Ask a bunch of scientists what’s wrong with their field, as Vox did a few years ago, and one thing nearly all of them will name is the funding process. You might think that top scientists at top universities are paid by those universities for the research they do, but for the most part, you’d be wrong: Nearly all academic researchers in the sciences rely on outside grants in order to pay salaries, buy their equipment, and run their experiments.
Those grants end up powerfully shaping the academic sciences. By some estimates, many top researchers spend 50 percent of their time writing grants. Interdisciplinary research is less likely to get funding, meaning critical kinds of research don’t get done. And scientists argue that the constant fighting for funding undermines good work by encouraging researchers to overpromise and engage in questionable practices, overincentivizing publication in top journals, disincentivizing replications of existing work, and stifling creativity and intellectual risk-taking.
A new biomedical research institute, called the Arc Institute, announced on Wednesday as a nonprofit in collaboration between Stanford, UC Berkeley, and UC San Francisco and funded by some of the biggest names in tech, is meant to address some of those problems — and show what could be a better way to fund science.
Arc is “an institutional experiment in how science is conducted and funded,” Patrick Collison, CEO of Stripe and one of the Institute’s funders, told me. Researchers get eight-year grants to do whatever they want, instead of three-year grants tied to a specific project.
The institute is also hiring for people who want to work on improving key biological research tools instead of on conducting experiments and writing papers. It’s an expensive approach that can, even at best, only solve the problems with our current system for a tiny fraction of the researchers affected by it. But its founders hope it can at least show that solutions are possible — and inspire further experimentation.
The key idea behind Arc is that the current system for biomedical scientists has enabled a lot of great research. But there’s an overwhelming sentiment among scientists that some key work is falling through the cracks.
Top investigators “spend over half of our time fundraising. The rest of that time we’re shouldering serious managerial, operational, logistical things,” Patrick Hsu, an assistant professor of bioengineering at the University of California Berkeley and one of Arc’s founders, told me.
The system “doesn’t empirically seem to enable people to pursue what they themselves think is their best idea,” Collison said.
Last year, Collison was involved in funding a major scientific grantmaking endeavor called FastGrants — an effort to get money out to researchers doing critical work on Covid-19. With the funds, they sent a survey. One question included was “if you have the same amount of funding but you could use it however you want, and it was stable, would you change your research program,” Silvana Konermann, an assistant professor of biochemistry at Stanford and the new executive director of Arc, told me.
Eighty percent of the scientists who answered said they would change their research program substantially.
Arc is a bet on doing exactly that. The researchers it brings on as core investigators will get less restricted, eight-year grants. The idea is that they’ll spend less time fundraising and feel secure to switch course and focus on whatever they believe is most essential, giving them what Konermann said is the “flexibility and freedom to pursue the research they’re the most passionate about and take risks and take on projects that might fail.”
The institute is also hoping to better support those scientists with “technology development centers” that focus on inventing better tools and processes for biomedical research. It’s the kind of work that is essential to science but that doesn’t lead to academic papers and that is therefore not seen as a viable career path.
With the current system, “there aren’t long-term careers for the people doing it,” Hsu told me. “It should be possible to work for 20 years on refining some super-useful technology” — for cutting and pasting DNA, for example, or something similarly fundamental. Right now there aren’t good career options in the academic biosciences for people who want to work on slow, slight engineering improvements on a key problem. It’d be nice to change that.
Arc will operate out of Palo Alto, California, where Stanford is located, as an independent nonprofit collaborating with Stanford, Berkeley, and UCSF — all institutions with highly regarded biomedical programs.
It’ll be expensive — part of why no one has fixed biomedical science is that it isn’t cheap. The venture has funding commitments from — along with Patrick Collison (who is Konermann’s husband) and his brother and Stripe co-founder John Collison — cryptocurrency billionaire Vitalik Buterin, and Dustin Moskovitz and Cari Tuna (whose ambitious philanthropy crops up frequently here on Future Perfect), among others.
If you’re a tenured professor doing academic research in many scientific fields, you’ll be responsible for running a lab. That means you’ll need to hire graduate students and postdocs, as well as pay for equipment, supplies, and publication fees.
Universities pay a share of these expenses, but the bulk is expected to come from research grants. In the US, most of those grants come from the federal government: Biomedical research is funded by the National Institutes of Health (NIH), and much other research is funded by the National Science Foundation (NSF).
Grant applications tend to be long and complicated, and often require “preliminary work” on a project, which means a lab has to already commit significant resources toward it before they can even apply for future work on it to be paid for, Hsu told me.
And the number of grant applications to institutions like the NIH has been going up, while funding hasn’t necessarily kept pace, so grant rejection rates are sky-high. In the 1970s when the program got started, the share of NIH grant applications that are approved was about 35 to 40 percent — now it’s about 20 percent. That means most of the time and effort on writing grant applications is effectively wasted.
The process isn’t just time-consuming — it’s also stunningly capricious. One study found very little correlation between how a grant was scored in the NIH process for grant approvals and whether the research it produced was eventually cited.
Another study, looking at high-quality proposals, found there was virtually no agreement on their merits — two researchers might come to vastly different conclusions about whether the grant should be approved. Another analysis looked at successful grants and found that 59 percent of them could have been rejected due to random variability in scoring. Watching how NIH grants were reviewed was “very eye-opening for me,” Hsu told me — and not in a good way.
Then there’s the fact that grants tend to expire after a few years. As John Pooley, a neurobiology postdoc at the University of Bristol, told Vox, that inclines researchers toward tackling something that they can address in a few years. But some of the most important problems in medicine will take much longer to crack.
To be clear, a lot of incredible biomedical research is happening in academic laboratories today. The US leads the world in biomedicine, and key inventions like CRISPR happened in US laboratories. While the system certainly needs improvement, no one wants to lose what makes US biomedicine already lead the world.
“It’s not that the current model is really bad for everyone — I think the current model actually works really well for some people,” Konermann told me. “The hope is not that everything will be like Arc, but that each of these models will have their own downsides and their own upsides,” making for a “healthier overall ecosystem.”
Of course, even if there are flaws in the current system, Arc might not necessarily address them. For one thing, it’s really small and not an option for most biomedical researchers, no matter how much they’d like it. For another, Arc represents a bet on what conditions produce good science — and while their case for their bet is compelling, they could be wrong that giving scientists more flexibility and more autonomy actually produces the lifesaving cures we hope for from biomedicine.
A wide range of ideas have been aired for how to fix the scientific grant process, from lotteries to limiting applications to one page. There have been private attempts to do better — like FastGrants, which aimed to get out Covid-19 research money in 48 hours instead of weeks or months, and which has moved more than $50 million to date.
But one key thing, Collison emphasized to me, is simply having more options. There’s nothing wrong with the NIH grant process as one way that researchers can secure money for an idea. But when it’s the only way, any science that doesn’t fit neatly into the NIH process won’t get done at all.
Arc is one more option, but it will only be able to fund a few core investigators and a limited number of full-time careers working on developing new tools in its technology development centers.
That’s why Collison says one measure of Arc’s success — alongside the more obvious ones, like whether its researchers report that they’re able to focus on science, and whether they discover new things that make the world a better place — is whether “other funders, other institutions, other stakeholders in our prevailing systems, are compelled to themselves pursue other experiments and other models.” If Arc works, then that raises new questions — what else might work? What other potential is going untapped in the current system?
It’s an experiment that the beneficiaries of biomedical research — which is to say, everyone — ought to watch with interest.