OUR RESEARCHERS FACE MOUNTING CHALLENGES

These rapid changes demand that our institutions adapt. As will be addressed in Chapter II, significant portions of our federal funding apparatus remain anchored to outdated assumptions. This is not an indictment of the talented scientists and grantmakers who staff these bureaucracies, but a product of institutional inertia, born from the lack of market selection pressure that drives constant experimentation.

NSF, for instance, still organizes itself primarily around academic disciplines, much as it did in the 1950s, and channels resources overwhelmingly to a single type of performer, the university-based, principal investigator-led research group. Many federal programs are still built around the linear model, making the assumption that basic research happens in academia while development happens in industry. Academic incentives often penalize rather than reward partnerships that cross institutional boundaries. Agencies face little external pressure to adapt, even as the scientific landscape transforms around them, preserving processes essentially unchanged for decades.

Within academia itself, well-documented inefficiencies compound these structural problems. Administrative burdens on researchers have grown over time; one study found that investigators spend nearly half of their federally funded research time on paperwork rather than on research or teaching.17 Universities also extract significant overhead from researchers, which funds a mix of legitimate shared infrastructure and growing administrative bloat.

Effective indirect cost rates at NIH-funded institutions average more than 40%,18 even though those same institutions frequently accept 10 to 15% overhead from private funders.19 The assumptions embedded in our federal grantmaking institutions, and the creeping inefficiencies within the universities they predominantly serve, reinforce one another. While researchers themselves recognize the need for renewal, few Administrations have had the will and mandate to pursue transformative change.

A TIME OF URGENT SCIENTIFIC NEED

On the global stage, America no longer stands alone as the unchallenged leader in science and technology. The United States risks losing its first position to competitors that are quicker to update their models of research funding, maintain greater institutional flexibility, and can build tighter connections between different sectors of their scientific enterprise.

Over the past twenty years, China’s R&D spending has surged from a negligible fraction of U.S. levels to full parity on a purchasing-power-adjusted basis by some metrics (Figure 2). This is a situation America did not face even during the Cold War, when the Soviet Union’s economy was appreciably smaller than our own.20 Beyond the sheer scale of its spending, Beijing is treating scientific research capacity as a central pillar of global competition, elevating decision-making to the highest level and taking a tightly integrated approach to partnerships across its sectors. This all-of-society approach has allowed China to surge resources toward basic scientific research, advance rapidly in technologies of key national interest, and drive improvements in its innovation process, where entrenched interests have resisted change. China is not the only country reforming its scientific enterprise. The United Kingdom, for example, has restructured its funding system to address internal inefficiencies, including creating a metascience unit to collect data and develop novel funding mechanisms. Norway has organized its national research council around a portfolio-based model, with boards that allocate funding across thematic priorities drawn from the government’s long-term research plan, alongside disciplinary portfolios. With new ideas bubbling up across the world and within Figure 2: For the first time, the United States faces a peer-level competitor in R&D spending on a purchasing power parity (PPP)-adjusted basis.21

GROSS DOMESTIC EXPENDITURE ON R&D 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2024 1,200 1,000 800 600 400 200 0 Billions of Current PPP Dollars United States EU 27 China

our own vibrant philanthropic, metascience, and academic communities, it falls on the United States as the world’s premier scientific power to take a hard look at how we can accelerate our own scientific machine as well.

REGAINING LEADERSHIP

Beyond strengthening science, we also have a duty to ensure that its downstream benefits accrue to the American people first.

For decades, the United States has funded large cohorts of foreign students and tolerated technology transfer abroad under lax security standards. As of 2024, temporary visa holders accounted for around half of U.S. doctoral graduates in computer science and mathematics with confirmed postgraduation plans.22 This reliance on foreign talent sidelines American students, a deep domestic talent pool that remains under-supported by its own government. At the same time, we train extraordinary global talent at enormous expense, only to lose this effort when foreign governments recruit them to build up their own technological enterprises. The interest of vast scores of Americans in using taxpayer dollars to invest in our STEM pipeline has, consequently, eroded.

America’s over-reliance on foreign students also creates a security challenge our institutions are ill-equipped to address. While our R&D ecosystem is now less vulnerable to exploitation due to Presidential actions taken in the first Trump Administration, there is an urgent need for a comprehensive approach to research security.23 Our competitors have learned to exploit a deep asymmetry between open and closed scientific systems, using our education system as an entry point into our innovation base while building domestic programs that undermine global scientific norms. The United States conducts research openly, publishes freely, and shares methods transparently. These values are central to scientific progress. However, there is a critical difference between sharing information on one’s own terms and being exploited; the parts of a scientific enterprise that are not shared are often its comparative advantages. These are precisely what our competitors work hardest to extract, leveraging human capital educated in our universities, harvesting data from our papers while restricting access to their own, and scaling breakthroughs first achieved in our laboratories on their factory floors. For too long, the United States has believed that technological leadership can be secured solely by the discoveries made in America, regardless of whether those discoveries are then translated into products, capabilities, and industries here or on foreign soil. The consequences of this unguarded openness, in laboratories and in markets, have been severe. The United States pioneered many key enabling technologies for Extreme Ultraviolet (EUV) lithography,24 yet the only company capable of manufacturing EUV lithography machines today is headquartered in Europe. We pioneered lithium-ion batteries, yet Asian firms dominate global supply chains, and by extension, battery chemistry research. We developed advanced manufacturing techniques that now underpin factories abroad. Discovery without domestic manufacturing leaves America paying the research bill while rivals develop the process improvements and capture the economic, strategic, and knowledge returns.

THE PRESIDENT’S CHARGE

As our competitors copy and adapt our scientific machine for themselves, sustained leadership requires us to keep innovating. Standing still while the world changes is not stability. The Manhattan Project succeeded not only because of brilliant physicists, but because we built new institutions capable of coordinating the activity of thousands across the basic science and immense engineering challenges needed to build the bomb. The Apollo Program reached the Moon because Figure 3: The share of U.S. doctorates awarded to temporary visa holders with definite postgraduation commitments has doubled over four decades, rising from roughly 20% to 40% in natural sciences and engineering, and from 10% to 20% in social, behavioral, and health sciences.25

PROPORTION OF DOCTORATES AWARDED TO TEMPORARY VISA HOLDERS 1980 1984 1988 1992 1996 200 2004 2008 2012 2016 2020 2024 50% 40% 30% 20% 10% 0% Proportion of Temporary Visa Holders Natural Sciences & Engineering Social and Applied Life Sciences

NASA could marshal resources and talent in ways no university or company could match. Each represented not merely new funding, but fundamentally new ways of organizing scientific work.

We must face the reality that more innovation is happening in industry than ever before, and that the balance and nature of work shared across the federal, university, and corporate pillars of the national research enterprise have shifted. We must acknowledge, too, that the linear model of discovery and technical progress no longer holds. The interplay between basic research, regular professional science, and commercialization is far more complex than had been assumed. Meanwhile, the context in which the American innovation enterprise operates has become global, and therefore intensely vulnerable; the technologies we invent rely on production chains that stretch around the world, and they are subject to theft by near-peer competitors. Finally, the technological context in which science is conducted and shared has been radically transformed by the internet and current information technologies, now changing even more with the rise of AI.

Throughout our history, Americans have reinvented, reformed, and refounded our institutions when the moment demanded it. Each generation of American scientists, inventors, and pioneers has seized the opportunity to expand the frontiers of knowledge. On the occasion of America’s 250th anniversary, we must remember that ours is a Republic defined by courage, innovation, and exploration. Just as scientific inquiry demands that we revise our theories when evidence contradicts them, evidence of scientific slowdown and serious competition from abroad should spur us to experiment with new systems, new models, and new ways of funding, conducting, and translating research. Vannevar Bush’s pioneering spirit calls us to do what he would surely do today: reimagine the entire enterprise for our time.

THE SCIENTIFIC MACHINE IS GETTING BOGGED DOWN

For the better part of a century, one of America’s most decisive advantages has been the ability to harvest novel discoveries for the prosperity of the American people. This advantage stemmed from the strength of our post-war innovation ecosystem: our universities, our national laboratories, and the symbiotic relationship between federal funding and research. As Bush wrote in 1945:

A nation which depends upon others for its new basic scientific knowledge will be slow in its industrial progress and weak in its competitive position in world trade, regardless of its mechanical skill.26 However, our scientific dominance today is at risk. While our capacity to drive breakthroughs in basic science remains the envy of the world, as described in Chapter I, competitors are closing the gap. And while federally funded science continues to generate a high return on investment for our taxpayers, a growing body of work provides evidence that, across many fields, our researchers are fighting against an increasingly calcified system that has driven up the cost of scientific progress over time.27

Despite massive funding increases in biomedical research since the 1990s, the rate of significant breakthroughs appears to have slowed, drug approvals have flatlined, and the enterprise’s productivity, hampered by growing burdens, has declined.28 It has become common to speak of “Eroom’s law” (Moore’s law in reverse) describing the predictable decline in the number of new drugs approved per billion dollars spent. Since 1950, pharmaceutical R&D efficiency, as measured in new drugs per billion dollars, has fallen roughly eighty-fold in inflation-adjusted terms, halving approximately every nine years.29 And although NIH’s budget has more than doubled since the 1990s, we have not seen a

proportional increase in breakthrough treatments, citation impact per dollar spent, or scientific productivity.30 As a funder and institution-builder, the Federal Government has fallen behind in creating environments where American scientists can do their best work.

Evidence shows that the inputs required to sustain past rates of improvement have increased sharply across many fields. A famous study demonstrated that sustaining the historical pace of Moore-style gains in transistor density has required a much larger workforce. Since the early 1970s, the number of researchers needed to double transistor density has risen more than eighteenfold, implying a 7% annual decline in “ideas productivity.” The pattern repeats in agriculture, where research effort has multiplied by factors of 3 to 25 since 1969, depending on the metric, while yield growth remains mostly flat; and in medicine, where the “years of life saved” per clinical trial peaked in the mid-1980s before falling sharply.31 Admittedly, slowdown in mature scientific subfields may be inevitable. One might argue that it is natural for the pace of progress to decline, once the proverbial low-hanging fruit has been picked. A slowdown could even be read as evidence of success. However, this intuition has repeatedly been proven wrong throughout modern history. And one would expect new mechanisms for sharing information, and new ways to compress scientific knowledge, to be countervailing forces that speed innovation.

The pattern of apparent stasis in a scientific field exploding into progress, opened by a new discovery and changes in scientific institutions, has repeated itself again and again. These punctuated equilibria are, in fact, the essential story of science. Max Planck famously had a professor tell him that physics was nearly as developed as mature fields like geometry, only for Planck’s own discoveries in quantum mechanics to completely reorient our understanding of the physical world. Many medical doctors believed their field was reaching perfection in the late 19th century, with one writing that “there cannot always be fresh fields for conquest by the knife.”32 Yet soon, the concurrent transformation of medical education and emergence of research hospitals created the institutional foundations for a broader understanding of disease and for modern medicine as we know it.33

Repeatedly, the tree has only looked bare from the current perspective; the fruits have not been exhausted at all. We had merely lacked the tools with which to pick them.

By adopting new social and material technologies, we can again accelerate the pace of discovery. It falls to us, as it fell to our predecessors, to imagine new machines capable of exploring the endless frontier. 15 Chapter II – Revitalizing America’s Science and Technology Enterprise

SLOWED BY GROWING FRICTIONS

Our first step is to strip away the frictions that keep our brightest minds from pursuing the ideas most likely to lead to transformative breakthroughs. Consider a young scientist with a promising proposal for federally funded research, and the decades of accumulated institutional bureaucracy she must navigate to seek support in today’s enterprise. She spends two to four months drafting the proposal, assembling preliminary data for the same project requiring funding, and navigating her university’s internal review process. If she applies to NIH in February, she will be lucky to learn whether she succeeded by the end of the year. Certain grants even have up to a 20-month lead time.34 That is almost as long as it took for the Boeing 747 “Jumbo Jet” to go from the drawing board to production.35 If she is awarded the grant, she will face mounds of paperwork. From 1991 to January 2025, the Federal Government imposed at least 270 new requirements on research grants, far outpacing efforts to reduce administrative burdens on researchers.36 Federally negotiated indirect cost rates now reach 50 to 60% of direct research costs at major institutions, a figure our scientist will have to bear in mind as she drafts her application.37 While effective rates often run closer to 40%, this remains a substantial tax on research budgets, shaping what she asks for. Some of this covers legitimate infrastructure she uses every day, but much of it funds administrative expansion at her university that has outpaced the growth of research itself.

Senior investigators can delegate paperwork to postdoctoral researchers, but our scientist runs a small lab and has no one to delegate to. She writes grant applications using hours that should have gone to experiments or mentoring students. This is a tax on innovation that does not appear in the federal budget but costs the nation dearly in foregone breakthroughs. These burdens also create a perverse incentive structure, in which scientists who excel at research administration leapfrog those who excel at research performance. The weight falls heaviest on the scientists America needs most.

THE INCUMBENCY TAX

The academy is a long and difficult road, leading to few stable positions. We should encourage early-career scientists at every step of the academic crucible, from undergraduate lab assistant to first faculty job, to pursue big and creative ideas. We should enable and empower scientists who persist into a research career to focus on breakthrough research from the start. But the data suggest we do not.

Between 1980 and 2008, the average age of NIH principal investigators rose from 39 to 51 (Figure 4), while the average age of new principal investigators rose from 36 to 42, exceeding the average age of Nobel Prize-winning contributions in related fields over a comparable period.38 These patterns lengthen feedback loops and bias careers toward safer, incremental projects during the long apprenticeship years. Our early-career scientist has seen this gradual graying of our research workforce, and is likely to adjust her ambitions accordingly. Our scientist may accept that these longer training cycles reflect increasing specialization, but these incumbency dynamics also dampen turnover at the frontier. The effect is encapsulated by Planck’s famous (mis)quote that “science advances one funeral at a time,” and is well documented.40 Studies show that when a star scientist in biomedicine passes unexpectedly, outsider contributions surge into the space the star’s network has informally dominated, and those outsider papers are then more likely to become highly cited.41 At the system level, researchers find that as disciplinary fields grow large, attention ossifies around a fixed canon. New papers are less likely to displace

Figure 4: The average age of first-time NIH investigators has increased consistently over the past four decades, from mid-thirties in 1980 to the early forties today, across all degree types.39

AVERAGE AGE OF R01–EQUIVELENT FIRST-TIME INVESTIGATORS 1980 1984 1988 1992 1996 2000 2004 2008 2012 2016 46 44 42 40 38 36 34 Average Age MD–PhD PhD only MD only

central ideas, and even highly cited ones tend to receive citations in a burst rather than through steady accumulation.42

WEAKENED MERITOCRACY

Our up-and-coming scientist is further discouraged by the corrosion of the merit principle that once made American research the envy of the world. What began with the NSF’s “broader impacts” criterion, which was a reasonable effort to ensure taxpayer funds benefit society, has evolved into a sweeping distortion of the selection process that reduces mobility for the best researchers. Between 2021 and 2024, the share of new NSF grants focusing on diversity, equity, and inclusion initiatives surged from fractions of a percent to more than a quarter.43 Our scientist has learned the new rules from principal investigators she has worked for, and is tempted to dress up her otherwise excellent technical proposals with ideological language to survive review. Until recently, NASA required research proposals to include plans for furthering “inclusion goals,” which were to be reviewed by review panels one-half composed of “diversity, equity, and inclusion professionals.”44

The American scientific establishment has placed politics above merit and performance before. In the 1920s, elite universities implemented quotas limiting Jewish enrollment. Harvard’s president worried about a “Jewish problem” as Jewish students grew from 6% to 22% of the student body. The methods are strikingly familiar: subjective criteria for “character” and “leadership,” diversity requirements designed to recruit from regions with fewer Jewish residents, and “holistic” reviews that obscured the actual basis for decisions. These quotas, maintained for decades, excluded some of the most talented minds in American science.45 We should learn from this history rather than repeat it. When qualified candidates are passed over for reasons unrelated to their scientific ability, when they are judged for who they are rather than for their ideas, we injure the cause of progress, discovery, and America’s scientific competitiveness.

MISALIGNED INCENTIVES

Even if the ambitious young scientist successfully navigates the selection gauntlet, she enters an academy whose incentives are severely misaligned with good scientific conduct.

Observing older peers, our scientist has learned that taking the long shot to challenge established paradigms may threaten her ability to deliver results and advance to tenure; yet by the time she achieves tenure, a substantial portion of her most creative years will be behind her. She notices, too, that the tenure and promotion system seems to reward quantity over quality. Cutting work into the smallest publishable units, which researchers call “salami-slicing,” often pays off more than making an ambitious attempt at transformative discovery. Our scientist knows a better way is possible. Private-sector laboratories and startups routinely give young researchers tremendous responsibility, and there they are afforded the opportunity to change the world. Research shows that when investigators receive longer-horizon support with tolerance for early failure, they produce portfolios with both more hits and more misses, the signature of genuine exploration.46 But project-tied, short-cycle grants dominate the federal landscape, and our scientist feels pushed toward safer, more “fundable” territories. These incentives also help explain why her peers now commonly reach professional independence only in their forties, not their early thirties as in previous generations.

The publication economy amplifies these problems. The young scientist quickly learns that there is a particular narrative pattern common to all articles published in the top journals. Meta-research has documented a structural reduction in novelty in papers and patents, which are less disruptive and less likely to reorient a field than in prior decades.47 Scientists eschew negative results, even though failed experiments often teach more than successes. Competitive environments amplify this positive-results bias, crowding out careful negative findings and tool-building that lack tidy narratives. Because the system pays by the paper, it under-invests in public goods like datasets, open-source code, and shared engineering infrastructure. The team science now required at the technological frontier gets abandoned for work that generates individual credit.

THE REPRODUCIBILITY CRISIS

As many leaders in American science agree, the research enterprise must learn to value and incentivize reproducibility studies and to hold its members responsible for failures in the scientific process. The reproducibility crisis, particularly acute in the social sciences, has undermined future research and public confidence. In one study that attempted to reproduce 100 psychology studies, fewer than 40 succeeded.48 The challenge extends across other fields as well. In Alzheimer’s research, a celebrated 2009 paper

in a top journal presented a promising path to treating the disease.49 By 2012, other researchers had demonstrated its irreproducibility, and internal reviews at the sponsoring pharmaceutical company terminated drug development based on its findings.50 Yet the paper accumulated more than 800 citations, misdirecting research priorities and federal funding for another decade.51 Its lead author became a university president, and the paper was only finally retracted 15 years after publication, shortly after the lead author resigned amid a broader investigation into data manipulation in his laboratories.52 The scientific review process worked, eventually, but far too late.

The good scientist knows that knowledge is probabilistic, that evidence accumulates gradually, and that uncertainty is inherent. But the system pressures researchers to deliver results in confident, discrete units. Journal editors want clear narratives. University press offices demand headlines. Nuance dies, as a result, in the race for attention. This failure has spilled over to political decision-making. With notable and still-undercelebrated exceptions, the scientific establishment during the COVID19 pandemic failed to recognize that science can only describe the world as it is, not the world that ought to be; while it informs policymakers, it cannot determine the best policy, or the tradeoffs that should be made. The scientific consensus to shutter schools demonstrated a profound inability to confront uncertainty or integrate knowledge across specializations. The best available evidence indicated that children were neither at high risk of the disease nor significant vectors of transmission. Conversely, the developmental costs of remote schooling and isolation were entirely predictable. Yet, scientific officialdom produced a “closedranks” response, preferring blind consensus over informed dissent. The scientific consensus failed to check its own work, communicate the limits of its own certainty, or remain skeptical of its own assumptions.

A LACK OF ACCOUNTABILITY

Even as our individual scientists continue to do world-changing work, performing miracles that save American lives and defend our homeland, these challenges reflect a systematic breakdown in the lines of accountability that align the scientific enterprise with the public interest. While repeating the mantra that science must be “independent” of politics, parts of academia have become highly dependent on government funding. But as Bush understood, federal support for science must be politically accountable. Accountability does not mean dictating how a research agenda is to be executed, nor turning away from the basic research that has long been the wellspring of American prosperity. Quite the opposite. It means ensuring that the system serves the researchers who are its lifeblood, rather than the entrenched interests that have accreted around them. Only those parts of the national research enterprise directly responsive to the political process can prioritize among the many potential avenues of inquiry, fund those that best reflect national priorities, and be checked when the scientific process breaks down. In a self-governing nation of laws and citizens, the Federal Government elected by the people shall have the ability to determine how to allocate public resources in the public interest. Within that framework, the independent role of federally funded scientists is to design and execute the research and experimentation program that achieves those objectives.

A properly accountable system is one that protects their freedom to do so. Originally intended to shield researchers from external political meddling, the invocation of “scientific autonomy” has too often been inverted. Autonomy, and indeed America’s culture of intellectual freedom, remains one of our most valuable scientific assets, but it has been cynically used as a shield against accountability, as scientific institutions are slow to police themselves, or even engage in political meddling of their own, often at the expense of our brightest and most energetic minds. The same scientists who receive grants often serve as the reviewers who dispense them, enforcing a consensus that perpetuates existing biases. Federal funding agencies, in particular, lack the feedback mechanisms found in the private sector. In venture capital or philanthropy, poor judgment faces the swift discipline of the market; bad bets lead to insolvency. In the federal sphere, absent deliberate action, there is no penalty for rejecting a breakthrough, nor for funding safe work that changes nothing.

A BETTER PATH FORWARD

One necessary step in restoring accountability to federally supported science is exercising better oversight of scientific funding so that research activities align with the best interests of the American people, through the intentional and principled allocation of capital. Our federal agencies distribute approximately $200 billion in annual R&D funding. Yet we have no systematic framework for identifying where those dollars could catalyze the greatest scientific returns, with deference instead given to the same incumbents that consume the funding. This process produces a portfolio that emerges by accident rather than intentional design.

Driving meaningful improvements in our portfolio allocation will require a coordinated effort, but it is not impossible. We already have proof that bold models can dramatically accelerate progress and deliver new scientific opportunities for our researchers.

Throughout history, Americans have developed various ways to support scientific progress that depart from the university-based, principal investigator-driven grant. These include Cold Spring Harbor Laboratory, a leading institution of biological research, founded in 1890, and the Institute for Advanced Study, founded in 1930, which gave luminaries like Einstein, Noether, Oppenheimer, and Gödel an opportunity to explore revolutionary ideas. Each provided a home for new kinds of science.

Another famous example is DARPA, which gave us GPS, the internet, stealth aircraft, and autonomous vehicles. Rather than relying on consensus-based review panels as the primary decision-making mechanism, DARPA gives individual program managers the power to make bold technological bets and actively curate teams to execute them. Congress has taken notice, creating the Advanced Research Projects Agency for Health (ARPA-H), Advanced Research Projects Agency–Energy (ARPA-E), and other similar agencies, collectively representing billions of dollars organized around the program-manager model rather than traditional peer review. Over the past decade, a growing community of researchers, philanthropists, and policymakers has turned the lens of scientific inquiry onto science itself, asking precisely how we can reduce frictions, align incentives, increase accountability, and inject more dynamism into the scientific enterprise. This field, often called metascience or the “science of science,” has begun generating rigorous evidence of what actually works, both by studying why models like DARPA or the Institute for Advanced Study succeed, and by running controlled experiments on new approaches.

Philanthropies and federal agencies are now deliberately applying these insights and piloting new organizational forms to address gaps the traditional academic system cannot fill. For instance, NIH has recently created a metascience office, and the new Directorate for Technology, Innovation and Partnerships (TIP) within NSF has begun experimenting with alternatives to traditional peer review, designing and executing controlled experiments in alternative funding mechanisms in partnership with metascience researchers.53 These scattered successes confirm the possibility of systematically redesigning how we organize, fund, and conduct science. Realizing that potential will require us to understand how the nature of scientific work has changed, what new institutional forms those changes demand, and what funding mechanisms can best support them.

The answers will not come from any single reform but from building the capacity for continuous experimentation, for applying the scientific method to the scientific enterprise itself.