In the most recent governmentwide Federal Employee Viewpoint Survey, federal employees rated their immediate supervisors at 81% positive. They rated their senior leaders at 63%. Eighteen points separate the person who approves their leave from the people who set their agency’s direction (OPM, 2024 FEVS Governmentwide Management Report, p. 6).
That gap is one of the most useful numbers available to a federal HR director. It is also one of the most consistently misread — and, right now, one of the most frequently misdated.
Start With the Vintage
These are 2024 figures. They come from the 2024 FEVS, administered to more than 1.6 million federal employees, with over 674,000 responses and a 41% response rate, and published in the Governmentwide Management Report (revised edition, April 2025). They are not current-year numbers, and no newer governmentwide equivalent exists.
OPM cancelled the 2025 FEVS outright — the first year the survey was not fielded in more than two decades (Federal News Network, August 2025; Government Executive, August 2025). Beginning with the 2026 cycle, OPM decentralized the instrument: agencies now design and administer their own annual surveys, and the number of required common questions dropped from 16 to 10 under a proposed rule amending 5 CFR Part 250, published in the Federal Register on July 2, 2026.
So 2024 remains the last apples-to-apples governmentwide read on federal workforce sentiment. Any 2026 brief presenting FEVS figures as current is presenting data two survey cycles old. State the year — being explicit about vintage protects a training investment case from being unwound by a single question in a budget review.
The Gap Is a Structure, Not a Bad Year
One year of data is an anecdote. Five years is a feature. Here are the governmentwide Employee Engagement Index subindices across the last five FEVS administrations (OPM, 2024 Governmentwide Management Report, p. 6):
| Subindex | 2020 | 2021 | 2022 | 2023 | 2024 |
|---|---|---|---|---|---|
| Supervisors | 80 | 80 | 80 | 80 | 81 |
| Leaders Lead | 62 | 60 | 59 | 61 | 63 |
| Gap (points) | 18 | 20 | 21 | 19 | 18 |
The gap never falls below 18 points and never exceeds 21 — across a pandemic, a change in administration, the shift to widespread telework, and a return-to-office push. (The gap row is our arithmetic on OPM’s published subindex scores.)
Note the shape of the movement. The Supervisors subindex barely moves: 80, 80, 80, 80, 81. Essentially all of the variation comes from Leaders Lead, which travels from 59 to 63. The gap widens and narrows because of what happens at the top, not because of anything happening in the supervisor relationship. That relationship is the most stable measurement in the federal workforce data set.
What the Gap Actually Measures
The two subindices are not measuring more and less of the same thing. OPM defines the Supervisors subindex as the interpersonal relationship between worker and supervisor — trust, respect, and support (items 48, 50, 51, 52, 54). Leaders Lead captures perceptions of the integrity of leadership and leadership behaviors such as communication and workforce motivation (items 57, 58, 59, 61, 62). One measures a relationship the employee is inside of. The other measures a relationship the employee observes from a distance.
The item-level 2024 results make the distinction concrete. Among the highest-scoring items governmentwide:
-
“My supervisor holds me accountable for achieving results” — 88%
-
“My supervisor treats me with respect” — 87%
-
“My supervisor supports my need to balance work and other life issues” — 85%
-
“My supervisor listens to what I have to say” — 83%
Among the lowest-scoring items governmentwide:
-
“Management involves employees in decisions that affect their work” — 48%
-
“In my organization, senior leaders generate high levels of motivation and commitment in the workforce” — 52%
-
“Management makes effective changes to address challenges facing our organization” — 55%
-
Satisfaction with “the information you receive from management on what’s going on in your organization” — 56%
The diagnosis writes itself. The high scores are about being seen, heard, and held to a standard by someone who knows you. The low scores are about involvement and information flow. Employees are not reporting that senior leaders lack competence or integrity. They are reporting that senior leaders are remote — decisions arrive without explanation, and employee input disappears without acknowledgment.
The Budget Implication Most Agencies Get Backwards
The reflexive response to a 63% leadership score is to buy executive leadership development. That inference is understandable and, in most agencies, wrong. Three reasons.
First, population math. Senior executives and their equivalents are a fraction of a percent of the federal workforce. The Leaders Lead subindex does not measure how well those executives lead; it measures a perception held by the roughly 99% of employees who rarely interact with them. Training the people being perceived moves the index only to the extent that the training changes what reaches everyone else.
Second, the deficit sits in transmission, not capability. The two weakest items — involvement in decisions at 48% and information from management at 56% — describe communication infrastructure. A leadership curriculum rarely repairs a broken cascade.
Third, the strongest available carrier is already trusted and already in the room. Supervisors sit at 81% and have held there for five straight years. Gallup’s workplace research finds that managers account for roughly 70% of the variance in team engagement — the single largest identified driver. In a federal context, the supervisor is the only leadership voice most employees encounter regularly, and it is a voice they already believe.
The practical conclusion, and this is GGS’s read of the data rather than a finding published by OPM: the highest-yield development dollar is the one that equips the supervisor tier to translate. Not general supervisory soft skills — those already test at 81%. Specifically: explaining senior decisions in the language of the work unit, carrying team input upward in a form that survives the trip, and closing the loop when a decision goes the other way. That is a narrow, teachable, measurable capability, and it targets the exact items scoring in the 40s and 50s.
Where the Gap Runs Widest
The 2024 data breaks out by agency size, and the pattern is clean (OPM, 2024 Governmentwide Management Report, p. 6):
| Agency size | Supervisors | Leaders Lead | Gap |
|---|---|---|---|
| Very small (<100) | 82 | 67 | 15 |
| Small (100–999) | 86 | 66 | 20 |
| Medium (1,000–9,999) | 84 | 67 | 17 |
| Large (10,000–74,999) | 83 | 63 | 20 |
| Very large (≥75,000) | 81 | 62 | 19 |
Supervisors hold between 81% and 86% at every size tier. Leaders Lead sits at 66–67% across the three smaller tiers, then drops to 63% and 62% at the two largest. The variable is organizational distance: what senior leadership communicates degrades as it crosses layers, and large agencies have more layers. For a large or very large agency, the leadership perception problem is substantially a scale problem — which again points the remedy toward the cascade rather than the executive suite.
On what the gap predicts, precision matters. OPM’s governmentwide report publishes the index scores; it does not publish an outcome model linking the supervisor-to-leader gap to attrition, time-to-hire, or mission performance.
The report publishes the subindex scores and their five-year trend and stops there. It contains no model relating the size of the supervisor-to-leader gap to turnover, time-to-hire, or mission results, and no other OPM publication supplies one. Any claim that a wider gap predicts a worse outcome is being imported from somewhere other than the federal data, and is worth treating with suspicion when it appears.
What is documented is the mechanism. Gallup’s panel analysis of the 2025 federal workforce reforms found that as federal employees became less likely to say they trust leadership, that their organization cares about their wellbeing, and that they feel connected to their culture, engagement and job satisfaction fell and burnout rose relative to comparable state and local employees. Once those perceptions were accounted for statistically, the federal-versus-state-and-local effect shrank considerably — the perceptions were the pathway, not a side effect. Most measures rebounded by Q4 2025 and held through Q1 2026.
Three Ways Agencies Misread This
1. Treating it as a verdict on leadership quality. It is a measure of proximity and visibility. An agency can have genuinely capable executives and a 62% Leaders Lead score at the same time, because the score is largely reporting on what employees can see from where they sit.
2. Treating 81% as solved. Five years at 80, 80, 80, 80, 81 is a plateau, not momentum — and a plateau in a period when the supervisor tier absorbed telework transitions, hiring freezes, and reorganizations. Stability under that load is a real asset, but one being drawn down rather than compounding. Gallup’s federal panel data through Q1 2026 shows the pressure has since increased.
3. Treating 2024 as current. This is now the expensive one. Under decentralization, each agency owns its own instrument, and only ten common questions are required governmentwide. Agencies that want a defensible five-year trend line in FY28 need to preserve the original FEVS item wording inside their own 2026 and 2027 surveys. Reworded items do not benchmark against the historical series, and a broken trend line cannot be repaired retroactively.
What to Do Before the FY27 Cycle
Pull your own agency’s 2024 Supervisors and Leaders Lead subindices from the OPM FEVS Dashboard and calculate your gap against the 18-point governmentwide figure. A wider gap indicates a cascade problem and argues for investment in the supervisor tier’s communication capability. A narrower gap with both scores low is a different diagnosis entirely and warrants a different budget.
Then check your 2026 agency instrument against the retired FEVS wording while there is still time to align it. The governmentwide baseline that made this analysis possible is not coming back in its previous form. The agencies that keep measuring the same things the same way will be the only ones able to prove what their development dollars bought.
Sources
-
U.S. Office of Personnel Management, 2024 Federal Employee Viewpoint Survey Results: Governmentwide Management Report (revised edition, April 2025). Subindex scores, five-year trend table, agency-size breakouts, and highest/lowest scoring items. https://www.opm.gov/fevs/reports/governmentwide-reports/governmentwide-reports/governmentwide-management-report/2024/2024-governmentwide-management-report.pdf
-
U.S. Office of Personnel Management, FEVS program page and FEVS Dashboard (agency-level results). https://www.opm.gov/fevs/
-
U.S. Office of Personnel Management, CHCO memorandum, “Annual Employee Survey Administration Requirements 2026 and Beyond.” https://www.opm.gov/chcoc/latest-memos/annual-employee-survey-administration-requirements-2026-and-beyond.pdf
-
Federal Register, “Agency Information Collection Request: Federal Employee Viewpoint Survey, OMB Control Number (3206-NEW),” July 2, 2026 (proposed changes to 5 CFR Part 250; required common questions reduced from 16 to 10). https://www.federalregister.gov/documents/2026/07/02/2026-13443/agency-information-collection-request-federal-employee-viewpoint-survey-omb-control-number-3206-new
-
Federal News Network, “After months of postponing, OPM opts to fully cancel 2025 FEVS,” August 2025. https://federalnewsnetwork.com/workforce/2025/08/after-months-of-postponing-opm-opts-to-fully-cancel-2025-fevs/
-
Government Executive, “OPM will forego FEVS in 2025, despite law requiring it,” August 2025. https://www.govexec.com/oversight/2025/08/opm-will-forego-fevs-2025-despite-law-requiring-it/407584/
-
Federal News Network, “OPM seeks to decentralize, revise governmentwide federal employee survey,” July 2026. https://federalnewsnetwork.com/workforce/2026/07/opm-seeks-to-decentralize-revise-governmentwide-federal-employee-survey/
-
Gallup, “What 2025 Federal Reforms Did to Worker Engagement,” Christos Makridis and Camille Lloyd, May 13, 2026. https://www.gallup.com/workplace/708866/2025-federal-reforms-worker-engagement.aspx
-
Gallup, “Managers Account for 70% of Variance in Employee Engagement.” https://news.gallup.com/businessjournal/182792/managers-account-variance-employee-engagement.aspx
-
Gallup, “U.S. Federal Employees Less Engaged Than the Rest.” https://news.gallup.com/poll/180206/federal-employeesless-engaged-rest.aspx