure 5-8.4, there has been an increase in new doctorates who work outside of academia and, in Figure 5-8.5, the number of master’s programs focused on providing professional skills to students with advanced scientific training at the master’s level. Yet most research uni-

FIGURE 5-8.4 Work sector of Ph.D.’s, by Field, 2006.

Note: Academia includes 4-year and other educational institutions. Private, Non-Profit includes self-employed. Government includes federal, state, and local government.

Source: National Science Foundation, National Center for Science and Engineering Statistics, Survey of Doctorate Recipients, 2006, in Characteristics of Doctoral Scientists and Engineers in the United States, 2006, Table 12. Available at: http://www.nsf.gov/statistics/nsf09317/content.cfm?pub_id=3920&id=2 (accessed April 22, 2012).

FIGURE 5-8.5 Total number of professional science master’s programs in U.S. universities, 1997-2011.

Source: Council of Graduate Schools, Professional Science Master’s Initiative.

versities have not adequately adapted to the new realities of these labor markets.

Doctoral Careers

First, job markets and careers for doctoral scientists and engineers have shifted since 1990, with more than 50 percent of new doctorates now working outside of academe. This shift has led to conversations about reforming doctoral education to better position new Ph.D.’s for the careers they will have by providing more information about career options and by providing opportunities to acquire, in addition to the knowledge of one’s field, skills that are useful for academic positions (teaching, grant writing, publishing, presentations) and positions in government, business, or non-profits (oral and written communication, project management, regulatory compliance, business ethics, and innovation.)⁵⁹

⁵⁹ William G. Bowen and Neil L. Rudenstine, In Pursuit of the PhD. Princeton: Princeton University Press, 1992. National Academy of Sciences, National Academy of Engineering, and Institute of Medicine, Reshaping the Graduate Education of Scientists and Engineers. Washington, DC: National Academy Press, 1995. Chris M. Golde, At Cross Purposes: What the Experiences of Today’s Doctoral Student Reveal about Doctoral Education. Released January 16, 2001. Available at: http://www.phd-survey.org/report (accessed September 19,

TABLE 5-8.2 Percent of Doctoral Programs that Track the Career Outcomes of Their Graduates, by Field, 2006

Source: National Research Council, A Data-Based Assessment of Research-Doctorate Programs (Washington, DC: National Academies Press, 2011). Available at: http://www.nap.edu/rdp/.

Few incentives, internal or external, motivate graduate programs to align doctoral education with evolving employment opportunities, whether regional or national in scope. In most universities, the size of doctoral programs is driven by a range of factors, including its research and undergraduate teaching missions (and the need for university teaching and research assistants), without much thought to labor market trends. At a minimum, research universities must require their doctoral programs to track their graduates. The NRC’s Assessment of Research Doctorate Programs, as shown in Table 5-8.2, found that between 60 and 83 percent of doctoral programs were tracking their graduates as of 2006, though many programs that only ask for first employment may have responded affirmatively to this question. It is likely that the percentage of programs and institutions that track students out to 10 years or more is much lower. Tracking data are a crucial starting point for understanding both the careers of program graduates and how programs should be better aligned to support those careers. With greater self-understanding, programs can then also increase interaction with current and prospective employers to better inform the content of their programs and develop internship

2011). Carnegie Foundation, Carnegie Initiative on the Doctorate. Available at: http://www.carnegiefoundation.org/previous-work/professional-graduate-education (accessed, September 19, 2011). The Pew Charitable Trust, Re-envisioning the Ph.D. Available at: http://depts.washington.edu/envision/index.html (accessed September 19, 2011). Woodrow Wilson National Fellowship Foundation, The Responsive Ph.D. Available at: http://www.woodrow.org/responsivephd/ (accessed September 19, 2011).

opportunities for students with employers in industry, government, and non-profits.

Master’s Education

The shift has at the same time also led to important experiments in science master’s education that have successfully trained students both more deeply in science and more broadly in the job skills they need in business, governments, and non-profits. Improving graduate education will help programs respond to the job market through new thinking about the curriculum and better communication between universities and employers. Addressing these concerns can also increase the appeal of graduate education to U.S. students who currently may be turned off at present by uncertainties in the length of time and outcomes of graduate education.

At the master’s level, the National Academies’ report Science Professionals: Master’s Education for a Competitive World argued that “strengthened master’s education in the natural sciences will produce professionals who bring scientific knowledge and also anticipate, adapt, learn, and lead where and when needed in industry, government, and nonprofit organizations.” Indeed, the report found that “exciting experiments in master’s education over the last decade—the Master of Biosciences (MBS) program at the Keck Graduate Institute of Applied Life Sciences and the Professional Science Master’s (PSM) initiative seeded by the Alfred P. Sloan Foundation—have shown that graduate education in these fields can prepare students for advanced science-based work in a way that is highly desired by employers.”⁶⁰ These students become professionals with both scientific knowledge and workplace skills for the practical application of that knowledge—that is, a new kind of scientist with multidisciplinary skills and experiences.” Graduates of PSM programs are in demand by banks, insurance and financial companies needing financial mathematicians; a maturing biotechnology industry needing middle managers with advanced scientific knowledge and broader business skills; computer services corporations that require technical employees with business and customer skills; and state and federal agencies needing science- and technology-savvy staff with interdisciplinary training.⁶¹

The national capacity for interdisciplinary, employer-focused professional science master’s programs is likely far higher than at present. The number of PSM programs has now grown, from 1997 to 2011, to 239 nationwide. The America COMPETES Act of 2007 authorized the National

⁶⁰ National Research Council, Science Professionals: Master’s Education for a Competitive World. Washington, DC: National Academies Press, 2008, p. 2.

⁶¹ Ibid., p. 3.

Science Foundation to create a new program of grants to 4-year institutions for the creation or expansion of science master’s programs and the American Recovery and Reinvestment Act (ARRA) of 2009 provided 1 year of appropriations for this program, which was able to fund 21 grants out of 214 applications. There is room for further effort.

Consequently, the Commission of the Future of Graduate Education in the United States, in its report The Path Forward: The Future of Graduate Education in the United States, recommended that “the federal government should authorize a new federal competitive grant program across agencies to build capacity at universities to inspire innovation in master’s degree programs and responsiveness to workforce needs.” The report suggests that “universities would propose innovative new master’s programs or reinvigoration of existing programs, including professional master’s programs.” Furthermore, “each successful program would be required to demonstrate maintenance of enrollment, completion rates, and job placement outcomes, as well as ongoing involvement by employers to ensure that programs produce graduates for local, state, regional, and national workforce needs. Programs will be required to secure at least two thirds of program funding from sources other than the federal government.”⁶²

Actors and Actions—Implementing Recommendation 9:

•  Research universities: Research universities should engage in efforts to improve education for all students at all levels in the United States by engaging in outreach to K–12 school districts and undertaking efforts to improve access and completion in their own institutions.

•  Research universities: Research universities should assist efforts to improve teacher education and preparation for K–12 STEM education and improve undergraduate education, including persistence and completion in STEM.

•  Federal government, states, local school districts, industry, philanthropy, universities: All stakeholders should take action—urgent, sus-

⁶² Council of Graduate Schools and Education Testing Service, Commission on the Future of Graduate Education in the United States, The Path Forward: The Future of Graduate Education in the United States, April 2010. Available at: http://www.fgereport.org/rsc/pdf/CFGE_report.pdf (accessed February 12, 2011).

tained, comprehensive, intensive, and informed—to successfully increase the participation and success of women and underrepresented minorities across all academic and professional disciplines and, especially, in science, mathematics, and engineering education and careers.

Budget Implications

Increasing federal support for programs that enable the participation and success of women and underrepresented minorities in STEM disciplines has already been stated as a priority by both the America COMPETES Act and the Office of Science and Technology Policy. The committee supports the investments recommended for these purposes by these efforts.

Expected Outcomes

Our people are our greatest asset. Improving the educational success of our citizens at all levels improves our democracy, culture and society, social mobility, and both individual and national economic success. As career opportunities in science, technology, engineering, and math continue to expand at a rapid pace, recruiting more underrepresented minorities and women into STEM careers and ensuring that they remain in the pipeline is essential and strategic not only for meeting the workforce needs of an increasingly technological nation but also for obtaining the intellectual vitality and innovation necessary for economic prosperity, national security, and social well-being that such diversity brings.

Discussion

Research universities should become more fully engaged in the effort to improve the nation’s educational systems and academic careers for all students and at all levels, but particularly in science, technology, engineering, and mathematics, and particularly for women and underrepresented minorities. Especially given the uncertainty in the future participation of international students and scholars in U.S. doctoral education, discussed in Recommendation 10, it is critical that we also address the need to develop a more robust domestic talent pool. For each of the topics discussed below, we cannot stress enough the importance of a commitment from institutional leadership to achieving these goals and to creating an environment conducive to achieving them. To engage faculty interest, clear goals must first be articulated at the top, so that there is a broad commitment by the research university—including, in particular, its research, graduate, and professional education programs (not just its

school of education)—to addressing the challenges facing K–12 education, as well as continuing to give a high priority to undergraduate education, particularly in STEM disciplines.

Research Universities and Educational Reform

Research universities have an obligation to play a key role in reforming and improving education in the United States in general, a critical goal for our nation as we seek to bolster our global competitiveness, grow our economy, and improve the lives of individuals and families. To advance this effort, research universities and their faculty can pursue several avenues that will have broad benefits. First, they may expand their outreach programs to assist public schools, particularly those that have large numbers of disadvantaged students. Faculty can assist in the development of high-quality educational curricula. Universities may join with business and others to establish high-quality learning environments as a top national priority. Second, they must also help meet the national goals of increasing college degree attainment. Here they have much work to do. In most states, the share of undergraduate students at public research universities that come from these underrepresented groups (people of color, students from relatively lower-income families, first-generation students) is less than the share of undergraduate students at public institutions in general. Our research universities must turn this around.

In Coming to Our Senses, the College Board elaborated common-sense strategies for helping to accomplish goals for improving access and persistence rates, including the following:⁶³

•  Clarify and simplify the admissions process to encourage more first-generation students to apply.

•  Provide more need-based grant aid while simplifying and making financial aid processes more transparent to minimize student debt, and at least keep pace with inflation; make financial aid processes more transparent and predictable; and provide institutions with incentives to enroll and graduate more low-income and first-generation students.

•  Keep college affordable by controlling college costs, using available aid and resources wisely, and insisting that state governments meet their obligations for funding higher education.

•  Dramatically increase college completion rates by reducing dropouts,

⁶³ College Board, Coming to Our Senses. Available at: http://advocacy.collegeboard.org/sites/default/files/coming-to-our-senses-college-board-2008.pdf (accessed September 19, 2011).

easing transfer processes, and using “data-based” approaches to improve completion rates at both 2- and 4-year institutions.

Research universities must participate in this effort, supporting these goals and strategies.

Research Universities and STEM Education

A recent PCAST report stressed the importance of STEM education: “The success of the United States in the 21st century—its wealth and welfare—will depend on the ideas and skills of its population. These have always been the nation’s most important assets. As the world becomes increasingly technological, the value of these national assets will be determined in no small measure by the effectiveness of science, technology, engineering, and mathematics (STEM) education in the United States. STEM education will determine whether the United States will remain a leader among nations and whether we will be able to solve immense challenges in such areas as energy, health, environmental protection, and national security.”⁶⁴

Research universities have an important, perhaps even more critical, role to play here. Rising Above the Gathering Storm, for example, recommended that the nation “annually recruit 10,000 science and mathematics teachers by awarding 4-year scholarships and thereby educating 10 million minds.” (Box 5-9.1 describes this recommendation in detail.) Research universities can be very instrumental in this vein by expanding their efforts to train qualified K–12 teachers in STEM disciplines by developing and replicating successful science teacher-training programs, such as UTeach, raising very substantially the quality of the teaching workforce. The Association of Public and Land-Grant Universities has developed the Science and Mathematics Teacher Imperative (SMTI) that also helps these institutions undertake this effort.⁶⁵ SMTI is driven by the commitments of 125 university presidents in 43 states whose institutions presently prepare more than 8,000 science and mathematics teachers annually, and there are hopes to link this to the new Undergraduate STEM Education Initiative just launched by the Association of American Universities, discussed below.

Research universities, along with our nation’s liberal arts colleges that

⁶⁴ President’s Council of Advisors on Science and Technology, Prepare and Inspire: K–12 Education in Science, Technology, Engineering, and Mathematics (STEM) for America’s Future. Available at: http://www.whitehouse.gov/sites/default/files/microsites/ostp/pcast-stemed-report.pdf (accessed September 19, 2011).

⁶⁵ See http://www.aplu.org/smti (accessed September 19, 2011).

prepare a disproportionate share of those who go on to earn doctorates in science and engineering, must also continue to invest in and enhance undergraduate STEM education to ensure that students are prepared for the twenty-first-century economy, for study at the graduate level, and for the life-long learning process that will be needed to be successful after graduation. As recommended in the National Academies’ Expanding Underrepresented Minority Participation: America’s Science and Technology Talent at the Crossroads, there are many well-documented approaches to strengthening our STEM pipeline (for all students, including minorities), including summer science programs that engage high school students, undergraduate research experiences, improved academic mentoring, career counseling, peer study groups, and activities designed to promote social integration.⁶⁶ It is also important to address financial concerns that may pose a disincentive to study in STEM fields. These could be addressed through a range of

⁶⁶ National Academy of Sciences, National Academy of Engineering, and Institute of Medicine, Expanding Underrepresented Minority Participation: America’s Science and Technology Talent at the Crossroads. Washington, DC: National Academies Press, 2010.

options, including the scholarship program recommended by Rising Above the Gathering Storm or other means, such as loan forgiveness for those who continue in STEM careers.

Reforming the first 2 years of undergraduate STEM education is critical. A new study by the President’s Council of Advisors on Science and Technology recommends the following:

•  Catalyze widespread adoption of empirically validated teaching practices;

•  Advocate and provide support for replacing standard laboratory courses with discovery-based research courses;

•  Launch a national experiment in postsecondary mathematics education to address the math preparation gap;

•  Encourage partnerships among stakeholders (high school and college; 2-year and 4-year institutions; majority- and minority-serving institutions; academia and business) to diversify pathways to STEM careers; and

•  Create a presidential council on STEM education with leadership from the academic and business communities to provide strategic leadership for transformative and sustainable change in STEM undergraduate education.⁶⁷

As the first two items suggest, we need a strategic focus on reshaping first-year courses in the sciences. For far too long, they have been large lecture courses used to “weed out” students. The focus must be shifted to student learning, support, and encouragement.

We are also looking forward to a new 5-year initiative of the Association of American Universities to improve undergraduate STEM education. This initiative will develop an analytical framework for assessing and improving the quality of STEM teaching and learning, particularly in the first 2 years of college. It will establish a demonstration program at a subset of AAU institutions to implement the framework; explore mechanisms that institutions and departments can use to train, recognize, and reward faculty members who want to improve the quality of their STEM teaching; and work with federal agencies to develop mechanisms for rewarding and promoting these efforts as well.⁶⁸

⁶⁷ President’s Council of Advisors on Science and Technology, Engage to Excel: Producing One Million Additional College Graduates with Degrees in Science, Technology, Engineering, and Mathematics, February 2012. Available at: http://www.whitehouse.gov/sites/default/files/microsites/ostp/pcast-engage-to-excel-final_feb.pdf (accessed February 22, 2012).

⁶⁸ See http://www.aau.edu/policy/article.aspx?id=12588 (accessed September 19, 2011).

Women and Underrepresented Minorities in STEM

A recent study by the National Academy of Engineering recommended that “all participants and stakeholders in the science and engineering community (industry, government, institutions of higher education, professional societies, and others) should place a high priority on encouraging women and underrepresented minorities to pursue careers in STEM fields and on facilitating their participation and success, addressing field-specific issues evidenced by differential rates of completion by gender and race or ethnicity among STEM fields. Increasing diversity will not only increase the size and quality of our scientific and engineering workforce, but it will also introduce diverse ideas and experiences that can stimulate creative approaches to solving difficult challenges. Although this is likely to require a significant increase in investment from both public and private sources, increasing diversity of our scientific and engineering workforce is clearly vital to the future of the nation.”⁶⁹

First, research universities must work to increase the success of women in STEM by examining ways to increase their success as faculty. COSEPUP’s Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering found that while there are increasing numbers of women entering the STEM pipeline, their loss from that pipeline was not due to lack of talent, but rather a consequence of unintentional biases and outmoded institutional structures that are hindering the access and advancement of women.⁷⁰ Noting that “the United States can no longer afford the underperformance of our academic institutions in attracting the best and brightest minds to the science and engineering enterprise,” the report recommended “transforming institutional structures and procedures to eliminate gender bias” and the following actions:

•  Trustees, university presidents, and provosts should provide clear leadership in changing the culture and structure of their institutions to recruit, retain, and promote women—including minority women—into faculty and leadership positions.

•  Deans and department chairs and their tenured faculty should take responsibility for creating a productive environment and immediately implement programs and strategies shown to be successful in minimizing the effect of biases in recruiting, hiring, promotion, and tenure.

•  University leaders should work with their faculties and department

⁶⁹ National Academy of Engineering, Engineering Research and America’s Future: Meeting the Challenges of a Global Economy. Washington, DC: National Academies Press, 2005.

⁷⁰ National Academy of Sciences, National Academy of Engineering, and Institute of Medicine, Beyond Bias and Barriers: Fulfilling the Potential of Women in Academic Science and Engineering. Washington, DC: National Academies Press, 2007.

chairs to examine evaluation practices to focus on the quality of contributions and their impact.

•  Professional societies and higher education organizations have a responsibility to play a leading role in promoting equal treatment of women and men and to demonstrate a commitment to it in their practices.

•  Federal funding agencies and foundations should ensure that their practices—including rules and regulations—support the full participation of women and do not reinforce a culture that fundamentally discriminates against women.

•  Federal agencies should lay out clear guidelines, leverage their resources, and rigorously enforce existing laws to increase the science and engineering talent developed in this country.

Of particular importance in driving cultural changes, research universities should conduct regular audits of institutional culture and practices regarding gender in the faculty so that issues can be brought to light and acted upon. The report provides a model for doing so.

More recently, the National Academies’ Gender Differences at Critical Transitions in the Careers of Science, Engineering, and Mathematics Faculty found that while there has been some improvement for women at key career transition points at research universities, underrepresentation of women continues (see Figure 5-9.1) and important actions remain to be undertaken. Importantly, and in response to the kinds of audits recommended in Beyond Bias and Barriers, some research universities have made progress in hiring and advancing women. Gender Differences, through surveys of departments and faculty of research universities, found that women who applied for STEM faculty positions were at least as likely as men to be hired. The report also found that women who came up for tenure review were also at least as likely as their male counterparts to be granted tenure. This is good news, to be sure, but it does not free research universities, their leadership, and programs of responsibility. Two key areas that all of these actors must continue to act on are (1) recruitment, so that the numbers of women in the hiring pool can be increased, and (2) retention, so that the numbers of women who eventually do come up for tenure review also grow and begin to match the overall numbers of women who are coming up in the pipeline.⁷¹

As we cannot detail here all that must be undertaken to increase the success of women in STEM, we strongly recommend key actors pay care-

⁷¹ National Research Council, Gender Differences at Critical Transitions in the Careers of Science, Engineering, and Mathematics Faculty. Washington, DC: National Academies Press, 2010.

FIGURE 5-9.1 Representation of women in faculty positions at Research I institutions by rank and field in 2003.

Source: National Research Council. Gender Differences at Critical Transitions in the Careers of Science, Engineering, and Mathematics Faculty. Washington, DC: National Academies Press, 2010. Table S-1.

ful attention to the detailed actions provided in the important reports discussed on how to achieve these broad goals.

Research universities also have, along with a range of other actors, a strong role to play in increasing the participation and success of underrepresented minorities in STEM. While African Americans, Hispanics, and Native Americans comprise 27 percent of the U.S. population, they represent just 9 percent of the college-educated U.S. science and engineering workforce. And what makes this especially worrisome, the groups that are most underrepresented in science and engineering are the fastest-growing groups in the country. As seen in Figure 5-9.2, these groups will comprise about 45 percent of the U.S. population.

The National Academies’ Expanding Underrepresented Minority Participation argues that underrepresentation of this magnitude is due to increasing underproduction of underrepresented minority scientists and engineers at every level. This report notes that in 2007, as shown in Figure 5-9.3, “underrepresented minorities comprised 38.8 percent of K–12 public enrollment, 33.2 percent of the U.S college age population, 26.2

FIGURE 5-9.2 U.S. population by race/ethnicity, 1990-2050 (2010-2050 projected) Source: United States Census Bureau.

percent of undergraduate enrollment, and 17.7 percent of those earning science and engineering bachelor’s degrees. In graduate school, underrepresented minorities comprise 17.7 percent of overall enrollment, but are awarded just 14.6 percent of S&E master’s degrees and a miniscule 5.4 percent of S&E doctorates.”⁷²

Expanding Underrepresented Minority Participation examines how students become scientists and engineers and the problems of underrepresentation across the entire educational pathway from preschool to graduate school. Based on this assessment, the report outlined six principles for action:⁷³

1.   The problem is urgent and will continue to be for the foreseeable future.

2.   A successful national effort to address underrepresented minority participation and success in STEM will be sustained.

3.   The potential for losing students along all segments of the path-

⁷² National Academy of Sciences et al., Expanding Underrepresented Minority Participation, p. 38.

⁷³ Ibid., pp. 7-8.

way from preschool through graduate school necessitates a comprehensive approach that focuses on all segments of the pathway, all stakeholders, and the potential of all programs, targeted or nontargeted.

4.   Students who have not had the same level of exposure to STEM and to postsecondary education require more intensive efforts at each level to provide adequate preparation, financial support, mentoring, social integration, and professional development.

5.   A coordinated approach to existing federal STEM programs can leverage resources while supporting programs tailored to the specific mis-

FIGURE 5-9.3 Enrollment and degrees, by educational level, race/ethnicity, and citizenship, 2007.

Sources: U.S Department of Education, National Center for Education Statistics, Digest of Education Statistics, 2008, Table 41; National Science Foundation, National Center for Science and Engineering Statistics, Women, Minorities, and Persons with Disabilities, Tables A-2, C-6, E-3, and F-11; National Science Foundation, National Center for Science and Engineering Statistics, Science and Engineering Degrees, 1966-2006, Table 3.

sions, histories, cultures, student populations, and geographic locations of institutions with demonstrated success.

6.   Evaluation of STEM programs and increased research on the many dimensions of underrepresented minorities’ experience in STEM help ensure that programs are well informed, well designed, and successful.

Box 5-9.2 outlines the six broad recommendations from the report that address important issues across the educational pathway of laying an academic foundation in reading and arithmetic, preparation in science and mathematics, motivation for STEM education careers, access to and affordability of higher education, and academic and social integration. We strongly recommend that K–12 and higher education institutions as well as other actors pay careful attention to the detailed actions provided in this significant report on how to achieve these broad recommendations.

As a priority for the short term, the report recommended the nation

FIGURE 5-9.4 Percentage of 2004 freshmen at 4-year institutions who aspire to STEM majors who then completed STEM degrees in 4 and 5 years, by race/ethnicity.

Source: University of California Los Angeles, Higher Education Research Institute, Degrees of Success: Bachelor’s Degree Completion Rates among Initial STEM Majors, January 2010. Available at: http://www.heri.ucla.edu/nih/downloads/2010%20-%20Hurtado,%20Eagan,%20Chang%20-%20Degrees%20of%20Success.pdf (accessed April 22, 2012).

focus on undergraduate completion in STEM. Citing new data from the Higher Education Research Institute at UCLA, displayed in Figure 5-9.4, the report argues that the nation needs to take action to address significantly lower 4- and 5-year completion rates in STEM of underrepresented minorities relative to those of whites and Asian Americans. Since underrepresented minorities who matriculate at 4-year institutions aspire to a STEM degree as their peers, these lower completion rates represent both a challenge and an opportunity if we can implement actions that we know from experience work in sustaining the persistence and completion of underrepresented students.

Expanding Underrepresented Minority Participation, therefore, recommended policies and programs that seek to increase undergraduate re-

tention and completion of underrepresented minorities in STEM through strong academic, social, and financial support. It strongly recommends financial support from the federal government for underrepresented minorities that allows them to focus on and succeed in STEM by joining it to programs that simultaneously integrate academic, social, and professional development. It also recommends a federal program modeled on the NSF’s ADVANCE Program that would fund efforts to change institutional cultures in colleges and universities so that they are more supportive of underrepresented minorities.

The report concludes by arguing that all of the nation’s higher education institutions—including research universities—must play a role in implementing this priority action. It argued that while diversity of institutions is an asset, “currently, only a small number of institutions” are playing the role that all must play. It notes that these institutions “are diverse and can be found among all institutional types and categories; they are successful because they are doing something special to support the retention and completion of underrepresented minority undergraduates in the natural sciences and engineering. Their actions can be replicated and when they are, with a focus on both numbers and quality, it will pay off significantly.”⁷⁴ The report identifies the importance of leadership in creating a positive institutional environment for minority integration and success; practical steps that can be taken to increase the completion of minorities (making student success a priority, tracking student achievement, identifying choke points such as course availability, and improving course transfer); key elements for successful program development (resources and sustainability, coordination and integration, focus on the pipeline and transition points, program design execution, and evaluation); and proven, intensive interventions for underrepresented minorities in STEM (summer programs, research experiences, professional development activities, academic support and social integration, and mentoring).

Actors and Actions—Implementing Recommendation 10:

•  Federal government: Federal agencies should ensure that visa processing for international students and scholars who wish to study or

⁷⁴ Ibid., p. 8.

conduct research in the United States is as efficient and effective as possible, consistent also with homeland security considerations.

•  Federal government: As we benefit from the contributions of highly skilled, foreign-born researchers, the federal government should also streamline the processes for non-U.S. doctoral researchers to obtain permanent residency or U.S. citizenship in order to ensure that a high proportion remain in the United States. The United States should consider taking the strong step of granting residency (a Green Card) to each non-U.S. citizen who earns a doctorate in an area of national need from an accredited research university. The Department of Homeland Security should set the criteria for and make selections of areas of national need and of the set of accredited institutions in cooperation with the National Science Foundation and the National Institutes of Health.

•  Federal government: Engage in the proactive recruitment of international students and scholars.

Budget Implications

There is no additional cost.

Expected Outcomes

The United States has benefited significantly over the last half-century and more from highly talented individuals who have come to the United States from abroad to study or conduct research. Today, there is increasing competition for these individuals as students or researchers both in general and from their home countries. It is in the interest of the United States to attract and keep individuals who will create new knowledge and/or convert it to new products, industries, and jobs in the United States.

Discussion

The federal government should also strongly encourage the continued study and work of international graduate students and postdoctoral scholars in U.S. science and engineering through improvements in visa, residency, and citizenship processes. As James Duderstadt has noted, “Aging populations, out-migration, and shrinking workforces are seriously challenging the productivity of developed economies throughout Europe and Asia. Yet, here the United States stands apart because of another important demographic trend: immigration. As it has been so many times in its past, America is once again becoming a highly diverse nation of immigrants, benefiting immensely from their energy, talents, and

hope.”⁷⁵ In fact, today, one-quarter or more of new high-tech companies launched in the United States are founded by immigrants.⁷⁶ Attracting such talent to the United States is particularly important in knowledge-intensive, high-skill areas such as science and technology. Here, American research universities are extraordinary assets, since the world-class quality of their programs attract the best and brightest from around the world as students and faculty. The attractiveness of U.S. research universities for non-U.S. doctoral students and researchers is still a relative strength of American research universities. As seen in Figure 5-10.1, temporary residents earn a significant percentage of doctorates from U.S. institutions in key fields, including 27 percent in the life sciences, 42 percent in the physical sciences, and 55 percent in engineering. Moreover, this has been a significant benefit to U.S. research universities and, by extension, to the United States generally, as we have often drawn the very best students from overseas. These highly trained individuals are the best affirmation of U.S. academic leadership, and many of them are the sparks for continued domestic innovation and economic growth in our highly competitive global community.

However, trends can reverse. In the late 1990s, doctoral students from Taiwan and South Korea, the leading countries of origin, peaked both in number and in the percentage that stayed in the United States following degree receipt. That is, fewer came and of those who did, an increasing proportion returned home due to increases in opportunities there. They were replaced by India and China as the leading countries of origin.⁷⁷ As the growing strength of Ph.D. programs, research opportunities, and

⁷⁵ James J. Duderstadt, Higher Education in the 21st Century: Global Imperatives, Regional Challenges, National Responsibilities, and Emerging Opportunities, September 1, 2007. Available at: http://milproj.ummu.umich.edu/pdfs/2008/Glion%20VI%20Globalization.pdf (accessed March 22, 2012).

⁷⁶ Vivek Wadhwa, AnnaLee Saxenian, Ben Rissing, and Gary Gereffi, America’s new immigrant entrepreneurs: Part I (January 4, 2007). Duke Science, Technology & Innovation Paper No. 23. Available at: SSRN: http://ssrn.com/abstract=990152. This report found that 25.3 percent of the engineering and technology companies started in the United States from 1995 to 2005 had at least one foreign-born founder. The “New American” Fortune 500, A Report by the Partnership for a New American Economy, June 2011, available at: http://www.renewoureconomy.org/2011_06_15_1, found that close to “20 percent of the newest Fortune 500 companies—those founded over the 25-year period between 1985 and 2010—have an immigrant founder.” American Made: The Impact of Immigrant Entrepreneurs and Professionals on U.S. Competitiveness, A joint study by National Venture Capital Association, Stuart Anderson (National Foundation for American Policy), and Michaela Platzer (Content First, LLC) (available at: http://www.nvca.org/index.php?option=com_content&view=article&id=254&Itemid=103), found that “40 percent of U.S. publicly traded venture-backed companies operating in high-technology manufacturing today [2005] were started by immigrants.”

⁷⁷ Peter H. Henderson et al., Doctorate Recipients from United States Universities: Summary Report 1995. Washington, DC: National Academy Press, 1996.

FIGURE 5-10.1 Doctorate awards to temporary visa holder by major field of study, 2009.

Source: National Science Foundation, National Center for Science and Engineering Statistics, Doctorate Recipients from U.S. Universities, 2009, (NSF 11-306). Arlington, VA: National Science Foundation, December 2011. Table 20. Available at: http://www.nsf.gov/statistics/nsf11306/ (accessed December 10, 2011).

incentives increase in India and China over the next decade, will future trends for their students follow the pattern we have seen for South Korea and China? How long will it take to see this trend play out? Recent trends in the number of international graduate student applications, admissions, and enrollment can be seen in Figure 5-10.2, and the number of doctorates awarded to non-U.S. students on temporary visas can be seen in Figure 5-10.3. These trends show significant oscillation and uncertainty about future directions.

The United States should make enhancements to immigration policy that would encourage talented international graduates from programs in science and engineering to remain in the United States and allow the country to benefit from the investment in their graduate education. Rising Above the Gathering Storm addressed this issue head-on by arguing for improvements in visa processing for international students and scholars; providing a 1-year automatic visa extension to international students who receive doctorates or the equivalent in science, technology, engineering, mathematics, or other fields of national need at qualified U.S. institutions

FIGURE 5-10.2 Year-to-year percentage change in international student participation in U.S. graduate education, 2003 to 2004 through 2009 to 2010.

Source: Council of Graduate Schools, Findings from the 2011 CGS International Graduate Admissions Survey, Phases III: Final Offers of Admission and Enrollment, November 2011. Available at: http://www.cgsnet.org/ckfinder/userfiles/files/R_IntlEnrl11_III.pdf (accessed April 22, 2012).

to remain in the United States to seek employment; instituting a new skills-based, preferential immigration option; and reforming the system of “deemed exports” (see Box 5-10.1).

Yet current immigration policies continue to seriously constrain the valuable flow of international talent so critical to the economic prosperity of our nation.

•  The process of obtaining most classes of temporary visas needed to come to the United States contains costs, delays, and uncertainties, though this has improved since Rising Above the Gathering Storm was published.

•  There are application fees and separate wait times for obtaining an interview and a determination. Around one-quarter of those who apply for student visas are rejected. While this rate is believed to be much lower for accepted applicants to research universities, it is still reported as an issue. While some rejections and delays are due to security concern, most are because the student was unable to prove that they have no intent to stay in the United States.

•  An increasing number of international conferences have been placed and held outside of the United States to avoid visa problems. The

need to recruit internationally and to have frequent visits from foreign researchers has also made this a factor in the placement of some research laboratories.

•  Stories of faculty and students being stranded abroad with visa problems, whether common or rare, become oft-repeated horror stories that affect decisions of others to come to the United States.

•  Restrictions on what research may be undertaken by foreign students and scholars in the United States affect both decisions to come to the United States and decisions whether to stay—this has improved since publication of Rising Above the Gathering Storm, but restrictions remain.

•  Foreign researchers are sometimes excluded from a research activity due to rules, or uncertainty about the rules, that pertain to sensitive areas, restricted exports, or the terms of a specific research grant.

•  While allowed to work as research assistants on federal grants,

FIGURE 5-10.3 Science and engineering doctorates awarded by U.S. institutions to non-U.S. citizens on temporary visas.

Source: National Science Foundation, National Center for Science and Engineering Statistics, Numbers of Doctorates Awarded in the United States Declined in 2010 (NSF 12-303), November 2011. Available at: http://www.nsf.gov/statistics/infbrief/nsf12303/ (accessed December 10, 2011).

foreign students are not usually eligible for federal fellowships and traineeships.

•  Many job opportunities after graduation are restricted to U.S. citizens. Application for U.S. citizenship usually requires 5 years after receiving a Green Card. Time as a student or with a temporary work visa does not count.

Put in the simplest of terms, the United States must address these issues both to ensure that we can capitalize on the flow of international students and scholars and to provide our nation with the talent we need as we make progress on our goals listed under Recommendation 9 in increasing the participation of women and underrepresented minorities in key fields. First, we must ensure that visa, residency, and citizenship processes are as efficient as possible. Second, we must reform the temporary work authorization visa process (H-1B visas). Third, we must, as a priority, be more proactive, both by recruiting students, postdoctorates, and scholars and by following the practice of other nations such as Canada in encouraging the immigration of international students by attaching a Green Card to every doctorate in science and engineering.

This page intentionally left blank.