Crippling America's Innovation Economy

PUBLISHED ON PROJECT SYNDICATE ON APRIL 4, 2025

In its war on universities and government-funded research, Donald Trump's administration is targeting precisely the institutions that made the United States the world’s leading technological, military, and economic power. With China appropriating the same innovation model for itself, its destruction could not come at a worse time.

CAMBRIDGE – The commanding heights of America’s innovation economy are under siege. President Donald Trump’s administration is threatening world-leading research universities like Harvard, Columbia, and Princeton with a loss of federal money unless they bend to its political demands, and funding for HIV/AIDS, cancer, and other urgent research is on the chopping block. Many leading scientists at the Food and Drug Administration (FDA), the National Institutes of Health (NIH), and other agencies are being hounded out of public service.

It took three generations to build America’s innovation economy, but Trump seems hell-bent on fatally weakening it in as many months. The institutions that he is targeting are key components in an ecosystem that has made the United States the world’s leading technological, military, and economic power for decades. The US owes its dominant status to a diverse range of collaborating entrepreneurs drawn from the public and private sectors and civil society. It was they who won the Cold War, spawned the digital and biotech revolutions, and advanced the frontiers of knowledge and invention across the entire spectrum of potential innovation.

THE AMERICAN WAY

This is a quintessentially American story. From the early days of the republic, as Fred Block of the University of California, Davis notes, economic development “depended on an ongoing partnership between the government and business. Government provided necessary infrastructure such as roads, canals, railroads, and harbors, and helped train the labor force and build the society’s technological capabilities; government agencies worked to facilitate the diffusion of productive innovations in agriculture, industry, and services.”

A central aspect of this partnership was the US patent system, the foundation of which was laid in Article I of the US Constitution. It was radically more accessible than the British patent system, which had evolved from the royal prerogative to grant monopolies, and thus fueled a “market in patents” that accelerated the process of turning inventions into economically significant innovations.

During the Civil War, Abraham Lincoln’s administration mobilized public resources to fund land-grant universities and agricultural research (through the Morrill Act), and to subsidize the transcontinental railway. An institutional connection between national security and government support of science was first established in March 1863, when Lincoln chartered the National Academy of Sciences, a private nonprofit organization that would “investigate, examine, experiment, and report upon any subject of science or art” at the request of any government department.

This was preceded by the key innovation that transformed manufacturing during the nineteenth century: the introduction of interchangeable parts, which standardized the components of everything from sewing machines and typewriters to rifles and bicycles. Production on a much larger scale was now possible, as first demonstrated at the US Army’s Springfield, Massachusetts, armory in the 1830s. From there, precision manufacturing machinery diffused across New England and the Old Northwest, ultimately enabling the mass production of consumer and industrial goods – most notably automobiles.

In the early twentieth century, as continued innovation came increasingly to depend on scientific advances, the great monopolies of the Second Industrial Revolution – AT&T, DuPont, GE, IBM, RCA, Xerox – established research laboratories to maintain their competitive edge. But World War II was an even more important turning point in the development of America’s innovation ecosystem. The creation of the Office of Scientific Research and Development transformed the American innovation ecosystem, bringing an array of new players into the game: the research universities.

THE MODERN US INNOVATION SYSTEM

One sign of the OSRD’s lasting impact is America’s unrivaled leadership in scientific research and discovery. After long lagging behind Europe, the US quickly took the lead after WWII – as measured, for example, by its disproportionately large share of Nobel Prizes (an achievement aided by the exodus of Jewish scientists from Nazi Germany).

But this metric hardly captures the extent of the OSRD’s economic contribution. What the agency really did, Daniel Gross and Bhaven Sampat show, is introduce the “largest R&D shock” in US history: 

“First, WWII R&D kicked off the postwar growth of technology clusters (counties x technologies) around the country. … Second, this sustained growth benefited from, but did not depend on, postwar federal R&D investment. Instead, our evidence suggests OSRD catalyzed self-sustained agglomeration, including firm in-migration, entry, and growing spillovers across inventors and technologies. Third, … these changes were accompanied by growth in local industrial employment and firm creation in related high-tech industries. Finally, … wartime R&D had permanent effects on the direction of US innovation, which pivoted toward electronics and communications.”

A crucial factor was the decision to channel federal R&D funding through universities, rather than exclusively into government laboratories – a policy that endured throughout the Cold War. Only the Manhattan Project was managed directly by the military, and even it grew out of a research project at Columbia University. Following the Sputnik shock of 1957, the US government not only renewed its commitment to funding scientific research, but also rolled out new institutional innovations to support the broader innovation ecosystem.

Specifically, the National Defense Education Act of 1958 funded enormous increases in university research capabilities and graduate schools, while the creation of the National Aeronautics and Space Administration (NASA) opened an additional channel for state-funded, university-based research. These came on top of the National Science Foundation, established in 1950 to support basic research, thus complementing the US Defense Department’s focus on applied research.

Then, in 1971, President Richard Nixon announced the War on Cancer, which massively increased the NIH budget just as the science of genetics was laying the groundwork for a new era of innovation in biotech and genetic engineering. Equally important, in 1980, the Bayh-Dole Act transferred ownership of the intellectual property generated by federally funded work to the universities where it occurred. And this coincided with the emergence of a professional venture capital industry capable of identifying and supporting inventions with the potential to produce commercially or clinically significant innovations.

Thus, government-funded university researchers and VC-backed startups forged a new path to the frontier of innovation, compensating for the retreat of the great twentieth-century corporations, whose research labs had atrophied as monopoly profits disappeared, and as stock buybacks came into wider use. While the startups demonstrate the commercial and clinical potential of inventions, established firms (acquirers) drive the scale-up process.

True, the extent to which this “division of innovative labor” can fully replace the old corporate research laboratories remains an open question, as evidenced by the slowdown in productivity growth (relative to the postwar period) following the temporary surge of the dot-com bubble in the late 1990s. Nonetheless, from WWII through the early years of the current century, US leadership in science and technology was essentially unchallenged. Japan’s own explosive postwar rise as a manufacturing power turned on innovations in industrial organization, an export-led industrial policy, and domestic deployments of lean manufacturing techniques. But it never came close to displacing US leadership at the technological frontier, and it would ultimately miss the migration of value from hardware to software.

THE INNOVATION SYSTEM DIVERGES

In biotechnology and medical science more generally, the modern innovation system continued to function through President Joe Biden’s administration. Its contributions were obvious in the extraordinary response to the COVID-19 pandemic, when advances in mRNA technology allowed for multiple, effective vaccines to be delivered faster than anyone thought possible. Yet Trump’s reluctance to take credit for Operation Warp Speed – the government program that created the incentives for rapid vaccine delivery – was a sign of what was to come.

In information technology, however, the small cohort of tech giants that emerged after the dot-com bust achieved massive cash flows, which in turn reflected their market power. In 2022, the five leading tech firms invested more than $200 billion in R&D, exceeding the federal government’s R&D expenditures by about 25%. Their central focus, of course, is AI. Generative AI, the fourth generation of attempts to build “machines that can think,” requires enormous computing resources to train models on datasets as large as the internet itself. While some studies have demonstrated the technology’s potential to raise productivity in specific applications, the promise (or threat) of “artificial general intelligence” remains a hotly contested topic.

Big Tech’s research focus remains substantially oriented toward serving its own digital platforms and business models. But in other domains, such as energy-related technologies (from batteries to nuclear fusion), the innovation pathway from government-funded university-based research through VC-backed startups proved relevant through 2024. Any “Greentech Revolution” to follow on the digital and biotech revolutions was going to be rooted in, and dependent on, the same old productive relationship between federal agencies and academia.

But in 2025, this longstanding partnership faces a new existential risk. With China moving from being a “fast follower” (imitator) to an innovation leader in its own right, Trump and Elon Musk’s assault on both pillars of the American innovation system – the state and academia – could not come at a worse time. In fact, most of their behavior is arguably indistinguishable from what a paid Chinese agent would seek.

BURN IT ALL DOWN

Those who believed that Trump’s second term would resemble his first have already been thoroughly discredited. As Michael Lewis documented in his 2018 book, The Fifth Risk, first-term Trump was wholly unprepared for the job. But this time, he entered office with a cadre of loyalists fully committed to implementing Project 2025’s 922-page Mandate for Leadership, a detailed roadmap for gutting the American state.

The purported objectivity and rigor of Project 2025 is undercut by two animating missions that are expressed repeatedly and with an emotional valence that contradicts the veneer of dispassionate analysis. The first imperative is to eradicate DEI (diversity, equity, and inclusion) initiatives root and branch wherever they can be found; the second is to eliminate any and all institutional and programmatic references and responses to climate change. One of the many ironies of the second Trump administration is that a multi-pronged attack on institutions in the name of “free speech” begins by restricting and punishing expression.

As Trump 2.0 has gotten under way, implementing Project 2025’s agenda has been accompanied by some of the incoherent flailing that we all remember from Trump 1.0. The most obvious example is Musk’s so-called Department of Government Efficiency (DOGE), which recently fired, and then tried to rehire, key personnel at the National Nuclear Security Administration.

This move directly contradicted Project 2025’s program, according to which “[t]he U.S. nuclear arsenal needs to be updated and reinvigorated if we are to be able to deal effectively with threats from China, Russia, and other adversaries. As a semi-autonomous agency, the NNSA has the primary responsibility for researching and designing new nuclear warheads and for ensuring that the existing nuclear arsenal is still potent. These efforts require significant funding and scientific knowhow.”

Trump and Musk’s assault on the innovation economy has several overlapping components. The administration’s plan to close the Department of Energy’s climate-response programs is fully in keeping with the Project 2025 roadmap. But such efforts have been only haphazardly complemented by DOGE’s ad hoc attacks on departments and agencies, most of which are targeted on various “anti-woke” grounds. Many of the recent mass firings and contract cancellations have been executed by inexperienced operatives who conducted no evaluation of the consequences.

THE WAR ON MEDICAL SCIENCE AND PUBLIC HEALTH

The most immediately destructive element is the attack on the NIH’s role in supporting medical science. Rage at the Biden administration’s fumbling efforts to fashion effective responses to the pandemic has been transformed into a broad cancellation of research on infectious diseases. Here, too, those responsible for US policy have no idea what they are doing, as demonstrated by the targeting of grant proposals containing taboo words like “diversity,” even where the diversity in question is that of the genomes that drive evolution.

Meanwhile, Secretary of Health and Human Services Robert F. Kennedy, Jr., is defunding all mRNA-related research, confirming that he remains wedded to his anti-vaccine conspiracy theories even as a measles outbreak threatens the country’s children, already claiming the life of a six-year-old girl in Texas.

The new NIH leadership’s very first move was to reduce payments for indirect costs to 15% of all research grants, thus eliminating institutions’ option of negotiating indirect cost recoveries (ICR). But any life-sciences research project depends on critical, expensive technologies such as gene-sequencing machines and cryogenic electron microscopes, and these infrastructures are quite different from the “administrative overhead” that the administration is supposedly targeting.

In a new working paper, three leading scholars of the economics of innovation offer a detailed analysis of the NIH’s ICR policies over time. They show that actual effective rates of ICR are substantially lower than nominal negotiated rates (36% versus 54%) – owing to various caps and exclusions from direct costs – and that they have not risen over time. The authors then consider the implications of the new administration’s policy to cap ICR at 15%. While the NIH itself estimates that payments to research universities would fall by $4 billion, they add that:

“… the institutions that would experience the largest reductions in total research revenues with a 15% rate are those linked to the most commercially valuable private sector patents. Finally, all of the institutions directly responsible for patents on multiple drugs since 2005 have (effective and negotiated) indirect cost rates at least twice as high as 15%, and would experience large funding declines with the proposed change. While we cannot estimate the causal impact of a 15% rate, the data suggest the incidence would be broad geographically and by institution type, but would disproportionately fall on institutions with the most links to private sector innovation and drug development.”

The attack on the research institutions has now been matched by massive layoffs at their federal partners: NIH, the FDA, and the Centers for Disease Control.

THE END OF EXCELLENCE

Yet another front in the administration’s attack is the selective targeting of specific universities, most notably Columbia University. The Trump administration recently canceled $400 million of federally funded research grants without any due process, then told Columbia that it may “negotiate” a restoration of funds, provided that it first take steps to address antisemitism on campus. Columbia has chosen to acquiesce rather than fight.

In a recent commentary about the “government’s attack on Columbia,” Princeton University’s president, Christopher Eisgruber, underscored the link between academic freedom and the role of research universities in the American innovation system since the start of the twentieth century: “The government’s successful collaboration with American universities depended on its respect for academic freedom, which, for decades, presidents and legislators from both political parties largely observed. That freedom attracted the world’s finest scholars and facilitated the unfettered pursuit of knowledge.”

But this arrangement came with “a huge risk,” Eisgruber added, because if the “government ever repudiated the principle of academic freedom, it could bully universities by threatening to withdraw funding unless they changed their curricula, research programs, and personnel decisions.” And now it has happened: “The attack on Columbia is a radical threat to scholarly excellence and to America’s leadership in research.”

Already, many American universities are resorting to hiring freezes and reductions in graduate admissions. Moreover, foreign students have every reason to worry that if they visit home, they may not be allowed back into the US. In this context, it is worth noting that around 57% of STEM (science, technology, engineering, and mathematics) postdocs in the US are foreign nationals: in this environment, it is easy to imagine a sharp decline in demand for American academic and research jobs.

A measure of the impact of this program of unfettered disruption was recently reported by Nature: “More than 1,200 scientists who responded to a Nature poll – three-quarters of the total respondents – are considering leaving the United States following the disruptions prompted by Trump. Europe and Canada were among the top choices for relocation.” Inducing the talent that drives the innovation system to vote with their feet is a critical step toward liquidating the system itself.

A CONSPIRACY OF CULTS

Trying to “sane-wash” the Trump/Musk agenda is a fool’s errand. Rather than a coherent strategy, what we find is a diverse set of players, more or less poorly aligned with one another but all seeking to take advantage of the opportunity created by Trump’s second presidency. All are leveraging Trump’s cult status and his MAGA (“Make America Great Again”) base, regardless of whether they are true believers.

Whereas true believers define themselves by their utter devotion to the charismatic leader, up to and including participation in or support for the violent insurrection at the US Capitol on January 6, 2021, the most conventional grouping, comprising of mainstream business and financial leaders, simply want tax cuts and deregulation. Dutifully following the president’s will and whims is worth the price.

Then there are the economic nationalists, represented by Peter Navarro, who align with Trump’s seemingly instinctive protectionist yearnings. In this case, their program of tariffs runs directly against the interests of the corporate leaders.

More cult-like are the Silicon Valley “tech bros,” enraged by any assertion of state authority that would constrain the unbounded development and deployment of their innovations: Marc Andreessen’s “Techno-Optimist Manifesto” is an exemplary statement of the creed. A subset leads the crypto gang, whose culture of grifting – shared by Trump and his family – does not present a clear and present danger in and of itself, given the utter lack of economically significant crypto applications. The real threat would be if they gained Trump’s backing for integrating crypto, with its unregulated fragilities, into the conventional financial system.

At the most extreme fringe, the tech bros imagine that the achievement of artificial general intelligence would render scientists redundant. With the instantaneous integration of discovery, invention, and deployment, the entire innovation system would become obsolete.

Musk, the world’s richest person, is a cult unto himself whose propensity for massive, unprogrammed cuts in expenses and personnel – first tear it down, then see what, if anything, needs to be rebuilt – animates his leadership of DOGE. But Musk’s focus on gaining access to the reservoirs of personal data at the Treasury and the Social Security Administration suggests a more narrow and self-interested agenda. These unique datasets can be mobilized to train his own, lagging large language model, known as Grok, in the competition for leadership in AI.

Trump’s single most irresponsible cabinet member, Kennedy, is, like Musk, revered by his followers. But given the epidemiology of the infectious diseases to which Kennedy has given free rein, the cost in lives in the US and abroad may mount with sufficient speed to discredit him while Trump is still in office.

What all these disparate groups share is the libertarian impulse than animates Project 2025. Since the establishment of the FDA and the Federal Trade Commission more than a century ago, Congress has entrusted regulatory authority to an array of independent agencies, deliberately removed from direct presidential control. The desire to escape regulatory oversight – to “deconstruct the administrative state,” as the nationalist Steve Bannon has put it – jibes with Trump’s desire to expand the president’s authority to include control of congressionally mandated independent agencies. These may include the institutions on which the financial system’s integrity and stability depend: not only the Securities and Exchange Commission, but also the Federal Reserve.

As for Trump himself, his campaign of vengeance against individuals and institutions who have opposed him or sought to bring him to justice could at any moment compromise or even contradict the interests of those who helped bring him to power. In this environment, there can be no ideological consistency, such as that reflected in Project 2025, or any continuous strategic program. Each group’s members know that they must protect themselves as best they can. For the time being, the concept of a “public good” is gone and with it the world from which the US innovation system evolved. When all are playing zero-sum games, the aggregate outcome can only be negative.

MAKING CHINA GREAT AGAIN

As if nothing had changed in the American innovation system, on March 26 the White House published a letter (signed by Trump) to Michael Kratsios, the new director of the White House Office of Science and Technology Policy, celebrating the achievements of the American innovation system:

“Scientific progress and technological innovation were the twin engines that powered the American century. … The triumphs of the last century did not happen by chance. As World War II drew towards a close, President Franklin D. Roosevelt wrote a letter like this one to his science and technology advisor, Vannevar Bush, charging him to explore new frontiers of the mind for the sake of national greatness and pioneer science in peacetime. Dr. Bush’s response laid the groundwork for the uniquely successful American partnership of Government, industry, and academia that built the greatest and most productive nation in human history.”

And yet, the administration’s reign of disruption is crippling research universities’ ability to serve as productive partners in innovation, thus threatening the very system that Trump purports to celebrate.

Like all follower countries, China has done everything it can to appropriate the leading power’s IP. Yet China’s most strategically significant appropriation has been the American model of innovation based on public-private collaboration, mediated by research universities. The final irony is that the Trump administration is rapidly clearing the path for America’s successor to become the new global leader at the frontier of science and technology. The Chinese will be forever grateful.