RUSSELL MUIRHEAD is the Robert Clements Professor of Democracy and Politics at Dartmouth College.
NANCY L. ROSENBLUM is the Senator Joseph Clark Professor of Ethics in Politics and Government Emerita at Harvard University.
They are the authors of Ungoverning: The Attack on the Administrative State and the Politics of Chaos.
As U.S. President Donald Trump returns to the White House, it already seems clear that his second term will look little like his first. Many of Trump’s first-term appointments distanced themselves from his views and even denounced him. “I picked some people I shouldn’t have picked,” Trump lamented on Joe Rogan’s podcast last November. “Disloyal people.” This time, Trump is ready. The president and his allies are convinced that he was let down in his first term by those around him and by the bureaucracy of the federal government. They will not let that happen again.
On his first day in office, Trump restored Schedule F, an employment category that he had created in the closing months of his first term that strips civil servants of protections, allowing them to be fired at will. Trump has convinced many of his followers that a “deep state” thwarted his first term and robbed him of the 2020 election. Now, the putative deep state will be expunged, along with the expertise and procedures that make effective administration possible. Trump will judge both appointees and civil servants by one criterion: loyalty, defined not by commitment to a programmatic agenda but by unquestioning obedience to the president.
According to reporting by The New York Times, the administration’s transition team has been asking applicants for government posts in multiple agencies, including in the intelligence services and the Department of Defense, about their views on the events of the January 6 attacks on the U.S. Capitol. And they have asked applicants pointedly about the 2020 presidential election. Those who condemn the attack or believe that Biden won the race appear unlikely to receive posts.
Staffing the federal government with sycophants is not about ensuring loyalty to an agenda. It is about ensuring submission to the president. And it serves to amplify what we call “ungoverning”: the degradation of state capacity and the substitution of unchecked personal will for the difficult, necessary business of shaping, implementing, and assessing policy for the nation. The administration will sideline experts and circumvent regular processes of information gathering and consultation. In so doing, it will degrade state capacity; the premium Trump places on personal loyalty will result in confounding his ability to govern.
Ungoverning is radical and rare in the annals of political history. There are simply not that many examples of states that have been systematically degraded and dismantled by individual rulers or parties. In some cases, those in power manage to substitute kleptocracy and state violence for administration; for instance, in Venezuela, where first Hugo Chávez and then Nicolás Maduro destroyed a prosperous and functioning state.
In the United States, ungoverning—which is distinct from conservative small government and from deregulation and privatization—is novel. Critics on both sides of the aisle have tried to cut through bureaucratic red tape in the past, but historically, officials of both parties and the general public have accepted the administrative state as indispensable to the fulfillment of public needs.
The question now is whether those around Trump will work to check his appetite for ungoverning. As subject-specific experts are replaced by flatterers, the government’s ability to achieve lasting, large-scale results shrinks. As process is shunted aside in favor of one person’s will, the state’s ability to gather accurate information and make effective judgments corrodes; its capacity to design, refine, and implement policy disappears. Ultimately, ungoverning makes the strongman weak.
At the heart of democratic governance—and the office of the presidency—is the idea of administration. The institutions and scope of the current administrative state can be traced to the New Deal era of President Franklin Roosevelt. But the importance of building a state with administrative capacity goes back to the country’s founding. When opponents of the Constitution insisted that state governments would be closer to the people—and more loved by them—than any government operating on a national scale, Alexander Hamilton, the first secretary of the treasury, responded that the aim of the government was not to reflect the passions of local majorities but to practice good administration.
A state with administrative capacity pursues its goals by first systematically and accurately gathering information on current conditions—as Hamilton tried to do on the national economy in his 1791 “Report on Manufactures” and as the Bureau of Labor Statistics and every other agency does now. The next step is to envision a future that is more prosperous, freer, and more secure. And, finally, to design a plan for applying resources, on a large scale and over a long period, to bring that future into being. The administrative state shapes, implements, and enforces every law passed by Congress and every executive order made by the president. As Hamilton wrote in Federalist, no. 72 (of The Federalist Papers), “the administration of government, in the largest sense, comprehends all the operations of the body politic.”
The machinery of government that Trump now inherits comprises about 4,000 political appointees and three million public servants who respond to natural disasters and military emergencies, pursue long-term goals set by Congress and the president, and do much else. For Trump, the problem with expertise and process is that they limit the scope and constrain the exercise of his power. Specialized knowledge is a source of authority, as Trump knows, which is why he cannot bear to stand alongside a knowledgeable official, even one of his own appointees, without asserting his own superior expertise. At a press briefing during the COVID-19 pandemic, he suggested to Deborah Birx, the White House coronavirus response coordinator, that irradiating COVID-19 patients’ bodies with ultraviolet light and injections of disinfectant might offer a cure. “I’m not a doctor,” the president said, pointing to his head. “But I’m a person that has a good you-know-what.”
Trump expects that selecting for loyalty will allow him to govern more effectively in the second term, to push through his agenda. But in truth, prioritizing loyalty over capability will undermine his administration. Unless he is able to yield to the expertise of his appointees and civil servants, his own policy aims will be frustrated. Trump’s problem is not that he requires loyalty to his agenda, on tariffs, on immigration, on foreign policy. It is that he demands personal loyalty—or what John Bolton, Trump’s longest-serving national security adviser in his first term, has called “fealty, a medieval concept implying not mere loyalty but submission.”
To be sure, some of Trump’s cabinet picks are loyal primarily to his program. Take his new border czar, Tom Homan, a former border patrol agent who served as head of the Immigration and Customs Enforcement deportation branch under President Barack Obama and acting director of the agency under Trump. “I’ve seen hundreds of policies come and go,” he told The Wall Street Journal earlier this month. “I know what policies worked and what policies don’t work.” He is devoted to Trump’s immigration policies, but he has also met with Republican members of Congress to moderate expectations about the practicality of mass deportation. For those who support Trump’s policies, this is the kind of appointment that promises to convert Trump’s rhetoric into reality. For those who oppose Trump’s policies, this is what makes Trump’s second term more foreboding than the first.
An effective president has to find and empower people in the mold of Homan—professionals with deep expertise capable of translating broad goals into workable policies and implementing them. This is what “administration” means. But Trump’s insistence on personal submission, and his distrust of the very administrative state he is charged with directing, is decidedly anti-administration. As Russell Vought, the prospective director of the Office of Management and Budget, has said: “We want the bureaucrats to be traumatically affected. When they wake up in the morning, we want them to not want to go to work because they are increasingly viewed as the villains.” This is also the orientation of Kash Patel, Trump’s nominee to head the FBI who last September promised to close the agency’s Washington headquarters and convert it into a museum of the “deep state.”
Then there’s the so-called Department of Government Efficiency, or DOGE—helmed by the political newcomer Elon Musk. The electric car and spaceship mogul originally promised to cut at least $2 trillion from a federal budget of nearly $7 trillion, to lay off two million federal employees, and to slash thousands of regulations. To shrink the budget, civil service, and regulatory scope of government on this scale is an enormous task, and DOGE has recruited leaders from Silicon Valley to develop the plan. Staffed mostly by individuals with little or no experience in governance, DOGE will be unlikely to realize the massive restriction in the scope of the federal government that its founders promise. To do so would require both intimate knowledge of the federal budget, the work of federal agencies, and buy-in from a working coalition of legislators, not to mention the backing of the American public that may not be ready to accept a government that does vastly less than people are accustomed to.
Ungoverning makes the strongman weak.
If Trump is to succeed in realizing the policies he espouses, his administration will need to make appointments on the basis of qualities more salient to effective governance than personal loyalty. Effective governance requires empowering public servants with experience and specialized knowledge. It means allowing for a process by which they can form and select from alternative policy options, detail plans for their implementation, and then act on those plans. It means staying true to clear goals so that appointees can advance the agenda without squandering time and talent kowtowing to the president.
Those who oppose Trump’s policies may welcome a chaotic anti-administration that cannot make good on the promise of mass deportation or across-the-board tariffs. But Trump will not oversee an administration that is simply ineffective. Rather he will orchestrate a comprehensive degradation of the administrative institutions of government—the processes of decision-making and the consultation of expertise that effective policy of any sort requires. Instead of a government capable of policymaking and implementation, the U.S. government will revolve around the personal will of the president.
Governing is not about domination and submission. As Hamilton observed, it is about undertaking “extensive and arduous enterprises for the public benefit, requiring considerable time to mature.” The Biden administration aspired to this, as exemplified by the Infrastructure Investment and Jobs Act and the CHIPS and Science Act, both of which are massive, long-term projects that will take a generation to reveal their full effects. If Trump harbors similar or grander aspirations, his administration will likewise need to act like an administration.
There is a danger for Trump, and for the country, beyond the convulsions of arbitrariness, the dismissal of knowledge and process, and the specter of draconian measures. The president, his party, his followers, and friendly media spent close to a decade denouncing the conspiratorial “deep state” and threatening to bring it crashing down. The result has been a wholesale delegitimation of the institutions that underpin the government’s administrative capacity. Delegitimation goes beyond distrust, a measure of which is always warranted, as vigilance is a duty of citizens in a democracy. For Trump and many of his allies and supporters, the machinery of government has no legitimacy—and compliance with its strictures is unnecessary unless they happen to align with the leader’s will.
During Trump’s first term and throughout the Biden administration, right-wing resistance to Democratic officials and policies, sometimes in the form of violence and threats, was frighteningly commonplace. Targets included election officials, local school boards, health-care workers, and teachers. In October 2024, reports of armed individuals threatening Federal Emergency Management Agency personnel delivering aid to hurricane victims in North Carolina forced the agency to relocate assistance teams and pause some operations.
Citizens who lament Trump’s return to the presidency are not likely to respond with violence as some of Trump’s followers did on January 6, 2021. Organized civil society will look to the regular political arena and to the courts; it will abide by the norms of a peaceful democratic opposition. But Trump’s followers are another story. The blanket clemency Trump awarded on the first day of his presidency to all the January 6 insurgents, including the leaders of the Proud Boys and the Oath Keepers militia groups, signals that violence against Trump’s opponents is acceptable.
Even if political violence does not break out, the continuous disruptions of ungoverning create unpredictability and uncertainty and, for many Americans, insecurity and dread. The effects of ungoverning are institutional but also personal and individual. This matters because vulnerability is infantilizing, even paralyzing. It taxes both personal and collective agency and degrades the vital moral underpinnings of liberal democracy. Trump’s administration will come to an end, eventually, as will his “Make America Great Again” movement. Afterward, the American people will have to rebuild what he destroyed: a government able to administer the country in the interests of its people.