OpenFeds Analysis

DOGE Tracker: Separating Signal from Noise

The Department of Government Efficiency promised $2 trillion in savings. The actual number is far smaller — but the workforce impact is real. Here's what the data actually shows.

Data: OPM FedScope Dec 2025·Last updated: March 2026

🎯The Promise vs Reality

When DOGE was announced in late 2024, the rhetoric was bold: eliminate waste, cut bureaucracy, save trillions. Elon Musk initially claimed $2 trillion in potential savings. The final tally will be significantly less — but the workforce impact has been substantial.

Promised Savings

$2T

Initial DOGE target

Claimed Savings

~$160B

DOGE's own estimate (disputed)

Verified Savings

~$35-50B

Independent estimates (CBO, GAO)

Positions Affected

~217K

Eliminated, frozen, or restructured

DOGE promised $2 trillion in cuts. Even by their own generous math, they've found $160 billion. By independent estimates, real savings are closer to $35-50 billion. That's still significant — but it's 2% of what was promised.

The gap between promise and reality isn't surprising. Most federal spending is mandatory — Social Security, Medicare, Medicaid, interest on debt — and untouchable without legislation. Discretionary spending, where workforce costs sit, is only about 25% of the budget. You can't save $2 trillion from a $1.7 trillion discretionary budget.

📊By the Numbers

Based on OPM FedScope data comparing January 2025 to December 2025, here's what actually happened to the federal workforce:

Net Separations (2025)

335.2K

Jan-Nov 2025

vs 2024 Same Period

201.1K

Jan-Nov 2024

Excess Separations

+134.1K

Above normal attrition

⚠️ Important Caveat

Not all excess separations are DOGE-driven. Some reflect normal turnover variation, pandemic-era deferred retirements finally happening, and voluntary departures accelerated by uncertainty. We estimate 60-70% of the excess is directly attributable to DOGE-related actions (RIFs, hiring freezes, early retirement incentives, agency restructuring).

Breakdown by Separation Type (Estimated):

~45,000

Reduction in Force (RIF)

Formal layoffs with bumping rights

~55,000

Deferred Resignation / Early Retirement

Voluntary with incentives

~65,000

Hiring Freeze Attrition

Vacancies not backfilled

~25,000

Agency Restructuring

Reorganizations, consolidations

~15,000

Probationary Terminations

New hires in first year released

~12,000

Normal Excess Attrition

Above-baseline voluntary departures

🏛️Hardest-Hit Agencies

DOGE's impact wasn't evenly distributed. Some agencies were targeted for dramatic restructuring while others were barely touched. The pattern reveals priorities: non-defense discretionary agencies bore the brunt.

USAID

~85%

Near-total shutdown. From 4,000+ to ~600 core staff. The most dramatic single agency reduction.

Department of Education

~50%

From 4,200 to ~2,100. Reflects stated goal of 'returning education to the states.'

EPA

~35%

From 14,600 to ~9,500. Regulatory enforcement dramatically reduced.

IRS

~25%

From 90,000+ to ~68,000. Ironically, IRS employees generate $5-12 in revenue per $1 of cost.

HUD

~20%

From 6,300 to ~5,000. Housing program administration consolidated.

OPM

~30%

From 2,300 to ~1,600. The agency that manages other agencies' workforces cut its own.

✅ Agencies Largely Spared

VA (451K → ~445K, -1%), DoD civilian (minimal impact),DHS (actually grew for border enforcement), DOJ (law enforcement protected). The pattern is clear: agencies aligned with administration priorities were protected; regulatory and social program agencies were cut.

⚙️How They Did It

DOGE used every tool in the federal workforce reduction playbook — and invented some new ones:

Hiring Freeze

Government-wide freeze on new hires, with exceptions for national security and law enforcement. Natural attrition (retirements, resignations) reduced headcount without any firings. This alone accounts for ~65,000 positions.

Deferred Resignation Program

Offered federal workers 7-8 months of pay to resign immediately. About 55,000 accepted — many of them senior employees whose institutional knowledge is now lost. Cost: ~$5-7B in severance. Savings: ~$6-8B/year in ongoing salary.

Reduction in Force (RIF)

Formal layoffs governed by 5 CFR Part 351. Complex process involving competitive levels, bumping rights, and retention registers. Legal challenges have slowed many RIFs.

Agency Restructuring

Reorganizations that eliminated entire offices. USAID effectively shuttered, CFPB scaled back, and several small agencies merged or defunded.

Probationary Employee Terminations

New employees in their first year (probationary period) can be fired with minimal process. ~15,000 were terminated.

Return-to-Office Mandate

Full-time in-office requirement for all federal workers. While framed as productivity policy, the clear intent was to induce voluntary resignations from remote workers.

💰The Savings Question

The central claim — that DOGE would save taxpayers trillions — requires scrutiny. Here's an honest accounting:

CategoryDOGE ClaimIndependent Est.Notes
Workforce reduction$25B/yr$15-20B/yrSalary + benefits savings, net of severance
Contract terminations$50B$10-15BMany contracts reinstated after service disruptions
Fraud reduction$75B$3-8BMost 'fraud' claims were reclassified improper payments
Lease/real estate savings$8B$2-4BMost leases have multi-year obligations
IT/systems consolidation$5B$1-2BEarly stage, most savings future
TOTAL (Annual)~$160B~$35-50B

Even $35 billion in annual savings is real money — about $100 per American per year. The question isn't whether DOGE saved anything. It's whether the savings justify the disruption to services that real people depend on.

⚠️Unintended Consequences

Workforce reductions always have second-order effects. Some are becoming visible:

Brain Drain

The deferred resignation program disproportionately attracted senior, experienced employees — exactly the people hardest to replace. Average tenure of departed workers: 18 years. Average tenure of remaining workers: 8 years.

Contractor Backfill

Agencies are already replacing lost capacity with contractors at 2-3x the hourly cost. The net savings will be lower than projected once contractor spending is included.

Service Degradation

Social Security processing times up 40%. VA claim backlogs increasing. IRS phone wait times doubled. Passport processing times tripled in some offices.

Legal Costs

Over 60 lawsuits challenging DOGE-related actions. Legal defense costs, court-ordered reinstatements, and back pay awards will eat into savings.

Institutional Knowledge Loss

Many departed employees were the only people who understood critical systems. Several agencies report 'key person risk' — systems no remaining employee can maintain.

Recruitment Damage

Federal job applications down 45% year-over-year. Top university recruitment programs report declining interest. The government is becoming an employer of last resort for technical talent.

⚖️The Verdict So Far

DOGE gets both too much credit and too much blame. Here's our data-driven assessment:

✅ What DOGE Got Right

  • • Forced a conversation about government efficiency that was overdue
  • • Identified genuine redundancy in multi-layered oversight structures
  • • Exposed contract spending that had grown unchecked for decades
  • • Savings of $35-50B/year is meaningful — not trivial
  • • Demonstrated that the federal workforce is not untouchable

❌ What DOGE Got Wrong

  • • Promised $2T, delivered ~$35-50B (a 97% miss)
  • • Used a machete where a scalpel was needed
  • • Lost critical institutional knowledge through voluntary exits
  • • Damaged federal recruitment for a generation
  • • Many "cuts" will be backfilled by more expensive contractors
  • • Service degradation affecting millions of Americans

The federal government genuinely needs reform. Accountability is important. But sustainable reform requires precision — identifying which positions add value and which don't, then making targeted changes. DOGE's approach of broad cuts followed by "figure it out" is the governmental equivalent of performing surgery with a chainsaw. Some of the growths needed removing. But the patient is bleeding.

We'll continue tracking the data as it comes in. Bookmark this page — we update it monthly.

Explore DOGE Impact Data

See which agencies, occupations, and states are most affected.