Read our COVID-19 research and news.

Jurisdictions that are part of NSF’s EPSCoR program (gray states not eligible). Other colors indicate year of entry. Orange, 1980; navy blue, 1985; red, 1987; yellow, 1992; cream, 2000; pink, 2001; dark green, 2002; dark blue, 2003; bright green, 2004; li

Jurisdictions that are part of NSF’s EPSCoR program (gray states not eligible). Other colors indicate year of entry. Orange, 1980; navy blue, 1985; red, 1987; yellow, 1992; cream, 2000; pink, 2001; dark green, 2002; dark blue, 2003; bright green, 2004; li

National Science Foundation

NSF research program for 'have-not' states faces growing criticism

Representative Bill Foster (D–IL) surprised some of his colleagues last week when he proposed killing a long-running program at the National Science Foundation (NSF) intended to lift up states at the bottom of the research funding heap. Although his amendment was defeated 232 to 195, the large number of “yeas” is the latest indication that the Experimental Program to Stimulate Competitive Research (EPSCoR) may need to do some serious soul-searching to remain viable.

Foster, the only Ph.D. physicist in the U.S. House of Representatives, is normally a staunch supporter of NSF and federally funded research. But Foster says he’s fed up with less-populated states getting far more dollars back from the U.S. government than what they pay in taxes. NSF’s EPSCoR is a small but egregious example of that phenomenon, he says. It allows smaller states that get relatively little NSF research funding to tap a $160 million pot that is off-limits to researchers in states, including Illinois, that get a bigger share of NSF research dollars.

“From a scientific point of view, it’s hard for me to understand why someone in the Texas panhandle should not have access to the same research funds that someone in the Oklahoma panhandle can have,” Foster told ScienceInsider the day after the vote, noting that Oklahoma is an EPSCoR state but Texas is not. “And why should scientists at Brown University (in Providence, Rhode Island)—and I’m not saying anything against them, it’s a great school—be eligible for benefits that are not available to researchers in states like New Jersey and New York and Massachusetts? … It’s simply because Oklahoma and Rhode Island happen to have fewer people” getting NSF grants because their population is so small, an outcome that makes them eligible to participate in EPSCoR.

Geographic factors have always been at the heart of EPSCoR, begun in 1979. In 1977, members of the House science committee grilled then–NSF Director Richard Atkinson on the fact that most of NSF’s money went to the more populous states and institutions on either the East or West Coasts, and accused the agency of having a geographical bias. “A number of members took us to task, and if you looked at the numbers it was hard to refute that criticism,” says Ray Bye, who headed NSF’s government affairs office at the time and then spent 2 decades as Florida State University’s chief lobbyist before retiring last year. “Everybody was looking for a solution that would address that problem while still retaining some aspects of a competitive research program.”

The answer was EPSCoR, a concept that has since spread to NASA, the departments of Energy and Agriculture, and the National Institutes of Health. Under NSF’s current rules, a state is eligible for EPSCoR funds if it receives less than 0.75% of the money—$6 billion this year—that the agency spends annually on research. Some 31 jurisdictions—28 states and Puerto Rico, Guam, and the Virgin Islands—fall below that line. North Dakota is at the bottom, receiving just 0.1% of NSF’s annual research spending. California ranks first, with 13% of the total pie, and Foster’s home state of Illinois is fourth, at 5.6%.

Agency officials consider a low percentage as prima facie evidence that scientists in those states need help in competing successfully for NSF funding. So NSF sets aside money each year for EPSCoR competitions aimed at improving their scientific readiness. A solicitation out this week, for example, offers EPSCoR jurisdictions a chance to receive up to $20 million over 5 years to support faculty, build research infrastructure, and provide better training opportunities at institutions in their state.

But Foster says that logic is flawed. A low amount of NSF funding may simply mean that there aren’t very many academic scientists in those states who apply for and receive NSF awards, he notes. A better metric, he says, would take into account a state’s total population, and calculate how much NSF funding each state gets per person.

Per state funding “is a completely irrational basis for doing anything,” Foster says. “The idea of ensuring that federal research dollars should be reasonably spread around is an arguably useful goal. But it should be done per person, not per state.”

Joseph Danek, who ran NSF’s EPSCoR program for many years and now works for a Washington, D.C., consulting firm, The Implementation Group, says that NSF has looked at “multiple formulas for eligibility” over the years but that the state has always been the essential unit of measurement. “The goal has always been to ensure that every state had a reasonable level of scientific capacity,” Danek says. “In the eyes of Congress, every state ought to be engaged in research, and students in each state should have the opportunity to learn in a research environment.”

Foster’s amendment, part of a 2016 House spending bill covering NSF and several other federal agencies, would have barred NSF and NASA from using any of the money for EPSCoR. The overall spending bill passed over objections from most Democrats about cuts to research and social welfare programs, and despite a veto threat from the Obama administration. And even though Foster’s amendment was defeated, it garnered the support of half of all House Democrats and 40% of Republicans, the latter group disregarding a plea by Republican leaders to oppose the amendment.

The vote resulted in some very unusual political alliances. Backing Foster were Representative Eddie Bernice Johnson (D–TX), the ranking member of the House science committee, and Representative Jerry McNerney (D–CA), a former mathematician. Although strong supporters of NSF research, both hail from non-EPSCoR states. At the same time, several liberal Democrats sided with Representative Lamar Smith (R–TX), the conservative chair of the science committee and a bête noire for much of the U.S. scientific community because of his criticism of NSF grantsmaking practices, in voting to preserve EPSCoR.

Foster may also have picked a ripe target. Two recent outside evaluations of EPSCoR have urged NSF to rethink the program, pointing to a “drift” in its rationale, eligibility criteria that allow more than half the states to participate, and a philosophy that enables states to remain in the program indefinitely.

“The 0.75% criterion fails to account for population and other critical aspects of research capacity and competitiveness,” concluded a 2013 report by the U.S. National Academies on all federal EPSCoR programs. “New graduation and eligibility criteria should be developed and implemented that could consider population, state commitment, proposal success rates per faculty member, total research funding, progress to date, and financial need.”

A report last fall on NSF’s EPSCoR program, by the IDA Science and Technology Policy Institute in Washington, D.C., said that NSF has never defined a key phrase in its 1950 mission statement that undergirds the program. “Deciding on the exact nature of the “undue concentration of research and education” that EPSCoR is intended to reduce is essential to predict which jurisdictions are mostly likely to benefit,” the report says. A clear definition, it notes, will allow NSF to choose “a quantitative indicator … to measure and track progress toward the goal.”

Foster isn’t very hopeful that NSF will take up the challenge. “I’ve detected no willingness to restructure the program,” he told ScienceInsider. Asked for the agency’s response to the two studies, an NSF representative said “steps to be taken in response to the recommendations have been outlined, and will be discussed with appropriate stakeholders. We’re not able to share details at this time.”

Heartened by the number of votes his amendment received, Foster says he plans to try again next year. “This is much a larger issue than science policy,” he says. “The real question is whether the federal government should be in the business of redistributing wealth to equalize the economic status of every state, including states where not many people, for whatever reason, have chosen to live. That type of redistribution is a distortion of our economy.”