The Department of Justice has fired more than 100 immigration judges in roughly a year, brought in 140-plus replacements with thinner records, and is now publicly advertising what an internal description casts as “deportation judges.” The official rationale is a 3.7 million case backlog. The unofficial rationale, visible in the firing pattern and the new hiring criteria, is faster removals with fewer asylum grants. This is what it looks like when an executive branch reshapes a quasi-judicial system to guarantee outcomes.
What Just Changed Inside the Bench
Acting Attorney General Todd Blanche framed the firings as a quality push on slow rulers, but the Washington Post review of the firing pattern found something more selective. The judges removed disproportionately came up under Democratic administrations, had previously represented immigrants in private practice, and granted asylum at higher rates than colleagues. The administration is not weeding for speed. It is weeding for outcome.
The April 10 round was the most explicit. Six judges were fired, including Roopal Patel of the Boston Immigration Court and Nina Froes of the Chelmsford Immigration Court, both of whom had recently ruled in favor of pro-Palestine activists in deportation cases the administration was watching closely. The union representing immigration judges puts the total fired or forced out at 113. EOIR has hired roughly 140 new permanent and temporary replacements over the same window, a turnover bigger than any in the system’s modern history.
Immigration courts are not Article III courts. The Justice Department runs them. The attorney general can fire the judges with relatively few constraints. That is the structural feature the current DOJ is exploiting. There is nothing illegal about it. There is also nothing accidental about which judges are leaving and which are arriving.
The Replacements Look Different
A separate Washington Post investigation into the new hires found that two-thirds of the new immigration judges have no immigration-law experience listed in their published biographies, a sharp break from prior years when the bench mostly came from immigration-law backgrounds on either the prosecution or defense side. Many of the new hires came from immigration-enforcement and prosecutorial backgrounds at ICE, CBP, or US Attorney offices. The training program for incoming judges has been cut from roughly five weeks to three.
The DOJ has also raised compensation for the new openings: up to $207,500 a year, with 25% signing bonuses for postings in Democratic-led states such as California where the recruiting has been hardest. Applications for the latest round are due May 22. None of that is unusual on its face. What is unusual is what the DOJ is selecting for inside that recruitment funnel.
The pattern that emerges, looking at the firing data and the hiring data side by side, is a bench that will hear more cases per judge, with less time per case, more often referred from prosecutors and enforcement agents who used to argue against the very reliefs the bench is now adjudicating. Immigration lawyers describe this as the “deportation judge” model, a label the DOJ rejects but which describes the throughput the system is being engineered to produce.
Speed Has a Cost That Doesn’t Show Up in the Backlog Number
The 3.7 million case backlog is real. So is the human cost of asylum seekers waiting four to six years for a hearing. The administration’s pitch is that anyone arguing against the overhaul is defending a status quo that fails everyone, including the people in the queue. That pitch lands with parts of the public for a reason: the system is not working.
But “fail faster” is not the same as “fix.” The backlog grew under multiple administrations because Congress underfunded the courts, immigration policy churned every four to eight years, and asylum demand exceeded any conceivable bench’s throughput. None of that is solved by replacing experienced judges with prosecutors fast-tracked through three weeks of training. What it produces is decisions made under time pressure, with less expertise, by adjudicators selected partly for outcome. The appeals will pile up. The federal courts that do have lifetime tenure will end up reviewing those appeals, and the opinion editorials from former immigration judges argue this is exactly the predictable outcome.
Who Bears the Risk
The cases in front of these courts are not abstractions. Asylum claims, withholding of removal, Convention Against Torture protection, cancellation of removal: each one is an evidentiary fight where the cost of a wrong decision is removal to a country where the person may be in real danger. Speed-optimized adjudication, by people new to the area of law, raises the rate of wrong decisions in both directions. Some grants that should not have been granted. More denials that should not have been denied. The denials produce removal orders, which the government acts on. The grants do not unwind as easily.
The legal community has been clear about the institutional risk. The American Immigration Lawyers Association describes a “systematic dismantling” of due process inside the immigration courts. That language is sharper than the polite version most bar associations use, and it is sharper because the union of immigration judges itself has said the same thing in milder form. When the people who run the system from the inside and the people who litigate against it from the outside agree the floor has fallen out, the dispute is not partisan. It is about whether a quasi-judicial system retains its quasi-judicial character or becomes an enforcement extension.
For broader context on the DOJ’s institutional trajectory, our coverage of the Bondi firing and Blanche’s elevation to acting AG traces how this current immigration-court push came together inside a department that has already absorbed a year of restructuring at the top.
The Move Congress Is Not Making
Article I of the Constitution gives Congress the power to set the rules of the federal courts and, by extension, the immigration courts that the executive runs as an administrative bench. There is a long-pending proposal to convert immigration courts into Article I tribunals, which would give the judges some independence from DOJ political control and stop exactly this kind of mass firing. That proposal has been on a slow burn for years. It is not on a current legislative path. The reason it is not on a path is that one party benefits when the bench is responsive to enforcement priorities, and the other party benefits when it is responsive to humanitarian priorities, and the median member of Congress would rather not write a system that constrains either.
The current overhaul makes the case for Article I status more vivid than any reform paper has. It also makes it less likely in the near term, because the political price of taking on the issue right now is higher than letting the courts process the consequences case by case.
What to Watch
Three signals matter in the near term. First, federal court rulings on individual immigration-court cases where the asylum decision looks plainly wrong on the record; the federal bench’s tolerance for fast-track outcomes is the real ceiling on how far this overhaul can go. Second, whether asylum grant rates fall sharply in the new bench’s data versus the historical baseline, which would convert a structural critique into a quantitative one. Third, whether any of the fired judges file wrongful-termination or whistleblower complaints that surface internal communications about how the firing list was assembled.
The longer-term signal is whether Congress moves on Article I status during the next budget fight. If it does, the current overhaul becomes a transitional moment. If it does not, the immigration bench enters a new equilibrium in which the executive’s outcome preferences are simply what the law produces. That is a different country, quietly, without anyone voting on it as such.