Summary

  • The Center for American Progress attributes rising National Labor Relations Board dismissal rates to a combination of severe staffing losses and new administrative protocols.
  • The agency recorded a 34.7 percent dismissal rate for union-filed unfair labor practice charges and a 67.4 percent rate for worker-filed charges during the first 16 months of the Trump administration.
  • The board experienced a 345-day loss of its decision-issuing quorum alongside a net workforce reduction exceeding 10 percent.
  • The new general counsel issued directives emphasizing settlement over litigation and requiring charging parties to submit substantial supporting documentation within two weeks of filing.

The Center for American Progress, in an analysis of more than 40,000 National Labor Relations Board cases, reports a 34.7 percent dismissal rate for union-filed unfair labor practice charges during the first 16 months of the Trump administration—a 14.2-percentage-point increase from 2024—and a 67.4 percent dismissal rate for worker-filed unfair labor practice charges over the same period, a 10.7-percentage-point increase. The analysis attributes the rise to a stack of overlapping causes: a 345-day loss of board quorum following the firing of Gwynne Wilcox, the first Black woman to serve on the board; a net workforce reduction of more than 10 percent, approximately 150 employees lost against eight added; a Bloomberg-reported 23-percent understaffing rate at regional board offices; February 2026 settlement-first guidance from administration-appointed General Counsel Crystal Carey, a former attorney at the union-avoidance law firm Morgan Lewis; and December 2025 intake protocols requiring charging parties to submit substantial supporting documentation within two weeks of filing. The board declined to comment on the analysis.

Documented causal factors and operational disruptions

The documented causal factors for the increased dismissal rates span structural disruptions, staffing levels, and procedural directives. The board lost its decision-issuing quorum for 345 days following the removal of board member Gwynne Wilcox. This 345-day figure is consistent across the Guardian, multiple legal-trade publications, and the Center for American Progress analysis. Regarding staffing, a February 2026 Bloomberg report found field operations were understaffed by 23 percent. The agency lost approximately 150 workers in 2025 and added eight, representing more than 10 percent of the total workforce. These figures are drawn from the February 2026 Bloomberg report and are repeated in the Guardian and the Center for American Progress analysis.

Administrative directives also changed during this period. Crystal Carey, a former attorney at the union-avoidance law firm Morgan Lewis, was appointed general counsel. In February 2026, she issued guidance emphasizing case resolution through settlements rather than litigation. In December 2025, the agency enacted intake rules requiring charging parties to submit substantial supporting documentation within two weeks of filing. Labor lawyers reportedly cited docketing system glitches as a factor in case dismissals, though the source article does not quantify the contribution of these technical issues.

Broader labor metrics shifted concurrently. National Labor Relations Board-overseen union election filings declined 30 percent in 2025, a figure corroborated by the Center for American Progress, the Guardian, Ogletree’s Fiscal Year 2025 petition statistics brief, and Supply Chain Brain. Separately, the administration has sought to cancel collective bargaining agreements covering more than 1 million federal workers, an effort that operates outside the unfair labor practice process.

Competing explanations for the dismissal surge

Multiple explanations compete to account for the surge in dismissals, each carrying distinct diagnostic signatures. The discretionary policy hypothesis posits that Carey’s settlement-first guidance and the December 2025 two-week evidence rule are discretionary policy choices operating at the case-intake stage where dismissal rates are measured. The observable signature for this hypothesis is dismissals concentrated at intake rather than after investigation, with the rise proportional to the share of cases failing the documentation threshold. Center for American Progress analyst Aurelia Glass stated the procedural changes “at their core make it easier for charges to be dismissed.”

The resource-collapse hypothesis suggests that a 10-percent workforce loss combined with a 345-day decision-making shutdown would force regional directors to triage by dismissing rather than investigating. The observable signature is a rise correlated with backlog growth and case-processing time, rather than with the documentation-rule threshold.

The null hypothesis, or charge quality hypothesis, proposes that the underlying pool of incoming charges may have deteriorated in quality, meaning a similarly staffed agency would have dismissed a similar share. The observable signature is that a 30-percent decline in union election filings may have shifted the composition of filed unfair labor practice charges, with the remaining charges disproportionately filed by individual workers without union legal backing and therefore structurally more vulnerable to the new two-week rule. The article does not report data controlling for charge composition.

The docketing-glitch hypothesis suggests systemic docket errors would be expected to produce a roughly random distribution of dismissals across case types and over time, rather than the uniform directional increase in both union and worker charges that the article reports. This hypothesis is consistent with the data but not diagnostic of it, and the contribution remains unquantified.

The working hypothesis synthesizes these mechanics, suggesting the available evidence is most consistent with a compounded effect of the resource-collapse and discretionary-policy mechanisms. Under this view, the loss of a quorum and severe understaffing created a structural imperative to reduce docket size, while the new two-week documentation protocol provided the procedural mechanism to execute that reduction efficiently. This conclusion is tentative and not adjudicated by the data the article reports.

Stakeholder interests and channel structure

The stakeholders in this dispute hold distinct interests operating through specific channels. Workers and unions possess primary interests in enforcement access, particularly during organizing drives when retaliation is most common, the integrity of the certification process, and the statutory mandate of the National Labor Relations Act. Center for American Progress analyst Aurelia Glass framed the dismissals as threatening workers “who depend on the NLRB to be able to enforce these laws.” The best alternative to a negotiated agreement for these parties is severely constrained; the board is the exclusive federal forum for many unfair labor practice claims, and the alternative channels of public pressure, economic strikes, and legislative action carry high costs and uncertain outcomes. Collective action through organizing remains possible but is itself contingent on board-run elections, which fell 30 percent in 2025.

National Labor Relations Board leadership and the administration demonstrate documented conduct through the appointment of Carey, the issuance of settlement-first guidance, and the adoption of December 2025 intake protocols. The criteria invoked for these actions include resource constraints, budgetary realities, and the substantial evidence standard at the intake phase. The best alternative to a negotiated agreement for this coalition is maintaining the procedural framework as constituted through continued executive-branch appointments and guidance issuance.

Employers hold interests in the minimization of regulatory burden, predictable case resolution, and the avoidance of protracted litigation. A settlement-first regime and a higher evidentiary bar at intake align with these interests by reducing the expected cost and duration of defending unfair labor practice charges. The employer-side interest is not voiced in the source article.

Structurally, the parties lack a shared negotiation channel; the dispute is conducted through administrative procedure rather than integrative bargaining.

Second-order effects and remedy divergence

The second-order effects of these dismissal rates involve deterrence recalibration and divergent remedy implications. Under standard deterrence models, if the expected cost of committing an unfair labor practice approaches zero due to high procedural dismissal rates, rational actors may increase the frequency of the behavior. The agency may achieve its immediate goal of docket reduction at the cost of diminishing the deterrent function of the National Labor Relations Act, potentially generating a larger, unmeasured volume of unfiled or unresolvable labor disputes.

Furthermore, the discretionary-policy reading, the resource-collapse reading, and the charge-quality reading each point to different remedies, namely procedural reversal, resource restoration, or no remedy. The data the article reports does not adjudicate between these competing diagnostic readings.

Methodological gaps and source representation

Evaluating the analysis requires examining methodological gaps and source asymmetries. The Center for American Progress analysis reportedly draws on 40,000 cases, but the article does not specify how the analysis defined dismissed. Administrative closures, withdrawals, and settlements are typically counted together with merit dismissals in board reporting; if the analysis is treating all non-investigated closures as dismissals, the rate reflects the agency’s intake posture more than its substantive judgments. Additionally, the article does not report appeal outcomes, settlements, or reinstatement statistics for cases that did proceed, which is the data series that would let a reader distinguish a substantive drop in enforcement from a procedural drop in initiation.

A baseline gap also exists. The article does not supply a pre-2024 dismissal-rate baseline stratified by filing type, so the claim that the observed rates are elevated relative to a properly-resourced agency cannot be independently checked from the source materials. This relates to the risk of input versus filter conflation. A 30-percent decline in union election filings might correlate with a shift in the types of unfair labor practice charges being submitted. The analysis reportedly does not control for changes in the volume or nature of charges filed, risking the conflation of a change in the input profile with a change in the processing filter.

Source representation asymmetry shapes the available information. Center for American Progress analyst Aurelia Glass is quoted at length, stating, “Workers who are trying to organize unions already really face an uphill battle because employers, they really get away with a slap on the wrist, even when they do break the law.” Conversely, the board declined to comment, meaning its institutional view is not represented. The article’s selection of subject matter and omission of the employer-side interest in case-quality filtering is consistent with the labor-side orientation of the Center for American Progress. Symmetric application of the analytical standard would require engaging with the possibility that some portion of the dismissed caseload was non-meritorious, and that a higher dismissal rate could in principle reflect better intake triage rather than worse enforcement. The article does not engage with that possibility, and the source materials do not supply the data that would let a reader resolve it.

Unresolved diagnostic requirements

Adjudicating the diagnostic questions raised by the competing hypotheses requires several unresolved data points. These include reason-code breakdowns distinguishing merit dismissals from administrative closures, withdrawals, and settlements. They also include appeal-disposition data for cases that proceeded past intake, and a pre-2024 dismissal-rate baseline stratified by filing type. Furthermore, charge-composition data is needed to control for the shift in unfair labor practice filings that may accompany the 30-percent election decline, alongside a quantified estimate of the docketing-glitch contribution to total dismissals. Finally, the board’s institutional view remains absent and unrepresented in the current reporting.

Analytical techniques used in this piece

This analysis applies the methods below. Each links to a short, plain-English explainer you can read and reuse.

Differential Diagnosis
Lists the candidate explanations for a symptom and rules them out one by one.
Principled Negotiation
Works a negotiation from interests, options, and objective criteria rather than positions.
Red-Team Assessment
Models a capable adversary probing a plan for the seams they would exploit.
Bayesian Reasoning
Starting from base rates and updating beliefs proportionally as evidence arrives.
Supply & Demand
Price and quantity settle where what buyers want meets what sellers will offer.