Across the federal government during the first week of October 2026, thousands of branch chiefs, division directors, and senior career executives sat down in conference rooms to carry out an instruction that violates the foundational premise of professional management. Their task was not to assess how well their teams had executed the statutory responsibilities assigned to them by Congress. It was to take existing, written performance appraisals—appraisals grounded in measurable tasks, from processing patent claims to monitoring infectious disease—and forcibly reduce the marks of employees who had met or exceeded every standard expected of them.

The administrative mechanism driving this process is a binding memorandum issued on September 21, 2026, by Scott Kupor, the director of the Office of Personnel Management. The directive, titled FY 2026 Standardized Distribution of Rating Levels and FY 2026 Performance Awards Guidance for Federal Employees, arrived nine days before the September 30 close of the fiscal year. It orders agency heads to ensure that no more than 40 percent of career General Schedule, Senior Professional, and Senior Executive Service personnel receive either Level 5 (“Outstanding”) or Level 4 (“Exceeds Fully Successful”) performance ratings. If an office has ten employees who each performed exemplary work during the past twelve months, at least six of them must, as a matter of mathematical decree, be pushed down to Level 3 (“Fully Successful”) or below.

To enforce compliance, agencies were instructed to convene internal “calibration programs” immediately following the fiscal year close. Between October 1 and October 7, 2026, those calibration panels began auditing and altering rating sheets across the civilian bureaucracy. The exercise represents the formal importation of corporate “stack ranking” into the General Schedule, dismantling nearly half a century of administrative safeguards designed to ensure that federal civil servants are evaluated on objective merit rather than managerial quotas.

The Anatomy of Forced Calibration

The institutional machinery behind this transition was assembled through formal rulemaking earlier in the summer. On July 7, 2026, the Office of Personnel Management published a final rule in the Federal Register (91 FR 41521, RIN 3206-AP06), which went into effect on August 6. The rule repealed a longstanding provision of the Code of Federal Regulations, 5 CFR § 430.208(c), which since the aftermath of the Civil Service Reform Act of 1978 had explicitly barred federal agencies from establishing forced distributions or rating quotas.

For more than four decades, the governing standard under the Civil Service Reform Act (codified at 5 U.S.C. § 4302(c)(1)) demanded that performance appraisal systems include criteria that “to the maximum extent feasible, permit the accurate evaluation of job performance on the basis of objective criteria related to the job in question.” In 1979 and again in 1995, OPM concluded that preestablished rating distributions were fundamentally incompatible with merit principles because they graded employees relative to one another rather than against their assigned responsibilities. The 2026 regulatory overhaul inverted that logic, authorizing OPM to impose “a standardized distribution of some or all rating levels” and permitting agencies to rank workers comparatively.

To prevent workers from contesting these forced downgrades, the administration simultaneously instituted 5 CFR § 430.208(k). This clause stripped federal employees of the right to challenge their performance rating of record through negotiated grievance procedures under their collective bargaining agreements. An employee whose evaluation is downgraded during this week’s calibration panels cannot file an administrative grievance through their union. Their sole recourse is an informal, non-binding agency reconsideration request submitted within sixty days.

The consequences of this arithmetic exercise extend far beyond annual bonus checks. In August 2026, OPM finalized revisions to federal Reduction in Force retention regulations under 5 CFR Part 351, which took effect on September 2. Under the new formula, performance ratings effectively displaced tenure and seniority as the primary factor governing who survives agency downsizing. A Level 5 evaluation awards a worker seven retention points on agency registers; a Level 4 awards five points; a Level 3 awards only three points. Because performance scores are averaged over a rolling multiyear window, an arbitrary downgrade imposed during the current October calibration cycle will depress an employee’s retention standing for up to four years, making them substantially more vulnerable to pending workforce reductions.

The Official Defense and Its Premise

Director Kupor and his aides have defended the 40 percent quota as a necessary corrective to chronic “ratings inflation.” Kupor, a venture capitalist who served as managing partner at Andreessen Horowitz before taking the helm at OPM in July 2025, has argued that the General Schedule had deteriorated into an administrative environment where meaningful distinctions were rarely drawn. In a July 20, 2026, interview on Fox News’s The Ingraham Angle, Kupor pointed out that across 2.2 million civilian workers, only 0.3 percent were ranked below expectations in recent appraisal cycles, while approximately 70 percent received ratings of Level 4 or Level 5.

“Inflated ratings carry no meaningful signal about who is performing at an exceptional level, deprive agencies of the information needed to direct the largest rewards to the employees who have earned them, and dilute awards across the workforce at the expense of top performers,” Kupor wrote in his September 21 memorandum. “Distribution caps are intended to address these concerns by establishing a governmentwide ceiling on the share of employees who may receive the highest rating levels.”

The institutional diagnosis behind this argument is not entirely without substance. Anyone who has observed large bureaucratic structures knows that supervisors often take the path of least resistance, awarding elevated marks to avoid confrontation, paperwork, or administrative conflict. Spreading performance awards evenly across three-quarters of a staff can blunt the incentive for extraordinary exertion.

Yet the policy solution adopted by OPM mistakes a training and supervisory deficit for a structural necessity. It assumes that the distribution of actual human competence within a highly specialized, selective workforce naturally conforms to a predetermined statistical distribution. In doing so, it confuses the statistical tool with the underlying reality.

The Lessons of Stack Ranking

The intellectual model for OPM’s initiative does not originate in the civil service. It is a direct copy of the “vitality curve” popularized during the 1980s by Jack Welch at General Electric. Welch mandated that GE business units divide employees into a rigid distribution: the top 20 percent, the middle 70 percent, and the bottom 10 percent, with the bottom tier routinely terminated. The management method, commonly called “stack ranking” or “rank and yank,” was widely adopted by corporate leadership during the 1990s and 2000s, most famously at Enron and Microsoft.

The trouble with importing this corporate model into the federal civil service in 2026 is twofold: first, the private sector has spent the past decade abandoning it because it corroded internal operations; second, the fundamental operating conditions of public institutions make it uniquely toxic to governance.

The corporate history is instructive. In November 2013, Microsoft eliminated stack ranking after extensive internal research revealed that forced distribution had crippled the company’s engineering culture. When engineers knew that only a fixed percentage of their team could receive high marks, they actively avoided collaborating with exceptionally talented peers, declined to share critical information, and sabotaged collective projects to protect their individual rankings. Two years later, in 2015, General Electric itself phased out Welch’s vitality curve, recognizing that forcing knowledge workers onto a fixed bell curve discouraged long-term innovation and damaged trust between managers and their teams.

Where the corporate comparison completely breaks down, however, is in the nature of public service. A commercial firm produces consumer goods or financial returns; underperforming business units can simply be closed or spun off. A federal agency administers public trusts and constitutional duties that demand seamless, long-term cooperation and rigorous adherence to law. Air traffic controllers, wildfire dispatchers, food safety inspectors, and patent examiners do not generate commercial profit margins; they protect life, property, and constitutional processes through collective vigilance.

Consider the practical absurdity of applying a 40 percent cap to NASA’s Artemis 2 flight crew, which completed a historic lunar flyby in April 2026. Under OPM’s forced distribution mandate, NASA management could not award all four astronauts a Level 4 or Level 5 appraisal, regardless of how flawlessly they executed their flight profile. By definition, two members of that four-person crew would have to be marked down to Level 3 to satisfy an administrative ceiling designed in Washington.

The distortion is equally visible among front-line technical professionals. At the US Patent and Trademark Office, patent examiners work under quantitative production goals established in collaboration with the Patent Office Professional Association. An examiner who processes 120 percent of their statutory quota with an unblemished quality record has fulfilled every objective standard for an Outstanding rating. Yet under the 40 percent cap, if other colleagues within the same art unit also met that benchmark, managers are forced to arbitrarily reclassify fully documented excellence as merely average.

Eradicating the Merit System

The deeper danger of OPM’s directive lies in how it intersects with the history of the American civil service. In 1883, Congress passed the Pendleton Act to end the spoils system, which had treated the federal workforce as political plunder and culminated in the assassination of President James A. Garfield by a disgruntled patronage seeker. The enduring insight of the Pendleton Act was that a democratic republic requires an enduring, competent, and nonpartisan civil service whose tenure is insulated from political whim.

Nearly a century later, President Jimmy Carter signed the Civil Service Reform Act of 1978 to modernize personnel management while reinforcing those merit principles. The 1978 law established clear statutory rights: workers must be judged against objective standards published in advance, and they retain the statutory right under 5 U.S.C. § 7121 to grieve arbitrary or unfair supervisory actions through binding arbitration.

The Kupor memorandum, coupled with the July final rule, creates an administrative back door around those statutory protections. When a manager must downgrade six out of ten employees regardless of their actual performance, the selection of who receives the protected higher ratings inevitably becomes subjective. Without objective criteria to differentiate among qualified workers, ratings inevitably track personal favoritism, managerial comfort, or political conformity.

As the Coalition to Protect America’s National Parks noted in its formal submission to OPM during the public comment period, forced ranking introduces a pervasive institutional chill. In an agency where natural resource management, environmental assessments, and public safety require transparent reporting of inconvenient scientific facts, forcing staff to compete against one another for an artificially scarce pool of survival ratings incentivizes silence over candor.

Everett Kelley, national president of the American Federation of Government Employees, captured the stakes succinctly when the rules were promulgated: “These changes are a major departure from how agencies currently assess the performance of their employees and are a violation of our negotiated contracts that stipulate how workers are evaluated and disciplined. This is yet another illegal attempt by this administration to eviscerate our nation’s civil service and make working conditions so untenable that federal employees will quit.”

The Structural Remedy

The institutional resistance to this policy has already moved to the federal courts. On August 24, 2026, four major labor unions—AFGE, the American Federation of State, County and Municipal Employees, the National Federation of Federal Employees, and the International Federation of Professional and Technical Engineers—filed suit in the US District Court for the Northern District of California (AFGE et al. v. OPM, Case No. 3:26-cv-08835). The complaint challenges the legality of the final rule, arguing that forcing ratings into a quota violates the plain text of 5 U.S.C. § 4302(c)(1) and that stripping negotiated grievance rights runs afoul of 5 U.S.C. § 7121.

Legal challenges, however, address only the symptoms of administrative distress. What would a genuinely functional approach to federal performance management look like?

A serious reformer would begin by recognizing that the federal government does not suffer from an excess of objective standards, but from an underinvestment in basic management capacity. Dealing with genuine underperformance requires time, documentation, and supervisory courage under existing statutory mechanisms—specifically Chapter 43 and Chapter 75 of Title 5. Federal supervisors rarely receive comprehensive, ongoing training in how to establish precise, measurable milestones, how to counsel struggling employees, or how to build legally defensible performance improvement plans. Facing heavy administrative workloads, many choose to sign off on acceptable ratings rather than embark on months of procedural documentation.

Imposing an arbitrary statistical ceiling does not solve that managerial failure; it merely absolves managers of the responsibility to exercise sound judgment. It replaces deliberate, evidence-based supervision with an algorithmic meat-cleaver.

A healthier civil service requires three structural commitments: First, performance standards must be co-designed with front-line practitioners and based on observable, mission-critical outcomes rather than comparative quotas. Second, agencies must invest in rigorous supervisory academies that train career managers how to document underperformance, provide constructive feedback, and execute adverse actions when warranted under the law. Third, statutory due process and grievance rights must be preserved not as bureaucratic luxuries, but as indispensable feedback loops that keep agencies focused on competence and insulated from personal bias or political patronage.

Agencies have until November 20, 2026, to submit formal waiver applications to Director Kupor if they believe their statutory mission cannot withstand the 40 percent ceiling. But the damage is already unfolding in the calibration panels convening this week. When an administrative system tells public servants that performing their duties flawlessly is no longer enough to secure a fair evaluation, it does not elevate excellence. It institutionalizes cynicism, breaks down collaborative enterprise, and weakens the civic foundations of the American state.