ANALYSIS: Washington Is Also Discarding the State’s Memory

A government can lose control

The number that refuses to remain abstract

214,000. This number doesn’t simply represent departures listed in a column. According to the Federal Harms Tracker as of August 18, 2026, it refers to civil servants who have left through voluntary or involuntary measures designed to downsize the U.S. government. The total is contentious, evolving, and open to debate regarding its scope. The wound it highlights, however, cannot be dismissed: an institution can lose employees much faster than it can pass on the knowledge they possess.

The Office of Personnel Management offers another perspective—one that is broader and more official. Its June data show a net decrease of 271,566 federal employees since January 20, 2025, with 139,928 participants in the deferred resignation program. These data sets do not count exactly the same things. Adding them together would be incorrect. Yet comparing them reveals the same trend: the federal government has contracted at an exceptional rate.

The Invisible Cutback

We see an empty office. We do not see the procedure that no one else understands, the case file whose progress depended on accumulated experience, the legal exception known only to a single team, or the network of contacts that made it possible to avoid a three-week wait. Operational memory is not a filing cabinet. It lives in people, habits, and channels of communication.

When this memory disappears without being passed on, the promised savings on payroll costs may come back in the form of delays, errors, and degraded services.

You can eliminate a position in a single day. You cannot recreate eleven years of knowledge by the following Monday.

Cutting isn’t a single action

Resignations, layoffs, attrition

The public debate seeks a single total. Yet the mechanisms overlap: layoffs, forced reductions, retirements, contract terminations, ordinary departures, and deferred resignation programs. OPM counts all personnel changes. The Federal Harms Tracker isolates departures and links them to reduction policies. OpenFeds reconstructs the sequence yet again, agency by agency. Three maps, three legends.

This difference matters. An administrative separation does not always mean a position has been eliminated. A laid-off employee may be reinstated. An agency may recall, rehire, or transfer staff. An announced resignation may not take effect for months. The exact figure for any given day is therefore never the final word. But statistical uncertainty does not negate the contraction; it requires us to describe it honestly.

The Resignation That Organizes Absence

The OPM memorandum of January 28, 2025, was clear on the program’s mechanics: volunteers’ tasks were to be reassigned or eliminated, and employees were then to be placed on paid administrative leave until their departure. That is the political point. The measure did not merely support individual choices; it ordered agencies to redistribute or eliminate work.

The word “voluntary” refers to the employee’s signature. It does not adequately describe the pressure, the uncertainty, or the institutional decision surrounding it.

The method diffuses responsibility across a thousand personal choices, even though the initiative comes from above.

Memory has an age

Eleven years don’t fit in a manual

At the OPM, more than half of those who left between December 2024 and March 2026 had at least eleven years of service, according to the Government Accountability Office report summarized by Government Executive. Eleven years mean system changes, revised rules, crises weathered, and mistakes acknowledged before they could be repeated. This knowledge rarely exists in a single document.

A healthy organization can prepare for this transition. It creates overlap, documents procedures, provides training, and maintains teams stable enough for newcomers to understand why their predecessors did things a certain way. A rapid downsizing reverses this order: first comes the announcement, then the handover—if time permits. Knowledge then becomes collateral damage that no one factored into the budget.

Human capital is not just a nice metaphor

The GAO had already warned, well before this wave, that gaps in critical skills could threaten OPM’s ability to meet its objectives. The agency tasked with helping the entire government manage its employees thus carried its own vulnerabilities. Abruptly dismantling this linchpin does not merely simplify the bureaucracy. It weakens the very place where the bureaucracy learns to repair itself.

A government without memory does not automatically become leaner. It becomes more dependent on crisis mode, consultants, improvisation, and reversible decisions.

Public knowledge is costly to build precisely because it cannot be purchased ready-made.

The OPM: A Laboratory of Contradiction

One-third Fewer

Between December 2024 and March 2026, the OPM reduced its workforce by 35%. Nearly 60% of the departures recorded during this period were through the deferred resignation program; involuntary layoffs accounted for 10% of the cuts. The agency maintains that it can continue to deliver its services. This assurance must be heard. It must also be put to the test.

Because the OPM is not a ministry detached from everyday life. It administers essential aspects of insurance programs, oversees retirement benefits, regulates hiring practices, and sets rules that affect millions of public employees, former employees, and their families. When its operations falter, citizens don’t hear the word “restructuring.” They simply wait.

Modernization as a Challenge

In its budget justification for 2027, the OPM is banking on artificial intelligence and IT modernization to accomplish more with fewer staff. This isn’t absurd. Many federal procedures need to be simplified. But a new tool doesn’t automatically know which exceptions protect a retiree, why a rule was created, or when an anomaly warrants human judgment.

Automation can lighten the load. It does not replace the responsibility, experience, or judgment that budget cuts take away.

Technology becomes dangerous when it is asked not to aid memory, but to make us forget that we have let someone go.

Retirement Knocks on the Door

A Retirement That Never Comes

Government Executive has documented former civil servants who were still not receiving their retirement checks months after leaving their jobs. A former IRS employee recounted having to dip into her savings, her 401(k) plan, and her credit cards while waiting. This account does not prove that every delay stems from budget cuts. It illustrates what an administrative delay means when the promised income is missing.

OPM’s retirement services division lost 16% of its staff between fiscal years 2024 and 2026. At the same time, the wave of departures automatically increased the number of cases to process. Fewer hands. More requests. This is a much more concrete chain of causality than any slogans about efficiency.

The cost shifts to the individual

A public savings can become a private debt. The agency spends less; the former employee pays interest. The budget looks better; the household postpones a purchase, withdraws savings, and endures the anxiety. This shift doesn’t always show up in political calculations because it isn’t listed under the eliminated federal budget line item.

Public services never disappear for free. When Washington stops shouldering a burden, someone else somewhere else often ends up shouldering it alone.

Waiting is the most discreet cost of a slimmer government.

FEMA Highlights the Risk Before a Disaster

Departures Without Sufficient Analysis

The GAO found that more than 4,300 employees left FEMA during fiscal year 2025, representing a 55% increase in separations year-over-year. Its most serious finding, however, is not the number itself. The oversight agency asserts that FEMA made workforce reduction decisions in 2025 and 2026 without basing them on an analysis of the capacity required to fulfill its mission.

FEMA prepares for, responds to, and helps rebuild after disasters. Its effectiveness is measured precisely on the days when everything else breaks down. A team may seem oversized during a lull and become insufficient when multiple emergencies strike. Cutting staff without modeling these peak demands is not prudent. It is banking on the next disaster being a mild one.

Preparedness isn’t a spectacle

The work that prevents a crisis from worsening attracts few cameras. Revised plans, contracts in place, local relationships maintained, drills repeated: none of this has the glamour of an announcement of budget cuts. Yet it is precisely this invisible infrastructure that allows the government to avoid having to figure out its own mission in the midst of an emergency.

An emergency response system cannot be judged solely by its actions when it intervenes. It must be judged by what it preserves before it is called upon.

You don’t appreciate the value of an umbrella during a week of sunshine.

Defense Also Pays the Price for the Void

173 programs to monitor

In a June report, the GAO concluded that the reorganization of the Office of the Director, Operational Test and Evaluation had reduced its capacity to oversee operational and live-fire tests of the 173 weapons programs on its list through 2025. Here again, the departure of employees does more than just reduce headcount. It increases the workload of those who remain.

This office does not manufacture weapons; it verifies that they function under realistic conditions. Cutting back on evaluation may create the illusion of speeding up the process. But the speed gained before testing risks turning into delays, higher costs, and vulnerability after the weapons enter service. Efficiency without oversight is a self-certifying promise.

Muscle and Nerve

Washington readily speaks of military power in terms of budgets, platforms, and ammunition. It sometimes overlooks the administrative nerve that determines whether a capability is ready, safe, and suitable. This nerve is less photogenic than an airplane. Its absence reveals itself at the worst possible moment, when the error can no longer be corrected by a memo.

A nation can purchase more equipment while losing the ability to understand what it is buying, what it is testing, and what risks it faces.

Power without memory is a force that no longer knows exactly where to set its foot.

The Myth of Uniform Fat

The government is not a monolith

Federal waste does exist. There are outdated programs, absurd rules, overlaps, and poorly designed positions. Refusing to acknowledge this would turn the defense of public service into a corporatist reflex. Donald Trump is right to demand measurable results. Where his administration fails is when it treats complexity as evidence of redundancy.

A federal workforce is not a homogeneous bloc. It includes meteorologists, engineers, retirement administrators, inspectors, scientists, prosecutors, procurement specialists, and emergency responders. Cutting based on headcount fails to distinguish between unnecessary paperwork and rare expertise. The percentage becomes a blind, indiscriminate axe.

Reform requires insight

Serious reform requires a mapping of work, workloads, skills, risks, and future needs. The GAO’s findings on FEMA indicate precisely the opposite: major decisions preceded the capacity analysis. A policy that announces its outcome before studying the mission is not an assessment. It is a preference disguised as a method.

Political courage isn’t about cutting everywhere. It’s about identifying what must go and standing by what absolutely must remain.

An axe doesn’t become a scalpel just because you give it a dashboard.

The official number tells a different story

271,566 fewer employees, but not 271,566 layoffs

OPM data shows a net change of 271,566 fewer employees since Trump’s return. This figure includes both hires and departures. It does not mean that 271,566 people were laid off. This clarification does not weaken the finding; it makes it actionable. Attrition, retirements, and deferred resignations can reduce headcount just as surely as a termination letter.

OPM also notes that a change in how the Department of War processes data prevented certain components from submitting their June 2026 figures. The official total, therefore, has its own limitations. A government undergoing restructuring must be able to measure what it is becoming. Here, even the snapshot requires a footnote.

The Pitfall of Competing Totals

The Federal Harms Tracker cites 214,000 targeted departures and 467,057 total separations from January 2025 through June 2026. OPM reports a distinct net decrease. OpenFeds offers its own reconstruction. These numbers are not at odds with one another; they answer different questions. Confusing them creates an easy scandal. Distinguishing them reveals a deeper problem.

We still don’t know the final tally for each program, but we already know that the reduction goes far beyond a mere marginal cleanup.

Uncertainty about the exact size of the shortfall does not mean the floor is intact.

Those who remain inherit the void

The workload doesn’t just vanish

When a task isn’t officially eliminated, it migrates. A colleague takes over the file. A team absorbs the urgency. A manager signs off on more decisions with less time to understand them. The institution then maintains the appearance of service by drawing on the human reserves of those who remain. These reserves are not infinite.

Fatigue increases the risk of error, but the danger isn’t limited to burnout. Overload leads to shortcuts: fewer checks, less mentoring, less documentation, and greater dependence on a few indispensable individuals. The next resignation then hits harder. One round of cuts sets the stage for the next.

The Silence of Lost Skills

A lost skill doesn’t always set off an alarm. Sometimes, no one notices the absence until an unusual case comes up. The processing time lengthens. The decision gets escalated too high up. The user calls back. Then the organization discovers that it once had an answer that is now nowhere to be found.

True administrative collapse isn’t always spectacular. It often begins with a simple question that no one knows how to answer quickly enough anymore.

What leaves with a civil servant doesn’t wear a badge after they’re gone.

A lean government can cost more

Cutting payroll, paying for contracts

An administration deprived of in-house expertise often has to buy back what it once knew how to do. Consultants can provide useful expertise. They can also become the external memory of a government that has laid off its own staff. The contract then replaces the salary, sometimes with less continuity, less direct control, and greater dependence.

It cannot be claimed that every cut will result in a higher bill; the data does not support that. But the history of public administration makes the mechanism plausible: if the mission remains and internal capacity disappears, someone will do the work—or no one will. In either case, the initial savings are incomplete.

The Cost of Rebuilding

Recruiting after a purge takes time. Empowering, training, transferring knowledge, and rebuilding relationships with states, suppliers, and beneficiaries takes even more. An administration can refill positions. It cannot instantly regain the trust or the accumulated expertise it has lost. Legal reversibility therefore does not guarantee operational reversibility.

The government can correct an organizational chart. It cannot order lost experience to return on a chosen date.

Public memory is slow to build and terribly quick to disperse.

The citizen becomes the sensor

When the service measures the cut

The debate over staffing levels will remain abstract until delays, decisions, or errors affect enough people to become visible. Retirees waiting for their checks are already sensors. Communities that depend on FEMA will become sensors at the next crisis. Service members and taxpayers become sensors when the evaluation of an arms program reveals a loss of capability.

This way of discovering the consequences is unfair. It turns users into guinea pigs for a reform they didn’t design. If the service holds up, the government claims success. If it breaks down, the citizen first absorbs the damage, then waits for an investigation to determine the cause.

Delay as Policy

A delay is never neutral for those who depend on a payment, a permit, assistance, or a decision. It can mean lost interest, a deferred rent payment, a lost contract, or a suspended reconstruction project. None of these effects, on its own, proves national failings. Together, they delineate the territory where the state becomes a tangible reality.

Efficiency is not measured by the number of closed offices, but by the number of lives that move forward without being held back by a closed door.

Citizens do not receive a reduction in staff. They receive a response—or they do not.

The Executive Branch and Accountability

Trump Chose Speed

The reduction of the civil service is a deliberate political choice of the second Trump administration. Its defenders see it as a correction of a bloated and overly autonomous bureaucracy. This ambition deserves better than to be caricatured. Above all, it deserves to be judged on its methodology: what missions, what competencies, what net savings, what timelines, and what risks?

When the GAO notes a lack of sufficient analysis at FEMA or reduced capacity in weapons evaluation, the administration cannot respond solely with intentions. To govern is to bear the foreseeable consequences of a decision. The fight against bureaucracy does not create an exception to this rule.

The Lack of Transparency

Government Executive reports that the GAO requested documents from OPM regarding its reorganization and the reasons for closures or consolidations. The agency had not provided this information or responded to written questions prior to the report’s publication, but subsequently submitted additional information. This delay prevents oversight from keeping pace with the speed of reform.

A government that demands performance from its civil servants must accept that its own reform be measured, documented, and challenged to the same degree.

Trust does not arise from a promise of efficiency. It arises from proof that one has accounted for what might be broken.

The counterargument must remain intact

Yes, the government can reform itself

Defending institutional memory must not become a sanctification of every position. Some processes are obsolete. Layers of approval slow things down without providing protection. Well-established technologies can reduce repetitive work and make services more accessible. Refusing any cuts would be just as irresponsible as making cuts without a proper assessment.

OPM claims to have confidence in its ability to manage the annual insurance enrollment period. It promises that digitizing pension processes will shorten processing times. These claims deserve a chance to prove themselves. A successful reform can indeed make a small team more effective than a large, poorly equipped organization.

But the proof comes after the slogan

The question, therefore, is not a choice between stagnation and demolition. It is about demanding that every cut be tied to a task that has been successfully eliminated, simplified, or automated. As long as delays persist and oversight agencies report missing analyses, the benefit remains a hypothesis, while the risk is already well-documented.

Cutting may be a form of governance. But cutting before knowing what still stands is simply gambling with the service provided to others.

A reform becomes credible the day it can explain not only what it takes away, but also what it protects.

The real test begins now

Counting What’s Missing

The coming months will reveal the difference between real gains and superficial savings. We’ll need to track retirement timelines, FEMA preparedness, defense assessments, errors, appeals, rehires, and external contracts. No single indicator will suffice. It’s the combination of these factors that will reveal whether the system has slimmed down or simply grown weaker.

We’ll also need to keep the numbers separate. 214,000 is not 467,057. A net decrease is not the total number of layoffs. A legal reinstatement does not necessarily erase a termination. Rigor is not a favor done for Trump. It is what allows us to attribute to him exactly what is his due.

The Last Chair

At the end of this story, there is no abstract war between bureaucrats and reformers. There is a chair that remains empty at the very moment someone needs to know the rule, understand the exception, open the file, and respond. If no one can do so, the citizen may never know which exit program created the void.

They will only know that the government, once capable of remembering on their behalf, now asks them to wait while it searches for its own memory.

214,000 departures. And how many answers left with them?

Signed, Maxime Marquette, columnist

Columnist’s Transparency Box

Editorial Stance

I am not a journalist, but a columnist and analyst. My expertise lies in observing and analyzing the geopolitical, economic, and strategic dynamics that shape our world. My work consists of dissecting political strategies, understanding global economic trends, contextualizing the decisions of international actors, and offering analytical perspectives on the transformations that are redefining our societies.

I do not claim to possess the cold objectivity of traditional journalism, which is limited to factual reporting. I strive for analytical clarity, rigorous interpretation, and a deep understanding of the complex issues that affect us all. My role is to make sense of the facts, situate them within their historical and strategic context, and offer a critical analysis of events.

Methodology and Sources

This text respects the fundamental distinction between verified facts and interpretive analysis. The methodological rule is consistent: factual information is published only if it is supported by a verifiable source, and the sources actually used in this article are listed under “Sources,” never here.

Categories of primary sources used by the publication, when applicable: official press releases from governments and international institutions, public statements by political leaders, reports from intergovernmental organizations, and dispatches from recognized international news agencies.

Types of secondary sources: specialized publications, internationally recognized news media, analyses by established research institutions, and reports from sector-specific organizations.

When an article cites statistical, economic, or geopolitical data, it comes from data-producing institutions (intergovernmental organizations, central banks, national statistical institutes), and the specific institution is listed under “Sources.”

Nature of the Analysis

The analyses, interpretations, and perspectives presented in the analytical sections of this article constitute a critical and contextual synthesis based on available information, observed trends, and expert commentary cited in the sources consulted.

My role is to interpret these facts, contextualize them within the framework of contemporary geopolitical and economic dynamics, and give them coherent meaning within the broader narrative of the transformations shaping our era. These analyses reflect expertise developed through continuous observation of international affairs and an understanding of the strategic mechanisms that drive global actors.

This article describes a situation documented as of its publication date, not a prediction: subsequent developments may alter these perspectives. No updates are promised in advance; when an article is corrected or supplemented, the change is dated within the text.

ANALYSIS: Washington Is Also Discarding the State’s Memory