In June, the head of the National Science Foundation’s (NSF’s) engineering directorate left the agency to return to academia. Next winter the agency will bid adieu to the assistant NSF directors who run its math/physical sciences and geosciences programs. But the departures of Pramod Khargonekar, Fleming Crim, and Roger Wakimoto, respectively, after only a few years at NSF are not some kind of mass protest against government service. In fact, these academic scientists agreed to be short-termers before coming to the agency’s Arlington, Virginia, headquarters in 2013.NSF believes that having academics spend anywhere from 2 to 4 years at the agency before returning to campus helps it identify and fund cutting-edge research. Last year 176 scientists, about 28% of NSF’s scientific workforce, were on such temporary assignments.
But NSF’s heavy use of these so-called rotators is unique among government research agencies. And some members of Congress are worried that the government is paying too steep a price for that management strategy.
Rotators are much pricier than regular federal workers because they retain their academic salaries—which are usually higher—when they join the government. In 2012 the average surcharge was $36,500 per rotator, and those in executive positions -- the heads of divisions, offices, and directorates, including the trio that left this summer -- earn 50% more than their federal counterparts. Legislators also fear that the constant churn weakens institutional memory and program oversight.
At a hearing last year before the science committee he chairs, Representative Lamar Smith (R–TX) warned that “if NSF is not capable of handling this type of program, then maybe we should consider legislation that limits the use of rotators.” And Representative Don Beyer (D–VA), whose district includes NSF’s headquarters and thousands of federal employees, noted that rotators “can cause problems among the rank and file employees [because] they are not necessarily trained managers.”
NSF officials hope to blunt such criticism by becoming less reliant on rotators at the uppermost levels. In particular, NSF is thinking about asking Congress for special hiring authority, a mechanism that would allow the agency to bring in senior scientists at salaries above what they can typically offer federal workers. Other research agencies, including the National Institutes of Health, NASA, and the Defense Advanced Research Projects Agency (DARPA), already have such hiring authority.
At a 9 August meeting of NSF’s oversight body, the National Science Board, NSF Director France Córdova said special hiring authority would allow NSF to keep top managers for longer than the 4 years allowed under the Intergovernmental Personnel Act (IPA) that governs their service. That longer tenure, she explained, would alleviate concerns about program instability and inexperienced administrators.
Rotators are rare at most other federal agencies and don’t hold executive positions, an NSF survey found. The exception is DARPA, which touts its extensive use of “limited-term” employees as a major reason for its success in developing new technology for the military.
NSF’s apparent preference for rotators in senior management—for example, all six directorate heads are rotators—has attracted the attention of its inspector general (IG), an independent watchdog over agency practices. A new IG report notes that the salary gap between rotators and regular employees is widest at NSF’s executive level. For example, the highest paid executive rotator earned $440,000 last year, and three more topped $300,000. Some 22 of 27 executive-level rotators exceeded the $183,000 maximum salary for permanent federal employees—which is also what the NSF director earns.
The disparity is putting a growing strain on NSF’s budget, IG notes. The extra salaries and benefits cost NSF $8.9 million in 2015, up 37% from 2012, when only 21 rotators held executive positions.
The IG report also questioned NSF’s commitment to recouping costs. Universities often pay faculty a 9-month salary, with the expectation that federal grants will provide so-called “summer” salaries. So when rotators take a job at NSF, the agency is supposed to use the maximum federal pay rate to calculate what they are owed for the additional 3 months. Instead, NSF uses a straight extrapolation based on their current salaries. Using the right formula, says IG, would have saved NSF $222,000 last year.
Universities are also supposed to pick up at least 15% of the rotator’s salary and fringe benefits. That cost sharing is based on the idea that an institution benefits from having a faculty member temporarily at NSF, both because of the added prestige and the knowledge and connections they bring when they return. But 17 of the universities didn’t chip in a dime, IG found, and only five paid the suggested 15%. “There’s room for a better cost share,” Joanne Tornow, a senior NSF manager who heads an internal task group looking at the IPA issue, told board members.
Having special hiring authority would not necessarily save NSF money because the new hires would still be earning higher salaries than regular government managers, Tornow acknowledges. But NSF officials see the new hiring authority as expanding their management toolbox. “It’s not an either/or situation,” Tornow says. “I think NSF has been pretty successful in attracting the best talent. But we want to have multiple options.”
Getting Congress to change the rules, says Córdova, “is a long process, but we have a defensible argument.” And after board member Geri Richmond urged the agency to be more aggressive in selling the idea to legislators, Córdova said she’s ready to lead the charge.
“One of the main goals is to get the narrative straight and have everybody at NSF on board with the importance of this,” Córdova told the board. “Then I think a next step is to put front and center [the value] of having a mix of rotators and federal staff."