Capital Gains Tax Gap: Requiring Brokers to Report Securities	 
Cost Basis Would Improve Compliance if Related Challenges Are	 
Addressed (13-JUN-06, GAO-06-603).				 
                                                                 
For tax year 2001, the Internal Revenue Service (IRS) estimated a
tax gap of at least $11 billion from individual taxpayers	 
misreporting income from capital assets (generally those owned	 
for investment or personal purposes). IRS did not estimate the	 
portion of this gap from securities (e.g., stocks, bonds, and	 
mutual fund capital gains distributions). GAO was asked for	 
information on (1) the extent and types of noncompliance for	 
individual taxpayers that misreport securities capital gains, (2)
actions IRS takes to reduce the securities tax gap, and (3)	 
options with the potential to improve taxpayer voluntary	 
compliance and IRS's ability to address noncompliant taxpayers.  
For estimates of noncompliance, GAO analyzed a probability sample
of examination cases for tax year 2001 from the most recent IRS  
study of individual tax compliance.				 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-06-603 					        
    ACCNO:   A55461						        
  TITLE:     Capital Gains Tax Gap: Requiring Brokers to Report       
Securities Cost Basis Would Improve Compliance if Related	 
Challenges Are Addressed					 
     DATE:   06/13/2006 
  SUBJECT:   Capital gains or losses				 
	     Noncompliance					 
	     Personal income taxes				 
	     Securities 					 
	     Taxpayers						 
	     Voluntary compliance				 
	     Tax gap						 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-06-603

     

     * Report to the Committee on Finance, U.S. Senate
          * June 2006
     * capital gains tax gap
          * Requiring Brokers to Report Securities Cost Basis Would Improve
            Compliance if Related Challenges Are Addressed
     * Contents
          * Results in Brief
          * Background
          * Individual Taxpayers Frequently Misreported Their Capital Gains
            or Losses from Securities Sales, Often Because They Misreported
            the Securities' Basis
               * Individual Taxpayers Frequently Misreported Their Capital
                 Gains or Losses from Securities Sales
               * Misreported Basis Was a Primary Type of Noncompliance That
                 Caused Taxpayers to Inaccurately Report Their Capital Gains
                 or Losses from Securities Sales
          * IRS Attempts to Reduce the Individual Capital Gains Tax Gap for
            Securities through Enforcement and Taxpayer Service Programs, but
            Various Challenges Limit Their Impact
               * IRS Attempts to Reduce the Individual Capital Gains Tax Gap
                 for Securities through Enforcement and Taxpayer Service
                 Programs
               * Various Challenges Limit the Impact IRS Programs Have on
                 Reducing the Individual Capital Gains Tax Gap for Securities
               * The Extent to Which IRS Enforcement Programs Have Reduced
                 the 2001 Capital Gains Tax Gap for Securities Is Not Known
          * Reporting of Cost Basis Could Reduce the Individual Capital Gains
            Tax Gap for Securities, but Implementation Challenges Would Need
            to Be Addressed
               * Increasing Examinations of Taxpayers with Securities Sales
                 Could Reduce That Portion of the Tax Gap but at the Expense
                 of Not Covering Other Areas of Noncompliance
               * Enhanced Taxpayer Services Might Improve Taxpayers'
                 Voluntary Compliance, but the Impact of Any Changes Would Be
                 Hard to Gauge
               * Information Reporting of Adjusted Basis Could Reduce the
                 Capital Gains Tax Gap for Securities
               * Expanding Basis Reporting Involves Implementation Challenges
                 That Would Need to Be Addressed
          * Conclusion
          * Matters for Congressional Consideration
          * Recommendations for Executive Action
          * Agency Comments and Our Evaluation
     * Scope and Methodology
     * Comments from the Internal Revenue Service
     * GAO Contact and Staff Acknowledgments
     * Related GAO Products

Report to the Committee on Finance, U.S. Senate

June 2006

CAPITAL GAINS TAX GAP

Requiring Brokers to Report Securities Cost Basis Would Improve Compliance
if Related Challenges Are Addressed

Contents

Tables

Figures

June 13, 2006Letter

The Honorable Charles Grassley Chairman The Honorable Max Baucus Ranking
Minority Member Committee on Finance United States Senate

Every year, a gap arises between the tax amount that taxpayers pay
voluntarily and on time and the amount they should pay under the law. For
tax year 2001, the Internal Revenue Service (IRS) estimated a gross tax
gap of $345 billion.1 IRS estimated that it would eventually recover $55
billion of the gross tax gap through late payments and IRS enforcement
efforts, leaving a net tax gap of $290 billion.2 The tax gap arises when
taxpayers fail to comply with the tax laws, whether intentionally or
unintentionally. Because of their noncompliance, the burden of funding the
nation's commitments falls more heavily on taxpayers who voluntarily and
accurately pay their taxes. In light of the size of the tax gap and the
nation's budget deficit, Congress has held several hearings seeking to
identify how the gap can be reduced. Given its size, even small or
moderate reductions in the net tax gap could yield substantial returns.

One type of noncompliance that contributes to the tax gap occurs when
individual taxpayers do not accurately report gains or losses from
transactions involving capital assets, which generally refers to property
owned for investment or personal purposes, on their tax returns. Taxpayers
generally determine their capital gains or losses by subtracting the
basis, which is generally the price they paid for an asset, from the gross
amount of proceeds they received from its sale. IRS estimated that for
2001, individual taxpayers' failure to accurately report their capital
gains income

accounted for at least $11 billion of the gross tax gap for that year.3
This amount is due to taxpayers understating their capital gains or
overstating their capital losses, both of which reduced the amount of
taxable income they reported. Although IRS has not estimated the amount of
the capital gains tax gap attributed to specific types of capital assets,
it has estimated that in recent years, securities transactions have
accounted for the majority of individuals' capital gains and losses.4
Securities transactions include the sale of securities-stocks, mutual
funds, bonds, and options-and capital gain distributions from mutual
funds.5 Securities transactions may be executed through third parties,
such as brokers.

To address your long-standing concerns about the tax gap, and particular
concern about the tax gap from individual capital gains tax noncompliance
for securities, this report responds to your request for information on
(1) the extent of and primary types of noncompliance that cause individual
taxpayers to misreport capital gains from securities, (2) actions IRS
takes in attempting to reduce the individual capital gains tax gap for
securities and any challenges that IRS faces with these actions, and (3)
options with the potential to improve taxpayers' voluntary compliance for
reporting securities gains and losses and IRS's ability to find
noncompliance related to the individual capital gains tax gap for
securities.

To provide information on the extent of and primary types of noncompliance
that cause individual taxpayers to misreport capital gains from
securities, we reviewed a probability sample of case files selected from
the nearly 46,000 randomly selected individual tax returns from tax year
2001 that IRS reviewed or examined through the National Research Program
(NRP), IRS's most recent study of individual taxpayer compliance. We used
the results of our case file review along with data from IRS's
examinations of the tax returns from NRP to make estimates for the entire
population of individual taxpayers. We present information on the extent
of noncompliance by estimating the percent of noncompliant taxpayers. We
did not estimate the portion of the capital gains tax gap specific to
securities. We could not provide a meaningful estimate of the tax gap for
securities because (1) of the 1,017 cases in our sample, we only received
849 complete cases by the time we completed our review, (2) the cases we
received included too few taxpayers who misreported securities
transactions (when selecting our sample, we could not determine which
cases included misreported gains and losses from securities as compared to
other types of capital assets), and (3) taxpayers misreported a wide range
of dollar amounts from the transactions. Since our estimates are based on
a sample, we express our confidence in our estimates as a 95 percent
confidence interval, plus or minus the margin of error indicated along
with each estimate in the report, which is the interval that would contain
the actual population value for 95 percent of the samples we could have
selected. To address the question of what actions IRS takes in attempting
to reduce the individual capital gains tax gap for securities and related
challenges, we reviewed documents from IRS's enforcement programs and IRS
publications that address capital gains. We also interviewed IRS officials
knowledgeable about the subject. To identify options with the potential to
improve taxpayers' voluntary compliance for reporting securities gains and
losses and IRS's ability to find noncompliance related to the individual
capital gains tax gap for securities, we reviewed our prior reports,
documents from IRS's enforcement programs, IRS publications that address
capital gains, and industry reports on securities holdings and information
reporting.6 We also spoke with IRS officials and representatives related
to the securities industry. For further discussion of our scope and
methodology, see appendix I. We conducted our review from June 2005
through May 2006 in accordance with generally accepted government auditing
standards.

Results in Brief

For tax year 2001, an estimated 38 percent of individual taxpayers who had
securities transactions failed to accurately report their capital gains or
losses from the transactions (8.4 million out of 21.9 million taxpayers),
often because they misreported the securities' cost basis. A greater
percentage of taxpayers are estimated to have misreported gains or losses
from their securities sales (36 percent) than misreported their capital
gain distributions from mutual funds (13 percent). One reason for this
difference could be because taxpayers must determine what portion of
income from securities sales is taxable whereas taxpayers need only add up
their capital gain distributions and enter the amounts on their tax
returns. We were not able to determine the total amount of capital gains
income from securities that taxpayers misreported or the securities tax
gap because the cases we reviewed included too few misreported securities
transactions and we did not receive other cases in time to include them in
our review. However, we found that around half of taxpayers who did not
accurately report their securities sales were estimated to have
misreported at least $1,000 of capital gains or losses. Also, around half
of the taxpayers who misreported their gains or losses from securities
sales did so because they failed to accurately report the securities'
basis, sometimes because they did not know the securities' basis or failed
to take certain events into account that required them to adjust the basis
of their securities. Additionally, around 9 percent of taxpayers with
securities sales misreported whether their gains or losses were short-term
or long-term.

IRS attempts to reduce the individual capital gains tax gap for securities
through enforcement and taxpayer service programs; however, various
challenges limit the impact these programs have on reducing this tax gap.
IRS uses enforcement programs to contact selected taxpayers it believes
may have inaccurately reported capital gains or losses. IRS's automated
enforcement programs largely rely on matching tax returns filed by
taxpayers to information returns provided by brokers that report
taxpayers' gross proceeds from securities sales. IRS also examines tax
returns by reviewing taxpayers' records of their securities transactions.
Additionally, IRS offers various taxpayer services intended to help
taxpayers comply with capital gains tax obligations, such as publications
describing how to determine tax liabilities from selling securities. The
challenges that limit the impact these programs have on reducing the tax
gap for securities include the relatively small portion of taxpayers with
securities transactions that IRS contacts through its enforcement programs
and the lack of information on the basis of securities sold, which IRS
needs to verify most gains or losses, and the difficulty in communicating
capital gains tax reporting requirements. Although IRS assesses additional
taxes through its enforcement programs, neither IRS nor we know the extent
to which these assessments reduced the 2001 capital gains tax gap for
securities, in part because the gap itself is not known.

Expanding information reporting on securities sales to include cost basis
has potential to improve taxpayers' voluntary compliance and help IRS find
noncompliance related to the capital gains tax gap for securities. On one
hand, some of this potential exists if IRS were to change its enforcement
and taxpayer service programs, such as by examining more tax returns or
enhancing guidance related to securities gains or losses. However,
examinations can be costly, and taxpayers may not know about or use the
guidance. On the other hand, basis reporting to taxpayers and IRS would
help taxpayers to voluntarily comply and would reduce their burden in
computing capital gains or losses. IRS research has repeatedly shown that
taxpayers' compliance is strongly related to the extent to which their
income is subject to information reporting. Basis reporting also would
provide information to help IRS verify securities gains or losses and
target enforcement resources to noncompliant taxpayers. However, such
basis reporting would raise challenges and trade-offs. Many of the
challenges can be mitigated to some extent. For example, tracking and
reporting basis would increase brokers' costs, but decisions about the
scope and details of the reporting could constrain the increase. Further,
taxpayers' costs would be reduced, and many brokers already provide some
form of basis information to taxpayers. The challenges arising when
brokers do not know the basis for securities purchased in the past could
be mitigated by only reporting basis for future purchases, which would
somewhat delay the full impact that basis reporting would have on reducing
the capital gains tax gap. Although IRS has broad authority to require
information reporting for securities sales, it may not have the authority
to require all of the actions that would be needed to implement cost basis
information reporting.

This report includes matters Congress may want to consider to reduce the
capital gains tax gap for securities. Specifically, Congress could require
brokers to report to both taxpayers and IRS the adjusted basis of
securities that taxpayers sell and whether the gains or losses were short-
or long-term, and direct IRS to work with brokers and related parties to
develop rules that seek to cost effectively mitigate some of the
challenges associated with requiring basis reporting. We are also making
recommendations to IRS on clarifying guidance related to reporting capital
gains and losses. In commenting on a draft of this report, the
Commissioner of Internal Revenue agreed with our recommendations.

Background

Individual taxpayers generally realize gains or losses when they sell
capital assets, which are generally defined as properties owned for
investment or personal purposes and outside the normal course of a
taxpayer's trade or business. In recent years, IRS studies show that the
majority of capital asset transactions and capital gains and losses were
for securities transactions, including sales of corporate stock, mutual
funds, bonds, options,7 and capital gain distributions from mutual funds.8
For example, in 1999, the latest year for which IRS published data on
capital assets sales, an estimated 91 percent of capital asset
transactions, 62 percent of capital gains, and 79 percent of capital
losses were from securities transactions.9 Also, over the past two
decades, individual ownership of securities assets held outside of
retirement accounts has increased.10 According to the Federal Reserve
Board, the percentage of families that own stock, mutual funds, and bonds
outside of retirement accounts increased from 25 percent in 1983 to a high
of 42 percent in 2001, before falling to 38 percent in 2004.11

When taxpayers sell or otherwise receive income from securities, they must
report the transactions on their federal income tax returns. For
securities sales, taxpayers are to report the dates they acquired and sold
the asset; sales price, or gross proceeds from the sale; cost or other
basis of the sold asset; and resulting gains or losses on Schedule D to
the individual tax return-Form 1040. Taxpayers are to report this
information separately for short-term transactions and long-term
transactions. Taxpayers also are to report the total amount of their
capital gain distributions from mutual funds, which are always considered
to be long-term transactions. Taxpayers are to report their overall gains
or losses from securities sales, capital gain distributions, and other
capital gains on the Form 1040 tax return itself.

Generally, a taxpayer's gain or loss from a securities sale is simply the
difference between the gross proceeds from the sale and the original
purchase price, or original cost basis.12 However, before taxpayers can
determine any gains or losses from securities sales, they must determine
if and how the original cost basis of the securities must be adjusted to
reflect certain events, such as stock splits, nontaxable dividends, or
nondividend distributions. For example, figure 1 shows how a taxpayer
would need to adjust the basis of a stock following a stock split to
accurately determine the resulting capital gain or loss when the stock is
sold. In this example, if the taxpayer fails to properly adjust the basis
of the stock to account for the split, he or she will incorrectly report a
capital loss from the sale.

Figure 1: Example of How Failure to Adjust Basis Can Lead to Misreporting
a Capital Gain or Loss from a Securities Transaction

Taxpayers who buy and sell the same stock or mutual fund shares at various
times can determine basis in a number of ways. Taxpayers can specifically
identify the groups of shares they want to sell. For example, if a
taxpayer buys a group of 10 shares of stock in one year for $1 per share
and another group of 10 shares of the same stock in the next year for $2
per share, and then sells 10 shares of the stock, the taxpayer can choose
to sell the stocks with either the $1 or $2 cost basis.13 If taxpayers
cannot identify which shares they sold among many they bought on varying
occasions, they must report the basis of the securities they purchased
first as the basis of the sold shares. Except for mutual fund shares,
taxpayers cannot use the average cost of securities they purchased at
various times to determine basis.

When taxpayers sell securities through a broker, that broker is required
to file Form 1099-B with IRS and the taxpayers to report a description of
the security, sales date, number of shares sold, and gross proceeds from
the sale, along with other information.14 Brokers are not required to
report the cost or other basis of the sold security or, with the exception
of regulated futures contracts, the resulting gain or loss from a security
sale. Capital gain distributions from mutual funds are to be reported on
Form 1099-DIV.15

The rate at which income from securities is taxed depends on how long
taxpayers held a security before sale and taxpayers' regular income tax
rates. Securities assets sold after being held for 1 year or less are
considered short-term and taxed at the taxpayers' regular income tax
rates. Assets sold after being held for more than 1 year are considered to
be long-term and are generally taxed at maximum rates of 5 percent or 15
percent, depending on the taxpayer's regular income tax rates.16 Capital
gain distributions from mutual funds are always taxed as long-term gains.
Taxpayers can deduct capital losses against their capital gains, and any
excess losses can be deducted against ordinary income up to a limit of
$3,000 ($1,500 for married taxpayers filing separately), beyond which the
losses can be carried over to offset capital gains or ordinary income in
future tax years.

Individual Taxpayers Frequently Misreported Their Capital Gains or Losses
from Securities Sales, Often Because They Misreported the Securities'
Basis

Thirty-eight percent of individual taxpayers who had securities
transactions misreported their securities gains or losses for tax year
2001. A greater percentage of taxpayers misreported their securities sales
(36 percent) than misreported their capital gain distributions (13
percent), and most of the misreported securities transactions exceeded
$1,000 of capital gain or loss. Taxpayers often misreported their capital
gains or losses from securities sales because they failed to accurately
report the securities' basis.

Individual Taxpayers Frequently Misreported Their Capital Gains or Losses
from Securities Sales

For tax year 2001, individual taxpayers frequently misreported their
capital gains or losses from the securities they sold. Overall, an
estimated 8.4 million of the estimated 21.9 million taxpayers with
securities transactions misreported their gains or losses.17 Table 1 shows
the estimated percentages of taxpayers who misreported their securities
sales and capital gain distributions, overall and by the securities'
holding period.

Table 1: Estimated Percentage of Individual Taxpayers with Securities
Transactions Who Misreported the Gain or Loss from One or More
Transaction, Tax Year 2001

                                        

                       Estimated percentage of     
                      taxpayers who misreported    
                          their transactions       
       Type of             Short-term transactions    Long-term           All 
     transaction                                   transactions transactionsa 
Securities sales                             28           31            36 
Capital gain                               N/Ab           13            13 
distributions                                                
All securitiesc                              28           32            38 

Source: GAO analysis of IRS data and examination case files.

Note: Percentage estimates have sampling errors of (+/-) 7 percent or
less.

aFor securities sales, "all transactions" includes those for which we
could not determine whether the holding period was short-term or
long-term.

bCapital gain distributions are always considered long-term transactions.

c"All securities" includes taxpayers who misreported both securities sales
and capital gain distributions.

Table 1 shows that a higher estimated percentage of taxpayers misreported
a securities sale than a capital gains distribution. Overall, an estimated
7.3 million out of an estimated 20.3 million taxpayers misreported their
securities sales compared to the estimated 1.2 million out of an estimated
9.1 million taxpayers who misreported their capital gain distributions.18
One reason taxpayers may misreport securities sales more frequently is
that taxpayers must compute the portion of their sales proceeds that
constitutes a gain or loss, whereas taxpayers need only add up their
capital gain distributions from information returns they receive and enter
the amounts on their tax returns. Table 1 also shows that individual
taxpayers are estimated to have misreported their short-term securities
sales about as often as their long-term sales. In addition, our analyses
showed the following:

o Of those taxpayers who misreported securities sales, an estimated 97
percent misreported gains or losses from the sales of stocks and mutual
funds while an estimated 5 percent misreported bonds, options, or
futures.19

o Individual taxpayers misreported securities sales more frequently than
other types of income, such as wages and salary, dividend income, and
interest income. Respectively, an estimated 10 percent, 17 percent, and 22
percent of taxpayers with these types of income misreported the income.20

We were not able to estimate the capital gains tax gap for securities
because the cases we reviewed included too few misreported securities
transactions and taxpayers misreported a wide range of dollar amounts from
the transactions, among other reasons (see app. I). However, we were able
to determine the direction of the misreporting. For securities sales, an
estimated 64 percent of taxpayers underreported their income from
securities (i.e., they understated gains or overstated losses) compared to
an estimated 33 percent of taxpayers who overreported income (i.e., they
overstated gains or understated losses).21 For both underreported and
overreported income, some taxpayers misreported over $400,000 in gains or
losses. Also, as shown in table 2, around half of taxpayers who did not
accurately report their securities sales were estimated to have
misreported at least $1,000 of capital gains or losses (that is, taxpayers
not in the less than $1,000 categories).22

Table 2: Distribution of the Estimated Amount of Net Misreported Capital
Gains Income From Securities Sales by Misreporting Taxpayers, Tax Year
2001

                                        

       Net misreported amount            Percentage of misreporting taxpayers 
Overreporting taxpayers         
Less than $1,000a                                                       19 
$1,000 to $9,999                                                        15 
$10,000 and greater                                                      5 
Underreporting taxpayers        
Less than $1,000                                                        27 
$1,000 to $9,999                                                        19 
$10,000 and greater                                                     14 

Source: GAO analysis of IRS data and examination case files.

Notes: Percentage figures do not sum to 100 because of rounding.
Percentage estimates have sampling errors of (+/-) 8 percent or less.

aCategory includes taxpayers that misreported securities sales in a way
that had no effect on the gain or loss from the sales.

In terms of income levels, the distribution of taxpayers who misreported
gains or losses from securities sales and capital gain distributions did
not vary greatly from the income level for all individual taxpayers for
tax year 2001, as shown in table 3.

Table 3: Estimated Distribution of Individual Taxpayers Who Misreported
Capital Gains or Losses from Securities Transactions and All Individual
Taxpayers by Adjusted Gross Income, Tax Year 2001

                                        

     Adjusted gross    Percentage of misreporting           Percentage of all 
         income                         taxpayers        individual taxpayers 
Less than $25,000                           51                          46 
$25,000 to $49,999                          20                          25 
$50,000 to $99,999                          20                          20 
$100,000 or greater                          9                           9 

Source: GAO analysis of IRS data and examination case files.

Notes: For misreporting taxpayers, estimates have sampling errors of (+/-)
8 percent or less. For all individual taxpayers, estimates have sampling
errors of (+/-) 0.3 percent or less.

Misreported Basis Was a Primary Type of Noncompliance That Caused
Taxpayers to Inaccurately Report Their Capital Gains or Losses from
Securities Sales

Based on information in the files we reviewed, a primary type of
noncompliance that caused taxpayers to inaccurately report their capital
gains or losses from securities sales in tax year 2001 was misreporting
the basis of the securities they sold. Table 4 shows the estimated
frequency of the types of noncompliance that caused taxpayers to misreport
capital gains or losses from their securities sales.23

Table 4: Estimated Frequency of Types of Noncompliance That Caused
Individual Taxpayers to Misreport Capital Gains or Losses from Securities
Sales, Tax Year 2001

                                        

         Type of noncompliance           Estimated percentage of misreporting 
                                                                    taxpayers 
Misreported basis of security sold                                      49 
Failed to report sale                                                   44 
Misreported sale proceeds                                               12 
Misclassified holding period                                             9 
Other                                                                    9 

Source: GAO analysis of IRS data and examination case files.

Notes: Estimates in this table do not include the results of our review
for four cases where we could not determine the type of noncompliance that
caused taxpayers to misreport securities sales. The "Other" category
includes taxpayers who misclassified capital income as other types of
income or vice versa or made mathematical errors. Some taxpayers
misreported more than one security sale or misreported a sale because of
more than one type of noncompliance. Percentage estimates have sampling
errors of (+/-) 9 percent or less.

For taxpayers who misreported basis, a greater percentage failed to
accurately report basis for long-term securities holdings (35 percent of
taxpayers who misreported securities sales) than for short-term holdings
(21 percent).24 Taxpayers who failed to report securities sales altogether
did not report short-term and long-term securities sales at a similar rate
(20 percent and 22 percent, respectively, of taxpayers who misreported
securities sales).25

Although we were able to determine the percentage of taxpayers who failed
to accurately report their securities sales because they misreported basis
(49 percent), we could not develop reliable estimates on the reasons for
this type of misreporting because most of the NRP examination case files
did not provide sufficiently descriptive information. However, of the 133
taxpayers who misreported basis from the 849 case files we reviewed, we
were able to determine that 32 taxpayers misreported basis for the
following reasons:

o Taxpayers did not have records of their securities purchases (16
taxpayers). Although during examinations, IRS was able to obtain basis
records for some of these taxpayers from their brokers, for 9 taxpayers,
basis records could not be obtained. For these taxpayers, IRS examiners
considered basis to be zero and treated all gross proceeds amounts as
capital gains.

o Taxpayers used original cost basis instead of adjusted cost basis (6
taxpayers).

o Taxpayers did not understand how to determine basis (5 taxpayers).

o Taxpayers reported basis information that was incorrectly determined by
a tax return preparer (4 taxpayers).26

o One taxpayer reported inaccurate basis information provided by a broker.

Of taxpayers who failed to report their securities sales altogether, an
estimated 28 percent were estimated to have failed to report capital
losses.27 By not reporting losses, these taxpayers potentially failed to
offset other capital gains or deduct their losses against other types of
income they reported. Likewise, some of these taxpayers who failed to
report capital losses exceeding $3,000 did not carry over these losses to
offset capital gains or other income in future tax years. Although in most
cases we could not determine why taxpayers did not report these losses,
some taxpayers told IRS examiners that they did not know they had to
report losses. In addition, IRS officials said some taxpayers might not
report their capital losses because they worry that their returns will be
examined if they overstate their losses. Also, the officials told us that
taxpayers might want to avoid the burden of filing a Schedule D or the
cost of paying someone to prepare their returns in cases where filing
Schedule D would make the difference between self preparing and using a
paid preparer.

As also shown in table 4, taxpayers failed to accurately report their
securities sales because they misreported the amount of their sale
proceeds (12 percent) or misclassified the securities' holding period (9
percent). However, the case files did not contain enough information to
explain why taxpayers made these errors. Also, the responsible officials
we interviewed at IRS could not provide explanations for why taxpayers
might have made these errors.

IRS Attempts to Reduce the Individual Capital Gains Tax Gap for Securities
through Enforcement and Taxpayer Service Programs, but Various Challenges
Limit Their Impact

IRS uses both enforcement and taxpayer service programs in attempting to
reduce the individual capital gains tax gap for securities. IRS checks the
accuracy of tax returns through its enforcement programs and contacts
taxpayers who may have inaccurately reported their securities gains or
losses. IRS also offers service programs to provide taxpayers with
assistance in fulfilling their capital gains tax obligations. However,
these programs face challenges that limit their impact on reducing the
capital gains tax gap for securities. Although IRS assesses additional
taxes for securities income through its enforcement efforts, neither IRS
nor we know the extent to which these assessments reduced the 2001 capital
gains tax gap for securities.

IRS Attempts to Reduce the Individual Capital Gains Tax Gap for Securities
through Enforcement and Taxpayer Service Programs

Consistent with its overarching philosophy that a combination of
enforcement and service efforts are essential to tax compliance, IRS
attempts to reduce the individual capital gains tax gap for securities
through its programs that enforce the tax laws and that seek to help
taxpayers voluntarily comply with the laws. IRS uses its enforcement
programs to check the accuracy of filed tax returns and contacts taxpayers
who have potentially made errors or inaccurately reported capital gains

information on their returns.28 Aspects of IRS's enforcement programs
related to capital gains income for securities appear in table 5.29

Table 5: IRS Enforcement Programs and Types of Securities Capital Gains
Tax Noncompliance They Can Detect

                                        

             IRS program                      Capable of detecting            
Math Error                      Data reported inconsistently between       
                                   Schedule D and Form 1040                   
Automated Underreporter (AUR)   Inaccurately reported gross proceeds from  
                                   securities sales and capital gain          
                                   distributions                              
Automated Substitute for Return Taxpayers who received proceeds from       
(ASFR)                          securities sales but did not file tax      
                                   returns                                    
Examination                     All forms of capital gains noncompliance   
                                   for securities                             

Source: IRS.

Math Error, AUR, and ASFR are automated enforcement programs. IRS uses the
Math Error program to check filed tax returns for internal inconsistencies
or mathematical errors, and contacts taxpayers, including when the errors
result in a tax change. Through AUR, IRS computers match the amounts of
capital gains proceeds that taxpayers report on their tax returns and that
brokers report on information returns. If this matching indicates that
taxpayers may have underreported their sale proceeds for securities and
IRS cannot resolve the discrepancies based on available information, IRS
may send notices asking taxpayers to explain the discrepancies or pay any
taxes assessed. When IRS determines through ASFR that taxpayers for whom
IRS received information returns on the sale proceeds for securities
failed to file tax returns, it may create tax returns for the taxpayers
and assess tax liabilities.

During examinations, IRS uses information from third parties as well as
from taxpayers to determine if taxpayers have accurately reported their

capital gains or losses.30 Examiners also may use other resources, such as
online services, to help them determine the basis of taxpayers'
securities. IRS assesses additional taxes if it determines that taxpayers
have underreported their capital gains income from securities.

IRS's taxpayer service programs provide taxpayers with information,
support, and assistance to help them understand and fulfill their capital
gains tax obligations for securities. For example, IRS produces
publications that explain how to report capital gains or losses and
provide examples of how to determine adjusted basis.31 IRS also provides
Web-based information and telephone, written, or face-to-face assistance
at Taxpayer Assistance Centers on how to accurately report capital gains
and losses.

Various Challenges Limit the Impact IRS Programs Have on Reducing the
Individual Capital Gains Tax Gap for Securities

IRS's enforcement and taxpayer service programs face limitations in
reducing the individual capital gains tax gap for securities. In addition
to resource constraints that limit how many cases of potential
noncompliance are pursued, table 6 summarizes the main limitations each
program faces.

Table 6: IRS Enforcement and Taxpayer Service Programs and Their Principle
Limitations on Reducing the Individual Capital Gains Tax Gap for
Securities

                                        

      IRS program                     Principle limitations                   
Math Error        Not intended to verify if taxpayers have accurately      
                     reported their capital gains tax liabilities for         
                     securities                                               
AUR               Lack of basis information from brokers prevents AUR from 
                     verifying the accuracy of reported capital gains or      
                     losses from securities sales                             
ASFR              Lack of basis information from brokers prevents AFSR     
                     from accurately determining how much of taxpayers' gross 
                     proceeds from securities sales is taxable                
Examination       Capital gains are too complex and time consuming to      
                     examine through correspondence                           
                                                                              
                     Face-to-face examinations are resource intensive and     
                     cover a small percentage of taxpayers with capital gains 
Taxpayer services Taxpayers may not use the services                       
                                                                              
                     Taxpayers may not understand information IRS provides    

Source: IRS.

As table 6 shows, IRS cannot use its automated programs to fully verify
the reported capital gains or losses from securities sales because it does
not receive basis information from brokers. Also, according to IRS
officials, a lack of basis information reduces productivity because IRS
spends resources contacting taxpayers for whom it ultimately does not
assess additional taxes. For example, for tax year 2002, the latest year
for which IRS has complete data, IRS did not assess additional taxes for
around 46 percent of the taxpayers it contacted through AUR to address
potentially misreported securities sales.32 By comparison, this "no tax
change" percentage was around 20 percent for AUR contacts for all other
types of income for 2002.33 For ASFR, IRS officials said that the lack of
basis information hampers IRS's ability to determine which taxpayers with
gross proceeds from securities sales should have filed tax returns and to
productively pursue those taxpayers who did not file.

Given that IRS does not receive basis information from brokers, it can
only verify the accuracy of the basis and gains and losses that taxpayers
report for their securities sales by examining these individuals' tax
returns. IRS does not examine these taxpayers' returns through
correspondence because it believes the returns are too difficult and would
take too much time to examine. IRS can only verify the accuracy of the
reported basis and gains and losses from securities sales through
face-to-face examinations. However, these examinations are resource
intensive and only cover a small percentage of individual taxpayers. For
example, in fiscal year 2004, IRS conducted approximately 200,000
face-to-face examinations34 for the 130 million individual taxpayers that
filed tax returns in 2003, including the estimated 22.7 million taxpayers
that filed a Schedule D with their tax returns.35 Even when IRS selects
individual taxpayers to examine face-to-face, IRS often places a greater
focus on issues it believes are more productive than securities sales,
such as business income or the sale of personal or business real property,
according to an IRS official responsible for examination planning.

In providing taxpayer services, IRS faces challenges in communicating
information to taxpayers on complying with capital gains reporting
requirements. Taxpayers may not use the services IRS offers or may not
understand the information that IRS provides. For example, IRS recently
changed the instructions for filing Schedule D to include language that
specifies taxpayers must include the details of all their capital gains
transactions when filing their tax returns. Although IRS included this
language to clarify an existing reporting requirement, some taxpayers and
tax practitioners perceived that the instructions required taxpayers to
report each capital asset transaction on Schedule D itself and not on
attached brokerage statements, as otherwise allowed. This misconception
required IRS to clarify on its Web site that taxpayers could continue to
report the details of their transactions on attached statements as long as
all transactions were included and they reported aggregate information on
Schedule D.

The Extent to Which IRS Enforcement Programs Have Reduced the 2001 Capital
Gains Tax Gap for Securities Is Not Known

Through its enforcement programs, IRS assessed additional taxes for
taxpayers who misreported their securities gains and losses for tax year
2001; however, neither IRS nor we know the extent to which these
assessments reduced the securities tax gap for that year. IRS has not
estimated the portion of the capital gains tax gap attributed to
securities for tax year 2001, and we were not able to estimate this
portion of the tax gap from our review of NRP case files. Likewise, IRS
does not have complete information on the amount of additional taxes it
assessed for taxpayers who underreported their income from securities
sales for 2001.

Through AUR for tax year 2001, IRS assessed around $190 million in
additional taxes for securities sales and around another $5 million for
capital gain distributions, and refunded over $8 million to taxpayers who
overreported securities income.36 For tax year 2001 examinations, IRS does
not have complete data for the amount of taxes it assessed for misreported
capital gains or losses. IRS maintains a database that tracks examination
results by the type of issue examined, such as capital gains or losses.
However, prior to October 2004, the database only captured examination
results for around 60 percent of individual examinations, according to IRS
officials.37As such, the database does not include all capital gains
noncompliance that IRS identified in tax year 2001 examinations. Even when
it includes such noncompliance, the database does not distinguish between
misreported capital gains income from securities versus other capital
assets. Likewise, the database does not specify the portion of additional
tax assessments that is attributable to misreported capital gains

income versus other types of noncompliance.38 Finally, IRS does not
maintain data on additional taxes assessed and collected because of
capital gains noncompliance through the Math Error or ASFR programs.

Reporting of Cost Basis Could Reduce the Individual Capital Gains Tax Gap
for Securities, but Implementation Challenges Would Need to Be Addressed

Expanded reporting of cost basis information has the potential to reduce
the individual capital gains tax gap for securities. Making administrative
changes to IRS's compliance programs that address capital gains also has
some potential to reduce the tax gap, but enforcement programs can be
resource intensive and taxpayers do not always use IRS's taxpayer service
programs. With such limitations, these changes likely would not
significantly boost taxpayers' voluntary compliance involving securities
sales. Information reporting of adjusted cost basis to taxpayers and IRS
would likely help reduce the tax gap from securities sales by improving
taxpayers' voluntary compliance and IRS's ability to cost effectively
address noncompliant taxpayers. Consistent reporting of basis information
would involve challenges that would need to be, and to some extent can be,
mitigated.

Increasing Examinations of Taxpayers with Securities Sales Could Reduce
That Portion of the Tax Gap but at the Expense of Not Covering Other Areas
of Noncompliance

IRS could seek to reduce the capital gains tax gap for securities by
increasing examination coverage of taxpayers with gains or losses from
securities, either by considering them when selecting taxpayers to examine
through correspondence or by increasing face-to-face examinations of these
taxpayers. Conducting more of each type of examination could increase the
amount of taxes assessed for misreporting securities income. However,
absent an increase in resources or access to basis information, which
would help IRS better target its resources toward truly noncompliant
taxpayers, focusing on taxpayers with securities gains or losses would
divert IRS's examination resources away from other productive areas of
noncompliance, according to IRS officials. An increased focus on
securities sales could reduce the capital gains tax gap, but a diversion
of resources could result in greater noncompliance for other types of
income. Moreover, although increasing examination coverage could induce
taxpayers who are misreporting willfully to voluntarily comply, expanded
coverage would not significantly affect voluntary compliance for taxpayers
who make mistakes while trying to comply, such as taxpayers who made
errors calculating basis, according to an IRS research official who has
studied the impact of enforcement on taxpayer compliance.

Enhanced Taxpayer Services Might Improve Taxpayers' Voluntary Compliance,
but the Impact of Any Changes Would Be Hard to Gauge

Addressing capital gains tax noncompliance for securities sales by
enhancing IRS's taxpayer service efforts might improve taxpayers'
voluntary compliance by helping them to better understand and fulfill
their capital gains tax obligations for securities. However, the effects
of any additional guidance that IRS might develop, for example on
reporting losses or on resources for determining basis, would be tempered
by challenges similar to those previously discussed, such as taxpayers not
using or understanding information IRS provides. Although IRS attempts to
generally ensure tax compliance through its service efforts, IRS
researchers have found it difficult to determine the extent to which
taxpayer services improve compliance among taxpayers who want to comply.
As such, it is hard to know if these improvements to IRS's service efforts
would have a substantial impact on taxpayer's reporting compliance for
securities sales.

Regardless, IRS's instructions for reporting capital gains and losses and
related guidance do not contain some information related to the causes for
taxpayers misreporting the basis of securities they sold or failing to
report sales at all-the leading types of noncompliance when taxpayers
erred in reporting capital gains and losses. In many cases, we could not
determine and IRS did not know exactly why taxpayers made these errors.
However, some taxpayers did not know they had to report gains or losses
and others did not understand how to determine basis. One counterintuitive
situation existed among the cases we reviewed, that is some taxpayers did
not report losses, which generally help them by lowering their tax
liabilities. IRS's instructions for filing Schedule D direct taxpayers to
report their capital gains or losses but the instructions do not clarify
the appropriate use of capital losses to offset capital gains or other
income. Further, although IRS provides guidance on how to calculate basis
for a variety of securities transactions, the instructions to Schedule D
do not contain guidance on resources available to taxpayers and tax
practitioners to determine basis for securities. Some examples of
resources taxpayers might use to determine the basis of their securities
holdings include brokers, tax preparers, or Web sites for companies that
issue stocks or other information. Providing taxpayers more information on
the benefits of reporting losses and resources available to them on
calculating basis would be consistent with IRS's responsibility to ensure
that taxpayers pay the right amount of tax. Further, compared to other
steps such as enforcement actions, providing additional guidance to
taxpayers would be a low cost option to potentially increase their capital
gains reporting compliance. Finally, any improvement in taxpayers'
compliance due to better guidance would reduce IRS's enforcement expenses
related to capital gains.

Information Reporting of Adjusted Basis Could Reduce the Capital Gains Tax
Gap for Securities

According to IRS officials and some representatives related to the
securities industry, taxpayers would likely report their gains or losses
from securities sales more accurately and at a reduced burden if brokers
consistently provided them with the adjusted basis of the securities they
sold. Likewise, basis reporting would allow IRS to verify taxpayers'
securities gains and losses through its automated enforcement programs and
take more efficient enforcement actions to address noncompliant taxpayers,
according to IRS compliance officials. The likely increase in taxpayers'
voluntary compliance and in the productiveness of IRS enforcement actions
resulting from basis reporting would likely substantially reduce the
capital gains tax gap for securities.

Taxpayers would benefit from basis reporting because, in many cases, they
would not have to track and compute the adjusted basis of the securities
they sold. Therefore, basis reporting would likely reduce the chance that
taxpayers who had not been tracking their adjusted basis would misreport
it for securities they sold. Also, if taxpayers received basis information
from their brokers for the securities they sold, they would enjoy a
reduced burden in filing Schedule D with their tax returns because, in
many cases, they would not need to make basis calculations on their own.

For taxpayers, the greater accuracy and reduced burden of reporting basis
that would result from basis reporting would likely improve their
voluntary compliance. As shown in figure 2, taxpayers tend to accurately
report income that third parties report on information returns because the
income is transparent to taxpayers as well as to IRS. For example,
individual taxpayers misreport nearly twice the percentage of their income
from sources subject only to some information reporting, which is the case
with income from securities sales now, compared to income subject to
substantial information reporting, such as income from dividends and
interest, and which would be close to the case for securities sales if
basis were consistently reported, according to an IRS research official.
Also, as discussed previously, based on our file review, taxpayers were
much less likely to misreport capital gain distributions (13 percent),
which are similar to dividends and are subject to substantial information
reporting, compared to income from securities sales (36 percent), for
which information reporting only covers gross proceeds but not cost basis.
The smallest percentage of misreporting is for wage and salary income, for
which substantial information reporting exists and taxes are withheld by
taxpayers' employers.

Figure 2: Individual Net Income Misreporting Categorized by the Extent of
Income Subject to Information Reporting, Tax Year 2001

Cost basis reporting would also benefit IRS, to the extent the reporting
was complete and accurate. IRS could use basis information to verify
securities gains and losses through its automated enforcement programs and
could more effectively allocate its enforcement resources to focus on the
most noncompliant taxpayers. For AUR and ASFR, IRS officials told us that
basis information would allow it to more precisely determine taxpayers'
income for securities sales and would allow it to identify which taxpayers
who misreported securities income have the greatest potential for
additional tax assessments. IRS's examination program could similarly
benefit. Specifically, IRS officials told us that receiving cost basis
information might enable IRS to examine noncompliant taxpayers through
correspondence because it could productively select tax returns to
examine. Also, having cost basis information could help IRS identify the
best cases to examine face-to-face, making the examinations more
productive while simultaneously reducing the burden imposed on compliant
taxpayers who otherwise would be selected for examination. As a result of
all these benefits, basis reporting would allow IRS to better allocate its
resources that focus on securities misreporting across its enforcement
programs.

IRS has endorsed the concept of matching information returns to tax
returns for the purpose of identifying unreported income since the 1960s
and Congress has created a number of statutes requiring information
reporting for various types of income or taxpayer information.39 The
related GAO products section at the end of this report provides references
to selected GAO reports related to information reporting.

We previously discussed the notion of basis reporting to help reduce
capital gains tax noncompliance in our May 1994 report on the tax gap.40
Also, based on discussions we had with officials from IRS's Taxpayer
Advocate Service when we initiated our review, the National Taxpayer
Advocate recommended that brokers be required to track and report cost
basis for stocks and mutual funds in her 2005 Annual Report to Congress.41
In March 2006 a bill was introduced in the U.S. Senate and in April and
May 2006 bills were introduced in the House of Representatives that would
require brokers to report taxpayers' basis for their securities
transactions.42

Expanding Basis Reporting Involves Implementation Challenges That Would
Need to Be Addressed

Expanding information reporting on securities sales to include basis
information would involve challenges for brokers and IRS. There are
various ways to mitigate each challenge. Tables 7 and 8 list some major
challenges for brokers and IRS, respectively, as well as some ways to
start mitigating the challenges. Discussion after the tables covers some
issues to consider when evaluating these mitigation strategies.

Table 7: Challenges to Brokers Associated with Basis Reporting and How the
Challenges Could Be Mitigated

                                        

          Challenges to brokers              Ways to mitigate challenges      
Implementing systems to track and    o Although the following do not       
report basis involves monetary costs directly mitigate costs for all       
                                        brokers,                              
                                                                              
                                        o Many brokers and mutual funds       
                                        already track and report basis to     
                                        many taxpayers, which could help form 
                                        a foundation for expanded basis       
                                        reporting                             
                                                                              
                                        o Brokers could leverage existing     
                                        systems that track and report gross   
                                        proceeds to taxpayers and IRS         
                                                                              
                                        o Congress or IRS could provide an    
                                        appropriate effective date that would 
                                        allow brokers that lack such systems  
                                        to develop them                       
Brokers may not be able to determine o Brokers could report on those       
basis for some securities            securities transactions not affected  
transactions because of complex tax  by complex tax laws                   
laws                                                                       
                                        o Tax laws on selling securities      
                                        could be simplifieda                  
                                                                              
                                        o Absent tax law changes, IRS could   
                                        develop consistent reporting rules in 
                                        concert with those who report         
Brokers may not know basis for       o Brokers could use an existing       
securities purchased through another system that allows them to transfer   
broker                               basis information when taxpayers move 
                                        their securities holdings from one    
                                        broker to another                     
Brokers may not know basis for       o Companies that directly issue stock 
securities purchased through         could track and report basis and use  
companies that directly issue stock  the basis transfer system             
(e.g., employee stock purchase       
plans)                               
Brokers may not know basis for older o Brokers could track and report      
securities                           basis prospectively (i.e., only for   
                                        securities purchased after a          
                                        particular date)                      
Brokers that do not know the basis   o Prospective reporting would likely  
may rely on taxpayers to provide     produce fewer cases in which the      
basis without any verification       broker does not know the basis        
(e.g., for stocks received as gifts)                                       
                                        o Brokers could indicate on the       
                                        information return if the basis       
                                        information came from taxpayers       
Brokers cannot always obtain timely  o These companies and the securities  
adjusted basis information from      industry in concert with IRS could    
companies that issue stock and       develop a system to timely make such  
engage in corporate events (e.g.,    information available on corporate    
mergers, acquisitions)               events that affect basis              

Source: GAO.

aTax code simplification is a method through which some believe tax
compliance could be enhanced. See GAO, Understanding the Tax Reform
Debate: Background, Criteria, & Questions, GAO-05-1009SP (Washington,
D.C.: September 2005).

Table 8: Challenges to IRS Associated with Basis Reporting and How the
Challenges Could Be Mitigated

                                        

            Challenges to IRS                Ways to mitigate challenges      
Expanding IRS's computer system      o Cost to implement system would be   
capacity to store and use additional outweighed by increased tax revenue   
data on basis involves monetary      resulting from higher voluntary       
costs                                reporting compliance (although such   
                                        funds would not be IRS's to directly  
                                        use)                                  
                                                                              
                                        o Funds could be budgeted to cover    
                                        these costs                           
IRS systems may not be able to       o Brokers could report aggregate      
process and match basis for each     adjusted basis for all securities     
securities sale reported on          sold for a taxpayer on the            
information returns and on Schedule  information return while reporting    
D of the Form 1040 (including any    adjusted basis for all sales on       
attachments on the securities sold)  annual statements provided to         
                                        taxpayers                             
IRS may still encounter taxpayers    o Brokers could report aggregate      
that misclassify the holding period  basis and gross proceeds for          
for their securities sales           short-term and long-term transactions 
                                        separately on the information return  
Taxpayers may improperly report      o Allow taxpayers to use the average  
basis when they sell portions of     costs of their securities holdings to 
their holdings in a security that    determine basis for securities beyond 
they purchased on multiple occasions mutual funds                          
                                                                              
                                        o Taxpayers could indicate the method 
                                        they will use to determine basis when 
                                        their security is sold and brokers    
                                        then would report the method selected 
                                        and the related basis amount on the   
                                        information return                    

Source: GAO.

Although not all inclusive, the strategies discussed above could help
mitigate many of the challenges facing brokers and IRS if information
reporting were expanded to include cost basis. However, the strategies
also involve a number of trade-offs that would need to be considered in
terms of the costs and burdens associated with basis reporting for
taxpayers, IRS, and brokers, and the impact on reducing the capital gains
tax gap for securities.

Representatives from the securities industry we interviewed said that
brokers would incur additional costs to develop and maintain systems to
track and report basis, although they did not provide precise costs.
However, we were also told that almost all of the largest brokers directly
provide basis information to a significant portion of their clients, and
many smaller brokers provide basis to a significant portion of their
clients through outsourcing. Also, representatives of the mutual fund
industry estimated that 80 to 90 percent of mutual funds provide average
cost basis information to their shareholders. Likewise, from a societal
perspective, the cost that brokers would incur in reporting basis
information would be offset to some extent by the reduced costs to
taxpayers in researching, calculating, and reporting basis, or paying a
return preparer to perform such services. However, some brokers may pass
on the costs of reporting basis information to their customers. Further,
decisions about the scope and details of basis reporting, as further
discussed below, could constrain how much brokers' costs would increase.43

Also, representatives from the securities industry told us that their
ability to provide taxpayers and IRS with accurate basis information would
be challenged when taxpayers move their securities holdings from one
broker to another. Some brokers use a system to transfer basis among one
another, but the system is not used by all brokers. In addition, brokers
do not always track and transfer basis in a consistent manner; that is,
some track original cost basis while others track adjusted cost basis.
Without a system through which all brokers transfer standardized basis
information, the effectiveness of basis reporting would be limited.

Additionally, brokers do not always know or may be challenged in
determining the basis of taxpayers' holdings. For example, some taxpayers
may hold securities that they purchased long ago or received as a gift,
for which neither they nor their brokers know the original purchase dates.
In these cases, brokers cannot know the basis of the securities. However,
this challenge could be mitigated to a large extent if brokers were to
track and report basis prospectively, that is, only for securities
purchased after a specified future date. The trade-off to prospective
basis reporting, however, is that it would not help some taxpayers report
basis for securities they owned before brokers began to report basis,
which for a period of time would limit the impact basis reporting would
have on reducing the tax gap. Also, prospective reporting would be
complicated in cases where a taxpayer held a security prior to the
specified date and then purchased additional shares of the same security
after the specified date. Brokers would likely incur some additional costs
to separately account for shares of stock purchased before and after the
specified date for prospective reporting on information returns.

Likewise, it is difficult for brokers to determine basis for some
complicated securities transactions, according to representatives of the
securities industry. For example, when taxpayers sell stock for a loss and
then buy shares of the same stock within 30 days, they are prohibited from
claiming a loss on the original sale. For these sales, known as wash
sales, basis is difficult for the broker to determine because the taxpayer
is required to add the disallowed loss from the wash sale to the basis of
the subsequently purchased stock. The difficulty in determining basis for
wash sales is compounded when taxpayers sell a stock at a loss through one
broker and then buy the same stock within 30 days from another broker. In
this case, the second broker would not know of the wash sale the taxpayer
executed through the first broker and would not know to adjust the
taxpayer's basis accordingly. We only found two cases through our file
review where taxpayers had misreported basis because of wash sales.
Regardless, transactions such as wash sales may be too complex for brokers
to feasibly report basis. Excluding these transactions from basis
reporting, however, would further reduce the impact of basis reporting on
closing the securities tax gap.

For IRS, having basis information, along with gross proceeds information,
for each of a taxpayer's securities sales would best enable the agency to
check whether taxpayers properly reported their capital gains and losses.
However, storing and making use of such information would be challenging
because of the costs and difficulty involved in storing and computer
matching the large volume of information that transactional reporting
would entail. However, if brokers were to report only aggregate basis
amounts to IRS for all of a taxpayer's transactions, the costs and
difficulties of storing and using the information for matching would be
reduced. Aggregate reporting would also reduce the costs to brokers of
reporting basis to IRS, although they could still report basis for all
transactions to taxpayers.

Another complication for IRS and brokers is that taxpayers can choose
among various methods for reporting basis in cases where they sell some of
their shares of a security they purchased on multiple occasions. Taxpayers
may choose to report basis in a different way than brokers would otherwise
choose because taxpayers can (1) specifically identify which shares they
sell among many they hold and report basis for those shares; (2) use the
basis of the first shares they bought; or (3) in the case of mutual funds,
use the average cost of the shares they own.44 Taxpayers could indicate
the method they chose to determine basis when they sell their securities,
and brokers then could report the method selected and the related basis
amount on information returns. However, this additional layer of tracking
would likely add to costs to taxpayers, brokers, and IRS. Although this
challenge could be alleviated if taxpayers were required to report basis
in a consistent manner, this requirement would end taxpayers' ability to
determine basis in the most advantageous manner for their particular tax
situations.

Given the number of decisions that would need to be made in conjunction
with basis reporting, IRS may not be able to require such reporting given
its current authority. Although IRS has long had the authority to require
information reporting related to securities, an official from IRS's Office
of Chief Counsel told us that IRS may not have the authority to require
all of the actions that would be needed to implement cost basis
information reporting, such as regulating a system through which brokers
transfer standardized basis information. Therefore, it may be difficult
for IRS to implement cost basis information reporting without further
statutory authority.

Representatives from the securities industry told us that in order to
implement basis reporting, a set of rules would need to be developed to
clearly establish, for example, what types of securities transactions
would be covered by any requirement and how a system to transfer basis
would be standardized. These representatives thought their input could be
helpful in designing any set of rules.

Conclusion

Although neither IRS nor we know the size of the tax gap related to
securities sales, tens of millions of taxpayers hold securities outside of
their retirement accounts and, according to our analysis of IRS data, an
estimated 36 percent of taxpayers who sold securities in 2001 erred in
reporting their gains and losses (an estimated 7.3 million out of an
estimated 20.3 million taxpayers). Of those erring, an estimated 64
percent underreported their income and 33 percent overreported income.
Also, an estimated 9 percent of individual taxpayers who sold securities
misclassified their holding periods, either reporting short-term holdings
as long-term, or vice versa. Enhancing IRS's current enforcement and
service efforts is an option for addressing these compliance problems, but
the most effective tool for improving taxpayers' compliance levels has
long been information reporting and tax withholding. Individual taxpayers
misreport nearly twice the percentage of their income from sources subject
only to some information reporting-which is the case for securities income
now-compared to income subject to substantial information reporting. Also,
given that the tax consequences associated with the holding period of
securities are significant, broker reporting on this specific issue,
whether as part of basis reporting or separately, would help taxpayers
apply the proper tax rules to their gains or losses and help IRS in
identifying compliance problems.

Extending information reporting for securities sales to include basis
information is not a simple and straightforward proposition. The manner in
which basis reporting is designed would affect how the costs of basis
reporting are distributed among taxpayers, brokers, and IRS, and the
extent to which basis reporting would close the securities-related tax
gap. In addition, although IRS has the general authority to require basis
reporting, IRS officials were not certain the agency had sufficient
authority to regulate how such reporting is implemented, such as
regulating a system through which brokers transfer standardized basis
information.

In the event that brokers were required to report basis for securities
purchased as of a specific future date, some taxpayers may continue to
misreport their gains and losses from the securities holdings they
currently hold. For these taxpayers, additional guidance on reporting
basis and gains or losses for securities sales could be a low cost way to
help them voluntarily comply with their tax obligations. For example, an
estimated 28 percent of taxpayers who failed to report their securities
sales had losses. Clarification of IRS's instructions for Schedule D on
the appropriate use of capital losses to offset capital gains or other
income could be a means to help ensure that taxpayers do not disadvantage
themselves when they experience losses from their investments. Also, given
the complexity involved in determining some securities' basis because of
events such as stock splits, guidance on the resources available to
taxpayers on determining basis, such as utilizing brokers, or services
offered by companies that issue stocks or other information available on
Web sites, could help improve taxpayers' ability to determine their
securities' basis.

Matters for Congressional Consideration

In order to reduce the capital gains tax gap for securities, Congress may
want to consider requiring brokers to report to both taxpayers and IRS the
adjusted basis of securities that taxpayers sell and ensuring that IRS has
sufficient regulatory authority to implement the requirement. Either in
connection with requiring basis reporting or separately, Congress could
also require brokers to report to taxpayers and IRS whether the securities
sold were short-term or long-term holdings. Additionally, Congress could
direct IRS to work with brokers and related parties to develop rules that
seek to mitigate some of the challenges associated with requiring basis
reporting.

Recommendations for Executive Action

To assist taxpayers in accurately reporting their capital gains and losses
from securities, in the instructions to Schedule D the Commissioner of
Internal Revenue should (1) clarify the appropriate use of capital losses
to offset capital gains or other income and (2) provide guidance on
resources available to taxpayers to determine their basis.

Agency Comments and Our Evaluation

In written comments on a draft of this report, which are reprinted in
appendix II, the Commissioner of Internal Revenue agreed with our
recommendations. He also concurred that for some securities, basis
reporting involves unique challenges and noted that IRS is committed to
working with industry stakeholders to develop cost effective methods to
mitigate such reporting challenges. IRS also provided comments on several
technical issues, which we incorporated in this report where appropriate.

As agreed with your offices, unless you publicly announce its contents
earlier, we plan no further distribution of this report until 30 days from
its issue date. At that time, we will send copies to the Chairman and
Ranking Minority Member, House Committee on Ways and Means; the Secretary
of the Treasury; the Commissioner of Internal Revenue; and other
interested parties. Copies will be made available to others upon request.
This report will also be available at no charge on GAO's Web site at
http://www.gao.gov .

If you or your staff have any questions, please contact me at (202)
512-9110 or [email protected] . Contact points for our Offices of
Congressional Relations and Public Affairs may be found on the last page
of this report. Key contributors to this report are listed in appendix
III.

Michael Brostek Director, Tax Issues Strategic Issues Team

Appendix I  Scope and Methodology

To provide information on the extent of and primary types of noncompliance
that cause individual taxpayers to misreport capital gains from
securities, we performed a number of activities that relied on data from
IRS's National Research Program (NRP). Through NRP, IRS selected and
reviewed a stratified random sample of 45,925 individual income tax
returns from tax year 2001. The NRP sample is divided across 30 strata by
the type of individual tax return filed and income levels. IRS accepted as
filed some of the NRP returns, accepted others with minor adjustments, and
examined the remainder of returns either through correspondence or
face-to-face meetings with taxpayers. If IRS examiners determined that
taxpayers misreported income for any aspect of the selected tax returns,
they adjusted the taxpayers' income accordingly and assessed additional
taxes.

IRS captured data from taxpayer returns and examination results in the NRP
database, including capital gains income. However, the data on capital
gains do not indicate the type of capital asset for which taxpayers
reported gains or losses or for which examiners made income adjustments.
Therefore, to obtain information on the extent and primary types of
capital gains tax noncompliance specific to securities, we selected a
statistical sample of NRP examination files to review.

The sample we selected contained 1,017 cases spread across 90 substrata,
defined by replicating each of the 30 NRP strata across 3 GAO substrata.
The first GAO substratum consisted of examination cases for which the
adjustments to capital gain income the examiners made had the largest
impact on the total amount of these adjustments for all taxpayers when
weighted for the entire population of individual taxpayers. We focused on
cases with the largest adjustments, in weighted terms, because including
these cases would improve the level of confidence of any estimates of the
total amount of capital gains income adjustments for securities. Because
our sample is a subsample of the NRP sample and is subject to sampling
error, we added cases, when applicable, to ensure that each of the 30 NRP
strata in this GAO substratum contained a minimum of 5 cases. In total, we
selected 290 cases for the first GAO substratum, and these cases accounted
for around 75 percent of the total capital gains adjustments in NRP when
weighted for the population of individual taxpayers.

The second substratum consisted of 187 cases for which IRS did not
identify misreported capital gains income when it reviewed or examined the
tax returns. We included these returns as part of our sample to verify
that the NRP examinations had correctly recorded when taxpayers were
compliant with respect to reporting capital gains and losses. We selected
these cases at random and in proportion to the NRP sample through an
iterative process, ensuring that a minimum of 5 cases and a maximum of 15
cases was included in each of the 30 NRP strata.

The remaining 540 cases that constitute the third GAO substratum were
selected from cases for which IRS examined taxpayers' capital gain income.
We selected these cases at random and in proportion to the number of NRP
returns for which IRS examined capital gains income, ensuring that we
selected a minimum of 5 cases for each NRP stratum. For one stratum, we
only included 2 cases because they were the only cases in the
corresponding NRP stratum.

Of the 1,017 cases we selected for our sample, we reviewed 849 cases. We
did not review the remaining 168 cases because either IRS did not provide
the files in time to include in our review (164 cases) or the files did
not contain examination workpapers essential to determining if examiners
made adjustments to taxpayers income from securities (4 cases).1 Based on
an analysis of the response rates by the 90 GAO substrata, we concluded
that the missing cases did not bias our analyses. We requested the cases
at two points, in late-December 2005 and late-January 2006, and
periodically checked on the status of our requests with IRS. We were only
able to review cases that arrived by April 21, 2006 in order to meet our
agreed upon issue date for the report.

We reviewed each selected case file to determine if the taxpayers reported
securities transactions on their returns or if examiners discovered any
misreported securities transactions. For returns where examiners
discovered misreported income from securities transactions, we determined,
when possible, the related security type, holding period, adjustment
amount, and reason for the adjustment, along with other information. We
recorded all determinations on a data collection instrument (DCI) that we
developed.

To ensure that our data collection efforts conformed to GAO's data quality
standards, each DCI that a GAO analyst completed was reviewed by another
GAO analyst. The reviewers compared the data recorded on the DCI to the
data in the corresponding case file to determine whether they concurred
with how the data were recorded. When the analysts differed on how the
data were recorded, they met to reconcile any differences.

We input the data we recorded on the DCIs into a computer data collection
program. To ensure the accuracy of the transcribed data, each electronic
DCI entry was compared to its corresponding paper DCI by analysts other
that those that electronically entered the data. If the reviewers found
any errors, changes were made to the electronic entries, and the entries
were reviewed again to ensure that all data were transcribed accurately.

The estimates we included in this report were based on the NRP database
and the data we collected through our file review and were generated using
statistical software. All computer programming for the resulting
statistical analyses were checked by a second, independent analyst. Our
final sample size was large enough to generalize the results of our review
or had margins of error small enough to produce meaningful estimates in
terms of percentages of taxpayers who were noncompliant in reporting
capital gains from securities transactions. However, we could not produce
meaningful estimates of the total amount of net misreported capital gain
income from securities or determine the securities tax gap, in part
because (1) in selecting our sample, we could not distinguish which cases
included misreported securities transactions as opposed to misreported
transactions for other types of capital assets, (2) some cases with large
amounts of misreported capital gains or losses were due to noncompliance
for assets other than securities, (3) 53 of the cases we requested from
IRS from our first substratum, which represented a large percentage of the
total amounts of misreported capital gains or losses, were not provided in
time to include in our review, and (4) taxpayers misreported a wide range
of dollar amounts from the transactions.2 We discussed our estimates with
IRS officials to obtain their perspectives on the results of our analysis.

Because we followed a probability procedure based on random selection, our
sample is only one of a large number of samples that we might have
selected. Since each sample could have resulted in different estimates, we
express our confidence in the precision of our particular sample's results
as a 95 percent confidence interval, plus or minus the margin of error
indicated along with each estimate in the report. This interval would
contain the actual population value for 95 percent of the samples we could
have selected.

We assessed whether the examination results and data contained in the NRP
database were sufficiently reliable for the purposes of our review. For
this assessment, we interviewed IRS officials about the data, collected
and reviewed documentation about the data and the system used to capture
the data, and performed electronic testing of relevant data fields for
obvious errors in accuracy and completeness. We compared the information
we collected through our case file review to corresponding information in
the NRP database to identify inconsistencies. Based on our assessment, we
determined that the NRP database was sufficiently reliable for the
purposes of our review.

We also used IRS's Statistics of Income (SOI) file for individual
taxpayers, which relies on a stratified probability sample of individual
income tax returns, to develop estimates for categories of individual
taxpayers on adjusted gross income, the percentage of individual taxpayers
that used paid tax preparers, and the number of taxpayers that filed a
Schedule D with their tax returns for tax year 2003. We compared our
analyses against published IRS data to determine that the SOI database was
sufficiently reliable for the purposes of our review.

To provide information on actions IRS takes in attempting to reduce the
individual capital gains tax gap for securities and on challenges that IRS
faces with these actions, we reviewed documents from IRS compliance
programs as they related to capital gains and interviewed IRS officials
knowledgeable about the subject. We reviewed documentation for IRS's
enforcement programs that address capital gains and reviewed IRS
publications and other documents that provided information on how to
accurately report capital gains and losses. To provide additional
information on IRS's compliance programs and identify challenges IRS faces
in using these programs to reduce the individual capital gains tax gap for
securities, we interviewed IRS officials from various areas of the agency,
including the enforcement, taxpayer service, and research functions.

To identify options with the potential to improve taxpayers' voluntary
compliance for reporting securities gains and losses and IRS's ability to
find noncompliance related to the individual capital gains tax gap for
securities, we reviewed prior GAO reports and other documents on capital
gains reporting and compliance such as those from IRS compliance programs
and industry reports on securities holdings and information reporting. We
also spoke with IRS officials and numerous representatives from, and
related to, the securities industry. At IRS, we spoke with officials from
various areas of the agency, including the enforcement, taxpayer service,
and research functions. Additionally, we spoke with officials from the
Taxpayer Advocate Service and members of IRS's Information Return Program
Advisory Committee (IRPAC).3 We also spoke with representatives of the
Securities Industry Association; Investment Company Institute, which
represents the mutual fund industry; Bond Market Association; American
Banking Association Securities Association; American Institute of
Certified Public Accountants; and the American Bar Association to get
their perspectives on capital gains tax noncompliance, ways to reduce
noncompliance, and any challenges related to reducing noncompliance and
how those challenges could be mitigated.

Appendix II  Comments from the Internal Revenue Service

Appendix III  GAO Contact and Staff Acknowledgments

Michael Brostek, (202) 512-9110 or [email protected]

In addition to the contact named above, Wes Phillips and Tom Short,
Assistant Directors; Jeff Arkin; Susan Baker; Candace Carpenter; Keira
Dembowski; Fred Jimenez; Matthew Keeler; Donna Miller; John Mingus;
Franklin Ng; Karen O'Conor; Cheryl Peterson; Sam Scrutchins; Jay Smale;
and Jennifer Li Wong made key contributions to this report.

Related GAO Products

Tax Gap: Making Significant Progress in Improving Tax Compliance Rests on
Enhancing Current IRS Techniques and Adopting New Legislative Actions.
GAO-06-453T . Washington, D.C.: February 15, 2006.

Tax Gap: Multiple Strategies, Better Compliance Data, and Long-term Goals
Are Needed to Improve Taxpayer Compliance. GAO-06-208T . Washington, D.C.:
October 26, 2005.

Tax Compliance: Better Compliance Data and Long-term Goals Would Support a
More Strategic IRS Approach to Reducing the Tax Gap. GAO-05-753 .
Washington, D.C.: July 18, 2005.

Tax Compliance: Reducing the Tax Gap Can Contribute to Fiscal
Sustainability but Will Require a Variety of Strategies. GAO-05-527T .
Washington, D.C.: April 14, 2005.

Tax Administration: More Can Be Done to Ensure Federal Agencies File
Accurate Information Returns. GAO-04-74 . Washington, D.C.: December 5,
2003.

Tax Administration: IRS Should Continue to Expand Reporting on Its
Enforcement Efforts. GAO-03-378 . Washington, D.C.: January 31, 2003.

Tax Administration: IRS Can Improve Information Reporting for Original
Issue Discount Bonds. GAO/GGD-96-70 . Washington, D.C.: March 15, 1996.

Reducing the Tax Gap: Results of a GAO-Sponsored Symposium. GAO/GGD-95-157
. Washington, D.C.: June 2, 1995.

Options Reporting to IRS. GAO/GGD-95-145R . Washington, D.C.: May 5, 1995.

Tax Gap: Many Actions Taken, But a Cohesive Compliance Strategy Needed.
GAO/GGD-94-123 . Washington, D.C.: May 11, 1994.

Tax Administration: Computer Matching Could Identify Overstated Business
Deductions. GAO/GGD-93-133 . Washington, D.C.: August 13, 1993.

Information Reporting. GAO/GGD-93-55R . Washington, D.C.: July 22, 1993.

Tax Administration: Information Returns Can Improve Reporting of Forgiven
Debts. GAO/GGD-93-42 . Washington, D.C.: February 17, 1993.

Tax Administration: Overstated Real Estate Tax Deductions Need to Be
Reduced. GAO/GGD-93-43 . Washington, D.C.: January 19, 1993.

Tax Administration: Federal Agencies Should Report Service Payments Made
to Corporations. GAO/GGD-92-130 . Washington, D.C.: September 22, 1992.

Tax Administration: Approaches for Improving Independent Contractor
Compliance. GAO/GGD-92-108 . Washington, D.C.: July 23, 1992.

Tax Administration: Benefits of a Corporate Document Matching Program
Exceed the Costs. GAO/GGD-91-118 . Washington, D.C.: September 27, 1991.

IRS Needs to Implement a Corporate Document Matching Program.
GAO/T-GGD-91-40 . Washington, D.C.: June 10, 1991.

Tax Administration IRS Can Improve Its Program to Find Taxpayers Who
Underreport Their Income. GAO/GGD-91-49 . Washington, D.C.: March 13,
1991.

Tax Administration: Expanded Reporting on Seller-financed Mortgages Can
Spur Tax Compliance. GAO/GGD-91-38 . Washington, D.C.: March 29, 1991.

IRS' Compliance Programs to Reduce the Tax Gap. GAO/T-GGD-91-11 .
Washington, D.C.: March 13, 1991.

IRS Can Use Tax Gap Data to Improve Its Programs for Reducing
Noncompliance. GAO/T-GGD-90-32 . Washington, D.C.: April 19, 1990.

Tax Administration: Information Returns Can Be Used to Identify Employers
Who Misclassify Workers. GAO/GGD-89-107 . Washington, D.C.: September 25,
1989.

Tax Administration: Missing Independent Contractors' Information Returns
Not Always Detected. GAO/GGD-89-110 . Washington, D.C.: September 8, 1989.

Tax Administration: IRS' Efforts to Establish a Business Information
Returns Program. GAO/GGD-88-102 . Washington, D.C.: July 22, 1988.

The Merits of Establishing a Business Information Returns Program.
GAO/T-GGD-87-4 . Washington, D.C.: March 17, 1987.

(450428)

transparent illustrator graphic

www.gao.gov/cgi-bin/getrpt?GAO-06-603.

To view the full product, including the scope

and methodology, click on the link above.

For more information, contact Michael Brostek at (202) 512-9110 or
[email protected].

Highlights of GAO-06-603, a report to the Committee on Finance, U.S.
Senate

June 2006

CAPITAL GAINS TAX GAP

Requiring Brokers to Report Securities Cost Basis Would Improve Compliance
if Related Challenges Are Addressed

For tax year 2001, the Internal Revenue Service (IRS) estimated a tax gap
of at least $11 billion from individual taxpayers misreporting income from
capital assets (generally those owned for investment or personal
purposes). IRS did not estimate the portion of this gap from securities
(e.g., stocks, bonds, and mutual fund capital gains distributions).

GAO was asked for information on (1) the extent and types of noncompliance
for individual taxpayers that misreport securities capital gains, (2)
actions IRS takes to reduce the securities tax gap, and (3) options with
the potential to improve taxpayer voluntary compliance and IRS's ability
to address noncompliant taxpayers. For estimates of noncompliance, GAO
analyzed a probability sample of examination cases for tax year 2001 from
the most recent IRS study of individual tax compliance.

What GAO Recommends

To reduce securities capital gains noncompliance, GAO suggests that
Congress consider requiring brokers to report adjusted basis to taxpayers
and IRS and requiring IRS to work with the industry to develop cost
effective ways to mitigate reporting challenges. GAO also recommends that
IRS clarify its guidance on reporting capital gains and losses.

In commenting on a draft of this report, IRS agreed with our
recommendations.

GAO estimates that 38 percent of individual taxpayers with securities
transactions misreported their capital gains or losses in tax year 2001. A
greater estimated percentage of taxpayers misreported gains or losses from
securities sales (36 percent) than capital gain distributions from mutual
funds (13 percent). This may be because taxpayers must determine the
taxable portion of securities sales' income whereas they need only add up
their capital gain distributions. Among individual taxpayers who
misreported securities sales, roughly two-thirds underreported and roughly
one-third overreported. Furthermore, about half of these taxpayers who
misreported failed to accurately report the securities' cost, or basis,
sometimes because they did not know the basis or failed to adjust the
basis appropriately.

IRS attempts to reduce the securities' tax gap through enforcement and
taxpayer service programs, but challenges limit their impact. Through
enforcement programs, IRS contacts taxpayers who may have misreported
capital gains or losses and seeks to secure the correct tax amount. IRS
also offers services to help taxpayers comply with capital gains tax
obligations, such as guidance on how to determine securities' gains and
losses. Challenges that limit these programs' impact include the lack of
information on basis, which IRS needs to verify most gains and losses, and
uncertainty as to whether taxpayers use or understand the guidance.

Expanding the information brokers report on securities sales to include
adjusted cost basis has the potential to improve taxpayers' compliance and
help IRS find noncompliant taxpayers. IRS research shows that taxpayers
report their income much more accurately when it is reported to them and
IRS. Basis reporting also would reduce taxpayers' burden. For IRS, basis
reporting would provide information to verify securities gains or losses
and to better target enforcement resources on noncompliant taxpayers.
However, basis reporting would raise challenges that would need to be
addressed. For instance, brokers would incur costs and burdens-even as
taxpayers' costs and burdens decrease somewhat-and many issues would arise
about how to calculate adjusted basis, which securities would be covered,
and how information would be transferred among brokers. However, industry
representatives said that many brokers already provide some basis
information to many of their clients and some use an existing system to
track and transfer basis and other information about securities. Many of
the challenges to implementing basis reporting also could be mitigated.
For example, many of the challenges could be addressed by only requiring
adjusted basis reporting for future purchases, and by developing
consistent rules to be used by all brokers. To the extent that actions to
mitigate the challenges to basis reporting delay its implementation or
limit coverage to only certain types of securities, the resulting
improvements to taxpayers' voluntary reporting compliance would be
somewhat constrained.
*** End of document. ***