How does the availability of resources affect what science policy is possible? How are priorities set in the face of scarcity?
In May of 2016, the National Science and Technology Council stood up the Machine Learning and Artificial Intelligence subcommittee to advise the Executive branch on national strategies for the development of AI systems within the Federal Government, private industry, and abroad. A key emphasis of the subcommittee’s seminal report, “The National Artificial Intelligence Research and Development Strategic Plan” was the need for future federal investment in research areas where private investment is unlikely. In particular, the report highlighted how investments in basic research for AI systems, including developments in AI “vision and cognition” for general purpose systems, constituted high-risk and long-term investments that conflicted with Industry’s penchant for quick returns on investment. As such, the report described the federal funding of basic AI research as an irreplaceable component of the Nation’s success in developing and implementing AI systems.
As a matter of science policy, few issues are as ubiquitous throughout the various fields of science as the Federal funding of research and development. For scientists working at the bench or in the field, the allocation (or not) of federal funding has the most immediate impact on the sustainment of their current work, the contributions to their field, and the benefit to society (not to mention their own livelihood). But beyond these direct impacts of Federal R&D funding, the allocation of the country’s monies carries with it broader significance to the Nation’s priorities, the greater R&D funding ecosystem, and the role of Science in shaping what the world looks like.
Interpreted as a bellwether for the Nation’s priorities, the Federal budget can be understood as an opportunity for the electorate and/or its leaders to put the country’s money where its mouth is in terms of investing in the values or necessities of the country. That said, nearly three-fourths of the country’s federal budget has traditionally been automatically allocated towards entitlement programs, paying off the Nation’s interest of its debts, and other mandatory expenses, the total amount of funding at the government’s discretion to allocate is significantly limited. Just as in all other cases of scarcity, the allocation or limitations on funding are always made within a context and not necessarily out of blatant favoritism or disdain for one agency or priority over others.
Another significant consideration for interpreting Federal funding of R&D is the greater R&D funding ecosystem as well as the unequal impacts of funding throughout the different stages of discovery. That is, discovery of underlying scientific principles (“Basic Research“), the application of those principles for practical purposes (“Applied Research“), and developing technologies to use or sell as products (“Development“).
Generally speaking, Basic, Applied, and Development research each constitute different aims of research and in turn are affected differently by fluctuations in funding levels. To the extent that these three forms of research exist on a continuum, the ratio of investment to pay-offs also correlates with Basic Research being the most time, risk, and cost intensive relative to Development where pay-off can be more immediate and assured, albeit with limited applications over time.
As annual funding trends bear out, Industry investment is more prevalent in Development whereas Federal funding makes up a majority of investment in Basic and Applied Research.
That said, Federal funding is a significant but far from the sole contributor to the Nation’s R&D efforts. Per the Congressional Budget Office, which advises Congress on the budget, Industry R&D funding is routinely three times the amount of the Federal government.
What’s more, there are contentious arguments surrounding the relationship of Federal funding levels for R&D and Industry contributions. according to analyses of the Congressional Budget Office, Federal and Industry funding may rise and fall together as the funding from one may produce research that increases the value of the other’s investment. Counter arguments, however, have suggested the opposite as investments from either side may encourage “free-riding” whereby one side’s investment may allow the other side to benefit from the other’s research without contributing any finances of its own. Similarly, though less understood, the complexity of Federal and Industrial funding of R&D is also prevalent globally as other nations, and their industries, strategically participate in R&D funding.
Alternative Means of Supporting Innovation
A final consideration for assessing Federal budget support for R&D innovation are the alternative mechanisms not explicitly captured in R&D budget allocation amounts that nevertheless contribute to private and public innovation. Several examples of such alternatives are captured in the table to the right and are routinely considered by Congress and the CBO throughout the budget and appropriations processes.
References and Further Reading
- Congressional Budget Office. (2014). Federal Policies and Innovation.
- Savage, J. D. (2000). Funding science in America: Congress, universities, and the politics of the academic pork barrel. Cambridge University Press.
- Kennedy, J. V. (2012). The sources and uses of US science funding. The New Atlantis, 3-22.
- Fortin, J. M., & Currie, D. J. (2013). Big science vs. little science: how scientific impact scales with funding. PloS one, 8(6), e65263.
- Gauchat, G. (2015). The political context of science in the United States: Public acceptance of evidence-based policy and science funding. Social Forces, 94(2), 723-746.