OpenFeds Analysis
$4.5 Billion for Nothing: The True Cost of the Deferred Resignation Program
The “Fork in the Road” email promised efficiency. Instead, the government paid 76,000 employees full salary and benefits to sit at home for eight months. Here's the full accounting.
📧The Deal
On February 4, 2025, the Office of Personnel Management sent what it called the “Fork in the Road” memo — the first-ever mass email to all ~2 million federal civilian employees. The offer was straightforward:
- ✅ Resign voluntarily with a deferred effective date of September 30, 2025
- ✅ Keep full pay and benefits through the end of the fiscal year
- ✅ Placed on administrative leave — no work required
- ✅ Continue TSP matching and annual/sick leave accruals
- ❌ No severance — resignation is voluntary, so no severance entitlement
- ❌ No rehire rights — unlike a RIF, no placement on reemployment lists
The initial deadline was February 6 — just two days. A federal judge temporarily blocked the deadline, and it was extended to February 12. In the end, approximately 76,000 employees — about 3.8% of the workforce — accepted.
📊The Numbers
The DRP cost roughly $59,000 per employee — which is what you'd pay for a year of a GS-7 salary. Except these employees produced nothing.
The $4.5 billion figure, estimated by the Partnership for Public Service, includes base salary, locality pay, and standard benefits (health insurance, TSP matching, leave accruals). It does not include the fully-loaded cost with overhead — which would push the number significantly higher.
👤Who Took It — And Who Didn't
The 3.8% acceptance rate was lower than the administration hoped. But the composition of who accepted matters more than the raw number:
Retirement-eligible employees
Many acceptees were already planning to retire within 1–2 years. The DRP gave them a paid runway to their planned exit. For these workers, the government essentially paid 8 months of salary it would have paid anyway — but got zero productivity in return.
Early/mid-career employees with options
Workers with in-demand skills (IT, data science, legal) used the DRP as a golden parachute to jump to private sector jobs — often at higher pay. The government lost talent it had invested years in developing.
Who stayed
Employees without good external options or nearing pension milestones generally declined. Ironically, the workers most likely to be underperforming had the least incentive to leave a guaranteed paycheck.
The DRP created a classic adverse selection problem: the people most likely to accept were the ones the government least wanted to lose.
⚖️The Alternative: What RIFs Would Have Cost
The administration's argument for the DRP was that it was cheaper and faster than formal Reductions in Force. Let's test that claim:
| Factor | DRP | RIF (76K workers) |
|---|---|---|
| Direct payment | $4.5B (8 months salary) | ~$1.2B (est. severance) |
| Time to execute | 2 weeks | 3–6 months |
| Legal risk | Lower (voluntary) | Higher (bumping rights, appeals) |
| Productivity during process | Zero for 8 months | Normal until separation date |
| Who leaves | Self-selected (adverse selection) | Seniority-based (retains top performers) |
The math is clear: the DRP cost roughly 3.75x more in direct payments than equivalent RIFs would have. The speed advantage was real — but at a $3.3 billion premium, it's fair to ask whether “fast” was worth it.
🏛️The Verdict
The DRP achieved its goal of quickly shrinking the federal workforce without the procedural burden of RIFs. But it did so at an extraordinary cost — $4.5 billion to pay people to not work — and created adverse selection that likely pushed out the wrong people.
Was there fat to cut in the federal workforce? Almost certainly. Was paying $4.5 billion in administrative leave the most efficient way to do it? The data says no.