Funding for the social and behavioral sciences is again under attack in the United States Congress. House Majority Leader Eric Cantor speaking at the American Enterprise Institute declared: “Funds currently spent by the government on social science– including on politics of all things– would be better spent helping find cures to diseases.” Speaker Boehner when he was Minority Leader wanted to limit the NSF’s focus to the “hard sciences.” Senator Coburn has called for defunding Social, Behavioral and Economic Sciences (SBE), and in the current fiscal year he succeeded in passing an amendment limiting what the National Science Foundation’s Political Science Program is authorized to fund. Most recently in March of this year Lamar Smith, the new Chair of the House Space, Science and Technology Committee (HSST) wrote, on behalf of his majority, to House Budget Committee Chair Paul Ryan:
[T]he NSF request for the Social, Behavioral and Economic Sciences (SBE) is over $259 million in FY 2013…. The Committee is concerned that the Administration has lost sight of NSF’s core mission in support of the physical sciences when so much funding is provided for SBE. Several recent studies conducted by NSF’s SBE funding have been of questionable value, and something our nation can ill-afford. These SBE funds are better spent on higher priority scientific endeavors that have demonstrated return on investment for the American taxpayer.
He followed this up in late April with a letter to Cora Marrett, the Acting Director of NSF, questioning five SBE funded studies and seeking to review the confidential peer assessments they had received.
The common threads behind these attacks are the claims that many of the studies funded by the NSF’s SBE Directorate are not worth the money they cost, and that even the worthwhile projects do less to advance national interests than similar expenditures in the physical, chemical and biological sciences. It is difficult to know whether these attacks stem simply from ignorance of the research NSF’s SBE Directorate has funded and the returns from that research, or whether they are motivated by animus toward politically uncongenial findings that are mistakenly seen as frequent outcomes of SBE research. Most likely both play a role.
If it is difficult to know the precise motivations that underlie the seemingly coordinated attack on the SBE budget, it is easy to see the bad faith behind elements of the attack. Chairman Smith’s statement above makes it sound as if NSF is devoting a substantial portion of its resources to its SBE Directorate. In fact the SBE 2012 allocation accounted for only 3.6% of NSF’s total budget and 4.4% of its research budget, not including research supported by the Education Directorate. The proportion appears to be somewhat less in 2013. Every other NSF research directorate has a budget that is well above twice SBE’s, and some have budgets that exceed SBE’s by four times or more. Another kind of bad faith is shown in the most detailed recent attack on specific NSF-funded research projects, which are in an NSF-focused report produced by Senator Coburn’s office in 2011. Responding to this report the HSST Democratic staff showed it to have been compiled from brief proposal abstracts and newspaper stories, with no evidence of any effort to contact the principal investigators (PIs) responsible for the questioned research to learn more about either their projects’ topics or the findings that emerged. When contacted by the HSST Democratic staff, PIs on 46 of the 50 studies that Coburn found questionable, many of which involved the SBE sciences, said their research had been mischaracterized, and even the four who said their projects had been reasonably well characterized thought that what had been said about their work was incomplete or misleading. Moreover, despite giving some lip service to the interest or value of some SBE funded research, those who would eliminate funding for SBE from the NSF budget do not propose to increase funding for the social and behavioral sciences in other agencies. The net result would be devastating. Estimates are that the SBE budget accounts for 62% of all federal funding for basic research in the social and behavioral sciences.
Of course, if the critics were right and SBE research did not return dividends commensurate with the money NSF spent on it, it would not matter that SBE was the source of almost 2/3 of federal basic research funding for the SBE sciences or that these sciences would be devastated if SBE was not supported in the NSF budget. The money should not be spent. But the situation is quite the reverse. Research in the SBE sciences is crucial for the nation’s well-being, and money spent to support basic social and behavioral science research has in the last 20 to 25 years contributed far more to the country’s well-being, in both dollars and non-monetary outcomes, than NSF has spent on social and behavioral science research since its inception. For example, developments in auction theory, reflecting decades of SBE grant support, coupled with the ability to test theoretical predictions through experimental economics, a subfield SBE grants nurtured, enabled the FCC to employ innovative auction rules when it allocated cell phone spectra, with the result that to date an estimated $60 billion has been returned to the U.S. treasury. A good portion of these returns would not have been captured had traditional approaches to auctioning government resources been used.
As with basic research in other sciences, returns to early investments in the SBE sciences are often unpredictable even when they are eventually great. One line of research began more than 50 years ago with a six page article funded by the Office of Naval Research that focused on the “toy problem” of how best to pair up couples in a closed group of men and women. The research on this and other seemingly fanciful or minimally important problems continued for decades with SBE support for advancing the theory. One result is that today there are algorithms that allow exchanges between living kidney donors and patients whom the donors would otherwise not have donated to or even met. Here is how this happens. If Mr. A.’s blood type does not match that of his wife, Ms. A, but does match that of Mr. B who also needs a kidney, and Ms. B’s blood type doesn’t match her husband’s but does match Ms. A’s, each donor can be alerted to the other’s need and secure their spouse’s wellbeing by donating to the other’s spouse. To date over 3000 people’s lives have been improved or saved thanks to this advance. As or more surprising is one outcome of work begun by a political scientist who, more than three decades ago, staged two computer tournaments to determine what strategies for playing the Prisoner’s Dilemma game would yield the highest average returns over the long run. Today, the results of those tournaments provide clues for innovative ways to attack cancer.
Much social and behavioral science research, often made possible only by SBE support, has had similarly valuable if less surprising returns. Improvements in eye witness identification, due to Law and Social Science, Social Psychology and Perception, Action and Cognition grants have diminished the likelihood that innocent people will be caught up in the criminal justice system and searches for true criminals prematurely terminated. Linguistic Program and Documenting Endangered Languages grants are preserving for future study the vocabularies and structures of rapidly disappearing languages. Decision, Risk and Management Science Grants have led to better ways to accurately communicate the dangers associated with weather and other emergencies. ‘Gold Standard” surveys supported for decades by SBE’s Sociology, Economics and Political Science Programs and more recently by a separate allocation have turned a telescope on America, fostered a multi-billion dollar survey industry, provided data used and referenced by scholars and journalists around the world, and influenced policymaking in many spheres. The list of those using SBE research and drawing on the expertise of social and behavioral scientists is also impressive. The military, the intelligence community, health care providers, businesses of all sorts and even the Congress of the United States regularly turn to SBE scientists for advice. Both the list of important accomplishments and the list of spheres of influence could go on and on.
With this list of accomplishments and the attention given to the results of SBE research, it is difficult to discern why an attack on funding NSF’s SBE Directorate is now occurring. Likely reasons are not clear. Anyone who thinks the SBE sciences are “soft” is unfamiliar with the complex mathematics and rigorous research designs that, depending on the area, characterize the largest proportion of SBE grants. Anyone who thinks SBE research is unduly political is also mistaken. Although some associate the social and behavioral sciences with findings that support or motivate left of center causes, most SBE sponsored research has no political implications whatsoever, and research results that have such implications are used to advocate policies across the political spectrum. Still social and behavioral science research, including studies sponsored by SBE, often highlight the ethnic discrimination and economic inequalities that persist in American society. Perhaps those seeking to defund SBE would rather not know this information than have to confront it. Regardless of whether the attack on SBE stems from ignorance, politics or both, ending or substantially cutting back on NSF’s support of basic research in the social and behavioral sciences will harm the nation and further diminish our country’s worldwide leadership in science.