Large study of scientists who move their labs reveals how location drives productivity

From ScienceMag:

The mantra “location, location, location” isn’t just about real estate. For life scientists, more than 50% of their productivity can be attributed to the institution where they work, according to a new study that tracked the publications of researchers as they moved during their career. The findings, published this month as a National Bureau of Economic Research working paper, play into an active and long-running debate over how to allocate limited research funding—and whether to implement policies that prevent grant dollars from being concentrated at a handful of big-name universities.

The study quantifies a phenomenon that’s probably familiar to many academics, says University of California, Berkeley economist Carolyn Stein, who wasn’t involved in the new study. It’s easy to imagine that “you can pick someone up and move them to a more productive place, and it will make them more productive.” Still, the magnitude of the observed effect is striking, she says. “The role of luck and path dependence in science is maybe larger than I’ve completely appreciated.”

The new study compiled data for about 300,000 U.S.-based life scientists who published between 1945 and 2023. Boston-area researchers had the highest productivity—publishing two or three times more papers per year in 15 journals that cover basic life science research, including Cell, Nature, and Science, compared with researchers in many other metropolitan areas. When a researcher moved from a less productive institution to one with higher average productivity, they became more productive as well, according to a “wandering scholar” analysis that included about 38,000 scientists who had a publication record from before and after moving between institutions.

The team wasn’t able to pinpoint what institutional characteristics led to productivity gains. Lead author Amitabh Chandra, an economist at Harvard University, notes that “it could be something about resources, facilities, graduate students.”

A previous study backs that up: Faculty at top U.S. universities who work in fields where collaboration and co-authorship are the norm were more productive in large part because they led larger lab groups, researchers reported in a 2022 paper published in Science Advances. “It’s this labor that increases faculty productivity, not really inherent characteristics of the faculty themselves,” says Aaron Clauset, a computer scientist at the University of Colorado Boulder and senior author on the 2022 study.

Chandra and Harvard colleague Connie Xu began the new study well before President Donald Trump’s administration took over and halted the flow of research funding to Harvard. But the findings indicate how much research could be lost if funding isn’t reinstated. “Entire countries produce less than what Harvard produces,” notes Chandra, whose analysis found that his institution publishes 3.6% of the global output of top life science papers—the most of any single institution in the world. “When we turn off the funding to one of these large producers … the implications are colossal.”

Recent political developments aside, the paper also plays into a decadeslong policy discussion about how to allocate federal grant money. Should funders aim to maximize output—resulting in a concentration of grants at relatively few elite institutions—or would it better serve the public to spread funding around more broadly? The new study makes plain the benefits of favoring the elites: “If … funders are choosing between two equally productive scientists, one at an institution whose average research output is twice the other’s,” Chandra and Xu write, “funders could get more than 50% more research by prioritizing a scientist at the more productive institution.”

But Chandra acknowledges it’s also “perfectly valid” to base funding decisions on priorities other than research output, such as reducing funding disparities. He hopes the data in the new study can provide solid numbers to inform the debate. “The point in our paper is not that you should not spread the money around, but you should know why you’re spreading the money around.”

Others say the argument against further enriching highly productive intuitions is already clear. “Those institutions are in a position to make faculty that much more productive … because they’ve received large amounts of research support over many, many decades,” says University of Vermont Vice President for Research and Economic Development Kirk Dombrowski, who also serves as board chair for the EPSCoR/IDeA Foundation, a nonprofit that promotes science in underfunded states. Allocating resources to them disproportionately would “reflect more of the historical inequities that have created differences,” he says.

The National Institutes of Health (NIH) has made some steps to try to address concerns that researchers at elite universities benefit from reputational bias in the review process. In January, the agency changed its grant review procedures to try to de-emphasize the importance of a researcher’s expertise and institutional resources. Reviewers can note potential concerns in those areas, but they’re no longer given a numerical rating. “You should judge [a proposal] on the quality of the grant that’s in front of you,” says Sharlene Day, a cardiologist and physician scientist at the University of Pennsylvania and former chair of an NIH study section.

NIH and the National Science Foundation (NSF) also have long-running programs, called IDeA and EPSCoR, respectively, that reserve a portion of their budget for projects in states that receive the least funding through traditional funding tracks. As the prospect of major funding cuts looms, some worry about the future of such programs. But in comments last month at a Senate hearing about the upcoming fiscal year’s budget, NIH Director Jayanta “Jay” Bhattacharya voiced his support. “It’s absolutely vital that NIH investments are geographically dispersed,” Bhattacharya told West Virginia Senator Shelley Moore Capito (R), whose state receives less than 1% of NIH grant dollars annually. “In my mind, it’s probably … less funded than it ought to be.”

Typically, about 94% of NIH’s budget goes to researchers in just 27 states, points out Prakash Nagarkatti, an immunologist at the University of South Carolina and former university administrator. But there’s evidence that when grants go to researchers in other, less well funded states, “they are really productive, and they publish quality papers.” His own research, published in PLOS ONE in 2023, shows that in underfunded states, the research community publishes more research articles and garners more citations per million dollars in federal grant funding than those in states with greater federal support.

But spreading out the funding makes sense regardless of recipients’ output, Nagarkatti adds. Additional funding for hiring and training a Ph.D. student, for example, could boost the pipeline of locally produced researchers and result in work that solves regional problems. “Overall, every state gets benefit[s], rather than few states.”

Read More