In his second term, Trump is pressing on. The US Forest Service will be relocated to Utah amidst a downsizing of research talent.

The US Department of Agriculture (USDA) recently announced that the Food and Nutrition Service (FNS) would reorganize its 16 nutrition assistance programs into five regional hubs, to be located in Dallas, Denver, Indianapolis, Kansas City, and Raleigh. This represents a relocation of most of its 1,200 employees from its current headquarters in Alexandria, VA and its seven regional offices.

The Trump Administration has framed relocations like this as a way to bring federal services closer to the people they serve. Critics say that it is an effort to induce more resignations, move federal jobs to red states, and closer to conservative stakeholders who can more easily influence federal employees.

In the case of FNS, what is true? Will these new offices be closer to people receiving SNAP benefits? SNAP is the largest of FNS’s 16 nutrition assistance programs and the one most directly tied to low-income household eligibility, making it a reasonable proxy for where FNS services are most needed.

So, let’s take a look at the data.

FNS’s Current and Future Office Locations

The current FNS structure includes seven regional offices plus the headquarters in Alexandria, VA, which together are located in counties with approximately 3 million households, about 437,000 of whom receive SNAP benefits—a recipiency rate of 14.2 percent. (Data for Robbinsville, NJ, the seventh current office location, were unavailable, so both figures are slight undercounts. Technical Notes are below that explain the data used here.)1

Under the reorganization, those seven offices consolidate into five new hubs plus headquarters. The cities in that new configuration collectively serve about 1.8 million households, of whom approximately 176,000 receive SNAP—a recipiency rate of 9.7 percent, lower than the current footprint.

Source: American Community Survey; Tom McGovern. Notes: According to the USDA, the Atlanta office will provide retailer and compliance operations.

I also merged these city-level data with 2024 presidential election results, by way of context. Most major cities lean Democratic, so the partisan skew in the table below is not surprising. Worth noting, however, is that all five of the new FNS hub cities are located in states that Donald Trump carried in 2024. Only Denver—and Arlington, VA, if included as a skeleton-crew location—sit in states Kamala Harris won.

Share

What If We Chose Differently?

Obviously, I don’t know how USDA arrived at these five cities. There may be real estate, administrative, or labor market considerations that made them preferable to alternatives.

But the question is worth asking: If the goal is to locate offices closer to SNAP recipients, what would a different set of cities look like?

If we optimized for the sheer number of SNAP-receiving households, we’d end up with the largest cities in the country—New York, Los Angeles, Chicago, Houston, and so on. That’s a somewhat crude measure. More interesting is what happens if we instead choose cities with the highest share of households receiving SNAP. (Alternatively, we could choose more rural areas that have high shares of SNAP recipients, but those areas tend to have smaller populations, so cities remain the more useful comparison).

Source: American Community Survey.

The table below shows the 50 largest US cities ranked by SNAP recipiency rate. The top five—Detroit, Fresno, Philadelphia, Milwaukee, and Cleveland—together serve about 1.6 million households, roughly 430,000 of whom receive SNAP benefits, a recipiency rate of 27.6 percent.

The five new FNS hub cities, highlighted in orange in the table, all rank in the bottom 20 cities by this measure. Even two of the current office cities—Boston and Chicago—reach more SNAP recipients, and at higher rates, than most of the new hubs.

Source: American Community Survey; Tom McGovern.

What the Data Say

The stated rationale for relocating FNS regional offices is to bring services closer to the people they serve. But the data tell a more complicated story. The five new hub cities—Indianapolis, Dallas, Denver, Kansas City, and Raleigh— collectively reach fewer households and a lower share of SNAP recipients than the offices they replace. Meanwhile, cities like Detroit, Fresno, Philadelphia, Milwaukee, and Cleveland—which rank among the highest in SNAP recipiency—are absent from the new map entirely.

None of this is to say the new locations are wrong. Real estate availability, workforce considerations, and operational logistics all matter in decisions like this. And proximity of an office to recipients is only one way to measure how well a federal program serves its constituents.

But if the goal is to be close to the people who rely most on nutrition assistance, the data suggest the new configuration is a step in the wrong direction. The cities with the highest need are largely not the cities getting the offices. The results reinforce suspicions that the relocation is for political reasons, fitting a pattern where Trump has shut off federal resources from blue states. Indeed, 30 of the 50 largest cities shown above are in states Trump carried in 2024, with 8 of the remaining 20 located in California. And already, three-quarters of USDA workers have said they won’t relocate, which is consistent with previous agency moves and will likely have consequences for the quality of processes and services the agency offers.

Dr. Jonathan Schwabish is an economist, author, and data communications expert, and founder of PolicyViz, whose four books and popular PolicyViz Substack newsletter help researchers, analysts, and policymakers communicate data and evidence more clearly and effectively.

The FAFSA form used to be a poster child for administrative burdens, creating a hurdle that families had to overcome if they wanted to get federal financial aid to help pay for college. Many states, as well as most colleges and universities, also use FAFSA information to determine eligibility for state and institutional financial aid.

Some 17-18 million students complete the form each year. Once you add parents and spouses who must provide information for their kids or partners, approximately 40 million people interact with FAFSA annually. So, it should be a no-brainer to make the process simple.

Instead FAFSA has historically been a barrier to entry for many students, especially for the first generation and low-income students who stand to gain the most from higher education. Ask current or former students about their FAFSA experience, and the modal response are groans when they recall the hassles involved. The fearsome reputation of FAFSA as a hurdle to overcome, rather than a helping hand on the way to college, meant students counted themselves out before trying to complete the form.

I work at the National College Attainment Network (NCAN), where most of our member organizations focus on helping first generation and low income students go to college. One of our members, uAspire estimated that simplifying FAFSA would free up one-third of the organization’s advising time. That is time they can use to give the students they serve more help, or to help more students. I used to run hours-long training sessions for college advisors, like those working at uAspire, on the basic FAFSA process, and how to avoid common mistakes, because there were so many places students would get tripped up.

NCAN’s research into why students did not complete the FAFSA showed that 15% were deterred by the length and complexity of the process. That research, using data from Trellis Strategies annual Student Financial Wellness Survey, also showed that nearly half of the students not applying for aid, thought they would not qualify. It was because of these barriers to the students our members serve, that NCAN has long pushed for a simpler, easier FAFSA.

Are you ready for some good news? The lobbying of NCAN members and student groups, and resulting policy proposals and administrative efforts actually made a difference. FAFSA is working better than ever before.

Today, students can complete their FAFSA in as little as 15 minutes. NCAN’s FAFSA Tracker, which uses Department of Education data for completions by high school to create state and national completion rates, shows we are on track for a record completion rate for the high school class of 2026.

As of May 1st, the percentage of high school seniors completing FAFSA has increased by 11.7 percentage points over 2025.

Compared to 2024, the FAFSA process is succeeding across the board, both for high schools with low and high-income students, white students as well as students of color, in rural and urban locations.

Share

Why Does FAFSA Simplification Matter?

I spent the first six years of my career working in financial aid, primarily running outreach and communications for UCLA’s financial aid office. Trying to help students and families understand our excessively complicated financial aid system, and seeing how complex the process was, is why I ended up moving to federal policy work. It was clear that things like the FAFSA, and many other higher education systems and supports, needed fixing at a macro level

You don’t have to work in financial aid for long to see all the cracks in the system, and FAFSA used to be a Grand Canyon sized crack. When you spend a lot of time helping high schoolers apply for aid, you notice all the friction points that make it harder. If students—especially first-generation students— first experience the financial aid system as burdensome, it becomes much harder to convince them that college is for them.

FAFSA Was a Mess: What Changed?

Passing laws is only the start when it comes to changing how real people experience a public policy change. Implementation matters, a great deal, and the implementation part of FAFSA simplification was so bumpy that it would make a cobblestone road look smooth. The significant improvement, compared to 2024 and 2025, is partly the story of a well-intentioned policy change resulting in an implementation failure, followed by an impressive turnaround.

During his first term, President Donald Trump signed the FUTURE Act into law in 2019 and the FAFSA Simplification Act in 2020. Implementing those laws required a lot of changes to the FAFSA form, the underlying eligibility calculations for the Pell Grant, and most importantly, massive updates to the technology infrastructure that underpins the whole federal financial aid system.

The changes mean that students now only answer 38 questions today, compared to over 100 previously. The eligibility formulas for the Pell Grant were changed, increasing the number of students that qualify for the primary federal grant program that helps low-income students pay for college. Finally, data sharing between ED and the Internal Revenue Service (IRS) was enhanced, reducing how many students have to send their tax returns into schools for verification—an audit process that was hugely burdensome for students, families, and college financial aid offices.

The redesign, and the need to completely rebuild and update all the backend technology that supports the FAFSA, led to serious delays in the form (usually open on October 1 each year) being released in 2023 and 2024, along with a host of glitches. On top of the delayed launches, students and families faced frequent system outages, students could not make updates to their information for months after initial submission, and colleges had to submit any changes for students individually, rather than by sending large batch files to Department of Education.

A reform process intended to make FAFSA easier instead generated headlines like this one from the New York Times.

Recounting every implementation challenge would require an essay of its own. The National Association of Financial Aid Administrators (NASFAA) has a handy timeline of the issues, for those who want to know more.

A highly curated list of what caused all the problems includes:

• A lack of focus on implementation from the Biden administration.

• Massive tech debt at the Department of Education’s’s Office of Federal Student Aid (FSA) that had to be fixed. Tech debt is the accumulation of choosing the fastest solutions to a problem over the best long-term solutions. Over time, those choices make it harder and harder to keep things running efficiently.

• Federal technology vendors tasked with upgrading FSA’s systems messing up.

• Congress’ unwillingness to properly fund FSA, in part because of antipathy towards the Biden administration’s attempts to forgive federal student loan debt for some borrowers.

So how did things get fixed? The Biden administration brought in outside help. Jeremy Singer, President of the College Board, was asked to lead the charge. Singer, who has significant experience implementing major IT infrastructure projects, brought in a team of colleagues, including Aaron-Lemon Strauss, who remains at FSA as the Executive Director for FAFSA, after Singer returned to the College Board.

Statecraft has an excellent podcast episode with Singer that delves deep into all the issues and how they were solved. One of the main takeaways is how important it was to have someone consistently in charge of the process. You can’t simply outsource such an important function without having the in-house capacity to make sure the product works. For me, that also shows why career staff are so vital to making government work for everyday people.

Singer, Lemon-Strauss, and the dedicated hard-working team of career federal workers deserve most of the credit for turning FAFSA around.

Building on that turnaround required FSA to maintain consistent staffing of knowledgeable experts . The consistency that the FAFSA team has provided, even as administrations switched, ensured that the work begun during the Biden administration was completed in the Trump administration.

This is one area where the Trump administration maintained state capacity, and can now take some of the credit for a major policy and tech innovation that is making life easier for millions of Americans.

How is FAFSA Simplification Helping Students?

Reducing barriers to college means more students apply and enroll. The benefits of removing FAFSA as an administrative burden are already showing up in application and enrollment data. On top of what looks like a record FAFSA completion year, Common App shows that the number of college applicants for the 2026-27 academic year is 4% higher than the same time last year.

We know that high school seniors who complete the FAFSA are 84 percent likelier to enroll in college immediately after high school. For students from the lowest income quartile, that number leaps to 127 percent likelier to head straight to college.

The data linking FAFSA completion and increased college enrollment rates are correlative rather than causal. But, the link is so strong that it is almost certain that FAFSA completion is helping students see college as a viable path after high school.

The biggest increases in college applications are coming from the students we want to help: first-generation, historically underrepresented and coming from communities with the lowest median incomes. People with fewer resources struggle more with administrative burdens. And so making the financial aid process easier for these students shows that a well-designed universal process can level the playing field and reduce inequality.

Most importantly, more students are eligible for the maximum Pell Grant than ever before, because of eligibility changes that were part of FAFSA simplification.

1.7 million more students were eligible for maximum Pell Grants in the 2025-26 FAFSA cycle, based on NCAN’s analysis of data from the office of Federal Student aid. That increase is a 27 percent jump compared to the 2023-24, the last application cycle before the FAFSA Simplification Act changes were implemented. There has also been an over 1,600% (yes, you are reading that correctly) jump in the number of students eligible for the minimum Pell Grant. These kinds of changes make it clear that removing administrative barriers makes an enormous difference in who gets support, but as importantly whether people who are eligible for support are able to ask for help.

Source: NCAN https://www.ncan.org/Web/Web/News/FAFSA-Simplification-Yielded-1.7-Million-Additional-Pell-Eligible-Students.aspx

FAFSA has come a long way from Lamar Alexander’s unwieldy paper form. There is still more to do when it comes to college access. Fixing FAFSA is only a piece of the puzzle.

Low-income students, and students of color still enroll in and complete college at lower rates than their wealthier peers, so making it easier to apply for aid should be the starting point for continued efforts to narrow the attainment gap between the students who have the most and those with the least.

What fixing the FAFSA shows, even if it took much too long, is that administrative burdens are not inevitable. Barriers can be removed or lowered, when consistent advocacy and political will lead to needed legislative changes, and smart hard working public servants are given the support to effectively implement those changes.

Subscribe now

They’re sure to be disappointed. There’s no going back to politics-as-usual, because it was dead long before Trump. As I show in my new book, The Right-Wing Idea Factory, the rise of dark money and, even more, social media put new forces in charge of American politics.

The grownups aren’t in charge, and there’s no chance that they will be any time soon. In the wild-west rodeo that defines American politics, the top job goes to whoever can stay on top of the bucking bull for at least an election cycle or two. But no one is going to stay there for long.

That’s because of two things. One is that Americans don’t trust any grownup for very long, because we quickly become convinced that they won’t deliver. The other reason rests in the new technology. Some analysts have held out hope that social media can bring people together. It’s become instead a wedge driving us farther apart.

Let’s look at each of these reasons in turn.

No president can deliver on big economic promises

At the core of most elections are big promises on the economy, as James Carville preached to Bill Clinton in the 1992 presidential election. “It’s the economy, stupid,” he said. He could add, “It’s always the economy, stupid.”

But the simple fact is that no president can truly deliver on promises about the economy. It’s become global, beyond anyone’s control, as Trump has discovered about the price of gas at the pump. The Federal Reserve is the closest thing we have to a steering wheel, but the president doesn’t rule the Fed, and the Fed itself is frank about the limits on its own power.

Promising to bring down the high price of eggs and beef and beef is tempting. So, too, is the pledge to put people to work. The president can’t really do much about that, however—except screw it up, with decisions that drive inflation up and employment down. And that’s driven trust in government on a steady downward track. From a high point following the 2001 terrorist attacks, trust in government to do what is right, always or most of the time, has slid from 60 percent to 17 percent at the end of 2025.

In February, Gallup found in February that people ranked government as the country’s biggest problem. Confidence in the presidency is embarrassing at 30 percent, but Congress is at the bottom of confidence in American institutions, at just 10 percent.

A little more than a decade ago, a poll found that Congress was less popular than a root canal, colonoscopy, head lice, and Genghis Khan. Nothing has changed since. It’s hard to see how any grownups can turn this around any time soon.

That translates into growing gloom about the future. Gallup found in February that 59 percent of Americans anticipate high-quality lives in five years, the lowest in the survey’s 20-year history.

The false promise of the social media bridge

The unifying promise of social media as a virtual town square, if it ever really existed, has been dashed.

Consider the most popular podcasts, for example. Of the top five on Spotify, three lean to the right, with The Joe Rogan Experience consistently at the top. One is news, produced by The New York Times, though many on the right would suggest it’s left-leaning. Of course, almost no one listens to the other side.

Those on the right, on the other hand, are sure they know what the left is up to: pushing their values down Americans’ throats, in ways that run against traditional American values. One group on the left has started a podcast listening club to tune in to what the right is saying, but as co-founder Nina Harris told an interviewer, “I’m really, really scared, and I just want to figure out what’s happening.” With a gulf that large, it’s little wonder that true crime podcasts are the only safe middle ground.

The sad fact is that Americans don’t much like each other. Among more than two dozen countries in a recent Pew Research Center survey, the US was the only country where a majority of people (53 percent) saw their fellow citizens as morally bad. Compare that with Canada, where 92 percent of the people saw others as morally good.

In a different Pew survey, a growing number of Republicans saw Democrats as immoral (an increase from 47 percent in 2016 to 72 percent in 2022). Democrats felt the same way about Republicans (with an increase from 35 percent to 63 percent). A majority of people who identify with each party saw those in the other party as close-minded, dishonest, and unintelligent.

Media floods Americans with choices, and they choose to go their own very different ways. The closest thing to a unifying social event is the Super Bowl but, in 2026, it only brought about a third of Americans to their televisions, and the Bad Bunny half-time show became a partisan spark point.

Religion has lost its bridging role. In 1952, 75 percent of Americans said that religion was “very important” in their lives, according to Gallup. In 2025, it was 47 percent. Over the same period, the number who said it was “not very important” grew from 5 percent to 28 percent. Now, even among those who are within the same religious denomination, there is partisan disagreement, as a Catholic vice-president entered a war of words with the first American pope.

Even families have become less important as social anchors. The marriage rate has declined, from 69 percent of those over 18 who are married to 50 percent in 2021. Pew concluded a decade ago, “As a result of these changes, there is no longer one dominant family form in the U.S.”

Instead, we increasingly choose to live with others who mirror our views and take our cues from others with whom we already agree. Most Americans live in bubbles, “with virtually no exposure to voters from the other party,” Jacob R. Brown and Ryan D. Enos found. It isn’t necessarily that people check the party registration of neighbors before they move in, but people with similar views tend to seek out similar homes and amenities.

As a result, in terms of where we live and what we listen to and what we believe, Americans are becoming increasingly segregated—by political party. That’s one of the reasons that so many incumbents hail from safe state legislative and congressional districts. Their party leaders draw boundaries to protect their members, and the clumping of the party faithful is making that easier.

Over the last dozen years, Republican state legislative chambers have tended to stay red. The same is true for the Democrats, although they’ve been locked out of more states. That’s one of the big reasons why presidential elections have increasingly boiled down to a small handful of swing states. And all of that will increase with the great redistricting war of 2026, aimed at gerrymandering congressional district boundaries.

So we can’t count on grownups to rescue us. Social media is uber-powerful, but it’s only widening the problem of polarization.

Share

Who can trust those who don’t deliver?

And then there is the problem deeply embedded in the way Republicans and Democrats both approach government itself. Democrats dream big, but they typically fail to think through how to deliver. That was the problem with the launch of Obamacare, where only six people were able to enroll on the first day. The Obama administration eventually straightened out the problems, but not before the problems of the launch became an anchor that dragged the program down.

And, convinced that government has gotten way too big, Republicans have devoted their energy to bulldozing programs and people. The Trump administration DOGEd its way through government. Elon Musk’s grand experiment slashed more than 400,000 jobs and left the government down 272,000, after federal agencies discovered they needed to refill the positions that had been eliminated.

But cutting government doesn’t deliver the things that people want government to do, from delivering weather warnings on time to making TSA checkpoints run smoothly to catching terrorists before they can act. In fact, the director of the US Office of Personnel Management, Scott Kupor, admitted that “sometimes you over-restructure,” and that the administration now needs to hire people back to fill some of the jobs it eliminated. It’s hard to rebuild on the rubble of what’s been destroyed.

In fact, Trumpism has turned the administration of government upside down. For my take on this, see Trumpism and the Future of Public Administration (Cambridge University Press, 2026), which is available as a free download from May 8 through May 20.

The one thing that progressives and conservatives share is a blind spot for actually making government work. That, as it turns out, not only drives the public’s distrust of government but also points the most important way out of this mess.

In a July 2024 survey by the Partnership for Public Service, “corrupt” was the most common word that respondents used to describe the federal government. And among 18-34 year-olds, the fast-growing and increasingly critical swing vote, just 15 percent trusted the federal government. That’s lower than any other age group. And it’s a worrying sign that the future will be even worse than our present moment.

Escaping the death spiral

The way out of this mess, message-meister Frank Luntz found in a June 2024 collection of focus groups and surveys, begins by recognizing that people want a government that is honest, trustworthy, transparent, in touch, accountable, and respectful, without excuses for why things go wrong.

Of those Luntz surveyed, 24 percent thought that the government did an “inadequate job” of serving the people; 18 percent thought the job was “terrible”; and 16 percent believed that the government was “simply incapable of doing anything at all.”

That’s why we neither grownups nor social media can put Humpty Dumpty back together again. It’s hard to trust a government that people don’t believe is working for them.

But amid this boiling cauldron of problems, there’s one silver lining. Overall, 78 percent of Americans want a federal government that functions effectively, and that’s one thing on which Democrats (89 percent) and Republicans (82 percent) agree. Luntz found that people wanted a government that made them feel respected, appreciated, valued, safe, and secure. People trust a government that convinces them that policymakers have their interests at heart—and that they’re working to produce results that matter to them.

A March 2026 survey reinforces that conclusion. Americans worry most about “the cost of housing, groceries, electricity, and everyday expenses” (79 percent) and are concerned that “political division and dysfunction in Washington” are keeping us from getting there (62 percent). At the top of those they blame: “politicians who’d rather fight than fix problems” (69 percent). What they want: “honesty and integrity” (76 percent) along with “a determination to get things done—even when that means working with the other party” (72 percent). This is precisely the strategy that Francis Fukuyama is leading with a team through Reform for Results. (Full disclosure: I serve on the executive committee of that group.)

And that suggests part one in the escape from democracy’s death spiral: building a government focused laser-like on producing results that matter to people, not entangled in self-aggrandizing politics. That’s a tall order, of course, but we do know how to do it. Luntz found that people like Social Security, Medicare, Medicaid, the TSA, FDA, SNAP, and roads and bridges, because they see—and appreciate—the results.

But delivering them in the era of mega-change is an even taller order, with everything changing at the speed of AI and with government lagging in the tech revolution.

That leads to part two: instead of relying on the grownups, we need to bring in the young ‘uns. We need fresh energy, fresh ideas, fresh technological skills, and fresh ideas about weaving networks.

The Pentagon, for example, for the last decade has run a “Hack the Pentagon” competition, where ethical hackers are turned loose on the department’s IT systems to find vulnerabilities. In the first competition, one hacker found a problem in just 13 minutes. Many of the most successful players had little or no college education.

But many of these successful players don’t have the standard credentials to check off the boxes in the federal government’s hiring process, so they drift off to the private sector. This is starting to change, with the federal government relaxing educational credentials for tech workers.

Uncle Sam is saying, “I want you!”—but “I can’t get you.” These are precisely the people the federal government needs to build a government that produces results, but it struggles to lure them in, especially because the federal government’s hiring system is in desperate need of repair. It takes three times longer for the federal government to hire than a typical private company, and in that time, smart young ‘uns, especially those weighed down by student loans, get snapped up before the federal government can grab them.

Uncle Sam, though, has an even bigger problem. The Partnership for Public Service found that two-thirds of 18-34 year-olds believed that a federal job “is an opportunity to have a positive impact on my community”—but an equal number reported they never considered a federal position. The more politicians trash the government we elect them to lead, the harder it’s going to be to convince ambitious young ‘uns to come to work for the feds.

So we know that it’s foolish to wait for grownups to save us, or to expect that social media will suddenly start drawing us together instead of splitting us apart. But we also know that if the government focuses more on producing results in a trustworthy way, it can draw in the young ‘uns who are the key to escaping the democracy death spiral.

The first step is to stop demonizing the very government we want. It’s an easy, tempting target. But the more we take aim and fire at it, the more we tighten democracy’s death spiral.

Each party, in its idea factory, is producing new ideas and ideologies. The ideas are pushing the parties—and the country—apart, and there’s no question that the right is far better at this than the left, which is having a hard time generating ideas on which anyone agrees. That’s sure to make the death spiral worse.

But there’s hope, captured in an exchange of letters between Thomas Jefferson and John Adams, who had the fiercest, ongoing political battles in American history. In their last years, however, they came to a consensus around the basic values of America and their hopes for its future.

Jefferson wrote a late-night letter to Adams. “What a Colossus shall we be. I like the dreams of the future better than the history of the past. So good night!” Adams replied, “May we be a Barrier against the Returns of Ignorance and Barbarism. What a Colossus Shall We Be!” Their agreement on the goals for the future, Adams concluded, led to a prophecy that “you and I Shall Soon meet and be better Friends than ever.” Meet again they did, when they both died on the same day, 50 years after America’s first Independence Day.

Their shared search for America’s promise pushed aside the differences that had divided them for decades. That’s not a bad thought, as we near the 250th anniversary of that day.

Donald F. Kettl is Professor Emeritus and Former Dean at the University of Maryland School of Public Policy. He’s the author of the new book, The Right-Wing Idea Factory (Oxford University Press).

I like this graph.

I’ve used it in articles, classes and presentations to the public. It shows how registration rates between Black and White voters varied in Louisiana since Reconstruction. In a single image, it tells the story of political power and discrimination. Black voters had power, briefly, then it was taken from them. A series of policies that were more or less explicit in their discriminatory purpose worked as intended, disenfranchising Black voters, and some poor White voters along the way.

This ended with the Voting Right Act of 1965, which removed the ability of governments to put in place facially neutral but clearly discriminatory practices, and in doing so ratcheted up Black registration in Louisiana until it matched White voters. This effect was immediate and enduring across the South.

The VRA increased Black political representation…

…which in turn increased Black economic power by opening up previously barred avenues, like public employment.

The reason I like these graphs is that they are clarifying. They cut through the noise. Not much ambiguity about them. They tell you the story of Reconstruction, Redemption, Jim Crow, and the Civil Rights era. There might have been political squabbles of the day, most obviously an extraordinary realignment of the two major political parties on racial issues. But the big picture is clear: different eras allowed the use of power either toward equality and honoring the constitution, or toward inequality and discrimination.

Being able to step back and see what is happening is easier in retrospect, of course, But we should be able to see what is in front of us now. The Louisiana v. Callais decision from the Supreme Court decision effectively guts what was left of the VRA. So lets be clear about what is happening.

What did the Supreme Court do to our democracy?

What has been the impact of the Roberts Court on democracy? Wealthy people have more political power than everyone else. Partisan actors have more power to reduce voter choice. Black people and other minority groups are offered fewer protections and less visible political representation.

If this seems like hyperbole, lets review the key decisions.

• With Citizens United in 2010, the Roberts Court killed any meaningful effort to regulate the flow of money in politics.

• The Roberts Court refused to take on gerrymandering when Justice Kennedy was on the court, and then in 2019 said that partisan gerrymandering might be “contrary to democratic principles” but there was nothing they could do about it.

• With the Shelby County decision in 2013, it eliminated the pre-clearance section of the VRA, which meant that districts with a record of discriminating against Blacks would no longer need to get permission from the DOJ for their changes. With the 2021 Brnovich case, they further limited the conditions under which laws that made it harder to vote could be considered as discriminatory.

• Now with Callais, SCOTUS gave state the green light to redistrict majority Black districts out of existence. The Court insisted it is not overturning the Voting Rights Act, but it made the bar for demonstrating such actions are illegal so high as to be impossible.

What happens next?

Immediately after the Callais decision, multiple Southern states moved to change their voting maps to eliminate remaining majority Black districts. People have already started to vote in Louisiana primaries, but the Governor declared a state of emergency to stop the process in order to redraw maps.

Florida’s existing redistricting efforts got a boost. Tennessee, Alabama and Mississippi officials have pushed their own plans. In all, eight Southern states could redistrict over a dozen seats by eliminating Black-majority districts.

Some of this appears a little too on-the-nose, as when the state leader by the name of Shad White calls for eliminating the only majority Black district in Mississippi.

Tennessee moved first, carving up Memphis to ensure the city could not meaningfully choose its representative in Congress. This led to an extraordinary image of Justin Pearson, a Democratic member of the Tennessee legislature, being denied entry by the Sargent-at-Arms for a legislative meeting that is redrawing the Congressional district he is running in. The picture could be from 1956, or 1886, but it is from 2026.

This response to Callais was entirely predictable. After Shelby, several states changed their laws, leading to almost 100 more restrictive election laws being adopted. Voter roll purges became more common. Black eligible voters became significantly less likely to vote because of Shelby. The racial turnout gap between White and Black voters grew almost twice as quickly in parts of the country that were previously governed by the VRA.

The Shelby decision was based on the premise that race no longer mattered to electoral laws. That premise was proven wrong, but the court moved ahead this week to continue to ignore race. They also blessed the state that kicked of the redistricting wave. Despite a detailed lower court ruling showing that the Texas map was racially gerrymandered, the Supreme Court gave it the go-ahead.

Democratic states like Virginia and California responded to Texas with their own redistricting plans. The pattern seems clear: our country will be made up of institutions that are designed to not accurately represent voter preferences. Parts of the country defined as red and blue will look even more so, ignoring the sizable number of voters who are not represented by the dominant parties. There will be fewer Black faces in Congress, especially from the parts of the United States that once enslaved Black people.

Share

Looking for clarity

Political scientists like to talk about polarization and commentators talk about culture wars because that’s the coin of the realm in their respective fields. But those are not really an explanation for the evolution of the Supreme Court, and I think that’s probably reflective that they are generally overused explanations for shifts in power in America.

Here, I’m drawing a bit here from a point made by the political scientist Jake Grumbach.

The graph I showed you at the top of this piece came from the American Political Science Review. Over the last decade, US domestic political scientists found themselves turning to their comparative colleagues or American history to find useful concepts to explain what is going on. (I am as guilty as anyone in this regard). We took democracy for granted, and struggled to explain what was happening right in front of us.

I think there is an equivalent problem in the commentariat. We are living in an era of unprecedented concentration of wealth and power, an era of extraordinary democratic backsliding. When I read blogs on Substack or op-ed pages, this is featured, but not with the sense of frequency and alarm one might expect. People have set up entire new media operations on the premise that now, now is the perfect time to punch left. (The left has plenty of problems, but attacking your democracy is not one of them). Or now is the time to argue that wokeness is the reason for the backlash we are seeing. People should write what they want, but at some point these perspectives are simply failing to engage in real criticism of power or defend American democracy.

Legacy media that have been captured by the emerging billionaire class are not going to challenge these broader trends. The Washington Post said that concerns about the Texas gerrymander was exaggerated, but the response from Virginia, which unlike Texas was at least approved by the public, was a “power grab by Democrats.”

Polarization or culture war explanations don’t explain what has happened to the Supreme Court in recent decades. The banal explanation is that very wealthy groups organized a project to capture the institution through a series of investments into college campuses and the legal profession, such as with the creation of the Federalist Society. They closely vetted SCOTUS candidates, and invested heavily in political spending that was further enabled by the Court. These groups opposed the VRA and elevated those who embraced an anti-democratic view of America. John Roberts made his opposition to the VRA clear as a lawyer with the Reagan administration. Alito might be a culture war crank, but that is not why he and the other five Republicans signed on to a project of weakening American democracy.

What are the solutions?

Step back and look for clarity: it sure looks like Judges nominated by the Republican Party are willing to overturn laws when it helps the Republican Party protect and extend its power. If you take this point seriously, what are the solutions? Vote harder is not a great recommendation if the Court is making voting processes less representative.

The VRA was crowning achievement of the Civil Rights movement, the product of a civil war, constitutional amendments, and intense political and social battles. Its intent was clear. And it was swept away, bit by bit, by a court whose majority was nominated by a President who lost the popular vote and which is historically unpopular.

Justice Kagan’s opinion on the case explained that the Voting Rights Act:

was born of the literal blood of Union soldiers and civil rights marchers. It ushered in awe-inspiring change, bringing this Nation closer to fulfilling the ideals of democracy and racial equality. And it has been repeatedly, and overwhelmingly, reauthorized by the people’s representatives in Congress. Only they have the right to say it is no longer needed—not the Members of this Court. I dissent, then, from this latest chapter in the majority’s now-completed demolition of the Voting Rights Act.

The VRA was eliminated not because it was no longer necessary, but because it had succeeded in challenging power structures. A court that will do this will eliminate other programs designed to make America more democratic and more fair.

The political scientist Jon Ladd responded to the decision by suggesting we should adjust our expectation that any pro-democracy agendas would be deemed constitutional. If Democrats were able to pass the John Lewis Voting Rights Act, or a national gerrymandering ban, would the Supreme Court allow such policies? Would the Court find a way to block proportional representation? Would they allow the extension of statehood to DC or Puerto Rico?

One of the great cautionary tales of American politics is that Roosevelt’s effort at court packing, which is seen as a dangerous overreach. But today, the conditions are different. “The Supreme Court itself has shown itself to be the enemy of democracy” according to legal scholar Rick Hasen.

Under these conditions, court reform becomes a logical solution. (For a contrary view Steve Vladeck argues that Congress can still make the Supreme Court accountable). It would require more Democrats and more of the general public to buy into the notion that defending democracy requires radical action.

But when the opponents of democracy are radical, the defense has to be clear-eyed. We look at countries like Brazil and South Korea, where leaders have been imprisoned for anti-democratic actions, or Hungary, where there is a purge of Orbanists from government. They recognize the urgency of their moment. Do we?

I was at a conference in Bucharest in Romania last week, a setting which reminds you about the fragility of democracy. On a discussion of democratic backsliding, I proposed the following test:1

Remember when you read a book or watched a movie about the 1930s, and you asked yourself “what would I have done under these conditions?” The answer is: whatever you are doing now.

Work requirements – or more accurately, paperwork requirements – are simply an alternative way for Trump to achieve his real goal: dismantle housing assistance, without receiving Congressional permission to do so. This includes the Housing Choice Voucher program and Project-Based Rental Assistance.

A few months ago HUD unveiled a plan to impose time limits and add work requirements on rental assistance. The Center on Budget and Policy Priorities estimates that 3.7 million people, including almost two million kids, could lose this support as a result, in a program that only helps about 10 million people. That seems like a pretty effective way to tip vulnerable families into homelessness.

OMB Director Russell Vought has argued that housing support is simply “subsidized irresponsibility.” But as you consider Vought’s judgment, bear in mind he has taken $15 million in development aid from the poorest people in the world, and another $5 million from funding to protect American consumers…to pay for his personal security detail.

If anything, we don’t do nearly enough to ensure children have a roof over their heads. Current estimates are that over 4 million children experience homelessness each year. Only 1 in 4 of those eligible for housing assistance actually receive support.

One can’t overstate the importance of these benefits. They reduce poverty, substantially reduce the risk of homelessness, and have downstream positive effects on people’s health and well-being. Overall, 70 percent of housing assistance recipients are children, the disabled, or older adults.

And the proposed changes come at a time when housing needs have skyrocketed, with ‘worst case’ housing needs (such as low-income families spending more than 50 percent of income on housing) rising by over 30 percent in the last 20 years. But rather than expand housing opportunity, the Trump administration is simply going to cut access to the program.

Housing assistance is already notoriously difficult to obtain. Because there isn’t enough funding to meet needs, most people who ultimately receive these benefits wait years before they get help. Wait times vary across places, but in Alabama the average wait time is 5 years, compared to 2 years in Arizona, 4 years in the District of Columbia, and 3 years in Georgia. Those receiving housing assistance are also tightly monitored, and face bias from landlords.

Work requirements don’t work

Scott Turner, the head of HUD, has justified the proposed rule as enabling self-sufficiency. But thats not what the research says. The main takeaways of this research (described in more detail below) are:

• Work requirements do not increase labor force participation, because people who can work are already working.

• Work requirements cause eligible individuals lose access to public services even if they are working because lots of people struggle with the paperwork.

• While work requirements might exempt some groups (this plan excludes people with a disability or with very young children for example), if these groups also have to demonstrate their exemption status via onerous paperwork, they are still at risk of losing support.

• The housing authorities that administer these programs are already understaffed and overwhelmed with managing existing onerous requirements.

The rule is structured as a waiver, meaning that it allows but does not force owners and housing agencies to adopt work requirements. But as CBPP points out HUD “could pressure housing agencies and private owners to adopt them, including through illegal tactics like those the Trump Administration has used to impose its policy priorities and preferences in other areas.”

Sounds bad? Here is something you can do

HUD is seeking public comment on this new policy. You can write a comment here. No log in. You can enter text, or upload a document.

Deadline is midnight on May 1. You don’t need to read the rule or be an expert. You can use research cited in this blog if you feel like it. Take some of the time you would have spent complaining on social media and do this instead. You can do this in 5 minutes.

The administration has to record support and opposition for the rule, and reach each comment. Make your voice heard! Do you know someone who might care about this policy? Share with them.

Share

Want more information? Here is a simple guide about the public comment process. Here are more resources about this particular issue that provide comment templates.

One of us (Don) submitted a comment that quoted parts of our recent literature review in the Journal of Economic Perspectives:

Work requirements assume a population of individuals who could be employed, but choose to rely on welfare benefits instead. Moreover, they assume a population that is not already subject to work requirements under other welfare programs. In practice, this population is very small. Additional work requirements do little to spur additional labor force participation. For example, just 6 percent of those on Medicaid are nonemployed and nonelderly clients who would not qualify for an exemption from work requirements, and one-third of those 6 percent are retired (Guth et al. 2023). There is little evidence that work requirements in the Supplemental Nutrition Assistance Program have substantially increased labor force participation. However, such requirements have caused vulnerable groups, such as the homeless and those with chronic health conditions, to lose benefit access (Cook and East 2024; Cuffey, Beatty, and Mykerezi 2022; Gray et al. 2023; Han 2022; Ndumele et al. 2024). For this reason, some safety net advocates characterize work requirements as “work reporting requirements,” to communicate that the consequential aspect of the regulation is the paperwork burdens, not on actual labor force participation.

When Arkansas adopted Medicaid work requirements, 95 percent of those targeted by the program were already employed or should have been exempted due to disability, but the work requirements in the state reduced Medicaid enrollment by 12 percentage points (Sommers et al. 2019). About one-third of those who lost coverage, who were disproportionately those with lower education, were simply unaware of the new requirements. The work requirements did not increase labor force participation. However, half of those who lost Medicaid coverage reported serious problems paying off medical debt, 56 percent delayed care due to cost, and 64 percent delayed taking medications because of cost (Sommers et al. 2020).”

An Administration of Burdens

There are some pretty clear signals that the Trump administration sees burdens as a way to shrink the safety net. Adding work requirements to food, health or housing supports is not just cruel, it is unnecessary and ineffective in encouraging work for the very good reason that people already have strong incentives to work because of the the Earned Income Tax Credit.

Here is another example of the Trump administration’s use of burdens to block access. Automatic renewal is a useful way to manage the hassles of safety net programs, reducing burdens both on state officials and clients, and improving program integrity. But the Trump administration is planning to limit automatic renewal according to the New York Times: “By one estimate, that would remove six million more people, including 1.8 million children, from SNAP.”

In other domains the sheer cruelty of the regime of burdens is hard to fathom. This ProPublica article will make you simultaneously sad and outraged: another proposed Trump policy, initiated by DOGE, would punish up to 400,000 people with disabilities for living with their parents by reducing their Supplemental Security Income benefits.

The story follows families who would be impacted, including Shy’tyra Burton, who was born two months prematurely, and suffers from significant intellectual disabilities. Her Dad works as a sanitation worker, earning about $2,000 a month to take care of her and her siblings. Her SSI would be cut by about a third under the planned change, losing about $330 per month.

The policy would affect people with with Down syndrome, dementia and other disabilities. It would also overwhelm these people, who are typically living with their parents because they cannot manage the demands of life on their own, under a wave of administrative burdens:

If enacted, the change will require intellectually disabled young people like Burton as well as very elderly people to file extensive monthly reports if they want to continue their benefits even at the reduced level. They’ll have to provide details about the property where they live: whether it’s leased or owned, as well as the names of anyone in the home, and whether any of these people has any new income or assets. They’ll also have to include documentation of all household bills and expenses, showing how much they do or don’t contribute personally, as well as financial documents such as bank statements and any pay stubs.

Burton will likely have to make an appointment and report in person at a Social Security field office any time her father’s hours or wages change even slightly; any time she and he switch up how they split utility bills; and any time an adult sibling spends even a few nights at the house and helps her with living expenses. If she doesn’t, she could later receive bills accusing her of having been overpaid by Social Security.

If you have a child with a disability, a large part of your brain is centered on how to ensure they can live a life with dignity and security in a country where the safety net seems to be designed to trip you up. Even with the current supports her father says “I’m still barely managing” to keep his household above water. And now a presidential administration, one where senior figures take turns to loudly emphasize their Christianity, would drown them. Jen Pahlka is fond of saying “paperwork favors the powerful” — the corollary is that “paperwork punishes the powerless.”

More on work requirements

As news erupted across the country on December 4, 2024 that the UnitedHealthcare C.E.O. was killed in New York City, alongside rightful condemnations of violence was an outpouring of frustration with delays and denials of coverage, predominantly at the hands of private health insurers.

To casual observers of the American health care system, it might seem puzzling that in the world’s wealthiest nation, there would be such widespread grievances about barriers to prescribed medical care. But to many Americans – especially those with chronic medical challenges – this only scratched the surface of health care burdens and the intensely political processes driving these barriers.

Every day across America, people learn that they cannot pick up prescribed medication or schedule a test or procedure because their insurer has not provided prior authorization, or required pre-approval for prescribed care.

In prior authorization’s nascency in the 1960s, it was quite narrowly circumscribed, such as requiring physicians to certify the necessity of a continued hospitalization. As health care inflation and soaring Medicare costs intensified cost containment pressures, America expanded its reliance on managed care. With it came new incentives: coverage decisions were made not simply by prescribing physicians (who in fee-for-service arrangements had some degree of perverse incentives to overprescribe), but by entities with fiduciary responsibilities to shareholders and obligations to contain costs and maximize profits.

Utilization guardrails proliferated, ostensibly to mitigate waste. What began as certification of hospital lengths of stay expanded to prior authorization for costly drugs (and even some less costly ones), high-tech imaging, and many surgeries and procedures. What may have begun as an effort to curb overprescribing has resulted in the delay and denial of medically necessary care, along with the imposition of burdensome processes to reverse them. That is the story I tell in my new book, Coverage Denied: How Health Insurers Drive Inequality in the United States, drawing on interviews, a nationwide survey, and administrative data.

The Prevalence of Coverage Denials

Because health insurance is fragmented across public and private payers, precise denial rates are difficult to determine and consequently, most analyses focus on slices of the market. For example, KFF finds that roughly 1 in 5 claims are denied each year by Affordable Care Act marketplace plans, though KFF’s analysis of prior authorization denials by Medicare Advantage plans revealed a lower denial rate of just under 8 percent. And Optum found through analysis of hundreds of millions of claims that the denial rate has increased from 9 percent in 2016 to 12 percent in 2023.

Surveying 1,340 patients nationwide, I found that 36 percent of survey respondents experienced at least one, typically more than one, coverage denial. These denials were for wide-ranging care, from prescription drugs to behavioral health to imaging to surgeries, highlighting the breadth of this practice in modern health care delivery.

Denials can occur pre-treatment (through denial of prior authorization) or post-treatment (a claim denial). Both are destabilizing for the patient and burdensome to rectify, but their effects are felt quite differently: prior authorization can keep care out of reach, while claim denials can jeopardize financial security and drive medical debt. While I do not observe pronounced disparities in susceptibility to being denied in the first place, the administrative and financial burdens of denial fall inequitably, with patients from marginalized backgrounds bearing the brunt of this managed care tool.

The Complexity of the American Health Care System

“Nobody knew health care could be so complicated” understandably elicited derision in a nation whose health care system is defined by complexity, fragmentation, and opacity that all too often strains the capacity of patients and providers alike. Public and private health insurance intertwined through delegated governance, and each insurer maintains its own (and annually changing) formulary and coverage rules, creating prescribing confusion and discontinuities when patients switch plans.

Consequently, all too often, physicians are left prescribing in a “black box,” unsure whether a treatment will be covered, require prior authorization, or be placed on a high cost-sharing tier. While some electronic medical record systems provide real-time information, access is uneven and inconsistencies across insurers reduce the extent to which the information yielded is actionable.

When opacity prevails, coverage barriers follow. And while initial denials are not the final word, what unfurls next is layered administrative burden on patients and physicians alike, contributing to health care rationing by inconvenience.

Share

Navigating Red Tape in Modern Medicine

The American health care system’s complexity, combined with Americans’ low health insurance literacy, can be a recipe for disaster, producing widespread confusion and consternation as well as delayed or foregone care, not to mention distrust of the health care industry. Frequent denials and cumbersome appeal processes exemplify administrative burden, or the learning, compliance, and psychological costs that combine to make the experience of policy feel onerous.

Denials of coverage can generally be appealed, but at a cost to both patients and physicians. Patients face a number of informational disadvantages that can help explain why KFF reports such low rates of patient appeals in the setting of marketplace plans’ claim denials (only less than 1 percent of which are appealed!) and in the setting of Medicare Advantage prior authorizations (only 11.5 percent of which are appealed).

Patients must know they can appeal, believe it is worthwhile, and understand how to appeal. This information may at first blush seem quite simple, but in practice, it is anything but. And it makes it all the more troubling that an anonymous claims reviewer for Elevance observed, “We’re told denying things is okay because people can appeal.”

Surveying 1,340 U.S. patients, I found that the two most common reasons for not appealing were lack of awareness that appealing was an option and the belief that they would not win. While health insurers must notify patients about the right to appeal, there can be knowledge gaps between patients and their insurers when information is conveyed in fine print or at a very high reading level relative to the average American. For example, while the average American adult reads at around an eighth grade level, ChatGPT estimated that UnitedHealthcare of Arizona’s 13-page appeal process information packet was written at the 13^th grade level, raising the odds that patients – especially those from marginalized backgrounds – may fall through the cracks and lose access to health care.

Doubting one’s ability to win is understandable when imagining the difficulty of an ordinary individual challenging the determination of an insurance giant armed with substantial experience and legal counsel (conjuring arguments illuminated in Marc Galanter’s seminal 1974 analysis of why the “haves” come out ahead of “one-shotter” plaintiffs). What’s more, when asked to estimate the proportion of health insurance appeals that are successful, most survey respondents estimated a rate under 20 percent, when in truth, it’s around a coin flip, with various estimates between 40 percent and 60 percent.

Even once overcoming the informational burdens, the layered process can be onerous on a good day, and patients are rarely having their best day when navigating these insurance processes. What’s more, physicians themselves are mired in burden or must maintain substantial staffing support to offset these burdens (though there are obvious inequities in access to this means for shifting burden).

The physician may need to dedicate some clinic time to play phone tag with the patient’s insurer to engage in a “peer to peer” review of the prescribed care, though the insurer-employed physician is rarely in the appropriate specialty. When this does not result in a reversal of the denial, there is a formal appeal process within the insurer, which may require the submission of additional documentation, some of which may have already been submitted. Ensuing requests for additional documentation may be highly time-sensitive, potentially producing denials due to lack of timely response.

One physician interviewee reflected:

It’s frustrating because you know you’re going to win in the end. You know the criteria. You know the patient qualifies for it. Yet you’re battling week after week, delay after delay for what you know they qualify for because you know the policy.

Patients may need to coordinate among their physician’s office, their insurer, and potentially a specialty pharmacy. They may need to be on hold during the middle of the day, competing with other responsibilities. Patient interviewees spoke of the desperation with which they talked with their insurer, submitted documentation, and took note of every conversation and reference number, even while lying in a hospital bed as one patient, Samantha did amid her protracted battle for a lifesaving bone marrow transplant. Reflecting on this experience, she said:

I learned from an early age, if you want to live, you have to be able to fight these battles. You have to have a level of literacy that is master’s degree or higher in public health or a legal background in order to navigate this. You have to get creative and look for back doors.”

It is little wonder why I find that less affluent patients are less likely than their counterparts to appeal at all, and why marginalized and sicker patients are less likely to prevail in this endeavor. So, while coverage barriers themselves are a widespread American experience, like so many other facets of burden, their effects are felt inequitably, leaving far too many behind, entangled in the red tape of modern medicine as marginalized patients also become more likely to delay or forego medical care (often, though not always with adverse health consequences) and to delay non-medical purchasing to accommodate these unexpected insurance barriers.

Thus, amid America’s successive political choices to increase reliance on managed care, risk has been shifted from health insurers to patients (as well as their physicians) in ways that profoundly shape health and financial security, and that widen existing inequities.

Where We Go from Here

When one patient interviewee recounted her trials navigating access to a life-saving bone marrow transplant, she observed that she had to be her own lawyer multiple times a day and asked, “How is someone’s grandmother supposed to do this? How is someone with four children supposed to navigate this? The deck is stacked against you at every single turn.”

So, where do we go from here?

• California has led the charge in working to strengthen oversight of health insurers’ use of artificial intelligence (AI) tools with which to process claims and prior authorizations, restoring human oversight before the issuance of denials. Enforcement of plain language health insurance communication could reduce knowledge gaps, especially among non-native English speakers and those with lower educational attainment.

• Some states have sought to pare back prior authorization requirements for those with sufficiently low reversal rates of their prescribing decisionmaking, though these policies are administered at a granular level and thus may not deliver on their promise.

• Massachusetts is working toward streamlining and reducing reliance on prior authorization processes across many areas of medicine including primary care. However, any state insurance reforms will be confined to state-regulated health plans due to the constraints of Employee Retirement Income Security Act (ERISA) preemption of laws that “relate to” the self-insured health plans in which roughly 100 million Americans are enrolled. Thus, national reform is necessary to comprehensively improve health care access, though federal policies currently tilt toward expanding prior authorization.

If the public outrage in full view in December 2024 revealed anything, it is that experiences of health insurance barriers are anything but aberrations. Rather, they are symptomatic of America’s policy choices to house health insurance delivery in the hands of actors mindful of cost containment and profit maximization, and too often place paperwork between patients and their prescribed care. In Coverage Denied, I show how we arrived at this moment, who gets left behind, and what is required to achieve a health care system in which access to care does not depend on one’s stamina in overcoming red tape.

Miranda Yaver is an Assistant Professor of Health Policy and Management at the University of Pittsburgh, where she holds additional appointments in Political Science and the School of Public and International Affairs.

It is not enough for the broligarchy to have untold wealth. They also wanted political power. Now they have that. And they want to be esteemed as philosopher kings, while simultaneously shielded from criticism. Well, that’s a taller order.

When they roll out their political manifestos, we should take them seriously, if not literally. Public statements by these actors, while often couched in statesmanlike or visionary terms, offer insights into a growing power elite: what they like, what they hate, their enemies, what they felt are entitled to.

These manifestos are relevant for governance, given Silicon Valley’s power via financial resources and platform control, and their embeddedness into government service delivery via grants, tax subsidies and contracts. They are no longer simply contractors, or one more donor class, but a powerful political class onto itself.

For example, when DOGE was announced, your best bet for understanding its purpose and impact was to pay attention to Elon Musk’s anti-government conspiracy ramblings on X, which fueled the elimination of entire agencies. Or Marc Andreessen’s manifesto, which predicted an attack on government services. These manifestos are, of course, riddled with unfounded claims, contradictions and blind spots. Andreessen bemoaned regulatory capture, while embarking on exactly such a program. But even, and perhaps especially, the blind spots are telling.

Some in the broligarchy see themselves as heroic historical figures, the kind of men (its always men) who are on a heroes journey that involves quashing some foes along the way. What if you are the foe? At some point this becomes a sovereignty problem, when the people empowered to manage your government have their own distinct values they are pursuing using the guise of public power and virtue.

The Palantir manifesto

And so we get to Palantir, a tech company that provides support for US government intelligence, immigration, policing, defense, tax preparation, health, revenue collection and logistical functions. Given that range of competencies, they are at the heart of concerns about tech surveillance under an increasingly authoritarian government.

A senior person in the Biden administration assured me that Palantir were seen competent deliverer of contracts. They got stuff done, including with pandemic era logistics in health care. The company grew under Democratic and Republic administrations, but the billion dollar contracts and eye-popping stock valuations did not start to arrive until the second Trump administration. They benefited from a fertile combination of massive increases in immigration and military spending, the rise of AI, and the downsizing of the federal government.

The way in which Palantir grew caused unease within its own ranks. A year ago, some former Palantir employees wrote an open letter about the violation of the company’s prior ethical guardrails.

A Code of Conduct was crafted to uphold democracy, preserve the spirit of free scientific inquiry, and ensure responsible AI development. Guardrails were set to prevent discrimination, disinformation, and abuses of power. These principles have now been violated, and are rapidly being dismantled at Palantir Technologies and across Silicon Valley

The company’s leader Alex Karp released a manifesto, which is actually just excerpts of the “The Technological Republic” written by Karp with help from Nicholas W. Zamiska. I’m not going to go through every point of the manifesto (some of which I am sympathetic to), but do want to focus on a couple of big points

Share

Can an anti-fascist become a techno-fascist?

A takeaway point is that on the whole, the manifesto’s vision of Palantir is that of a US government and its tech allies as dominant players, unconstrained by accountability. This has caused alarm bells to go off. For example, in a series of posts on Bluesky, Cas Mudde, one of the leading scholars of authoritarianism, was blunt:

The Palentir statement on its desired “Technological Republic” is one of the scariest things I have seen in a while. It is a call for a world dominated by an authoritarian U.S., generated by AI (both the statement and the world), run by tech-surveillance companies. Technofascism pure! Clearly, this disqualifies Palentir as a business partner for any other country than the U.S., not just in defense and security related sector but ANY sector — as information is power. European countries should not just start new collaboration with Palentir, they should divest it ASAP! Democrats should develop an actionable agenda of democratic reform in case they return to power. This cannot be limited to institutional refoms, but must include reigning in the power and wealth of technofascist companies and individuals.

Is it possible that the underlying book, with more room to caveat and explain, is less objectionable. Not so, according to John Ganz, who slogged through the whole thing:

The book is extremely creepy: It becomes clear in the course of reading this “Technological Republic” the authors propose is essentially some kind of merger or acquisition of the United States government by Silicon Valley, a state run by an engineering elite that would be empowered to “ruthlessly” pursue “outcomes.” It’s a proposal for a kind of tech oligarchy: “no public "oversight for me, surveillance for thee.” I contend it’s a work of reactionary modernism.

The manifesto makes a straightforward case that government and Silicon Valley power must be fused in order to defend a peaceful and prosperous way of life for Americans. If we pause with AI we are merely allowing our adversaries gain an upper hand:

The question is not whether A.I. weapons will be built; it is who will build them and for what purpose. Our adversaries will not pause to indulge in theatrical debates about the merits of developing technologies with critical military and national security applications. They will proceed.

The manifesto argues Silicon Valley has invested its efforts in frivolities rather than American military capacity — unlike Palantir! Western values, which are superior to others, must be defended with power:

Some cultures have produced vital advances; others remain dysfunctional and regressive. All cultures are now equal. Criticism and value judgments are forbidden. Yet this new dogma glosses over the fact that certain cultures and indeed subcultures . . . have produced wonders. Others have proven middling, and worse, regressive and harmful. We must resist the shallow temptation of a vacant and hollow pluralism. We, in America and more broadly the West, have for the past half century resisted defining national cultures in the name of inclusivity. But inclusion into what?

“Defending the West” is part of Palantir’s DNA. It was never a company that sought to please consumers, but viewed itself as aligning with US military interests. It also works with Israeli military, extending “the West” to a political rather than geographical concept.

Since “The Technological Republic” was published, Karp has gone from being a partial critic of Trump, to a defender. He once declared his “biggest fear is fascism” and that “I always thought if fascism comes, I will be the first or second person on the wall.” So how is he now a Trump donor? How did an anti-fascist become someone seen as authoring a techno-fasicst manifesto?

One of Trump’s extraordinary skills is to recognize and recruit those whose greed and vanity outstrip their principles, and who are happy to shout righteously for a cause they once abhorred. Deregulation, praise and massive contracts are a powerful cocktail to help you forget your commitments, especially when your supposed allies in liberalism are now critical of you. “Being unpopular pays the bills” Karp told a biographer in response to liberal outrage.

Some speculators think Palantir is wildly overvalued due to a Trump bubble where earnings are way out of proportion with stock price, and have tried to short it. Trump praised Palantir when its stock tumbled recently. Right now, the company’s prospects, or at leasts its stock price and leader’s personal wealth, feel increasingly dependent on MAGA’s vision of the world.

But it is uncomfortable to label yourself a sell-out, and so a billionaire anti-fascist write a manifesto about how, yes, they are still defending democracy, and no, they are not working for a fascist. In a thoughtful column, Michelle Goldberg noted that the German politician who was the primary subject of Karp’s dissertation went from being a fascist in the dissertation to a speaker of forbidden truths in “The Technological Republic.” Trump seems to have made the same transition for Karp.

As Palantir becomes engaged in domestic surveillance and immigration enforcement for a government that has repeatedly violated the law, “defending the West” starts to look indistinguishable from bog-standard right-wing ethno-nationalism. As part of a somewhat bizarre critique of his one-time mentor Jurgen Habermas, Karp argued that the structure of public discourse cannot be rational but “must be rooted in a more corporeal and traditional — and indeed national and cultural — source.”

His biographer said that Karp’s views on the West have changed:

Judging by his own words … he does not see multiracial, pluralistic democracy as the thing about the west that should be defended...he sees it much more as just a collection of countries bound by a shared Judeo-Christian heritage, and, to varying degrees, by an attachment to free enterprise. That’s kind of where he is, I think. And it can lead you down some pretty dark paths.

Goldberg’s profile of Karp notes that at one point his reputation as a Democrat, avowed anti-fascist, and bi-racial student activist worked well to dampen Democratic hostility to Palantir when they were in power. Now they are not, Karp has more reason to embrace a different version of his identity. Goldberg concludes: “I was left suspecting that the best explanation for Karp’s journey — as for most of his right-wing billionaire compatriots — is the vulgar materialist one.”

Which public servants deserve accountability, which deserve power and praise?

One striking element of the manifesto is it’s incoherence. At turns it is assertive and dominant, and at other points wheedling in its pleas for grace and tolerance. But if you make some simple assumptions about who the manifesto is talking about, it becomes clearer.

Lets take the following points together since they touch on public service:

8. Public servants need not be our priests. Any business that compensated its employees in the way that the federal government compensates public servants would struggle to survive.
9. We should show far more grace towards those who have subjected themselves to public life. The eradication of any space for forgiveness—a jettisoning of any tolerance for the complexities and contradictions of the human psyche—may leave us with a cast of characters at the helm we will grow to regret.
11. Our society has grown too eager to hasten, and is often gleeful at, the demise of its enemies. The vanquishing of an opponent is a moment to pause, not rejoice.

18. The ruthless exposure of the private lives of public figures drives far too much talent away from government service. The public arena—and the shallow and petty assaults against those who dare to do something other than enrich themselves—has become so unforgiving that the republic is left with a significant roster of ineffectual, empty vessels whose ambition one would forgive if there were any genuine belief structure lurking within.

19. The caution in public life that we unwittingly encourage is corrosive. Those who say nothing wrong often say nothing much at all.

Again, this is incoherent until you realize that Karp is presenting you with two types of public servants. Civil servants purged under Trump, and the tech elites who venture into government. The former deserve your scorn. The latter — the Alex Karps of the world, the Silicon Valley contractor class — merit your understanding and praise.

Few groups in American public life have suffered more material or reputational damage in the past year than federal civil servants. More than 400,000 employees have left the federal government. For many, they left a dream job, pushed out by people who knew less than they did. They worked under a government where the President declared them to be the deep state, government leaders promised to put them “in trauma” and the President’s supporters often exposed them to harassment campaigns. They were demeaned by bosses who opposed the mission of their organization.

Karp does not have much grace or understanding for this group of public servants. They are overpaid priests. The echo of “will no one rid me of this meddlesome priest?” may not be intended, but its there. As for the claim that government employees are overpaid, generally, white collar government workers are not overcompensated compared to the public sector, and in some cases forego a significant wage premium. We also spend much more money on contractors than employees, so if you wanted to save money...

Another way to look at this is that billionaires have joined a war on professional administrative class of public servants precisely because of their proximity to formal mechanisms of accountability. Civil servants are sworn to uphold the constitution, report wrongdoing to officials, follow the rules. They are in most respects more accountable than our politicians to ethical and legal requirements, and more difficult to buy off. This makes government slower and more irritating for sure, and puts those officials at odds with the Musks, Andreessens, and Karps of the world, whose vision and interests benefits from reduced oversight and accountability.

I do have some sympathy for the point that procurement processes are unwieldy and political appointees are generally held to a very high standard of scrutiny that often discourages them from participating in government. But frankly, its Democrats who impose those standards, which Trump has abandoned. The Biden administration struggled to get tech elites to join government because of stringent ethical and conflict of interest barriers. The Trump administration has none. Elon Musk made clear that for the chosen few, conflicts of interest no longer exist. New tech hires can stay with their current companies that work with the government.

So the plea for grace for those who enter the public square is not about rules or formal constraints here, but public criticism of power. It has never been easier or potentially more lucrative from the Silicon Valley set to jump into government. The only penalty they face is a social one. While Karp is angry that certain cultures have not been harshly enough criticized, criticism of the consolidation of financial, technological and political power happening in America is off the table.

Public criticism offers one form of accountability. If people violate our laws and norms, should they held accountable? The manifesto’s answers is “yes…for some, not us.” Violent criminals? Yes. Civil servants? Yes. White collar criminals? Well, lets not be hasty? Tech billionaires working for government? That would be “corrosive” for the country.

I’m reminded of the Simpsons character Artie Ziff, Marge’s prom date who later becomes a billionaire. When he asks that Marge not share with others his unwelcome prom night groping, he puts his request for silence in the most virtuous terms possible:

We need to be nicer to the broligarchy….or risk damaging the entire country.

More seriously, how will future Democratic administrations respond to an organization that might have genuinely valuable capacities, but who is not shy about pursuing a hostile ideological agenda and bristles at the slightest accountability.

The merging of people and companies, profit and values

Until relatively recently, Palantir was seen as secretive, criticized more for its connections to founder Peter Thiel than anything Karp said. However, the manifesto is presented as that of Palantir’s, hosted on their website, promoted by their social media. There is no meaningful distinction between the leader, Karp, and the company’s policies, which is an odd thing for a publicly traded company.

Generally its healthy to separate our sometimes weird personal opinions from the organizations we work for. There is a reason that contractors give money to both party and try to limit their ideological stances. They don’t want their values to chase off business. Once those values go beyond internal guidance for employees, and instead become political manifestos, they demand allegiance or opposition.

For example, should a tech company have opinions on military and foreign policy? The manifesto talks about the need to rearm Germany and Japan. Why? Some at the company surely have opinions on this. Others might different opinions. Thats fine. But why should we care about what Palantir thinks?

Similarly, Palantir calls for national military service. This is a common and unfortunate affliction of middle-aged and older men, akin to an interest in the Roman Empire or World War II. Its simply not something you see many men in their 20s promote for obvious I-would-prefer-not-to-die reasons. Charitably, it reflects a sense of a desire for creating shared meaning among the young that the elderly perceive is lacking, or to coerce the elite into sharing the cost of military conflict. Uncharitably, it is the old telling the young how to live and risk their lives.

It is easy to mock Palantir on this. How many of them completed their military service? The may sincerely believe that they are part of the greater military effort via their defense work. Indeed, the Trump administration made some of their executives officers in the Army Reserve. But all of that is beside the point. Why should Palantir have an opinion on this and why should anyone take it seriously?

There are other examples:

The limits of soft power, of soaring rhetoric alone, have been exposed. The ability of free and democratic societies to prevail requires something more than moral appeal. It requires hard power, and hard power in this century will be built on software.

As a statement about foreign power, this seems pretty misguided. The US has dissipated its soft power to an extraordinary degree in a rapid period of time. Are we better off for it? We are, as we speak, engaged in a period of hard power and the public does not seem to like its elements — needless wars, tariffs — or its outcomes — higher prices, weakening alliances.

The problem with the manifesto is that Palantir’s worldview and business model are impossible to separate. A world where soft power has real and lasting impact is simply less profitable for a company like Palantir relative to a world where we blow a lot of stuff up. A world featuring an AI arms race is more profitable than a world with AI regulation. A world where Silicon Valley polices domestic crime is more profitable than a world that constrains surveillance on the public. A world with fewer meddlesome government employees is a world where Palantir takes on the administrative reins of the state.

In each of these cases, Palantir takes a stand that is bravely consistent with its bottom line. In an interview with Wired, Karp says that Palantir has put principles before profits, by refusing to work with China and Russia. But I’m skeptical: how exactly can a tech security firm win contracts from the US government while also working with its primary adversaries? When Palantir gets into a fight with the US government akin to Anthropic’s current battle about the use of Claude for violent or unconstitutional ends, they will deserve them more credit.

When the subtext defeats the text

The subtext of the manifesto here is so overwhelming that the text can barely stand up for itself. For example when Karp says:

The pervasive intolerance of religious belief in certain circles must be resisted. The elite’s intolerance of religious belief is perhaps one of the most telling signs that its political project constitutes a less open intellectual movement than many within it would claim.

…a purely textual interpretation might be that he is commenting on the most powerful man in the world attacking the most visible religious figure in the world. Thats in the news, right? Trump attacks the Pope! A pretty big deal!

But of course, no-one believes that is the intent. Palantir is operating on a series of shared assumptions: that religious intolerance is only a project of the left, despite increasingly open attacks on non-Christian religions from the right; that motivated erosion of individual rights, such as abortions, are not a form of intolerance worth commenting on.

A student who once engaged in Black campus politics now argues that wokeness is a “pagan religion view” inculcated by campuses: “a new religion with sacrifices. Who’s the sacrifice? Me, I’m the sacrifice.” Again, this mix of the powerful preening as victims while blaming woke campuses is bog-standard right-wing fare.

I don’t want to be too critical. Here is a piece of solid advice:

The psychologization of modern politics is leading us astray. Those who look to the political arena to nourish their soul and sense of self, who rely too heavily on their internal life finding expression in people they may never meet, will be left disappointed.

Yes. Yes! YES!

Small problem. Its hard to think of any segment of society that has been more guilty of engaging in this than the tech bros. Elon Musk cooked his brain online. Marc Andreessen and others seemingly struggled with being in virtual spaces where their ideas and status were challenged, and rewired their political beliefs as a form of cope.

The Palantir manifesto reflects the worldview not of deep philosophical consideration but of a rotten information environment, one where pluralism and accountability are bad, and coercive power is the necessary and profitable price of freedom. To what degree should our government endorse and support this worldview?

“Taxes are what we pay for a civilized society” according to Chief Justice Oliver Wendell Holmes Jr.. But the billions of hours and hundreds of billions of dollars Americans spend on tax reporting are the price we pay for a dysfunctional government unwilling to invest in public interest capacity.

In addition to tariff- and war-induced inflation, Americans can also thank President Trump for allowing this time tax to grow. Trump killed Direct File, a free tax reporting option that IRS had built. Binyamin Appelbaum of the New York Times summed up how I and a lot of other people who care about government services feel:

He has destroyed things that are more important than Direct File, but this one sticks in my craw. It was a straightforward way to make life a little better for a lot of Americans. It was a step toward the kind of easy-to-use, efficient, high-tech government services that everyone claims to want. It worked. And now it’s gone.

What Appelbaum is pointing out that Direct File is not just about whether you save some money at tax time. It is also about whether American government can deliver modern public services in a digital age. One member of the Direct File team I spoke to noted their initial surprise to learn that the US was lagging so far behind other countries when it came to digital tax reporting:

I think that we should have greater ambition to actually do things on the Internet in our government. Just because we’re big and complex and multiple time zones doesn’t mean it’s not possible. So I kind of came with that attitude and was stunned, absolutely stunned to find out that the IRS didn’t offer a free tax filing thing.

Over the last few months I’ve been talking to more than 20 people who worked on the Direct File project, everyone from the top of the policy food chain — like former Treasury Secretary Janet Yellen and IRS Commissioner Danny Werfel — to the engineers and designers who worked directly on building the tool itself. This tax season, I’m sharing what I learned. For more detail, you can read a Better Government Lab working paper.

Direct File showed that government could build a high quality tax-reporting digital tool

The idea of reducing tax reporting burdens is not new. Presidents since Reagan made commitments to build something like Direct File before Biden delivered. The logic was clear: you should not have both pay taxes, and pay to complete your legal obligation to report your taxes.

You will never guess who also proposed an IRS app to help for Americans file their taxes.

When I asked those who helped build Direct File about its legacy, a consistent response is that they showed it could be done. Former IRS Commissioner Werfel said “it’s a proof point that the IRS itself can deploy user-friendly paradigm shifting technology in a short turnaround if the preconditions are set up for it to do so.”

The arguments about the technical feasibility about government provided tax-reporting are over. Direct File was small, but it was real and it worked and could have been expanded to tens of millions of taxpayers. Instead, it was killed under DOGE and Trump.

Users loved Direct File:

In 2024, 86% of users said it increased their trust in IRS, not the most beloved of organizations. The Direct File “net promoter score” (a measure of customer satisfaction and loyalty) is 84, well above that of TurboTax, and even Apple. In user surveys conducted by Code for America, 94% prefer Direct File and state equivalents to their previous filing method and 84% said they were “very satisfiedˮwith another 14% being “satisfied” with the product.

Why was Direct File killed?

Without the technical arguments, what are the reasons not to do Direct File? The Trump administration points out that it was not widely used. This is true, but the criticism is a little rich since the Trump administration made no effort to promote the new product it inherited.

The limited use also reflects a strategic choice by the Direct File team to take a “start small, fail small” approach. This was in response to the long shadow of Healthcare.gov, the Obama-era website that rolled out on day 1 without sufficient testing and immediately and very publicly collapsed. Direct File did the opposite — lots of testing, lots of discovery of where users found problems, lots of iterative fixes, improvements, and gradual additions of capability. This approach recognized the political reality that a single big failure would have killed the project. It also matches best practice with how to build a digital product.

It seems more than a little ironic, then, that Direct File was filled by people who said they were bringing tech innovations to government. The fate of Direct File under DOGE was initially unclear. As noted above, Musk talked about building a product that sounded a lot like Direct File, but he also pledged to get rid of the tech unit 18F when someone pointed out they were involved in Direct File

It could be that Musk and DOGE’s intolerance of technologists who did not come from their stable contributed to Direct File’s demise, despite the fact that the Direct File team was doing the sort of thing DOGE said it was going to do: use tech skills to build new products in-house, without relying on expensive contractors.

What we do know is that DOGE officials in IRS initially praised the Direct File team, and reassured them the project would continue — until right before it was cancelled. They changed their tune after meeting with representatives from the private tax preparation industry. Soon after, Direct File was killed.

Intuit, the maker of Turbotax, and other players in the private tax industry poured extraordinary resources into preventing any public option for tax filing. By one measure, Intuit had more lobbyists than the entire Direct File team had employees, and ramped up its spending as Direct File was launched.

The Trump administration says “Direct File had low overall participation and relatively high costs and burdens on the federal government, compared to other free filing options.” I literally wrote the book about administrative burden in the public sector and this has it completely backwards. It is a calculation that only makes sense if you do not value the time and money of citizens, but regard the government doing anything to help the public as a burden.

The average American spends about $290 to file. They also spend about 13 hours, often giving the government information it already has, which works out to another $486 worth of time assuming the average BLS hourly wage.

The Trump administration says that Direct File was too expensive. Compared to what? Direct File cost IRS $16 million in its final year of operation. Between 2021-2025 Intuit also got almost $500 million dollars in federal tax credits.

Share

Public interest capacity in a democracy

And here I want to make some points about state capacity. I think Direct File tells us about capacity in a couple of ways. First, it reflects whether government can provide digital services that helps the public in a political economy where private sector actors can both make money from those services, and use some of that money to lobby the government to kill a public option.

The Trump administration is building state capacities in certain domains, such as immigration enforcement and surveillance. But if we think about public interest capacity as the ability of the state to generate classic public goods with widely distributed benefits, the Trump administration is largely failing. It is not an exaggeration to say that in order to facilitate a tax cut, your government added administrative burdens to the safety net so that both administrators and the public will spend more time and energy to get less and less from a dysfunctional system. They see value in breaking rather than fixing public services.

Public interest capacity matters to the credibility of democratic systems. The ability of government to competently help the public broadly reinforces liberal democratic systems. Historically, states with administrative capacity centered on merit and impartial public services are less likely to experience democratic breakdowns relative to governments driven by cronyism. The failure to deliver basic public services appears to fuel anti-government populism, even if the populists who benefit end up delivering worse services because they don’t take governance seriously.

The capacity to build digital services

The second capacity point is that Direct File showed us that government can change how it builds such services. The proof of concept that they offered was not just about tax reporting, but about different ways of operating. One team member said: “Direct File was also demonstrating a way of working to the rest of government. This was bringing the US government into the 21st century.”

For people who study or have worked in government there is an emerging conventional wisdom that current modes of operating simply do not work: they are dominated by contractors, cost too much and often don’t deliver. Making it worse, governments traditionally structure contracts in ways that did not allow iterative improvements, and do also not have internal capacity to manage complex tech products built by its contractors.

A member of the Direct File Team described the choice between building internal capacity and

the traditional IRS way: we create a bunch of requirements and we toss it over a fence to a vendor who then builds something and comes back to us in six months, and we see what the product is. We argued pretty hard that that’s not the approach we should be taking. And in fact, we should be building it ourselves in-house…to show that you can build good government technology that doesn’t have to be built by vendors, that it could be built internally within an agency…showing that you can actually transform how this agency works, helping to drive larger changes that would have impacts beyond just Direct File and could help modernize some of the processes and the ways of working.

There is also a better sense of what government should do more of: take ownership of more products in-house, led by smart technologists who actually know how to build, and constantly iterate and improve. Jen Pahlka has labeled a product approach. Direct File represents what is the most visible and ambitious version of the product approach I’ve seen in American government.

Merici Vinton, who helped to lead the Direct File project wrote:

The approach we took to Direct File is the blueprint for future government delivery — multidisciplinary teams, focused on a clear user outcome, delivering in an iterative manner — and while our project might be dead now, you can’t kill a good idea whose time has come.

Private contractors will still play a role in the product approach, but they cannot be in the driver’s seat. For public-facing products in particular, the building needs to incorporate human centered design — constantly looking at user experience, finding and removing pain points. One Treasury employee contrasted working on another project with contractors where:

if you wanted to improve the product, it was this awful game of telephone where you're like totally dependent on, on like a chain of telephone to these contractors who like, may or may not be an A team….They're not really that accountable to you or to the ultimate mission. And it's also just like a kludgy way of operating. Conversely, like the promise of Direct File was to make it a center of gravity not just for itself, but for modernizing the IRS more broadly and for pulling in other parts of the tax system that also needed to be brought into the tax 21st century.

The hollowing out of state capacity

This will be the first tax season after more than 28,000 IRS employees were fired or left, more than 1 in 4 employees, and wants to cut 4,000 staff. Secretary Yellen pointed to the negative effect on IRS capacity: “There is no more money for the IRS. DOGE came in and ruined it.”

After adjusting for inflation, the IRS budget is 40% below its 2010 level, even as it has taken on more responsibilities and serves more taxpayers. And to be clear, this does not save America money: it blows a hole in the deficit. According to the Yale Budget Lab:

The IRS reductions from funding and layoffs have likely resulted in about $861 billion in decreased revenue. The layoffs from actions like DOGE alone have likely resulted in $597.8 billion in decreased revenue over the 2026-2035, and the clawback in IRS funding of $20 billion has likely resulted in $262.8 billion in decreased revenue over the same period.

It remains to be seen how much of an effect this downsizing will have on tax services, but when you are asking HR staff to answer customer service hotlines, the expectations for world-class service will be low.

The cuts had had a very clear effect on Direct File team, which was dismantled, resulting in many skilled technologists leaving government. In all, IRS eliminated about 40% of its tech staff, mostly IRS tech employees who worked outside of Direct File.

One former IRS employee told me that the DOGE official who killed Direct File expressed surprise that people were quitting, asking the team “Who’s gonna do the work?” Because the Direct File team were generally new IRS hires, they were probationary, and so had little job protections, with a widespread expectation that they would be fired as part of broader elimination of probationary employees pursued by DOGE. “No no one ever bothered to talk to them or reverse that sentiment.”

Many who worked on Direct File drew a sharp contrast with DOGE and their approach to building tech products. One point of distinction was DOGE’s seeming disinterest in public interest goals and of the public itself: “if you do not think government has a responsibility to serve people, I think it draws into question how good are you going to be at making government work better for people if you just don’t believe in that underlying principle”

Towards an attitude of wisdom about government reform

Every year I have my students read Karl Weick on sensemaking. One thing I hope sticks with them is what he calls an “attitude of wisdom”: the ability to combine knowledge and doubt, humility with a desire to challenge the status quo. Wisdom involves experience, but also questioning and self-questioning, a willingness to be wrong and adapt.

I thought about the attitude of wisdom when I interviewed Direct File team members. They had to understand the intricacies of the tax code as well as build a good digital product, and noted the tendency of DOGE to underestimate the challenges of managing public services. One put it like this:

I think that anyone who comes into government and wants to make change goes through a period of, like, understanding just how difficult it is to accomplish and just how just what the scope and scale of complexity that you’re dealing with…someone should patent a law that states that the easier a project looks in government, the harder it is to do in reality. The justification being: if it looks really easy and common sense, it means that 20 people have already thought about it and tried it before you, and they’ve all failed because it’s still around…They don’t know what skills are necessary to do this work because I don’t think they understand the complexity of the work. And meanwhile are creating an environment where the people who do know how to do that work either are being forced out by being fired or are not interested in sticking around.

The leads me to propose a paradox of government reform: the simpler a potential change appears, the more likely that it has not been implemented because it features deceptive complexity that others have tried and failed to resolve.

It is also the case that there are often lots of well-pickled myths that contribute to government sclerosis: we can’t do something because we assume we can’t do something, or because one overly-cautious lawyer said we can’t do it years ago.

It can be both true that outsiders underestimate the degree of complexity in changing government, and that insiders overestimate the barriers to change. Sometimes people in government resist change because of an attachment to the status quo, or because the have not tested alternatives. And sometimes they waste a lot of time not learning from failed reforms of the past, repeating the same mistakes.

And this is where an attitude of wisdom would help is knowing the difference between these two scenarios. It is knowing to ask what were the original statutes or rules that generate the constraint — maybe they actually don’t do so. And sometimes it is trusting employees that constraints are real. How do you know the difference?

In the case of the Direct File employees, I don’t think anyone could reasonably accuse them of being attached to the status quo. This is why their testimony of the need to master the complexity of their policy setting feels so striking and credible to me. Maybe DOGE will get around to building great things, but they haven’t so far, even as they pushed out the people that were. They were impatient, and assumed they knew better.

Can we have nice things?

Yes, we can have nice things. We can have a government that builds cool stuff that helps the public. The thing I’ve heard again and again from tech workers in government is the tech is the easy part. Changing the systems in which they work — including procurement and personnel systems — is the bigger lift.

Fixing a political economy where politically powerful actors want them to fail may feel impossible, but the Direct File team gave those actors fewer fig leaves to hide behind. They made the barriers to public interest capacity clearer. One described their hope that Direct File would return in the future:

my hope is that we work to raise the expectations of the political class that make those decisions and that we are fully empowered to do that and that there’s enough of a mass of people to do it as well… that we have really completely reimagined organizations that are just like able to deliver with ambition and scale of Direct File.

A relatively simple account of Direct File is that one presidential administration dedicated political resources and commitment to building a state capacity in tax reporting, while another opposed it. A more sophisticated account considers the role of private interests. The two arguments can be combined: political commitment to public interest capacity was needed to overturn private interest opposition, but the capacities produced are vulnerable to removal if political support is not broad-based.

Long-term private opposition to a public tax option was able to take advantage of a shift to an administration run by officials who doubted the value of public interest projects, and ended the Direct File after consulting with private sector opponents and with little attention to feedback from the public. The Direct File team built a great public interest tool. Then the CEO of an Andreessen-Horowitz-backed start-up, his hedge fund boss, and the private companies that want to kill competition for their products got together and ended it.

Back to the Free File Future

What happens now? Democrats are pushing to reinstate Direct File, and canny Presidential hopefuls might float it in their campaigns to offer a “get shit done” contrast to Trump. The contrast is easy to make, because the Trump administration’s best idea is to go back to the future.

Last year the One Big Beautiful Bill Act directed IRS to look for private partnerships to provide free filing. Folks, we tried this already. It is called the Free File partnership. It has worked poorly for the very good reason that Intuit and other companies ensured that the product was lousy, and used it to redirect customers to their fee-charging products. Indeed, Intuit repaid $141 million to low-income customers for charging them for services advertised as free, and left the Free File partnership after IRS dropped a non-compete clause that blocked Direct File.

Secretary Yellen pointed to the problems of working with the private tax industry as a motivation for Direct File: “And I think the experience maybe we’d had in the past with Intuit or others developing software that the IRS then didn’t control was a negative past experience that suggested doing it in house would be a better way to go.”

And this is the Trump administration solution? Because in America we can’t do the obvious, proven thing to fix problems. We must replay past mistakes, at the public expense, because it is in someone’s private interest to do so.

P.S. Viktor Orban lost his bid to hold power in Hungary yesterday in such a landslide than any attempt to contest the election was not feasible. I wrote in 2024 about how Orban had eroded democracy in Hungary. Pour one out for the right wing intellectuals who Orban paid to boost his political standing and bring these ideas to America: they just lost a generous patron.

While never mentioning Trump or political parties, the show’s second season offers an almost-real time commentary on how policy changes are affecting health — including ICE agents arresting an ER nurse, a MAHA subplot involving turmeric poisoning,1 rural hospital closures, research grants cancelled, and chronic burnout faced by ER workers, our last line of defense in a broken US health care system.

It also provided a window into what will cause millions of Americans to lose their health insurance– as Trump’s One Big Beautiful Bill rolls out in 2027.

As last week’s episode unfolded, an adolescent boy is brought into the ER having trouble breathing. He has asthma. His frantic mother explains that she lost Medicaid because renewal forms were sent to her old address, and not forwarded. The family is on the hook for $400 a month for the most affordable medication, but that still does not allow them to purchase the ones he was prescribed. Meanwhile, she is stuck in an administrative rabbit hole — tracking down tax returns and wage documentation that is hard to get given the gig nature of her work — in order to restore their coverage.

This one boy’s story represents a huge problem with Medicaid, one that is – by design– going to cause the largest loss of health insurance coverage in US history. As states begin to implement new Medicaid work requirements in January of 2027, the paperwork hurdles, which already cause chaos for millions of Americans, are about to get a lot worse.

Is this realistic? Yes. Larry Levitt, the Executive Vice President for Health Policy at KFF explained that they had consulted with the show to get the details right while also proposing some other storylines The Pitt could adopt.

How bad is it?

Losing Medicaid coverage–among those still eligible–is a longstanding problem. In The Journal of Economic Perspectives, we provided a specific example:

By the age of five, Abel Sewell had survived cancer and was receiving monthly blood tests to ensure his leukemia had not returned. The tests were covered by TennCare, Tennessee’s Medicaid program, until, at a regular doctor’s visit, his mother discovered that the coverage had lapsed. Abel’s mother spent months fighting to get their coverage restored, ultimately taking out a second months fighting to get their coverage restored, ultimately taking out a second mortgage on their home to manage the mounting health care debt that came from being uninsured. The family very much wanted health services, and were willing to endure significant hardships to get it. Between 2016 and 2019, almost 250,000 children in Tennessee lost coverage. What happened? The Sewells, like many others, said they never received the TennCare renewal forms. Even those who did receive the renewal packets often struggled to complete the 47 pages. Failure to return forms accounted for 67 percent of those who lost coverage, and it seems likely that many of them did not receive the forms due to outdated mailing addresses.

These types of administrative burdens significantly reduce health insurance coverage. It’s incredibly common for people to get kicked off Medicaid–and then to quickly reenroll. For example, among children in 2018, nearly half of those losing coverage regained it within a year—and most were likely still eligible when they lost it.

People need to be aware when it’s time to renew their benefits, and also understand how to actually get through the process. Similar to the mother in the Pitt, because low-income individuals tend to move a lot, beneficiaries often don’t receive notifications that they need to renew.

Knowing you need to renew is just one part of what makes the process so difficult. A recent survey of Medicaid beneficiaries found widespread problems. Almost 60 percent of people renewing their coverage reported problems such as long waits for help, confusion over which documents to submit, and state agencies losing submitted forms and documentation.

There is no shortage of horror stories. Florida was sued by Medicaid beneficiaries who had their coverage terminated, but received notices of that termination which were impossible to decipher. One mother of two received a notification of coverage loss for her entire family, which buried the notice on page 8 of a 12 page document. Her 2 year old, with cystic fibrosis, needs constant medical care. A review of her recorded calls to a help-line noted that she received “conflicting, confusing, and inaccurate information.”

In perhaps one of the more dystopian cases, a blind woman lost her Medicaid coverage because the state sent digital notices to her, instead of her caregiver who handled her paperwork. She couldn’t read the notice. She was one of many eligible beneficiaries who lost coverage, including many postpartum mothers whose insurance is supposed to be maintained for 12 months after the birth of a child.

The coverage loss not only has implications for beneficiaries, it has large implications for overall program costs. People moving on and off the program substantially increases administrative costs, since new enrollments are more onerous than renewing existing clients. By one estimate, churn costs between $400-$600 per person.

Share

It’s going to get a lot worse

The problem is only going to get worse. Trump’s One Big Beautiful Bill, which added work requirements to Medicaid, will be rolled out in January of 2027.

While on the surface, work requirements may seem benign, the reality is that it will tie people up in paperwork that will lead to massive coverage loss. The details of work requirements also do not reflect the reality of a gig or service economy, where hours fluctuate, and documentation might be sparse.

Almost 90 percent of those subject to the work requirements are either exempt or are working. Current projections are that most of the people who will lose coverage will, in fact, still be eligible. In addition to figuring out how to document their employment, a complicated task for people with work lives dominated by unstable employment, multiple jobs, and gig work, beneficiaries will need to recertify twice a year, instead of the current practice of once a year. Prior evidence points to the damage that will result, both in the administrative costs to process these applications, as well as the increased risk that people will erroneously lose coverage.

The Trump Administration is undermining the most feasible solution

There is a way to reduce the coverage loss that comes during renewal processes. States are legally require to attempt automated (ex parte) renewals by drawing on administrative data — such as state tax records — that can verify eligibility. Instead of filling out forms and providing documentation, the state can simply verify your eligibility, automatically reenroll you, based on data they already have on hand.

Indeed, by July of 2024, over half of Medicaid renewals were effectively automated. This compared to less than one-quarter or renewals just a year earlier. The significant improvement was instigated by a series of reforms and capacity supports implemented by the Biden Administration.

The use of these automated processes has the potential to offset significant amounts of coverage loss associated with the new work requirements. But the Trump administration has already made it harder for states to automate renewals by rolling back a series of state administrative flexibilities. Between June and July of 2025 alone — when many policy flexibilities started to expire — ex parte renewals dropped by 6 percentage points.

Conservative organizations have started to target automated Medicaid renewals, arguing it’s “fraud by design.” Conservative states are responding, proposing laws to prevent or limit their Medicaid agencies from doing automated renewals, including bills in Texas and Louisiana.

More broadly, the Trump Administration’s focus on fraud, including in the Medicaid program, will incentivize states to make application and renewal processes harder. The unprecedented withholding of $259 million in Minnesota’s Medicaid funding, despite no meaningful evidence of large scale fraud in the program, is a warning to other states. Indeed, the head of CMS, Dr. Mehmet Oz, has since threatened other states, including Maine, California, Florida, and New York.

We don’t make everyone who gets help paying for their insurance face these burdens

If you’re tempted to think, “Well, it’s not ideal, but sometimes we have to exert a bit of effort to get help,” you might want to consider how most private insurance works. Americans with employer-based health insurance receive large tax subsidies to reduce the costs of that coverage, comparable in scale to spending on Medicaid and CHIP, but don’t face equivalent administrative burdens. Indeed, the role of government is so invisible that most people don’t even realize that it’s subsidizing their insurance in the first place.

If you have employer based coverage, imagine engaging in a process similar to that of Medicaid beneficiaries to obtain your employer based subsidy–complicated applications, confusing requirements for documentation, and hit or miss communication when you need to renew your benefit annually.

What’s coming in January 2027

As states stare down the January 2027 deadline to implement Medicaid work requirements, there is growing concern about their capacity to actually implement them —and the consequences for the public if they fail.

Even with well designed administrative procedures, it will be very difficult to implement work requirements without coverage loss among the eligible. But states are making clear they are struggling to meet the current timeline — and the federal government is doing very little to help. They still haven’t released guidance to states regarding the specific rules they will need to follow to implement them, leaving states guessing, even as the deadline rapidly approaches.

The consequences for the public will be devastating. The story dramatized on The Pitt will become an all-too common occurrence across the American health care system.

As Supreme Court Justices debated whether Trump’s claim that the plain text of the Constitution grants people born in America citizenship, it represents just one more episode in a shameful history of demonizing immigrants. The attitude of the Justices at oral arguments suggests that Trump is destined to lose, despite his unprecedented decision to appear in the courtroom. But the fact that the case went so far despite the plain text of the constitution, ample precedent, and the Fourteenth Amendment framers’ intent, is astonishing.

In defending Trump v Barbara, the case about the constitutionality of Trump’s Executive Order stripping birthright citizenship, Solicitor General John Sauer’s case opens with: “The ‘main object’ of the Citizenship Clause was to grant citizenship to freed slaves and their children, whose allegiance to the United States had generally been established through generations of parental domicile.”

In the oral arguments, Justice Alito went further, suggesting that illegal immigration was not a problem at the time of the Fourteenth Amendment, implying that the lack of a historical record discussing the exemption of immigrants from birthright citizenship is not a problem for Trump’s arguments. Justice Gorsuch stated that there were no immigration laws in the 1860s.

These claims are historically inaccurate.

The Fourteenth Amendment framers meant also to clarify the legal citizenship of free Black people. The under-defined citizenship of the large population of free African Americans before the Civil War had made this group vulnerable to a range of restrictionist state-level migration laws and other forms of racial discrimination. While policymakers of another era may not have thought about immigration in the same way they do today, they were clearly determined to limit the internal migration of free Black Americans, laws that the Fourteenth Amendment helped to wipe away.

In oral argument, the government and the ACLU had different interpretations of the original public meaning of the birthright citizenship clause. Even if Trump’s executive order is defeated, it provides us with an opportunity to revisit the origins of birthright citizenship. While Sauer emphasized an “intent to remain” as a key feature of citizenship, for Black Americans their ability to move around was circumscribed by state laws. Indeed, citizenship was extended to all Black Americans because their internal migration and settlement was previously restricted by state laws before their legal citizenship was clarified.

Today, Americans witness the human misery of the federal government’s ability to run mass migrant detention centers and to commit due process and equal protection violations of immigrants and citizens. The scope and scale of immigration enforcement is enabled by the US government’s control of immigration and the Department of Homeland Security’s eye popping $178 billion dollar budget. This runs contrary to an older and mostly forgotten history of when the federal government did not always control migration. My book Migration and the Origins of American Citizenship—African Americans, Native Americans, and Immigrants explores the history of migration before the federal government took over from the states in 1888. What form did state migration laws take and what was it like for the people who had to live under these regimes of control?

My research on immigrant experiences spans the colonial period until 1888, identifying the “administrative burdens” they encountered on the way. The use of these administrative burdens is also a clue about how state capacity was used. States with limited administrative capacities used burdens to create pretexts for racial profiling and harassment of politically disfavored groups. This stands in contrast to today, where the federal government is investing in massive new capacities to the same end.

Historical Colonial and State Migration Laws

The US has never been a nation with open borders, even before there was a United States. Phalanxes of laws were on the books as early as the colonial period to recruit and import scarce labor and to ban unwanted migrants (for example Quakers, Catholics, convicts). Other laws were in place to regulate the reluctant labor of European indentured servants and, later, enslaved Africans by requiring paper passes to police their movements. Colonial laws were then carried forward into the US after the founding and replicated by the states in the early republic to the late nineteenth century.

Before the control over international and interstate migration transferred to the federal level in 1888, nineteenth century restrictions on the international and internal mobility and settlement of free Black people was locally managed. Slave and free state governments had laws on the books restricting the mobility and for their ability to stay in the place of their choosing of free Black Americans. Lacking the manpower and financial resources to round up and deport the unwanted population of free Black people, both slave and free states passed laws that created administrative burdens to prevent the formation of large free Black communities.

These racially discriminatory laws were responses by state governments to the increased manumission of enslaved people in the early national period, and their rejection of the formation of large free Black communities. States legislatures claimed that free Black people were incapable of economic self-sufficiency and more prone to criminal activity. Slave states did not want a large free Black population forming that might interact with the enslaved population leading to possible insurrections.

Sub-national management of migration was the norm for a century after the US founding until as late as 1888. Why did that configuration of power last that long? There were decidedly political reasons why states wanted to keep migration policy for themselves. Colonies and then northeastern seaboard states were collecting revenue from head taxes or bonds on poor, sickly, or disabled people who they believed could not economically take care of themselves and would become burdens on their communities. Colonial era laws taxed heavily sick, disabled, or poor people. Those ancient laws were the forerunners of what US immigration law today, which contains a ban based on one’s likelihood to become “a public charge.”

Slave states loathed the idea of federal control over migration because they feared the possibility that the President and national government could deport all their enslaved property. Before the Civil War, any changes made to state laws about voluntary migration would also mean changes to slavery laws as states had regionally-specific reasons for preserving migration laws for themselves. The Supreme Court repeatedly upheld state migration laws based on the legal doctrine of state police powers, which gave states the authority to manage policy areas having to do with health, safety and morals.

In the first three decades of the nineteenth century, free African Americans’ interstate mobility became more difficult as slave and free states passed interlocking laws to restrict their internal migration and settlement. These laws were a blend of outright bans on new Black migration into the state, exile laws mandating African Americans who became free to leave the state (and not return) by a deadline or be re-enslaved, or statutes mandating compliance with administrative burdens to simply stay in a state.

Historian Samantha Seeley explains that Illinois, Indiana, Iowa, Ohio, Michigan, and Wisconsin attempted “to form a bloc that would not accept African American migrants.” Some of the legal requirements for Black people to stay were grafted onto vagrancy and poor laws, whose main purpose was to prevent the legal settlement of indigent people who might drain public coffers. These laws existed at a time when there were no government safety net anti-poverty programs. The additional intent of poor and vagrancy laws, though, was to minimize formations of large free Black communities.

Share

“Show Me Your Papers” Laws

Administrative burdens can be racialized. One way that free and slave states alike policed free African American’s mobility was to require them to produce paper documents to prove their right of presence. The requirement of paper documentation was a way during the existence of legalized chattel slavery that white people could distinguish between free Black people and enslaved ones.

Because of the need to police and surveil all African Americans’ mobility and settlement, Black people’s liberty could be as fragile as the pieces of paper they were forced to carry. At a time when there was no xerox or digital technology, free African Americans transported around their former enslavers’ wills manumitting them, affidavits of clerks from slave jurisdictions verifying their freedom, and similar documents. Few had birth certificates. These papers could be lost, stolen, damaged, or simply disbelieved by white people with power. Most white people faced no such burdens, although poor people (of any race) were sometimes required to carry passes and faced heavier scrutiny if they moved around too much or tried to remain in a place where local authorities did not deem them to be legally settled.

Administrative Burdens to Stay in a Jurisdiction

Today we assume the legal right to enter a jurisdiction also means the right to stay there for a time. Before the Reconstruction Amendments, the right to mobility and ability to remain were two different rights—both controlled by state laws.

Slave and free states required that free African Americans go to the local clerk’s office to register themselves and bring proof of how they came to be free in order to legally remain in their jurisdictions. Other states required the payment of exorbitant fees, for example “from $500 to $100” in Virginia before 1793 to stay. Additionally, places like the District of Columbia in 1821 had a law requiring a “license” to stay in addition to producing three “respectable white inhabitants” to verify a free Black American’s financial independence.

As historian Kate Masur writes, “the nation’s internal borders were far more significant for Black people than they were for white people.”

With a crazy quilt of exile and settlement laws, many states used their limited power to restrict Black mobility and their ability to remain. Herd and Moynihan define administrative burdens as the costs that people encounter when they search for information and public services, comply with rules and requirements, and experience the stresses, loss of autonomy, or stigma that some from such encounters. Lacking the administrative capacity to deport a large free African American population or to surround the states’ borders with border patrol agents, states instead use administrative burdens to hamstring their mobility and their desire to set roots or to move around to pursue employment opportunities.

Although there is little evidence that these laws were widely enforced, they provided a pretext for racial profiling and the harassment of all Black people. James Forten, an African American man born free, wrote in Letters from a Man of Color, on a Late Bill Before the Senate of Pennsylvania in 1813 that he objected to state registration and document production requirements for Black people. He argued that such laws would provide the police and slave patrols with excuses “to apprehend any black, whether a vagrant or a man of reputable character who cannot produce a certificate that he has been registered.”

States thus used administrative burdens to make the lives of Black Americans stressful and to cast suspicion on their mere presence. Legislators’ ultimate hope was that African Americans would not remain.

It was not until the Civil War concluded and the Reconstruction Amendments were passed that immigration transitioned to the federal level and immigration laws came to be nationalized. With slavery invalidated by the Thirteenth Amendment, slave states stopped fighting with the federal government for control over international and internal migration.

The end of the Civil War also led to Supreme Court invalidations of northeastern state laws restricting poor, disabled, and sickly migrants. The decoupling of slavery from voluntary migration laws enabled the Supreme Court in a series of cases to strike down northeastern state migration state laws it had previously routinely upheld.

Northeastern states now feared that they would be overrun by people who would become economic drains and started lobbying the US government to assume control over immigration. The federalization and nationalization of immigration controls would start in 1875 and not be completed until 1888.

All the state laws restricting Black mobility and settlement were possible before the Reconstruction Amendments because all African Americans’ (including native-born ones) legal citizenship was under-defined. Although northeastern states claimed Black men as state citizens, the hierarchy between state and US citizenship was unclear.

And here we return to birthright citizenship. The Fourteenth Amendment’s birthright citizenship clause not only states unequivocally that African Americans are US citizens but as national citizens they are entitled to due process and equal protection of the laws.

The Reconstruction Amendments did not come about because of white noblesse oblige. African Americans had long fought back against racist laws that were enabled by their under-defined legal citizenship. They mounted grassroots resistance by organizing through networks of Black churches, newspapers, and conventions to challenge their oppression.

As the Reconstruction Amendments invalidated slavery and confirmed the US citizenship of Black Americans, the state-level migration restrictions were invalidated. Today, immigration controls have transferred to the federal government, with far more administrative capacity than the states ever had.

The federal assumption of immigration power has not meant more freedom. Although grassroots activism by immigrants and their supporters are underway, the challenge to overturn unjust immigration enforcement is larger given the sweep and orbit of federal power now.

Federal Control of Immigration Today

The cost of transitioning immigration controls to the federal government can be measured in profound human suffering. We know today that the Department Homeland Security has, as of February 2026, 68,289 non-citizens in detention. The Transactional Records Access Clearing House that obtains data directly from the federal government, also reports that of those in detention, 73.6% have no criminal convictions. (A violation of immigration law is civil in nature.)

More disturbing than the sheer number of people in immigration detention are the squalid conditions. Sometimes migrants are held up to a year without having the opportunity to see an Immigration Judge. This network of federal migrant detention camps have resulted thus far in 2026 in 11 deaths. In 2025, 31 deaths which was “a two decade high” died in Immigration and Customs Enforcement (ICE) custody.

Lest US citizens believe they are immune to DHS’ dragnet; there are disturbing reports of ICE using technology called Mobile Fortify to verify people’s immigration status with facial scans. The technology, which was never authorized by Congress to be used in the interior of the US, can result in wrongful arrests.

Like free Black people before the Civil War who carried the burden of proof to demonstrate that they were not enslaved people to any white person who might ask, the burden of proof today has been borne disproportionately by US citizens and immigrants who are people of color as ICE aggressively tries to meet daily arrest quotas.

ProPublica’s investigation found at least 170 US citizens who had been assaulted and/or detained by ICE. Nicole Foy writes, “Immigration agents do have authority to detain Americans in limited circumstances. Agents can hold people whom they reasonably suspect are in the country illegally. We found more than 50 Americans who were held after agents questioned their citizenship. They were almost all Latino.”

Today we are told by the former head of the Department of Homeland Security that US citizens should be prepared to show proof of citizenship in the interior. The US has returned to a prior era as a “show me your papers” state, selectively enforced against people of color. The past is not quite past, but the form of government power has taken on new dimensions.

Dr. Anna O. Law is the Herbert Kurz Chair in Constitutional Rights at the City University of New York, Brooklyn College. She researches, writes, and teaches in US constitutional law, US immigration policy history, legal history, and American political development. She is the author of Migration and the Origins of American Citizenship—African Americans, Native Americans, and Immigrants

The platform is Polymarket. Its closest competitor, Kalshi, operates on the same basic premise: users bet real money on real-world outcomes, from sports results to economic indicators to elections. Together, the two platforms have become the dominant players in an industry that didn’t meaningfully exist for ordinary Americans five years ago.

Increasingly, these prediction sites seem to be everywhere (including on Substack). So who is regulating them? In theory, the answer is the Commodity Futures Trading Commission (CFTC). It is also, at the moment, a commission-in-transition with only one member. While a fully functioning CFTC normally has five commissioners, the agency asserting exclusive jurisdiction over a multi-billion dollar industry is, currently, a single person.

Polymarket launched in June 2020. Less than two years later, the CFTC fined it for operating without CFTC registration and blocked it from operating in the United States. The platform spent the next three years operating offshore, officially closed to American users. Then, in October 2024, a federal court ruled against the CFTC in Kalshi v. CFTC, finding that the agency had improperly blocked election betting contracts. Polymarket moved quickly and in July 2025 it acquired QCX, a CFTC-registered derivatives exchange, accelerating its return to the U.S. market by leveraging an existing registration, with official re-entry secured by late 2025. A U.S. app soft-launched via waitlist in December 2025.

Around the same time, Donald Trump Jr. joined its advisory board. (He is also a strategic advisor with Kalshi.) Under such circumstances, few believe the new entities will be subject to close regulatory attention from Trump’s government, creating an incentive for states to jump in.

The logic behind prediction markets is unlike a poll or a pundit. Instead, the bettor has to publicly put money on the line about their expectations and will lose hard cash for being wrong. This should incentivize accuracy, but only if one assumes the money on the line reflects genuine belief rather than privileged access.

Evidence that it may reward privileged access to information is accumulating.

On Polymarket, eight recently created accounts, all made around March 21, collectively wagered close to $70,000 on a ceasefire occurring. If a deal is struck before March 31, those accounts could walk away with over $800,000 in winnings. A separate account, opened just before the US attacked Iran on February 28, placed a successful bet predicting those strikes as well. That same account has made no other wagers.

Kalshi also wants you to know it won’t engage in disquieting, illegal or corrupt behavior because of federal regulation!

But death markets are occurring, and become all the more likely in times of conflict.

One concern is whether bets involving the use of military force might encourage bets by well-informed insiders. A user pocketed over $400,000 on bets placed before Nicolás Maduro was taken from Venezuela, another wagered on Ayatollah Khamenei’s removal from power before his death in Tehran. Here, it appears the market did not aggregate crowd wisdom. Rather, it rewarded someone who appears to have known something the crowd didn’t.

Legislators have been approaching the problem from two directions simultaneously. The End Prediction Market Corruption Act, introduced in early March, targets what can be traded. It would ban prediction market contracts on sports events entirely, essentially siding with the states that have argued these are unlicensed gambling operations. The PREDICT Act, announced only a couple weeks later, by a bipartisan group of lawmakers, targets who can trade: it would prohibit members of Congress, the president, vice president, and others in the executive branch from participating in prediction markets involving political events.

In response to these government actions, both Kalshi and Polymarket announced new rules aimed at banning insider trading on their platforms. For instance, Kalshi announced that regulations would include banning political candidates from trading on their own campaigns and prohibiting athletes from trading contracts related to their sport.

But the category of people with relevant non-public information extends well beyond elected officials and professional athletes. Coaches, military officers, intelligence analysts, diplomatic staff, campaign workers, bankers, regulatory officials – anyone with advance knowledge of the outcomes being traded on is a potential insider. How either platform intends to identify, monitor, or enforce rules against that universe of people remains unclear.

The question of who shouldn’t be trading on these platforms is easier to answer than the question of whether anyone can trust what the platforms are actually showing. When prediction markets are cited as evidence of what the public believes, and the odds are being moved by actors with non-public information, the “future of news” starts to look less like a forecasting tool and more like an information asymmetry problem with a financial instrument attached. For example, Polymarket calls itself “News 2.0” despite pumping out misleading content.

Before stepping down in early 2025, outgoing CFTC chairman Rostin Behnam declared the agency was well positioned to be the regulator for “digital commodity assets.”

But is it?

The problems outlined above, insider trading, manipulation, and macabre markets, are not simply enforcement failures. They are symptoms of a federalism problem: a new industry that arrived faster than the laws written to govern it is now caught between a federal regulator claiming exclusive authority and states insisting it is theirs to police. Neither framework was designed with this in mind.

Share

The Lineage of Prediction Markets

In a recent Wall Street Journal piece, current CFTC Chairman Michael Selig wrote:

The Commodity Futures Trading Commission for decades has overseen regulation of prediction markets — or event contracts, as we refer to them — that help market participants hedge risk, aggregate information and test hypotheses about future outcomes.

Prediction markets have existed since the late 1980s. The Iowa Electronic Markets (IEM) began in 1988. It was founded by three University of Iowa professors who wanted to know if they could predict polling data better than pollsters. Known as the "Iowa Political Stock Market," it was originally created for the 1988 Presidential election. From the start, the platform ran into legal friction with gambling laws and securities regulations prompting it to restrict participation to the University of Iowa community.

Ordinarily, futures contracts fall under CFTC jurisdiction via the Commodity Exchange Act (CEA). The IEM was no exception in theory. In practice, the CFTC chose not to regulate it due to its academic focus and the small scope. Instead, it issued a series of no-action letters, a formal mechanism by which the agency signals it will not pursue enforcement action against a particular entity.

In 1992 and 1993, those letters permitted the IEM to expand beyond the University of Iowa to other academic institutions, but under strict conditions: participation was capped at 2,000 traders per market, with a maximum investment of $500 per participant. The CFTC wasn’t overseeing the IEM. It was formally looking the other way, on the condition that the platform stayed small and academic.

It still operates today, still capped at $500 per participant, still run as a nonprofit academic research project by the University of Iowa, and still operating under a CFTC no-action letter rather than normal registration. Its director, University of Iowa professor Thomas Gruca, describes its purpose in terms that would be unrecognizable to Polymarket’s investors:

The purpose is not to make money, we’re nonprofit, we’re even non-revenue. Our purpose is to see if different groups of people can put their heads together and solve really hard forecasting problems.

The professors affiliated with the IEM are aware that prediction markets like Kalshi and Polymarket use them as a justification for existing. Finance professor and Iowa Electronic Markets board member Thomas Rietz has noted that while Kalshi and Polymarket prioritize opening many markets to increase site traffic and betting for revenue, “The difference is we are designing our markets and our contracts specifically to make the most accurate forecast possible.”

In March 2026, Yahoo Financereported that Polymarket recorded a single-day notional trading volume of $478 million, with their politics category accounting for over $200 million of that activity. The IEM has no equivalent figure because there is no equivalent: a $500-capped nonprofit research tool operating under a regulatory carve-out and a billion-dollar commercial platform are not the same thing and were never designed to be governed the same way.

When Michael Selig says the CFTC has overseen prediction markets “for decades,” he is technically correct about the genealogy of these platforms. But he is describing something radically different in scale, scope, and purpose. The original version was a university research project explicitly exempted from normal regulation because it was small and academic. What exists now is a multi-billion dollar commercial platform trading on nuclear war, assassinations, and elections.

The Category Problem: States vs the CFTC

One of the fundamental problems with regulating something like Polymarket is that there is no agreement on what it actually is. The answer determines everything: who has jurisdiction, what rules apply, how users are protected, and whether the platform should exist in its current form at all.

Platforms like Polymarket brand themselves as a kind of digital commodity market offering contracts not gambling. Michael Selig says that prediction markets function the same way as other futures contracts with standardized derivatives obligating parties to exchange assets at prices set on future dates. Hedging against something like weather, the argument goes, is not betting against the house in the way that sportsbook companies do.

The CFTC’s legal authority here is genuinely broad. As per the CFTC:

Under the plain language of the Commodity Exchange Act, event contracts are “swaps.” They are derivative instruments that allow two parties to speculate on future market conditions without owning the underlying asset. In the wake of the 2008 financial crisis, Congress expressly granted the CFTC comprehensive authority over any such contract based on a commodity. The statutory definition of “commodity” is extraordinarily broad and includes practically all goods, articles, services, rights and interests except for onions (due to a history of market manipulation) and movie box-office receipts (because of Hollywood lobbying).

But states argue that prediction markets are a loophole for state gambling regulations, particularly on sports. Several states including Tennessee, New York, New Jersey, Nevada, and Massachusetts have taken Polymarket and Kalshi to court, arguing that event contracts on sports are effectively gambling and are therefore subject to state jurisdiction and taxed differently than financial markets. User eligibility requirements differ as well: prediction markets generally allow users 18 and older, while state gambling is restricted to those 21 and older.

Massachusetts won an injunction against Kalshi, temporarily banning the firm from operating sports-related contracts in the state. In response, Kalshi argued that the state was “trying to block [its] innovations by relying on outdated laws and ideas.” Polymarket filed its own lawsuit against the state, seeking to avoid what it called “imminent and irreparable harm” from the enforcement of state gambling laws against a federally regulated derivatives exchange.

From the CFTC’s perspective, this is not a gambling question. It is a jurisdictional one that states have no authority to be involved in. It has referred to state governments as “overzealous” in their attempts to “undermine the agency’s exclusive jurisdiction over these markets.” The agency has pushed this position in court, filing an amicus brief asserting exclusive jurisdiction over event contract markets. Multiple rulings on the issue have been split, with a federal court in Tennessee leaning toward CFTC jurisdiction while courts in Nevada side with the state.

But the jurisdictional fight obscures a deeper problem that neither side is asking.

In a standard commodity futures market, the underlying asset, such as corn, oil, an interest rate, exists independently of the contract written against it. Political prediction markets are different. When enough money moves on an election outcome, the market doesn’t just reflect political reality, it potentially shapes and reshapes it. Campaigns track them. Donors read them as signals. News outlets and pundits refer to prediction markets as reasonable measures of public opinion and political expectations. That feedback loop is not something the Commodity Exchange Act was written to address. The question of whether a Polymarket or Kalshi-type prediction market should exist at all is still a question to be reckoned with.

Let’s say the answer is yes – how can one be sure they are operating fairly?

These platforms rest on a core premise: that markets aggregate real information through real financial stakes. However, there’s evidence that challenges this.

Researchers at Columbia University found that roughly 25% of all trading on Polymarket over the past three years consisted of “wash trading.” This is an illegal form of market manipulation where users rapidly buy and sell the same contracts to artificially inflate activity. The figure varied by category: 17% of election-related trading, 45% of sports-related trading. It was made possible by Polymarket’s decision not to charge transaction fees, its use of pseudonymous crypto wallets, and its operation on a public blockchain. These are structural choices that, intentionally or not, created the conditions for manipulation at scale. Kalshi does not operate on a blockchain, so its trading data is not available to outside researchers. Whether the same problem exists there is unknown.

Regulatory Momentum, Sort Of

In early February 2026, the CFTC assembled an Innovation Advisory Committee, describing its goal as ensuring “the CFTC’s decisions reflect market realities so the agency can future-proof its markets and develop clear rules of the road for the Golden Age of American Financial Markets.” Alongside executives from traditional financial institutions like Nasdaq, the committee includes executives from FanDuel, DraftKings, and Gemini, as well as executives from Polymarket and Kalshi.

The platforms the committee is meant to help regulate are represented on the committee itself.

On February 23rd, six Senate Democrats issued a letter to Michael Selig urging the CFTC to prohibit prediction contracts that correlate with or incentivize geopolitical outcomes including assassination, death, terrorism, and war. Two days later, the CFTC’s Enforcement Division issued an advisory emphasizing that all trading practices prohibited under the Commodity Exchange Act apply equally to prediction market contracts. On March 9th, Selig announced that he had instructed CFTC staff to prepare an advance notice of proposed rulemaking on prediction markets.

These actions have the appearance of regulatory momentum. But it is worth being precise about what each one does.

The Senate letter asks a narrow question about the most obviously objectionable markets. The enforcement advisory reminds the industry that existing fraud rules apply, which was already true. The advance notice of proposed rulemaking is the earliest and most tentative step in a regulatory process that typically takes years. Meanwhile, the CFTC’s most concrete action to date has been battling in several court cases not to scrutinize the platforms but to ensure that the states trying to constrain them cannot.

That is not incidental. The rulemaking process is not simply reactive enforcement. It is an attempt to lock in federal jurisdiction over prediction markets by defining them as federally regulated financial instruments, with the downstream effect of foreclosing state regulatory claims entirely.

Neither the Senate letter nor the rulemaking process is asking the question this piece has been raising: what does it mean for democratic life that a private platform, backed by venture capital and advised by political insiders, has become the infrastructure through which Americans form probabilistic beliefs about their own political future?

The institution claiming sole authority to govern it has, so far, spent most of its energy making sure no one else can.

Lets see where we are with the shutdown.

• Trump urged the Senate to keep the Department of Homeland Security closed, refusing to offer concessions on Immigrations and Customs Enforcement or Customs and Border Protection, and demanding the Senate pass his mass disenfranchisement bill, the SAVE Act.

• Amidst a growing airport crisis, Trump declared he will pay Transportation Security Administration employees on Thursday, March 26. He does not specify any legal means to do so in a memo released the next day, as a record 12% of TSA agents called in sick. He later claimed the power to pay all DHS agents affected by the shutdown.

• At 2.20 am Friday morning, the Senate passed a bill that would fund all parts of DHS, except for ICE and Border Protection. Most people breathe a sigh of relief. Airport crisis appears to be over, ICE/CBP issues to be resolved later. Senate goes into recess.

• The Senate bill would pass handily in the House…except Mike Johnson does not give it a vote. Late Friday night, the House passes its own bill with Trump’s backing to fund all of DHS until late May, no compromises on immigration. The Senate won’t pass this bill, which would require Democratic agreement. Similar proposals have been rejected for weeks.

• The House also goes into recess. Meaning no-one is around to actually sort out the differences. The shutdown appears set to continue for a couple more weeks at least.

Now remember, ICE and CBP don’t actually need new money. They are funded for multiple years because of the One Big Beautiful Bill Act. Their employees are being paid during the shutdown. This is really about whether a) they will face some nominal accountability for their interactions with citizens and noncitizens, specifically on whether ICE can wear masks and be allowed warrantless entry into people’s home and b) whether other parts of DHS will be funded, with TSA being the obvious pain point.

So here is where the institutional betrayal comes in. Mike Johnson had a pretty clear choice:

A) Reclaim the power of the purse for Congress by passing broadly popular legislation.

B) Allow the President not only to dictate appropriations, but also to engage in a potentially illegal scheme to pay TSA employees to deal with public anger.

Johnson chose B.

We have been here before. During the last major shutdown, Trump claimed that he could decide that members of the military would be paid. There was no major uproar or legal appeal — after all, who is against paying the troops? — but it set a precedent. DHS is sitting on a mountain of money. So Trump could make the same dubious claim that he just needs to move a bit of it around, and the shutdown is over!

Except, that is not how it works. Either the shutdown — a lapse in appropriations means the executive must lay down its tools — is real, or its not. Either the legislative branch determines appropriations, or it doesn’t. The idea that it does is central to our constitutional scheme of government. During the last shutdown, Matt Glassman wrote:

Trump and, especially, [OMB Director Russ] Vought are proposing theories of executive spending that would essentially undo late 17th century English settlements between King and parliament, and upend the 1787 constitutional dynamic. But the ability to not spend money appropriated by law is nothing compared to the ability to ignore purpose restrictions on money. The most bedrock feature of appropriations—indeed, the reason the word appropriation exists—is because the English parliament won the ability to condition their grants of supply to the King on three things: duration, amount, and purpose.

In effect, Trump is making the case that he can both a) impound funds he does not want to spend, and b) spend funds that have not been appropriated. Both of these are incredibly dangerous threats to checks and balances. For example, we have credible evidence that the Trump administration uses partisanship to selectively direct money away from blue states when it comes to contracts, infrastructure spending, safety net spending because of fraud allegations, and even emergency funding requests.

All of which to say, Trump keeps reminding us of why its a bad idea for a single person to control how public money is spent, a point that the framers were especially aware of when they designed the American constitution.

Share

In this specific case, its not that the President cannot pay federal employees during a shutdown, but he has to have a clear appropriations source. He has that for ICE and CBP, no dispute. He does not for TSA.

The Trump memo has not explained what that appropriations source is. It just says “to use funds that have a reasonable and logical nexus to TSA operations to provide TSA employees with the compensation and benefits.” And so, some lawyers at DHS and OMB were probably asked, “well, what is the most plausible appropriations source that we could claim here?” And these are lawyers who were selected for their loyalty to Trump, and so they will come up with something, or write a couple of paragraphs about Article II powers and not get into the details.

The memo raises some uncomfortable questions for Trump, beyond the obvious one of “where is the money coming from?” Another is “if you had this power, why are you waiting until now to pay the TSA employees?” Are we in a country where the King only pays his employees after they walk off the job, are living in cars, or become dependent on food banks?

For the rest of us, the DHS dispute raises bigger questions about whether certain checks and balances still exist. I’m not sure anyone wants to hear this constitutional nit-picking. We all want our airports to be functional again. And Trump redirecting DHS previously-appropriated money to non-ICE and CBP activities might be considered a win for Democrats.

But it extends the DOD shutdown precedent, and that is dangerous. If a President can point to any unspent funds for an agency, he can reappropriate it for what purposes he wants. We are increasingly operating a form of government where a) we routinely have government shutdowns, and b) the President can selectively end shutdowns for parts of the government he likes, while maintaining them for the part he dislikes. That is is a system of government where Congress has little functional power. As Glassman noted this “reduces Congress to a single, negative power: the right to not appropriate any money.”

The Georgetown University Congressional legal scholar Josh Chafetz wrote:

It's not just that unilaterally paying TSA agents is illegal. It's that Trump has now repeatedly undermined the core of the congressional power of the purse. He's not spending money that has been appropriated by law (impoundments) and he's spending money that has not been appropriated.

These actions aren't just illegal, although they are that. They aren't just unconstitutional, although they are that. They are anti-constitutional: they strike at the core of one of the principles that allows our entire constitutional order to function.

From the Founding through the present day, the power of the purse has been understood as one of Congress's most potent tools against the executive: the executive can't do anything without money, and he can't spend money without Congress. If Trump is able to get away with ignoring that, with arrogating the entire power of the purse to himself, then Congress has pretty much no way of checking him.

Also worth noting that, in Plutarch's telling--the telling that the Founding generation would have known intimately--it was Caesar's raiding of the treasury and threatening of the tribune Mettellus, who tried to defend it, that was one of the clearest signs that he intended to become a dictator.

And this is where we get back of Johnson. In a myriad of ways, Johnson has completely failed in his job of protecting the institution he is tasked with leading. There is little meaningful oversight coming from Congress, and Johnson encourages Trump’s grabbing of legislative power.

In this case:

• Trump created a problem via poor management of government services, which triggered a shutdown which worsened other government services.

• Trump offers a likely-illegal and constitutionally suspect solution that further weakens Congress.

• The Senate same up with a solution that would have solved the problem via constitutional means, while reducing the scope of the conflict.

• Mike Johnson signs on for Trump’s solution, rejecting the popular and constitutional option.

Shutdowns are dumb. No other rich country assumes that if politicians are at a budget impasse, public employees and the public should be punished via a loss of paychecks and services respectively. Indeed, such budget impasses typically trigger an election in other countries, giving politicians a real incentive to pass budgets on time.

Shutdowns are not a core constitutional requirement. A government lawyer invented this problem in 1980 through an interpretation of the Antideficiency Act of 1870. Some have argued that another government lawyer could simply offer a different interpretation that shutdowns are not required. I’m not a constitutional lawyer, and it clearly would be preferable for Congress to amend the Antideficiency Act than to leave it to the executive branch. But anything seems better for a functional and constitutional government than the emerging dynamic: shutdowns are routine, and also an opportunity for the President to amass and selectively use legislative power.

A while back ago I had the good fortune to visit Norway to give a talk at Bergen University.

Some takeaways: Beer in Norway is expensive. How expensive? Like imagine President Trump had bombed half of the breweries, and now we are relying on our strategic beer reserve.

I had to sell a kidney in order to damage my liver by purchasing this flight of beer at the cozy Schouskjelleren in Oslo. Totally worth it.

Oslo had a surprisingly active Octoberfest which was in full swing when I visited. This seemed a bit odd, since it was late October, by which time Octoberfest has usually wrapped up, and also, its Oslo.

I talked to some Germans in Oslo, some of whom thought it was all in good fun. One did not, asking if I’d seen the logo, which he said would be considered racist if it was any other group than Germans. And, you know, he’s not entirely wrong.

With a limited schedule my plan was to see the best of Norway in a single day. Which apparently is completely doable with a day trip called Norway in a Nutshell. This takes you between Oslo and Bergen. In theory, you can stop and do overnights along the way, but who has time for that?

I love train travel, so valued the chance to see some Norwegian countryside from the comfort of my seat, even if the weather was a little gloomy…

…and even snowy in parts.

The big plus of the Norway in a Nutshell is that you get not only some scenic train routes, you can also go Fjording, hopping off your train and directly onto a sleek electric boat.

Apparently spring is a good time to go, when these waterfalls are fed by snowmelt.

The Fjords are genuinely cool!

My trip ended in Bergen, where I strongly endorse Pingvinen (below), a gastropub where the gastro actually part actually matches the quality of the beers. Book ahead if going. A local bar that had more actual locals than college students was Folk & Røvere.

Among the souvenirs on offer are Norwegian sweaters. These are expensive (the Dale of Norway run between $200-600). Depending on your point of view, these could either be a) a treasured family heirloom, b) the most expensive gift for clothes moths you could come up with. Given I spent my life savings on a round of beers, I was forced to pass on this opportunity.

There you go: short trip, few insights! Regrets? If I had more time, I would have ventured further north up past the Arctic circle. But we always have to leave something for the next time! Comments are open, so please add your Norwegian tips.

If you made it this far, you may as well check out previous travel writing on the best train ride in the world, Ghent, Utrecht, hacks to manage Italy, the west coast of Ireland, what to do in Portland, ME, and AI-enabled beer drinking in Copenhagen.

If more people support my half-assed travel writing, I promise to stop doing my scrupulous coverage on the demise of the American empire.

Anyway, enjoy more Scandiscrolling:

That, in a nutshell, is what is happening to your federal government right now. A new data point shows just how bad things have gotten:

➡️ Federal employee engagement and satisfaction is only 32% and more than 58% of employees say they are less engaged than they were last year

➡️ Only 7.5% of survey respondents agreed that political leaders generate high levels of motivation for the federal workforce

➡️ Only 1 in 4 employees believe they can report wrongdoing without retribution

➡️ 36.5% of the federal workforce say their work unit provides worse quality services compared to last year

All of these metrics are going not just in the wrong direction but sinking like stones.

Its worth taking a moment to explain the survey before I unpack the results. Every year, the Partnership for Public Service — a nonpartisan nonprofit whose core mission is making government work — publishes a ranking of the best places to work in the federal government, based on employee surveys. Historically, they worked directly with the U.S. Office of Personnel Management (OPM) who collected and shared data from the Federal Employee Viewpoint Survey (FEVS). This occurred consistently across four administrations — including the first Trump term — and was a widely accepted measure of which agencies were functioning well or poorly.

Last August, OPM canceled the 2025 survey entirely, despite it being required by law. Officials promised to bring it back eventually. But it is hard to ignore the timing: in the year of the most aggressive campaign against federal employees ever, the Trump administration simply chose not to ask those employees how they were doing.

The Partnership pressed on. It designed and administered its own version — the Public Service Viewpoint Survey — collecting responses from 11,083 current federal employees. It complemented this analysis with qualitative focus groups of employees, from which I will quote from below.1 The results, released this week under the title Federal Public Service in Peril, are extraordinary.

A staggering collapse in employee engagement

The headline figure is a government-wide Employee Engagement and Satisfaction Index Score of 32 out of 100.2 To put that in context: prior Best Places to Work scores, even at poorly rated agencies, rarely dropped below 50.

This is not a dip. It is a collapse.

Here are the comparisons between 2024 and 2025 for larger agencies. The Department of the Army registered the highest score among large agencies — at just 48% out of 100 — with only 9% saying Secretary Pete Hegseth’s political team generates high levels of motivation. Every other large agency scored lower, some dramatically so. The average score by agency in 2024 is about 70% and 29.5% in 2025 — just over a 39 percentage point decline.

The picture at mid-size agencies is even worse. In 2024, the average score is 73% and in 2025 it is just over 25% — an astonishing 48 percentage point decline.

Elizabeth Linos is the Emma Bloomberg Professor of Public Policy and Management, and Faculty Director of The People Lab at the Harvard Kennedy School, which focuses on government employees and performance. It is hard to think of anyone more qualified to comment. She wrote the following:

By any historical measure, the new data released yesterday by the Partnership for Public Service is documenting the worst employee engagement and workforce sentiment I’ve ever seen for the federal government. To put this into perspective: in a typical year, agencies work hard to get their engagement scores from the high 60s to the mid-70s or even into the 80s (if you’re NASA). This year, the average is 32. No federal agency is a “best place to work” at this point.

A record of leadership failure

If the engagement numbers are at a record low, the assessment of federal leadership is even worse. At Health and Human Services — which had been a top-five agency in prior years — only 2.6% of employees said Secretary Robert F. Kennedy Jr.’s political team generates high levels of motivation. At Treasury, just 6.6% said the same of Secretary Scott Bessent’s team. At DHS, 8.9%. These are not unfavorable ratings. They are near total rejections.

Will any agency leaders will be called before Congress to explain why their workforces feel leaderless and under siege? Probably not. This is a Congress that has shown little appetite for protecting even its own institutional prerogatives, let alone protecting the government workforce.

Republicans have long favored the argument that government should be run more like a private business. It is a fair question to turn back on them: what private organization, having presided over a collapse of this magnitude would be considered well-managed? What board of directors would look at these numbers and declare the strategy a success?

Share

A demoralized workforce is not an effective one

Over the last couple of years, reporters have sometimes asked me how the public will see the effects of politicization and cuts in government capacity. The answer is sort of unsatisfying because it is hard to connect a specific personnel decision to a specific outcome. But I usually say:

• look for measurable declines in metrics of service provision for agencies that have direct interactions with the public (veterans services, IRS, TSA, Social security) and

• gaps becoming apparent during an emergency or after a scandal or undeniable failure.

The Iran war is exposing some of those gaps. The State Department and FBI have shed employees with specialized expertise on Iranian energy and counterintelligence. The Department of Defense gutted the office working on civilian casualty minimization — a fact became newly salient following a U.S. airstrike on a girl’s school in Iran. The National Security Council’s director for Iran, Nate Swanson, laid out exactly how Iran would respond to a bombing attack — unfortunately he was fired last year at the direction of online nutjob Laura Loomer. That expertise might be useful right about now.

Another way of assessing performance is to ask the federal employees if they are able to deliver. While these assessments are subjective, they offer insider knowledge from non-political actors and I think they are credible enough that I’ve used them in peer-reviewed research. And the federal employees closest to the work are telling a very different stories from the Trump appointees who insist that everything is fine.

A decimated workforce is less able to produce. So too is a demoralized one — especially when that work depends on expertise, judgment, and institutional knowledge that is hard to measure and impossible to mandate. From Elizabeth Linos:

And all the evidence tells us the same thing: an engaged workforce is critical to effective service delivery. When engagement collapses, performance follows. Under these conditions, it’s hard to see how we sustain services in the near term—let alone rebuild a functional public sector over time.

The President of the Partnership for Public Service, Max Stier, agrees:

Significant research, including our own, has shown that when employee engagement suffers, our government’s ability to provide essential services to the public declines. No government can serve or protect the public effectively with such rampant dysfunction.

Undermining public service motivation

There is a well-established body of research on what scholars call “public service motivation” — the idea that many government employees are driven not primarily by pay, but by a genuine commitment to public service. In the survey, more than 95% said that that it was important to them that their work contributes to the public good.

That motivation is a powerful predictor of recruitment, retention, satisfaction, and performance. It is also fragile. For example, one of my most cited research pieces pointed out that such motivation declines as employees get older, with the implication being that public organizations need to work actively to maintain one of their very few advantages over the private sector by rekindling the desire to serve.

Another important research finding here is that low person-organization fit weakens public service motivation — as people no longer see the organization working toward its public service mission, they grow disaffected and demotivated. The new survey data provides some hints this is happening. While 78% of employees still say they identify with the mission of their agency only 52% say they feel a strong sense of attachment to their agency.

Over the past year, the opposite has happened. Federal workers have been publicly characterized as the enemy. The Office of Management and Budget director stated explicitly that the goal was to “put them in trauma.” By this metric, the Trump administration is succeeding, but their desire to traumatize the federal public service runs directly at odds with the public interest in a competently run government.

Agency heads have actively opposed the core missions of the agencies they lead. A generation of newer employees has largely disappeared, while decades of institutional knowledge walked out the door every time experienced hands decided they simply had enough.

If you know people who work in federal government, none of this is surprising. The survey results are consistent with their accounts, consistent with the mass exodus of employees, and consistent with the open letters that workers across multiple agencies have published to raise the alarm about conditions inside their institutions.

An IRS employee described work conditions as “The beatings will continue until morale improves.” A Department of Education employee shared:

I’m a combat veteran and if I had to choose to relive this past year or go back to a combat zone, at this point, I think the combat zone would be a lot easier to deal with.

A scared workforce is bad for accountability

Another factor that jumps out from the survey is the growing pattern of politicization. Federal employees were already concerned about whether they can report wrongdoing without fear of retribution. This has gotten even worse. Only 1 in 4 employees now feel their organization doesn’t tolerate arbitrary action or political coercion. Roughly 3 in 4 employees no longer feel safe reporting suspected legal or regulatory violations.

An IRS employee said:

I’ll be honest, [at the] political level, I have minimal confidence that if we reported anything, any misbehavior, that it would be taken care of. In fact, quite honestly, I would fear retribution if we reported an ethical violation of a political level person in the IRS. I would do it anonymously. I’d probably do it through the union. I would not do it through the normal channels.

A Customs and Border Protection employee echoed this point:

I can tell you actively nothing happens [when incidents are reported]. The management above is too scared to do anything. And I mean raise it up, say anything about the incident, have a conference to air out the issue. It’s truly a dead silence, full stop. No one’s going to do anything. And that’s on low-level stuff. On high-level stuff, you might as well be talking to a brick wall for all the answer or response you’re going to get.

These are your employees

In any large organization of millions of people, you will find poor performers. That is true in government as it is everywhere. But the federal workforce, on the whole, is composed of people doing their best under increasingly difficult circumstances to keep the machinery of government functioning. Many of them are extraordinarily talented individuals who could have earned far more in the private sector and chose not to, because they believed in the work.

They are not your enemies. They are not the deep state. They are your employees — funded by taxpayers like you, delivering public services to the people like you. We have a profound interest in ensuring that those employees are capable, supported, and motivated to do their jobs well. What is happening right now is the opposite of that. The data is no longer anecdotal. It is right there: 32 out of 100. And the consequences, when they fully arrive, will not be popular and will not be easy to reverse — regardless of which side of the political aisle you sit on.

You can find a fantastic summary of this issue from my colleague Just Wolfers. Please share widely!

Related from Can We Still Govern?

Leaning into his favorite trope, Trump noted that immigrants “have been cheating for years” when Stephen Miller claimed that “if all of this theft were stopped, it would be enough to balance the budget. The extraction of wealth from American taxpayers to people who don’t belong here is the primary cause of the national debt.”

This is not just untrue, it is a lie on the scale of Elon Musk claiming he would eliminate the deficit. Absolutely fantastical, a kind of fraudulent fraud claim, one untethered from the surly bonds of reason and evidence. The reality, according to an analysis from the libertarian Cato Institute, is the opposite: immigrants have significantly reduced the deficit, contributing a fiscal surplus of $14.5 trillion between 1993-2023.

Trump especially pushed the false claim that Somalis had stolen $19 billion in Minnesota. $19 billion? A Department of Justice prosecutor suggested that half of the state’s Medicaid spending of $18 billion might be fraudulent, but offered no underlying data for the estimate. Indeed, the federal government’s own data points to Minnesota being effective at controlling waste in its Medicaid program. But Trump has repeated his fabricated $19 billion claim repeatedly regardless.

Trump used such claims to justify the withholding of $259 million in Minnesota’s Medicaid funding, which is how the Administration had kicked off Trump’s “War on Fraud.” While a lawsuit is underway, this type of strategic and illegal withholding of funds — almost unheard of prior to the current Trump Administration — has become all too common.

The task force created by a new executive order only mentioned blue states — in addition to Minnesota, California, Illinois, New York, Maine, and Colorado — reinforcing the suspicion that the war in fraud is effectively a war on blue state safety nets. Fraud feels like the new DOGE: an attempt to target disfavored programs and states wrapped in the gauzy language of good government. Indeed, its no coincidence that Elon Musk also used “fraud” as a justification for his attack on government, even as he found precious little.

Trump’s emphasis on fraud in the safety net maligns popular programs — Medicaid has a 90 percent approval rating — but will do little to help Americans who are drowning in medical debt. Americans across the political spectrum worry more about health care costs than any other cost-of-living concern. Three-quarters of the public say these costs will impact their vote in the midterms. In the meantime, the Administration keeps rolling out one bad idea after another — including most recently an ACA plan with a $31,000 annual deductible — that does little to address peoples underlying concerns.

Trump is betting, instead, on his ability to use fraud as a trope to gut Medicaid, by claiming cuts are simply removing fraud and waste. The underlying hypocrisy is staggering. Trump is yelling about fraud, even as his administration goes on an unchecked crime-spree. Withholding $250 million in Minnesota’s Medicaid funding will not only do nothing to prevent fraud — it will turn eligible beneficiaries into the victims of Trump’s fraudulent fraud squad.

How do we know this is just a political ploy?

Is the Trump administration serious about fraud? This is, after all, arguably among the most corrupt presidential administrations in history. Conflicts of interest are the norm. Overspending on national security appears to be out of control. The Department of Defense is “struggling to spend” its additional $500 billion from US taxpayers that the Administration demanded. But they will give it the old college try, spending almost $100,000 on a piano, $26,000 for a violin, and $21,750 for a flute.

Kristi Noem spent $200 million on ‘top tier’ luxury jets — that included bars and bedrooms, and handed another $220 million in no-bid contracts to firms that had connections to her. Her alleged paramour, Corey Lewandowski, is accused of of shaking down multiple DHS contractors for kickbacks, with such behavior apparently an open secret in the White House.

Kash Patel has turned the FBI into his own personal Make-A-Wish Foundation, hiring UFC fighters, using FBI planes to party with US hockey players in Italy, with buddies for golf and hunting trips, or to meet up with his girlfriend who now has her own FBI security detail.

President Trump has also pardoned more than 70 people convicted of fraud.

• Philip Esformes stole $1.3 billion from Medicare and Medicaid via fraudulent billing. Trump commuted his 20 year sentence after 14 months served.

• Lawren Duran was part of a $205 million Medicare fraud scheme but longer needs to repay $87 million in restitution because of Trump’s pardon.

• Trump nominated Benjamin Landa as ambassador to Hungary, despite the fact that a federal audit found more than $31 million in Medicare overpayments in his nursing homes.

This is just the tip of the iceberg: Trump’s pardons eliminated $700 million in restitution and fines.

Its never been a better time to pursue some types of fraud. Some of Trump’s pardons have gone to crypto operators accused of ripping off investors, but whose investment into Trump’s crypto coincided with their legal and regulatory problems disappearing. Also, the federal agency designed to protect the public from fraud by private companies is effectively dead. The CFPB is now a zombie regulator, having dropped dozens of lawsuits that would valued at more than three billions dollars.

If the Trump administration was really interested in pursuing fraud in Minnesota, it would support prosecutors trying to track down those who broke the law. Instead, it has undermined such prosecutions. Trump’s ramp-up on immigration in Minnesota meant that federal lawyers had little chance to keep their focus on fraud. A slew of Department of Justice prosecutors based in Minnesota, including Joseph Thompson, the lead prosecutor working on fraud cases, quit after they were pushed to investigate Renee Good and her wife after Good was killed by an immigration agent.

The Minneapolis Police Chief said: “When you lose the leader responsible for making the fraud cases, it tells you this isn’t really about prosecuting fraud.” The New York Times reported that Thompson “grew frustrated in recent weeks as the immigration surge became a distraction for the office’s work on fraud, undermining the goal the administration said it was trying to pursue.”

Putting Minnesota in Perspective

There are other indicators that Trump is more interested in treating fraud as a means to score political points rather than something to be minimized. The letter from the administration to Minnesota, which provided the basis for blocking funding, cites no evidence of current (or recent) fraud from federal audits of Minnesota’s Medicaid program. It simply asserts that there is “rampant” fraud.

The reality is that Minnesota has faced challenges with fraud, particularly in its federally funded child care and pandemic era nutrition programs. But it has been broadly effective at managing its Medicaid health services. What limited fraud it did have, which nearly every state has to manage, was being addressed.

Since 2024, the state has made significant organizational changes, including hiring an Inspector General that has a decade of experience prosecuting Medicaid fraud, as well as ramping up its investigative and oversight capacity. It also had complied with federal demands by submitting a Corrective Action Plan. (Meanwhile, at the federal level, about two-thirds of federal government Inspector Generals whose job it is to detect fraud have been pushed out by Trump).

The more telling point is simply that the Minnesota audit data that the federal government tracks shows the state’s effectiveness at minimizing Medicaid payment errors — a broader measure of whether the state is ‘wasting’ tax payer money. Minnesota’s error rate is just 2 percent. For comparison, the rate in Tennessee is 5.6% and 9.7% in Alaska.

Share

Is there waste in Medicaid?

Let’s step back. Trump wants you to believe that fraud is endemic in Medicaid, a massive problem. Is this true?

The 2025 federal audit of Medicaid found that about 94% of its payments are paid accurately — and most of the remaining error reflects improper documentation rather than actual payment error. By contrast, a similar estimate for private health insurers is 80%. In other words, the payment error rate for Medicaid is a fraction of that of private health insurers.

The reality is that it is difficult to avoid any errors. Payment rate determinations are complex and vary based on a wide range of factors. Moreover, most health care providers spend significant amounts of administrative resources managing the paperwork and administrative costs designed to keep these errors lower.

What’s the actual goal?

Fraud has a precise legal definition, one which the Trump administration largely ignores. This definition reflects intention to deceive rather than an accident or incompetence. And this sort of behavior is relatively rare.

The simple reason Trump focuses on fraud is because it resonates with the public — it’s an effective political tool.

Over half of the public thinks that fraud is a significant problem in Medicaid. Maybe this just reflects the firehouse of claims about fraud made by conservatives, who are inclined to think that fraud is undertaken by government employees. But aggregate public opinion data reflects a somewhat more nuanced view. Few people think that beneficiaries are responsible for fraud — instead half blame private insurers that participate in these programs, as well as health care providers.

And these beliefs roughly align with the facts. To the extent that there is Medicaid fraud — it’s perpetrated by larger companies and providers, not the public or government employees. For example, Kaiser Permanente reached a $556 million settlement with the Department of Justice over Medicare Advantage fraud. Chances are that you have not heard of this story, because it doesn’t fit with Trump’s preferred narrative of fraud. But even in Minnesota, the pattern of organized providers stealing resources, rather than individual beneficiaries, appears to hold true.

The vulnerability to private organizations engaged in fraud is a natural function of a health and aging system where the government pays private and nonprofit organizations to provide services. We have chosen not to have government directly provide these services, and constantly look for cheaper alternatives, like home based services. This is also why the Republican beliefs in the survey cited above that government officials are responsible for fraud doesn’t make sense. In an article in Health Affairs, experts warn that targeting, for example, home-based service providers:

jeopardizes services that have become foundational to our country’s modern long-term care system. It would undo more than 40 years of bipartisan federal policy designed to rebalance that system away from institutional care toward less restrictive care provided in the homes and communities where disabled people and older adults want—and have a civil right—to live.

The Trump administration seems to think if it succeeds in pushing its fraud narrative it can justify what are, in reality, just cuts to the existing program that individual beneficiaries rely on.

This comes in the form of direct cuts, such as the $259 million denied to Minnesota, but also in more subtle ways. The Trump administration has been using fraud claims to drown the public in paperwork, making it harder for them to access services. The One Big Beautiful Bill Act introduces a wave of new administrative burdens in Medicaid and SNAP, which will dramatically reduce the use of these programs.

States can try to innovate to manage around these federal government constraints by, for example, expanding automatic renewal when they have reliable data showing that Medicaid beneficiaries are eligible. In fact, they are required to do so by law. But Trump-aligned Medicaid opponents, such as the Foundation for Government Accountability, label such efforts to reduce burdens as “fraud by design”.

Basically, any system that is not burdensome must be fraudulent! If this sounds outlandish the Trump executive order on fraud is basically a laundry list of new administrative burdens with zero acknowledgement that this will predictably reduce access to these programs.

Throwing the baby out with the bathwater

With Minnesota, Trump is targeting a state that has both been in compliance with federal fraud reduction efforts and is not an outlier on its performance in Medicaid.

By de-emphasizing legal prosecutions to favor the collective punishment of all Minnesotans, Trump is punishing the eligible claimants who need care, while reducing the legal risks for the actual perpetrators of fraud.

This won’t get better anytime soon. The Trump administration seems to have persuaded itself that fraud is the key to re-establishing its flagging midterm hopes. For example, fraud is the nominal justification for the SAVE Act, a bill that would likely disenfranchise millions of Americans, and which Trump has argued is essential to Republicans winning elections. If the economy keeps slowing, gas prices keeps rising, and the war with Iran becomes a quagmire, expect to hear a lot more about fraud.

Since the Pendleton Act of 1883, the U.S. federal government has rested on a simple promise: professionalism, merit-based recruitment, independent oversight. Over time, U.S. federal law enforcement became a global reference point—effective, technically sophisticated, built to serve the law rather than a leader. And it traveled. For decades, officers from across the world sought training through U.S. programs such as the FBI’s National Academy and the Justice Department’s ICITAP.

Now that model is collapsing — and ICE is the tip of the spear. A federal judge in West Virginia called it “a regime of secret policing.” Political scientists are applying secret-police criteria to ICE and keep finding the same warning signs: political targeting, arbitrary arrests, concealed identities, operations outside judicial oversight. Tom Homan, Trump’s border czar does not see himself as bound by the law: “We are not stopping. I don’t care what the judges think.”

So how does the world’s role model of professional law enforcement come to be accused of building a rapidly-growing secret police?

That answer is not primarily about ideology—or the particular nature of a single leader. The machinery of building a secret police operates with disturbing predictability, relying on a recognizable organizational structure, and set of career incentives.

We have spent the last decade studying how authoritarian security organizations are built, staffed, and sustained. We asked, who does the dirty work of these regimes – and why?

Our new book Making a Career in Dictatorship traces the career trajectories of more than 4,000 officers in Argentina’s dictatorship-era security apparatus and pairs that evidence with case studies from Nazi Germany, Stalin’s Soviet Union, and The Gambia. What we found contradicts what most people assume about how violent secret police organizations emerge.

The Problem Isn’t Extremists. It’s That You Don’t Need Them.

Most people assume that repressive organizations are filled with true believers — ideological extremists who genuinely want to harm others, or at minimum sadists and sociopaths for whom the work is personally gratifying. The logic of this view is that the way to build a secret police force is to find the worst people and give them badges.

Our research tells a different story.

When we combed through the personnel archives of Argentina’s Intelligence Battalion 601 — the secret police unit that orchestrated the disappearance, torture, and killing of thousands during the country’s so-called Dirty War — we were not looking for monsters. We were looking for patterns. And the pattern we found was strikingly mundane: the officers who joined Battalion 601 had, in the main, performed worse than their peers at the military academy. They had graduated toward the bottom of their cohorts. They had stalled in the lower and middle ranks. They were men whose regular career paths had quietly closed.

These were not the most extreme officers in Argentina’s army. They were the most stuck.

And herein lies the key insight. The Argentine army maintained a rigorous, century-old meritocratic promotion system — Prussian in design, consistent across political regimes, based on performance at each career stage. This system did exactly what meritocratic systems are supposed to do: it identified and advanced the most capable officers. But it did something else too, something less discussed. It reliably produced a large pool of men who did not make the cut — men who underperformed early, fell behind their cohorts, and faced the prospect of forced early retirement under the army’s unforgiving up-or-out rule.

In management theory, this is the Peter Principle at work: people rise to their level of incompetence and then stop rising. In most organizations, those people simply plateau. Military “up or out” systems, including in the US, eventually push out those who fail to advance.

But in Argentina in the 1970s, the military dictatorship offered another option: a parallel unit that needed staffing, valued loyalty over competence, and offered career-pressured officers a second chance. The dirty work of state terror — kidnapping, torture, disappearing people — was psychologically repugnant enough that high-performing officers with smooth career trajectories had every reason to avoid it. But for the men at the bottom of the cohort, it was a ladder.

A ladder that paid off in terms of higher positions at the end of the career, more salary, and better pensions. The worse an officer’s academic record, the more likely he was to join the secret police. Once inside, the worst performers were assigned to the most brutal departments, where the work was most repugnant and the career reward for doing it most valuable.

This is what we call the detouring logic: career-pressured officers “detour” through repressive units not because they are fanatics, but because the detour is the only viable path upward. The regime does not need to recruit extremists. It only needs to create the right organizational conditions — and then let ordinary career anxiety do the rest.

Share

The Institutional Playbook

So what does the playbook to build a secret police force look like in practice? The sequence is familiar to most authoritarian leaders—and once you know it as well, you can spot it in real time.

First, find two pyramids

An existing institution — in this case, federal law enforcement or the military — provides the talent pool. It already contains, by the logic of any competitive promotion system, a substantial number of career-pressured officials: people who have plateaued, who feel passed over, who sense their professional options narrowing.

The second pyramid is the new or repurposed unit — the one that will be staffed with willing enforcers. ICE has existed for more than two decades, but it is now being massively expanded. Its budget tripled under the “One Big Beautiful Bill,” to a level larger than all other federal law enforcement agencies combined. The agency is hiring over 10,000 new agents. It is, structurally, a rapidly growing second ladder — and it needs to be filled.

Second, resource the second pyramid generously and lower barriers to entry

The second pyramid has to offer something the first cannot: advancement opportunities and the prospect of status gains for people whom the first pyramid has passed over. This means positions, bonuses, and a relaxed vetting process.

ICE’s $50,000 signing bonus, reduced training requirements from 13 weeks to 6, and the admission of recruits under looser standards, including some found to have disqualifying criminal histories or gang affiliations, are not implementation failures. They are the point.

As former ICE instructor Ryan Schwank testified before Congress, “no matter how badly a cadet does at those practical exams, no matter how many mistakes they make, no matter how egregiously they violate the law during a practical, we graduate them.” They are not trying to hire the best. They are trying to hire whoever is willing.

Third, squeeze the first pyramid

To enlarge the pool of career-pressured recruits available for the second, the authoritarian makes the professional environment in the first pyramid less secure and less attractive.

DOGE-driven cuts, mass dismissals of career officials, the elimination of entire bureaus and the firing of inspector generals and watchdogs: these generate the anxious, displaced personnel that the second pyramid needs. The officials in other law enforcement agencies compelled to work with DHS via agreements or presidential directives without ever leaving their home organizations, including the FBI agents reassigned to immigration enforcement; the career civil servants who no longer know whether they will have a job next month; the U.S. Army officers who asked for too big of a bonus — these are, functionally, the labor supply for the alternative structure being built alongside them.

Fourth, signal impunity loudly and publicly

Career-pressured officers are not making ideological commitments. They are making career bets. For the detour to be worth taking, they need to know they will be protected when they push legal and ethical limits. This signal must be credible, and it must be public — because it needs to be heard by everyone calculating whether the bet is worth making.

The Trump administration has delivered this signal with unusual clarity. Sweeping clemency for more than 1,500 January 6 defendants, including people convicted of violent assaults on police, tells everyone inside the security apparatus exactly what the rules are: act in the regime’s interest to avoid consequences. Indeed the regime will prosecute people deemed to have impeded such officials, even if it cannot successfully attain a conviction.

The same message is built into enforcement oversight and training. At the ICE academy, ex-instructor Schwank reported, he had been shown a memo authorizing agents to enter homes without a judicial warrant — contradicting both the Fourth Amendment and the academy’s own written materials — and told to teach it verbally, without notes, without changing the curriculum.

And when agents cross the line, no one is watching. Between March and December 2025, DHS’s civil-rights office received nearly 6,000 complaints. It investigated three percent. Its staff was cut from 147 to fewer than 40. The detention ombudsman’s office went from 118 to 5. In a year with 32 deaths in custody, the office investigated one. When ICE agents killed US citizens in Minneapolis, the regime investigated the victims. For every agent in the field, the message is unmistakable: the accountability rules of the first pyramid no longer apply in the second.

Meritocracy Will Not Save Democracy

The playbook works, of course, only if the raw material is there — only if the first pyramid already contains enough people under sufficient career pressure to be tempted by the second. And here we must confront the most uncomfortable part of our argument.

The conventional wisdom holds that professional, merit-based institutions are firewalls against authoritarianism. If promotions depend on competence rather than loyalty, the thinking goes, the bureaucracy will be committed to rules and law rather than to any particular leader. The whole case for civil service reform — from the Pendleton Act onward — rests on this premise. The logic of the highly professionalized American military system that underpins civil-military relations likewise follows this logic.

Our research tells a different story. The Argentine army maintained a rigorously meritocratic promotion system through democracies, dictatorships, and everything in between. It was explicitly designed as an apolitical professional body. And yet this same institution produced both mass repression and repeated coups.

Not despite its meritocratic structure. Partly because of it.

Meritocracy, by design, creates winners and losers. The more rigorous and competitive the promotion system, the larger the pool of losers who do not advance — ordinary professionals with ordinary career anxieties, facing a system that has told them, in effect, that their best years are behind them. When a leader offers them a second chance — a new structure, new positions, different criteria for advancement — the offer is not primarily ideological. It is professional. It speaks directly to the most basic human desire to have a future and be valued.

This is why the deprofessionalization of American federal agencies is not a paradox. It is a strategy. And it does not require extremism to succeed. It requires only that enough career-pressured people accept the terms on offer — and that enough top performers exit, retire early, or keep their heads down.

“Anybody who can retire is retiring,” one veteran FBI agent told the New Yorker recently. “Because you don’t know when you’re going to be fired.” That sentence captures much of the dynamic our research describes. The first pyramid is being hollowed out from the middle. The second is being built in its place — faster, looser, and staffed by people whose professional futures have narrowed.

To be clear, we do not yet have a systematic public breakdown of who exactly is joining ICE. The recruitment pool extends well beyond career-pressured insiders from the military or law enforcement, including individuals who never made it into the first pyramid in the first place. For them, too, this presents a rare chance to pursue a “career, not a job.” Well-paid work with government benefits offers career prospects otherwise unattainable for many.

Still, the fragmentary evidence we do have points in a familiar direction. According to Matthew Elliston, an assistant director at ICE Enforcement and Removal Operations, the positions are sought after “[e]specially amongst those who are currently in law enforcement”: current or former officers, corrections staff, fired federal workers, and veterans, many framing the move as a career opportunity.

At one of the many job fairs, for example, one Army serviceman, applying to become a deportation officer, explained to NPR that he expected to retire soon and wanted to use the ICE job to continue “doing kind of the same thing I’ve been doing.” Asked about the controversy surrounding the agency, he said: “I’m the guy that just executes at this point. So whatever they want, they tell me to do it, I go do.”

Trump Is Playing with Fire

There is one more thing you need to know about the career-pressured officer we have been describing. This playbook has a second edge—one leaders rarely anticipate.

In our book, career pressure produces two rival solutions to the same career problem. One is the described detouring: officers demonstrate loyalty through repression because a coercive assignment offers the only ladder left. The other is forcing: when careers collapse and exits close, some officers decide the best way to salvage their future is to remove the leadership that made them expendable—by conspiring against the regime rather than serving it.

That is why Trump is playing with fire when he weaponizes career pressure. The same pressure that can fill a growing coercive apparatus with willing enforcers can also manufacture a coup risk. And if the regime fast-tracks yesterday’s losers, it also threatens yesterday’s winners. When promotion and prestige are suddenly rerouted, even high performers can become angry stakeholders—raising incentives for moves that destabilize the leadership that rewrote the ladder.

How We Got Here

We began with a puzzle: how does a country with strong democratic institutions, a professional civil service tradition, and independent law enforcement arrive at a moment where a federal agency is being compared, seriously, to a secret police force? A federal agency whose acting director likens its detention-and-deportation system to “[Amazon] Prime, but with human beings”?

The answer is not a mystery. It is a process. A recognizable one, with documented historical precedents, and a specific set of institutional moves that drive it forward. While every setting is different, it would be foolish not to observe the warning signs from history.

What our decade of research has shown — in Argentina, in Germany, in the Soviet Union, in The Gambia, and in the global data across authoritarian regimes since 1945 — is that the raw material for repressive organizations is not ideology. It is career anxiety. It is the entirely ordinary human desire to have a future, to be valued, to find a way forward when the regular path has closed.

Understanding that is not cause for despair. It is cause for precision. The moves that create these conditions can be named, traced, and — while the institutions that could reverse them still exist — interrupted.

We did not write a book about the United States. We wrote a book about the logic of career pressure under dictatorship. But our book about the past may also describe the present, and future, of America.

Historically, we have assumed that public sector procurement processes involved bland contractors with no real ideology beyond profit-seeking. They might do a bad job, or cheat the government, but not because they were opposed to the government. Indeed, this is part of why governments (and especially conservative governments) turned to contracting: they trusted contractors to be more responsive than civil servants.

But what if that is not true? The Trump administration does not think its true. It is cancelling many more contracts from contractors based in blue states than red states. Most recently, it labeled Antrophic as “woke” for not violating red lines about the use of AI to kill and surveil humans, declaring it a supply chain risk, effectively imposing sanctions on the company from within the US, while cancelling a $200 million contract.

These episodes reflect Trump’s partisanship, paranoia and tendency towards retribution more than any coherent philosophy. Antrophic CEO Dario Amodei isn’t fundamentally opposed to the Trump administration and it is disturbing that a private contractor is the one trying to insist on such values. But the example should be causing Democrats to rethink how they assess procurement.

Gabe Menchana offers a thoughtful critique of those supporting the Anthrophic decision, whose view is: “No government could, would, or should tolerate a private vendor dictating the terms of how it fulfills its core functions.” Henry Farrell drew out some implications for the future:

if the Trump administration actually uses the Defense Production Act or similar measures against Anthropic, it’s going to mark a big shift in the political economy of state-private actor relations in the US…If Elon Musk or Alex Karp are at all capable of sober reflection, they might realize that this change is likely not in their interests, regardless of whether the Trump administration loses or wins the next presidential election.

Maybe you are wondering “how big of a deal is procurement and contracts to current modes of governing?” The answer: a very big deal.

• The federal government spends about twice as much on procurement (about $750 billion according to some estimates) than it spends on federal employees (about $360 billion).

• State and local governments spend about twice as much on procurement as the federal government.

Historically, however, we have given private contractors and the processes that hire and manage them a fraction of the political and analytical attention we direct to government employees. One exception is Paul Verkuil's Outsourcing Sovereignty, which warns of the practical and constitutional problems of relying on private actors performing public tasks, especially in the domain of national security.

The Abundance-themed debate about state capacity has helped to change this, surfacing long-simmering problems. This is new and important! I cannot tell you the number of times I’ve been in conversations with government officials where they might raise concerns about certain contractors but refuse to name them, as if they were Lord Voldemort.1

I would characterize this discourse as having two characteristics:

• Broad bipartisan agreement that procurement processes are important but broken, undermining state capacity.

• No clear consensus or models about what to do about it. Which is especially important for tech since tech products are a both a major item of procurement spending and also the means by which governments plan to improve their capacity.

Share

Beyond principal agent models

One limitation of the welcome attention to procurement is that it arrives within the standard political economy logic of principal agent theory: Procurement is a market transaction where the agent (the contractor) uses information asymmetry to exploit the principal (government).

The logic of a market approach is that the agent exploits for market gains: to sell more to the principal than they need, to charge more, to sell a shoddy product and so on. If agents have broader ambitions to reshape government policy are market-related: to curry favor, reduce competition, to establish conditions of rent-seeking and monopoly.

Principal agent logic is useful, but framing government procurement only in market terms misses something, which is that procurement with government is not like selling products between two private actors. It involves the building and maintenance of state power to undertake critical public functions.

So an additional way to think about procurement is via the frame of sovereignty. Doing so helps to put the stakes and dilemmas of procurements in more stark terms, perhaps to the point that people whose eyes naturally glaze over when they hear about procurement (most of us, lets be honest) actually pay attention. It does not mean abandoning the logics of principal agent, which I rely on below, but putting them in a broader framework of sovereign power.

So lets consider three specific ways that a sovereignty perspective might reshape our sense of how procurement matters.

1. Sovereignty to do things

At some fundamental level, if a government can’t get core tasks done, can it maintain sovereignty?

Andrew Greenway asks this question, comparing the ennui of weakened state capacity (including, but not limited, to poor procurement practices) to the feeling that drove Brexit:

The other argument for state capacity is to maintain and enhance sovereignty; or more plainly, a feeling of control. As geopolitics becomes increasingly febrile, more careful thought is going into how national governments maintain their ability to exercise power within their domain. The Brexit campaign got ahead of that nagging sense of loss, playing effectively on voters’ sense of decisions being made elsewhere. It crisply tied the loss of control and agency people many felt in an increasingly globalised world directly to a supranational entity who really did have some power to bind Britain’s hands in ways she may not have independently chosen. If you voted for Brexit because you were persuaded by the repatriation of sovereign power - the ability of government to ‘take back control’ - it must be frustrating to find yourself with a state that still does not appear to be in control.

This is sort of sovereignty vibes — a sense of inability to do basic tasks — triggers a questioning of the state itself. The Brexit comparison is apt. The Brexit slogan was “take back control.” If you look at the abundance discourse, sovereignty vibes are there, reflected in a sense of loss of ability to provide safe and affordable communities with good public services.

Depending on your political perspective, maybe this sense of helplessness reflects the poor performance of public organizations and justifies more outsourcing. But it is federal policy to contract out if there are viable commercial providers and it has been for decades. Contracting out is a central feature of our current model of governing, just as much as bureaucracy bashing has been a feature of our political rhetoric. Over time, both parties have been willing to grow the contractor state while they limit or cut the number of federal employees. So its impossible to engage in complaints about public services while excluding contractors

And in some cases, private actors blur the role between contractor and rent seeker. The US seems to be unable to build a simple free tax filing team that other countries have. This is not due to a lack of technical capacity: under the Biden administration, IRS proved it was possible with Direct File. In no small part, this is because of private vendors like Intuit who both use their political clout to block IRS from building such a system, and “partnered” with the IRS to deliver a subpar product designed to send most taxpayers to its their for-profit services.

Using a different example: as state governments respond to the onerous demands that Trump’s policies place upon administering safety net services like SNAP and Medicaid, they may find that their ability to do so is limited by private vendors who run those systems. Luke Farrell has labeled this the means-testing industrial complex.

2. Sovereignty as national security

The majority of federal government procurement goes to the military. A hollow state where the state only contracts and never builds is vulnerable to the limits and preferences of private actors, and this point becomes especially salient in times of conflict. ).

The additional wrinkle that other countries face is that they are not just dealing with private vendors, but with the government that can use them as a instrument of political pressure. The US government freely accepts this logic when it comes to other countries, even when it involves non-military companies, like TikTok and Huawei.

But now, the US is the major source of digital and military procurement sovereignty concerns for other countries. In Europe, this concern is very much shared among both researchers of technology, and policymakers. Professor Francesca Bria emphasizes:

Europe faces the urgent task of reclaiming its sovereignty in the technology sector…The dominance of US-based tech giants over critical components of Europe’s digital ecosystem – ranging from chips and computing power to AI and digital platforms – has led to what can be described as a new form of digital colonialism.

The majority of military weaponry for non-US NATO members comes from the US. And these countries are not just buying weapons, but an entire US-controlled support system to maintain these increasingly digital systems.

As NATO members view the US with growing suspicion, they also start to think about their supply systems in the context of sovereignty. Mark Carney has ordered a review of purchasing F-35s from a country where the President has threatened Canadian sovereignty. The Portuguese Minister of Defense said:

This ally of ours, which for decades has always been predictable, may bring limitations in use, maintenance, components, everything that has to do with ensuring the aircraft will be operational.

These sovereignty concerns run both ways. The US is dependent on allies for production of its weapons, and for parts of its immigration enforcement procurement. But it is the actions of the Trump administration that has caused the relationships to fray and these sovereignty concerns to become much more real.

Using a more non-military example, the digital sovereignty movement has made the case that a government that cannot guarantee national control of its digital systems is not truly sovereign. The position of the United States is that other countries cannot be allowed to control digital platforms their citizens use, or establish data sovereignty.

The Government of Canada (GC) has had reason to think about this, and framed it in the following terms:

For the GC, digital sovereignty is defined as the ability of the GC to exercise autonomy over its digital infrastructure, data and intellectual property. It is the capacity to operate effectively and make independent decisions about digital assets, regardless of where technologies are developed, hosted, or supported.

Digital sovereignty relies on the GC’s shared ability to govern, access, and secure its digital systems so that programs and services can continue without interruption. It is a collective responsibility across government to keep those systems reliable, resilient, and available. It is impossible for the GC to obtain a state of complete digital sovereignty, known as digital autonomy, due to the absolute interconnected nature of the digital world.

Dominance of military and digital services, and control of the global financial system gives the United States the ability to dictate terms for now. Declining political stability in the United States, and the naked antagonism of the Trump administration towards its erstwhile allies, means that ability becomes a threat to others.

As the Trump administration switches America from an ethos of “speak softly and carry a big stick” to one that is closer to “post crazy threats and occasionally act upon them” it pushes other countries commit to the hard work of building systems that make them less dependent on the US. This was, in effect, the message of the Mark Carney speech at Davos, but is easier said than done.

3. Sovereignty when the contractor is your ideological enemy

Thus far, the motivations of private actors can be explained by following a) profit maximization or b) the dictates of its government when it engages with other governments.

There is a third motivation: c) the contractor has its own ideological agenda. That is to say, the contractor has their own view of politics that goes beyond their business interests, and will use their government position to pursue those goals.

One way in which we see this happen is contractors explicitly align with one political party to the degree they can no longer be viewed as good faith actors for the other. Another way is contractors who operate independent of the state, because they have amassed so much power.

The obvious example here is Elon Musk. Musk is an explicitly white nationalist political force now, aligned with Republicans, but implementing his politics by supporting the far right in at least 18 countries around the world. Within the US, he was Trump’s largest donor during the 2024 campaign, and used his position in government to dismantle parts of government that progressives have championed, often in legally questionable ways.

The U.S. government has developed a multi-layered dependence on Musk’s companies, particularly in aerospace, national security, and digital infrastructure. If it wants to send astronauts to space, or bring them home, it increasingly must turn to Musk. The DOD depends upon Musk for both Starlink for internet service, and for military-use satellites. Grok has been embedded into government, including the Pentagon. Most recently, HHS is sending users to Grok as part of its Make America Healthy Again campaign.

This seems fine for Republicans, who are happy to subsidize Musk and use US diplomatic and economic power to push foreign governments to unconditionally purchase Musk’s products and accept Musk’s platforms that are structured to upend the politics of their countries.

But what about the next Democratic President? The scale of Musk’s power makes it hard to dislodge him from government, even as the extreme nature of his views make it unsustainable to treat him as a neutral contractor.

Musk is just one example. A huge chunk of Silicon Valley bowed down to Trump, ranging between the begrudging, the opportunistic, and the enthusiastic. Professor Francesca Bria has referred to the emerging: “Authoritarian Stack“— “a network of firms, funds, and political actors turning core state functions into private platforms.” She points out that the key actors in the Silicon Valley right — including Anduril, Palantir and venture capital funders like AndreessenHorowitz — are closely connected and working with one another. They also, crucially, share an ideology that places them as empire-makers able to compete with sovereign states.

Another example: Larry Ellison joined phone calls about overturning the 2020 election, and Trump has enabled him and his son to expand their media empire via TikTok, Paramount (including CBS News) and now Warner (including CNN). Also: HBO, Nickelodeon, MTV, Comedy Central, BET, TNT.

The financial backing for the Ellison family’s consolidation of media power comes largely from the fortune Larry Ellison made with Oracle, which provides the backbone of cloud computing and data management for the US government and military. Mikey Dickerson, the first administrator of the US Digital Service noted on LinkedIn:

We got used to seeing a giant, unnecessary Oracle deployment stapled onto the side of every government program, burning millions of tax dollars in O&M [Operations and Maintenance] per year. If you asked why, you got eye rolls and basically "if you don't put in a few million dollars for Oracle, they will protest and sue infinitely many times and stop the program dead.

With Oracle, the government is funding a business closely aligned with one political party, which has shown a willingness to use newly acquired media to tilt the playing field toward their preferred regime.

How do you run a government, when the government is run by your enemy?

Remember that the critique of “the deep state” is that civil servants were using their position to favor one party over the other. The logic is that we need a relatively neutral workforce that follows the law rather than its own preferences. Shouldn’t this logic should also apply to contractors, who are more brazen in melding their political preferences and organizational powers?

Why does this matter? Going back to principal agent theory, there is an underlying assumption that government always holds power over the contractor. Can’t a new government just swap out one contractor for another? For some services, yes. It’s unlikely that the pop-up firms with ties to Kristi Noem will do so well winning DHS contracts in the future now that she is gone. Simple grafters are easily replaced. But in other cases, such as military or tech procurement, replacement is very costly for a number of reasons.

Complex contracts are often for multi-year products that are not easily abandoned without great costs. For digital services, in particular, the contractor can build lock-in in a variety of ways: contracts that are expensive to break, learned helplessness on the government side, and the stacking of tech dependencies that makes it difficult to withdraw one system without affecting the other.

For moguls like Musk and Ellison, the costs of pushing out tech vendors may be even harder. They are massive political donors, and they can turn their media platforms against the government of the day.

If nothing else, one lesson has to be that debates about outsourcing, or increasingly insourcing, have to be based on considerations that are broader than a reluctance to hire new government employees, or imagined efficiencies. Some of these tasks that vendors perform are not so complex that skilled government digital teams could not perform them.

The reality is that government bureaucrats are far more accountable than embedded vendors, and building a nonpartisan and ethical state capacity is in the interest for anyone who wants to govern. But that requires a willingness to build excellence in government, and to recognize the risks of having government run by people who often oppose government.

On February 18th, the Kansas legislature overrode Governor Kelly’s veto to enact Senate Bill 244. The new law has four major components that drastically affect the lives of trans Kansans.

1. Sex is defined as binary, immutable, and determined at birth.

2. As a result, the driver’s licenses and birth certificates of Kansans who had previously changed the sex marker on those documents are now invalid.

3. The law also requires that individuals use restrooms that correspond with their birth sex with violations carrying civil penalties of up to a $1,000.

4. Finally, private citizens can file lawsuits if they feel “aggrieved by” the presence of a trans person (or someone they believe to be trans) in a public restroom. In effect, it deputizes bathroom bounty hunters, much like the law passed in Odessa, Texas and a bill advancing in Idaho.

Three days before the legislation went into effect, an unsigned letter from the Division of Vehicles was mailed to hundreds of transgender Kansans informing them that driver’s licenses that do not list sex at birth are invalid effective February 26. Independent reporter Erin Reed was sent a copy of one of these letters.

Gender marker changes had been allowed on driver’s licenses in Kansas since 2007 as a matter of public policy. But in 2023, the legislature passed a “Women’s Bill of Rights” that defined girls and women as those who produce ova and boys and men as those whose biological systems fertilize ova. Kris Kobach, the Attorney General, tried to apply that definition to the gender marker on driver’s licenses. A Kansas appellate court found that this statute was not clear on whether or not it applied to driver’s licenses and allowed gender marker changes to resume. SB 244 eliminated the ambiguity.

Share

From Administrative Burden to Criminalization and Erasure

What is happening in Kansas, alongside the now-stalled SAVE Act, an example of what are sometimes described as “gendered burdens”: experiences of coercive and controlling state actions that regulate gendered bodies, labor and identity.

Trans people have long faced specific administrative burdens when interacting with state agencies. As Isaac demonstrated in his dissertation, “The Cost of Being Trans: Administrative Burden, Citizen-State Interaction, and Transgender People in the US,” in the area of identity documents, trans people face multi-layered learning, compliance, and psychological costs. If your main complaint about getting identity documents is that DMV lines are too long, it’s worth unpacking all that is involved for trans people.

In 24 states, individuals must find a trans-friendly medical provider who will write a letter. In Kansas, for example, before the recent change, the letter had to attest that one underwent, “appropriate clinical treatment for change of gender” or that gender reclassification was appropriate, and file that letter with the Division of Vehicles, along with the requisite fee. To legally change their name, they would have to go before a judge, and in many states have that name change published in a newspaper.

Some states have eased these burdens; twenty-one states now allow for “self-attestation” of gender for driver’s licenses—no supporting documentation required. However, even that process can still be very expensive, on average costing at least $250. For many trans people, this is an untenable amount of money. The 2016 US Trans Survey found that 35% of individuals who wanted to update their documents had not done so because of cost.

None of this accounts for the additional burden of navigating the process: trans people must become familiar with the laws, regulations, and forms, find office locations, and take time off of work. There are also psychological stressors: anticipated stigma, stress, and negative interactions with frontline workers who can be naive at best and hostile at worst. Many survey and interview participants from Isaac’s dissertation had stories of bad interactions. When a trans woman who had medically transitioned applied for SNAP, a caseworker threatened to call the police after she presented her license with an M on it.

If administrative burdens are the costs of interacting with government to receive a service, SB 244 raises those costs to a punitive extreme. With this law, the costs of compliance have escalated dramatically: residents face penalties both for having complied with the prior law and possibly for attempting to bring their documents into conformity with the new one. The law retroactively converts prior compliance into a violation.

Individuals who carefully followed the rules in place until just last week now face criminal penalties for having the “wrong” gender marker on their identity documents. Simply driving to the Division of Vehicles to change a license back to one’s birth sex becomes “operating a vehicle without a valid credential,” a Class B misdemeanor, punishable by up to six months in jail, a $1,000 fine, and a 90-day license suspension.

More troubling, officials are required to keep a list of everyone who reverts the gender marker on the driver’s license back to their birth sex. According to an email obtained by independent journalist Marissa Kabas, Ken Selk, Driver’s Services Manager for the Kansas Department of Vehicles, instructed employees to track people complying with SB 244.

The law’s costs don’t end at the Division of Vehicles. A valid driver’s license is not merely a credential to operate a vehicle; it is one of the primary identity documents people use to navigate daily life. Kansas legislators are compelling individuals to accept a state-mandated sex designation that contradicts their lived identity. Now applying for a job, picking up certain medications, going through security, or having a document notarized—any routine interaction that requires ID—carries the risk of discrimination, harassment, and violence.

The psychological costs are well documented: twenty-two percent of respondents to the 2022 US Trans survey said they had experienced harassment or violence when they presented an ID with a name or gender marker that did not match their presentation, with Black and Indigenous trans people experiencing statistically higher rates. SB 244 forces individuals who had brought their documents into alignment back into the very mismatch that the state, until last week, had helped them resolve.

An Architecture of Surveillance and Control in a Federalist System

SB 244 didn’t pass in a vacuum. Kansas was a pioneer in this area, but since 2023 fifteen more states have passed laws mandating a statewide definition of sex. Even before Texas passed its statute last year, it had already stopped permitting changes to sex designations on driver’s licenses. Like Kansas, the Texas Department of Public Safety is keeping a list of individuals who try to change the gender marker. Even as these states deny the existence of transgender identity, they are building administrative systems to track trans people. What appears to be definitional housekeeping is, in practice, an expanding architecture of surveillance and control.

Trump’s January 2025 executive order did for the federal government what Kansas had begun at the state level: mandating a uniform definition of sex across all agencies. (The executive order goes far beyond identity documents—from education to the military to prisons to medical care—but identity documents are our focus here.) These efforts have political appeal because they seem to formalize what most people already believe: that legal sex is a single, stable fact—the same on every document, in every jurisdiction, at every agency.

For trans people, it has never worked that way. There is no single, objective definition of sex that operates across law and policy in the United States. As Paisley demonstrated in his book, Sex Is as Sex Does, legal sex is whatever an entity whose decisions are backed by the force of law says it is. Paired with frontline worker discretion, in a federal system with fifty states and five inhabited territories, different governments can and do say different things. Even as the Trump administration and sixteen state legislatures try to impose a single definition, federalism makes that goal all but impossible. On the question of driver’s licenses alone, twenty-two states and the District of Columbia let people attest their gender (M, F, or X); some states require proof of medical transition; others, like Kansas and Texas, use only birth sex.

No document made that variation more navigable—or its loss more consequential—than the passport. For years, it was the most reliable document for trans people in hostile states, an escape valve in the form of a federal document that represented their lived gender even when their own state government didn’t allow them to change it on state-issued documents, or created insurmountable administrative burdens. That divergence was not an anomaly, it was a consequence of American federalism, in which different jurisdictions define and administer legal categories for their own ends.

Trump’s executive order shut that escape valve. The State Department stopped allowing gender marker changes on passports and reverts people to their birth sex when they renew them. In November, the Supreme Court stayed a preliminary injunction blocking this policy, finding that the Trump administration was likely to prevail in a challenge brought by the ACLU.

The variation is not new—federal and state definitions of sex and gender have sometimes aligned and sometimes diverged. In fact, the picture is messier still: even within the same level of government, agencies had different rules for sex classification. And across fifty states, any alignment between a given state and federal agency was incidental, not by design.

What is new is the coordinated effort under Trump and “red” states to close down that variation. When Kansas attaches penalties to its definition and the federal government reverses decades of passport policy, the already limited room individuals once had to piece together workable identity documents across jurisdictions narrows even further. It’s no wonder many transgender people are moving out of states with hostile policies toward ones thought to be more welcoming. In a recent survey from the Movement Advancement Project, nearly one in ten transgender respondents reported they had moved to a different state because of anti-trans policies and politics, and forty-three percent said they had considered moving for the same reason.

However, not everyone can just pick up and move to another state. For those who remain, document mismatches open their holders to accusations of fraud: if the gender marker on the ID conflicts with their appearance, the assumption is that they are using identification that doesn’t belong to them. Florida has taken this a step further. The state doesn’t just prohibit changing gender on state IDs. An official from the Florida Department of Highway Safety and Motor Vehicles announced in a rescission memo that anyone with a different sex marker than the one assigned at birth is “misrepresenting” their gender, which constitutes fraud. A Texas law introduced last year and reintroduced this January goes further still, creating the criminal offense of “gender identity fraud.”

The “fraud” framing has consequences beyond criminal exposure. Not having an ID that matches one’s appearance creates a subordinate form of citizenship. Kansas is one of 13 states that requires photo ID for voting, and several counties have special elections next week. For trans Kansans without passports reflecting their lived gender, the ability to vote now lies with election officials. An understanding frontline worker might not make a fuss when the person in front of them appears “to be a man” (an issue that merits much more than a blog post) but has an F as the sex marker on his ID. However, given the experiences of trans people, they have little reason to count on such understanding. Upcoming elections will show just how much the new law dissuades trans people in Kansas from voting.

Subordinated Citizenship

Problems accessing correct identity documents are not just a gender issue. Black and Latino people are roughly twice as likely as white people to lack government-issued identification, and Black and Indigenous trans people are less likely than white trans people to have updated identity documents.

Gendered administrative burdens operate within existing racialized hierarchies, casting those who fall outside racialized norms of femininity and masculinity as suspect. As more states consider following Kansas’ lead, trans people—particularly Black and Indigenous, immigrant, and disabled trans people—are pushed to the margins of citizenship. By deputizing citizens to sue trans people over restroom use, the Kansas statute institutionalizes vigilantism.

This citizen surveillance ends up ensnaring cisgender women; last year alone there were multiple cases of women deemed “too masculine” to belong in a women’s bathroom. This is especially true for Black cisgender women, whose femininity is compared against ideals of white womanhood. That vigilantism is unlikely to remain confined to restrooms. In a political climate saturated with claims of election fraud, the sporadic reports of voter intimidation at polling places to monitor for (nonexistent) voter fraud will only increase.

SB 244 exemplifies policymaking by other means. It shows that defining and classifying “sex” is not technocratic housekeeping but an exercise of political power. Here, that power is wielded through legislation that humiliates, deters, and possibly disenfranchises those it targets. As similar bills proliferate across the country, these new burdens do not combat the fraud they purport to address. Instead, they function to punish difference under the guise of administrative order.

After 9/11 there was broad support for invading Afghanistan because the country hosted the attack’s mastermind. In 2003 there was less support for what turned out to be the false claim that Iraq had weapons of mass destruction. They didn’t, but Saddam stuck a belligerent tone and kicked weapons inspectors out. In 2026…Iran was at the negotiating table and we invaded…for reasons.

A couple of days before the attack Daniel C. Kurtzer, a Professor of Middle East Policy Studies at Princeton summed up the lack of clarity:

The bottom line is we don’t really know why the U.S. appears to be preparing for war with Iran. Trump has not yet briefed Congress or spoken to the American people about a possible military action that could inflame regional tensions and spark a bloody conflict.

Clarity has not emerged since the bombs started dropping. Democratic Senator Mark Warner was briefed by the White House House as senior member of the Senate Intelligence Committee. He said: “What was the imminent threat to America? I don’t know the answer.”

So lets review the various reasons the Trump administration and its supporters have articulated for a likely illegal war they are pursuing without Congressional permission.

1. Iran was working on nuclear weapons

Trump discussed the potential for Iran acquiring nuclear weapons in his state of the union address. How close was Iran? We don’t know. Trump negotiator Steve Witkoff said Iran was “probably a week away from having industrial-grade bomb making material.” Trump himself claimed Iran was just “a few weeks away” from having a nuclear weapons, before deciding that “two weeks” was the right number.

But last June the Trump administration said it had “obliterated” Iran’s nuclear facilities, and cited Israeli officials who said the nuclear program was set back by “years.” How do you go from years to weeks in months?

There is also the point that Trump inherited a credible and carefully developed agreement to prevent Iran from acquiring weapons that was working, but tore up that agreement because it was negotiated by Obama. Recreating that agreement might have been on the table, but we will never know.

At the State of the Union, Trump also said that Iran was “working to build missiles that will soon reach the United States of America.” But this was exaggerated and not an imminent threat. According to the New York Times:

Three American officials with access to current intelligence about Iran’s missile programs said that Mr. Trump exaggerated the immediacy of the threat posed to the United States. One official said some intelligence analysts were concerned that top aides have inflated the threats or that intelligence was being selectively presented or distorted as it was sent upward.

White House aides exaggerating threats to spur a war they already wanted has a very 2003 feel to it. Say what you will about the Bush administration, but when they sold a Middle-Eastern war, they took some pride in selling the lie.

2. Iran interfered in US elections!

Trump has complained that Iran interfered in US elections. Well, yes they did, but to no effect. And so did China and Russia, and we are not attacking them. Indeed, we are attacking anyone who implies that Russia tried to influence US elections.

A bit more backstory. Stanford academic researchers and government investigators who identified the role of Iran’s election interference in 2020, which included social media manipulation, were pilloried by the Twitter Files brigade as censors, hauled before Congress, and sued by Stephen Miller’s American First Legal. CISA, the part of the federal government that monitored election interference, was gutted by the Trump administration and the Stanford Internet Observatory was shut down.

I’m inclined to believe if election interference was enough of a reason to attack a foreign country, the administration and its allies would not be punishing people who identified election interference as a problem.

Kate Starbird, who was one of the researchers targeted, summed it up this way:

The real story is that foreign governments try to interfere with elections in the U.S. all the time. In 2020 and 2024, the U.S. had private and public infrastructure in place to catch them, call them out, and mitigate the damage. Now, those mechanisms are gone, intentionally undermined and dismantled by the Trump administration.

Share

3. Iran actually started the war 47 years ago

Ok, I don’t really understand this either. The US clearly started the conflict, along with Israel, start the conflict by sending warships to the region and then bombing the shit out of Iran.

But what if, and bear with me here, the words “war” or “conflict” could mean whatever you wanted? With this linguistic freedom, we can say that Iran started the war by being an antagonist of the US, and having some low-level conflict with it. Here is what Hegseth said:

The United States did not start this conflict, but we will finish it. If you kill or threaten Americans anywhere in the world—as Iran has—then we will hunt you down, and we will kill you.

At a press event, Hegseth quipped:

It turns out the regime who chanted 'death to America' and 'death to Israel' was gifted death from American and death from Israel. This is not a so-called 'regime change war,' but the regime sure did change.

Another version, articulated by Republican House Committee is that Iran started a “forever war” with the 1979 revolution (don’t ask about US engagement in Iran before 1979!)

It this sounds sort of nuts, be assured that for some Americans, the Iranian revolution, including its kidnapping of American hostages, felt like a declaration of war. It is the justification being fed to the troops, in fact the only justification that the Commander of CENTCOM invoked in a message to the troops: “Since 1979, the Iranian Regime has killed and wounded thousands of Americans, and the continue to threten Americans and our friends.”

This messaging is also very visible in the media. “Trump is ending a 50 year war” is a recurring talking point among Fox hosts and guests. The Free Press, is broadly supportive of the attack, presenting the it as “the end” of “our 50-year war with Iran.”

Bari Weiss, who has editorial control of both The Free Press and CBS News is also pushing the 50 year war at CBS.

If you did not know the US was in a 50 year war with Iran, no-one is judging you. You had a lot of other wars to keep track of. And don’t ask about anything that happened before 1979, like the American-supported coup of a democratically elected leader in 1953, and the subsequent installation of a military official as Shah that fueled the anti-American theocratic regime that took control in 1979.

Subscribe now

4. Iran about to attack the US

According to Trump-mouthpiece and CNN commentator Scott Jennings, Iran was actually just about to attack the US. While the US had warships at its door, Iran had planned a preemptive strike. We stopped them in the nick of time!

Senior officials gave other reporters the same message. I wonder if these officials were the same administration officials that were claimed there was credible intelligence that Iran would have nuclear weapons next week, or that it would be soon firing missiles at the United States. But here is the thing: intelligence officials who briefed members of Congress said there was no imminent threat. So you can believe the anonymous political appointees making the case for war to their preferred media outlets, or the actual intelligence officials talking to Congress.

By Monday, the White House was undercutting this justification, saying that if Israel attacked, Iran would then attack American targets, and that was the imminent threat. In other words, we were getting our retaliation in first. When pressed by reporters, both Mike Johnson and Marco Rubio presented this explanation, coining what might become one of the defining quotes of this war. Rubio said:

There absolutely was an imminent threat and the imminent threat was that we knew if Iran was attacked, and we believed they would be attacked [by Israel], that they would immediately come after us…We went proactively in a defensive way.

This is, by any measure, a very different argument than the claim that Iran was about to attack US targets, since it is contingent on the previously unspecified assumptions that Israel was going to attack which would trigger an Iranian retaliation. It also takes for granted that Israel’s decision to take action was not something that the US could have blocked, but instead obliged the US to join the attack. Other reporting from the New York Times portrays Netanyahu as pushing Trump’s decision:

During a meeting at Mr. Trump’s Mar-a-Lago estate in December, Mr. Netanyahu had asked for the president’s approval for Israel to hit Iran’s missile sites in the coming months. Two months later, he got something even better: a full partner in a war to topple the Iranian leadership.

5. Regime change

In announcing his attack, Trump said: “To the great, proud people of Iran, I say tonight that the hour of your freedom is at hand. When we are finished, take over your government. It will be yours to take.” This would be good news for the people of Iran, who have been plainly unhappy with their brutal rulers. But they also have reason for caution.

Trump also offered immunity for members of the regime who lay down their arms, but its not really his immunity to offer if the people of Iran are to determine their future. In a social media post, Trump said that he hoped that the Islamic Revolutionary Guard Corps and other security forces would just merge with the people they have been killing, which seems a tad unrealistic:

Hopefully, the IRGC and Police will peacefully merge with the Iranian Patriots, and work together as a unit to bring back the Country to the Greatness it deserves,” the president added. “That process should soon be starting in that, not only the death of Khamenei but the Country has been, in only one day, very much destroyed and, even, obliterated.

After Trump previously encouraged protestors to rise up, tens of thousands were murdered by the regime. And if they look at Venezuela, or even in America, Iranians might see a Trump administration that does not see its mission as encouraging democracy.

In Venezuela, a corrupt but pliable regime is preferred to the opposition who won the last election. Trump has described Venezuela as a template for Iran. Apart from the fact these are two very different countries and forms of government, the Israeli bombing of political leaders seems to have taken out likely successor candidates, including targeting members of the opposition. Trump said:

We don't know who the leadership is. We don't know who they'll pick. Maybe they'll get lucky and get someone who knows what they're doing ... We don't know who is leading the country now. They don't know who's leading. It's a little like the unemployment line.

Does this sound like a plan?

The Iranian hardliners likely assume that Trump will not put boots on the ground, and they can hold onto power by offering concessions about external military and terrorism threats. Once the crisis has passed, they can reassert their brutal control over the Iranian people. Trump has already agreed to talk to the new leadership.

Beyond the Iranian people, advocates of regime change also see it as generating a new age of peace in the Middle East. See, for example, Republican Senator Lindsey Graham:

I fervently pray that the long-suffering people of Iran will have their oppression ended soon. I also fervently pray that we’re on the verge of a new dawn in the Middle East, with historic opportunity for lasting peace and prosperity. As to our allies in Israel, President Trump and all under his command, your bravery has set in motion the end of evil and darkness, and the beginning of the light. Well done.

This is, to put it kindly, wishcasting that sounds very much like the neo-conservative rationales for war in Iraq. Once Iraq is toppled, democracy would blossom and the region would stabilize. It seems like we are always just one expensive and bloody war away from stabilizing the Middle East! No downsides to the US!

In 2003, the wishcasting was accompanied with some efforts of nation building, poorly planned. But it appears that no such effort exists now. When Graham was challenged on Meet the Press Trump plan for Iranians, he snapped:

No, its not his job or my job to do this. How many times do I have to tell you? Our job is to make sure Iran is no longer the largest state sponsor of terrorism, to help the people to reconstruct a new government. No boots on the ground. You know the sign [that says] “you break it, you own it”? I don’t buy that one bit.

Here, Graham appears to be referring to Colin Powell’s Pottery Barn Rule that he articulated to Bush: “If you break it, you’ve bought it.” The Trump rule, it would appear is, “you break it, they own it.”

6. Protecting Americans

Speaker Johnson articulated some of the above reasons but also said that he was briefed that “that military action may become necessary to protect American troops and American citizens in Iran.” What? America is bombing Iran to protect American citizens in Iran? There is no public data on the number of US citizens living in Iran, but they are presumably less safe because their government is bombing them. There are hundreds of thousands of American citizens living in the Middle East, especially in some of the cities that Iran is bombing right now. Are they safer? Will they be safer in the years to come in a region where many are outraged about the US-led attacks?

7. Its not actually a war, you dummy,

A couple of administration advocates such as Senator Markwayne Mullin and Representative Paulina Luna and even Speaker Mike Johnson have gone to the airwaves to insist its not a war. After all, wars have to be approved by Congress! And there are no troops on the ground! Its just a military action!

The White House actually sent out talking points to advocates emphasizing that “Operation Epic Fury” was merely “a major combat operation.”

But the “not a war” nonsense is undermined by the fact that Trump and Hegseth are calling it a war. And can you imagine what would happen if the US was attacked via an intense bombing campaign that killed dozens of political leaders, and people insisted we were not at war?

What happens when you don’t have a clear rationale

None of these feel like legitimate reasons, and the multiplicity of the reasons makes things worse. To be fair, its not unreasonable to have multiple goals with a military engagement, but Trump’s tendency to exaggerate and undercut his rationales with contradictory statements creates a sense of incoherence. And the fact that Trump did not present his case to Congress further weakens the legitimacy of his efforts.

The reality is that this is a war of choice. Trump wanted to pursue it, and Middle Eastern allies who are sick of Iran’s funding of terrorists encouraged or went along with the attack. According to reports, both Israel and Saudi Arabia, Iran’s regional foes, urged Trump to attack. In part, this is because Iran is in a weakened state — already attacked by Israel, losing allies like Syria, and with its terrorist arms decimated — not because it is an imminent risk.

There might be a bunch of other, less articulated reasons, such as oil extraction (like Venezuela) and investment opportunities. Trump has also undersold the risks presented to him by intelligence and military advisors, preferring instead political appointees who are either pro-war or afraid of challenging Trump and the optimistic claims of Netanyahu.

According to the New York Times, Congressional briefs made no mention of regime change the White House discussed, while Trump distorted the military advice he received:

Mr. Trump’s top military adviser, Joint Chiefs of Staff Chairman Gen. Dan Caine, told the president that a war could lead to significant American casualties. Days later, Mr. Trump told the public that his military adviser had been far more reassuring. He wrote on Truth Social that General Caine had said that any military action against Iran would be “something easily won.”

“We think Iran is a bad actor without an immediate threat to the US” or “our regional allies really wanted this” are not especially compelling reasons for action. This has a couple of consequences.

First, there is not public buy-in for the war. It is worth remembering that the pattern of Middle Eastern wars is that they start as popular, and then become calamitous as time goes on. This one starts as unpopular because there is no compelling justification to rally around the flag.

Just 1 in 3 of those polled on the say of the strikes support the attack according to G. Elliott Morris. This is actually an improvement from previous polls. For example, 2 weeks ago, when just 1 in 5 supported a US attack. More Republicans backing their President, but most Americans are unconvinced.

For Iranians on the ground, they now have to wonder, if the Trump administration has their best interests at heart. His track record, and mixed messages, make Trump unable to provide a credible commitment.

The Global Supply Chain Model

Recently, I wrote a piece called “Private Chains, Public Harm” that illustrated how ICE is dependent on global supply networks often involving foreign-owned firms. The global supply chain underpinning ICE’s operations operates at the intersection of public authority and market forces: it is structured in part by national oversight through DHS and in part by private, profit-driven incentives.

That hybrid structure expands the government’s operational reach and flexibility, but it also creates dependencies. When the United States relies on foreign firms to carry out controversial enforcement policies, while simultaneously straining relations with those firms’ home governments through economic pressure or military threats, it exposes itself to political and economic backlash.

However, this dependence is not confined to private firms. It holds true for subnational governments.

The Limits of Federal Budgetary Power

The 3,000-agent ICE operation in Minnesota that began in November and led to the detention of 4,000 undocumented immigrants and the deaths of two American citizens pushed Congress to rein in ICE through a series of demands that are driving the current partial government shutdown of DHS:

1. DHS agents cannot enter private property without a judicial warrant.

2. Immigration enforcement agents are prohibited from wearing masks or face coverings

3. DHS agents must clearly identify themselves by agency, badge number, and last name, including verbal identification upon request.

4. Federal funds may not be used for enforcement operations near schools, churches, polling stations, or medical facilities.

5. Prohibit racial profiling.

6. Establish a standardized use-of-force policy and require additional training.

7. Ensure state and local jurisdictions can investigate and prosecute potential criminal conduct or excessive force.

8. Require all detention facilities to meet uniform baseline standards.

9. Mandate body cameras.

10. Standardize uniforms and equipment to align with civil enforcement norms.

These strategies rely on two traditional congressional tools: statutory reform and budgetary leverage. Both presume that ICE’s enforcement capacity flows primarily from federal authority and federal funding.

Yet ICE’s operations are not sustained by appropriations alone. While prior funding, including the more than $70 billion allocated last summer with the passage of the One Big Beautiful Bill Act, has ensured the agency’s continued functioning, these resources are embedded in private contracts and intergovernmental agreements.

Much of this funding is available through 2029. Additional funding for DHS explicitly authorizes support and reimbursement for state and local participation in homeland security efforts. This includes reimbursing state and local agencies for their involvement in ending the presence of criminal organizations, combating human smuggling and trafficking networks, and supporting immigration enforcement, further embedding local actors into federal operational networks.

State and local governments may enter into 287(g) agreements with ICE for several reasons. Beyond financial incentives, these agreements enhance local law enforcement capacity by deputizing officers to carry out certain federal immigration enforcement functions, such as identifying and processing noncitizens. They can also serve as a political signal, demonstrating a jurisdiction’s commitment to cooperating with federal authorities. Finally, 287(g) participation allows local agencies to influence enforcement priorities and procedures under federal supervision, giving them some control over how immigration enforcement intersects with local policing.

In this way, ICE’s capacity is co-produced. Federal power may authorize enforcement, but it does not execute it alone. Recognizing this distributed architecture shifts the analytic frame: ICE depends on supply chains, subcontractors, and cooperative agreements that extend well beyond Washington.

This dependency creates chokepoints, and these chokepoints are often located at the state and local levels.

Share

Federalism as a Choke Point

Looking to Congress alone to discipline ICE assumes that its operational capacity is wholly centralized. However, American federalism distributes not only legal authority; it distributes the material and administrative conditions under which policy is executed. States are not powerless simply because they are smaller or lack national jurisdiction. It is their relative autonomy makes them both powerful partners as well as potential points of weakness in a multi-level network.

This structural feature was recognized in the earliest days of the republic. In Federalist Paper No. 46, James Madison remarked that if states sought to resist federal encroachments, they would possess “the advantage in the means of defeating such encroachments,” because unpopular federal measures would still depend on execution “by means on the spot and depending on the State alone.” Madison’s observation was not merely rhetorical. It described the reality of federalism and policy goals: even when federal authority is supreme in law – which it is as per Article VI, Clause 2 of the U.S. Constitution – many of the federal government’s aims depend on subnational infrastructure in practice.

That remains true today.

Federalism therefore creates both opportunities for cooperation as well as the structural leverage necessary for states to undermine federal government objectives.

Consider New Jersey Governor Mikie Sherrill signing an executive order prohibiting immigration officers from accessing non-public portions of state-owned property without a judicial warrant and barring the use of state property as staging grounds for enforcement operations. The order does not invalidate federal authority. It changes the conditions under which that authority is exercised.

Section 287(g) agreements

States and local governments cannot nullify federal immigration law, nor can they directly cancel federal contracts. But they can regulate access to state-owned facilities, withdraw personnel, restrict the use of state and local resources, and terminate cooperative agreements. Each of these actions alters the terrain on which enforcement occurs. A substantial share of the intergovernmental cooperation concerning ICE occurs through section 287(g) agreements between the federal government and state and local actors.

Established with the passage of the Illegal Immigration Reform and Immigrant Responsibility Act of 1996, section 287(g) agreements allow state and local law enforcement agencies to enter into formal agreements with ICE:

• The Jail Enforcement Model focuses on screening and processing noncitizens who are potentially removable and who have been arrested on criminal charges by state or local law enforcement.

• The Task Force Model operates as a force multiplier, enabling state and local officers to exercise limited immigration authority during routine policing activities under the supervision of ICE.

• The Tribal Task Force Model similarly extends limited immigration enforcement authority to tribal law enforcement agencies, pursuant to Title 25 U.S.C. § 2804, with ICE oversight.

• The Warrant Service Officer program authorizes ICE to train and deputize state and local officers to serve and execute administrative immigration warrants within their correctional facilities.

These agreements, whether jail enforcement, task force, tribal task force, or warrant service officer models, effectively expand federal capacity by embedding immigration authority within existing local law enforcement structures. They expand operational reach without adding more federal personnel.

According to the ICE website:

“As of February 20, 2026 4:22 pm ICE has signed 1,427 Memorandums of Agreement for 287(g) programs covering 40 states. These include [Jail Enforcement Model] agreements with 152 law enforcement agencies in 32 states, 287(g) [Warrant Service Officer] agreements with 465 law enforcement agencies in 35 states, and 287(g) [Task Force Model] agreements with 809 agencies in 35 states.

In addition, there are 12 pending 287(g) applications that include 10 task force model agreements and 2 warrant service officer agreements.

Because this capacity is delegated, it can also be withdrawn. Under the Trump administration, 287(g) agreements have increased. Yet state and local law enforcement retain authority over whether their own agencies participate.

The state-local relationship can block or coerce 287(g) agreements. Some states like Florida have taken a firmly cooperative stance on 287(g) agreements. In 2022, the state enacted CS/SB 1808, requiring “each law enforcement agency operating a county detention facility to enter into a specified agreement with the United States Immigration and Customs Enforcement to assist with immigration enforcement.” This appears to require county jail operators to participate in 287(g) models.

Conversely, in Virginia, Governor Abigail Spanberger signed an executive order enabling her administration to remove Virginia State Police from a previously established 287(g) agreement, thereby retracting state-level participation in federal enforcement efforts, and joining other Democratic-run states that have implemented bans.

When states withdraw support, end jail contracts, or limit information sharing, ICE retains formal authority but loses operational efficiency. Its agents must locate individuals without local data, transport detainees longer distances, and navigate unfamiliar communities without the guidance of local law enforcement.

Local Governments in Multi-level Governance

Many of the agencies entering into 287(g) agreements are at the county or city level. Local governments add another layer of institutional complexity. In American constitutional doctrine, local governments are often described as “creatures of the state,” meaning their authority ultimately derives from state law. Maryland demonstrates how states can structure this layered authority. For instance, Governor Wes Moore is set to sign legislation prohibiting both state and local agencies from entering into 287(g) agreements and terminating existing ones.

In many jurisdictions, local governments exercise substantial discretion over policing practices, facility access, and contractual relationships. That discretion can either reinforce or complicate federal enforcement. Where these local agencies are granted greater autonomy, many are using it to weaken the interdependent institutional relationships that ICE depends on for enforcement.

After Maryland took steps to end 287(g) agreements in the state, local governments including Baltimore, Howard, Prince George’s, and Montgomery County, have also moved to adopt legislation creating uniform rules to limit how local law enforcement engages with ICE.

Additionally, according to reports from Sludge, a nonprofit project that tracks money in politics, Hartford, Connecticut is ending an ICE-related contract involving the Hartford Parking Authority. In Philadelphia, city council members have sought to bar municipal agencies from collaborating with ICE. These measures do not somehow overturn federal immigration law, but they do adjust access to local resources and personnel, thereby affecting how enforcement unfolds in practice.

Across these examples, the pattern is consistent. Federal immigration enforcement is not a self-contained hierarchy executing policy from Washington downward. It is a layered system that depends on cooperation embedded within a federal structure. Federalism creates multiple points at which enforcement can be enabled, reshaped, or constrained.

Rejecting or Reversing 287(g) Agreements is a Real Way to Limit DHS

The debates surrounding ICE are often framed as solely battles over law, funding, or presidential authority, but enforcement is produced through the discretionary actions of bureaucracies embedded across the federal system. Just as global supply chains expose material dependencies beyond U.S. borders, American federalism embeds domestic dependencies that shape whether federal policies function in practice.

This means state and local governments are not passive implementers but active actors. For critics of ICE, this institutional reality broadens the terrain of accountability making mayors, city councillors, state legislators, and governors potential focal points of public protest. While elected officials have occasionally used federal funding as leverage over states, a pattern that has occurred across administrations, the reputational risks associated with local involvement in immigration enforcement mean that public organizations may seek to distance themselves from ICE operations, particularly as these programs become increasingly toxic.