Concerns About the National Science Foundation’s Current Priority-Setting Process
The recent focus on NSF’s setting of priorities among large facility projects continues a long-running discussion of the best way for NSF to support such undertakings. In the June 12, 2002, letter to National Academy of Sciences President Bruce Alberts that led to the present study, six senators stated that “funding requests by the Foundation for large facility projects appear to be ad hoc and subjective.” The letter directed the National Academies to “review the current prioritization process and report to us on how it can be improved.”
In responding to this charge, the committee found that a number of concerns have been expressed by policy makers and researchers about the process used to rank large research facility projects for funding. First, the ability of new projects to be considered for approval at the NSB level has stalled in the face of a backlog of approved but unfunded projects. Second, the rationale and criteria used to select and set priorities among projects for MREFC funding have not been clearly and publicly articulated. Third, there is a lack of funding for disciplines to conduct idea-generation and, once ideas have some level of approval, there is a lack of funding for conceptual development, planning, engineering, and design—information needed to judge adequately whether a project is ready for full funding. Those concerns have eroded confidence among policy makers and the research community that large research facility projects are being ranked on the basis of their potential returns to science, technology, and society.
Large research facility projects have become too complex, expensive, and numerous to handle with procedures that may have sufficed in the
past. NSF has improved the administration of the MREFC program over the last decade, but further improvements are necessary and possible.
In response to the first concern, in its 2003 report Science and Engineering Infrastructure for the 21st Century: The Role of the National Science Foundation, the NSB examined the status of the nation’s science and engineering infrastructure and concluded that
there is an urgent need to increase Federal investments to provide access for scientists and engineers to the latest and best S&E infrastructure, as well as to update infrastructure currently in place.
The NSB recommended that NSF “increase the share of the NSF budget devoted to S&E infrastructure,” and one subject of emphasis was large facility projects.
In its report, the NSB noted that large facility project needs identified 5-10 years ago have not yet been funded although the scientific justification for the facilities has grown. Although the FY 2003 appropriation for the MREFC account was $148 million, an annual investment of $350 million for several years would be needed to address the backlog of projects.
As an example of the second concern, although the FY 2004 budget request did set priorities among projects, there was no justification for giving one project a higher priority than another as requested by Congress. It is only in the followup correspondence between NSF and the House Science Committee (see Appendix F) that the reasons become clearer. The lack of transparency has eroded confidence among policy makers and the research community and increased concerns that projects are not being chosen solely on the basis of merit. This lack of confidence in NSF’s priority-setting process has increased the risk that large facility projects will be funded for reasons other than their potential returns to science, technology, and society.
NSF’s support of large research facilities has undergone important improvements since the MREFC account was established in FY 1995. For example, NSF has institutionalized and publicized procedures governing the preparation and review of proposals.1 However, the committee has received a number of comments regarding the lack of funds for the development of new projects. NSF should consider a more adequate process for ensuring the financing of such activity.