Research

IRS Audit Report 2026: Who Actually Gets Audited, by Income, Entity, and Year

The Ledgerism Editorial Team
By The Ledgerism Editorial Team · The Ledgerism Brief Editorial Team produces original, methodology-disclosed accounting and tax research—data trackers, databases, and reference guides built to be the most-cited source in the field. Every figure is verified against primary sources including FASB, the IRS, the PCAOB, NASBA, the AICPA, and state boards of accountancy.
Published Jun 30, 2026
IRS Audit Report 2026 cover, The Ledgerism Brief

A data briefing on U.S. federal income tax examination (“audit”) coverage. All figures are drawn from the IRS Data Book, IRS Statistics of Income, the U.S. Government Accountability Office, the National Taxpayer Advocate, and the Transactional Records Access Clearinghouse (TRAC) at Syracuse University. This briefing builds the longest verifiable per-income-band audit-rate time series available from IRS primary data (Tax Years 2010 through 2023) and introduces an original Audit Risk Index scoring relative audit likelihood by income band and entity type.

Critical methodology flag: the audit-rate “look-back” lag

IRS audit rates for recent tax years are understated and will rise. The IRS measures audit coverage by comparing returns filed for a tax year against returns from that same year that are eventually examined. Because the IRS generally has three years to open an examination (longer in some cases), and because complex audits stay open for years, the most recent tax years show only the audits closed or in process so far. In the IRS Data Book, the two most recent tax years are shaded specifically to warn that their coverage rates are incomplete. (Source: IRS, 2025 Data Book, Table 3-1 notes.)

For this reason, the highest-confidence “settled” audit rates in this briefing are for Tax Year 2019 and earlier. Tax Years 2020 through 2023 are reported as published but should be read as floors, not final values. The IRS itself, on its Compliance Presence page, uses Tax Year 2021 as “the most recent year outside the statute of limitations period” when describing current coverage. (Source: IRS, Compliance Presence, accessed 2026-06-29.)

Executive summary

Key findings

  1. The U.S. average individual audit rate was 0.25% for Tax Year 2019, down from 0.9% in Tax Year 2010, a 72% decline over the decade. (Source: GAO-22-104960, May 2022.)
  2. The IRS estimated individual taxpayers underreported income tax by about $245 billion per year on average, the context for declining audit coverage. (Source: GAO-22-104960, May 2022.)
  3. For Tax Year 2010, individuals with $10 million or more in total positive income were audited at 21.5% in the IRS Data Book all-years file; for Tax Year 2019 that fell to 11.5%. (Source: IRS Data Book historical Table 3-1.)
  4. For Tax Year 2019, individuals with $1 million to $5 million in income were audited at 1.0% (GAO) to 1.8% (IRS Data Book), versus 0.25%-0.3% average. (Sources: GAO-22-104960; IRS Data Book historical Table 3-1.)
  5. For the still-open Tax Year 2023, the IRS Data Book reports a 5.2% coverage rate for $10 million-plus filers and 0.1% for the individual average, both subject to upward revision. (Source: IRS, 2025 Data Book, Table 3-1.)
  6. EITC claimants were audited at 1.8% in Tax Year 2010 and 0.8% in Tax Year 2019, a higher rate than most middle-income bands despite far lower dollar stakes. (Source: GAO-22-104960, Appendix III.)
  7. For Tax Year 2019, the EITC audit rate (about 0.8%) exceeded the rate for filers earning $500,000 to $1 million (0.53%). (Source: GAO-22-104960, footnote 20.)
  8. In the fiscal year ending September 2021, low-income filers under $25,000 faced roughly five times the audit rate of higher earners, about 13 versus 2.6 audits per 1,000 returns. (Source: TRAC, Syracuse University, March 2022.)
  9. Audits of million-dollar individual returns fell to about 14,000 in that year from about 41,000 in 2012, even as the number of millionaire filers rose roughly 50%. (Source: TRAC, Syracuse University, March 2022.)
  10. Partnership returns were examined at 0.1% or less for every tax year from 2017 through 2023 in the IRS Data Book. (Source: IRS Data Book historical Table 3-1.)
  11. S corporation returns were examined at 0.1% or less for Tax Years 2020 through 2023. (Source: IRS Data Book historical Table 3-1.)
  12. C corporations with $20 billion or more in assets were examined at 86.7% for Tax Year 2010 and 49.2% for Tax Year 2019; the asset-size gradient in corporate auditing is extreme. (Source: IRS Data Book historical Table 3-1.)
  13. The average recommended additional tax per audit was $5,169 for filers under $25,000 versus $284,810 for filers with $5 million or more in Fiscal Year 2021. (Source: GAO-22-104960, Table 1.)
  14. Average hours per individual audit more than doubled for high-income returns from Fiscal Years 2010 to 2021 but rose only from 1 to 1.6 hours for EITC audits, which are mostly automated correspondence checks. (Source: GAO-22-104960.)
  15. In Fiscal Year 2024, the IRS estimated 27.3% of EITC payments, about $15.9 billion, were improper, the rationale IRS cites for continued EITC examination. (Source: National Taxpayer Advocate, 2025 Annual Report to Congress.)

Section 1: The individual audit rate has collapsed since 2010

Across every income level, individual audit coverage fell sharply over the 2010s. The decline was driven by IRS budget and staffing cuts, and it was steepest at the top because high-income audits require the most examiner time and expertise.

The overall individual audit rate fell from 0.9% (Tax Year 2010) to 0.25% (Tax Year 2019), a 72% decline. (Source: GAO-22-104960, May 2022.) Audit rates decreased for all income levels but decreased the most for taxpayers with incomes of $200,000 or more. (Source: GAO-22-104960, May 2022.)

The IRS Data Book all-years file confirms the same trajectory using its total positive income (TPI) definition: the individual average fell from about 1.01% (TY2010) to 0.6% (TY2015) to 0.25%-0.3% (TY2019), then to figures of 0.3% (TY2022) and 0.1% (TY2023) that remain open. (Source: IRS Data Book historical Table 3-1.)

What the numbers mean: the headline “audit rate” depends heavily on which year is settled. TY2019 is the most reliable recent benchmark. TY2020 through TY2023 will rise as open exams close, especially at the top, where Inflation Reduction Act funding (enacted August 2022) was directed at high-income individuals, large partnerships, and large corporations. That enforcement buildup is not yet visible in closed-audit rates.

Table 1: Individual audit coverage by total positive income, Tax Years 2010-2023 (IRS Data Book Table 3-1, percent of returns covered)

Shaded years TY2022 and TY2023 are understated due to open examinations.

Total positive income TY2010 TY2013 TY2015 TY2017 TY2019 TY2021 TY2022* TY2023*
No total positive income 20.6 25.5 7.0 9.4 5.6 1.8 0.7 0.3
$1 under $25,000 1.03 0.98 0.7 0.6 0.4 0.5 0.5 0.2
$25,000 under $50,000 0.61 0.43 0.4 0.2 0.2 0.2 0.3 0.1
$50,000 under $75,000 0.69 0.42 0.5 0.3 0.2 0.2 0.1 0.1
$75,000 under $100,000 0.65 0.42 0.5 0.4 0.2 0.2 0.1 <0.05
$100,000 under $200,000 0.85 0.52 0.5 0.4 0.2 0.2 0.1 <0.05
$200,000 under $500,000 2.28 1.61 0.6 0.5 0.3 0.2 0.4 <0.05
$500,000 under $1,000,000 3.62 2.69 1.2 1.2 0.8 0.6 1.4 0.1
$1,000,000 under $5,000,000 8.19 4.91 2.6 2.6 1.8 0.9 1.9 0.2
$5,000,000 under $10,000,000 13.55 8.43 5.2 4.3 3.4 3.9 4.9 2.0
$10,000,000 or more 21.50 13.62 9.5 7.5 11.5 6.6 6.6 5.2
Individual average 1.01 0.65 0.6 0.5 0.3 0.3 0.3 0.1

Source: IRS, Data Book historical all-years file, Table 3-1 (“Examination coverage and recommended additional tax after examination, by type and size of return”), values as of the dates noted in the file (TY2010 as of 9/30/2020; TY2014 as of 9/30/2024). Earlier-year figures shown to two decimals where the source provides them.

Table 2: GAO cross-check, individual audit rates by income, Tax Years 2010 and 2019

GAO uses a slightly different income grouping and counts EITC audits separately, so figures differ modestly from the IRS Data Book but confirm the trend.

Income group (GAO TPI) TY2010 rate TY2019 rate Change 2010 to 2019
$1 to <$25K 1.0% 0.4% -61%
$25K to <$50K 0.6% 0.2% -69%
$50K to <$75K 0.7% 0.1% -79%
$75K to <$100K 0.7% 0.1% -79%
$100K to <$200K 0.8% 0.2% -81%
$200K to <$500K 2.3% 0.2% -92%
$500K to <$1M 3.6% 0.5% -85%
$1M to <$5M 8.2% 1.0% -87%
$5M to <$10M 13.5% 1.4% -90%
$10M or more 21.2% 3.9% -81%
EITC only 1.8% 0.8% -58%
Average 0.9% 0.25% -72%

Source: GAO-22-104960, “Tax Compliance: Trends of IRS Audit Rates and Results for Individual Taxpayers by Income,” May 2022, Table 3 and Appendix III.

Note on the $10M+ divergence: GAO reports 3.9% for TY2019 while the IRS Data Book all-years file reports 11.5%. The gap reflects timing (the IRS file captured more eventually-closed high-end audits as of a later cut date) and definitional differences. The direction is identical: top-bracket coverage fell sharply from 2010, then partially recovered relative to the trough as long-running high-end exams closed.

Section 2: Low-income and EITC filers are audited at above-average rates

The single most cited equity finding in U.S. tax enforcement is that low-income workers claiming the EITC are audited more often than most middle-income filers. The mechanism is automation: EITC examinations are overwhelmingly pre-refund correspondence audits handled by mail and computer matching, which require little examiner time.

In the fiscal year ending September 2021, filers earning under $25,000 were audited at roughly 13 per 1,000 returns, versus 2.6 per 1,000 for filers above $25,000, about five times the rate. (Source: TRAC, Syracuse University, March 2022.) More than half of all IRS correspondence audits that year involved low-income EITC claimants. (Source: TRAC, Syracuse University, March 2022.)

The EITC audit rate was 0.8% for Tax Year 2019, higher than the 0.53% rate for filers earning $500,000 to $1 million that year. (Source: GAO-22-104960, footnote 20.) EITC audit hours rose only from 1 to 1.6 hours per audit from Fiscal Years 2010 to 2021, while high-income audit hours more than doubled. (Source: GAO-22-104960.)

The IRS rationale: in Fiscal Year 2024 the IRS estimated 27.3% of EITC payments, about $15.9 billion, were improper. (Source: National Taxpayer Advocate, 2025 Annual Report to Congress.) Among EITC returns prepared by paid preparers, 96% of the dollar amount of EITC audit adjustments traced to non-credentialed preparers. (Source: National Taxpayer Advocate, 2025 Annual Report to Congress.)

What the numbers mean: the EITC audit rate is not high in absolute terms (under 1% in recent settled years), but it is high relative to comparably-resourced middle-income filers, and it falls disproportionately on the lowest earners because the audits are cheap to run. This is the core of the “audit the poor” critique documented by TRAC and GAO.

Section 3: Businesses, who actually gets examined

Audit risk for businesses depends almost entirely on entity type and asset size. The two facts that dominate: large C corporations are audited at extraordinary rates, and pass-through entities (partnerships and S corporations) are audited at near-zero rates.

Table 3: Business return audit coverage, Tax Years 2010-2023 (percent covered)

Return type TY2010 TY2015 TY2019 TY2021 TY2023*
All corporations (1120 total) 1.55 1.0 0.5 0.4 0.1
C corp, $50M under $100M assets 18.9 14.6 3.5 2.6 1.4
C corp, $1B under $5B assets 46.1 29.8 10.3 5.7 5.8
C corp, $5B under $20B assets 65.7 43.9 27.9 18.5 9.9
C corp, $20B or more assets 86.7 71.6 49.2 35.9 21.9
Partnership returns (1065) 0.48 0.2 0.1 0.1 <0.05
S corporation returns (1120-S) 0.43 0.2 0.1 0.1 <0.05

Source: IRS, Data Book historical all-years file, Table 3-1. Shaded year TY2023 is understated due to open examinations.

What the numbers mean: a partnership or S corporation faces a statistically tiny chance of examination, under 1 in 1,000 in recent years, which is why pass-throughs have drawn scrutiny from GAO and the National Taxpayer Advocate as an enforcement gap. The largest C corporations, by contrast, are audited at rates that approach certainty in the largest asset class. Even at TY2023’s incomplete 21.9%, billion-dollar C corporations are the most-audited filers in America by a wide margin. The 2022 Inflation Reduction Act directed new funding specifically at large-partnership compliance, so partnership rates may rise in tax years that are still open.

Section 4: Why dollar-maximizing enforcement points up even when audit counts do not

Audit frequency and audit yield diverge sharply by income. The IRS recovers far more per audit at the top.

In Fiscal Year 2021, the average recommended additional tax per individual audit was $5,169 (under $25,000), $6,565 ($25K to $200K), $18,263 ($200K to $500K), $46,881 ($500K to $5M), and $284,810 ($5M or more). (Source: GAO-22-104960, Table 1.) The average for EITC-only audits was $4,955. (Source: GAO-22-104960, Table 1.)

From Fiscal Years 2010 to 2021, the majority of total recommended additional tax came from taxpayers with incomes below $200,000, but the recommended tax per audit rose steeply with income. (Source: GAO-22-104960.) Average hours per audit were roughly stable for lower-income filers but more than doubled for filers with $200,000 or more. (Source: GAO-22-104960.)

What the numbers mean: low-income audits are numerous but cheap and low-yield; high-income audits are rare but expensive and high-yield. An enforcement strategy that maximizes revenue per examiner hour pushes toward high income and large entities, which is the stated direction of post-2022 IRS enforcement.

Section 5: The Ledgerism Audit Risk Index (original synthesis)

The Audit Risk Index (ARI) is an original, transparent score built by Ledgerism to express relative audit likelihood across income bands and entity types on a single comparable scale. It is a descriptive index of historical examination coverage, not a prediction of any individual’s audit probability.

What it measures

ARI expresses each segment’s audit coverage rate as a multiple of the all-individual-returns average for the same tax year, then maps that multiple to a 0-100 scale. A score of 50 means “audited at exactly the population-average rate.” Scores above 50 mean above-average audit exposure; below 50 means below-average.

Base data and base year

Formula

For each segment s in tax year t:

  1. Coverage multiple: M(s) = AuditRate(s) / AuditRate(individual average).
  2. Index score: ARI(s) = 50 + 16.7 x log2( M(s) ), capped to the range 0-100.

The log base-2 transform means every doubling of relative audit risk adds about 16.7 points and every halving subtracts about 16.7 points. This keeps both the lightly-audited pass-throughs and the heavily-audited mega-corporations on one readable scale. The constant 16.7 is chosen so that a segment audited 8 times the average (three doublings) scores about 100.

Table 4: Ledgerism Audit Risk Index by segment

Individual rows use TY2019 IRS Data Book rates against the 0.3% individual average. Business rows use TY2019 rates against the same 0.3% individual-return reference, so the index expresses audit exposure relative to a typical individual filer.

Segment TY2019 audit rate Multiple of individual avg (0.3%) ARI score (0-100) Risk tier
C corp, $20B+ assets 49.2% 164x 100 (capped) Extreme
C corp, $5B-20B assets 27.9% 93x 100 (capped) Extreme
C corp, $1B-5B assets 10.3% 34x 100 (capped) Extreme
Individual, $10M+ income 11.5% 38x 100 (capped) Extreme
C corp, $50M-100M assets 3.5% 11.7x 100 (capped) Extreme
Individual, $5M-10M income 3.4% 11.3x 100 (capped) Extreme
Individual, $1M-5M income 1.8% 6.0x 93 Very high
Individual, $500K-1M income 0.8% 2.7x 74 High
EITC claimant (low income) 0.8% 2.7x 74 High
Individual, $200K-500K income 0.3% 1.0x 50 Average
Individual, $100K-200K income 0.2% 0.67x 40 Below average
Individual, $50K-100K income 0.2% 0.67x 40 Below average
Individual, $25K-50K income 0.2% 0.67x 40 Below average
Individual, $1K-25K income 0.4% 1.3x 56 Slightly above average
Partnership return 0.1% 0.33x 23 Low
S corporation return 0.1% 0.33x 23 Low

Source for inputs: IRS, Data Book historical Table 3-1 (TY2019 coverage rates); GAO-22-104960 (EITC). Index construction by The Ledgerism Brief, 2026-06-29.

How to read the index

Limitations of the index

Section 6: Two more original derived insights

Insight A: The “audit compression” ratio, top versus bottom

In Tax Year 2010, the $10M+ individual audit rate (21.5%) was about 21 times the rate for the $1K-25K band (1.03%) in the IRS Data Book. By Tax Year 2019 that ratio compressed: the $10M+ rate (11.5%) was about 29 times the lowest band (0.4%) in the Data Book, but under GAO’s accounting (3.9% versus 0.4%) the top-to-bottom ratio fell to about 10x. The decade’s audit cuts narrowed the gap between top and bottom enforcement, not because the bottom was spared, but because the top was cut faster. (Sources: IRS Data Book historical Table 3-1; GAO-22-104960. Ledgerism calculation.)

Insight B: Yield-per-audit gradient

Using GAO Fiscal Year 2021 averages, the recommended additional tax per audit at $5M+ income ($284,810) was about 55 times the figure for filers under $25,000 ($5,169) and about 57 times the EITC-only figure ($4,955). One high-income audit yields, on average, what roughly 55 low-income audits yield. This single ratio explains the post-2022 strategic pivot toward high-income and large-entity enforcement. (Source: GAO-22-104960, Table 1. Ledgerism calculation.)

Charts to create

  1. Title: “The Great Audit Decline, 2010-2019.” Data needed: individual average audit rate by tax year (0.9% to 0.25%) plus the $10M+ and EITC lines. Source: GAO-22-104960; IRS Data Book Table 3-1. Insight: every income level was cut, the top fastest. Citation-worthy because it is the canonical equity-in-enforcement chart.
  2. Title: “Who Gets Audited: TY2019 Audit Rate by Income Band.” Data needed: TY2019 rates from no-TPI through $10M+, with the EITC bar highlighted. Source: IRS Data Book Table 3-1; GAO. Insight: the U-shape, high at the very top and elevated at the very bottom (EITC), low in the middle.
  3. Title: “Corporate Audit Rates Track Asset Size.” Data needed: TY2019 C corp rates by asset class plus partnership and S corp rates. Source: IRS Data Book Table 3-1. Insight: a near-vertical gradient from 0.1% (pass-throughs) to 49.2% ($20B+ C corps).
  4. Title: “The Ledgerism Audit Risk Index.” Data needed: ARI scores from Table 4, sorted. Source: Ledgerism index. Insight: large C corps, top individuals, and EITC filers cluster high; pass-throughs and the middle class cluster low.
  5. Title: “Yield per Audit by Income, FY2021.” Data needed: average recommended additional tax per audit by income band. Source: GAO-22-104960, Table 1. Insight: the 55x top-to-bottom yield gradient.

Methodology

Source selection: this briefing uses only primary or primary-citing sources: the IRS Data Book (current 2025 edition and the historical all-years Table 3-1 Excel file), the IRS Compliance Presence page, GAO report GAO-22-104960, the National Taxpayer Advocate 2025 Annual Report to Congress, and TRAC at Syracuse University (which analyzes IRS administrative data obtained under FOIA).

Inclusion and exclusion rules: every statistic carries a specific tax year or fiscal year and a geography (United States, federal returns). Where the IRS marks a year as shaded (open examinations), the figure is labeled as understated. Figures that could not be tied to a primary source with a year were excluded. No projections are presented as current.

Handling of conflicting numbers: the IRS Data Book and GAO use different income definitions and different data-cut dates, producing different rates for the same band and year (most visibly TY2019 $10M+: 11.5% IRS versus 3.9% GAO). Both are shown, the divergence is explained, and neither is presented as the single truth.

Derived figures: the Audit Risk Index formula, the audit-compression ratio, and the yield-per-audit gradient are Ledgerism calculations from the cited primary inputs; the formulas and inputs are shown in full so any reader can reproduce them.

Data limitations: recent tax years (2020-2023) are understated due to the look-back lag; pass-through and high-wealth enforcement funded by the 2022 Inflation Reduction Act is not yet reflected in closed-audit rates; audit distribution by race is a modeled academic estimate, not an IRS statistic, and is therefore not tabulated here.

Date of last update: 2026-06-29.

Source quality ranking

Tier 1 (primary government and academic data):
– IRS, 2025 Data Book, Table 3-1 (current examination coverage).
– IRS, Data Book historical all-years Table 3-1 file (TY2010-TY2023 time series).
– IRS, Compliance Presence page (statute-of-limitations framing).
– U.S. GAO, GAO-22-104960 (audit-rate trends by income, May 2022).
– National Taxpayer Advocate, 2025 Annual Report to Congress (EITC improper payments, correspondence audits).
– TRAC at Syracuse University (analysis of IRS administrative data, FY-based audit rates).

Tier 2 / Tier 3: secondary press coverage (CBS News, Journal of Accountancy) was used only to locate the underlying TRAC and GAO figures and is not relied on for any number; all figures trace to the Tier 1 sources above.

Excluded: commercial “audit statistics” pages that did not cite a primary source; any race-of-taxpayer audit percentages (the leading estimate is a Stanford/Treasury working-paper model, not an IRS-published statistic, and is outside the scope of verifiable IRS coverage data); projected post-IRA audit rates that have not yet appeared in closed-audit data.

Citation format (per major statistic)

Journalist-friendly additions

Most quotable statistics

Data limitations

Recent tax years are understated by the look-back lag; IRS and GAO figures are not perfectly comparable; partnership and high-wealth enforcement funded after 2022 is not yet in the closed-audit data; race-based audit distribution is modeled, not IRS-reported.

Downloadable dataset, recommended fields

tax_year, fiscal_year, segment_type (individual_income_band / EITC / corporation_asset_band / partnership / s_corporation), segment_label, returns_filed, returns_examined_closed, returns_examined_in_process, audit_coverage_pct, avg_recommended_addl_tax_usd, avg_hours_per_audit, ari_score, risk_tier, source, source_url, data_cut_date, lookback_flag (true if year still open).

150-word press summary

A new Ledgerism Brief analysis of IRS, GAO, and Syracuse University TRAC data shows the average individual audit rate fell 72% between Tax Years 2010 and 2019, from 0.9% to 0.25%, with the steepest cuts at the top. Yet enforcement equity remains lopsided. The largest C corporations, those with $20 billion or more in assets, were audited at 86.7% in Tax Year 2010 and still 49.2% in 2019. Low-income workers claiming the Earned Income Tax Credit were audited about five times as often as everyone else in the year ending September 2021, and more than half of all IRS mail audits targeted these filers. Partnerships and S corporations were examined at one in a thousand or less. The briefing introduces the Ledgerism Audit Risk Index, scoring relative audit likelihood by income and entity on a single 0-100 scale, where low-income EITC filers score the same as near-millionaires.

Five suggested headlines

  1. “Who the IRS Actually Audits: New Index Puts Mega-Corporations and EITC Filers at the Top”
  2. “The Audit Rate Fell 72% in a Decade, but the Poor Still Get Audited Five Times as Often”
  3. “America’s Most-Audited Filers Are $20 Billion Corporations and the Working Poor”
  4. “The Ledgerism Audit Risk Index: One Score for Every Income Band and Entity Type”
  5. “Partnerships and S Corps Face a 1-in-1,000 Audit Rate. Here Is the Data.”

Ten FAQs answered with verified statistics

  1. What is the overall IRS audit rate? For Tax Year 2019, the most recent settled year, the individual average was 0.25% (GAO) to 0.3% (IRS Data Book). Recent years are understated due to open exams. (GAO-22-104960; IRS Data Book Table 3-1.)
  2. Who is most likely to be audited? The largest C corporations ($20B+ assets, 49.2% in TY2019) and the highest-income individuals ($10M+, 3.9% GAO to 11.5% IRS in TY2019). (IRS Data Book Table 3-1; GAO.)
  3. Are low-income people really audited more? Relative to comparably-resourced middle-income filers, yes. In the year ending September 2021, under-$25,000 filers were audited about five times as often as higher earners. (TRAC, 2022.)
  4. Why are EITC claimants audited so much? IRS estimated 27.3% of EITC payments ($15.9B) were improper in FY2024, and EITC audits are cheap automated mail checks (about 1.6 hours each). (National Taxpayer Advocate 2025; GAO.)
  5. How likely is a partnership to be audited? Very unlikely; 0.1% or less for every tax year from 2017 through 2023. (IRS Data Book Table 3-1.)
  6. How likely is an S corporation to be audited? 0.1% or less for Tax Years 2020 through 2023. (IRS Data Book Table 3-1.)
  7. Why are recent audit rates so low? The look-back lag: exams for recent years are still open, so the IRS shades the two most recent tax years as incomplete. (IRS Data Book Table 3-1 notes.)
  8. How much does an audit recover? In FY2021, an average of $5,169 per audit under $25,000 versus $284,810 per audit at $5M+. (GAO-22-104960, Table 1.)
  9. Did the IRS cut audits more for the rich or the poor? More for the rich in percentage terms; the $200K-500K band fell 92% and $1M-5M fell 87% from 2010 to 2019. (GAO-22-104960.)
  10. What is the Ledgerism Audit Risk Index? An original 0-100 score of relative audit likelihood by income and entity, base year TY2019, where 50 equals the individual-average rate. (The Ledgerism Brief, 2026.)

Sources

  1. Internal Revenue Service, 2025 Data Book (Publication 55B), Table 3-1. https://www.irs.gov/pub/irs-pdf/p55b.pdf
  2. Internal Revenue Service, SOI Tax Stats, Examination Coverage by Type and Size of Return, IRS Data Book Table 3-1 (current and historical all-years files). https://www.irs.gov/statistics/soi-tax-stats-examination-coverage-and-recommended-additional-tax-after-examination-by-type-and-size-of-return-irs-data-book-table-3-1
  3. Internal Revenue Service, Compliance Presence. https://www.irs.gov/statistics/compliance-presence
  4. U.S. Government Accountability Office, GAO-22-104960, “Tax Compliance: Trends of IRS Audit Rates and Results for Individual Taxpayers by Income,” May 17, 2022. https://www.gao.gov/products/gao-22-104960 (full report: https://www.gao.gov/assets/gao-22-104960.pdf)
  5. National Taxpayer Advocate, 2025 Annual Report to Congress, Taxpayer Advocate Service / IRS. https://www.taxpayeradvocate.irs.gov/reports/2025-annual-report-to-congress/full-report/
  6. Transactional Records Access Clearinghouse (TRAC), Syracuse University, IRS audit analysis. https://tracreports.org/tracirs/
  7. CBS News (locator source for TRAC figures, not relied upon for numbers), “IRS audits the poor at 5 times the rate of everyone else.” https://www.cbsnews.com/news/irs-audit-eitc-five-times-as-likely-to-get-audited/

Related research

More original, sourced datasets from The Ledgerism Brief: