OpenFeds Analysis

Federal Remote Work: The $30 Billion Office Space Question

The federal government owns or leases 376 million square feet of office space. Post-pandemic, average utilization is just 25%. That's tens of billions in wasted real estate — or a case for permanent telework. Or both.

Sources: OPM, GSA, GAO·Last updated: March 2026

📊The Telework Numbers

Before the pandemic, about 25% of eligible federal employees teleworked at least occasionally. At the pandemic's peak, that surged to near 100% of eligible positions. As of late 2024 (pre-DOGE mandate), the landscape looked like this:

Telework Eligible

~1.1M

53% of civilian workforce

Regular Telework

~850K

At least 1 day/week

Full Remote

~228K

100% telework approved

Avg Days Remote

2.4/week

For telework participants

Nearly half the federal workforce — primarily blue-collar, law enforcement, healthcare, and field positions — was never eligible for telework. A VA nurse can't treat patients from home. A Border Patrol agent can't patrol from a couch. The telework debate is really about the ~1 million white-collar federal workers, mostly in DC and major metro areas.

Who Teleworks Most (by Agency)

  • Patent & Trademark Office: ~95% remote. Flagship telework program since 2012.
  • SEC: ~80% hybrid/remote. Knowledge work that translates well to remote.
  • GSA: ~75% remote. Shed most of their own office space — practicing what they preach.
  • OPM: ~70% hybrid. Ironic, given they now enforce the RTO mandate.
  • Treasury/IRS: ~55% hybrid. Large call center workforce works from home.

The Patent & Trademark Office has had a full telework program since 2012. Their output — patent examinations per examiner — is 15% higher than pre-telework. If one agency proved the model works, why isn't it the standard?

USPTO Annual Report, 2023

🏢The Real Estate Problem

The federal government is the largest office tenant in the United States. And most of that space sits empty.

Federal Office Space

376M sq ft

Owned + leased

Annual Cost

~$9.6B

Rent, operations, maintenance

Avg Utilization

25%

Post-pandemic, pre-RTO mandate

GAO has reported federal buildings as "high risk" since 2003 — not because they're falling down (though many are), but because the portfolio is poorly managed. The government owns buildings it doesn't need, leases space at above-market rates, and struggles to consolidate.

Metro AreaFed Office SpaceUtilizationAnnual Cost
Washington, DC Metro105M sq ft22%$3.2B
New York18M sq ft28%$890M
Atlanta12M sq ft31%$340M
Philadelphia11M sq ft26%$310M
Dallas-Fort Worth9M sq ft35%$220M
Denver8M sq ft29%$240M
San Francisco7M sq ft19%$380M
Chicago7M sq ft24%$210M

The federal government spends $9.6 billion per year on office space used at 25% capacity. That's $7.2 billion in wasted space. You could fund NASA's entire Artemis program with what we spend heating empty federal buildings.

The arithmetic is straightforward: if telework is permanent for 50%+ of the eligible workforce, the government could shed 40-50% of its office portfolio and save $3-4 billion annually. If everyone returns to the office, utilization improves but you still have a maintenance backlog exceeding $30 billion. There's no version of reality where the current real estate portfolio makes sense.

📈The Productivity Evidence

Does federal telework actually reduce productivity? The evidence is mixed but mostly favorable:

USPTO (2012-2023)

✅ Positive

Patent examinations per examiner up 15%. Attrition down 30%. Most-cited success case.

GSA Telework Study (2022)

✅ Positive

No measurable decline in output metrics. Employee satisfaction up 25%. Office costs down 40%.

OPM Federal Viewpoint Survey (2024)

⚠️ Self-reported

84% of teleworking employees reported being 'as productive or more productive' at home.

GAO Report GAO-23-105609

❌ Mixed

IRS phone answer rates declined during max telework. Processing times increased for paper returns.

SSA Inspector General (2023)

❌ Negative

Processing times for disability claims 18% longer with remote staff vs in-office. But sample size concerns.

Stanford WFH Research (Bloom, 2024)

✅ Positive

Hybrid work (2-3 days remote) shows no productivity loss and significant attrition reduction across sectors.

The honest summary: hybrid work (2-3 days remote) appears to maintain productivity while reducing costs and attrition. Full remote is more variable — it works well for focused individual work (patent examination, tax processing) but less well for collaborative or supervisory roles. The blanket RTO mandate ignores this nuance.

🔙The Return-to-Office Mandate

In January 2025, the administration ordered all federal employees back to the office full-time. The stated reason: accountability and productivity. The unstated reason: inducing voluntary attrition to reduce headcount without formal layoffs.

Employees Affected

~850K

Regular teleworkers pre-mandate

Voluntary Departures

~35K

Attributed to RTO mandate

Compliance Rate

~70%

As of Feb 2026

The Quiet Part Out Loud

Administration officials have privately acknowledged that the RTO mandate is partly a workforce reduction tool. When remote workers are told to commute 2+ hours daily to an office where their team doesn't sit, many choose to quit. This achieves headcount reduction without the legal requirements of a formal RIF. It's legal. It's also dishonest.

If the goal is productivity, study the data and implement evidence-based policies. If the goal is headcount reduction, be honest about it. Using RTO as a stealth layoff tool wastes everyone's time — and the $7 billion in empty office space proves nobody actually needed those desks.

💰The Locality Pay Paradox

Federal locality pay was designed to match local cost of living. A GS-13 in San Francisco gets ~44% locality on top of base pay; one in Birmingham gets ~18%. The question nobody wants to answer: should full-time remote workers in low-cost areas keep their high-cost locality pay?

ScenarioGS-13 Step 5 PayLocality
DC office worker$131,89033.26% DC locality
DC-locality, remote from rural Virginia$131,890Same — gets DC rate
If adjusted to 'Rest of US'$115,20017.46% base locality
Annual savings per employee$16,690If locality matched actual location

If even 100,000 remote workers are receiving DC locality pay while living in lower-cost areas, that's ~$1.7 billion per year in excess compensation. This isn't about punishing remote workers — it's about paying people based on where they actually live, which is the entire point of locality pay.

🎯The Smart Path Forward

The data supports a clear set of conclusions:

Evidence-Based Federal Telework Policy:

  • 1.Default to hybrid (2-3 days in-office) — the evidence supports this for most knowledge workers.
  • 2.Allow full remote for proven roles — patent examiners, tax processors, call center staff.
  • 3.Adjust locality pay to actual location — if you live in Kansas, you get Kansas rates.
  • 4.Sell/shed 40% of office space — use the savings for IT infrastructure and facility improvements.
  • 5.Measure output, not attendance — badge swipes don't measure productivity.

The return-to-office mandate is policy by ideology, not evidence. But the pre-mandate status quo — where workers collected DC locality pay from their beach house in North Carolina — wasn't right either. The answer is in the middle, guided by data. As usual, neither extreme is willing to go there.