In a rare show of bipartisanship, the Senate Commerce Committee last week voted 24–4 to advance legislation that would massively increase funding for technology-oriented research at the National Science Foundation. The bill, known as the Endless Frontier Act, is a particular favorite of Senate majority leader Chuck Schumer (D., N.Y.). But the basic idea behind it has broad support across party lines, particularly as a way to respond to China’s growing technological prowess. It may get a vote on the Senate floor as soon as this week, and in some form it has a pretty good shot at ultimately getting enacted.
But “in some form” is no small caveat. The committee markup made clear that the particular shape and scope of the bill are very much in flux. They may well change again on the floor, and surely will in negotiations with the House, where a related but different measure has already been taken up.
It’s a good thing that the details remain unsettled. The idea behind the legislation is a good one, broadly speaking. But legislation cannot stop at speaking broadly, and in several important respects the Senate bill risks mistakes that could do real harm.
The bill has many parts, but at its core as originally envisioned has been a proposal to inject more than $100 billion into the budget of the National Science Foundation over the coming five years to support technology-oriented research in areas such as artificial intelligence, semiconductors, quantum computing, biotechnology, energy research, and others. That kind of funding would be a massive boost to the agency’s budget, which was $8.5 billion this past year.
The NSF supports vital basic research, and providing it with more resources to do that makes a lot of sense. But that kind of immense infusion of money would be transformative in ways that have to be thought through. Three questions in particular are worth considering in advance. They involve the culture of the NSF, the existing array of federal-research investments, and the pace of new funding. The first two are clearly already on the minds of some legislators, but the third and most crucial seems not yet to be.
Science or Technology?
The first question has to do with the effect of the proposed funding increase on the National Science Foundation’s core mission. The architects of the Senate bill want to channel a lot of its resources toward the creation of a new directorate for technology research at NSF. At this point, the NSF has seven research directorates: biological sciences, computer science, engineering, geosciences, mathematical and physical sciences, social sciences, and education and human resources. All of these divisions focus mostly on basic rather than applied research. In this century, they have delved a bit more into translational research and some technological applications, but they still stay out of real mission-directed sorts of R&D work of the kind envisioned in the bill.
My American Enterprise Institute colleague Tony Mills, the Right’s foremost scholar of science policy, has noted that this emphasis on basic science has always been the essence of the NSF’s mission, but that there have been efforts to change that ever since the agency’s founding in the earliest years of the Cold War. Those efforts have been moved by a sentiment like the one that now prevails in Congress: To maintain our edge in the world, we need strategically directed technological innovation, not just advances in theoretical research. But the NSF is built on the understanding that technological innovation has to be rooted in basic science, and that advances in basic science are more in need of the kind of investment that only public support can sustain.
The enormous new investment in applied research and technology that is envisioned in the Endless Frontier Act presents itself as less transformational than those past efforts to remake the NSF. It is intended to come alongside the agency’s existing commitments to basic science, not in place of them. But the culture of the institution would unavoidably be transformed by that kind of infusion, and it is crucial to ask whether this is desirable.
The NSF is distinct for its commitment to basic research, and over the years it has supported lines of inquiry that could not have been justified in terms of practical application but that nonetheless proved enormously important both for the advancement of knowledge and for the development of applications that might not have been envisioned in advance. If it were gradually transformed into a funding agency for strategically prioritized technological projects, it would lose its unique purpose and distinct value. It is far from clear that the gain in technological development could make up for that loss, or whether a large investment in technology-oriented research might not be better housed elsewhere in the federal bureaucracy to avoid deforming the NSF’s essential work.
The Right Agency?
This points to the second question legislators should consider: Wouldn’t this immense new investment in the NSF be duplicative of efforts already underway elsewhere in the federal government, or at least be better placed elsewhere?
The most obvious place where such efforts are underway is in the system of 17 national laboratories. These labs conduct their own research, which NSF does not, but they also support outside research and would like to do more of that. Although they are nominally operated by the Department of Energy, the labs work on a broad range of research in the natural and physical sciences, and their emphasis in many cases is on precisely the kind of strategically significant technological advancement that the new bill envisions. They exist to direct R&D toward national goals, which is exactly what the Endless Frontier Act is trying to encourage, but not at all what the National Science Foundation exists to do.
The system of national labs is also designed strategically in a different sense: The labs are spread throughout the country in a way that has given them real political prowess in Congress. There is one, for instance, in Morgantown, W.Va. — a fact we can be sure is well known to the Senate’s most powerful member just now. Others are distributed in influential states in a pattern that is likely to prove important.
The funding envisioned by the Endless Frontier Act might not only be duplicative of some of what these labs do, it might also be irresistibly attractive to the labs, and given their political prowess we should not be surprised if they end up gobbling up large portions of it by the time the bill gets a final vote. Last week’s committee markup already provided some serious evidence of that. Resisting that tendency would require a real case for the need to direct this money through a new funding channel at the NSF rather than through existing ones elsewhere — and it’s not clear that any such case has been made, or could be.
In this respect, the committee markup moved the bill in a good direction, even if not quite for the best reasons. Pressure from some senators (especially those with national labs in their states) led to passage of an amendment dividing the massive infusion of dollars between the proposed new technology directorate and more traditional NSF investments in basic science, and then the former portion was again divided between that new NSF directorate and the existing system of national labs. This was an example of legislative bargaining rooted in classic interest-group pressures plainly improving a legislative proposal by better rooting it in the real world.
When the bill reaches the Senate floor, however, there may well be some pressure from its original architects to restore its scope and focus. We can be sure the question will come up again in negotiations with the House. Legislators thinking through these pressures should consider not only the political power of the labs but also the logic of the existing structure of federal R&D investment. Getting the balance right could make the difference between transformational investments and a lot of wasted public dollars.
Too Much Too Fast?
The third question that legislators need to wrestle with is the most important, and also the most complex and counter-intuitive. It has to do with the pace of increased funding. A number of champions of the Endless Frontier Act have pointed to the doubling of the National Institutes of Health budget over a five-year period in the Bill Clinton and George W. Bush years as a model for what they hope to do. But that model offers cautionary lessons as well as encouraging ones.
The NIH budget was doubled between 1998 and 2003, in an effort spanning four Congresses and two presidents, and spearheaded by a broad and bipartisan group of legislators. Those members of Congress, and both Clinton and Bush, saw the rapid doubling of the budget as a means of expressing the country’s commitment to medical research, and spurring vital biomedical innovation. It was an impressive and effective effort, and it has surely borne some fruit since that time.
But in retrospect, it is clear that the doubling of the budget happened much too quickly, and that this did genuine harm to the very biomedical-research enterprise it was intended to help. The steep growth in spending, roughly 15 percent per year for five years, built expectations and momentum that set the NIH up for disappointment when the doubling was done. Academic research institutions used the added funds to start new projects, support many more graduate students, and establish more programs and labs. But these all required continuing support beyond the five-year doubling period, which meant that once the steep increases were done and the NIH budget returned to normal growth rates of about 3 percent per year, an enormous portion of the (now much larger) budget went towards continuing support for projects established in the fat years, and not enough was left to fund new ventures or support the careers of all those new scientists.
The resulting pain has worked through the system by now, and at this point you’d surely have to say that the doubling of the budget was very beneficial on the whole. But that pain was real, and led to real setbacks and deformations in American biomedicine that the physical sciences should not now want to emulate.
This was first brought home to me in the mid 2000s. I was a policy staffer in the Bush White House working on health issues, and so it often fell to me to meet with delegations of prominent research scientists interested in talking about public support for their work. The University of California system was especially good at getting their scientists on our schedule at the White House, but many other elite research institutions did too. And from the outset I was stunned by the tone of these meetings: These prominent and generously funded researchers were bitter and angry. They had nothing but complaints about the support they were getting from the NIH, and they were also deeply worried about the prospects of their most promising projects and students.
At first, I couldn’t understand it. The work these researchers were doing was not only lavishly funded but much more lavishly funded than it had been just a few years earlier. The Office of Management and Budget has a very fine-grained database of NIH funds, so I got into the habit of going into these meetings with an OMB chart showing the trajectory of public dollars supporting the work of the particular researcher I would be meeting with. When he or she launched into a tirade, I would take that chart out of my notebook and point to what was often on the order of a 70 percent increase in funding over the prior few years for the very work this person had come to talk about.
I thought this was very clever. But the scientists were rarely fazed by it, and went right on complaining. And while at first I took this to be evidence of a staggering sense of entitlement, I came to realize over time that they were actually right. They had been encouraged by the rapid doubling of the budget to make all kinds of investments and launch a variety of projects, and when the doubling was over, they felt like they’d been thrown into a wall. The rhetorical and political logic of doubling the budget in five years failed to take account of how academic research actually works, and at least in the short term it was harming precisely the kinds of elite researchers and cutting-edge projects it was meant to help.
It’s hard to overstate the damage this did to the morale of American biomedical science in the mid 2000s. Researchers watching fledgling projects crumble became convinced they were being targeted by anti-science politicians even though those very politicians had just doubled their budgets.
I would argue, for instance, that this dynamic contributed enormously to the bizarre tone of the embryonic-stem-cell debates of that era. Cell biology had been a particularly prominent beneficiary of the doubling of the budget, and so was high among the victims of the sudden deceleration. And the resulting morale hit drove some prominent researchers to leap into politics in ways that abandoned all professional responsibility. They were soon being used by politicians to back up all kinds of unfounded claims about the state and the potential of stem-cell research, and allowed themselves to be dragged into stunningly unprofessional and just frankly disgraceful forms of politicization, dishonesty, and outright charlatanism — all the while using ill patients as props and lying to people who were desperate for hope. Two decades later, those false promises look worse than ridiculous, and the people involved have long since stopped talking that way, but they have never explained their behavior. The dynamic of the budget expansion doesn’t excuse that behavior. Nothing could excuse it. But it does, I think, help to explain some of it.
This lesson of that era applies directly to the Senate’s Endless Frontier Act and its companion legislative measures in the House. Here, too, Congress is looking to massively increase funding for a set of academic research enterprises over a five-year period in a way that fails to account for the basic structure and dynamics of academic research. Expanded funding should grow slowly but over a longer period of time, to allow for new projects to not only start but also be sustained, and to make it possible for new cohorts of researchers to find their places in the scientific community.
Slower growth would also make it possible for funding to reach beyond the small cadre of elite institutions that would be able immediately to digest a huge influx of dollars. Here, too, the NIH experience is illuminating. While the size of the average NIH grant grew significantly during the five-year doubling of the agency’s budget, the pool of institutions receiving funding hardly changed at all. As the Chronicle of Higher Education found in reviewing lists of NIH grantees in the wake of the doubled budget, of the 515 institutions receiving direct grants to investigators in 2003, only five had not been receiving such grants prior to 1998. Moreover, the portion of grants going to the top 100 research institutions in 2003 was 92 percent, exactly the same as it was in 1998. In other words, more NIH funding meant more money for the same relatively small group of ultra-elite institutions, and even there it contributed to a surplus of new projects and researchers that quickly became burdensome once the growth spurt ended.
Deformed growth of that sort could do real damage to the kinds of technology-oriented research that legislators are looking to spur with this bill. That doesn’t argue against the added funding they envision, but it argues for considering how that infusion of dollars will affect the research culture it seeks to strengthen.
Matching Means to Ends
The Endless Frontier Act is moving quickly. But even if the Senate approves a version of it this week, it will still need to be merged with a House bill and reconsidered at least one more time. As that happens, it’s important that legislators in both houses carefully think through the implications of how they’re designing this effort. They are right to want to give American R&D a boost, particularly with a wary eye toward China. But what form that boost should take will matter enormously.
While old-fashioned logrolling and interest-group politics has actually improved the original proposal already, further improvements will require serious attention to the nature of the research culture involved. Legislators should consider whether they want to fundamentally transform the character of the National Science Foundation, whether there might not be other more suitable channels for funding, and how the growth they envision could support and foster an expanded academic research enterprise rather than distort its incentives and undermine its morale.