Governance by Shakedown
In late February, a routine contract dispute between the Pentagon and the artificial intelligence (AI) company Anthropic escalated into something far more extraordinary. For months, Anthropic and the Department of Defense had been negotiating the terms under which the Pentagon could use the company’s large language model, including safeguards barring domestic mass surveillance and fully autonomous weapons. When Anthropic refused to relax those restrictions, the Trump administration responded with sweeping coercive measures: President Trump directed federal agencies to phase out Anthropic technology, and Defense Secretary Pete Hegseth designated the company a national security “supply-chain risk,” effectively excluding it from large portions of the defense contracting ecosystem.
The message was unmistakable: Accept the government’s preferred terms, or face a reputationally damaging designation and exclusion from business with the federal government. Anthropic sued, and a federal district judge granted a preliminary injunction, writing that “[n]othing in the governing statute supports the Orwellian notion that an American company may be branded a potential adversary and saboteur of the U.S. for expressing disagreement with the government.” Yet months later, the case remains in litigation, and Anthropic remains excluded from the list of AI firms approved to deploy their capabilities on the Defense Department’s classified networks.
The episode is unusual in its details, but it is only the most recent in a growing number of similar—and normatively troubling—cases. Across both domestic and international arenas, the administration has deployed discretionary executive authority aggressively as leverage against a wide range of targets, threatening or imposing harm as an instrument of coercive bargaining, and seeking changes in the target’s behavior—often in the form of a purportedly voluntary agreement, settlement, or “deal” acceding to the administration’s demands in return for relief from executive action.
Universities have faced threats to grants and tax status unless they accepted intrusive governance reforms. Major law firms have confronted executive orders tied to their representation choices and internal policies. Broadcasters are warned of regulatory retaliation unless they curb disfavored speech. Foreign governments face punitive tariffs unless they “voluntarily” consent to dramatic U.S. tariff increases in bilateral deals. And most famously, in 2019 Trump linked congressionally appropriated military aid for Ukraine to President Volodymyr Zelenskyy’s public announcement of politically useful investigations into Trump’s rivals. These episodes vary dramatically in substance, legality, and outcome. Yet they share a common structure.
Commentators have increasingly reached for mafia metaphors to describe this governing style. The Economist has described the president’s foreign policy as a “mafia-like struggle for global power”; Martin Wolf has characterized his trade and monetary strategy as “a protection racket”; and Anne Applebaum, surveying the administration’s tactics, observed: “This is not how a representative government works. It is how the mafia works.”
The metaphors capture something real. But “shakedown” deserves to be treated as more than invective. It names an analytic category: a distinct and recurring mode of executive rule, with a specific anatomy that can be identified, tracked, and studied across policy domains that are usually analyzed in isolation. Call it governance by shakedown.
The Playbook
Governance by shakedown has three defining elements.
First, the executive invokes a nominal legal or policy predicate—antisemitism, discrimination under diversity and inclusion guidelines, national security, trade reciprocity, supply chain risk—and uses it to activate a discretionary lever of state power against a specific target. Such levers include control over federal funding, contracting eligibility, regulatory enforcement, licensing authority, security clearances, prosecutorial discretion, and tariffs. The predicate may be real, exaggerated, pretextual, or fabricated; what matters is that it supplies the ostensible justification for executive action.
Second, that lever is used to impose, or credibly threaten, serious harm on the target: freezing grants, terminating contracts, excluding a firm from procurement markets, opening investigations, revoking licenses or clearances, imposing tariffs, suspending aid. In some cases the harm is imposed immediately; in others, the credible threat of its imposition is sufficient. The executive branch’s institutional capacity to impose substantial costs quickly, and often unilaterally, is what makes the threat bite.
Third, the threatened actor is offered contingent relief if it grants concessions sought by the administration. Those concessions may involve policy changes, public statements, contractual commitments, financial transfers, governance reforms, or withdrawal from disfavored activity. The resulting arrangement is typically framed as a voluntary settlement or a mutually beneficial “deal,” even though the concessions would never have occurred absent the preceding coercive pressure. Such bargains exemplify what legal scholar David Pozen has described as “regulation by deal.” The concept of governance by shakedown, in turn, captures the coercive process through which the executive seeks such deals—including cases in which the target resists and no bargain is ultimately concluded.
The three elements thus form a common sequence: A nominal predicate activates a discretionary lever; the lever imposes or threatens targeted harm; and relief is offered in exchange for concessions—sometimes producing a deal, sometimes provoking resistance instead.
What Governance by Shakedown Is Not
Three boundaries matter. First, governance by shakedown is not ordinary policymaking, in which the executive uses delegated authority to pursue broadly applicable objectives. Presidents routinely reinterpret statutes, reverse regulations, and shift enforcement priorities in ways that affect whole sectors. Whatever one thinks of such decisions, they operate generally rather than against a specific actor, and they are not paired with individualized demands for concessions.
Nor is it ordinary law enforcement—the administration’s own preferred description of many of these actions. Enforcement, like a shakedown, targets specific actors and often ends in settlement. But three features distinguish them. Ordinary enforcement generally adjudicates before it punishes: Investigation, findings, and an opportunity to respond typically precede the sanction. The shakedown inverts the sequence, imposing harm first and negotiating afterward. Ordinary enforcement selects targets by evidence of violation, applied evenhandedly, whereas a shakedown selects targets strategically—often by perceived enmity, vulnerability, or their value as examples. Ordinary enforcement, finally, tethers the remedy to the violation; in a shakedown, by contrast, the demands are disproportionate to and have at best a weak nexus to the predicate, serving less to remedy the alleged violation than to secure concessions aligned with the administration’s broader political priorities. Hence, the wider the gap between the stated predicate and the demanded concession, the more clearly an episode has moved from ordinary law enforcement into shakedown.
Nor, for that matter, are the administration’s actions simple retaliation, in which the machinery of government is directed against adversaries with no pathway to escape. The objective in retaliation is punitive rather than transactional. A retaliatory act says, in effect, “You will suffer because you opposed us.” A shakedown says, “You can avoid or end that suffering if you give us what we want.” Governance by shakedown occupies the middle ground: Like retaliation, it targets particular actors; unlike retaliation, it contains an offer. And it differs from ordinary hardball bargaining—a budget veto, a hard-nosed trade negotiation—because the lever is deployed outside its expected institutional purpose, converting authority granted for one end into leverage for unrelated concessions.
The Pattern, Case by Case
Even within this restrictive definition, the Trump administration’s many shakedowns, in both domestic and foreign affairs, are easy to identify. In addition to the Anthropic case above, consider the following cases.
Ukraine, 2019
The canonical case. The administration abruptly withheld nearly $400 million in congressionally appropriated military aid to a country at war and dependent on U.S. support. In the July 25, 2019, phone call, Trump asked Zelenskyy to “do us a favor though” by opening investigations into then Vice President Biden and the 2016 election. Ukraine’s leadership, highly vulnerable, moved toward compliance—drafting statements, planning a CNN announcement—until a whistleblower complaint and congressional investigation exposed the scheme. The aid was released, and the announcement never came. The House impeached the president, the Senate acquitted him, and the Government Accountability Office (GAO) concluded that the aid hold violated the Impoundment Control Act. In this case—and exceptionally—congressional guardrails kicked in just in time to prevent a successful shakedown; in Trump’s second term, that avenue has thus far failed as a meaningful constraint on the administration.
The Law Firms
In the opening months of the second term, the administration turned the same governance playbook inward. Executive orders targeted five elite firms—Perkins Coie, Paul Weiss, Jenner & Block, WilmerHale, and Susman Godfrey—suspending their security clearances, restricting building access, and directing reviews of the firms’ and their clients’ federal contracts. Four firms sued, won rapid interim relief within days, and ultimately won permanent injunctions from four different district judges, who converged on the central point: The executive orders unconstitutionally punished protected advocacy, association, and representation. (The consolidated appeals are now awaiting judgment in the U.S. Court of Appeals for the District of Columbia Circuit.) Paul Weiss took the other path, folding, in the words of Benjamin Wittes, “like a cheap suit.” Its chairman, describing the order as an “existential risk” to the firm, negotiated directly with the president and settled within a week, pledging $40 million in approved pro bono work. Eight more major firms then settled preemptively—without ever being formally targeted—bringing the administration’s total haul to nearly a billion dollars in committed legal services, plus commitments on hiring, DEI, and client representation.
The Universities
The campaign against higher education followed the same template at a greater scale. Invoking civil rights enforcement—including Title VI antisemitism claims, Title IX athletics disputes, and DEI investigations—the administration froze or threatened billions in research funding at eight elite institutions and sent warning letters across the sector. The demands reached far beyond the predicates: governance restructuring, admissions audits, monitoring, and payments. Columbia settled, paying roughly $221 million and agreeing to sweeping conditions and federal oversight. Brown, Cornell, Northwestern, and Penn reached their own agreements to restore funding. Harvard refused, sued, and won summary judgment from a district court that called the administration’s charges a “smoke screen”; the University of California, Los Angeles (UCLA) also resisted, and University of California faculty and unions successfully challenged in court the administration’s efforts to pressure the university into a proposed $1.2 billion settlement. Then, in October 2025, the White House generalized the model, offering nine universities a “Compact for Academic Excellence in Higher Education”: preferential federal funding in exchange for ideological and governance commitments—episodic coercion upgraded to standing conditionality.
The Broadcasters
The Federal Communications Commission’s chairman deployed licensing, enforcement, and merger-review leverage—along with public threats—against broadcasters airing disfavored speech. The clearest episode was the pressure on Disney and ABC over Jimmy Kimmel, whose show was suspended and then reinstated after intense public and commercial backlash. The case shows both that the shakedown can operate through informal regulatory threats rather than formal orders, and that targets can reverse course when audience pressure outweighs regulatory fear.
The Tariffs
Returning to the foreign policy realm, the administration scaled the playbook to the world. Invoking emergency and national-security authorities, it imposed a 10 percent global baseline tariff and country-specific rates reaching 20 percent for the European Union, 34 percent for China, and 46 percent for Vietnam—then offered relief through bespoke bilateral “deals” in which target countries would “voluntarily” accept higher U.S. tariffs in violation of World Trade Organization rules, one-sided market opening, and large purchase and investment pledges. Eighteen such deals were concluded before the Supreme Court invalidated the underlying tariff authority in Learning Resources v. Trump, threatening to bring yet another international shakedown to an end. The administration’s response, however, was not to unwind the deals but to reconstruct the leverage, pivoting to Section 122, 301, and other statutory authorities while insisting that partners honor agreements negotiated under duress. So far, those countries, faced with potentially massive new tariffs on pretexts such as forced labor, have indeed done so. In the administration’s terms, the shakedown has been tumultuous but highly effective and surprisingly durable—although whether it benefits the American people remains a separate question.
The Fed
The Federal Reserve constitutes an edge case—a variant with no explicit settlement on offer. In this case, the administration deployed Justice Department investigations and removal pressure directed at Governor Lisa Cook, and sustained public and legal pressure on then-Chair Jerome Powell. Here the demanded concession is not a signed deal or settlement but conformity to the president’s wishes—an implicit shakedown of an institution whose independence is the point of attack. In April, the Justice Department ended its investigation of Powell in order to unlock the confirmation of his successor Kevin Warsh, and the Supreme Court in June blocked the administration’s initial effort to remove Cook. In early August, however, the administration renewed its effort to fire Cook, making the shakedown—or at least intimidation—of the Fed an ongoing project.
Illegal, Yet Effective
Illegality is not part of the definition of governance by shakedown—some uses of executive leverage may be lawful though normatively troubling—but it is a recurring empirical feature of the cases. The GAO found the Ukraine aid pause unlawful. Four district courts held the law firm orders unconstitutional. Harvard and UCLA won significant judicial relief. The Supreme Court struck down the administration’s tariffs under the International Emergency Economic Powers Act and Trump’s initial effort to fire Cook from the Fed. Again and again, the administration has asserted executive authority in ways that federal courts ultimately rejected.
And yet the legal weakness of Trump’s executive actions did not make them ineffective. That is the puzzle at the heart of the phenomenon—and the key to understanding it.
Fast Pain, Slow Guardrails
The administration’s leverage is immediate: Aid can be frozen, grants canceled, security clearances suspended, clients frightened, tariffs imposed, merger plans threatened almost at once. Institutional guardrails, by contrast, are slower and uncertain. Congress may investigate, markets or publics may react, and legal rulings may eventually vindicate the target—but those protections often arrive only after the target has already suffered harm or faced a credible threat of escalation.
This temporal asymmetry explains why legally weak, even blatantly unconstitutional, measures can be effective instruments of coercion: The target must decide whether to accommodate now or gamble on relief later. Temporary restraining orders in the law firm and university cases mattered precisely because they shortened that gap; the months-long delay before the Supreme Court’s tariff ruling mattered because it did not. Governance by shakedown exploits not only executive discretion, but time: The administration can impose pain quickly, while constitutional guardrails work slowly.
Does It Work?
The record is mixed—and that mix is itself the finding. The administration repeatedly failed to secure universal compliance and frequently lost in court: Ukraine never announced the investigations; four firms and two universities resisted and won; Kimmel returned to the air.
But failure on those measures does not mean the shakedowns failed. The campaigns extracted nearly a billion dollars from the legal profession, hundreds of millions of dollars and structural concessions from universities, and “agreement” to asymmetric trade terms from much of the world. They also produced highly visible victories that Trump and the White House eagerly claimed: a “massive” trade deal with the EU, a “historic” settlement with Columbia, “Great News for America” when ABC suspended Kimmel, and, in the White House’s words, Big Law continuing to “bend the knee to President Trump.”
The most important effects, however, may lie beyond the named targets, in the diffuse, difficult-to-measure caution of every institution not yet targeted but keenly aware of the costs of crossing the administration. Governance by shakedown can lose in court, fail to obtain some ultimate objectives, and still succeed in extracting concessions, reshaping behavior, and spreading fear.
Recognizing the Shakedown Playbook
That is why the greatest danger lies not in any individual abuse, but in the possibility that the tactic becomes legible, repeatable, and normalized. Once targets understand that executive power may be used selectively against them, and once others observe the costs of resistance, the practice disciplines behavior well beyond the cases in which sanctions are actually imposed. Governance by shakedown thus operates as both coercion and demonstration: It extracts concessions from immediate targets while teaching everyone else to anticipate and avoid presidential retaliation.
Kim Lane Scheppele, writing about “autocratic legalism,” argued that the first task in confronting contemporary threats to constitutional democracy is to “stare into the face of the new autocracy to track in detail how it works,” because “we need to stop taking for granted that constitutions can defend themselves.” Recognizing the shakedown playbook—the pretextual predicate, the discretionary lever, the targeted harm, the proposed deal—is that first task. The pattern is visible across law firms, universities, broadcasters, tech companies, central bankers, and foreign governments. Making that pattern legible is the beginning of understanding how to recognize it, when it succeeds, when it fails, and how the gap between fast pain and slow guardrails might be closed.
