Net Tax Gap and Remittance Gap Estimates
Agency decision
Ask Donna
What actually matters in this document.
Text
Income Tax
Compliance
Research
Net Tax Gap and Remittance Gap Estimates
(Supplement to Publication 7285)
Department of the Treasury
Internal Revenue Service
Research Division
Publication 1415 (4-90)
Department
of the
Treasury
Internal
Revenue
Service
Income Tax
Compliance
Research
Net Tax Gap and Remittance Gap Estimates
(Supplement to Publication 7285)
April 1990
Roger L. Plate
Director, Research Division
Gary P. Bingham
Assistant Director, Research Division
This report supplements the estimates of the gross tax gap
published in March of 1988 (Publication 7285). It presents the
corresponding estimates of the net tax gap (i.e., net of
enforcement collections). This report also includes estimates of
the remittance gap, which were not available for inclusion in the
1988 report.
Dennis R. Cox
Chief, Compliance Analysis Group
Berdj Kenadjian
Chief Economist
(Compliance Estimates)
Alan H. Plumley
Project Economist
Suggested Citation
Internal Revenue Service
Income Tax Compliance Research:
Net Tax Gap and Remittance Gap
Estimates
(Supplement to Publication 7285)
Publication 1415 (4-90)
Washington, DC 20224
Net Tax Gap and Remittance Gap Estimates
Table of Contents
page
Executive Summary . . . . . . . . . . . . . . . . . . . . . . . .
iii
Chapters:
I. Tax Gap Concepts . . . . . . . . . . . . . .
A. Gross Tax Gap . . . . . . . . . . . . . . .
B. Net Tax Gap . . . . . . . . . . . . . . . .
C. How Much of the Net Tax Gap Could IRS Recover?
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
1
3
4
II. The Net Tax Gap Estimates .
A. Individual Income Tax . . .
B. Corporate Income Tax . . .
C. Trends . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
9
9
11
11
III. Sources and Methods . . . . . . . . . . . . . . . . . . . . .
A. Range of Estimates of the Gross Tax Gap . . . . . . . . . . . . . .
B. Net Tax Gap . . . . . . . . . . . . . . . . . . . . . . . . .
13
13
15
IV. Remittance Gap . . . . . . . . . . . . . . . . . . . . . . . .
17
Tables:
1. Range of Estimates of the Gross Legal-Sector Income Tax Gap and Voluntary
Compliance Rates, Selected Tax Years, 1973-1992 . . . . . . . . . . . .
2
2. Range of Estimates of the Net Legal-Sector Income Tax Gap,
Tax Years 1987, 1984 and 1981 . . . . . . . . . . . . . . . . . . .
10
3. Legal-Sector Income Tax Remittance Gap Estimates,
Selected Tax Years, 1973-1992 . . . . . . . . . . . . . . . . . . .
18
Charts:
1. Range of Estimates of the Gross and Net Legal-Sector Income Tax Gap,
Tax Year 1987 . . . . . . . . . . . . . . . . . . . . . . . . . .
3
IRS Publication 1415, 19 April 1990
.
.
.
.
.
.
.
.
.
.
.
.
ii
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Net Tax Gap and Remittance Gap Estimates
Executive Summary
The mission of the Internal Revenue Service (IRS) is "to collect the proper amount
of tax revenue at the least cost . . . in a manner warranting the highest degree of public
confidence in our integrity, efficiency and fairness." (See Document 6987.) To achieve
this purpose, IRS must determine the extent of noncompliance with the tax law and
regulations. IRS tax gap estimates are comprehensive measures of noncompliance that
convey significant information about the challenges faced in collecting taxes that are not
voluntarily paid. A proper understanding of the nature of the tax gap is important for the
development of future IRS programs and revenue initiatives.
In 1988, IRS published a series of estimates of the "gross income tax gap," which
is the amount of true income tax liability that is not voluntarily paid.1 The principal
purpose of this new report (which should be seen as a supplement to that gross tax gap
report) is to present IRS's estimates of the "net income tax gap," which is the gross income
tax gap less the amount of income taxes paid or collected as a direct result of IRS
enforcement activity. These net income tax gap estimates represent the amount of income
tax liability that will remain unpaid after all IRS enforcement activity has been completed
for the tax year in question. Recovering more of the gross tax gap would require either that
voluntary compliance improves or that IRS expand or improve its enforcement activities (or
some combination of both). Although we cannot estimate how much of the net tax gap
IRS could recover as a result of expanding its existing enforcement activities, we know that
a significant portion of it could be collected cost-effectively through a balanced strategy that
incorporates expanded and improved IRS programs (including taxpayer service, education,
and enforcement), as well as compliance-oriented tax law changes (intended, for example,
to reduce complexity and eliminate various opportunities for noncompliance). This report
discusses the various difficulties inherent in detecting and collecting the remainder of the
net tax gap, as well as the problems associated with estimating how much could be
collected cost-effectively.
It is difficult for IRS to know in all cases what a taxpayer's "true" tax liability is.
This is because examiners cannot always detect all noncompliance and—for a variety of
legal and administrative reasons—some of the amount that is recommended, but is not
assessed, may represent "true" noncompliance. Therefore, estimates of the tax gap are
subject to significant uncertainty. For this reason, our 1988 report included a set of
"alternate" estimates of the gross income tax gap in addition to our "primary" estimates.
For the "primary" estimates, figures based on examinations of tax returns reflected the full
amount of additional tax recommended by the examiners; for the "alternate" estimates, these
figures were adjusted to allow for the portions of recommended tax deficiencies that are
conceded in the appeals process or lost in litigation. We continue to show two sets of
estimates for this report, but instead of referring to them as "primary" and "alternate," we
show them as a range2 of estimates—the "higher" tax gap estimates corresponding to our
1
Internal Revenue Service, Income Tax Compliance Research: Gross Tax Gap Estimates and Projections for
1973-1992, Publication 7285, Washington, D.C., March 1988.
2
By "range" we simply mean the difference between the two sets of estimates. The "true" tax gap need not lie
IRS Publication 1415, 19 April 1990
iii
Net Tax Gap and Remittance Gap Estimates
old "primary" estimates, and the "lower" tax gap estimates corresponding to our old
"alternate" estimates. We have also re-estimated the old "alternate" estimates, because after
those estimates were published, we found that the data used to make the adjustments were
not current. The lower estimates given in this report, therefore, are based on more recent
information.3
In most cases, the difference between the higher and lower estimates is relatively
small. However, the uncertainty concerning true tax liabilities is particularly unsatisfactory
with respect to the estimates for large corporations. Since we have assumed (for lack of
data) that the examination program detects the entire gross tax gap for large corporations,
and since large corporations pay virtually all of their assessments, this implies a higher
estimate of the net tax gap which is almost exactly equal to the amount of tax deficiencies
that is recommended by Examination but conceded in appeals or lost in litigation. For the
same reason, the lower estimate of the gross tax gap for large corporations implies that the
net tax gap is close to zero. This problem of estimating "true" tax liabilities (which is the
basis for estimating the gross tax gap) for large corporations is the subject of several IRS
research projects. We expect that future estimates will reflect the results of these studies.
All of these estimates include only taxes due on incomes earned in the legal sector
of the economy. They do not include unpaid taxes due on illegally-earned incomes, nor do
they include unpaid tax liabilities involving other types of federal taxes, such as estate and
gift, employment, or excise taxes.4 We do not have current estimates of the tax gaps in
these other tax areas.5
Also included in this report are new estimates of a significant element of the gross
income tax gap: the remittance gap. This is the amount of tax which is voluntarily reported
by taxpayers but is not voluntarily paid—either because taxpayers fail to remit it with their
returns, or because their employers fail to remit it after it is withheld. We now have the
information needed to make these estimates, which were not included in the 1988 report.
For 1987, the remittance gap estimate is $9.3 billion. (See Chapter IV for details.) The
1988 report gave a "primary" estimate of the gross legal-sector income tax gap for tax year
(TY) 1987 of $84.9 billion and a corresponding "alternate" estimate of $71.2 billion (which
we have re-estimated to be $73.2 billion, as discussed above). Thus, with the remittance
gap included, our estimates of the total gross legal-sector income tax gap range from $82.5
billion to $94.2 billion. Our estimates of the corresponding net legal-sector income tax
gap for TY 1987 range from $60.7 billion to $72.3 billion.
within this range—especially if our assumptions concerning the detectability of tax deficiencies are not correct.
If these assumptions are realistic, then the only other source of uncertainty concerns the ability of the
examination-appeals-litigation process to arrive at the "true" tax liability. This is the uncertainty defined by the
range in our two sets of estimates. See Chapter III for a discussion of these two sets of estimates.
3
See Internal Revenue Service, Evaluation of the IRS System of Projecting Enforcement Revenue, Publication
1501, Washington , D.C., November 1989.
4
For fiscal year (FY) 1987, income taxes accounted for 55.8 percent of federal budget receipts. Employment
taxes accounted for another 35.5 percent, while excise and estate and gift taxes combined accounted for 4.7
percent. The remaining 4.0 percent came from customs duties and miscellaneous receipts. See Office of
Management and Budget, Historical Tables, Budget of the United States Government, 1990, pp. 25, 37.
5
However, we plan that our next tax gap report in this continuing series will contain estimates of the
employment tax and illegal-sector income tax gaps.
IRS Publication 1415, 19 April 1990
iv
Net Tax Gap and Remittance Gap Estimates
1
I. Tax Gap Concepts
A. Gross Tax Gap
The amount of tax that is due but not voluntarily paid is the gross tax gap. This is
comprised of the reporting gap, which is the amount of tax liability that taxpayers do not
voluntarily report on their returns, and the remittance gap, which is the amount that
taxpayers report on their returns as due, but which is not voluntarily paid—either because
they do not remit it with their returns, or because their employers fail to remit what they
withhold from their wages.6 The percentage of tax liability voluntarily paid is the
voluntary compliance rate (VCR).
Table 1 presents estimates of the gross legal-sector income tax gap and the
associated VCRs for selected tax years from 1973 to 1992. As shown in Table 1, our
estimates of the TY 1987 VCR range7 from 83.6 percent to 81.7 percent. We estimate
that taxpayers voluntarily paid $411.3 billion in legal-sector income taxes for TY 1987. Our
estimates of what they did not voluntarily pay—the tax gap—range from $82.5 billion to
$94.2 billion of their total income tax liability that year. The word "voluntarily" as used
here means without any IRS enforcement action (such as examination, collection, or
criminal investigation) taking place between the due date of the tax and the time of payment.
Fostering voluntary compliance with the Federal tax law is an important IRS
objective. IRS seeks to maintain and improve voluntary compliance both through customer
service activities and through enforcement programs.
Customer service activities account for approximately 40 percent of the IRS budget.
These activities include such functions as taxpayer service, returns processing, and the
administration of tax forms and publications. IRS customer service initiatives aimed at
improving voluntary compliance have, over the last five to ten years, included such
activities as expanded telephone assistance for taxpayers, improved taxpayer education
(especially for new businesses), simplified tax forms, efforts to obtain the views of
taxpayers concerning their needs, and more rapid and accurate processing of tax returns.
These activities recognize that most taxpayers want to comply with the law, and IRS efforts
in these areas are designed to make it as easy as possible for them to be fully compliant.
6
Note that neither the gross tax gap nor the remittance gap are synonymous with IRS's accounts receivable
inventory, which is the amount of tax, penalties and interest that has been assessed, but has not yet been paid at
a particular time. Unlike the tax gap, accounts receivable includes interest and penalties, as well as assessments
related to many tax years and several types of taxes (some of which we do not have corresponding tax gap
estimates for). Moreover, since a large part of the gross tax gap is never assessed, that portion will never
become a part of accounts receivable. All of the remittance gap passes through accounts receivable (since, by
definition, it is self-assessed), but the remittance gap relating to a particular tax year of liability is not necessarily
all present in accounts receivable at any one time.
7
The higher estimates of the VCR correspond to the lower estimates of the tax gap, and the lower estimates of
the VCR correspond to the higher estimates of the tax gap.
IRS Publication 1415, 19 April 1990
2
Net Tax Gap and Remittance Gap Estimates
Table 1. Range of Estimates of the Gross Legal-Sector Income Tax Gap1
and Voluntary Compliance Rates (VCRs)2, Selected Tax Years,
1973-1992
Gross Income Tax Gap (in $ billions)
Voluntary Compliance Rate (%)
Tax
Year
Total
Individual
Income Tax 3
Corporate
Income Tax 4
Total
Individual
Income Tax 3
Corporate
Income Tax 4
1973
27.9 to 32.0
22.0 to 22.7
5.9 to 9.3
83.9 to 82.0
83.0 to 82.6
86.6 to 80.3
1976
39.7
45.5
32.6
33.7
7.1
11.8
82.7
80.7
81.2
80.7
87.3
80.6
1979
61.0
69.0
51.5
53.2
9.6
15.9
81.9
80.0
80.3
79.8
87.2
80.4
1981
68.9
76.1
60.1
61.9
8.7
14.1
82.8
81.4
82.0
81.6
87.0
80.6
1982
64.6
70.5
57.5
59.1
7.1
11.4
82.9
81.7
82.2
81.8
87.1
80.8
1984
83.0
91.3
73.5
75.7
9.6
15.6
81.6
80.1
80.4
80.0
87.2
80.6
1986
96.1 105.6
86.0
88.8
10.1
16.8
81.0
79.5
79.7
79.2
87.7
81.1
1987
82.5
94.2
69.1
71.4
13.4
22.8
83.6
81.7
82.3
81.8
88.1
81.3
1988
84.5
96.8
70.3
72.6
14.2
24.3
84.0
82.0
82.7
82.3
88.1
81.2
1992
110.1 127.0
91.0
94.0
19.1
33.1
84.6
82.7
83.6
83.1
88.1
81.1
Some components do not add due to rounding.
1 The results of examinations of tax returns are used in estimating several parts of the tax gap. Where these
results are used, the higher estimates of the tax gap (and the corresponding lower estimates of the voluntary
compliance rate) are based on the amounts of additional tax recommended by IRS examiners, and the lower
estimates of the gap (and the corresponding higher estimates of the VCR) are based on the amounts of tax
ultimately assessed after all appeals and litigation. If we have correctly estimated the extent to which
examiners cannot detect all tax deficiencies, then the "true" tax gaps (or VCRs) lie between these two sets of
estimates. See Chapter III for a discussion of the two sets of estimates. The estimates include estimates of
remittance gaps, which were not included in Publication 7285.
2 The Voluntary Compliance Rate is what taxpayers voluntarily pay as a percentage of their total tax liability.
3 Includes insufficient remittance by individuals and underdeposit of individual income tax withheld by
employers.
4 Includes fiduciaries' income tax and unrelated business income tax of tax-exempt organizations. The VCRs for
all corporations combined fluctuates slightly—even though we have assumed constant compliance rates—
because of the changing mix of small, mid-size, and large corporations over time.
Improving voluntary compliance is also a major goal of IRS enforcement programs.
These programs seek to encourage voluntary compliance indirectly by: (a) improving the
subsequent voluntary compliance of the specific taxpayers subjected to IRS enforcement
programs; and (b) improving the voluntary compliance of taxpayers who are not directly
affected by enforcement actions, but who nevertheless may become more compliant due to
the general deterrent effect of IRS activity—the so-called "ripple effect." These indirect
revenue effects of enforcement are very difficult to measure; we do not currently have
estimates for them. We do, however, have estimates of the amount of additional tax which
is involuntarily paid as a direct result of IRS enforcement activities.
IRS Publication 1415, 19 April 1990
Net Tax Gap and Remittance Gap Estimates
3
B. Net Tax Gap
The net tax gap is the part of the gross tax gap that is not collected through IRS
enforcement activities. As shown in Chart 1, our estimates of the net legal-sector income
tax gap for TY 1987 range from $60.7 billion to $72.3 billion. The difference between
these estimates and the corresponding gross tax gap estimates for the same year represents
an estimated $21.9 billion of tax which will be collected as a result of current IRS
enforcement activities.8 IRS enforcement programs, although designed to encourage
individual and corporate voluntary tax compliance, do nonetheless yield significant
amounts of revenue directly. Some of these programs focus on taxpayers who file returns
(filers), and some on taxpayers who fail to file returns (nonfilers).
Chart 1. Range of Estimates† of the Gross and Net Legal-Sector Income
Tax Gap ‡ (in billions), Tax Year 1987
Gross Tax Gap
$94.2
Voluntary Payments of
Tax Year 1987 Income Tax
$411.3
Net
Tax Gap
$72.3
Higher
Estimates
Tax Year 1987 Income Tax Payments
Projected to Arise Directly From Enforcement
$21.9
Voluntary Payments of
Tax Year 1987 Income Tax
$411.3
Net
Tax Gap
$60.7
Lower
Estimates
Gross Tax Gap
$82.5
† The results of examinations of tax returns are used in estimating several parts of the tax gap. Where these
results are used, the higher estimates of the tax gap are based on the amounts of additional tax recommended by
IRS examiners, and the lower estimates are based on the amounts of tax ultimately assessed after all appeals and
litigation. If we have correctly estimated the extent to which examiners cannot detect all tax deficiencies, then
the "true" tax gap lies between these two estimates. See Chapter III for a discussion of the two sets of estimates.
‡ All amounts refer to income tax only. Other taxes (such as employment taxes) are excluded. Voluntary
payments were projected for the 1988 gross tax gap report, and have not been adjusted to reflect actual
payments. Payments arising from enforcement are projections of what will be collected from tax year 1987
liabilities, and exclude interest and penalties.
8
Enforcement revenue includes penalties and interest collected, but since the gross tax gap is measured in
terms of tax only, penalties and interest are not included in estimates of enforcement yield for the purpose of
estimating the net tax gap.
IRS Publication 1415, 19 April 1990
4
Net Tax Gap and Remittance Gap Estimates
The filer programs include: (a) examination of tax returns (which, in over 80
percent of the cases, involves face-to-face contact between IRS employees and taxpayers or
their representatives); (b) computer matching of tax returns against third-party information
documents (such as Form W-2 withholding statements filed by employers and Form 1099
interest statements filed by banks); (c) math error verification; (d) other computer-based
adjustments of taxes; and (e) collection of overdue taxes through computer-generated
notices and field operations.
IRS has two main nonfiler programs. The first identifies "stopfilers"—those who
had filed returns for a prior tax period, but not for the one tested. The second identifies
delinquent nonfilers through matches of third-party information documents with IRS
masterfile records of current-year tax return filings. The delinquent returns and taxes are
obtained mainly through computer-assisted correspondence, automated paperless
operations, or, generally as a last resort, field investigations.
C. How Much of the Net Tax Gap Could IRS Recover?
On its face, the net tax gap appears to be a source of additional federal revenue that
could be realized without raising taxes. A significant portion of the net tax gap could be
collected with a balanced strategy that incorporates expanded and improved IRS programs
(including taxpayer service, education, enforcement, and systems modernization), as well
as compliance-oriented tax law changes (intended, for example, to reduce complexity and
eliminate various opportunities for noncompliance). It is important to realize that IRS
enforcement alone is neither a cost-effective nor a socially acceptable method of collecting
the net tax gap. We do not know how much can be collected through enforcement, but the
nature of the noncompliance reflected in the tax gap estimates suggests that the potential net
revenue gains from enforcement alone are limited; the key is to pursue a balanced strategy.
The following sections contain a discussion of the nature of income tax noncompliance and
of the difficulties in trying to estimate the maximum portion of the net tax gap that could be
collected through additional enforcement activity.
The Nature of Noncompliance
The gross tax gap consists of some types of noncompliance (e.g., overstated
deductions) that are much easier to detect than others (e.g., unreported income that is not
covered by third-party information returns). In fact, almost half of the gross tax gap
estimate for TY 1987 among filers of individual income tax returns is associated with types
of unreported income that are very difficult to detect. These include income not reported by
"informal suppliers" (proprietors who operate with informal business styles, usually on a
cash basis), other non-farm proprietors (who may keep two sets of "books"), farmers
(who may understate cash income received for their produce), and skilled and professional
workers (some of whom "moonlight" on a cash basis).9 Although we have a reasonable
9
This is not to say that all of the tax gap among these taxpayers is not detectable. These are just some
examples of the hard-to-find types of noncompliance among individuals. We estimate what is not detectable by
multiplying the amount that is detected (without the help of third-party information documents) by a factor
IRS Publication 1415, 19 April 1990
Net Tax Gap and Remittance Gap Estimates
5
basis for estimating the total extent of such noncompliance (at least for individuals), it is
very time-consuming and costly to detect it on a case-by-case basis.
It is also difficult to collect some of what can be detected. Although most
taxpayers who are found to owe additional tax, penalties, and interest pay their assessments
fairly quickly in response to an IRS notice or two, some accounts are never fully
collected—especially if the taxpayers are unable to pay.
Moreover, much of the remaining noncompliance occurs in small amounts, spread
over a large number of taxpayers. Since our return selection systems are not perfect, we
would have to audit all returns in order to detect the maximum amount of noncompliance
possible. And, since enforcement costs are significant, most of these detectable amounts
are not cost-effective to pursue with existing enforcement programs—either to detect or to
collect if they are detected.
A significant portion of the net tax gap could be collected with a balanced strategy
designed to foster increased voluntary compliance (through improved taxpayer service,
enhanced taxpayer education, expanded enforcement, systems modernization, and tax law
changes) and to collect directly the revenue generated by the expanded enforcement.
Nonetheless, the net tax gap should not be viewed as a large bank account with which to
balance the budget.
Difficulties in Estimating
Maximum Cost-Effective Enforcement Yields
Current IRS enforcement programs operate at very favorable yield-to-cost ratios—
even at the margin. An expansion of most programs would result in eventual collections
amounting to at least four or five times the total cost of the expansion. It is clear that IRS
would have to expand enforcement programs dramatically before the marginal cost
exceeded the marginal revenue. While this assures us that IRS could cost-effectively
recover much of the net tax gap, it also makes it very difficult to estimate just how much.
Since a large expansion would put IRS enforcement levels well beyond current operational
levels, we have little information upon which to estimate the likely revenues and costs at
those levels. Research is under way to improve our estimates of the maximum amount that
can be recovered cost-effectively, but for now, even our best guesses would have
numerous weaknesses. The most important of these weaknesses are discussed below:
1. Unknown indirect effects: It is generally thought that, in addition to the direct revenue
gains that arise from IRS enforcement, enforcement activities also improve voluntary
compliance indirectly through deterrence. If so, then significant expansions of
enforcement programs would generate corresponding increases in revenues paid
voluntarily in subsequent years—both by those contacted by the enforcement activities
and by others who learn of the expanded enforcement. This is an area of research
receiving increasing attention, but there is still no consensus as to the magnitude of
derived from a special study. Refer to our 1988 gross tax gap report (Publications 7285 and 1415).
IRS Publication 1415, 19 April 1990
6
Net Tax Gap and Remittance Gap Estimates
these indirect effects. Estimates (outside IRS) have ranged from roughly the same
magnitude as the direct revenue effects to as much as seven times this magnitude. For
example, these estimates imply that if an expansion in IRS enforcement activity
produced $100 million of revenue directly from the taxpayers contacted, it would also
indirectly induce other taxpayers to pay an additional $100 million to $700 million
voluntarily. Although we should include indirect revenues in the yields we expect from
any expansion, we are unable to do so with any confidence.
2. Unknown taxpayer costs: When IRS contacts a taxpayer through one of its
enforcement programs, such as by auditing his or her tax return, the taxpayer bears
some unavoidable costs. The taxpayer must take the time to assemble records, attend
interviews, etc. Sometimes taxpayers elect to pay for professional assistance in dealing
with IRS enquiries. Currently we cannot estimate the magnitude of these
inconveniences and out-of-pocket outlays; ideally they should be included in the costs
of any expansion. (In any revenue estimates we might make, we would also need to
account for those cases in which some of these expenses would be deductible in
calculating taxable income.)
3. Uncertain estimates for large corporations: It is probable that significant additional
amounts of tax deficiency could be identified cost-effectively by examining the complex
returns of large corporations in greater depth.10 It is also likely that the amount
collected could be increased by improving the quality of the examination-appealslitigation process so that we assess a larger fraction of the tax deficiency that is
proposed. However, we currently have no data with which to make estimates of the
maximum cost-effective potential of these two types of improvements. We have
initiated projects that may help us make such estimates, but the results of these studies
are not yet available.
4. Speculative nature of yield curves: Even for individuals and small corporations, for
which we could estimate marginal yield curves based on data from the Taxpayer
Compliance Measurement Program (TCMP)11 and operational data, we could not
currently have a high level of confidence in estimates based on them because any such
curves would be highly speculative at coverage levels significantly beyond operational
levels. There are several reasons for this: TCMP results do not reflect what could or
would be recommended by operational examinations; we cannot adequately sort the
TCMP cases to reflect the order in which returns would be examined in operational
practice; and even the most recent TCMP results are always several years old—not able
to reflect completely the impact of changes in law and in taxpayer behavior. (The yield
curves used for resource allocation are not as susceptible to these problems, since
operational levels do not vary much from year to year.)
10
Note that since we lack a reasonable way to estimate how much more tax deficiency could be found among
large corporations, our large corporation gross tax gap estimates only include the amount detected in our
operational programs. When our special projects to quantify this additional gap are completed, we will adjust our
gross tax gap estimates accordingly. (All of this additional gap may not be cost-effective to recover, however.)
11 See Chapter III for an explanation of TCMP.
IRS Publication 1415, 19 April 1990
Net Tax Gap and Remittance Gap Estimates
7
5. Unknown impact of improved productivity: We currently have little information upon
which to estimate how much the productivity of the enforcement functions can be
improved—measured either in terms of the ratio of yield to cost or in terms of the
percentage of the tax gap that is detected, assessed, and collected. We could estimate
the revenue impact of expanding existing programs at the margin assuming no change
in the relative productivities of the enforcement functions, but we cannot reliably
estimate the maximum revenue impact of improving the quality of our enforcement
efforts—even at present coverage levels. Improving quality—collecting more of the tax
gap with existing resources—might include increasing the cost-effectiveness of
enforcement personnel and/or changing the allocation of resources between the various
functions involved in the enforcement process (e.g., Examination, Appeals, Tax
Litigation and Collection). Several new efforts have been suggested—with some now
reaching the implementation stage—that will yield more in revenue than they cost, but
we cannot estimate the potential revenue that other (as yet unknown) efforts might
realize.
6. Transitional effects: Even if the government did want to maximize the net revenue
gains arising from enforcement, it could not expand IRS enforcement to these levels in
one step. It would probably be very difficult to hire that many new agents at once, and
it would take a significant amount of time to train them once they were hired. Instead,
the expansion would occur gradually over a number of years. Each step in the
expansion would be supported by specific estimates (much more accurate than any we
could make of the maximum cost-effective amount) of the net revenue that would be
generated by that step alone. As the gradual expansion took place, valuable information
would be gained about the amounts of revenue actually obtained from the additional
effort, and about the actual cost of the effort. Moreover, from a budget standpoint, any
major expansion would incur significant transitional costs—such as the direct cost of
training new personnel, the lower productivity of these new people during their first
several years, and the forgone revenues associated with assigning experienced agents to
train and coach them.
Difficulties in Estimating Compliance Gains
From Improved Customer Service Programs
Experience suggests that IRS customer service activities have a positive impact on
voluntary compliance. To the extent that IRS can help taxpayers to reduce errors and
minimize the frustrations many experience trying to get help, we might realize significant
direct—and, perhaps, indirect—revenues. However, we are not yet able to quantify the
extent to which the net tax gap can be reduced by improving or expanding our current
efforts in this area.
IRS Publication 1415, 19 April 1990
8
Net Tax Gap and Remittance Gap Estimates
Conclusions
Two major conclusions can be drawn from this discussion:
•
A significant portion of the net tax gap (plus amounts of interest and penalties of
roughly the same magnitude as the tax collected) could be recovered cost-effectively by
expanding IRS enforcement efforts; and yet
•
The greatest challenge for closing the net tax gap lies in the potential for improving
voluntary compliance—by expanding and improving taxpayer service and education
programs, by changing the tax law to reduce complexity and to eliminate opportunities
for noncompliance, and by modernizing our systems.
Both of these areas are topics of on-going research. As for expanding IRS
enforcement efforts, projects are under way that would identify ways to improve the costeffectiveness of resource allocations to the various enforcement programs, and other
studies are proceeding that should improve our ability to estimate the maximum potential of
expanded enforcement. As for increasing voluntary compliance, we are developing ways
to improve and enhance taxpayer service and taxpayer education programs (although
estimating the revenue impact of such improvements may not be possible), and we are
studying the indirect effects of enforcement. These research projects will not only aid us in
making estimates for future tax gap reports, but—more importantly—will guide IRS efforts
directed at closing the net tax gap.
IRS Publication 1415, 19 April 1990
Net Tax Gap and Remittance Gap Estimates
9
II. The Net Tax Gap Estimates
We have estimated the net legal-sector income tax gap for Tax Years 1981, 1984,
and 1987. Table 2 presents detailed estimates for TY 1987, together with less-detailed
estimates for TY 1981 and TY 1984. Our higher estimate of the net tax gap increased from
$60.7 billion in 1981 to $72.3 billion in 1987, while our lower estimate increased from
$53.4 billion to $60.7 billion in the same period. This increase (which, for the higher
estimate amounted to $11.6 billion, or 19.1 percent), does not reflect a decline in
compliance; indeed, we estimate that the voluntary compliance rate rose slightly over this
period (see Table 1). Instead, the increase arose from the increase in tax liabilities during
the period. Moreover, the enforcement figures in Table 2 suggest that IRS is recovering an
increasing share of the gross tax gap through its various enforcement programs—from
20.2 percent of the higher estimate for 1981 to 23.2 percent of the corresponding estimate
for 1987. The sections below discuss the estimates for the individual income tax and the
corporate income tax separately, followed by a discussion of the trends in the estimates.
A. Individual Income Tax
For TY 1987, our estimates of the gross legal-sector income tax gap for individuals
range from $69.1 billion to $71.4 billion. The portion of this which will eventually be
collected through IRS enforcement programs (operated at currently anticipated levels) is
estimated to be $13.6 billion. Our estimates of the net tax gap for individuals, therefore,
range from $55.5 billion to $57.8 billion, as shown in Table 2. (See Chapter III for a
discussion of the relative merits of the higher and lower estimates.)
Table 2 presents some detail on the sources of enforcement revenue for the
individual income tax. Examination of tax returns and collection of the remittance gap are
the largest sources. For 1987, out of the total estimated enforcement tax yield of $13.6
billion in individual income tax, we expect the examination of returns and the collection of
the remittance gap (including both insufficient remittance of income tax by individuals and
underdeposit of withheld income tax by employers) to account for $4.7 billion each.
Matching of third-party information documents (mainly Forms W-2 or 1099) with tax
returns or with the IRS individual masterfile is the next-largest element of enforcement
revenue.12 IRS document matching programs for filers and nonfilers combined will
recover an estimated $2.3 billion of unreported 1987 tax liabilities.13 Correction of
taxpayer math errors is expected to recover another $1.0 billion of net additional
revenue.14 Finally, delinquency investigations of those who stop filing returns will yield
an estimated $0.7 billion in tax revenue.
12
Matches with the masterfile identify individuals who had not filed tax returns even though they received
significant amounts of income.
13 This estimate takes into account significant budget cuts that reduced the number of TY 1987 cases worked.
14 This is an estimate based on prior years. Preliminary data for 1987 suggest that the net amount collected by
the math-error program was actually negative for TY 1987, meaning that taxpayers made larger errors in the
government's favor than in their own. If this implication is borne out by further analysis—for example, of the
impact of the Tax Reform Act of 1986 and of certain administrative changes—then we will revise the tax gap
estimates accordingly in our next report.
IRS Publication 1415, 19 April 1990
10
Net Tax Gap and Remittance Gap Estimates
Table 2. Range of Estimates of the Net Legal-Sector Income Tax Gap 1
(in $ billions), Tax Years 1987, 1984, and 1981
(1)
(2)
(3)
Gross income
tax gap
(1) + (2)
(4)
Enforcement
yield
(tax only)
Gross income tax
reporting gap
Remittance
gap
(5)
Net income
tax gap
(3) - (4)
Tax Year 1987
Total
73.2 to 84.9
9.3
82.5 to 94.2
21.9
60.7 to 72.3
Individual Income Tax
• Filers
Examination of returns
Document matching
Correction of math errors
Other computer-based
adjustments
Insufficient remittance
61.2 to 63.5
54.0
56.3
7.9
5.0
69.1 to 71.4
59.0
61.3
13.6
9.9
4.7
1.2
1.0
55.5 to 57.8
47.2
49.5
• Nonfilers
Stopfilers
Document matching
0.2
2.8
7.2
• Employer underdeposit
of withheld income tax
Corporate Income Tax
•Small
•Mid-size
•Large
•Fiduciaries andUBIT 2
12.0 to 21.4
4.8
5.2
0.7
1.6
6.4
14.3
0.2
0.3
7.2
1.8
0.7
1.1
5.4
2.9
2.9
1.9
1.0
1.4
0.2
0.9
0.0
0.3
13.4 to 22.8
5.0
5.4
1.6
2.5
6.4
14.3
0.5
0.6
8.3
1.6
0.5
6.0
0.1
5.2 to 14.5
3.4
3.8
1.1
1.9
0.3
8.2
0.4
0.5
83.0 to 91.3
73.5
75.7
9.6
15.6
18.9
13.0
5.9
64.2 to 72.4
60.5
62.7
3.7
9.7
68.9 to 76.1
60.1
61.9
8.7
14.1
15.4
10.7
4.7
53.4 to 60.7
49.4
51.3
4.0
9.5
Tax Year 1984
Total
Individual Income Tax
Corporate Income Tax
72.7 to 81.0
64.2
66.4
8.6
14.6
10.3
9.3
1.0
Tax Year 1981
Total
Individual Income Tax
CorporateIncome Tax
59.7 to 66.9
51.8
53.6
7.9
13.3
9.2
8.3
0.8
Some components do not add due to rounding.
1 The results of examinations of tax returns are used in estimating several parts of the tax gap. Where these results are
used, the higher estimates of the tax gap are based on the amounts of additional tax recommended by IRS examiners,
and the lower estimates are based on the amounts of tax ultimately assessed after all appeals and litigation. If we
have correctly estimated the extent to which examiners cannot detect all tax deficiencies, then the "true" tax gaps lie
between these two sets of estimates. See Chapter III for a discussion of the two sets of estimates.
2 UBIT is unrelated business income tax of tax-exempt organizations.
IRS Publication 1415, 19 April 1990
Net Tax Gap and Remittance Gap Estimates
11
B. Corporate Income Tax
Our estimates of the TY 1987 gross legal-sector income tax gap for corporations
range from $13.4 billion to $22.8 billion. Of this, an estimated $8.3 billion will eventually
be collected as a result of IRS enforcement. Our estimates of the net corporate income tax
gap, therefore, range from $5.2 billion to $14.5 billion.
The corporate tax gap estimates are subject to more uncertainty than the estimates
for individuals. While the latter are based on special, intensive audits designed to measure
taxpayer compliance (under TCMP), the estimates for corporations are based mainly on the
results of operational audits (the only exception being the estimates for corporations having
assets under $10 million, which are based on TCMP results). Furthermore, the Tax
Reform Act of 1986 (TRA) will likely affect the corporate net tax gap in two ways—
neither of which is fully understood, given the lack of data. IRS estimates of the gross gap
assume that the amount of corporate noncompliance will grow in proportion to the growth
of tax liabilities brought about by TRA. However, the projections of revenue to be
recovered from examination of corporation tax returns assume that TRA will reduce
enforcement yields for a given expenditure of resources. Enforcement yields will be
reduced in part because TRA's lower tax rates reduce the tax liability associated with each
dollar's worth of unreported taxable income detected by revenue agents. Also, TRA's
repeal of the investment tax credit eliminates an item for which each dollar of adjustment by
examiners changes tax liability by a full dollar, rather than by just some fraction of a dollar.
C. Trends
Table 2 shows estimates of the gross and net legal-sector income tax gap for tax
years 1987, 1984, and 1981. For example, from 1981 to 1987 our higher estimate of the
gross tax gap increased by $18.1 billion, or 24 percent. For the same period, the estimate
of enforcement tax yield increased by $6.5 billion, or 42 percent. The difference in growth
rates reflects the fact that IRS enforcement revenues are growing not only because of the
growth of the tax gap, but also because of expansions and improvements in enforcement
programs. The corresponding net tax gap estimate grew by $11.6 billion, or only 19
percent, from 1981 to 1987.
Looking separately at the 1981-1984 and 1984-1987 periods, Table 2 indicates that
the estimated gross income tax gap increased during both periods—except for the lower
estimate, which declined slightly from 1984 to 1987. The estimated net tax gap declined
slightly from 1984 to 1987 after increasing substantially from 1981 to 1984. The estimated
gross and net tax gaps for individuals both declined from 1984 to 1987. In contrast, the
gross and net tax gap estimates for corporations both increased between 1984 and 1987.
These increases in the corporation estimates are due largely to the effects of tax reform. As
discussed above, TRA tended to increase corporate tax liabilities while simultaneously
reducing anticipated yields from audits of corporate tax returns.
IRS Publication 1415, 19 April 1990
12
Net Tax Gap and Remittance Gap Estimates
IRS Publication 1415, 19 April 1990
Net Tax Gap and Remittance Gap Estimates
13
III. Sources and Methods
A. Range of Estimates of the Gross Tax Gap
The gross tax gap is the difference between "true" tax liability and the amount
voluntarily paid. Since we are unable to determine "true" tax liability precisely, we
provided two sets of estimates of the tax gap in our March 1988 report: "primary"
estimates based on the amount of tax adjustments recommended by IRS examiners, and
"alternate" estimates based on the amount eventually assessed, after all appeals and
litigation have been completed.15 The difference between these two estimates is the
amount of the original recommendation which is conceded in the administrative appeals
process or lost in litigation. In this report we refer to the "primary" estimates as our
"higher" estimates, and to the "alternate" estimates as our "lower" estimates.
Results of examinations (audits) of tax returns are the fundamental data for major
parts of our gross tax gap estimates. For individuals and small corporations (those with
less than $10 million in assets), the examinations upon which our estimates are based are
special audits conducted under TCMP.16 For larger corporations, the examinations are the
regular operational audits of the tax returns of these corporations.17 At the conclusion of
each audit, the tax deficiency "recommended" by the examiner is recorded—on a special
checksheet in the case of TCMP audits, and in the Audit Information Management System
(AIMS) in the case of regular audits. Unfortunately, neither data source includes
information on the amounts eventually assessed on these audits; we have had to rely on
other data to estimate these amounts. Moreover, the "alternate" estimates of the gross tax
gap in our 1988 report were based on statistics which have since been found to be
incorrect. The corresponding estimates reported here have been revised to reflect more
reliable information. In this section, we describe the theoretical basis for these lower
estimates, discuss the relative merits of the lower and higher estimates, and briefly describe
the previous and current sources and methods for the estimates.
The theoretical attraction of the lower estimates is that, in many cases, the final
assessment arising from the examination may better reflect the "true" tax liability than does
15
These two sets of estimates of the tax gap should not be interpreted as two concepts of the tax gap. Since
there is only one "true" liability, and a definite amount is voluntarily paid, then there is only one "true" tax gap
amount. We provide two sets of estimates only because of uncertainty concerning the "true" liability, which—if
our assumptions concerning the detectability of tax deficiencies are correct—is probably somewhere between our
two estimates.
16 Under TCMP, IRS completes thorough, line-by-line examinations of several different types of tax returns.
For each type of return, IRS examines a probability sample taken from the entire population, allowing us to make
population estimates dealing with a wide variety of compliance issues. To compensate for the fact that even
TCMP examinations cannot detect all unreported income, we multiply the amount that is detected (without the
help of third-party information documents) by a factor derived from a special study. Refer to our 1988 gross tax
gap report (Publications 7285 and 1415) for more details.
17 Our gross tax gap estimates for these corporations account for the fact that not all large corporate returns are
examined. However, we assume (for lack of data) that operational audits of large corporations' returns detect all
tax deficiencies. That is, we "multiply" the amount that is detected by a factor of 1.0. (See the preceding
footnote.)
IRS Publication 1415, 19 April 1990
14
Net Tax Gap and Remittance Gap Estimates
the initial examination recommendation.
This may be so because, first, the
appeals/litigation process may be considered a review of the examiners' work, so that some
errors may be eliminated; and second, the eventual assessment may be considered to be the
true legal liability in the sense that IRS cannot attempt to assess more later, except in
unusual cases.
The lower estimates of the tax gap may seem most appropriate in the case of large
corporations since more than 90 percent of recommendations from operational
examinations are appealed by these taxpayers. Even here, however, it is not obvious that
the entire amount conceded or lost should be removed from the estimates of the gross tax
gap. There are various reasons for these concessions. For example, appeals officers
sometimes simply make mistakes in conceding examination recommendations. In this
case, the "true" liability is the original recommendation—supporting the higher estimates of
the tax gap. Examiners also make mistakes18 sometimes—proposing too much additional
tax—but there are several categories of reasons for concessions in the appeals/litigation
process which fall between these clear polar extremes, making it difficult to estimate where
the "true" tax liability lies. In addition to the uncertainty concerning what portion of
unassessed recommendations represent "true" tax deficiencies, it is possible that even the
higher estimates may understate the tax gap for large corporations. Since the examinations
of large corporations are not necessarily thorough in pursuing all potential issues on these
complex returns, substantial amounts of tax deficiency may not be recommended by the
examiners and, hence, may never be included in either the higher or the lower estimates.
For the individual and small corporate tax gap estimates, there are additional
complications. TCMP examinations, upon which these estimates are based, are controlled
as part of major research projects, in which the principal purpose of finding the proper tax
liabilities of the examined taxpayers is augmented by the purpose of estimating the total tax
liability of all taxpayers combined. TCMP examiners are specially trained to be
exhaustively thorough in finding all noncompliance, whether it reduces the taxpayers'
liabilities or increases them. Every examination is subjected to quality review to determine
that all relevant probes have been made and that the results are correct. While it is
unreasonable to assume that the TCMP checksheets have no errors whatsoever, it is not
unreasonable to assume that the errors are symmetrically distributed with an average of
zero; that is, that the aggregate estimate of the total tax liability is not biased. If all TCMP
cases were reviewed further by—for example—appeals officers, some errors would be
found and corrected, but the aggregate amount of noncompliance would, under this
assumption, not change significantly, since both positive and negative errors would be
corrected. However, in practice, appeals officers review only those TCMP cases which are
appealed by taxpayers (representing only about 20 percent of the recommended deficiencies
for individuals, and about 30 percent for small corporations). It seems reasonable to
assume that taxpayers whose examiners recommended too little additional tax will, in the
18
Actually, not all overstatements of the "true" tax liability are mistakes. For example, in some situations,
examiners will recommend the "same" tax deficiency against more than one taxpayer, even though it is clear that
only one of them owes the tax. This happens because the factual situations make it unclear which of them does
owe the tax, and it is tactically wise to recommend it against all taxpayers who might owe it, pending a final
determination in the appeals/litigation process. Here, the "true" liability is the amount assessed, and the
duplicated recommendation clearly is not part of the tax gap.
IRS Publication 1415, 19 April 1990
Net Tax Gap and Remittance Gap Estimates
15
aggregate, be less likely to appeal the recommended deficiency, and that those whose
examiners recommended too much additional tax will be more likely to appeal. If so, then
adjusting the overall TCMP estimates for the results of the appeals process would introduce
a bias into the tax gap estimates—that is, the lower estimates would understate the true
liability of the taxpayers.
Because of the uncertainties concerning the implications of the difference between
examination recommendations and final assessments, IRS is conducting research projects
to illuminate the matter. These projects involve dividing concessions in the appeals process
into the various categories of reasons for concession. We have projects for regular
operational examinations and for TCMP examinations. Pending the results of these
studies, we will publish two sets of estimates of the tax gap, as in this report.
The "alternate" estimates of the gross tax gap in our 1988 report were based on
information gathered by the Audit Integrated Reporting System (AIRS). This system
tracks each completed examination through to its final disposition (after the
appeals/litigation and collection processes have been completed). Our earlier estimates
were based on tabulations for examinations completed in FY 1982. After we published
those estimates, the 1982 tabulations were found to contain major anomalies. The 1972
tabulations are evidently accurate but, of course, very out of date. Over the past decade the
proportion of examination recommendations which taxpayers appeal has risen dramatically.
Meanwhile, the proportion of appealed deficiencies which is conceded has remained fairly
stable. It is clear, therefore, that the difference between examination recommendations and
final assessments has grown significantly since 1972. This situation is described more
completely in Publication 1501, Evaluation of the IRS System of Projecting Enforcement
Revenue. The lower estimates of the reporting gap shown above in Chapters I and II are
based on the "interim method" described in that report.
B. Net Tax Gap
Estimating the net tax gap is, in principle, a straightforward process given that the
gross tax gap has already been estimated. One simply subtracts from the gross tax gap
estimate the tax revenue (associated with the same tax year) that is eventually collected by
enforcement activities. However, the available data on enforcement yields are not
commensurate in various ways with the gross tax gap estimates. Therefore, the process of
estimating the net tax gap involves numerous adjustments to the enforcement yield statistics
to get figures that are conceptually comparable to the gross tax gap estimates. These
adjustments are discussed below.
IRS management information systems generally report data on enforcement
results—proposed adjustments, actual assessments, and actual collections—by fiscal year
of IRS operation. The tax gap estimates, on the other hand, of necessity relate to the tax
years in which tax liabilities are incurred. Thus, an important step in estimating the net tax
gap is to translate fiscal year enforcement results into the corresponding results by tax year
of liability.
IRS Publication 1415, 19 April 1990
16
Net Tax Gap and Remittance Gap Estimates
Collecting taxes that were not voluntarily paid takes time. The assessments of
additional tax IRS makes during a particular fiscal year relate to a number of earlier tax
years. Assessments made in FY 1987, for example, related to returns filed for tax years
1986, 1985, 1984, and even earlier—going back to the late 1970s. Reaching agreement
with some large corporations on the additional tax due may take longer than 10 years,
including time taken in appeals and, in some cases, litigation. In these cases, assessments
will fall behind incurred tax liabilities by more than a decade. Consequently, even if the
IRS management information systems were to report data by tax year of liability, we would
have to wait until the second half of the 1990s to determine enforcement tax yields related
to tax years in the late 1980s.
We translate the available data (and projections) on proposed adjustments or
assessments of additional tax (by fiscal year of recommendation or assessment) to estimates
for the corresponding tax years using typical distributions of the elapsed time between the
tax year of liability and the fiscal year of assessment. Revenue collections, in turn, often
lag behind assessments. For purposes of tax gap estimation, however, we need not
concern ourselves with these lags as long as we have estimates of the proportions of
assessments that will eventually be collected. These proportions, which we used to
translate assessments to eventual collections, vary by type of revenue-producing
enforcement program. For some collection programs, it was possible to derive a combined
proportion from a previous study of the individual net tax gap. In other cases, where
program-specific data were not available, the proportions were derived from management
information systems for the individual income tax. In the case of assessments and
collections that eventually arise from examinations, we based our estimates on the "interim
method" of estimating examination-related revenue described in Publication 1501.
Another problem is that IRS information systems frequently combine tax yields
with associated penalties and interest. To estimate the net tax gap, we must "back out"
these penalty and interest amounts. Since interest and penalties are not included in the
gross tax gap estimates, including them in the enforcement tax yield would understate the
resulting net tax gap estimates.
IRS Publication 1415, 19 April 1990
Net Tax Gap and Remittance Gap Estimates
17
IV. Remittance Gap
The gross legal-sector income tax gap estimates IRS published in March 1988
consisted of the tax liability reporting gaps of individuals and corporations. The estimates
did not include remittance gaps that occur when taxpayers report their tax liabilities, but the
full amount is not remitted voluntarily. In the case of the individual income tax, the
remittance gap has two main components: (a) insufficient remittance of tax balances due
when taxpayers file their returns; and (b) employer underpayment of income taxes they
withhold from wages.19 The latter component is part of the individual income tax
remittance gap since it represents income tax obligations which have been reported by
individuals but are not paid to the government. In the case of the corporate income tax,
only the first type of remittance gap applies.
Table 3 shows that, for TY 1987, the estimated gross remittance gap is $9.3 billion.
The largest component of this, an estimated $5.0 billion, is due to insufficient remittances
of individual income tax balances due at the time of filing. Estimated employer
underpayments of individual income tax withheld amount to $2.9 billion. The corporate
income tax remittance gap estimate is the smallest component at $1.4 billion. The total
gross remittance gap estimate grows from $3.6 billion in 1973 to $13.3 billion in 1992.
This growth and the growth of the net remittance gap were interrupted in 1982 as the
economic recession, and the tax rate cuts enacted in 1981, reduced income tax liabilities,
and, consequently, taxes due at the time of filing. A second interruption in the growth of
the gross and net gap totals was due to the Tax Reform Act of 1986, which lowered these
gaps for individuals by more than it raised them for corporations. Nonetheless, when
expressed as a percentage of total tax liability reported, the remittance gap declines slightly
from 1973 to 1992—from about 2.5 percent to about 2.2 percent of tax voluntarily
reported.20
The estimates of the gross remittance gap are based on special data extracts from
recent IRS masterfiles of individual and business tax records. For the portion due to
insufficient remittance of individual income tax balances due, the masterfile data were for
TY 1982 and TY 1983. Estimates for those years were extrapolated back to 1973 and
forward to 1992 using actual or projected values of total individual income tax due at the
time of filing. Actual values of tax due at the time of filing through 1985 came from IRS'
Statistics of Income (SOI) program. These values were extrapolated to 1992 using
projections of income tax to be reported on individual tax returns. Historically, the average
growth rate in the amounts of tax due at the time of filing is very close to the growth rate in
total tax liabilities reported by individuals. The estimates of employer underdeposit of
withholding are also based on a masterfile analysis for TY 1984 and TY 1985. Withheld
income tax is assumed, in this analysis, to have been not voluntarily paid if it was still
19
Payers of interest and dividends may also underdeposit some of the tax they are required to withhold under the
backup withholding provisions of the Interest and Dividend Tax Compliance Act of 1983. However, we have no
data on this element of noncompliance.
20 These percentages are for the total remittance gap, but the decline was similar for both the individual income
tax and the corporate income tax.
IRS Publication 1415, 19 April 1990
18
Net Tax Gap and Remittance Gap Estimates
Table 3. Legal-Sector Income Tax Remittance Gap Estimates
(in $ billions), Selected Tax Years, 1973-1992
1973
1976
1979
1981
1982
1984
1986
1987
1988
1992
Gross Remittance Gap
3.6
4.8
7.2
9.2
7.9
10.3
10.6
9.3
9.7
13.3
Individual income tax
3.0
4.1
6.4
8.3
7.2
9.3
9.5
7.9
8.3
11.4
1.9
2.6
4.1
5.2
4.2
5.8
5.3
5.0
5.3
7.5
Employer underdeposit
of withheld income tax 1 1.0
1.5
2.2
3.1
3.0
3.5
4.2
2.9
3.0
3.9
Insufficient remittance
by individuals
Corporate income tax
0.6
0.7
0.9
0.8
0.7
1.0
1.1
1.4
1.5
2.0
Net Remittance Gap
1.4
1.9
2.8
3.6
3.1
4.0
4.1
3.6
3.8
5.2
Individual income tax
1.2
1.6
2.5
3.3
2.8
3.7
3.7
3.2
3.3
4.5
0.8
1.1
1.8
2.3
1.8
2.5
2.3
2.2
2.3
3.2
Employer underdeposit
of withheld income tax 1 0.4
0.5
0.8
1.0
1.0
1.2
1.4
1.0
1.0
1.3
0.2
0.3
0.3
0.2
0.3
0.4
0.5
0.5
0.7
Insufficient remittance
by individuals
Corporate income tax
0.2
Some components do not add due to rounding.
1 Employers deposit withheld income taxes and social security taxes together. These figures represent only the
income tax portion of their underpayments.
overdue one month after the close of the quarter for which the employment tax return was
filed. For the corporate remittance gap, the masterfile analysis was for TY 1984 and TY
1985. The estimates for those years were extrapolated with methods similar to those used
for individuals.
Each of the components of the net remittance gap was estimated as a constant
proportion of the corresponding gross gap. The proportion used to estimate the net gap for
insufficient remittances of individual income tax was derived from an analysis of 1982-83
data pertaining to enforcement-related collections. The proportion for employer
underdeposits of withheld individual income tax was derived from an analysis of recent
accounts receivable data. The ratio of the net to the gross corporation income tax remittance
gap was assumed to be equal to the corresponding ratio for employer underdeposit of
withheld individual income tax.
IRS Publication 1415, 19 April 1990
This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.