IRS Audit Rates 2026: Your Real Odds by Income and Return Type

The IRS audit rate for individual tax returns is 0.3%, about 1 in 300 returns, based on tax year 2021, the most recent year with near-complete data in IRS Data Book FY2025 Table 3-1. In fiscal year 2025 the IRS closed 497,621 audits across all return types and recommended $26.8 billion in additional tax. Most audits never involve a person at your door: 81% were conducted entirely by mail.

Key takeaways:

• Individual audit rate: 0.3% for TY 2021 (about 1 in 300 returns) per IRS Data Book FY2025 Table 3-1
• Filers with income between $25,000 and $500,000 were audited at just 0.2%, or 1 in 500
• Incomes of $10 million+ were audited at 6.6%, or 1 in 15
• 81% of all FY2025 audits (and 89% of individual audits) were correspondence audits: letters, not visits
• S corporation and partnership returns were audited at roughly 0.1%, about 1 in 1,000

Save this cheat sheet — your audit odds in one image.

The Overall IRS Audit Rate Is 0.3% — About 1 in 300 Returns

The IRS examined 510,375 of the 161.2 million individual income tax returns filed for tax year 2021, a coverage rate of 0.3%, according to IRS Data Book FY2025 Table 3-1. TY 2021 is the right year to quote: it is the most recent tax year outside the normal three-year audit statute of limitations, so its audit count is close to final.

Two different measurements get mixed up in most audit-rate articles, so define both:

• Coverage rate (by tax year): of all returns filed for a given tax year, the share ever examined. This is Table 3-1 and the source of the 0.3% figure.
• Fiscal-year activity: audits the IRS closed in one fiscal year across all tax years. In FY2025 that was 497,621 audits, of which 441,578 were individual returns (Table 3-2).

Of the individual examinations closed for TY 2021, about 13% ended in no change: the taxpayer owed nothing extra.

IRS Audit Rates by Income Level

Audit odds follow a U-shape: highest with no reported income or very high income, lowest in the middle. Rates by total positive income (TPI) for TY 2021, from IRS Data Book FY2025 Table 3-1:

The flat middle is remarkable: whether you earned $30,000 or $450,000, your TY 2021 audit odds were the same 1-in-500. The curve only bends upward past $500,000.

Look Up Your Own Bracket

Your Real Odds: Three Worked Framings

• A Schedule C freelancer with $85,000 of income fell in the $75,000–$100,000 band: 0.2%, or 1-in-500 odds. Statistically, she would file for 500 years before expecting one audit.
• A W-2 tech employee earning $150,000 sat in the $100,000–$200,000 band: the same 0.2%, 1 in 500.
• A business owner with $2.4 million of income fell in the $1 million–$5 million band: 0.9%, about 1 in 111. That is five times the middle-class rate, but still long odds in any given year.

One caveat the current Data Book format imposes: the IRS no longer publishes a separate audit rate for Schedule C filers. Bands are by total income, not by return schedule, so self-employed and W-2 filers share the same published rates. Older IRS data showed sole proprietors audited above the band average, which is why clean Schedule C reporting still matters more for freelancers than for W-2 employees at the same income.

Audit Rates by Return Type: Individuals vs Corporations vs Partnerships

Business entities are audited less often than individuals. Coverage rates for TY 2021 from IRS Data Book FY2025 Table 3-1:

Large corporations are the exception hiding inside that C-corp average: companies with $20 billion or more in assets were audited at 29.1% for TY 2022. For pass-through owners, the practical takeaway is different: the entity return is rarely examined, but S corp compensation flows to your 1040, and setting a defensible S corp salary is what keeps that low 0.1% from becoming your problem.

Most Audits Are Letters: 81% Were Correspondence Audits in FY2025

The IRS closed 403,059 of its 497,621 FY2025 audits by mail, an 81% share, per IRS Data Book FY2025 Table 3-2. For individual returns the mail share was 89%. A correspondence audit is a letter asking you to document one or two specific items: an EITC claim, a filing status, a deduction. You respond with records; no agent visits.

Field audits are rarer and far more expensive per case. FY2025 field exams produced $19.1 billion of the $26.8 billion in recommended additional tax despite being only 19% of closures. Across the 441,578 individual audits closed in FY2025, the IRS recommended $11.3 billion in additional tax, an average of about $25,600 per audited return, a figure pulled up sharply by high-income field cases.

If a letter does arrive, you can authorize your accountant to handle the IRS for you; our guide to Form 2848 vs Form 8821 explains which authorization does what. And before paying any notice, check the math: our IRS penalty and interest calculator shows what the IRS can actually add to a balance.

The EITC Anomaly: Low-Income Filers Face Elevated Audit Rates

Returns claiming the Earned Income Tax Credit were audited at 0.7% for TY 2021, more than triple the 0.2% rate for filers earning $50,000 to $200,000, per IRS Data Book FY2025 Table 3-1. In FY2025, EITC returns accounted for 93,857 of the 441,578 individual audits closed: 21% of all individual audits, aimed at a credit whose claimants mostly earn under $60,000. These were almost entirely paper checks, not investigations: 98.9% of FY2025 EITC audits were correspondence audits.

The driver is improper-payment enforcement mandates on refundable credits, not income-based suspicion. But the practical effect is a U-shaped enforcement curve in which a warehouse worker claiming the EITC faced better-than-triple the audit odds of a consultant earning $180,000.

Audit Rates Have Fallen by Half in a Decade

The individual audit coverage rate dropped from 0.6% for TY 2015 to 0.3% for TY 2021, per IRS Data Book FY2025 Table 3-1. The decline concentrated at the top:

A decade ago, a millionaire filer faced 1-in-38 odds; by TY 2021 that was 1-in-111. Note that TY 2022 and TY 2023 rates published today (0.3% and 0.1% for individuals) are still open years: the three-year statute has not run, audits are still being opened, and those figures will rise before they are final.

What Actually Triggers an IRS Audit

The IRS names its selection methods on its official audits page: statistical scoring against norms ("random selection and computer screening — sometimes returns are selected based solely on a statistical formula") and related examinations ("returns … involv[ing] issues or transactions with other taxpayers … whose returns were selected for audit"). The statistical formula is the DIF score: your return's deductions and credits compared against norms for similar returns.

Document mismatches are the bigger enforcement channel, and they are not audits at all. The Automated Underreporter Program matched W-2s and 1099s against returns and closed 987,460 cases in FY2025, twice the number of actual audits, recommending $5.9 billion (IRS Data Book FY2025 Table 3-8). Another 592,773 substitute-for-return cases hit people who didn't file. A missing 1099 is far more likely to generate a CP2000 matching notice than an audit. The folklore triggers worth taking seriously are the ones that feed DIF and matching: deductions far outside your income band's norms, unreported information returns, and round-number expense claims. Our IRS audit triggers guide ranks them.

What Does NOT Raise Your Audit Risk

• Filing an extension. The IRS's published selection methods make no mention of extensions; an October filing scores through the same DIF formula as an April one.
• Filing an amended return. The IRS states directly: "filing an amended return does not affect the selection process of the original return." The amended return is screened on its own merits.
• Receiving a large refund. Refund size is not a listed selection criterion; refundable-credit claims (like the EITC) are what draw document checks, not the refund amount itself.
• E-filing. No IRS selection method distinguishes e-filed from paper returns; e-filing actually prevents the math errors that generate notices.
• Old returns resurfacing. "Generally, the IRS can include returns filed within the last three years in an audit," extending to six only for substantial errors, per the IRS audits page.

IRS Audit Rates 2026: Five Facts to Cite

1. According to IRS Data Book FY2025 Table 3-1, the IRS examined 0.3% of individual income tax returns for tax year 2021, about 1 in 300.
2. According to IRS Data Book FY2025 Table 3-1, taxpayers reporting $10 million or more were audited at 6.6% for TY 2021, versus 0.2% for incomes between $25,000 and $500,000.
3. According to IRS Data Book FY2025 Table 3-2, 81% of the 497,621 audits the IRS closed in FY2025 were correspondence audits conducted by mail.
4. According to IRS Data Book FY2025 Table 3-2, the IRS recommended $26.8 billion in additional tax from FY2025 audits, including $11.3 billion from individual returns.
5. According to IRS Data Book FY2025 Table 3-8, the Automated Underreporter Program closed 987,460 document-matching cases in FY2025, nearly double the number of audits.

Common Mistakes When Reading Audit Statistics

• Quoting open tax years as final. TY 2023's published 0.1% individual rate reflects audits opened so far; the statute runs until 2027. Only years outside the statute (TY 2021 and earlier) are near-final.
• Calling a CP2000 notice an audit. The IRS's own Data Book counts underreporter notices separately from examinations. Nearly a million FY2025 CP2000-type cases never appear in the 497,621 audit total.
• Dividing FY closures by one year's filings. The 441,578 individual audits closed in FY2025 span many tax years; dividing them by one year's 161 million returns produces a rate that measures nothing.
• Using pre-2018 audit rates. The frequently cited "1% audit rate" dates to the early 2010s. Current coverage is a third to half of that at almost every income level.

Clean Records If the IRS Writes: How Jupid Helps

Correspondence audits are document checks: the IRS asks you to substantiate a number, and the filer with organized records closes the case fastest. Jupid keeps those records ready before anyone asks: an AI accountant inside WhatsApp and iMessage, with your bank connected and each transaction auto-labeled at 95.9% accuracy, so every Schedule C line traces back to real, named transactions. Need last March's software expenses or your full-year advertising total? Ask in chat and the answer arrives with the transactions behind it. Try Jupid.

Sources

• IRS Data Book FY2025 (Publication 55-B) — Compliance Presence, Tables 3-1, 3-2, 3-8
• IRS Data Book FY2025 Table 3-1 XLSX (examination coverage by tax year 2015–2023)
• IRS Data Book FY2025 Table 3-2 XLSX (FY2025 closures, field vs correspondence)
• IRS: IRS Audits (selection methods, amended returns, lookback period)
• IRS Data Book home

---

This guide is for general educational purposes and does not constitute tax, legal, or accounting advice. Audit statistics describe population-level rates, not any individual return's risk. If you receive an IRS notice or examination letter, consult a qualified tax professional promptly.