Andrew Harrer/Bloomberg via Getty Images

How NSF cut 11% from its budget

Last fall’s divisive presidential campaign was still underway when Jim Olds, who leads the biology directorate at the National Science Foundation (NSF) in Arlington, Virginia, began worrying that the agency could soon be facing a serious budget crunch. It was already under a government-wide spending freeze, and Olds wanted to be prepared if things got worse. So he asked his staff to begin thinking about how to handle a 20% cut in the directorate’s $724 million budget.

“I picked what I thought was an extreme number,” he says. “The idea was to think about what would be necessary to get the science done in a very challenging environment. And we wanted to stay positive.”

That hypothetical exercise turned into an ugly reality this spring when NSF learned that the winner of that election, Donald Trump, planned to include an 11% cut to NSF’s $7.4 billion budget in his first full spending request to Congress. The news—which was publicly unveiled on Tuesday as part of the president’s $4.1 trillion budget for 2018—sent Olds and the heads of NSF’s other six research and education programs scrambling to erase big chunks of their portfolios without sacrificing NSF’s ability to fund the best new ideas.

It was a historic challenge: No U.S. president in NSF’s 67-year history had ever proposed giving the agency less than its current budget. (President Ronald Reagan may have come closest in 1981 when he tried to take back money Congress had already appropriated.) And Olds’ planning exercise was soon followed by final action on 2017 spending levels, which held NSF essentially flat.

NSF Director France Córdova responded to the White House’s directive by asking her senior managers to come up with a set of principles to guide their budget-cutting decisions. Maintaining capacity across all six research directorates and NSF’s education programs was paramount, Córdova explained at a Tuesday budget briefing. “You never know where the next discovery will come from.”

Another mandate, she said, was to allow NSF to continue funding the best unsolicited ideas from academic researchers, what she called its “core” programs. Maintaining support for cross-disciplinary research and interagency efforts—like NSF’s contribution to the Brain Research through Advancing Innovative Neurotechnologies initiative—were two other priorities.

Party like it’s 2008

Noting that the last time NSF operated on the $6.65 billion that Trump has requested in his 2018 budget was in 2008, Córdova and her deputies decided to look back at how NSF spent its money then for clues on how to craft the new budget request. “We were able to take a few things off the top,” Córdova said. The list included cutting in half the annual number of prestigious graduate research fellowships it awards. “Graduate Research Fellowships had doubled in recent years, to 2000, so we rolled it back to earlier levels,” she explained.

Those reductions still left NSF with an uncharted 9% cut. Cordova said she didn’t want to simply spread the pain equally across all programs. Instead, Olds and his colleagues were given the flexibility to decide what “core” meant to each of their disciplines, and to propose cuts that would preserve it.

Olds’s team had already decided that everything biology did should relate to the overarching theme of understanding the “rules of life.” That screen was applied to programs that serve the directorate’s main audience of individual investigators and those working in small groups. In addition, Olds said he used the results of an annual federal survey of employee satisfaction to think about ways to improve working conditions and streamline operations.

At NSF’s math and physical sciences directorate, the acting head said that defining core programs was a big challenge. “We started by asking people if they wanted us to take a common approach and the answer was a resounding no,” says James Ulvestad. “Sixty percent of astronomy’s budget goes to facilities, like telescopes, while mathematics has no facilities. So we allowed each division to define its core.”

James Kurose’s computing and engineering sciences directorate serves an equally diverse audience that includes those working as individual investigators; as members of larger, interdisciplinary teams; and as users of expensive machines. And Fay Cook, head of the social, behavioral, and economic sciences directorate, said that cross-NSF initiatives like Understanding the Brain also serve as platforms for the work of individual scientists in many different areas.

In the end, the directorates ended up taking relatively similar hits, from a 7.1% cut for biology to a 10.6% reduction for the geosciences. But Maria Zuber, head of the National Science Board, NSF’s presidentially appointed oversight body, says those similar percentages mask large variations.

“It wasn’t a one-size-fits-all,” says Zuber, vice president for research at the Massachusetts Institute of Technology in Cambridge. “That wouldn’t be strategic. [The director] empowered each directorate to look closely at its portfolio in deciding what to cut.”

How full is the glass?

The two biggest numerical losers in the request appear to be graduate research fellowships, down $86 million to $245 million, and EPSCoR, a program to help states with relatively small amounts of NSF funding, which would fall $60 million from its current level of $160 million. Even so, Córdova insists that training the next generation of scientists and maintaining geographic diversity across NSF’s portfolio were high priorities.

Despite all the reductions, which total $841 million, Córdova says the president’s 2018 request sends a positive message to researchers. “It’s not about what we can’t do, it’s about what we do for the American taxpayer,” she asserts. “I believe this administration thinks that basic research is important. And $6.65 billion is a solid investment.”

The next step will be to see how Congress reacts to the NSF proposal. In general, for example, both fellowships and EPSCoR have enjoyed bipartisan support. Fellowships are part of NSF’s investment in a tech-savvy workforce, a professed priority for every lawmaker. EPSCoR’s popularity stems from its ability to ensure that every state receives at least a bit of NSF funding even if it doesn’t host a top-tier research university.

Science advocates are concerned, however, that NSF lacks the political clout of its bigger basic research sibling, the National Institutes of Health (NIH), which can count on large patient advocacy groups and a robust pharmaceutical industry to advocate on its behalf. NSF also faces stiff competition within its congressional spending panel, which also decides funding for popular agencies such as NASA and the Department of Justice, and which must also deal with the Census Bureau’s need for additional funding to prepare for the 2020 decennial headcount.

Given those pressures, science lobbyists anticipate a very tough fight. “I worry that the science advocacy community will spend their time fixing the really bad cuts and go near silent on NSF,” says one, questioning the community’s resolve. Another advocate fears that even a vigorous campaign will fall short. “I’m pretty confident that the academic research community will think that 11% is a really bad cut and that panic will set in. The question is whether it will be possible to fully reverse the proposed cut.”