National holidays cost the United States about $325 billion a year, according to a widely repeated estimate — here is the arithmetic behind it

A figure keeps circulating that America's public holidays cost the economy roughly $325 billion a year in lost output. No federal agency publishes such a number — it has to be built from a workforce count, a GDP total and a holiday calendar, then checked to see whether it holds up.

The claim, as reported in 2022, treats each of the country's federal holidays as a day on which a slice of normal economic output simply does not happen, then adds up that daily hit across the year. That is a reasonable starting method, but every input in it is a judgment call — how much output actually pauses, and how much of it is genuinely gone rather than shifted to another day.

Rebuilding the arithmetic

Start with the two inputs the estimate needs: total output and the size of the workforce producing it. US nominal GDP for 2022 came to $25.46 trillion, per the Bureau of Economic Analysis. Average employment that year was roughly 158 million people, per the Bureau of Labor Statistics' household survey. Dividing one by the other gives GDP per worker of about $161,000 a year.

Spread that over a standard working year — 365 days minus 104 weekend days minus the 11 federal holidays observed in 2022, or about 250 working days — and each worker accounts for roughly $645 of output on an average workday. Multiply that by the 158 million-strong workforce and the 11 holidays, and a complete, economy-wide shutdown on every one of them works out to about $1.12 trillion a year.

That number is the theoretical ceiling, and it is obviously too high: retail, healthcare, transport, hospitality, utilities and emergency services do not stop on a federal holiday. Estimates of this kind commonly apply a 50% haircut to account for that, which brings the total to roughly $560 billion. The original $325 billion figure is lower still — equivalent to treating only about 29% of the full-stop maximum as genuinely lost. That is a defensible assumption, not a measured one.

Why the real figure is probably smaller still

The 11-holiday count assumes every worker gets every federal holiday off with pay, which is not how US employment actually works. The Bureau of Labor Statistics' benefits data shows the average private-industry worker receives closer to eight paid holidays a year, not eleven, and roughly one in four private-sector workers gets none at all. A meaningful share of the workforce is simply on the job regardless of the federal calendar.

Retail, healthcare, transport and hospitality do not close for a federal holiday — they are often at their busiest.

There is also the question of whether paused output disappears or just moves. A report due on a holiday Monday tends to land Tuesday instead; a factory run gets rescheduled rather than cancelled. A 2022 economic analysis of holiday scheduling — cited in nationally syndicated reporting on the topic in 2025 — found that when a federal holiday falls on a weekday instead of a weekend, quarterly GDP growth shifts by only about 0.08 to 0.2 percentage points. That is a far smaller real-world effect than a simple day-by-day multiplication implies, and it is the strongest evidence that estimates like $325 billion should be read as an upper-bound sketch rather than a precise loss.

How the US calendar compares

As of 2022, the US observes 11 federal holidays, the newest being Juneteenth, added in 2021. For international context, Pew Research Center's 2026 review of public holidays across 190 UN member states found the typical country observes about 13. On that measure the US calendar is slightly leaner than the global norm, which cuts against any suggestion that America's holidays are unusually costly by international standards.

What the $325 billion estimate assumes, in short:

• Every one of the 11 federal holidays is paid time off, economy-wide — in practice, many workers get fewer, and some get none.
• A fixed dollar value of output would otherwise be created on that calendar day, calculated from annual GDP divided across a standard working year.
• Only a partial share of that output — not the full amount — is treated as genuinely forgone rather than simply deferred.
• None of the paused output is later recovered — though the scheduling research above suggests some clearly is.

Frequently asked questions