Government is a major funder of scientific research and technology development. This report serves as a guide to some long-term federal investment trends, and updates previous budget guides and primers published by AAAS. An excerpt is adapted below. Download the full report.
Discretionary Spending Is Important Context for R&D
Virtually all federal R&D funding is contained within the discretionary budget, the portion of the budget determined through annual appropriations. As seen in Figure 1 below, discretionary spending – especially defense – once accounted for a greater share of the budget. Over time, the budget has come to be dominated by mandatory spending, which is made up mostly of the major entitlement programs – Social Security, Medicare and Medicaid – and is mostly on autopilot. The greatest drivers for this shift are an aging population and rising health care costs.
This matters for science because R&D doesn’t tend to change radically as a share of the discretionary budget. At the height of the Space Race, R&D was 17.4% of discretionary spending. But since the late 1970s, R&D has tended to fluctuate between 11 and 13% of discretionary spending (recent changes in what gets counted as R&D have pushed this share lower). In most years, as the discretionary budget goes, so goes the R&D budget.
The relevance of discretionary spending to science can be seen in Figure 2. In FY 2011, the discretionary budget began coming down first for nondefense and then later for defense as well. The Budget Control Act of 2011 further constrained spending. As Figure 2 shows, science agency budgets have been responsive to these changes in overall spending. At the same time, there have been outliers: defense basic science has remained a priority. The National Science Foundation was more protected early in this period, but has since fallen. More recently, energy R&D and the National Institutes of Health (NIH) have caught up. Note that this graph does not include emergency COVID-19 spending, which provided another $5 billion for NIH.
Major Recent Trends
In FY 2020, federal R&D reached an estimated $164 billion. This estimate comes with three caveats. First, it excludes emergency R&D for COVID-19. Over the course of 2020, NIH was provided with nearly $5 billion in emergency spending and other agencies with much more (see the COVID tab in the AAAS FY 2021 appropriations dashboard), though much of this was not for R&D. Official R&D estimates should be available sometime in the 2021 calendar year.
Second, the $164 billion figure corrects the official government figures from the FY 2021 budget request to include Department of Defense (DOD) R&D Management Support funding, known as the “6.6” account in DOD parlance. This funding, worth roughly $7 billion to $9 billion a year, was excluded from the official figures, but is included in AAAS estimates in accord with past practice, National Center for Science and Engineering Statistics methodology, and the Office of Management and Budget (OMB) A-11 budget preparation guidance.
Third, OMB adopted a new, narrower definition of “development” in FY 2017 and beyond. The result is that the amount of spending labeled “R&D” has been reduced by approximately $30 billion per year, mostly from DOD. This accounting change is visible in the sharp drop in Figure 3 below.
The past 20 years of federal R&D appropriations can be divided into four rough phases. In the first phase, from FY 1997 to FY 2004, federal R&D funding increased rapidly, by 44%. This rise was driven partly by increased defense R&D following the Sept. 11 attacks, but especially by the congressional effort to double the National Institutes of Health (NIH) budget. Certain other agencies like the National Science Foundation (NSF) and the Office of Science within the Department of Energy (DOE) also experienced substantial funding growth in this period.
The second phase ran from FY 2004 to about FY 2010 and mostly represents a plateau. Defense R&D remained elevated, while the picture was more complicated for nondefense agencies. Funding did increase for some, including NSF, DOE Office of Science, and the National Institute of Standards and Technology (NIST), all of which were prioritized for doubling by the Bush administration and the America COMPETES legislation. But this was offset by erosion of the NIH budget as appropriations failed to keep pace with inflation. The end of this second phase was punctuated by the generous but one-time funding boost in the American Recovery and Reinvestment Act (ARRA), which added over $18 billion in nominal R&D dollars in FY 2009, along with transiently generous annual appropriations in FY 2009 and FY 2010.
After FY 2010, funding entered a brief but jagged third phase of decline. As mentioned above, discretionary spending was cut dramatically over the FY 2010 to FY 2013 period, partly in reaction to the large deficits incurred during the financial crisis. These cuts, concurrent with a drawdown in war funding, resulted in substantially reduced R&D spending within defense and nondefense agencies, notwithstanding the Obama administration’s preferences. Note that accounting changes within DOE and NASA in those years make nondefense R&D appear to follow a steadier path than it did in reality when agency budgets were being cut as seen in Figure 2.
Phase four is the most recent, with multiple deals to raise discretionary spending underlying multiple large increases for science agency budgets, often with bipartisan support. These increases continued into the Trump years with a White House hostile to science funding. For instance, R&D appropriations in FY 2018 were the most generous in years.
For more, download the full report.