Distribution of IRS Audits by Income and Race
link to page 1 
Updated November 6, 2023
Distribution of IRS Audits by Income and Race
The Internal Revenue Service (IRS) enforces the federal tax
Responsibility Act of 2023 (P.L. 118-5). Negotiators also
code. Among the ways it does so is by conducting audits to
informally agreed to rescind $10 billion from the IRS
verify the information on a taxpayer’s return is accurate and
during both the FY2024 and FY2025 appropriations
the correct amount of tax has been paid. Some in Congress
processes. Secretary of the Treasury Janet Yellen directed
have expressed concern about who is audited and how that
the IRS not to use the enforcement funds in the IRA to raise
has changed over time.
audit rates above historical levels for individual taxpayers
with incomes below $400,000 annually.
Less than 1% of individual income tax returns are selected
for audit each year. The audit rate has fallen for all income
Distribution by Income
groups since 2010, but it has declined most for high-income
The IRS generally audits a larger share of high-income
taxpayers. While the IRS still audits a greater share of high-
taxpayers than those with lower incomes, as illustrated in
income filers than low-income ones, low earners who claim
Figure 1. However, those who claim the Earned Income
the Earned Income Tax Credit (EITC) face much higher
Tax Credit (EITC)—who typically have low incomes—are
audit rates than other taxpayers with similar incomes.
much more likely to face an audit than all but the highest-
Research also shows that the IRS is more likely to audit
income taxpayers. While the average audit rate for all
Black taxpayers than those of other races, even though the
domestic taxpayers with positive income who filed an
IRS does not collect information on taxpayer race.
individual income tax return in tax year 2019 was 0.29%,
the rate was 0.78% for EITC claimants.
Background on Audits
The IRS selects returns for audit using a variety of methods,
Figure 1. Audit Rates by Positive Income,
including determining the likelihood that there may be an
2019 Returns
error on the return.
Audits generally take one of two forms: correspondence
and field. The most common type of audit is a
correspondence audit, where an IRS examiner requests
information from a taxpayer by mail. Correspondence
audits typically target one or several tax items rather than
all the information provided on a return. In a field audit, an
examiner typically meets a taxpayer at their home or place
of business to review their records. Field audits typically
examine more tax items in a filing than correspondence
audits do. The IRS closed roughly 626,000 audits of
individual income tax returns in FY2022, 85% of which
were correspondence audits. The agency’s National
Research Program (NRP) also audits a small, semi-random
Source: Internal Revenue Service, IRS Data Book, Publication 55-B,
selection of filers thoroughly to study the nature and
March 2023, Table 17, https://www.irs.gov/pub/irs-pdf/p55b.pdf.
frequency of tax nonpayment.
Notes: Audits of EITC returns are included in the audit rates of
taxpayers with $1-25K and $25K-$500K of income. These figures
The IRS had audited 0.29% of 2019 individual income tax
were published before the three-year statute of limitations on audits
returns with positive income as of March 2023, a drop from
ended.
the 0.89% of filers that were audited in 2010. This decline
reflects a decrease in appropriations for enforcement at the
The IRS audited a smaller share of filers in 2019 than 2010
IRS. Outlays for enforcement, measured in inflation-
at every income level. According to the agency, it did so
adjusted dollars, fell by 26% from FY2010 to FY2022.
because of decreased funding (when controlling for
inflation), staff attrition, and major disruptions such as a
Congress made $46 billion available to the IRS for
government shutdown and the COVID-19 pandemic.
enforcement activities—in addition to normal annual
appropriations—through FY2031 as part of the $80 billion
The likelihood that high-income taxpayers will face an
for the IRS and related agencies in the law commonly
audit has fallen by more than the same decline in audit
referred to as the Inflation Reduction Act of 2022 (IRA;
probability for lower-income taxpayers. While the audit rate
P.L. 117-169). Congress subsequently rescinded $1.4
for those earning over $5 million fell by 66% for returns
billion of these total IRA funds through the Fiscal
https://crsreports.congress.gov
Distribution of IRS Audits by Income and Race
from tax years 2010 to 2019, the rate for those earning
selecting similar returns based on likely total tax
below $25,000 fell by 57%. The IRS says that resource
underpayment.
constraints have compelled it to focus more of its resources
on relatively simple, cost-effective audits, such as
According to Elzayn et al., approximately 14% of the
correspondence audits on EITC recipients.
disparity is attributable to the fact that black filers are more
likely to claim the EITC. Of the remainder, 78% is
According to the Government Accountability Office
attributable to higher audit rates among Black EITC
(GAO), the IRS audited taxpayers who claimed the EITC
claimants relative to other EITC claimants, while 8% result
more frequently than many other types of taxpayers because
from higher audit rates among Black non-claimants than
the EITC is often claimed improperly. From 2006 to 2008,
other non-claimants. This would imply that to reduce the
the IRS estimated that 42%-49% of tax returns that
race gap, the IRS should focus on reducing disparities in
included an EITC incorrectly claimed the credit.
audits of EITC claimants. However, preliminary analysis by
Hertz et al. (2023) suggests that auditing Black and non-
The IRS may also view audits of EITC claims as more cost-
Black EITC claimants at the same rate would have the same
effective than others. GAO estimated that the IRS collects
effect on reducing the audit race gap as auditing EITC and
roughly equal amounts of money per hour of auditing from
non-EITC claimants at the same rate.
EITC claimants and those with incomes between $5 and
$10 million. Audits of EITC returns—almost all of which
Hertz et al. acknowledged other potential causes of the
are correspondence audits—typically also require less time
audit race gap, including that tax preparers who submit
than those of high-income taxpayers, who are more likely to
many improper claims for the EITC—and whose returns
use sophisticated techniques that are difficult to detect.
may be subject to additional scrutiny by the IRS—have
Further, since most audits of EITC returns are done before
disproportionately Black client bases. Additionally, tax
the credit is issued, such audits may be more successful in
benefits such as the child tax credit and the EITC use
“protecting revenue.” Additionally, high-income taxpayers
different eligibility criteria to determine the number of
may be able to reduce the tax owed (abatement), lowering
children a filer can claim for each credit. These differences
the amount collected. Audit practices that emphasize cost
in eligibility criteria—and the criteria themselves—may be
efficiency may raise federal revenues but could lead to
particularly confusing in situations in which the child’s
unexpected disparate treatment among different types of
parents are unmarried, which is more common among
taxpayers.
Black households than others.
There may also be concerns with the higher audit rates of
Hertz et al. also identified within the IRS’s audit selection
taxpayers who claim the EITC. Low-income filers are less
system potential sources of “algorithmic bias,” meaning
likely to respond to IRS correspondence audits adequately
factors in the structure of the IRS’s technical systems that
or at all, meaning the IRS may incorrectly disallow the
favor non-Black filers in ways that are neither explained by
credit. These audits may also have longer-term
the selection objectives themselves nor the goals of the
implications. EITC claimants whom the IRS has audited
system. The IRS models the residency of children and their
appear to be less likely to claim the credit in subsequent
relationship to filers indirectly, and as the researchers
years, even if the audit found them compliant. Further,
found, imperfectly. Errors in this system appeared to raise
other research has found that audits of high-income filers
the audit risk for Black EITC claimants relative to others.
still result in more increased revenue per dollar spent than
The IRS previously over-selected returns for audits in
audits of lower earners, despite their higher up-front cost.
certain weeks, which contributed to the implicit racial bias.
The authors stated that improved scheduling has largely
Distribution by Race
resolved this issue.
The likelihood a taxpayer will face an audit also varies by
race. After imputing the race of taxpayers, Elzayn et al.
Recent IRS Action
(2023) used research data from NRP audits and found that
On September 18, 2023, Commissioner Werfel said the IRS
the IRS audits Black filers 2.9 to 4.7 times more often than
will audit a smaller share of taxpayers who claim the EITC
it does others, even though the IRS does not use race in its
and other refundable tax credits, and will instead focus
audit selection process. IRS Commissioner Danny Werfel
more resources on examining large partnerships and
said that the IRS independently investigated the same issue,
collecting owed taxes from high-income individuals. The
and that “while there is a need for further research, our
Commissioner also said the IRS has implemented changes
initial findings support the conclusion that Black taxpayers
to how it measures children’s residency that “testing
may be audited at higher rates than would be expected
suggests will increase the expected return on investment for
given their share of the population.” Studies to date have
cases selected while simultaneously reducing disparities.”
not publicly presented disaggregated results for non-Black
He added that the agency is piloting two alternative EITC
taxpayers of different races.
case selection processes. Taken together, these changes aim
to reduce the audit rates on lower-income taxpayers and
Elzayn et al. used several hypothetical methods of selecting
shrink the gap in audit rates between taxpayers of different
returns for audit to show that different audit selection
races.
models can select returns with different demographic
profiles. For example, selecting EITC returns for audit
Brendan McDermott, Analyst in Public Finance
based on the likelihood of a refundable credit overpayment
flags filings by Black households at a higher rate than
IF12521
https://crsreports.congress.gov
Distribution of IRS Audits by Income and Race
Disclaimer
This document was prepared by the Congressional Research Service (CRS). CRS serves as nonpartisan shared staff to
congressional committees and Members of Congress. It operates solely at the behest of and under the direction of Congress.
Information in a CRS Report should not be relied upon for purposes other than public understanding of information that has
been provided by CRS to Members of Congress in connection with CRS’s institutional role. CRS Reports, as a work of the
United States Government, are not subject to copyright protection in the United States. Any CRS Report may be
reproduced and distributed in its entirety without permission from CRS. However, as a CRS Report may include
copyrighted images or material from a third party, you may need to obtain the permission of the copyright holder if you
wish to copy or otherwise use copyrighted material.
https://crsreports.congress.gov | IF12521 · VERSION 2 · UPDATED
Updated November 6, 2023
Distribution of IRS Audits by Income and Race
The Internal Revenue Service (IRS) enforces the federal tax
Responsibility Act of 2023 (P.L. 118-5). Negotiators also
code. Among the ways it does so is by conducting audits to
informally agreed to rescind $10 billion from the IRS
verify the information on a taxpayer’s return is accurate and
during both the FY2024 and FY2025 appropriations
the correct amount of tax has been paid. Some in Congress
processes. Secretary of the Treasury Janet Yellen directed
have expressed concern about who is audited and how that
the IRS not to use the enforcement funds in the IRA to raise
has changed over time.
audit rates above historical levels for individual taxpayers
with incomes below $400,000 annually.
Less than 1% of individual income tax returns are selected
for audit each year. The audit rate has fallen for all income
Distribution by Income
groups since 2010, but it has declined most for high-income
The IRS generally audits a larger share of high-income
taxpayers. While the IRS still audits a greater share of high-
taxpayers than those with lower incomes, as illustrated in
income filers than low-income ones, low earners who claim
Figure 1. However, those who claim the Earned Income
the Earned Income Tax Credit (EITC) face much higher
Tax Credit (EITC)—who typically have low incomes—are
audit rates than other taxpayers with similar incomes.
much more likely to face an audit than all but the highest-
Research also shows that the IRS is more likely to audit
income taxpayers. While the average audit rate for all
Black taxpayers than those of other races, even though the
domestic taxpayers with positive income who filed an
IRS does not collect information on taxpayer race.
individual income tax return in tax year 2019 was 0.29%,
the rate was 0.78% for EITC claimants.
Background on Audits
The IRS selects returns for audit using a variety of methods,
Figure 1. Audit Rates by Positive Income,
including determining the likelihood that there may be an
2019 Returns
error on the return.
Audits generally take one of two forms: correspondence
and field. The most common type of audit is a
correspondence audit, where an IRS examiner requests
information from a taxpayer by mail. Correspondence
audits typically target one or several tax items rather than
all the information provided on a return. In a field audit, an
examiner typically meets a taxpayer at their home or place
of business to review their records. Field audits typically
examine more tax items in a filing than correspondence
audits do. The IRS closed roughly 626,000 audits of
individual income tax returns in FY2022, 85% of which
were correspondence audits. The agency’s National
Research Program (NRP) also audits a small, semi-random
Source: Internal Revenue Service, IRS Data Book, Publication 55-B,
selection of filers thoroughly to study the nature and
March 2023, Table 17, https://www.irs.gov/pub/irs-pdf/p55b.pdf.
frequency of tax nonpayment.
Notes: Audits of EITC returns are included in the audit rates of
taxpayers with $1-25K and $25K-$500K of income. These figures
The IRS had audited 0.29% of 2019 individual income tax
were published before the three-year statute of limitations on audits
returns with positive income as of March 2023, a drop from
ended.
the 0.89% of filers that were audited in 2010. This decline
reflects a decrease in appropriations for enforcement at the
The IRS audited a smaller share of filers in 2019 than 2010
IRS. Outlays for enforcement, measured in inflation-
at every income level. According to the agency, it did so
adjusted dollars, fell by 26% from FY2010 to FY2022.
because of decreased funding (when controlling for
inflation), staff attrition, and major disruptions such as a
Congress made $46 billion available to the IRS for
government shutdown and the COVID-19 pandemic.
enforcement activities—in addition to normal annual
appropriations—through FY2031 as part of the $80 billion
The likelihood that high-income taxpayers will face an
for the IRS and related agencies in the law commonly
audit has fallen by more than the same decline in audit
referred to as the Inflation Reduction Act of 2022 (IRA;
probability for lower-income taxpayers. While the audit rate
P.L. 117-169). Congress subsequently rescinded $1.4
for those earning over $5 million fell by 66% for returns
billion of these total IRA funds through the Fiscal
https://crsreports.congress.gov
Distribution of IRS Audits by Income and Race
from tax years 2010 to 2019, the rate for those earning
selecting similar returns based on likely total tax
below $25,000 fell by 57%. The IRS says that resource
underpayment.
constraints have compelled it to focus more of its resources
on relatively simple, cost-effective audits, such as
According to Elzayn et al., approximately 14% of the
correspondence audits on EITC recipients.
disparity is attributable to the fact that black filers are more
likely to claim the EITC. Of the remainder, 78% is
According to the Government Accountability Office
attributable to higher audit rates among Black EITC
(GAO), the IRS audited taxpayers who claimed the EITC
claimants relative to other EITC claimants, while 8% result
more frequently than many other types of taxpayers because
from higher audit rates among Black non-claimants than
the EITC is often claimed improperly. From 2006 to 2008,
other non-claimants. This would imply that to reduce the
the IRS estimated that 42%-49% of tax returns that
race gap, the IRS should focus on reducing disparities in
included an EITC incorrectly claimed the credit.
audits of EITC claimants. However, preliminary analysis by
Hertz et al. (2023) suggests that auditing Black and non-
The IRS may also view audits of EITC claims as more cost-
Black EITC claimants at the same rate would have the same
effective than others. GAO estimated that the IRS collects
effect on reducing the audit race gap as auditing EITC and
roughly equal amounts of money per hour of auditing from
non-EITC claimants at the same rate.
EITC claimants and those with incomes between $5 and
$10 million. Audits of EITC returns—almost all of which
Hertz et al. acknowledged other potential causes of the
are correspondence audits—typically also require less time
audit race gap, including that tax preparers who submit
than those of high-income taxpayers, who are more likely to
many improper claims for the EITC—and whose returns
use sophisticated techniques that are difficult to detect.
may be subject to additional scrutiny by the IRS—have
Further, since most audits of EITC returns are done before
disproportionately Black client bases. Additionally, tax
the credit is issued, such audits may be more successful in
benefits such as the child tax credit and the EITC use
“protecting revenue.” Additionally, high-income taxpayers
different eligibility criteria to determine the number of
may be able to reduce the tax owed (abatement), lowering
children a filer can claim for each credit. These differences
the amount collected. Audit practices that emphasize cost
in eligibility criteria—and the criteria themselves—may be
efficiency may raise federal revenues but could lead to
particularly confusing in situations in which the child’s
unexpected disparate treatment among different types of
parents are unmarried, which is more common among
taxpayers.
Black households than others.
There may also be concerns with the higher audit rates of
Hertz et al. also identified within the IRS’s audit selection
taxpayers who claim the EITC. Low-income filers are less
system potential sources of “algorithmic bias,” meaning
likely to respond to IRS correspondence audits adequately
factors in the structure of the IRS’s technical systems that
or at all, meaning the IRS may incorrectly disallow the
favor non-Black filers in ways that are neither explained by
credit. These audits may also have longer-term
the selection objectives themselves nor the goals of the
implications. EITC claimants whom the IRS has audited
system. The IRS models the residency of children and their
appear to be less likely to claim the credit in subsequent
relationship to filers indirectly, and as the researchers
years, even if the audit found them compliant. Further,
found, imperfectly. Errors in this system appeared to raise
other research has found that audits of high-income filers
the audit risk for Black EITC claimants relative to others.
still result in more increased revenue per dollar spent than
The IRS previously over-selected returns for audits in
audits of lower earners, despite their higher up-front cost.
certain weeks, which contributed to the implicit racial bias.
The authors stated that improved scheduling has largely
Distribution by Race
resolved this issue.
The likelihood a taxpayer will face an audit also varies by
race. After imputing the race of taxpayers, Elzayn et al.
Recent IRS Action
(2023) used research data from NRP audits and found that
On September 18, 2023, Commissioner Werfel said the IRS
the IRS audits Black filers 2.9 to 4.7 times more often than
will audit a smaller share of taxpayers who claim the EITC
it does others, even though the IRS does not use race in its
and other refundable tax credits, and will instead focus
audit selection process. IRS Commissioner Danny Werfel
more resources on examining large partnerships and
said that the IRS independently investigated the same issue,
collecting owed taxes from high-income individuals. The
and that “while there is a need for further research, our
Commissioner also said the IRS has implemented changes
initial findings support the conclusion that Black taxpayers
to how it measures children’s residency that “testing
may be audited at higher rates than would be expected
suggests will increase the expected return on investment for
given their share of the population.” Studies to date have
cases selected while simultaneously reducing disparities.”
not publicly presented disaggregated results for non-Black
He added that the agency is piloting two alternative EITC
taxpayers of different races.
case selection processes. Taken together, these changes aim
to reduce the audit rates on lower-income taxpayers and
Elzayn et al. used several hypothetical methods of selecting
shrink the gap in audit rates between taxpayers of different
returns for audit to show that different audit selection
races.
models can select returns with different demographic
profiles. For example, selecting EITC returns for audit
Brendan McDermott, Analyst in Public Finance
based on the likelihood of a refundable credit overpayment
flags filings by Black households at a higher rate than
IF12521
https://crsreports.congress.gov
Distribution of IRS Audits by Income and Race
Disclaimer
This document was prepared by the Congressional Research Service (CRS). CRS serves as nonpartisan shared staff to
congressional committees and Members of Congress. It operates solely at the behest of and under the direction of Congress.
Information in a CRS Report should not be relied upon for purposes other than public understanding of information that has
been provided by CRS to Members of Congress in connection with CRS’s institutional role. CRS Reports, as a work of the
United States Government, are not subject to copyright protection in the United States. Any CRS Report may be
reproduced and distributed in its entirety without permission from CRS. However, as a CRS Report may include
copyrighted images or material from a third party, you may need to obtain the permission of the copyright holder if you
wish to copy or otherwise use copyrighted material.
https://crsreports.congress.gov | IF12521 · VERSION 2 · UPDATED