Once Bitten, Twice Shy?
The Lasting Impact of IRS Audits on Individual Tax Reporting*
Jason DeBacker†, Bradley T. Heim‡,
Anh Tran§, and Alexander Yuskavage**
October 28, 2015

Abstract
This paper studies the impact of tax enforcement activity on subsequent individual taxpaying
behavior. We exploit three waves of randomized Internal Revenue Service (IRS) audits of
individual income tax filers during the 2006-2009 period to study both the short and long
run effects of audits on taxpaying behavior. Rich and confidential IRS data allow us to
discover several effects on post-audit taxpaying behavior. First, we find that taxpayers
respond to audits by increasing reported taxable income. This impact is stronger but more
quickly diminishing for incomes not subject to third-party information. Second, individuals
with higher income volatility are found to owe more tax, conceivably because of a larger
degree of asymmetric information between the filer and enforcement agency. Finally, we
find larger and more persistent responses among those who are less tax literate. Overall, the
IRS’s revenue gain from the long-term impact of audits is about twice as large as the revenue
gained directly from the audit itself.
Keywords: individual income tax, tax audit, tax evasion, tax avoidance
JEL Classifications: H24, H26

________________________________
* We thank Alan Plumley and John Guyton for helpful discussions, seminar participants at George Mason’s
SPGIA, Harvard Kennedy School, the Spring 2015 NBER Public Economics Program Meetings, and the 2014
National Tax Association Annual Meetings. The views expressed in this paper are those of the authors and do
not necessarily reflect the views of the U.S. Department of the Treasury or the Office of Tax Analysis.
† Middle Tennessee State University, T: (615) 898-2528, E: jason.debacker@gmail.com; ‡ School of Public and
Environmental Affairs, Indiana University, T: (812) 855-9783, E: heimb@indiana.edu; § School of Public and
Environmental Affairs, Indiana University, T: (812) 855-0563, E: trananh@indiana.edu; ** Office of Tax
Analysis, U.S. Department of the Treasury, T: (202) 622-0694, E: alexander.yuskavage@treasury.gov.

1

 1. Introduction
According to the Internal Revenue Service, the tax gap - the difference between taxes paid
and what taxpayers should be paying under law - was approximately $385 billion in 2006
(IRS 2012). The majority of the tax gap is associated with the individual income tax. To
increase tax compliance, the IRS audits about 1% of individual filers each year (TRAC
Reports, Inc. 2014). These audits have both immediate and long-term effects on tax filers.
The immediate effect is that the taxpayer must make up tax discrepancies uncovered during
the audits. The long-term effect is a deterrence of future behavior by the audited tax filers.
This study focuses on these long-term effects of audits by examining changes in the
taxpaying behavior of a large sample of tax filers who were randomly audited by the IRS.
In economics, the formal models of tax compliance date back at least to Allingham
and Sandmo (1972) who build on the earlier work on the economics of crime by Becker
(1968). While the theoretical literature on tax compliance has extended the seminal model of
Allingham and Sandmo (1972) in a number of ways, empirical work on how audits shape the
subsequent behavior of taxpayers is scant. More broadly, there have been very few rigorous
empirical studies on how audits or other enforcement activities can affect long-run behaviors
of different kinds.
Conceptually, there are two conflicting expectations about the effects of audits on
taxpaying behavior. The first, and perhaps more intuitive, is that experiencing an audit leads
taxpayers to revise their perceived audit probability up and therefore reduce their subsequent
noncompliance. Afterwards, for each year they do not experience another audit, taxpayers
may revise the perceived probability downward, encouraging noncompliance. Therefore the
post-audit tax payment trend would consist of an immediate increase followed by a decrease.
The second hypothesis predicts essentially opposite behavior. Taxpayers may, correctly or
2

 incorrectly, believe that auditors rarely come back immediately after an audit, and therefore
that it is safest to evade taxes immediately following an audit. As years pass, the risk that
auditors will return increases and thus it is best to reduce noncompliance. In this case, the
post-audit tax payment trend would be an immediate fall followed by an increase.
Empirically, two new studies using taxpayer data find support for both of these patterns
of behavior. DeBacker et al. (forthcoming) use U.S. tax data to show that corporations tend
to pay less tax following an audit. In contrast, Advani et al. (2015) use U.K. data to show that
self-employed individuals in the U.K. tend to pay more tax after an audit. Previous studies rely
mostly on lab experiments and can show either pattern.1 These conflicting expectations and
patterns suggest that the important question of the long-term impact of audits is not yet
resolved, either conceptually or empirically.
To tackle this question, we need an understanding of how taxpayers determine their
level of compliance. For example, there is the observation of why, given the very low audit
rates and penalties, people choose to evade so little tax. To explain this puzzle, some studies
have shown that moral and culture factors can play an important role in mitigating people’s
willingness to evade tax.2 Recently, Kleven et al. (2011) point out the importance of thirdparty information, which makes Danish taxpayers unable, instead of unwilling, to cheat.
Kleven and his coauthors highlight the problem of information frictions in the area of tax
compliance.
In this paper, we propose that information frictions may also help us to understand the
long-term impact of audits. In fact, tax authority usually has imperfect information about the
true income of a taxpayer. The tax authority makes attempts update this information from
Notable lab studies are Alm and Mckee (2004), Fortin et al. (2007), Kirchler et al (2007) and Kastlunger et al
(2009), and Choo et al. (2013).
2 Notable studies in this area include Alm, McClelland, and Schulze (1992), Andreoni, Erard, and Feinstein
(1998), and Feld and Frey (2002).
1

3

 two main sources, namely, third-party reporting and tax audits. While the information from a
third-party (e.g. an employer) tends to be updated annually, the audit information is updated
only when an audit is conducted. When audits are not conducted, information from previous
audits gradually becomes outdated. The speed of updating and outdating of income
information allows those who wish to evade tax the opportunity to strategically manipulate
their income reporting.
To test for the long-term effects of audits on tax reporting, we use data from the
Internal Revenue Service’s (IRS) National Research Program (NRP). The NRP began
conducting random audits of individual tax filers starting in tax year 2001, and began
conducting annual random audits starting in 2006. To these data, we merge returns from the
universe of filers from 2000 to 2012, allowing us to examine the impact of audits on
individual taxpaying behavior for a period of up to six years after an audit.3 These data are
instrumental in addressing the impact of legal enforcement on subsequent behavior for
several reasons. First, the IRS conducts intermittent audits and keeps systematic records of
them. Second, the IRS provides accurate data on subsequent tax payments in each year
following the audit, even when there is no audit. Third, these data comprise a panel of the
entire population of individual taxpayers over time, allowing for rigorous empirical analysis.
Since the treatment (an IRS audit) is randomized through the NRP, our empirical
strategy is straightforward. Using the sampling weights that the IRS uses to select individuals
for an NRP audit, we construct a nationally representative sample of audited individuals. We
pair this with a random sample of individuals drawn from the same population of tax filers.
Then, we compare the tax filings of these two groups before and after the audit year.

The IRS defines an audit as “a review/examination of an organization’s or individual’s accounts and financial
information to ensure information is being reported correctly, according to the tax laws, to verify the amount
of tax reported is correct.” (Internal Revenue Service (2014)).
3

4

 We begin by using this U.S. data to test for the main results that Kleven et al (2011) find
among Danish taxpayers. We find that an audit increases subsequent reported wage income
by 1.3% and Schedule C income by 14.2% in the US on average. Thus we find that the effect
of audit on incomes subject to third-party information reporting is much smaller than on
incomes that are not subject to third-party reporting. This result is consistent with Kleven et
al (2011). Further, we find that audits have a long-term effect on tax reporting. An audit
increases reported taxable income by an average of $1,185 per year, or equivalent to about
2.9% of the average taxable income. Thus the revenue gain from the audit adjustments
understates the true revenue gain (which accounts for changes in subsequent tax reporting)
by more than half when considering the five-year window after an audit.
Next, we study the persistence of the effect of the audit over time. The IRS data show
that, after an audit, reported Schedule C income increases sharply by over 16% but then falls
quickly in subsequent years. The effect of audit on reported Schedule C income in the fifth
year after audit is only about one third of that from the first year after audit. In contrast,
after an audit, reported wage income increase by only about 1% in the first year after audit,
but continues to increase in the following two years to around 2% and only gradually
declines.
We then consider the role of income volatility as another important factor in tax
evasion. To illustrate, consider that when the tax authority finds it harder to predict the
income of a plumber than the salary of a teacher, it is not only because the plumber may
have less income subject to third-party information reporting, but also because the plumber’s
income may be more volatile than that of a teacher. Such income volatility increases the rate
at which the information gleaned from an audit becomes outdated. Thus, two factors,
income volatility and third-party information, may correlate and confound previous studies
5

 that focus on the role of information reporting. We include measures of income volatility,
computed separately for across demographic groups, in our analysis and that find volatility
mitigates the persistence of the effects of audit on reported income. This empirical result
holds even when we control for third-party information reporting. The implication is that
the effects of audits on taxpaying behavior diminish more quickly for filer with higher
income volatility, holding constant the level for information reporting across those filers.
Further, we find variation in the effect of audit across filers correlated with measures of
tax literacy. While we cannot observe tax literacy directly, the results using several proxy
variables related to the filer’s understanding of the tax code are consistent with one another.
These results show that those with a better understanding of the tax code respond more
strategically to their audit experience.
The paper is organized into eight sections. Following this introduction, Section 2
describes the conceptual framework. Section 3 describes the data and Section 4 presents
information on tax compliance in our data. Section 5 presents our main empirical results,
and Section 6 reports results for particular income and deduction items and the relationship
between tax compliance and income volatility. In Section 7, we offer evidence that the
response to audit is affected to a great extent by the filer’s tax literacy. Section 8 concludes.

2. Conceptual Framework
This paper addresses the question of how individuals respond to an audit. Specifically,
we seek to understand how their taxpaying behavior changes in the years following an IRS
audit. To understand the mechanisms which guide our empirical analysis, we provide an
intuitive conceptual framework.

6

 The workhorse model for tax compliance is that of Allingham and Sandmo (1972),.
Several studies consider the impact of audit rates on individual taxpayer compliance. These
include Tauchen, Witte, Beron (1989), Dubin, Graetz, and Wilde (1990), and Witte and
Woodbury (1985), all of whom find that increases in audit rates increase compliance.
Marginal tax rates also impact evasion since they affect the value of misreported income.
Evidence of this has been provided by Clotfelter (1983) and Feinstein (1991). In addition,
Feldman and Slemrod (2007) find that documented income is much less likely to be evaded,
which has been corroborated in other studies such as Kleven et al (2011). Additional related
studies are those that show evidence of the indirect effects of audits that we are interested in.
These include Alm and Yunus (2010), who find a role for norms and learning in tax evasion
in the U.S., and Dubin (2007), who calculates the deterrent effect of audits.
One way in which we extend the Allingham and Sandmo (1972) model is to explicitly
incorporate changes in subjective expectations of audit.

To study the tax compliance

behavior of taxpayers following audit, one must incorporate changes in the expected of the
probability of audit conditional on having been audited. Conceptually, it is uncertain how
audits would affect subsequent illegal tax behavior. On the one hand, those evading tax may
update the audit risk in a Bayesian manner after experiencing an audit, and curb their
misbehavior. On the other hand, tax evaders may perceive that inspectors are unlikely to
come follow immediately and therefore increase their evasion. Given these two potential
opposing effects of an audit on the filer’s expectations of future audits, the total effect of an
audit on taxpayer behavior is ambiguous and an open empirical question.
Some interesting laboratory experiments lend some support to the second possibility.
Guala and Mittone (2005) and Mittone (2006) find that lab subjects increase tax evasion
immediately following an audit. Maciejovsky et al. (2007) also find their lab subjects to
7

 increase tax evasion following an audit, but to decrease it gradually over time to the pre-audit
level. Kastlunger et al. (2009) show that this behavior is caused mostly by misperception of
audit probability. Further, DeBacker et al. (2015) find evidence of such a hump-shaped
effect on tax aggressiveness in their examination of corporate audits.
A second important extension of the basic Allingham and Sandmo (1972) model that is
relevant to our study is the inclusion of different degrees of information reporting to the tax
authority. Kleven et al. (2011) conceptualize third-party reporting of income as affecting
taxpayers’ ability to be noncompliant, and provide theoretical and empirical evidence that the
degree of third-party reporting is a crucial determinant of the amount of taxpayer
compliance. Their results imply that responses to audit will tend to be larger for income
sources with less third-party reporting, for which the ability to be noncompliant is greater.
In reality, the tax enforcement agency has two main sources of information from which
is can try to determine the distribution of the filer’s true income. One source is from thirdparty income reporting and the other from audits. The key difference between these two
sources of information is that third-party information is updated with each filing, but audit
information is only updated when an audit is conducted. This leads to a gradual outdating of
audit information after an audit. The degree to which the amount the information from
audit diminishes over time is a function of the variability in the filer’s income. For example,
if small business income has larger year-to-year fluctuations than wage income, the
information gleaned from audit about the taxpayer’s business income becomes less
informative after a given amount of time than does the information about wage income.
Sanchez and Sobel (1993) show that in a model in which a fraction p of the population
are willing to underreport income if it maximizes their expected payoff, and where the
auditor conducts costly audits to maximize revenue subject to a budget constraint, the audit
8

 function results in cutoff rules, where certain segments of the population are audited at a
positive rate and others not at all. In such a model, a more widely dispersed income
distribution implies a lower cutoff point, with more scope for taxpayers to underreport
income than if the income distribution were more compact.
To illustrate this argument, consider Figure 1. Figure 1 shows the conditional income
distributions, as perceived by the auditor, for an income source that has a high variance. The
distribution is most highly concentrated in the year after audit as the information from the
audit is still relatively informative about the distribution of true income. However, as time
passes, the spread of the distribution increases as the information from the audit becomes
more outdated. This increase in the spread of the distribution results in more underreported
income as the reduction in information to the IRS results in more scope for taxpayer
noncompliance. Thus we expect to see reported income (denoted by Yr in Figure 1) decline
as the spread in the conditional income distribution increases in the number of years since
audit. For income sources that have less variance or volatility (or that are subject to more
third-party reporting) the degree to which audit information becomes outdated is much
lower. Thus we should expect a strong effect of audit on income sources with high variance
and little third-party reporting (where the information gleaned from audit is particularly
important), but that this effect declines more quickly than for sources of income with more
third-party reporting and less variance or volatility.
[Figure 1 about here]
To summarize, our conceptual framework implies three propositions we can examine in
the data. First, the direction of the effect of an audit on subsequent income reporting is
ambiguous. Second, to the extent that income reporting responds to audit, we will observe
more of a response among income sources that have less third-party reporting. Third, the
9

 responses to audit will be stronger and more transitory for those sources of income with less
third-party reporting and more variance or volatility.

3. Data
Our data come from three sources. We discuss each data source in turn and then the
process by which we merge the data and create our final sample.
First, we use data on audits from the IRS’s National Research Program (NRP).
Specifically, we use the taxpayer information generated by the audits conducted as part of
the NRP’s 2006, 2008, and 2009 waves.4 Taxpayer information includes taxpayer identifiers
(the social security number (SSN) of the primary filer), year of the audited return, and the
resulting adjustment to the tax return by line on the Form 1040. The NRP conducts audits
on a stratified, random sample of the filing population and includes in their sampling weights
to allow researchers to create population-representative statistics. Each of the 2006-2009
waves has approximately 15,000 observations.
Second, we use data from the IRS’s population of individual income tax returns.
These data include many items from the filer’s Form 1040 and the associated forms and
schedules, including all items on the front page of Form 1040 and the main line items from
most associated schedules. We use these data from the years 2000 to 2012.

Note we exclude the NRP waves from 2001 and 2007 and those conducted after 2009. Documentation
suggests that the sampling frame and intent of the 2001 wave was sufficiently different from later waves to treat
them as comparable. Further, constructing the control group for the 2007 NRP wave was complicated by the
fact that stimulus rebate checks were sent out in 2008, and that to be eligible for a stimulus check one must
have filed a year 2007 tax return. The resulting population of filers for tax year 2007 (who filed taxes in early
2008) was much different than in other years and did not closely resemble the weighted NRP sample. In
particular, there was an increase in the number of people who typically did not file a tax return. We attempted
to address this by using the methodology of Ramnath and Tong (2014) to identify those who filed only to claim
the stimulus check, but the random sample was still sufficiently different than the NRP sample. NRP waves
later than 2009 have been excluded to ensure that open audits were able to be completed before we drew our
data.
4

10

 Finally, we use data from the IRS’s Audit Information Management System (AIMS).
The AIMS data contain detailed information on all IRS audits (including NRP and non-NRP
audits) from 1996 to present. We use these data to augment the audit data from the NRP.
In particular, the AIMS data allow us to observe variables such as the date the audit began
and ended, the hours of examiner time put towards the audit, and examiner characteristics.
We construct our sample in the following way. We create a control group by
randomly selecting a 0.1% sample of filers by choosing a different set of 10 four-digit SSN
endings for each year 2006, 2008, and 2009.5 For each of these years, we then select all
primary filers who had one of these 10 four-digit endings from the universe of returns filed
that year and which met the general sampling criteria of the NRP. We create our treatment
group by finding the SSN of all primary filers in NRP waves from 2006, 2008, and 2009.
Finally, we pull all tax returns from the 2000-2012 period for filers in either our sample or
control groups6. Our final panel is thus comprised of a control group of randomly selected
filers from the years of the NRP waves (followed over time) and a treatment group of
randomly audited filers from the NRP waves (who are also followed over time). Creating
our control group in this way (by ensuring that that those in the control group filed a return
in the year the treatment group was audited) allows us to match attrition rates across
treatment and control groups.
Using the SSN of the primary filer, we are able to link returns across the three data
sources, the population files, the NRP, and the AIMS. Thus in our final panel, we have
detailed information on each tax return filed from 2000-2012. For the treatment group, we
also have detailed information on the characteristics of the audit and the adjustments to tax

The sample size is dictated by computational constraints.
We also restrict these pre- and post- audit observations to those which would be consistent with the NRP
sampling criteria.
5
6

11

 returns following audit, though we lack information on audits that are not closed by the time
we pull data from the AIMS database. As such, information from audits not closed by
October 2014 is missing in our sample. However, given that our last NRP wave is from
2009 and that well over 95% of audits are closed within two years, almost all audits have
been closed. Table 1 summarizes our sample, noting weighted observations in the base year
(i.e., NRP wave year) and across all years 2000-2012.7
[Table 1 about here]
While we do observe the date an audit was opened and closed, we do not know
when the filer was notified of the audit or the results of the audit. Thus we use as our timing
convention the number of years since the audited return was filed. For example, for the
2006 NRP wave, their tax year 2006 return was audited. Thus we consider their tax year
2007 return as being one year since the audited return was filed. We use this convention
throughout the paper. As a result, one would not expect a sharp increases in reported
income for all filers in a given NRP wave in a specific year since audit, since the duration of
audits and the time when filers were notified varies. However, since the vast majority of
audits are closed within two years, we do expect the effects of audits to fully materialize two
to three years after the audited tax year.
Throughout, all monetary variables are deflated to 2005$ and 99% Winsorized.
Winsorization of the data is necessary for dealing with outliers. The IRS does not edit most
of our data sources, and thus data entry and calculation errors by the filers or the IRS agent
entering the data are not uncommon. Winsorizing data inevitably removes some genuine

We use weights for both our randomly sampled control group and the treatment group. We weight the
control groups by giving each filer equal weight to sum to the total population of filers in the base year. We
weight the treatment groups using the NRP sampling weights. This gives us a number of weighted observations
approximately equal to the population of filer in the base year for the NRP sample. We then apply these
weights to the filing units for each year they are in the panel.
7

12

 variation from the sample; however, we think that Winsorization is acceptable to ensure that
extreme outliers from coding errors do not influence our results.

Further, the results

reported in this paper are robust to different levels of Winsorization.8

4. Tax Compliance in the U.S.
The IRS’s tax gap measure summarizes aggregate compliance with the U.S. income tax.9 In
2006, the last year for which the IRS reports the tax gap, the net gap was $385 billion
(Internal Revenue Service (2012)). This represents a compliance rate of 85.5%.

Non-

compliance with the individual income tax code represents the largest source of
noncompliance, accounting for $235 billion of the $385 billion gap.
Within the individual income tax, Internal Revenue Service (2012) shows that the
lowest compliance rates come from income with less documentation. For example, the
underreporting of business income, and in particular income from sole proprietors (as
reported on Schedule C of Form 1040), accounts for about half of the individual income tax
gap ($122 of $235 billion). Looking across income and deduction items, one can see the
pattern that compliance rates fall as withholding and third-party verification decline. Such a
pattern is also documented in Danish data by Kleven et al. (2011).
Our NRP data allow us to delve more deeply into the data than the tax gap statistics
provided by IRS. Table 2 documents the measures of compliance found in our NRP data.
Column 1 reports means by income and deduction sources and the fractions with those
sources of income. Columns 2-4 report audit adjustments. Column 2 shows the average
audit adjustment by income/deduction item and the fraction of those who report non-zero
values of that item for which there is a non-zero adjustment. Column 3 and 4 decompose
8
9

Results with different levels of Winsorization are available from the authors by request.
Note that the NRP data we use plays a large role in the IRS’s estimation of the individual income tax gap.

13

 the adjustments in Column 2 into underreporting of income (which results in upward
adjustments in income/downward adjustments in deductions) and over reporting of income
(which results in downward adjustments of income/upward adjustment of deductions).10
[Table 2 about here]
Consistent with the compliance results seen the aggregate IRS tax gap reports the
pattern that emerges is that those sources of income with the most documentation show the
lowest rates of tax non-compliance. We find that non-compliance rates (measured by the
fraction of filers with adjustments) are largest for Schedule C income, which has no
withholding and little third-party verification. Schedule C income is adjusted for about 73%
of those filers who are audited. The rate of non-compliance is lowest for wage and salary
income, which is adjusted for about 6.5% of audited filers. Underreported income is more
frequent than over reporting of income for all sources and is highest for Schedule C income.
The average amount of underreported Schedule C income is $8,483. This compares to a
mean of $8,401 for reported Schedule C income. Wage income is underreported by $1,233
compared to an average of $42,657. Overall, compliance rates are highest for wage income,
which has a high withholding rate. Compliance rates are also high for capital income (both
capital gains reported on Schedule E and capital income reported elsewhere), which has
third-party reporting on most items, but not withholding.

5. Effects of audits on subsequent reported income
With an understanding of our data and tax compliance in the U.S., we now turn to our
research question.

10

The objective of the paper is to understand changes in individual

Note that Column 2 is the weighted sum Columns 3 and 4.

14

 taxpaying behavior in response to audit. To ensure the robustness of the results, we will use
different methods, which include post-treatment difference, difference-in-difference and
within-filer estimations.

5.1 Post-treatment difference and difference-in-difference estimates
The randomized controlled trial nature of the NRP allows us to consider the effects of audit
on taxpaying behavior using either a simple post-audit difference or difference-in-differences
estimator. For outcome variables, we will look at the reporting of taxable income, wage
income, and Schedule C income. These three measures provide a nice contrast in terms of
the amount of third-party information the IRS has regarding each income source. For
example, most wages are subject to withholding, Schedule C income has very little
documentation, and taxable income is a broad measure of overall income, composed of
income with different reporting requirements and determined after deductions are reported.
Our difference-in-differences estimator of the effect of audit is thus:
!!!"#$!!"!!"#$% = !!,! − !!,! − !!,! − !!,! ,

(1)

where B denotes the treatment group (i.e., the NRP sample) and A the control group. The
subscripts 1 and 2 denote the pre-audit and post-audit periods respectively. For each, we
consider the mean over a span of 3 years. Thus the !!,! is calculated as the mean of the
income source of interest for the NRP sample over the three years after audit and !!,! is
calculated as the mean of the income source of interest for the NRP sample over the three
years prior to audit. The means for the control group are constructed in an analogous way.
[Table 3 about here]
Table 3 reports the post-treatment and difference-in-differences results. We present
the results in percentage terms because the income sources have very different mean
amounts. The top panel of the table shows that (reported) taxable income of the audited
15

 group increases by 3.8% when comparing the post-audit period to the pre-audit period,
while that of the control group increases only 0.9%. These figures imply that audits increase
taxable income by 2.9% on average ($1,185 per year). In the second panel of Table 3,
however, the effect of audits on reported wage income is only 1.3% on average ($590 per
year). This effect is consistent with the idea that it is more difficult to misreport income that
is also subject to withholding and third-party verification (Kleven et al. (2011)). The largest
effect of audits, in percentage terms, is on Schedule C income. The third panel in Table 3
shows this effect to be 14.2% ($1,156 per year). This large effect supports the view that it’s
relatively easier to manipulate Schedule C income than wage income, and that such income is
more responsive after an audit.

5.2 Persistence of audit effect
Given these results, a question of further interest is how individuals change tax
reporting over time after audit: do they increase reported income permanently, or does the
initial effect decline as time passes? Figure 2 plots the differences between the mean reported
incomes of the audited and control group. Reported taxable income increases in the first and
second years and after an audit and remains elevated even after six years. Adjusted gross
income (AGI) and wage income follow a similar pattern as taxable income.
As Figure 2 shows, the effect of audits on Schedule C income are strong in the first
couple of years after audit. Following the initial upswing in reported Schedule C income, it
then turns downward toward the pre-audit level. This result stands in contrast to the trends
in taxable income, AGI, and wage income, and suggests further that income with less-thirdparty reporting responds differently to audit. We thus delve more deeply into the varying
responses by type of income in Section 6.

16

 For each of the four income sources considers, it is apparent from the figure that the
pre-audit trends are similar across the NRP and non-NRP samples. Thus the common
trends assumption needed for identification is satisfied. We next consider models with
individual fixed effects, which will allow for the identification of the effects of audit from
within filer variation in reported income.
[Figure 2 about here]

5.3 Within-filer estimates
Because we have a panel of tax returns, we can examine changes in individuals’
behavior after an audit while controlling for time-invariant unobserved individual
characteristics. We first estimate an equation of the form
!"#$%&!" = !"#$%&'()%!" + !! + !! + !!" ,

(2)!

where !"#$%&!" denotes a measure of income for individual (taxpayer) i in year t; and
PostAuditit denotes that the individual was audited during our sample period prior to year t, !!
denotes an individual (taxpayer) fixed effect, and !! denotes a year fixed effect. In this
specification, identification of the effects of audit come from within filer changes in reported
income between the pre and post audit periods, net of trends in income common across the
treatment and control groups, which are accounted for by the year fixed effects.
[Table 4 about here]
Table 4 reports the results from regressions that estimate the effect of audits on
taxable income. Column 1 shows that audits increase reported taxable income by $1,109,
which is statistically significant at the 1% level. This result implies that, consistent with the
simple difference-in-differences tabulations above, individuals tend to report more income
after audit.
17

 Column 2 examines whether, in our estimation framework, the effects of audit differ
with the number of years since the audited tax year by estimating an equation of the form:
!"#$%&!" =

!
!!! !!

(!"#$%&'($!" ) ∗ (!!!"#$%!!"#$%!!"#$%) + !! + !! + !!" (3)

In this specification, the key explanatory variables are a series of dummy variables that show
the difference between the audited and control group from Year 1 through (at most) Year 6
after the audited tax year. This column shows that reported taxable income increases quickly
during the first two years after the audit, with an increase of around $1,200, and stays at this
level until at least Year 6, with all of the effects being statistically significant.

Figure 3

summarizes a further specification where we include dummies for the two years prior to
audit. Neither of these pre-audit dummies are significant.
[Figure 3 about here]
Because of our data, we are able to observe the adjustment to each filer’s return.
One might suppose that those filers for whom there is a positive adjustment to tax liability,
the response to audit is stronger than those for whom there was not adjustment. That is
indeed that case. We provide evidence for this in Figure 4, where we estimate Equation 3 on
taxable income separately for each of three groups: those with a positive adjustment to tax
liability following the audit, those with no adjustment, and those with a negative adjustment.
We see the strongest and most statistically significant response from the group with positive
adjustments.

Almost all the coefficients for the other two groups are statistically

insignificant and have much lower point estimates than for the positive adjustment group.
Thus, while we find that the non-compliant increase compliance after audit, we do not find
that the compliant increase non-compliance.

These results are consistent with what

Gemmell and Ratto (2012) using UK data.
[Figure 4 about here]
18

 6. Effect under third-party information and income volatility
We now examine whether the effects of an audit differ with the level of third-party reporting
by estimating separate impacts by the source of income or the type of deduction that is being
claimed. Then we examine whether the response differs for taxpayers with more or less
volatile income.

6.1 By Income Source
IRS data allow us to distinguish incomes from wages, Schedule C, D and E. Among those,
wage income information can be easily cross-checked with data reported by employers.
Information on incomes from Schedule C, D and E are harder to verify as little third-party
information is available. We will examine the audit effect on incomes from these sources by
both intensive and extensive margins.

6.1.1 Intensive Margin
In Table 5 (and Figure 5), we present estimates versions of Equation 3, where the
dependent variable is income of a particular type.

Note that Figure 5 presents the

coefficients for the post-audit year indicator variables in percentage terms to make the
responses more comparable across income sources with very different mean values. We also
restrict our sample to those who have non-zero amounts of income of that particular type in
the year of the audit, so these can be considered intensive-margin results.
[Table 5 about here]
[Figure 5 about here]

19

 In Column 1, we repeat the results from Column 2 in Table 4, in which the
dependent variable is taxable income. Column 2 presents results for total income.11 The
results in this column are slightly smaller than, though similar to, those for taxable income.
This is not surprising due to the fact that taxable income also incorporates the filer’s choice
of deductions and so provides for more opportunity to manipulate taxes through reporting.
In Column 3, the dependent variable is wages and salaries. Similar to the results
found above in the simple tabulations, this specification finds a small positive effect of an
audit on reported wages, with an increase of $330-530 in the first three years after audits.
However, the impact on wages dies out thereafter.
Column 4 presents results when the dependent variable is Schedule C (sole
proprietorship) income. As noted above, this source of income is not generally subject to
third-party reporting, and so may be easier for taxpayers to manipulate. Consistent with this,
the estimation results suggest that Schedule C income increases substantially after audit, by
over $1,000 in the first two years. Interestingly, that effect diminishes 3 years after the audit,
and is insignificant after four years. Further, in years 5 and 6, the estimated impact is actually
negative. These results suggest that taxpayers with sole proprietorship income may be more
careful in reporting income right after an audit, but may become more aggressive than they
were in the years prior to the audit over time.
In Column 5, which presents results for Schedule D income (capital gains and
losses), no significant effects of audits are found in any year. However, the results in
Column 6 for Schedule E income, which includes partnership, S corporation, and rental
income, mirror that for Schedule C income. Schedule E income, like Schedule C income, is
largely self-reported. A significant positive impact of audits is found in the first two years,
11

Total income is from Form 1040, Line 22. It is AGI with the above the line deductions added back in.

20

 with the effect diminishing in the third year, and no longer significant in the fourth year. In
years 5 and 6, the estimates turn negative, but (unlike for Schedule C income) are not
statistically significant in those years.

6.1.2 Extensive margin effects
To understand the extensive margin effects of audit, we estimate linear probability
models of the form:
!(! ≠ 0)!" =

!
!!! !!

(!"#$%&'($!" ) ∗ (!!!"#$%!!"#$%!!"#$%) + !! + !! + !!" , (4)

where !(! ≠ 0)!" is an indicator function equal to one if ! is not zero. The variable !
represents income from Schedule C, Schedule E, or wages and salaries. The estimates of
Equation 4 are reported in Figure 6, Panels A-C. Each panel reports results for three
groups, those who filed the schedule or, those who did not, and the full sample. The graphs
then plot the change in the likelihood of filing the given form by year since the audited
return was filed.
The extensive margin results for Schedule C and E filers are reported in Figure 5,
Panels A and B. These two graphs show sharp declines in the likelihood that a filer
continues to file the relevant schedule after audit. Thus, the increases in taxable income
following audit is only one effect of the audits on those with business income. The other
effect is to make them less likely to claim business income. There are at least three possible
reasons for this decline, though our data do not allow us definitively say how many filers are
affected by each of the three possibilities. First, audits may result in increases in reported
income and thus taxes. This pushes down the after-tax return on the business endeavors
and may cause the filers to forgo them. Second, the Schedule C or E may have been filed
for an activity that generated losses, which were used offset ordinary income elsewhere on
the return. An audit may have found such a losses to be illegitimate and thus the filer
21

 discontinued their use of those losses and the filing of the associated schedule. Finally, and
this is particularly relevant for Schedule C income, an audit may have found that the filer
should not have been filing as an independent contractor, but instead should have been
classified as an employee. The increase in compliance after an audit in this case would result
in fewer Schedule Cs being filed. Some evidence of this effect can be seen in Panel C. This
is the extensive margin effect of audits on filers’ wage income. For those who do not report
wage income in the year of audit, there is a strong increase in the probability they report
wage income in subsequent years, which mirrors the results for the reporting of Schedule C
or Schedule E income.
[Figure 6 about here]
Taken together, the intensive and extensive margin results are consistent with those of
Kleven et al. (2011) who point out that taxpayers’ compliance is strongly related to the ability
to be noncompliant, as this ability is greatest with self-reported income such as that reported
on Schedules C and E.

6.2 By Deduction Item
We next estimate variants of Equation 3 in which the dependent variables denote the
amounts of particular types of deductions that filers claim. Column 1 of Table 6 presents
the results for above the line deductions (i.e., deductions that figure into AGI) and Column 2
the results for total itemized deductions. Columns 3-5 present results for individual types of
itemized deductions (charitable contributions, state and local income taxes, and mortgage
interest). We also summarize the results in Figure 7, which plots the percentage changes in
the deduction items following audit. Columns 1 and 2 show that both types of deductions
decrease after an audit (which implies higher taxable income), with a larger decline for
itemized deductions. For both types of deductions, the effects continue at a high level for up
22

 to six years after the audited return was filed. As with the income source results, the results
for deductions are consistent with the idea that taxpayers manipulate their tax reporting
where they are able to do so. Thus we see larger effects of audit on the more malleable, and
less documented, itemized deductions category. Charitable contributions are estimated to
fall by around $270-460 after an audit, with the size of the impact being consistent across
years. Interestingly, although they are both subject to third-party reporting, state and local
taxes and mortgage interest are also estimated to fall after an audit, with the effect increasing
over time.
[Table 6 about here]
[Figure 7 about here]

6.3 By degree of volatility
The result that income sources such as Schedule C and Schedule E income have the largest
responses to audit is not surprising given the limited amount of third-party reporting on
these income sources. However, the transitory nature of the impact of audit on these income
sources and on the taxable income of those deriving income from these sources is not as
intuitive.
In addition to being subject to less third-party reporting, business income reported
on Schedules C and E differs from labor income (which makes up the majority of total
income) by its volatility. DeBacker, Panousi, and Ramnath (2014) document that business
income exhibits both a higher variance and more volatility than labor income. We believe
that the difference in income processes interacts with audits in an interesting way. In
particular, we posit that higher income volatility allows taxpayers to change reported income
from year to year more easily. The result is that the effects of audit are not as persistent for
business income as for labor income. For example, consider an audit f a filer with only wage
23

 income. Since wage income is less volatile, that filer may believe that the IRS can infer that
their wage income a given number of years in the future is within some range with a fair
degree of accuracy. Thus the effect of audit on wage income will be smaller, and potentially
more persistent. However, for the more volatile Schedule C income, the filer may believe
that the same number of years from the audit, the IRS has very little ability to predict what
the range of true Schedule C income is. Therefore the effect of audit on Schedule C income
is more transitory.
To test this hypothesis we include a measure of volatility into our models and
interact volatility with the audit indicator variable. We measure volatility by considering the
variance in each individual filer’s taxable income over time. We then group filers by filing
status, number of dependents, and schedules filed. Finally, we take averages over these groups
and use that as the measure of volatility for each individual in that group.

Table 7 presents the results of this model, separately for all filers and for those with
Schedule C income. In all cases, the interaction terms are negative, indicating that volatility
reduces the effect of audits on subsequent taxpaying behavior. This is consistent with our
hypothesis. Furthermore, the interaction terms show that the effect of volatility becomes
more negative over time, which means the effect of an audit is more short-lived for filers
with more volatile income.
[Table 7 about here]

7. The implications of tax literacy
As noted above, an NRP audit is random, and the letters NRP subjects receive
explicitly mention that they were selected for a random audit used for IRS research

24

 purposes. Despite this, we find significant impacts of audit on filers. One reason for this
may be that filers misunderstand the nature of the NRP audit. To test this, we look at the
responses to audit across several subsamples of the tax filing population. Each of these
subsamples is created based on characteristics that may be correlated with tax literacy or
understanding of tax enforcement.

These include income, age, the use of a paid tax

preparer, and the use of e-filing of the tax return. Although none of these characteristics is a
perfect proxy for tax literacy, consistent responses to audits across all these subsamples
would suggest that tax literacy may play a role in how filers respond to audits.

7.1 Responses by Filer Age
Figure 8 splits the sample according to the age of the primary filer in the year of the
audit.12 Here, the three oldest age groups (35-44, 45-54, and 55-64) appear to be the most
responsive after an audit, with estimated effects two years after audit in excess of $1,700,
while the impact for the youngest group two to four years after audit is around $750. These
results are in contrast to Kleven, et al. (2011) who find that the propensity to underreport
income falls with age.
[Figure 8 about here]
Considering the long-term patterns, we see that the youngest age group (25-34)
responds significantly differently than the older age groups to audit. In particular, this age
group does not show a return to the pre-audit trend in income reporting. Following an
audit, those aged 25-34 report higher incomes and this effect persists over the next six years
with increases over the entire window. In contrast, the older age groups show an increase in
reported taxable income in years two and three after the audit, but a decline after that. Thus
For observations in the control group, we split based on age in the tax year for which they were drawn as a
control observation.
12

25

 we find evidence that audits have more persistent deterrent effects on those who are
relatively new tax filers. This may reflect differences in how those with a short filing history
update there prior beliefs about IRS enforcement following an NRP audit as compared to
those with longer filing histories and more experience with IRS enforcement.

7.2 Responses by Filer Income
Next, we show how the responses to audit differ across income groups. These
results are shown in Figure 9. Income quintiles 1-4 show similar responses to audit, with
persistent increases in reported taxable income following audit. Amongst these, the lowest
income group has the strongest response to audit measured by the change in taxable income
reported after audit. Note that this is true despite this group having the lowest taxable
income at the time of audit. The top income quintile shows the opposite trend, with a
persistent decline in reported income after the initial increase in income in the year after the
audited return was filed. That those in the highest income quintile have such different
responses from those in the lower income quintiles is consistent with these filers being more
sophisticated and having a better understanding of the NRP audit process.
[Figure 9 about here]

7.3 Responses by Filing Method
We now condition samples on filing method and compare the responses to audit.
By filing method, we mean the use of a paid preparer to file the tax return and the use of efiling of a tax return. E-filing is highly correlated with the use of tax filing software. Our
assumption is that those using a paid preparer or e-filing tend to be more sophisticated in
their understanding of the tax code than those without, at least conditional on the personal
or software assistance they receive in filing.
26

 Figure 10, Panel A displays the results comparing those who used paid-preparers in
the year the audited return was filed to those who did not. In the first three years after audit,
these two groups have very similar responses, increasing reported taxable income by over
$1,000 more than the comparable group who was not audited. However, after year three the
post-audit trends diverge. In these years, the trend for those using a paid prepare moves
taxable income towards its pre-audit level. While for those who did not use a paid preparer,
the increase in reported taxable income is persistent.
[Figure 10 about here]
In Figure 10, Panel B, we see that a similar comparison can be made between those
who used e-filing in the year the audited return was filed and those who did not. In
particular, the changes in taxable income following audit are greater and more persistent for
those who did not e-file.
A word of caution about these results is in order. Over our sample period, there was
a sharp increase in the number of e-filed returns. Thus, those who e-file in 2006 (from
whom the effect 6 years from audit is identified) may be a very different group of filers than
those who e-file later in the sample period. Fixed effects do control for level differences in
income between these groups, but the composition of the groups may also differ on
unobservables. Differences in unobservables that lead to heterogeneous audit treatment
effects may skew the estimated effects of audit over time.

7.4 Summary of findings related to tax literacy
Although there exists no direct measure of tax literacy for the filers in our sample, in
each case above, the results are consistent with smaller, and more transitory, responses from
those with a better understanding of the audit process they are involved in: the young and
27

 inexperience respond more strongly and persistently than the old, those with low income
respond to a greater extent than those with high income, those using paid-preparers have
more fleeting responses than those who don’t, and those who e-file have less of the long-run
response than those who do not. While the proxies for tax literacy are imperfect, the
robustness of these results should suggest that there is a relationship between tax literacy and
the responses to audits.

8. Conclusion
Tax evasion is estimated lower global tax collections by about 18% and result in
about three trillions dollars in lost public funding worldwide (Murphy 2011). Among all
countries, the U.S. suffers the largest loss in absolute terms because of the sheer size of the
economy. One way to deal with this problem is through audits of tax returns. Auditing
affects tax compliance through deterrence effects, which impact all filers. Auditing also has
more direct effects on the audited. First, there is a static revenue gain when the auditors
discover noncompliance. Second, those audited tend to report higher taxable income in
subsequent years, resulting in further revenue gains. This study provides a rigorous
evaluation of the effectiveness of tax audits in reducing tax evasion, in the short and long
term.
We examine the impact of audits on subsequent individual taxpaying behavior using
data from the Internal Revenue Service (IRS) National Research Program on random audits
on tax returns from 2006-2009 matched to returns from the universe of filers from 2000 to
2012. It is worth noting that the effects we measure of the effects of a random audit. Filers
may respond differently to non-random audit; for instance, they can make some inferences

28

 about the auditing process from the fact that they were selected. However, because of this
selection, the effects of non-random audits are more difficult to measure. One argument
that suggests the effects of non-random and random audits may not be far apart is made by
Manoli and Turner (2014). They provided evidence from a randomized field experiment
showing that the content of contact between the IRS and tax filers is much less important
than the existence of this contact.
The results from a simple difference-in-differences specification indicate that an audit
increases reported taxable income by over $1,100 per year, or equivalent to 2.9% of the
average income. This effect is only 1.3% for wage income but is 14.2% for Schedule C
income. Further, we find that the impact of auditing on reported wage lasts over time while
it is fleeting for Schedule C income.
Similar results are found when controlling for individual fixed effects. These results
suggest that Adjusted Gross Income increases for at least 6 years after an audit.
Contributing to this increase, Schedule C and Schedule E income (which are not subject to
third-party reporting or withholding) tends to sharply increase after an audit, but this
increase diminishes (and turns negative) 5 or more years after audit, while the increase in
wage and salary (which is subject to third-party reporting and withholding) is considerably
smaller. In addition, above the line and itemized deductions both decrease significantly after
audit, and the decrease in deductions is apparent even among deductions (like state and local
taxes and mortgage interest) that are subject to third-party reporting.
There are several clear implications one can draw from the results of this study. First,
an audit of a randomly-selected individual tax filer increases reported taxable income by an
average of about $1,185, and this effect appears to persist for at least six years. Second,
audits produce more transitory effects on income sources that are subject to less third-party
29

 reporting. Third, income volatility provides an important shield for those willing to evade
taxes. Fourth, responses to audits are affected by the tax literacy of the filer. We believe that
extending this research to encompass both the direct and indirect effects of audit is an
important topic for future work.

30

 References:
Advani, Arun, Jonathan Shaw, and William Elming. 2015. "How long-lasting are the effects
of audits?" Working Paper, Tax Administration Research Center, UK.
Allingham, Michael G and Agnar Sandmo. 1972. "Income Tax Evasion: A Theoretical
Analysis." Journal of Public Economics, 1(3), 323-38.
Alm, James and Mohammad Yunus. 2009. "Spatiality and Persistence in Us Individual
Income Tax Compliance." National Tax Journal, 101-24.
Alm, James, and Michael McKee. 2004. "Tax compliance as a coordination game." Journal of
Economic Behavior & Organization, 54(3) 297-312.
Alm, James, Gary H. McClelland, and William D. Schulze. 1992. "Why do people pay
taxes?." Journal of Public Economics 48(1): 21-38.
Alm, James. 2010. "Testing Behavioral Public Economics Theories in the Laboratory."
National Tax Journal, 63(4), 635-58.
Andreoni, James, Brian Erard, and Jonathan Feinstein. "Tax compliance." Journal of economic
literature (1998): 818-860.
Becker, Gary S. 1968. "Crime and Punishment: An Economic Approach," Journal of Political
Economy, 76, 169-217.
Choo, Lawrence, Miguel A. Fonseca, and Gareth Myles. 2013. "Lab Experiment to
Investigate Tax Compliance: Audit Strategies and Messaging." HM Revenue and Customs
Research.
Clotfelter, Charles T. 1983. "Tax Evasion and Tax Rates: An Analysis of Individual
Returns." The review of economics and statistics, 363-73.
DeBacker, Jason Matthew; Vasia Panousi, and Shanthi Ramnath, 2014. “A Risky Venture:
Income Dynamics within the Non-Corporate Private Business Sector”, mimeo.
DeBacker, Jason, Bradley T. Heim, Anh Tran and Alexander Yuskavage, “The Impact of
Legal Enforcement: An Analysis of Corporate Tax Aggressiveness after an Audit,”
forthcoming, Journal of Law and Economics
31

 Dubin, Jeffrey A. 2007. "Criminal Investigation Enforcement Activities and Taxpayer
Noncompliance." Public Finance Review, 35(4), 500-29.
Dubin, Jeffrey A; Michael J Graetz and Louis L Wilde. 1990. "The Effect of Audit Rates on
the Federal Individual Income Tax, 1977-1986." National Tax Journal, 395-409.
Feinstein, Jonathan S. 1991. "An Econometric Analysis of Income Tax Evasion and Its
Detection." The RAND Journal of Economics, 14-35.
Feld, Lars P. and Bruno S. Frey. 2002. “Trust Breeds Trust: How Taxpayers Are Treated.”
Economics of Governance 3, 87-99.
Feldman, Naomi E and Joel Slemrod. 2007. "Estimating Tax Noncompliance with Evidence
from Unaudited Tax Returns." The Economic Journal, 117(518), 327-52.
Fortin, Bernard, Guy Lacroix, and Marie-Claire Villeval. 2007. "Tax evasion and social
interactions." Journal of Public Economics, 91(11): 2089-2112.
Gemmell, Norman and Marisa Ratto. 2012. "Behavioral Responses to Taxpayer Audits:
Evidence from Random Taxpayer Inquiries." National Tax Journal, 65(1), 33-58.
Guala, Francesco and Luigi Mittone. 2005. "Experiments in Economics: External Validity
and the Robustness of Phenomena." Journal of Economic Methodology, 12(4), 495-515.
Internal Revenue Service. 2012. "IRS Releases New Tax Gap Estimates."
Kastlunger, Barbara; Erich Kirchler; Luigi Mittone and Julia Pitters. 2009. "Sequences of
Audits, Tax Compliance, and Taxpaying Strategies." Journal of Economic Psychology, 30(3), 40518.
Kleven, Henrik Jacobsen; Martin B Knudsen; Claus Thustrup Kreiner; Søren Pedersen and
Emmanuel Saez. 2011. "Unwilling or Unable to Cheat? Evidence from a Tax Audit
Experiment in Denmark." Econometrica, 79(3), 651-92.
Maciejovsky, Boris; Erich Kirchler and Herbert Schwarzenberger. 2007. "Misperception of
Chance and Loss Repair: On the Dynamics of Tax Compliance." Journal of Economic Psychology,
28(6), 678-91.

32

 Manoli, Dayanand S. and Nicholas Turner, 2014. “Nudges and Learning: Evidence From
Informational Interventions for Low-income Taxpayers.” National Bureau of Economic
Research, NBER Working Paper 20718.
Mittone, Luigi. 2006. "Dynamic Behaviour in Tax Evasion: An Experimental Approach."
The Journal of Socio-Economics, 35(5), 813-35.
Sanchez, Isabel and Joel Sobel, “Hierarchical design and enforcement of income tax
policies,” Journal of Public Economics, March 1993, 50 (3), 345–369.
Tauchen, Helen V; Ann Dryden Witte and Kurt J Beron. 1993. "Tax Compliance: An
Investigation Using Individual Taxpayer Compliance Measurement Program (TCMP) Data."
Journal of quantitative criminology, 9(2), 177-202.
TRAC Reports, Inc. 2014. "Graphical Highlights of the Internal Revenue Service."
Witte, Ann D and Diane F Woodbury. 1985. "The Effect of Tax Laws and Tax
Administration on Tax Compliance: The Case of the Us Individual Income Tax." National
Tax Journal, 1-13.

33

 Table 1. Number of observations
Total observations
NRP Sample
All Years
Base Year
Postive Adj to Tax Liability
Negative Adj to Tax Liability
Random Sample
All Years
Base Year

4,364,655,406
404,252,738
164,338,287
29,778,483
4,335,270,000
405,187,000

Note: Observations are weighted; 2006, 2008, 2009
NRP waves

 Table 2. Summary of adjustments
Pre-audit Income Audit Adjustment Underreported Income Overreported Income
Taxable Income
$40,735
Fraction Non-zero
0.764
Adjusted Gross Income
$48,417
Fraction Non-zero
0.997
Deductions
$2,420
Fraction Non-zero
0.252
Wages and Salaries
$42,657
Fraction Non-zero
0.843
Sch C Income
$8,400
Fraction Non-zero
0.155
Sch D Income
$3,066
Fraction Non-zero
0.170
Sch E Income
$11,309
Fraction Non-zero
0.117
Note: NRP Audits, 2006, 2008, 2009 Waves

$4,031
0.598
$4,502
0.434
$58
0.471
$921
0.065
$7,116
0.733
$1,721
0.161
$4,512
0.482

$5,167
0.828
$5,893
0.812
$726
0.604
$1,233
0.774
$8,483
0.861
$3,133
0.649
$6,614
0.761

-$1,427
0.172
-$1,512
0.188
-$961
0.396
-$148
0.226
-$1,317
0.139
-$886
0.351
-$2,186
0.239

 Table 3. Double and triple difference with percentage changes
Column1

Taxable Income
Pre-audit
Post-audit
% Diff

NRP Sample Non-NRP, Random Sample % Difference
41,144.14
42,725.13
0.038

40,879.17
41,262.21
0.009

0.006
0.035
0.029

43,951.25
44,967.15
0.023

43,787.59
44,206.48
0.010

0.004
0.017
0.013

8,016.93
8,290.09
0.034

8,111.85
7,331.30
-0.096

-0.012
0.131
0.142

Wage Income

Pre-audit
Post-audit
% Diff
Sch C Income

Pre-audit
Post-audit
% Diff

Note: Pool all waves 2006, 2007, 2009. Calculate means over three years
before and three years after audit (to define pre and post-audit period)

 Table 4: Effect of audit on reported taxable income
(1)
Post-Audit
1 Year Post Audit

1109.329***
(184.312)

(2)

!
!

687.422***
(186.026)
2 Year Post Audit
1203.155***
(201.095)
3 Year Post Audit
1424.118***
(246.485)
4 Year Post Audit
1480.547***
(328.235)
5 Year Post Audit
1122.686***
(434.124)
6 Year Post Audit
1190.270**
(515.569)
Individual FE
yes
yes
Year of Tax Return yes
yes
Constant
34098.286*** 34098.639***
(137.019)
(136.963)
R-squared
0.005
0.005
N
4771427
4771427
Notes: * p<0.10, ** p<0.05, *** p<0.01,

 Table 5: Effect of auditing on different sources of income, for the pooled sample
Taxable IncomeTotal Income Wage Income Schedule C
1 Year Post Audit

792.761***
606.842***
(222.921)
(202.347)
2 Year Post Audit
938.812***
1036.045***
(239.565)
(222.467)
3 Year Post Audit
1015.539*** 1280.403***
(293.897)
(277.113)
4 Year Post Audit
803.839**
1363.217***
(390.969)
(367.693)
5 Year Post Audit
-83.959
939.992*
(511.201)
(500.293)
6 Year Post Audit
-147.849
1155.673*
(602.338)
(591.324)
Individual FE
yes
yes
Year of Tax Return yes
yes
Constant
39220.724*** 50411.774***
(151.694)
(152.286)
R-squared
0.009
0.008
N
3840937
4763014
Notes: * p<0.10, ** p<0.05, *** p<0.01,

331.796*
(181.705)
452.527**
(220.212)
522.173**
(262.200)
103.304
(355.860)
-43.082
(505.031)
3.137
(588.270)
yes
yes
41058.233***
(128.911)
0.009
3996502

1172.550***
(129.573)
1035.716***
(146.667)
477.465***
(154.856)
4.185
(196.367)
-723.781***
(261.934)
-745.640***
(284.947)
yes
yes
5903.185***
(72.712)
0.008
849576

Schedule D

Schedule E

-231.878
(146.579)
-86.169
(118.565)
135.764
(131.805)
284.549
(178.504)
145.590
(212.183)
133.558
(249.393)
yes
yes
5729.068***
(94.025)
0.040
956067

1011.329***
(277.308)
1498.063***
(305.301)
845.633***
(313.095)
356.594
(417.286)
-405.062
(518.825)
-288.744
(584.497)
yes
yes
8201.870***
(129.265)
0.006
695152

 Table 6: Effect of auditing on different deduction items, for the pooled sample
Total Adjustments
1 Year Post Audit

12.351
(26.780)
2 Year Post Audit
-75.077**
(30.054)
3 Year Post Audit
-132.297***
(31.941)
4 Year Post Audit
-169.009***
(42.228)
5 Year Post Audit
-261.008***
(57.608)
6 Year Post Audit
-290.208***
(58.736)
Individual FE
yes
Year of Tax Return yes
Constant
1023.370***
(13.565)
R-squared
0.031
N
1323803
Notes: * p<0.10, ** p<0.05, *** p<0.01,

Itemized Deductions Charitable Contributions State & Local tax Mortgage interest
-562.917***
(119.826)
-1406.712***
(128.412)
-1690.529***
(139.131)
-1845.593***
(172.254)
-2507.810***
(234.789)
-2012.379***
(255.428)
yes
yes
13934.544***
(57.258)
0.064
1931745

-267.326***
(26.915)
-358.449***
(29.488)
-361.669***
(31.393)
-356.883***
(40.350)
-463.804***
(55.961)
-436.051***
(60.477)
yes
yes
2130.039***
(13.775)
0.012
1607515

-39.360
(28.098)
-91.159***
(32.039)
-118.969***
(37.640)
-119.188**
(52.516)
-337.769***
(67.109)
-252.940***
(78.813)
yes
yes
3026.309***
(17.492)
0.021
1859451

82.500
(70.714)
-415.768***
(76.328)
-636.392***
(78.949)
-785.216***
(95.295)
-1069.046***
(127.859)
-920.873***
(132.118)
yes
yes
6506.456***
(36.246)
0.076
1551268

 Table 7. Effect of audit and income volatility on subsequent taxable income

1 Year Post Audit
2 Year Post Audit
3 Year Post Audit
4 Year Post Audit
5 Year Post Audit
6 Year Post Audit

All Filers

All Filers, Volatility Controls Sch C Filers

Sch C Filers, Volatility Controls

575.784***
(200.673)
1000.612***
(220.897)
1261.721***
(274.855)
1335.868***
(364.205)
942.338*
(496.192)
1146.724*
(586.239)

3492.192***
(572.575)
5781.773***
(630.918)
7303.227***
(708.186)
6558.445***
(894.466)
8842.014***
(1374.814)
6669.378***
(1355.481)
-523.944***
(163.790)
-667.382***
(173.935)
-1063.773***
(184.120)
-764.784***
(230.788)
-1395.788***
(323.981)
-1003.972***
(321.598)

-0.025***
(0.006)
-0.109***
(0.007)
-0.134***
(0.007)
-0.153***
(0.009)
-0.201***
(0.012)
-0.182***
(0.013)

3492.192***
(572.575)
5781.773***
(630.918)
7303.227***
(708.186)
6558.445***
(894.466)
8842.014***
(1374.814)
6669.378***
(1355.481)
-523.944***
(163.790)
-667.382***
(173.935)
-1063.773***
(184.120)
-764.784***
(230.788)
-1395.788***
(323.981)
-1003.972***
(321.598)

yes
yes
43524.824***
(279.661)
0.004
849576

yes
yes
0.445***
(0.003)
0.115
849576

yes
yes
43524.824***
(279.661)
0.004
849576

1 Since Audit*Volatility
2 Since Audit*Volatility
3 Since Audit*Volatility
4 Since Audit*Volatility
5 Since Audit*Volatility
6 Since Audit*Volatility

Individual FE
Year of Tax Return
Constant

yes
yes
49892.838***
(149.769)
R-squared
0.008
N
4762852
Notes: * p<0.10, ** p<0.05, *** p<0.01,

 Figure 1: Income Variance and the Long Term Effects of Audit

 

1 year after audit

 



 

-4

% Change in Reported Income
0
4
8
12
16

Figure 2: E↵ects of Audit on Reported Income

-10

-5
0
Year Relative to Audit Year
Taxable Income
Wages and Salaries

5

Adjusted Gross Income
Schedule C Income

2

 -500

Change in Reported Taxable Income
0
500
1000
1500

2000

Figure 3: E↵ects of Audit on Reported Taxable Income

-2

0

2
Year Relative to Audit Year

3

4

6

 -700

Change in Reported Taxable Income
200
1100
2000
2900

Figure 4: E↵ects of Audit on Reported Taxable Income by Audit Experience

0

2
4
Year Relative to Audit Year
Positive Adj.
Negative Adj.

4

6
Zero Adj.

 -11

Change in Reported Income
-4
3
10

17

Figure 5: E↵ects of Audit on Reported Income by Income Source

0

2
4
Year Relative to Audit Year
Total Income
Sch C Income
Sch E Income

5

Wage Income
Sch D Income

6

 Figure 6: E↵ects of Audit by Filer Type, Extensive Margin
(b) Likelihood of Filing Schedule E

-.14

-.21

Change in Prob File a Sch C
-.16
-.11
-.06
-.01

Change in Prob File a Sch E
-.11
-.08
-.05
-.02
.01

.04

.04

(a) Likelihood of Filing Schedule C

2
4
Year Relative to Audit Year
All filers
Non-Sch C filers

6

0

2
4
Year Relative to Audit Year

Sch C filers

All filers
Non-Sch E filers

Change in Prob Report Wage Income
0 .02 .04 .06 .08 .1 .12 .14

(c) Likelihood of Reporting Wage Income

-.02

0

0

2
4
Year Relative to Audit Year
All filers
No Wage Income

6

Wage Income

6

6
Sch E filers

 -20

% Change in Reported Deductions
-13
-6
1

Figure 7: E↵ects of Audit on Reported Deductions by Deduction Type

0

2
4
Year Relative to Audit Year
Above the Line
Charitable
Mtg Interest

Itemized
State and Local Tax

7

6

 -800

Change in Reported Taxable Income
-300 200 700 1200 1700 2200

Figure 8: E↵ects of Audit on Reported Taxable Income by Filer Age

0

2
4
Year Relative to Audit Year
Age 25-34
Age 45-54

8

Age 35-44
Age 55-64

6

 Change in Reported Taxable Income
-5100
-2600
-100
2400
4900

Figure 9: E↵ects of Audit on Reported Taxable Income by Filer Income

0

2
4
Year Relative to Audit Year
1st Income Quintile
3rd Income Quintile
5th Income Quintile

9

2nd Income Quintile
4th Income Quintile

6

 Figure 10: E↵ects of Audit on Reported Taxable Income by Filing Method

0

0

Change in Reported Taxable Income
600
1200
1800
2400

(b) E-filers vs. Non-E-filers

Change in Reported Taxable Income
300 600 900 1200 1500 1800 2100

(a) Filers Who Used Paid Preparers vs. Others

0

2
4
Year Relative to Audit Year
Paid Preparer

6

0

No Paid Preparer

2
4
Year Relative to Audit Year
E-Filers

10

Non-E-Filers

6