How Much Does a Government Shutdown Cost Per Day? Oct 21 Update

Last Updated on November 14, 2025 | Prices Last Reviewed for Freshness: January 2026
Written by Alec Pow – Economic & Pricing Investigator | Content Reviewed by CFA Alexander Popinker

Educational content; not financial advice. Prices are estimates; confirm current rates, fees, taxes, and terms with providers or official sources.

What changed once the shutdown went live (Day 21 snapshot)

For agency-by-agency status and source links, see the What’s open vs. closed table.

• • SEC (CorpFin) update: Oct 9 guidance confirms limited operations; new registrations effectively paused except emergencies.
  • Hill dynamics: Senate failed for the 8th time to advance a funding bill; stalemate extends into week 3.
  • Macro: Treasury says the shutdown is now visibly hitting the economy; “data darkness” complicates reads.

• • Workforce: Administration reports 4,108 federal layoffs; unions challenge in court. (Historically unusual vs. prior furlough-only lapses.)

On peak-stress days, stacked effects can push shutdown costs well above $1B, into the $1.2–$1.3B range.

Related day-rate calibration: Our breakdown of National Guard deployment costs per day shows how personnel, logistics, and ops line items stack on a true daily meter.

Why most statistics get the numbers wrong

Jump to sections

Most “shutdown cost” stats use a narrow lens or smooth out the worst days. Here’s why that misses real money:

• Budget isn’t value. Back pay restores salaries, not the services or same-day spending that didn’t happen.
• Averages hide peaks. A 35-day average can bury the few worst days when airports snarl or refunds stack up.
• Spillovers get missed. Reliability penalties, missed trips that aren’t rescheduled, and deferred purchases don’t show up on federal ledgers.
• Rates move fast. Rates can move within hours, and on a $37T base even a tiny change is real money.

The headline range is conservative; bad days are worse than the averages suggest.

We treat the headline range as conservative for five reasons that standard summaries often underweight:

1. Reliability penalties are real money. Airlines and contractors pay liquidated damages (contractual late fees) and schedule penalties that do not show up in simple value-of-time math. Even a 1 percent reliability surcharge on $1.0B of affected daily activity adds ~$10M/day. See USDOT BCA 2024 for time-value baselines; the surcharge is a reasonable add-on for reliability risk.
2. Refund behavior, not just float. Households who expect refunds often delay durable purchases. If $3B in refunds shift a day and 5–10 percent of recipients defer a $300 purchase, short-run consumption drag can be $45–$90M/day even though interest float is only ~$0.33–$0.41M (see IRS 3/1/2024, 3/22/2024).
3. Seasonality can swing costs by tens of millions per day. A lapse during peak travel or tax season magnifies items D and E in the waterfall. TSA passenger counts exceed 3.0M/day on peaks, which doubles delay costs relative to a 1.5M trough day (BTS/TSA).
4. Contractor payments are often non-recoverable. Back pay covers federal employees, but contractors typically cannot bill for idle hours, so item B skews permanent. GAO case write-ups show stop-work orders that strand labor and material (GAO-20-201).
5. Spread channel scales with debt stock. With public debt above $37T, even a +1 bp (0.01 percentage point) stress premium is about $10.2M/day, and +5 bp around $51.2M/day until confidence normalizes (Treasury Debt to the Penny).

Tiny spread moves on a $37T base compound into newsroom-worthy daily dollars.

What our adjusted daily range looks like

Anchors vs. our working range

Anchor	What it represents	Add-ons & resulting “working” range
~$314M/day (CBO) CBO 2019 analysis	Average day across a partial, 35-day lapse.	Reliability penalties ≈ $5–$15M + behavioral deferrals ≈ $20–$60M (off-peak) + small spread bumps ≈ $10–$20M Working range: ~$350–$500M/day
~$0.9B/day (S&P full-scope) S&P yardstick (via Axios)	Full-breadth shutdown rule-of-thumb.	Reliability penalties ≈ $10–$20M + peak-season behavior ≈ $45–$90M + spreads ≈ $10–$51M Working range: ~$950M–$1.05B/day on peak stress days

Notes: Add-ons are conservative and illustrative; they reflect penalties and behavior shifts that typical budget or GDP lenses underweight.

Why this talk now

The U.S. government shut down at 12:01 a.m. ET on Oct 1, 2025 and remained partially closed until Nov 12, 2025, when a funding deal ended the lapse after 43 days. During the shutdown, agencies operated under contingency plans; effects varied by office, with courts funded through fees until about Oct 17 and many regulatory reviews paused. The daily price tag mattered in real time for households and businesses, and the math below remains a useful scale check for future lapses as well.

The daily price tag matters because a shutdown does not simply shift federal paychecks forward. It dents travel spending around parks and museums, jams airport throughput, delays loans and refunds, raises restart friction, and, when brinkmanship drags, can widen credit spreads that feed into borrowing costs for everyone. Journalists and everyday readers want a clean, sourced number they can quote without oversimplifying the moving parts.

What counts as cost

There are three practical lenses. The budget lens captures direct outlays such as back pay for furloughed employees and overtime to clear backlogs. The GDP lens captures lost or delayed private activity such as tourism, small business credit, and airline operations. The people lens separates who receives back pay later and who does not. In 2013, the Office of Management and Budget reported about $2.5 billion in pay and benefits to furloughed workers for hours not worked, a budget cost that did not restore the services those hours would have produced.

For the anatomy of large civic-process bills (petition gathering, legal, compliance, media), compare our breakdown of the cost to mount a statewide redistricting ballot measure.

Not all losses are equal. Delayed tax refunds and some federal purchases tend to be recouped, at least on paper, while a missed trip to a national park, a canceled museum day, a dropped business meeting, or a contractor’s idle shift rarely comes back in full. During the 2013 lapse, national groups cited a U.S. Travel estimate near $152 million per day in travel-related losses, a sector anchor for non-recoverable activity (ABC News citing U.S. Travel).

“Back pay restores paychecks, not the services that would have happened.”

How Much Does a Government Shutdown Cost Per Day?

The Congressional Budget Office estimated the 35-day partial lapse that ended in January 2019 shaved $11 billion off output across two quarters, with $3 billion permanently lost. That back-of-the-envelope works out to roughly $314 million per day over the period, excluding hard-to-quantify spillovers such as reputation damage or attrition (CBO).

Earlier episodes point the same way. The 2013 shutdown imposed measurable payroll and operations costs on the federal side, and independent tallies of private sector hits ranged into the billions when tourism leakage and airport delays were added. S&P Global’s rule of thumb ahead of 2018 put a full shutdown near $6.5 billion per week (about $0.9 billion per day) if scope widened. Duration and breadth make harm non-linear.

“The longer it lasts, the worse each day gets.”

The daily cost waterfall

These expandable cards show formulas, example inputs and sourced anchors so readers can see the math and deduce plausible daily totals.

Static scenario table

Use this grid to describe range, drivers and irrecoverable shares in plain language.

This table translates component math into three recognizable news scenarios. The irrecoverable share rises with duration because tourism, contractor hours and attrition effects compound.

These brackets square with a realized $314 million per day average for 2019 and an ≈$0.9 billion per day full-scope yardstick if breadth and duration expand.

New quick-read tables

Grab defensible numbers fast, with sources linked in captions.

Per-household equivalents translate big GDP figures into everyday scale without implying a tax bill.

Spread math shows how brinkmanship, even without a lapse, can raise borrowing costs at scale.

Worked example for a single day

Consider a limited-scope day affecting a slice of the federal workforce and several high-visibility services.

Add refund float impacts and small-business credit timing, and the daily total plausibly approaches $300 million. This arithmetic is illustrative, not a forecast, but it mirrors ranges in CBO and S&P benchmarks.

Hidden add ons

Back pay does not include restart overtime or weekend work, which can run tens of millions across large agencies after a long pause. Procurement re-competitions and compliance checks add smaller but real administrative bills. States also pay to keep marquee parks open to shield local economies, as Arizona and Utah pledged in 2023, after earlier outlays during 2013.

For a concrete analogue of non-payroll rework (signage, seals, forms, IT, contracts), see our estimate of rebranding the Pentagon to the “Department of War”.

Real cases

In Jan 2019, a LaGuardia ground stop tied to staffing strain rippled across the system, illustrating how even partial lapses can spike delay costs for a day (Reuters).

Museums took visible hits. The Smithsonian reported closures beginning Jan 2, 2019, and national groups summarized larger nationwide losses for parks and cultural institutions (AAM).

Methodology

1. Top-down lens uses CBO’s realized GDP loss from the 2018–2019 lapse and S&P’s pre-lapse weekly yardstick to bound a per-day range for broad shutdowns.
2. Bottom-up lens adds sector components: federal payroll exposure, contractor idle time, aviation value-of-time, parks and museum spillovers, small-business credit timing, refund float, research pauses, spread effects and restart friction, each with a sourced anchor.
3. Assumptions stay conservative where uncertain. Value-of-time uses USDOT BCA 2024. Debt spread math uses Treasury Debt to the Penny and basis-point deltas.
4. Recoverability is labeled qualitatively. Budget back pay is paid later, but lost tourism, contractor hours and missed business meetings are often not recovered.
5. Rounding keeps component math readable. Per-household figures divide daily ranges by the latest total household count from FRED.

References

• Congressional Budget Office, 2019 partial shutdown effects
• Office of Management and Budget, 2013 shutdown report
• GAO-20-201, Selected Agencies, 2020
• GAO-20-377, 2020
• USDOT Benefit-Cost Analysis Guidance, 2024 update
• BTS, average daily number of people screened by TSA
• IRS filing-season stats, week ending Mar 1, 2024 and week ending Mar 22, 2024
• BEA/Census FT900, July 2025 trade
• Census FT900, methodology PDF
• U.S. Treasury, Debt to the Penny
• American Alliance of Museums, economy and jobs
• Axios summary of S&P estimate
• Reuters, LaGuardia ground stop, Jan 2019
• ABC News citing U.S. Travel, 2013
• FRED, Total Households

Answers to Common Questions

How was the daily range derived?

By combining a top-down view from CBO’s 2019 analysis with bottom-up sector anchors such as USDOT value-of-time math, park and museum revenues and S&P’s full-scope yardstick, then adjusting for recoverability.

Which costs are usually repaid later?

Federal employee pay and many delayed purchases show up later on budget ledgers. Tourism spending, contractor hours and missed business travel are rarely recouped.

Do markets always suffer during shutdowns?

Short lapses often show muted moves, but prolonged brinkmanship can widen spreads. With public debt above $37T, even small spread changes scale quickly.

Why include parks and museums in a daily number?

They convert foot traffic into local sales every day; closures or partial openings erase spending that rarely returns. Several states have paid to keep gates open for this reason.

Is the per household number fair?

It is a communication device, not a tax bill, created by dividing the national daily impact by household count so the scale is tangible.