The National Science Foundation (NSF) announced on October 1, 2008, in its new Proposal and Award Policies and Procedures Guide (PAPPG), that the agency will no longer limit salary support on NSF grants to effort expended during the summer months. The new PAPPG becomes effective January 5, 2009.1 As the agency explains, “[t]his change moves away from the concept of summer salary and allows for reimbursement of two months of salary per year whenever appropriate during the year.”2 NSF’s salary policy change brings the agency in step with other federal agencies, most of which allow nine-month faculty to work on their grants at any time during the academic year as well as during the summer.

The change also brings NSF salary policy into line with the practicalities of academic research. Most nine-month faculty are engaged in their research throughout the calendar year, and it is often neither practical nor beneficial to NSF research to confine that research to two summer months. Adherence to the NSF two-months summer salary policy has been complicated by the fact that the policy has not been consistently applied. Some NSF directorates have been amenable to rebudgeting salary support from the summer period to the academic year, whereas other directorates have not permitted this practice.

NSF’s former policy created a potential compliance problem for some institutions. If faculty who received “summer salary” actually performed work on their NSF grants during the academic year, but reported their effort as summer effort, there would be two potential effort reporting problems. First, the summer effort report would be technically inaccurate, to the extent that it reported NSF grant effort that did not actually occur during the summer. Second, if no effort report were generated for the academic year (because the payroll system did not detect any academic year salary support), there would be no effort report to support the compensation charged for the work done during the academic year.

In July 2007, the Council on Governmental Relations (COGR) and the Association of American Universities (AAU) wrote to the National Science Board to request that NSF amend its policy so that salary charges could be made whenever appropriate throughout the year.3 The new NSF policy still effectively limits compensation to no more than a “summer’s worth” of support (i.e., two months in most cases, allowing one month for vacation). This is consistent with the historical NSF funding philosophy, not adhered to by most other federal sponsors, that NSF should not support activity that is already part of a university appointment, or salary that the university has already committed to pay. The new NSF policy does, however, provide grantees with the flexibility to distribute the effort and salary charges throughout the entire academic year. A comparison of the old and new PAPPG language appears here.

Both the current and revised version of the Award and Administration Guide (AAG) explain that “[s]alary is to be paid at a monthly rate not in excess of the base salary divided by the number of months in the period for which the base salary is paid.”4 Thus, NSF-funded salary is still limited to two months’ pay as determined by the base salary, though that pay need no longer be associated with effort expended during the summer months.

It should be noted that the new NSF policy does not fully solve the effort reporting compliance problem described above. Although the new policy makes it clear that academic year effort is now permissible and reimbursable under NSF awards, it will still be necessary for NSF grantees to support all academic year compensation charges with appropriate effort reports. This will probably mean that nine-month faculty who until now have received only “summer salary” support, and have therefore been accustomed to sign effort reports only for the summer months, will now be expected to complete effort reports for the academic year, to document effort that they expend during the academic year. Institutions that require faculty to complete effort reports only once a year will not face this problem, since most annual effort reports do not require documentation of when during the year the effort was expended.