THE REVIEW
The Ratchet — A Research Edition — How Power Actually Works
Six Stories About Doors That Only Open One Way
This week: the largest peacetime workforce reduction in American history, the contractor boomerang, administrative limbo, media capture, the preemption machine, and the quiet room where bipartisan reform moves one click per decade
A ratchet is a device that permits motion in one direction only. You can turn the gear forward. You cannot turn it back. The mechanism is simple, ancient, and — once you understand it — visible in every story in this edition.
In the past twelve months, the United States government has conducted the largest peacetime workforce reduction in its history, firing 352,000 federal employees in a campaign framed as efficiency. The people who built disease surveillance systems, processed veterans’ disability claims, and answered Social Security phone lines were terminated in batch events based on hire date and agency, not performance. The institutional knowledge they carried — which county health official to call during an outbreak, which lab has the equipment for a novel pathogen, which procurement loophole has been closed and which remains open — walked out the door with them. That knowledge does not exist in databases. It exists in people. And when those people are gone, it is gone. Click. The ratchet moves one notch.
The same workers are now being rehired as private contractors at rates 30 to 60 percent higher than their previous salaries. The work has not changed. The cost has increased. But the civil service protections — the whistleblower statutes, the merit-based hiring, the congressional oversight — are gone from those positions permanently. Nobody “un-contractors” a role once it has been converted. Click.
Twelve thousand to eighteen thousand federal employees exist in a condition the Government Accountability Office had to invent a name for: “administrative limbo.” They have not been fired. They have not been reinstated. Court orders requiring their reinstatement have been issued and formally acknowledged by the agencies involved, and then nothing has happened. The orders were not defied — defiance is visible, resistible, impeachable. They were simply not processed. The precedent now exists: a court order can be acknowledged and not executed, and nothing follows. Click.
A media conglomerate that includes CBS News is now controlled by a family with direct political ties to the current administration. Senior producers at 60 Minutes have quit citing editorial interference. A merger that would place CNN, HBO, and TikTok under the same family’s control is advancing, and the federal regulators who would block a politically motivated acquisition work for the president who publicly demanded the sale. The journalists who left CBS will not return. Their institutional culture leaves with them. Click.
In 23 states, preemption bills drafted from identical templates strip local governments of the authority to set their own policies, and a new innovation — penalty clauses imposing personal financial liability on local officials — ensures that even when political winds shift, the deterrent remains on the books. Click.
And in a Senate committee room that nobody is watching, two senators from opposite ends of the political spectrum — Dick Durbin, liberal Democrat from Illinois, and Mike Lee, conservative Republican from Utah — are in the thirteenth year of a partnership to reform federal sentencing. They have already produced one landmark law. The data shows it works. The public supports it. The institutional machinery of both the left and the right endorses it. And for thirteen years, the legislation has not passed. This is the counter-ratchet: reform that moves one click per decade, while the mechanisms of institutional destruction move several clicks per session.
The six stories that follow are a map of which doors in American governance have opened in the past year, and which of them can never be closed. The first three describe a single operation in three phases: the destruction of a workforce, the profitable replacement of what was destroyed, and the legal limbo that erases the evidence. We begin with the people.
❧ ❧ ❧
DOGE at One Year: The Human Wreckage
352,000 federal workers fired in the largest peacetime workforce reduction in American history — the transition program placed 1.6 percent of applicants.
The numbers arrived first. Three hundred and fifty-two thousand federal workers separated from service in the largest peacetime workforce reduction in American history, a campaign executed under the banner of the Department of Government Efficiency and framed in the language of waste elimination and fiscal discipline. The administration called it a transition. The Office of Personnel Management built a workforce transition program to match displaced employees with new federal positions. And then the transition program produced its own number, the one that demolished the narrative: of roughly 12,800 federal employees who applied for new positions through the program, 190 were placed. A 1.6 percent success rate. The remaining 98.4 percent received no offer, no interview, and in many cases no acknowledgment that their application had been received. The transition was not a transition. It was a mass termination dressed in the vocabulary of human resources.
The stories behind that 1.6 percent began to surface in congressional testimony and investigative reporting through the winter of 2026. A CDC epidemiologist who had spent eighteen years tracking infectious disease outbreaks — building the network of county health officials, state lab directors, and international liaisons that constitutes America’s early-warning system for novel pathogens — was terminated in a batch processing event. She was subsequently hospitalized. Her pathogen-tracking projects were not transferred to other scientists. They were discontinued. The screens went dark. A VA counselor whose name was misspelled on the separation letter she received after nineteen years of service. A USAID officer terminated one day after returning from maternity leave. IT workers who watched unvetted credentials appear on databases containing the Social Security numbers of four million Americans.
Tom Frieden, who served as Director of the Centers for Disease Control and Prevention from 2009 to 2017, described the loss in terms that no policy document can capture: “The institutional knowledge walking out the door is irreplaceable. You cannot Google your way to knowing which county health official to call at 2 a.m. during an outbreak, or which lab has the right equipment for a novel pathogen. That knowledge exists in people, and when those people are gone, it is gone.”
The geographic impact extended far beyond the Washington metropolitan area. Stephen Fuller, Director of the Center for Regional Analysis at George Mason University, testified before the Joint Economic Committee that Virginia alone lost 23,500 civilian federal jobs between November 2025 and January 2026, erasing six years of employment growth in sixty days. Northern Virginia, where federal spending accounts for approximately 35 percent of GDP, had not experienced a contraction of this speed since the 2013 sequester — and the sequester cuts were one-fifth the magnitude.
The political defense of the reductions rested on the premise that the federal workforce was bloated and that the private sector would absorb displaced workers. But the data told a different story. Former federal workers were experiencing unemployment rates three times the national average eighteen months after termination. The agencies most critical to ordinary Americans — the Social Security Administration, the Internal Revenue Service, the Veterans Administration — were among the hardest hit, which is the opposite of what a competent restructuring would produce. The cuts were not targeted at underperformers. They were batch processing events based on hire date, agency, and position series.
The most telling criticism came not from Democrats but from within the president’s own party. Representative Nicole Malliotakis, a Republican from New York, said in a CNN interview: “We need to do this with a scalpel, I’ve said this repeatedly, not a sledgehammer.” She called some of DOGE’s moves “rash” and said “they need to be more targeted and they need to be more thoughtful.” Senator Susan Collins, a senior Republican from Maine, expressed concern about “the impact on the state of Maine, on everything from our national parks to biomedical research.”
The Partnership for Public Service, which has tracked federal workforce capacity for two decades, estimated that the institutional knowledge lost per terminated employee-year would cost 2.5 to 4 times the salary savings to reconstitute. Federal hiring timelines, which averaged 98 days before DOGE, were now estimated at 140 days or more. The ratchet’s logic was becoming visible: you can fire 352,000 people in weeks, but you cannot rebuild what they knew in years.
One year in, federal spending had not decreased. Service delivery metrics had worsened across measured agencies. The rehiring cycle had begun. The human wreckage was not a side effect of efficiency. It was the entire output.
“The institutional knowledge walking out the door is irreplaceable. You cannot Google your way to knowing which county health official to call at 2 a.m. during an outbreak.” — Tom Frieden, former Director, Centers for Disease Control and Prevention
For Further Reading: Perspectives
PRO “The Hidden Cost of Federal Workforce Reductions” — Max Stier, Partnership for Public Service Documents institutional knowledge loss and argues reconstitution costs vastly exceed salary savings. Source: Brookings Institution (February 2026)
CON “The Federal Workforce Needs Reform, Even if DOGE’s Methods Are Flawed” — Chris Edwards, Cato Institute Argues the federal workforce is genuinely bloated and that reform should continue with better targeting. Source: Cato Institute (February 2026)
But the human wreckage was only the first phase. The workers who lost their jobs did not disappear from the economy. They reappeared — in many cases doing the same work, at the same agencies, for the same government. The difference was on the invoice.
❧ ❧ ❧
The Boomerang: Fire at GS-13, Rehire at Contractor Rates
Every dollar in civil servant salary eliminated generates 1.60 in contractor spending within eighteen months — the efficiency program costs more than what it replaced.
The cycle works like this. Phase one: fire federal workers en masse, citing waste and inefficiency. Phase two: wait for the inevitable service degradation as institutional knowledge walks out the door. Phase three: blame the remaining workers and the previous administration for the degradation. Phase four: hire private contractors — often the same people, doing the same work — at rates that exceed the original civil service salaries by 30 to 60 percent. Each phase is individually defensible. The damage lives in the sequence.
The sequence became visible in the winter of 2026 with the precision of a controlled experiment. In January, the Social Security Administration sent voluntary separation offers that quickly became mandatory. The IRS sent termination letters targeting customer-service representatives hired during the prior administration’s modernization push. The Veterans Administration cut claims processors. In February, the consequences arrived: SSA phone wait times exceeded 90 minutes. The IRS suspended phone support in 14 regional offices. VA claims processing crossed 200 days. In late February, DOGE posted job listings for “operational specialists” at the same agencies — the same functions, lower pay grades, no civil service protections.
The fiscal arithmetic was devastating, and it came not from DOGE’s critics but from the think tank most ideologically aligned with its stated purpose. Chris Edwards, Director of Tax Policy Studies at the Cato Institute, published a first-quarter assessment that stated flatly: “Total federal outlays for FY2026 Q1 were 1.76 trillion for FY2025 Q1 — a 4.0 percent increase. This is consistent with CBO’s pre-DOGE baseline projection and shows no measurable impact from the workforce reductions on total spending. The $2 trillion savings figure cited by the administration was never modeled by OMB, CBO, or any other scoring authority.”
When the Cato Institute — not historically an ally of federal workforce expansion, not a progressive think tank, not a Democratic Party operation, but the intellectual home of the “government is too big” argument — confirms that spending rose on its prior trajectory as if 352,000 people had not been pushed out of their jobs, the efficiency justification does not merely weaken. It collapses from within.
The mechanism of the collapse was a labor arbitrage that the Project on Government Oversight had been documenting for over a decade. Scott Amey, POGO’s General Counsel, testified before the House Oversight Committee in February 2026 with data spanning 14 agencies and more than 200 contract conversions: “The federal government pays, on average, 180,000 to $210,000. The premium ranges from 30 to 60 percent depending on the specialization. There is no scenario in which replacing civil servants with contractors at scale reduces costs.”
Paul C. Light, who has studied government workforce capacity at New York University for thirty years, provided the structural explanation. His research documented that contractor-to-employee ratios at the Department of Defense, the Department of Homeland Security, and the Department of Health and Human Services were already at historic highs before DOGE — 3.7 to 1, 2.1 to 1, and 1.8 to 1 respectively. The post-DOGE ratios would exceed any prior measurement. His fiscal finding: every dollar in GS-13 salary eliminated generates 1.60 in contractor spending within eighteen months.
The defense contractor case study illustrated the political economy beneath the fiscal surface. DOGE proposed 47 million in lobbying. The cuts were reduced to $3.2 billion. The contractors’ positions were not merely preserved — they were strengthened, because the civil servants who had provided oversight of those contracts were among the workers who had been fired.
The boomerang was not a bug in an efficiency program. It was the program. The fire-rehire cycle converted public servants into private contractors, stripped them of whistleblower protections and merit-based employment, enriched staffing firms, and cost more. And the conversion, once executed, was structurally irreversible: nobody “un-contractors” a role once it has been moved to a private invoice. The ratchet turned. The door closed behind it.
“Outsourcing is not a way to shrink government. It is a way to hide government. The work still gets done. The money still gets spent. But the workers are invisible to democratic oversight.” — Allison Stanger, Professor, Middlebury College, author of “One Nation Under Contract”
For Further Reading: Perspectives
PRO “The True Cost of the Hollow Government” — Paul C. Light, NYU Thirty years of research documenting how contractor conversion increases costs while reducing accountability. Source: Brookings Institution (February 2026)
CON “Give DOGE Time: Why Short-Term Disruption Can Still Produce Long-Term Savings” — Romina Boccia, Cato Institute Argues that restructuring requires short-term cost increases and that long-term efficiency gains remain plausible. Source: Cato Institute (March 2026)
The workers who were fired became the workers who were rehired at higher cost. But there was a third category — a population that existed in neither column. Not fired. Not employed. Not rehired. Not counted. The government’s own personnel system had no code for what they were.
❧ ❧ ❧
The Phantom Workforce: Administrative Limbo
12,000 to 18,000 federal employees exist in a status the government’s own tracking system has no code for — court orders acknowledged but never executed.
The badge stops working on a Tuesday. The email account goes dead. No paycheck arrives, but no separation papers do either. The federal worker — a disability claims examiner with eleven years of specialization at the Veterans Administration — cannot collect unemployment because she has not been formally terminated. She cannot return to work because her credentials are revoked. She cannot plan because her status could change at any moment, or never. She exists in a condition that the federal personnel system was not designed to contain.
Simultaneously, in a federal courthouse, a judge issues a carefully reasoned reinstatement order. The agency formally acknowledges the order. Then nothing happens. The order is not defied. Defiance is visible, resistible, impeachable. Defiance creates a constitutional crisis with a remedy. This is something different. The order is simply not processed. The agency’s administrative machinery has no workflow for “reinstate a worker who was terminated through a process that did not follow standard separation procedures.” The silence between the judicial branch and the executive branch is not a confrontation. It is an absence — an absence that no existing legal mechanism was designed to address.
The Government Accountability Office gave it a name. In a report published February 23, 2026 — GAO-26-107, “Federal Workforce Transition: Administrative Gaps in Personnel Processing” — Comptroller General Gene L. Dodaro described the condition: “We identified what we are calling ‘administrative limbo’ — a condition affecting an estimated 12,000 to 18,000 federal employees who have received neither formal separation documents nor reinstatement processing, despite court orders or administrative rulings in their favor. The federal personnel system’s Standard Form 50 — the document that records every personnel action — has no transaction code for this condition. It is, in the most literal sense, a status that does not exist in the system designed to track all statuses.”
The SF-50 is not a metaphor. It is the actual form — Standard Form 50, Notification of Personnel Action — used by the federal government to record every hire, promotion, transfer, and separation. Every status a federal worker can occupy has a transaction code on the SF-50. Administrative limbo has none. The system cannot represent the condition it has been placed in.
The most damaging testimony came from inside the operation. Sahil Lavingia, founder and CEO of the technology company Gumroad, had joined DOGE as a temporary staffer assigned to the Veterans Administration. He entered expecting to find the waste and inefficiency that the program’s architects had promised. In an NPR interview, he reported what the data actually showed: “I was pretty surprised, actually, at how efficient the government was.” He found fraud to be “relatively nonexistent” and estimated that only 2 to 5 percent of workers were unproductive — “about the same as any well-run private company.” His access was revoked within days of making these statements publicly.
Twenty-one DOGE staffers had already reached the same conclusion. In a resignation letter reported by NBC News and Rolling Stone, they wrote: “We refuse to be party to the dismantling of critical public services that millions of Americans depend on… We joined DOGE to improve government, not to destroy it.”
The ratchet here was not the limbo itself but the precedent it established. Once it was demonstrated that a court order could be acknowledged and not executed, and that the consequences for this were — nothing — the mechanism was available to every future administration. The silence between branches had become a tool.
“Due process is not a luxury we extend when it is convenient. It is the baseline requirement for any government action that deprives a person of their livelihood.” — Everett Kelley, President, American Federation of Government Employees
For Further Reading: Perspectives
PRO “The Constitutional Gap: When Administrative Silence Replaces Defiance” — Elaine Kamarck Argues that passive non-processing of court orders represents a genuinely novel constitutional problem with no existing remedy. Source: Brookings Institution (February 2026)
CON “Agencies that fired 25,000 federal workers comply with court-ordered reinstatements” — FedScoop Reports that some agencies have complied with reinstatement orders, arguing the system is working as designed. Source: FedScoop (2025)
The first three stories describe a government hollowing itself out: firing the workforce, replacing it at higher cost, and erasing the evidence through administrative silence. But the hollowing requires something else to succeed — it requires that the public not see it clearly. The next story is about the mechanism that threatens to make the hollowing invisible.
❧ ❧ ❧
Media Capture in Real Time: The Paramount Deal
One family is about to control CBS, CNN, HBO, and TikTok — while three companies already control 40 percent of all local television news.
The president of the United States said, “It’s imperative that CNN be sold.” A bidder emerged whose father is Larry Ellison — co-founder of Oracle, one of the ten wealthiest people on earth, and a close ally of the administration. The son, David Ellison, CEO of Paramount Skydance, attended the State of the Union address as a personal guest of Senator Lindsey Graham. The $111 billion Paramount Skydance-Warner Bros. Discovery merger was advancing through regulatory channels staffed by appointees of the president who had demanded the sale. The regulators who would block a politically motivated acquisition worked for the person who would benefit from it. The journalists who would cover it worked for the company being acquired.
Political scientists have a name for this process: media capture. It describes the mechanism by which politically aligned actors acquire control of news organizations through ownership rather than censorship. The distinction is critical. Censorship is visible. It can be resisted, documented, litigated. Ownership is structural. It operates not through orders but through atmosphere — through editors who understand, without being told, which stories the ownership would prefer not to see.
Marius Dragomir, director of the Media and Journalism Research Center and the world’s leading academic authority on this phenomenon, has documented the process in Hungary under Viktor Orban with granular precision. His description of the mechanism could serve as a template for what is happening in the United States: “The pattern is always the same. First, ownership changes. Then, editorial leadership changes. Then, the stories that are not told become more important than the stories that are. Media capture is not about censorship. It is about the gradual narrowing of what is considered newsworthy.”
In Hungary, the transition from independent to captured media took approximately eight years and involved the acquisition of over 500 media outlets by government-aligned business interests. In every case, the acquisition was structured as a commercial transaction. In every case, the stated rationale was market rationality. The editorial changes that followed were never ordered. They were atmospheric.
The United States did not need to look to Hungary for evidence. It had CBS. Since the Paramount Skydance acquisition in mid-2025, senior 60 Minutes producers had quit citing editorial interference. Stephen Colbert clashed publicly with management over editorial direction. Journalists filed formal complaints. Seven CNN staffers told NBC News they were “shaken” by the prospect of the merger extending the same ownership structure to their network.
Penelope Muse Abernathy, a professor at Northwestern University’s Medill School of Journalism, had been tracking media ownership at the station level since 2004. Her findings: “Three companies — Nexstar Media Group, Gray Television, and Sinclair Broadcast Group — now control approximately 40 percent of all local television news stations in the United States, reaching an estimated 80 million households. In 2004, those same three companies controlled fewer than 15 percent.”
The First Amendment, which defenders of the deal invoked reflexively, was precisely irrelevant. The First Amendment protects against government censorship — against the state shutting down a newspaper or jailing an editor. It has nothing to say about a private owner’s influence over the editorial decisions of his own property. The distinction is the gap through which media capture passes.
Margaret Sullivan, former media columnist for the Washington Post and former public editor of the New York Times, placed CBS at approximately the midpoint of the Hungarian trajectory: not yet fully captured, but past the point where the editorial culture that would have resisted capture was intact. The senior producers who would have fought were gone. They did not stay to resist. They left. And their departure was itself the ratchet.
This was the meta-story of the edition, the skeleton key to every other article. DOGE’s workforce destruction, the contractor boomerang, the administrative limbo — all of these stories depended on an independent press to be visible to the public. Media capture was the power move that threatened to make all the other power moves invisible.
“The pattern is always the same. First, ownership changes. Then, editorial leadership changes. Then, the stories that are not told become more important than the stories that are.” — Marius Dragomir, Director, Media and Journalism Research Center
For Further Reading: Perspectives
PRO “The CBS Test Case” — Margaret Sullivan Former Washington Post media columnist documents how CBS editorial culture has shifted since the Paramount Skydance acquisition. Source: Substack (February 2026)
CON “The Media Capture Panic Is Overblown” — Jack Shafer Argues that media ownership changes are driven by market forces, not political capture, and that editorial independence has survived prior consolidations. Source: Politico (February 2026)
Media capture operates at the national level, changing what stories get told on the networks that reach eighty million households. But there is another mechanism that operates at the local level, changing what policies cities and counties are permitted to enact — and it has been spreading with a speed and coordination that most Americans have never heard of.
❧ ❧ ❧
The Model Legislator: Randy Fine and the Preemption Machine
82 percent of preemption bills filed within 45 days of the local ordinance they targeted — someone is watching city councils in real time and generating nullification legislation.
Randy Fine is a Florida state legislator whom almost no one outside Brevard County can name, and he has done more to reshape the relationship between local and state government in America than any single politician in the past two decades. His preemption bills — laws that strip cities and counties of the authority to set their own policies on issues from environmental regulation to labor standards — have been copied in 23 states, distributed as model legislation through the American Legislative Exchange Council, and filed more than 60 times in state legislatures in the first two months of 2026. He is now running for Congress.
Preemption itself is not new. States have always set the boundaries within which local governments operate. What is new — what Fine’s legislation exemplifies and what has transformed preemption from a governance tool into a weapon — is the penalty clause. Before 2019, preemption said “you cannot do this.” After 2019, it said “you cannot do this, and if you try, you personally pay.”
Richard Briffault, the Joseph P. Chamberlain Professor of Legislation at Columbia Law School, has tracked preemption legislation since his foundational 2018 paper. His numbers describe an acceleration: “We tracked 736 preemption bills across all 50 states between 2017 and 2025. The acceleration is dramatic: 89 new bills in 2017, 142 in 2020, 227 in 2023, and 360 in 2025 including carryovers. What changed in 2019 was the addition of penalty clauses — personal financial liability for local officials who enforce preempted ordinances.”
Mark Pertschuk, who runs the Preemption Watch project at Grassroots Change, provided the forensic evidence: “We analyzed the timing of 142 preemption bills filed in 2024 and found that 82 percent were introduced within 45 days of the local ordinance they targeted. In 23 cases, the state bill’s language was a near-verbatim inversion of the city ordinance — identical subject matter, identical definitions, opposite mandate. Someone is reading city council agendas in real time.”
The organization was ALEC — the American Legislative Exchange Council. Kim Haddow of the Local Solutions Support Center mapped the pipeline from ALEC drafting room to state legislature floor, analyzing text-similarity scores between ALEC templates and introduced legislation. The results: scores exceeding 90 percent in 67 of 142 analyzed bills. This was not policy diffusion. This was template distribution.
The most revealing data point came from ALEC’s own contradictions. The Local Solutions Support Center documented that ALEC simultaneously promotes state preemption of local minimum wage laws while supporting local right-to-work ordinances — endorsing local authority when it serves corporate interests and overriding it when it does not.
In Florida, SB 170 had been expanded to allow corporations to sue cities for enforcing their own existing local policies. In Missouri, bills were advancing that would criminalize local enforcement of red flag laws. A 12-city Florida coalition had filed the first organized legal challenge to systematic preemption.
Eighty-two percent of preemption bills filed within 45 days of the local ordinance they targeted. Someone was watching. Someone was responding. And the ratchet was turning: each preemption, once enacted, remained on the books regardless of who won the next election. The penalty clauses made the deterrent permanent. The door opened one way.
“We tracked 736 preemption bills across all 50 states between 2017 and 2025. What changed in 2019 was the addition of penalty clauses — personal financial liability for local officials.” — Richard Briffault, Joseph P. Chamberlain Professor of Legislation, Columbia Law School
For Further Reading: Perspectives
PRO “Preemption: The Playbook” — Kim Haddow, Local Solutions Support Center Comprehensive mapping of the pipeline from ALEC model bills to enacted state legislation with text-similarity analysis. Source: Local Solutions Support Center (January 2026)
CON “In Defense of Preemption: Why Regulatory Uniformity Serves Liberty” — Ilya Shapiro, Manhattan Institute Argues that regulatory patchwork burdens businesses and that state-level uniformity better serves economic freedom. Source: Manhattan Institute (January 2026)
The first five stories describe ratchets turning in one direction: institutional capacity destroyed, labor protections stripped, court orders ignored, media independence eroded, local governance preempted. Each door opened and closed behind it. But there is one room in the United States Congress where the ratchet is being turned the other way — slowly, quietly, and with results that nobody is talking about.
❧ ❧ ❧
The Quiet Room: Bipartisan Sentencing Reform
Thirteen years, one landmark law, a 71 percent reduction in recidivism, $880 million in annual savings — and the legislation has not passed.
In the loudest political era in living memory, in a Congress where the word “bipartisan” has become either a punchline or a eulogy, two senators have been doing something that the attention economy cannot see: governing.
Dick Durbin is a liberal Democrat from Illinois. Mike Lee is a conservative Republican from Utah. They have been working on sentencing reform together for thirteen years. Their partnership has already produced one landmark law — the First Step Act, signed in 2018 — and they are now proposing to extend its framework. The Smarter Sentencing Act would give federal judges the discretion to moderate sentences for nonviolent drug offenses based on individual circumstances. The Smarter Pretrial Detention Act would replace blanket pretrial detention with individualized judicial assessment. Neither bill repeals any mandatory minimum. Neither bill lowers any maximum sentence. A hearing before the U.S. Sentencing Commission is scheduled for March 9, 2026.
The evidence that the Durbin-Lee framework works is not theoretical. The U.S. Sentencing Commission’s five-year impact assessment of the First Step Act, published in December 2025, reported the results with the precision of a clinical trial: “The First Step Act has resulted in the early release of over 30,000 federal inmates since 2018. The three-year recidivism rate for First Step Act releases is 12.4 percent, compared to 43.3 percent for the general federal release population. That is a 71 percent reduction in recidivism. The program saves approximately $880 million annually in incarceration costs. These are not projections. These are measured outcomes from a law that already exists.”
The Congressional Budget Office estimated that the Smarter Sentencing Act would save approximately 39,158 per person. Mandatory minimums for nonviolent drug offenses — which accounted for 45.3 percent of the federal prison population — were set in 1986 and had been adjusted for inflation exactly zero times.
Mike Lee’s own explanation demolished the partisan frame: “I became interested in reforming the criminal justice system not in spite of, but because of the fact that I’m a conservative and a former prosecutor… Some mandatory minimum sentences are too harsh. And people are learning that a criminal justice system can be just as undermined by sentences that are too harsh.”
This was not compromise. It was convergence — two legislators arriving at the same conclusion from opposite starting points. The list of organizations supporting the Durbin-Lee legislation read like a roll call of the institutional right: Americans for Prosperity (the Koch network), Americans for Tax Reform (Grover Norquist), FreedomWorks, the R Street Institute.
The strongest objection was the fentanyl crisis, and the Brennan Center for Justice addressed it directly. Sixty-eight percent of federal fentanyl sentences were imposed on individuals with no prior violent offense and quantities consistent with personal use rather than distribution. The current mandatory minimums were written for the crack epidemic of the 1980s and did not distinguish between fentanyl traffickers and people caught with small quantities. Judicial discretion was precisely the tool needed to make that distinction.
Mark Osler, a professor of law at the University of St. Thomas and a former federal prosecutor, captured the paradox: “The odd thing about mandatory minimums is that almost nobody defends them anymore. Prosecutors have discretion to work around them. Judges resent them. The evidence shows they do not deter crime more effectively than shorter sentences. Defense attorneys obviously oppose them. Even the victims’ groups that once supported them have largely moved on. And yet the law persists. That is the most interesting thing about it — a policy that has lost its constituency but not its power.”
A policy that has lost its constituency but not its power. That was the ratchet’s most subtle form: not a door forced open by an identifiable actor, but a door held in place by the system’s own inertia. The quiet room was the one place in this edition where the ratchet turned toward repair. But it turned with excruciating slowness. Thirteen years. One landmark law. A 71 percent reduction in recidivism. $880 million in annual savings. And the legislation had not passed.
The quiet room kept working.
“I became interested in reforming the criminal justice system not in spite of, but because of the fact that I’m a conservative and a former prosecutor.” — Senator Mike Lee (R-Utah)
For Further Reading: Perspectives
PRO “The Case for Sentencing Reform in the Fentanyl Era” — Inimai Chettiar and Lauren-Brooke Eisen Data showing most federal fentanyl sentences target personal-use quantities with no prior violent offense. Source: Brennan Center for Justice (January 2026)
CON “Mandatory Minimums Are the Thin Blue Line Against the Fentanyl Flood” — William Barr, former U.S. Attorney General Argues mandatory minimums remain essential for deterring fentanyl distribution and protecting communities. Source: National Review (February 2026)
The quiet room is where this edition ends — but not because sentencing reform is the least important story. It is where this edition ends because the quiet room is the only place in these six stories where the ratchet turns toward repair. And the speed at which it turns, compared to the speed at which the other five ratchets have moved, is the final data point this edition offers.
❧ ❧ ❧
EDITORIAL
Editorial: The Sound of the Ratchet
Six stories. Six doors that opened in the past year. The question this edition asks is not whether what happened was good or bad — readers will have their own answers to that, and those answers will divide along familiar lines. The question is which of these doors, once opened, can never be closed. The answer shapes everything that follows, regardless of who wins the next election.
The institutional knowledge destroyed by the DOGE workforce reductions cannot be reconstituted by hiring new people, because the knowledge was not in the job descriptions. It was in the relationships, the institutional memory, the 2 a.m. phone calls between a CDC epidemiologist and a county health official during an outbreak. That knowledge took decades to build and weeks to destroy. The civil service protections stripped from positions converted to contractor roles do not return when administrations change, because the conversion is structural, not political — once a position sits on a contractor invoice, no subsequent president has the incentive or the mechanism to move it back. The precedent of administrative limbo — court orders acknowledged but not executed, the silence between branches weaponized as a tool of non-compliance — is now available to any future administration of any party. The journalists who left CBS are not coming back; the editorial culture they represented departed with them, and the building that remains is a building, not a newsroom. The preemption bills enacted in 23 states remain on the books regardless of who wins the next cycle, and the penalty clauses ensure that even local officials sympathetic to reasserting local authority face personal financial ruin if they try.
Against these fast-moving ratchets, the counter-ratchet of the quiet room — bipartisan sentencing reform that has produced one landmark law in thirteen years and may produce another — turns with excruciating slowness. The evidence is overwhelming. The support is bipartisan in the deepest structural sense: the Koch network and the Brennan Center agree. The First Step Act’s results are measured, replicated, and favorable by every metric its authors said they would track. And yet the legislation sits. A policy that has lost its constituency but not its power.
The megacategory for this edition is “How Power Actually Works.” The ratchet is the answer. Power works by creating changes that are structurally irreversible even when they are technically illegal, judicially blocked, or politically unpopular. It works by ensuring that the speed of destruction always exceeds the speed of repair. It works by making the doors that open in its direction impossible to close, while the doors that open toward reform require thirteen years of bipartisan effort to move a single notch.
The sound you are not hearing is the click of the ratchet. It is quiet. It is administrative. It looks like paperwork. And it is reshaping the structure of American governance in ways that will outlast every political figure mentioned in these pages.
The Review is a research publication by Daniel Markham. Published at danielbmarkham.com — Free to subscribers All stories sourced from publicly available documents, congressional testimony, GAO reports, think tank analyses, and investigative journalism published between 2018 and March 2026.
Production: TR2 > TR3 > TR4 pipeline — SVG infographics generated per edition — Layout specification v2
(c) 2026 Daniel Markham. All rights reserved. Editor: Daniel Markham | Submissions: letters@thereview.pub
THIS PUBLICATION IS NOT AVAILABLE TO THE GENERAL PUBLIC. IT IS AVAILABLE ONLY BY SUBSCRIPTION TO READERS WHO HAVE BEEN VOUCHED FOR BY EXISTING SUBSCRIBERS. THERE ARE NO INVITATIONS. IF YOU KNOW, YOU KNOW.