The Changing Nature of U.S. Basic Research: Trends in Funding Sources

Part two of a 3-part series: Part one, Part three

The first installment of SSTI’s series on basic research discussed the changing nature of federal research and development spending, namely that basic research as a share of federal spending on R&D has increased over time while making up a relatively small proportion of total spending compared to development and applied research. Furthermore, U.S. spending on research and development has failed to keep up as a share of total spending. Since peaking in 2009 with the first stage of funding from The American Recovery and Reinvestment Act of 2009 (ARRA), research and development funding at the federal level has begun to decrease. Recently released data from the National Science Foundation (NSF) now shows funds obligated by federal agencies for research and development and R&D facilities dropped from $141 billion in FY 2012 to $127 billion in FY 2013, a decrease of 9 percent, though increases are projected for FY 2014.

Writing in The New York Times, journalist Eduardo Porter offers deeper insight into federal spending on research and development, largely echoing many of the claims made by UK-based economist Maria Mazzucato, who argues that the risky investments made by governments in financing early scientific breakthroughs have had significant impacts over the course of the U.S. innovation history. Porter’s newest article serves as a valuable follow-up to his previous piece on what he considers the unstable foundation of America’s innovation system. In addition to noting stagnant funding for federal research and development as a share of GDP, American Innovation Rests on Rickety Foundation describes the private sector’s shift away from basic research spending in favor of acquiring innovation from places such as universities or startups and short-term economic thinking.

The second in SSTI’s series on basic research, this article focuses on the ways that the funding sources of R&D and basic research have changed over time. Ultimately, the nature of basic research – long horizons, unknown rewards, and high costs – indicates why some sectors, namely the private sector, may seek shorter term options. A better understanding of the changing nature of funding sources for basic research may be useful in understanding the current state of research and development in the U.S.

R&D and Basic Research Funding Sources:

The National Science Foundation (NSF) provides data on funding sources and performers of research and development through their National Patterns of R&D Resources series, last updated in December 2013.  The NSF classifies sources of research and development spending as federal, industry, university and colleges, other nonprofits, and other levels of government.

Source: National Science Foundation, National Patterns of R&D Resources: 2011–12 Data Update

As seen in the figure above, total spending on research and development has increased at an increasing rate in the United States since 1953, reaching $452.5 billion in 2012. From 1953 to 2012, more than 90 percent of total funding for research and development has come from either industry or federal funding sources. While federal sources made up the majority of funding between 1953 and 1978, more than half of total R&D spending has come from industry sources between 1981 and 2012. Other sources of funding for research and development include universities and colleges, other nonprofits, and other levels of government. Historically, these sources have combined to comprise less than 8 percent of total R&D spending.

While federal spending on R&D as a share of total R&D spending peaked at 67 percent in 1964, this proportion has decreased consistently since then, dropping to as low as 25 percent in 2000, and increasing to 30 percent in 2012. Industry investments in R&D have gone from comprising as little as 31 percent in 1963 to as much as 69 percent in 2000. In 2012, 63 percent of all R&D spending in the U.S. ($285 billion) came from industry sources. At $32.5 billion, other sources, including universities and colleges, other nonprofits, and other governments, collectively comprised approximately 7.2 percent of total R&D spending in 2012.

Despite recent trends, federal spending has historically remained an important part of total research and development spending. Within the private sector, R&D is seen as a necessary cost of doing business, an important part of driving future innovation and competitive advantage for a firm. Companies have moved away, however, from the Bell Labs and Xerox PARC model that afforded research scientists in the private sector more time (and funding) to focus on research that may not have direct application, at least in the short term. For this reason, the ways basic research are funded is worthy of additional exploration. 

Source: National Science Foundation, National Patterns of R&D Resources: 2011–12 Data Update

While federal sources made up just 30 percent of total research and development spending in 2012, the $39.4 billion they invested in basic research represents 52.6 percent of the $74.8 billion spent on total basic research that year. Furthermore, industry sources comprise 21.3 percent of total basic research funding in 2012, while universities and colleges represent 11.2 percent, nonprofits represent 11.8 percent, and other governments represent 3 percent.  In the graph above, these last three sources are combined into the “other” category, which has made up an increasing share of total basic research spending over time.

From 1953 to 2012, federal sources comprised more than half of all basic research spending in the United States in every year, though this share has decreased from the more than 70 percent share it held from 1964 to 1980 to 52.6 percent in 2012 – the smallest share on record for federal sources. During that same time period, universities and colleges have gone from representing just 1 percent of basic research expenditures in 1956 to 11 percent in 2012, while other nonprofit sources have gone from representing 6 percent to 12 percent of basic research spending. The share of basic research spending sourced from industry has, for the most part, declined since reaching its all-time high of 35 percent in 1956.

University of Washington professor Margaret Pugh O’Mara, in her recently released book, Cities of Knowledge: Cold War Science and the Search for the Next Silicon Valley, suggests that during the Cold War, political conditions allowed for science to remain an exclusive place where the ability to determine power and allocate funding for research was the belonged to a just a few institutions and individuals. This is reflected in the expansive role the federal government took in financing basic research during the 1960s and 1970s. From the early 1980s to the mid-1990s, the share of federal funding for basic research began to decline, while the share for industry sources and other sources began to increase. This period, where federal expenditures continued to increase in dollar terms, coincides with the early stages of the Bayh Dole Act, which incentivized involvement in the university commercialization process. As basic research spending at the federal level began to level off in the beginning part of the 2000s, the last decade has seen the rise in basic research funded at the university and college level.

Source: National Science Foundation, National Patterns of R&D Resources: 2011–12 Data Update

Eduardo Porter’s original article focusing on the foundation of America’s innovation system suggests that a main reason for concern is that the private sector is moving increasingly away from basic research. The chart above describes how, for each source, basic research as a share of its total R&D expenditures has changed since 1953. According to this chart, basic research has never played a significant part in the private sector’s total R&D spending. Early on in their history, universities and colleges, other nonprofits, and other governments began devoting more than half of their R&D spending to basic research. As described in the first installment of this series, the federal government increased the share of its R&D spending going to basic research through the early 2000s, when it has since began focusing more on applied research and development. While some may argue that the private sector should devote more spending to basic research, industry’s move away from basic research in favor of development and applied research is hardly a new phenomenon.

In Funding ≠ Innovation, a LinkedIn post written in response to Eduardo Porter’s articles in The New York Times, George Mason University professor Philip Auerwald notes the shift from science-based innovation to networked innovation makes the linearity of the invention-to-growth model increasingly obsolete. Instead, collaboration has become a necessary contributor to both the frontier of new knowledge and in the creation of new product combinations. In order to understand the ways in which networked innovation can occur, it is valuable to examine the ways in which R&D performers – necessary components of the network – have changed over time. In particular, the relationship between the funding sources described above and the performers they pay is worthy of additional attention.

The third installment of SSTI’s series on basic research will discuss how the areas where R&D and basic research are performed have changed over time.

Basic Research 2.xlsx