Internal Revenue Service: IRS Initiatives to Resolve Disputes Over Tax
Liabilities (Letter Report, 05/09/97, GAO/GGD-97-71).

GAO reviewed the Internal Revenue Service's (IRS) initiatives to resolve
tax liabilities disputes without litigation, focusing on IRS': (1)
design of these initiatives and taxpayers' use of them; and (2) plans
for evaluating the impacts of its new initiatives on the stated goals.

GAO noted that: (1) since 1990, IRS Appeals, Chief Counsel, and
Examination have implemented at least eight initiatives to improve
dispute resolution between IRS and taxpayers over certain tax issues;
(2) each of the initiatives apply to specific groups of taxpayers,
generally large corporations; (3) two of these initiatives seek to
prevent disputes, three seek to resolve disputes before they reach
Appeals, and two seek to resolve disputes in Appeals more quickly; (4)
only one initiative uses a neutral third person as a mediator to help
resolve disputes in Appeals; (5) generally, the goals of these
initiatives are to reduce the overall time, costs, and taxpayer burden
of dispute resolution; (6) in June 1996, IRS identified 276 taxpayers
that had used or were using 1 of IRS' 8 initiatives to resolve tax
disputes since 1990; (7) as of November 30, 1996, IRS data showed that
these taxpayers had used IRS' initiatives to resolve 209 disputes over
tax issues; (8) this is a small fraction of the relevant disputed tax
issues since 1990; (9) various reasons exist for the limited use of the
initiatives to date; (10) also, IRS officials said use of the
initiatives ultimately depends on the willingness of eligible taxpayers;
(11) IRS has established some performance measures intended to evaluate
the impacts of its initiatives on reducing the time, costs, and taxpayer
burden in dispute resolution; (12) GAO's analysis indicated that many of
these measures will not allow IRS to directly gauge the initiatives'
impacts on these goals; (13) Appeals has established some measures, such
as the level of taxpayer satisfaction, that are more directly related to
its initiatives' goals of reducing the time, costs, and burden of
dispute resolution; (14) IRS officials said they thought it was too
early to assess the impacts of all of their initiatives and was
difficult to obtain data that would isolate the impacts, particularly
when the issues being resolved are highly technical and can carry over
to future tax years; (15) IRS officials described ongoing efforts to
develop other measures, in conjunction with a special IRS task force, by
the spring of 1998; and (16) measures that more directly gauge the
impacts of the initiatives on their goals would help IRS determine,
after sufficient data are available a over period of time, whether and
the extent to which the initiatives had the intended effects of reducin*

--------------------------- Indexing Terms -----------------------------

 REPORTNUM:  GGD-97-71
     TITLE:  Internal Revenue Service: IRS Initiatives to Resolve 
             Disputes Over Tax Liabilities
      DATE:  05/09/97
   SUBJECT:  Tax administration
             Corporations
             Tax return audits
             Government collections
             Administrative remedies
             Taxpayers
             Income taxes
             Mediators
IDENTIFIER:  IRS Advance Pricing Agreement Program
             IRS Coordinated Examination Program
             
******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO report.  Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved.  Major          **
** divisions and subdivisions of the text, such as Chapters,    **
** Sections, and Appendixes, are identified by double and       **
** single lines.  The numbers on the right end of these lines   **
** indicate the position of each of the subsections in the      **
** document outline.  These numbers do NOT correspond with the  **
** page numbers of the printed 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.                                         **
**                                                              **
** A printed copy of this report may be obtained from the GAO   **
** Document Distribution Center.  For further details, please   **
** send an e-mail message to:                                   **
**                                                              **
**                                            **
**                                                              **
** with the message 'info' in the body.                         **
******************************************************************


Cover
================================================================ COVER


Report to the Chairman, Subcommittee on Oversight, Committee on Ways
and Means, House of Representatives

May 1997

INTERNAL REVENUE SERVICE - IRS
INITIATIVES TO RESOLVE DISPUTES
OVER TAX LIABILITIES

GAO/GGD-97-71

IRS Dispute Resolution Initiatives

(268656)


Abbreviations
=============================================================== ABBREV

  ACUS - Administrative Conference of the United States
  ADR - alternative dispute resolution
  AIR - Accelerated Issue Resolution
  APA - Advance Pricing Agreement
  CSP - Classification Settlement Program
  CEP - Coordinated Examination Program
  IRS - Internal Revenue Service
  SOI - Statistics of Income
  TEI - Tax Executives Institute

Letter
=============================================================== LETTER


B-260164

May 9, 1997

The Honorable Nancy L.  Johnson
Chairman, Subcommittee on Oversight
Committee on Ways and Means
House of Representatives

Dear Chairman Johnson: 

Each year, thousands of disputes arise between taxpayers and the
Internal Revenue Service (IRS) over billions of dollars in additional
taxes recommended by auditors in IRS' Examination Division
(Examination).  IRS eventually resolves most of these disputes over
tax liability without litigation through negotiations with taxpayers
in its Office of Appeals, but the resolution process can take years
and hundreds of staff hours for disputes over large tax amounts
because of the complex issues involved. 

Since 1990, IRS has made available to certain groups of taxpayers
several initiatives to provide alternative ways for resolving certain
tax disputes without litigation.  We are addressing this report to
you at your request because of your ongoing interest in IRS'
enforcement programs and IRS' use of performance measures in striving
to achieve its mission and program goals.  Our objectives were to (1)
analyze IRS' design of these initiatives and taxpayers' use of them
to resolve disputes between IRS and taxpayers over tax liability, and
(2) analyze IRS' plans for evaluating the impacts of its new
initiatives on the stated goals. 


   BACKGROUND
------------------------------------------------------------ Letter :1

The federal government and the private sector have long recognized
that litigation is costly, time consuming, and destructive of
cooperative relationships.  Congress intended that federal agencies
avoid these problems by offering prompt and inexpensive
administrative processes for resolving disputes; yet, over the last
30 years, agency processes have grown more formal, costly, and time
consuming. 

Seeking to counter this trend, the Administrative Conference of the
United States (ACUS) began in 1982 to encourage federal agencies to
use alternative dispute resolution (ADR) processes.\1 Because use of
ADR grew slowly among federal agencies, Congress passed the
Administrative Dispute Resolution Act in 1990 to explicitly authorize
and encourage agencies to use neutral third party ADR techniques
(app.  I describes these types of ADR techniques).\2

Well before the 1990 act, IRS offered an administrative dispute
resolution process through its Office of Appeals (Appeals) as an
alternative to litigation.  Otherwise, a taxpayer dissatisfied with
the tax adjustments recommended by an IRS auditor can either take the
dispute to Tax Court, where IRS' District Counsel initially transfers
the dispute to Appeals, or pay the additional taxes and claim a
refund in the U.S.  Court of Federal Claims or a federal district
court. 

Organizationally located in the Office of the Commissioner, Appeals
operates independently from IRS functions such as the Examination
Division, which performs audits to determine the correct tax
liability, and the Office of Chief Counsel, which litigates Tax Court
cases for IRS.  Its mission since 1927 has been to resolve tax
controversies without litigation on a basis that is fair and
impartial to both the government and the taxpayer and that will
enhance voluntary compliance and public confidence in IRS' integrity
and efficiency.  With a staff of about 2,150 employees, Appeals is
one of the oldest and largest dispute resolution organizations in the
United States. 

Appeals' process consists of an administrative review by an Appeals
officer and negotiations with the taxpayer, usually after an IRS
audit, over the tax treatment of one or more issues on the tax
return.  An Appeals officer must first review the relevant facts,
law, regulations, and court cases.  Then, through written submissions
from the taxpayer and one or more informal conferences, the Appeals
officer must assess the relative merits of the opposing views and
determine an acceptable settlement position for IRS.  Unlike
Examination, which is limited to applying the tax code, Appeals is
authorized to consider the hazards of litigation.  Thus, the Appeals
officer can negotiate with the taxpayer and make concessions to
arrive at a settlement that attempts to approximate the probable
results if the case were to be tried in court. 

This appeals process handles an average inventory of about 53,500
dispute cases.\3 Appeals' processing time for small cases (those with
less than $10 million in dispute) that are not docketed in the courts
averages about 8 months.\4 Larger, more complex cases average about
2.4 years to process.  While the larger cases comprise about 1
percent of all Appeals cases, they account for about 88 percent of
the tax dollars in dispute.  Most large cases come from IRS'
Coordinated Examination Program (CEP), under which IRS audits the
largest corporations.  Overall, Appeals has been resolving about 85
percent of its large cases. 

In addition to Appeals, which has the major role in dispute
resolution, two other IRS functions also have roles in dispute
resolution.  The Examination Division attempts to resolve
disagreements over additional tax recommendations with taxpayers
before they go to Appeals.  The Office of Chief Counsel also has a
role, particularly when Appeals' negotiations do not resolve the
disputes.  All three functions have developed initiatives to improve
the resolution of disputes over tax liability. 


--------------------
\1 ACUS was an independent federal agency established in 1964 to
promote efficient, adequate, and fair procedures in federal agencies. 
It was not funded in fiscal year 1995 and passed out of existence. 

\2 In recent years, Congress has encouraged federal agencies,
including IRS, to use ADR techniques instead of litigation or
adversarial administrative procedures.  The 1990 act serves as an
example of this encouragement.  This act terminated in October 1995,
but Congress reauthorized use of these techniques in an October 1996
act.  IRS' initiatives, for the most part, do not involve neutral
third parties, and thus do not include the full range of ADR
techniques encouraged by the act. 

\3 The average inventory for Appeals cases is based on the fiscal
year-end inventory for the 5-year period ending September 30, 1995. 

\4 Appeals divides its case workload into two basic categories: 
nondocketed and docketed.  Nondocketed cases are those protested
directly to Appeals by the taxpayer; docketed cases are those entered
on the calendar, called the docket, of the Tax Court and referred to
Appeals by the District Counsel. 


   RESULTS IN BRIEF
------------------------------------------------------------ Letter :2

Since 1990, IRS Appeals, Chief Counsel, and Examination\5 have
implemented at least eight initiatives to improve dispute resolution
between IRS and taxpayers over certain tax issues (app.  II describes
IRS' initiatives).  Each of the initiatives applies to specific
groups of taxpayers, generally large corporations.  Two of these
initiatives seek to prevent disputes, three seek to resolve disputes
before they reach Appeals, and two seek to resolve disputes in
Appeals more quickly.  Only one initiative uses a neutral third
person as a mediator to help resolve disputes in Appeals.  Generally,
the goals of these initiatives are to reduce the overall time, costs,
and taxpayer burden of dispute resolution. 

In June 1996, IRS identified 276 taxpayers that had used or were
using 1 of IRS' 8 initiatives to resolve tax disputes since 1990.  As
of November 30, 1996, IRS data showed that these taxpayers had used
IRS' initiatives to resolve 209 disputes over tax issues.\6

This is a small fraction of the relevant disputed tax issues since
1990.  Various reasons exist for the limited use of the initiatives
to date.  For example, the initiatives were relatively new and
generally target disputes with very large corporations for certain
types of issues, such as employment taxes.  Also, IRS officials said
use of the initiatives ultimately depends on the willingness of
eligible taxpayers. 

IRS has established some performance measures intended to evaluate
the impacts of its initiatives on reducing the time, costs, and
taxpayer burden in dispute resolution.  Our analysis indicated that
many of these measures will not allow IRS to directly gauge the
initiatives' impacts on these goals.  For example, Chief Counsel and
Examination both have as a measure the number of times the
initiatives were used.  Officials from these functions believe that
frequency of use means a reduction in time, costs, and burden.  But
knowing how often initiatives were used does not answer the question
of how effectively they reduced time, costs, and burden.  Appeals has
established some measures, such as the level of taxpayer
satisfaction, that are more directly related to its initiatives'
goals of reducing the time, costs, and burden of dispute resolution. 

IRS officials said they thought it was too early to assess the
impacts of all of their initiatives and it was difficult to obtain
data that would isolate the impacts, particularly when the issues
being resolved are highly technical and can carry over to future tax
years.  IRS officials described ongoing efforts to develop other
measures, in conjunction with a special IRS task force, by the spring
of 1998.  We recognize the challenges of developing measures and
evaluating the initiatives as well as the importance of proper timing
of the evaluations.  Even so, measures that more directly gauge the
impacts of the initiatives on their goals would help IRS determine,
after sufficient data are available over a period of time, whether
and the extent to which the initiatives had the intended effects of
reducing the time, costs, and burden of resolving tax disputes. 


--------------------
\5 Other IRS functions, such as Collection, have undertaken
initiatives for resolving disputes other than those involving the
amount of income tax to be assessed. 

\6 This figure does not reflect the unknown number of disputes over
transfer pricing issues that were avoided by using one of the
initiatives that focuses on those issues. 


   OBJECTIVES, SCOPE, AND
   METHODOLOGY
------------------------------------------------------------ Letter :3

Focusing on tax disputes between IRS and taxpayers, our objectives
were to (1) analyze IRS' design of its initiatives and taxpayers' use
of them since 1990 in resolving disputes over tax liability, and (2)
analyze IRS' plans for evaluating the impacts of its new initiatives
on their goals.  To gain perspective in doing work on both
objectives, we first reviewed the 1990 and 1996 acts, the
Congressional Record, and various ACUS publications including its
1995 report to Congress. 

To address both objectives, we first asked IRS officials to identify
initiatives begun since 1990 to help resolve tax disputes between IRS
and taxpayers.  They identified 11 initiatives, of which our review
included 8.  Tax Court Rule 124 was excluded because it is under the
jurisdiction of the United States Tax Court.  IRS' Ombudsman was
excluded because it has existed since 1988 and its scope of disputes
goes beyond issues of tax liability.  Simultaneous referral to
Appeals/Competent Authority was excluded because its stated purpose
was to enhance the competent authority process.  According to the
Assistant Commissioner (International), the competent authority
process addresses disputes between the United States and treaty
nations rather than disputes between the IRS and taxpayers.  Appendix
II briefly describes these additional IRS initiatives. 

To analyze IRS' design of its dispute resolution initiatives since
1990, we first learned about IRS' traditional dispute resolution
method--the appeals process--and the roles that the Examination and
Chief Counsel had in the dispute process.  To do so, we reviewed
published procedures and reports, and interviewed officials in these
functions at IRS' National Office.  To learn about IRS' new
initiatives, we reviewed related authorizations and procedures as
well as written comments on proposed initiatives from inside and
outside of IRS.  We interviewed IRS National Office officials in the
Examination Division, Office of Appeals, Office of Associate Chief
Counsel (International), and the Office of the Assistant Commissioner
(International). 

To analyze taxpayer use of IRS' initiatives, we asked IRS to identify
taxpayers that have used an initiative since 1990 by name and
identification number so we could develop profiles of those
taxpayers.  In June 1996, IRS provided that information on 276
taxpayers, usually very large corporations, that had used or were
using 1 of the 8 initiatives.  Using this information, we queried
IRS' Statistics of Income (SOI) database on corporate filers for
1993--the most recent data available.  Being a sample of all
taxpayers, this database had profile information on 209 of the 276
taxpayers.  We then identified the number of disputed tax issues that
were resolved under an initiative for the taxpayers through November
30, 1996.  Finally, we interviewed officials at the Tax Executives
Institute (TEI) and collected documentation on TEI's views of IRS'
initiatives.  Because TEI represents very large corporations--the
major users of IRS' initiatives--TEI's views provided insights on the
issue of taxpayer usage. 

To analyze IRS' strategy for evaluating the impacts of its new
initiatives, we reviewed ACUS guidance for evaluating ADR programs
and IRS functions' evaluation plans, questionnaires, and reports such
as the 1995 Appeals' Measurements and Standards Task Force report. 
We did not attempt to evaluate the impacts of IRS' initiatives
because they generally were too new and IRS data were not readily
available. 

We requested comments from IRS and TEI on a draft of this report.  On
March 17, 1997, we obtained comments from representatives of the IRS
Commissioner.  We received written comments from TEI on April 4,
1997.  As appropriate, we made changes in the report based on these
comments.  The comments and our evaluation of them are discussed
starting on page 27.  We conducted our review from March through
December 1996 at Washington, D.  C.  and our Kansas City Office in
Mission, KS, in accordance with generally accepted government
auditing standards. 


   IRS' INITIATIVES TARGET
   SPECIFIC TAXPAYERS AND ISSUES
------------------------------------------------------------ Letter :4

IRS has implemented eight initiatives since 1990 to improve
resolution of its disputes with taxpayers over certain tax issues. 
These initiatives attempt to meet goals related to reducing the time,
costs, and burden of dispute resolution.  IRS' design and timing of
these initiatives have, to date, limited taxpayers' use of the
initiatives. 


      IRS' INITIATIVES TO IMPROVE
      THE RESOLUTION OF DISPUTED
      INCOME TAX LIABILITY ISSUES
---------------------------------------------------------- Letter :4.1

Since 1927, IRS' Appeals function has offered taxpayers an
administrative process to resolve disputes over tax liability.  This
traditional process, while resolving most tax disputes without
litigation, can be time-consuming, costly, and adversarial.  In the
process, the Appeals officer acts more as an independent reviewer and
negotiator on behalf of IRS than as a neutral third party chosen by
the disputants to help design their own resolution.  As such,
Appeals' process is best characterized as settlement negotiations
with the taxpayer.\7

To improve the resolution of tax disputes between IRS and taxpayers,
IRS' Appeals, Chief Counsel, and Examination functions have
implemented at least eight initiatives since 1990.  One of these
initiatives--Appeals' mediation initiative--uses neutral parties to
help resolve disputes.  Two initiatives seek to prevent disputes,
three seek to resolve disputes before they reach Appeals, and two
seek to resolve Appeals cases more quickly.  Generally, the goals of
these initiatives are to reduce the overall time, costs, and taxpayer
burden of resolving disputes without litigation.  The following
summarizes the initiatives across the three functions (see app.  II
for a fuller description). 

  -- Using Neutral Third Parties:  In fiscal year 1996, Appeals
     completed a 1-year test of mediation procedures for nondocketed
     CEP cases.\8 Mediation has been designed to be an additional
     attempt to avoid litigation and to be available only after
     negotiations in Appeals are unsuccessful.  Once IRS approves a
     request for mediation, Appeals and the taxpayer are to select a
     neutral third party from inside or outside IRS as mediator and
     to enter into a written agreement on the issues to be discussed
     and the location and dates of the mediation. 

  -- Preventing Disputes:  In 1991, the Office of Chief Counsel
     (International) implemented its Advance Pricing Agreements (APA)
     Program to avoid disputes over intercompany transfer pricing
     issues.  Transfer pricing refers to the amounts that affiliated
     members of a multinational corporation charge one another for
     goods and services.  IRS developed the APA program to avoid
     transfer pricing disputes and the prolonged, expensive
     litigation that had been used to resolve the disputes.  Under an
     APA, IRS avoids such disputes by reaching a prospective
     agreement with the taxpayer on an appropriate transfer pricing
     methodology, the factual nature of the transactions involved,
     and the expected results of the methodology. 

In 1996, Appeals began to offer taxpayers the option of receiving IRS
valuations of art works for such purposes as estate and gift taxes
and the charitable contribution deduction on an income tax return. 
These procedures permit a taxpayer to have an art valuation for tax
purposes approved prior to filing the tax return, thus avoiding any
dispute during an audit.\9

  -- Resolving Disputes Prior to Appeals:  In 1994, Examination
     implemented the use of closing agreements between IRS and the
     taxpayer that were designed to extend the current resolution of
     a particular issue during a CEP audit to future audits of tax
     years ending prior to the date of the agreement.  IRS called
     this Accelerated Issue Resolution (AIR).  It avoids raising the
     same issue when those tax years are audited.  IRS audits nearly
     all tax returns filed by CEP taxpayers. 

In 1990, IRS gave limited authority to CEP case managers to accept
settlement offers on issues that "recur" or "rollover" across the tax
years being audited by applying a previous Appeals settlement with
the same taxpayer and issue.\10 In 1996, IRS gave limited settlement
authority to CEP case managers for particular issues in designated
industries.  These issues involve those for which (1) IRS' position
needs to be coordinated across its functions to promote consistent,
nationwide treatment, and (2) Appeals has published issue papers
containing settlement guidelines. 

  -- Resolving Disputes in Appeals:  In 1994, Appeals started
     accepting the early referral of key disputed issues before the
     end of a CEP audit in the hopes that concurrent processing would
     reduce total processing times and that early resolution of a key
     issue would help resolve related issues in Examination.  In
     1996, Appeals started a similar initiative for employment tax
     audit disputes. 


--------------------
\7 Unlike the 1990 act, the 1996 act did not include a reference to
"settlement negotiations" in the list of ADR techniques.  The
deletion was made to clarify Congress' intent to encourage use of
neutral third-party methods.  According to ACUS, settlement
negotiations do not use a neutral third party, and do not constitute
an "alternative" resolution method because agencies already had been
using them. 

\8 IRS has extended the test period for another year beginning on
January 13, 1997. 

\9 IRS has used a panel of outside experts since 1968 to evaluate
appraisals submitted by taxpayers to support the fair market value
claimed on federal income, estate, and gift tax returns for works of
art.  Disputes over the value of the art can affect tax liability. 

\10 A "rollover" issue arises from a taxable event that impacts more
than one tax period.  A "recurring" issue arises from separate or
repeated taxable events for which a taxpayer advances the same legal
position.  In 1996, IRS revised this limited authority to include any
CEP audit issue for which Appeals had previously settled the same
issue of the same taxpayer or of another taxpayer who was directly
involved in the transaction or taxable event. 


      TO DATE, USE OF INITIATIVES
      IS LIMITED
---------------------------------------------------------- Letter :4.2

As of November 30, 1996, IRS records showed that CEP and large
corporate taxpayers had used 7 of IRS' initiatives to resolve at
least 209 tax issues in dispute between IRS and taxpayers since
1990.\11 Compared to the tens of thousands of disputes we estimate
are raised annually by the audits of these taxpayers, the number of
resolutions achieved by IRS initiatives is small.\12

Several reasons, including IRS' design and timing of the initiatives,
help account for the limited use.  First, IRS generally limited use
of its initiatives to CEP taxpayers to date.  Although CEP and other
taxpayers with large disputes account for about 88 percent of the
dollars in dispute, they account for about only 1 percent of Appeals
cases.  Second, many of the initiatives began recently; three did not
start until 1996.  Third, IRS intended many of these initiatives to
initially have limited applications, as illustrated below. 

  -- Mediation may be requested only when the case is not designated
     by IRS for litigation, is not docketed before the United States
     Tax Court, or does not involve certain tax issues, and only
     after negotiations in Appeals have failed to resolve the
     dispute.\13

  -- Early referral may be used only when the referred issue (1) is
     not designated by IRS for litigation; and (2) could be expected,
     if resolved early, to help resolve related issues in
     Examination. 

  -- Certain initiatives covered unique tax issues; for instance, the
     APA program dealt only with transfer pricing issues, art
     valuation dealt only with art, and one early referral initiative
     dealt only with employment tax issues. 

  -- Limited settlement authority targeted issues that recur in CEP
     audits. 

In acknowledging the limitation on eligibility, IRS officials also
pointed out that the eligible population usually disputes very large
tax adjustments that take a lot of time to resolve.  If the
initiatives work, they could reduce the time to resolve disputes, as
well as related costs and burdens.  We agree that the potential for
such reductions exists.  Even if IRS finds that the initiatives
produce such reductions for some large dollar disputes, other
taxpayers, disputing thousands of issues annually, would be unlikely
to benefit from these reductions if they continue to not use the
initiatives or to be ineligible.  IRS officials said they plan to
expand the pool of eligible taxpayers and encourage more usage by
changing criteria such as user fees for an APA. 

IRS officials also noted that eligible taxpayers have the final say
on whether to use the initiatives.  Some taxpayers may be reluctant
to use them because they are comfortable with the traditional Appeals
process.  TEI officials said their members generally are confident of
Appeals' independence and ability to reach fair and practical
resolutions; a 1993 TEI survey indicated that over 80 percent of the
respondents were satisfied with the Appeals process.  As discussed in
our 1994 CEP report, IRS litigated relatively few CEP tax disputes
and only assessed 22 percent of the taxes recommended in CEP audits
after Appeals' settlement process.\14 TEI officials also said CEP
corporations may not yet be comfortable using mediation because it is
relatively new and IRS has rejected five of nine requests that did
not meet IRS' eligibility guidelines for mediation.\15 TEI suggested
that IRS mediation guidelines, while necessary, should not be too
restrictive and that IRS should better promote the use of its
initiatives. 

As for those that have used or were using the initiatives, we
analyzed the most recent IRS information about the types of users. 
As of June 1996, we found information on 209 corporations that had
elected to use an IRS dispute resolution initiative since 1990.\16
Appendix III contains tables that profile users by type of initiative
and the category of industry as well as the average amounts that the
users reported on their corporate income tax return for total assets,
total income, taxable income, and income tax. 

For example, 50 percent of the 209 corporations were manufacturers. 
Another 23 percent were involved in the financial, insurance, or real
estate industries, with banks being the most common users in this
category.  Further, the average amounts of total assets, total and
taxable incomes, and net tax liability by type of dispute resolution
initiative varied widely but were relatively large because nearly all
users were CEP taxpayers.  To illustrate, average total assets ranged
from about $7 billion to about $57 billion, and average taxable
income ranged from about $207 million to about $1 billion by type of
initiative. 


--------------------
\11 The 209 do not reflect the unknown number of disputes avoided
through APAs; as of November 30, 1996, IRS had completed 74 APAs. 
Appeals and Chief Counsel provided updated figures on the number of
disputed issues resolved using many of their initiatives as of
February 28, 1997.  They identified 18 more issues resolved by early
referral and early referral for employment tax issues and 82 issued
APAs. 

\12 IRS does not yet track the total number of disputed issues raised
by its audits.  Using IRS' data, we conservatively estimate that CEP
and other large corporation audits annually generate tens of
thousands of disputes.  IRS audits about 10,000 to 12,000 large
corporations per year, of which about 70 to 80 percent raise one or
more tax issues; in 1996, CEP audits raised an average of 17 issues. 
And, large corporations dispute many audit issues, often those
involving large tax amounts. 

\13 About 69 percent of Appeals cases are nondocketed.  The certain
issues include those that involve (1) specialized industries, (2)
coordination across IRS to ensure consistent treatment in the audit
or appeals process, and (3) tax treaties with other nations. 

\14 Tax Administration:  Compliance Measures and Audits of Large
Corporations Need Improvement (GAO/GGD-94-70, Sept.  1994). 

\15 Two requests did not involve CEP cases, two were premature, and
one involved docketed years not under Appeals' jurisdiction.  Of the
four approved requests, two have been completed as of November 1996. 

\16 Taxpayers involved in the 209 resolved disputes were not the same
209 taxpayers that IRS identified as having used 1 of the 8
initiatives, including APAs, and that we found in IRS' SOI sample; it
is coincidence that both populations total 209. 


   IRS' SELECTED PERFORMANCE
   INDICATORS WILL NOT MEASURE THE
   IMPACTS OF INITIATIVES ON ALL
   GOALS
------------------------------------------------------------ Letter :5

IRS' goals for its initiatives include reducing the time and costs
consumed by dispute resolutions and improving taxpayers' satisfaction
with the process.  The goals also address improvements to the
outcomes of the resolution process, including voluntary compliance
with the tax laws.  IRS officials also told us that an overarching
goal of the initiatives is to resolve more disputes without
litigation.\17 However, IRS' current performance indicators (or
measures) are not designed to directly gauge the impacts of the
initiatives on all of these stated goals, particularly the time and
costs. 

ACUS has provided guidance on possible ADR program goals and
performance measures (app.  IV summarizes ACUS' guidance on
evaluating ADR).  The goals include (1) reducing the time and costs
consumed by dispute resolutions; (2) improving the outcomes of the
resolution process, such as reducing the dispute inventory or
improving the rate at which disputes are resolved; and (3) improving
participants' satisfaction with the process and outcomes.  To
determine whether an ADR program is meeting its goals, ACUS guidance
advises ADR managers to collect and compare data for performance
measures under conditions with and without ADR. 

In addition, the 1993 Government Performance and Results Act provides
guidance on the need to have program performance measures that allow
an agency to demonstrate a program's effectiveness in achieving its
goals.  Performance measures that effectively identify whether a
program is achieving its stated goals either (1) directly measure
change (e.g., amount of time required to resolve disputes with and
without the initiative), (2) use a reasonable proxy for the goal
(e.g., taxpayer satisfaction with the initiative as an indicator of
reduced burden), or (3) provide the data needed to evaluate specific
research questions about the program and its effectiveness. 

Table 1 presents the goals and performance measures identified for
the various IRS initiatives. 




                                     Table 1
                     
                      IRS' Evaluation Measures for the Goals
                        of the Initiatives by IRS Function

                                      Goal(s)               Measure(s)
Function        Initiative(s)         --------------------  --------------------
Associate       Advance pricing       Improve voluntary     Number of advance
Chief Counsel   agreements program    compliance with       pricing agreements
(Intl),                               international tax
Office of                             laws and treaty
Chief Counsel                         provisions

                                      Reduce the (1) time   A comparison of
                                      and costs used to     average lapse time
                                      develop and resolve   and staff days to
                                      transfer pricing      complete an advanced
                                      issues, and (2) rate  pricing agreement
                                      of increase in        and to complete an
                                      resources used on     audit involving
                                      transfer pricing      transfer pricing
                                      issues                issues\a

Examination     Limited settlement    Improve rate at       Number of times
                authority;            which CEP taxpayers   initiatives are used
                accelerated issue     fully or partially
                resolution closing    agree with audit
                agreements            issues in
                                      Examination

                                      Reduce average
                                      calendar days for
                                      CEP audits

                                      Improve currency of
                                      CEP audits

                                      Improve voluntary
                                      compliance\b

                                      Reduce taxpayer
                                      burden

Appeals         Early referrals       Reduce processing     Whether Examination
                                      time in Examination   case managers and
                                      and Appeals           taxpayers perceive
                                                            that early referral
                                                            reduced audit hours
                                                            and calendar days in
                                                            Examination\c

                                      Improve the CEP       Whether Appeals
                                      agreement rate in     resolved the
                                      Examination           referred issues and
                                                            Examination resolved
                                                            related issues

                                      Improve taxpayer      Level of taxpayer
                                      satisfaction          satisfaction

                Mediation             Improve Appeals'      Rate at which CEP
                                      rate for settling     disputes are settled
                                      CEP disputes;         by Team Chiefs in
                                      resolve nondocketed   Appeals
                                      issues

                                      Reduce the costs of   No measure
                                      litigations           selected\d

                                      Improve voluntary     No measure selected
                                      compliance\b

                                      Improve taxpayer      Level of taxpayer
                                      satisfaction          satisfaction

                General               Reduce taxpayer       Number of taxpayers
                                      burden                using APAs /
                                                            Appeals' initiatives
--------------------------------------------------------------------------------
\a This is an interim measure.  In addition, Appeals, which is a
participant in Counsel's negotiation of an advance pricing agreement,
will ask Appeals participants whether the agreement reduced the time
needed to resolve issues from open tax years prior to the agreement. 

\b According to the Director, Office of Dispute Resolution and
Specialty Programs, improved voluntary compliance is an indirect goal
of Examination's and Appeals' initiatives.  He said that the
initiatives have a relationship to voluntary compliance in that the
taxpayers can choose to use an initiative to help resolve disputes. 
However, IRS has not identified a measure for voluntary compliance. 

\c Although not a selected measure of whether early referral reduced
Appeals' case/lapse time, Appeals is also asking appeals officers for
the number of hours used to resolve the early referral issue. 

\d Although Appeals has not selected a measure, Appeals officials
believe that a successful mediation avoids the costs of litigation. 

Source:  IRS data. 

As table 1 indicates, some of the measures selected by the three
functions do not directly gauge the impacts of the initiatives on
their stated goals.  For example, both Chief Counsel and Examination
plan to measure how often a taxpayer uses an initiative.  While this
is useful information, in isolation it does not demonstrate or
measure the effectiveness of the initiatives in reducing dispute
resolution time, costs, or taxpayer burden.  Also, initiatives in all
three functions include voluntary tax compliance as a goal but have
not included a related direct measure.  A prior director of the APA
program acknowledged that voluntary compliance is difficult to
measure; current IRS officials agreed and characterized taxpayers'
use of an initiative as an indicator of their desire to voluntarily
comply.  Although this may be true, signing any agreement does not
necessarily mean full voluntary compliance in the future.  Under the
APA program, IRS' Revenue Procedure 91-22 requires taxpayers to
submit annual reports that IRS can use in monitoring compliance. 

Further, one goal of early referral is to reduce total processing
time (Examination and Appeals).  However, Appeals' evaluation
planning documents show that the primary performance measure
addresses only Examination's processing time, and only the impact on
Examination processing time is reported to the National Director of
Appeals.  Appeals plans to collect data about the impact on
processing time by asking Examination case managers and taxpayers
(i.e., qualitative data) rather than collecting and comparing
quantitative data.  In March 1997, Appeals officials said that they
also will be tracking the number of hours that Appeals officers spend
on early referral issues and that their evaluations of early referral
will consider the effect on Appeals processing time. 

Table 1 also shows that Appeals' planned measures go beyond counting
the number of disputes resolved and link more directly with the
stated goals of its initiatives.  For instance, to assess the impact
on the goal of improving CEP taxpayers' satisfaction with the
examination and appeals processes, Appeals plans to query taxpayers
who use early referral or mediation about their satisfaction with
these initiatives. 

Officials from Chief Counsel and Examination indicated that they have
not completed their evaluation design efforts to directly measure the
impacts of their initiatives on all their goals for various reasons. 
For instance, they said (1) data on the time that field personnel
spend working on an APA are not always reliable because IRS has not
yet devised a method for distinguishing time spent by field personnel
on APA negotiations from time spent on pending examinations of the
same taxpayer, and (2) Examination tracks the time spent on the
entire audit and tracks certain large dollar issues but does not
track all issues nor the time spent by issue. 

Further, they pointed to the difficulty in evaluating the impacts,
partly because issues associated with initiatives such as APAs and
AIRs are often highly complex and extend years into the future.  As a
result, measuring the impacts of an initiative may have to wait
years.  They also pointed to the difficulty in isolating the impacts
of an initiative when goals such as those involving the number of
days to complete a CEP audit or the currency of the tax years being
audited can be affected by many other factors (e.g., availability of
information on an issue). 

Even with these concerns about current data and indicators, it
remains important for IRS to measure the effectiveness of the
initiatives in achieving their goals.  Without this information, IRS
cannot know whether the initiatives worked as intended (and if so, to
what extent they worked), need improvements, or merit being extended
to other issues or groups of taxpayers. 

In addition, concerns about the time and costs consumed by dispute
resolution in IRS make evaluation of each initiative important.  For
example, an article reporting a 1996 survey of CEP taxpayers,
although showing increased satisfaction with aspects of CEP audits,
showed continued dissatisfaction with Examination's ability to
resolve issues.  In the article, the Assistant Commissioner for
Examination said the satisfaction level was not as high as IRS had
hoped, particularly given the number of efforts to make CEP audits
more efficient and less time consuming.  These efforts include
Examination's dispute resolution initiatives. 

At the time of our work, the three IRS functions had not yet
evaluated the impacts of their initiatives using the measures and
goals.  In general, officials in these functions said the initiatives
were too new or used too infrequently to have sufficient data for
evaluation.  These officials told us in March 1997 that they have
made progress in developing more measures and tools to evaluate the
impacts of the initiatives on the goals.  They said they are working
with a special task force that is developing measures across IRS. 
These officials said they hope to have sufficient measures by the
spring of 1998 in order to more fully evaluate the initiatives.  Even
so, they believe that the initiatives have been helping to resolve
issues without litigation and at reduced time, costs, and burden. 

For example, Chief Counsel officials discussed interim and planned
measures such as (1) the average lapse time and staff days for
completing an APA, (2) the impact of the APA on audit cases opened
prior to it, (3) the quality of taxpayer information received while
developing the APA, (4) the dollars and number of issues
agreed/unagreed for cases with transfer pricing issues developed in
Examination, (5) taxpayer satisfaction, and (6) the dollars spent on
expert witnesses for litigations involving transfer pricing. 
Examination officials said they plan to develop a survey to collect
feedback from their staff and taxpayers and to measure time spent on
an initiative. 


--------------------
\17 IRS officials also told us that litigation is necessary at times
to ultimately resolve disputes over selected tax issues. 


   CONCLUSIONS
------------------------------------------------------------ Letter :6

For almost 70 years, IRS has relied on its Office of Appeals to
resolve most disputes over tax liability that arise after audits in
Examination without involving Chief Counsel in litigation with the
taxpayer.  Even so, Appeals' resolution of tax disputes with large
corporations can take years and hundreds of staff hours.  Because
these three IRS functions have a role in the disputes and their
resolutions, each function has started initiatives to avoid disputes
over tax liability or improve the resolution process.  Since 1990,
the three functions have started at least eight initiatives, and
Congress has encouraged federal agencies, including IRS, to use
neutral third parties in resolving disputes to help reduce the time
and costs.  One of the eight dispute resolution initiatives that IRS
has implemented since then uses a neutral third party to aid dispute
resolution. 

IRS' initiatives generally attempt to reduce the time, costs, and
taxpayer burden of the dispute resolution process by avoiding
disputes or improving the existing process.  IRS officials believe
that their initiatives have partially met these goals.  However, IRS
functions are not yet able to show whether their initiatives are
achieving these goals, partly because many of the performance
measures selected by the functions do not directly gauge the impacts
of the initiatives on these goals.  Some IRS officials have pointed
to the difficulties in evaluating whether the initiatives meet their
stated goals.  They cited problems in collecting reliable data and
isolating the impacts for highly complex tax issues that may involve
future tax years.  These difficulties, while real, do not prevent IRS
from developing indicators that more directly measure the impacts of
the initiatives on the time, costs, and burden of dispute resolution. 

IRS officials stated that a special IRS-wide task force is helping
the IRS functions to develop measures and an evaluation program for
their initiatives and that they believe that they are making
progress.  They said they hope to have sufficient measures during the
spring of 1998 to more fully evaluate the initiatives' impacts on the
stated goals.  Such efforts, if successful, would help IRS to
determine whether the initiatives worked as intended and how they
might be improved or expanded to other tax issues or groups of
taxpayers. 


   RECOMMENDATIONS
------------------------------------------------------------ Letter :7

We recommend that the IRS Commissioner hold the IRS functions
accountable, in conjunction with the special measures task force, for
(1) completing the development of performance measures that directly
gauge the impacts of the dispute resolution initiatives on their
stated goals and (2) setting milestones to measure these impacts. 
Using these measures, the functions should, after sufficient data are
available over a period of time, analyze whether each initiative
reduces the time, cost, and taxpayer burden of dispute resolution. 


   AGENCY COMMENTS AND OUR
   EVALUATION
------------------------------------------------------------ Letter :8

We obtained comments on a draft of this report in a meeting on March
17, 1997, with IRS officials that represented the IRS Commissioner. 
These officials included the Assistant Commissioner of Examination
and the Director of CEP, the National Deputy Director of Appeals and
the Director of the Office of Dispute Resolution and Specialty
Programs in Appeals, the Director and Deputy Director of the APA
Program in the Office of Associate Chief Counsel (International), and
representatives from the Office of Associate Chief Counsel (Domestic)
and the Office of Legislative Affairs. 

We discussed their comments on our findings, conclusions, and
recommendations as well as comments on technical aspects of the draft
report.  We summarize these comments in this section and made
technical changes, where appropriate, in the related parts of the
report. 

First, in discussing the type of IRS initiatives, IRS officials asked
us to describe other initiatives that fall outside the scope of our
work or that have recently begun.  We expanded our objectives, scope,
and methodology section to clarify the rationale for focusing on
eight initiatives and excluding three others--Tax Court Rule 124,
Simultaneous Referral to Appeals and Competent Authority, and IRS
Ombudsman.  We describe these and other initiatives that have begun
recently in appendix II. 

Second, in discussing the limited usage of the eight initiatives, IRS
officials said IRS plans to expand the pool of eligible taxpayers. 
These officials also noted that taxpayers ultimately choose whether
to use an initiative.  If this is true, any efforts to increase usage
by expanding the eligibility pool would not address those who are
eligible but choose to bypass the initiatives.  We added these
comments to the text of the report. 

Third, IRS officials stated that they have been working to develop
more measures to evaluate the impacts of the initiatives on the
stated goals.  They pointed to the difficulty in developing
comprehensive measures and evaluations, given disputed issues that
are highly complex, technical, and may need to be tracked for a
number of years before sufficient data are available for analysis,
particularly for the APA and AIR initiatives.  They believe that they
have made progress.  They pointed to examples of interim measures
they have been developing and measures they plan to develop through a
special IRS-wide measurement and evaluation task force.  They hoped
to have sufficient measures by the spring of 1998 to more fully
evaluate the initiatives' impacts on the stated goals.  They also
offered clarification about some existing goals and measures for
mediation and early referral.  We have incorporated these comments
and clarifications in the section of this report dealing with
measures and evaluations. 

In discussing our conclusions and recommendations, the IRS officials
asked us to more specifically account for the difficulty in
developing measures and evaluations for the initiatives.  They said,
for example, these initiatives address very complex issues that often
have to be tracked years into the future before enough data becomes
available for an evaluation.  They also asked us to account for their
efforts to further develop the measures.  Accordingly, we have
reworded our recommendations to recognize the work of the measurement
task force and the time frame needed to develop sufficient measures. 
We also have revised our conclusions to further recognize the
difficulty in developing measures for evaluation. 

TEI officials generally agreed with our description of the IRS
efforts to resolve disputes over tax liability, the appeals mission
and process, and the time to complete this process.  TEI noted that
its members are satisfied with the appeals process and the
independence of appeals staff, recognizing that such staff are IRS
employees and not neutral third parties.  While TEI favors
initiatives to resolve disputes without litigation at reduced time
and costs, TEI did not want any initiative to detract from the
historical role Appeals has played in providing an alternative to
costly, time-consuming litigation.  Given these views, TEI agreed
with our recommendation on performance measures, believing that
revising the measures would reflect the importance of dispute
resolution techniques and encourage their use. 

In addition, TEI stated that it believed the report could provide a
stronger endorsement of IRS Appeals and IRS' efforts to develop the
initiatives.  TEI also believed that IRS should do more to promote
use of the initiatives.  Given the scope and objectives of our work,
we did not take positions on these issues.  Although Appeals
generally plays a valuable role, we did not design our study to
evaluate Appeals; nor did we evaluate how well IRS developed the
initiatives and whether the initiatives worked well enough to merit
greater usage.  TEI also provided technical comments that we have
incorporated in the appropriate sections of the report. 


---------------------------------------------------------- Letter :8.1

We are sending copies of this report to the Subcommittee's Ranking
Minority Member, the Chairmen and Ranking Minority Members of the
House Committee on Ways and Means and the Senate Committee on
Finance, various other congressional committees, the Director of the


Office of Management and Budget, the Secretary of the Treasury, and
other interested parties.  We also will make the report available to
others upon request.  Major contributors to this report are listed in
appendix V.  If you or your staff have any questions on this report,
please contact me on (202) 512-9110. 

Sincerely yours,

Lynda D.  Willis
Director, Tax Policy and Administration Issues


WHAT IS ALTERNATIVE DISPUTE
RESOLUTION? 
=========================================================== Appendix I

Alternative dispute resolution, or ADR, is a name for a group of
problem-solving methods designed to resolve disputes consensually. 
These methods usually involve a neutral individual who works with the
disputing parties to help them find mutually acceptable solutions. 
The various ADR methods can be viewed as points along a continuum,
ranging from processes over which the parties have the most control
(e.g., conciliation, mediation) to processes over which they have the
least control (e.g., binding arbitration).  Here are some ADR
methods: 

Conciliation is the attempt by a neutral individual to reduce
tensions and improve communications among the parties so they can
agree on a process for resolving their dispute. 

Facilitation uses a neutral individual to assist the parties in a
meeting where the established process is used. 

Mediation uses a trained neutral individual to help the parties
negotiate a mutually agreeable settlement.  The mediator has no
independent authority and does not render a decision or opinion; a
decision must be reached by the parties themselves. 

Fact Finding is often used in technical disputes.  It uses a neutral
party with subject matter expertise to make findings of fact.  This
can be useful where disagreements about the need for or meaning of
data are impeding resolution, or where the disputed facts are highly
technical and would be better resolved by experts.  The fact-finder
usually prepares a report/advisory opinion based on an informal
presentation by each party and independent research. 

Early Neutral Evaluation uses a neutral individual with substantive
expertise to evaluate the relative merits of each party's case.  This
process usually involves an informal presentation to the neutral of
the salient points of each party's position.  The neutral provides a
nonbinding opinion that can give the parties a more objective
assessment of their positions, thereby increasing the chances that
further negotiations will be productive. 

The Settlement Conference uses a neutral individual, generally a
judge other than a presiding judge, to serve as a mediator and
neutral evaluator in a case pending before an agency tribunal.  The
settlement judge may give an informal advisory opinion.  If
settlement is not reached, the case continues before the presiding
judge. 

The Minitrial is a structured settlement process in which each side
presents a highly abbreviated summary of its case before senior
representatives of each party.  Following the presentations, the
senior representatives seek to negotiate a settlement.  A neutral
advisor sometimes presides over the proceeding, and can mediate or
render an advisory opinion if asked to do so. 

The Summary Jury Trial is a structured settlement process in which
each side has a limited time to present its case before a peer jury. 
The jury's verdict is advisory and is used to facilitate
negotiations. 

Arbitration uses a neutral individual to decide disputed issues after
hearing evidence and arguments from the parties.  The arbitrator's
decision may be binding on the parties either through agreement or
operation of law, or it may be nonbinding or advisory.  Arbitration
may be voluntary, or it may be mandatory and the exclusive means
available for handling certain disputes. 

Partnering is a process used in contracting to avoid or simplify
disputes.  At the start of a project, participants seek to identify
common goals and interests and establish clear lines of
communication.  The process may involve a joint workshop, managed by
a neutral facilitator, to develop a team charter; follow-up meetings;
and evaluation processes.  A partnering agreement usually includes a
commitment by the parties to use ADR to resolve disputes that arise
during a project. 


--------------------
\1 Federal Administrative Procedure Sourcebook, Administrative
Conference of the United States, 2nd ed.  (Washington, D.  C.: 
1992), pp.  227-229. 


IRS DISPUTE RESOLUTION INITIATIVES
========================================================== Appendix II

Table II.1 presents the dispute resolution initiatives begun by IRS
since 1990 that were included in our review.  The table presents
information on when the initiative was begun, a brief description,
the limits inherent in or placed on the initiative's use, and the
number of times taxpayers have used the initiative as of November 30,
1996.  Immediately following table II.1 are brief descriptions of
additional IRS initiatives not addressed by this report. 



                                    Table II.1
                     
                     IRS Dispute Resolution Initiatives as of
                      March 1996 to Resolve IRS and Taxpayer
                        Disputes Over Income Tax Liability

                                                                      Use as of
                          Year                                        November
Initiating  Initiative    Initiated    Description    Limits          1996
function    ------------  -----------  -------------  --------------  ----------
Appeals     Advance       1996         Permits        Limited to      2
            valuation of               taxpayers to   works of art.   completed
            works of art               obtain an IRS
                                       review of a
                                       taxpayer's
                                       valuation of
                                       a work of art
                                       before filing
                                       a return.

            Mediation     October      Permits        1. CEP cases    2
                          1995         taxpayers and  only after      completed
                                       Appeals to     negotiations    5 requests
                                       negotiate a    in Appeals.     denied
                                       settlement                     2 in
                                       assisted by    2. Not          process
                                       neutral        available for
                                       individuals    issues
                                       who have no    designated for
                                       authority to   litigation,
                                       impose a       docketed
                                       decision.      cases, or ISP,
                                                      ACIP, or
                                                      Competent
                                                      Authority
                                                      issues.

                                                      3. The
                                                      mediator(s)
                                                      are either
                                                      Appeals
                                                      employees, in
                                                      which case IRS
                                                      pays the
                                                      mediators'
                                                      costs, or
                                                      outside
                                                      parties, in
                                                      which case IRS
                                                      and the
                                                      taxpayer share
                                                      the expense.

            Early         March 1994   Allows         1. CEP cases.   23\a
            referral                   taxpayers
                                       whose returns  2. Does not
                                       are being      apply to
                                       examined to    issues
                                       request the    designated for
                                       referral of a  litigation or
                                       developed,     for which the
                                       unagreed       taxpayer has
                                       issue to       or intends to
                                       Appeals while  seek Competent
                                       Examination    Authority
                                       continues to   assistance.
                                       develop other
                                       issues.

            Early         March 1996   Allows         Limited to      2\a
            referral of                taxpayers      employment tax
            employment                 whose returns  issues.
            tax issues                 are being
                                       examined to
                                       request the
                                       referral of
                                       one or more
                                       employment
                                       tax issues to
                                       Appeals.

Counsel     Advance       March 1991   Taxpayers can  1. Limited to   74
            pricing                    request an     transfer        agreements
            agreements                 agreement on   pricing         in force\a
                                       the factual    issues.
                                       nature of
                                       intercompany   2. $5,000 user
                                       transfers, a   fee.
                                       transfer
                                       pricing
                                       methodology,
                                       and an
                                       expected
                                       range of
                                       results from
                                       the
                                       methodology.
                                       Agreement may
                                       be applied to
                                       prior tax
                                       years still
                                       in dispute.

Examinatio  Limited       November     Grants         1. Limited to   81
n           settlement    1990; Rev.   discretionary  CEP.
            authority     2, March     authority to
                          1996         Examination    2. The facts
                                       case managers  must be
                                       to accept      substantially
                                       settlement     similar;
                                       offers where   Appeals must
                                       a prior        have settled
                                       settlement     the subject
                                       has been       issue on the
                                       negotiated by  merits
                                       Appeals for    independent of
                                       the same       other issues;
                                       issue with     and legal
                                       the same       authority must
                                       taxpayer, or   be unchanged.
                                       of another
                                       taxpayer who
                                       was directly
                                       involved in
                                       the taxable
                                       event.

            Limited       March 1996   Grants         Applies to      2
            settlement                 authority to   issues
            authority                  Examination    coordinated
            for                        case managers  under the
            coordinated                to accept      Industry and
            issues                     settlement     International
                                       offers with    Field
                                       respect to     Assistance
                                       coordinated    Specialization
                                       issues within  Programs for
                                       the Industry   which Appeals
                                       and            has settlement
                                       International  guidelines.
                                       Field
                                       Assistance
                                       Specializatio
                                       n Programs on
                                       which Appeals
                                       has
                                       coordinated
                                       issue papers
                                       containing
                                       settlement
                                       guidelines or
                                       positions.

            Accelerated   October      Allows         1. Limited to   97
            issue         1994         district       CEP.
            resolution                 directors, in
                                       coordination   2. Not
                                       with other     available for
                                       IRS            issues subject
                                       functions, to  to an APA,
                                       execute        issues
                                       closing        involving
                                       agreements     employee plans
                                       with CEP       or exempt
                                       taxpayers      organizations
                                       that cover     or partnership
                                       both the tax   items under
                                       year under     Tax Equity &
                                       audit and      Fiscal
                                       unaudited tax  Responsibility
                                       years ending   Act of 1982,
                                       prior to the   issues for
                                       date of the    which
                                       agreement.     resolution
                                                      would be
                                                      contrary to an
                                                      IRS position,
                                                      or issues
                                                      designated for
                                                      litigation.
--------------------------------------------------------------------------------
\a IRS officials provided updated figures as of February 28, 1997, on
the use of some initiatives to resolve disputed issues.  They
identified 39 disputed issues resolved by early referral and 4
resolved by early referral of employment tax issues.  They also said
82 APAs were in force. 

Source:  IRS. 


   OTHER INITIATIVES TO RESOLVE
   TAX LIABILITY DISPUTES AND
   OTHER ISSUES
-------------------------------------------------------- Appendix II:1

1.  Tax Court Rule 124 -- In 1990, the Tax Court promulgated Rule 124
allowing voluntary binding arbitration of factual issues under the
direction of the Tax Court.  Arbitration has not been seriously
considered to resolve disputes between taxpayers and the IRS outside
of docketed Tax Court status. 

2.  Although no specific Tax Court mechanism for the use of mediation
has been developed, mediation can be used by the parties to resolve a
Tax Court case under existing Tax Court authority. 

3.  IRS Taxpayer Ombudsman/Advocate -- The Taxpayer Ombudsman was
authorized by Congress in 1988 to issue Taxpayer Assistance Orders to
stop certain actions to alleviate a hardship faced by a taxpayer. 
The Ombudsman was replaced by the Taxpayer Advocate, effective July
30, 1996.  The Taxpayer Advocate has broad authority to require
action or to stop an action with respect to a taxpayer. 

4.  Simultaneous Appeals/Competent Authority Procedures -- Taxpayers
may request competent authority assistance when they believe actions
of the United States, a treaty country, or both, will result in
taxation that is contrary to treaty provisions.  If the request is
accepted, the U.S.  Competent Authority generally will consult with
the appropriate foreign competent authority in an attempt to reach an
agreement that is acceptable to all parties.  According to the
Assistant Commissioner (International), the competent authority
process addresses intergovernmental disputes between the United
States and treaty nations rather than disputes between the IRS and
taxpayers and does not meet the definition of ADR. 

Revenue Procedure 96-13 established a new competent authority
procedure whereby taxpayers may seek simultaneous competent authority
assistance and Appeals consideration of the competent authority
issue.  This procedure coordinates the two processes and is intended
to reduce the time required to resolve disputes by allowing taxpayers
more proactive involvement.  Taxpayers may request simultaneous
Appeals/competent authority procedures in four situations:  after
Examination has proposed an adjustment, after the taxpayer files a
protest against the adjustment, after the dispute goes to Appeals,
and after a competent authority request has been made. 

5.  Environmental Cleanup Costs -- IRS is developing a revenue
procedure that will provide special procedures for requesting written
guidance on the tax treatment of environmental cleanup costs incurred
over several years.  The procedures will attempt to facilitate
resolution of issues involving the capitalization or deduction of
such cleanup costs for prior and future years of a single
environmental cleanup transaction. 

6.  Classification Settlement Program (CSP) -- The Commissioner
announced the CSP on March 5, 1996, to help resolve audit disputes
over the classification of workers as either employees or independent
contractors.  Under the CSP, IRS auditors will offer pro forma
settlements to taxpayers under audit using a standard closing
agreement developed in the National Office. 


PROFILE DATA ON TAXPAYERS THAT
HAVE USED IRS DISPUTE RESOLUTION
INITIATIVES
========================================================= Appendix III

Based on identifying information provided by IRS, we located 1993 SOI
data for 209 corporations that have used 1 of 6 IRS dispute
resolution initiatives as of June 24, 1996.\1 Table III.1 indicates
that the largest number of users by industry are manufacturers. 
Within IRS' industry codes, the heaviest users were banks (15
percent) and chemical companies (12 percent). 



                                   Table III.1
                     
                      Percent of 209 Corporations Using IRS
                      Dispute Resolution Initiatives by Type
                      of Initiative and Category of Industry

                             IRS dispute resolution initiative
            --------------------------------------------------------------------
                                               Early
                 Limited   Accelerated     referrals
Industry      settlement         issue           and                         All
category       authority    resolution     mediation           APA   initiatives
----------  ------------  ------------  ------------  ------------  ------------
Constructi            0%            0%            5%            1%            1%
 on
Manufactur            50            53            55            48            50
 ing
Transporta            13            15             4             1             6
 tion and
 public
 utilities
Wholesale             10             2             9            19            13
 trade
Retail                13             6             4             2             5
 trade
Finance,              13            21            23            26            23
 insurance,
 and real
 estate
Services               0             2             0             3             2
================================================================================
Total               100%          100%          100%          100%          100%
--------------------------------------------------------------------------------
Source:  GAO analysis using IRS data. 

As shown below, income varied substantially by industry for the 209
corporations.  The average income for all corporations was over $3
billion.  This is not surprising because mostly very large
corporations are given the opportunity to use IRS' dispute resolution
initiatives.  The table also shows that the average taxable income
was $443 million and average income tax paid was $76 million. 
However, 64 of the corporations (31 percent) had no taxable income
and paid no taxes.  Another 52 corporations (25 percent) paid no
taxes on their net income. 



                              Table III.2
                
                Assets, Income, and Tax Reported by 209
                     Corporations Using IRS Dispute
                         Resolution Initiatives

                         (Dollars in millions)

                                                Averag
                                Averag  Averag       e  Averag
                                     e       e  taxabl       e  Averag
                                 total   total       e  income   e net
IRS initiative                  assets  income  income     tax     tax
------------------------------  ------  ------  ------  ------  ------
Limited settlement authority    $9,700  $2,100    $247     $86     $77
AIRs                            23,200   4,700     704     246     112
Early referrals and mediation   56,700  10,200   1,200     435     186
APAs                             6,900   1,700     207      72      38
All corporations                16,200   3,300     443     151      76
----------------------------------------------------------------------
Source:  GAO analysis using IRS data. 

The use of tax credits and net operating losses of corporations using
IRS initiatives may explain why many of the corporations paid few
taxes.  Many of the corporations claimed tax credits.  For instance,
116 of 209 (56 percent) of the corporations claimed credits totaling
about $17 billion.  Most of these corporations either had or were
seeking APA agreements. 



                              Table III.3
                
                   Tax Credits and Net Operating Loss
                 Deduction Claimed by 209 Corporations
                Using IRS Dispute Resolution Initiatives

                                                     In millions
                                                ----------------------
                                        No. of       Total     Average
                                    corporatio      amount      amount
Type of credit                              ns     claimed     claimed
----------------------------------  ----------  ----------  ----------
Foreign tax credit                          84     $14,700        $175
General business credit                     85         914          11
Alternative minimum tax credit              31         779          25
Total credits                              116      16,600         144
Net operating loss deduction                57       4,200          74
----------------------------------------------------------------------
Source:  GAO analysis using IRS data. 



                              Table III.4
                
                   Tax Credits and Net Operating Loss
                  Deduction Claimed by 30 Corporations
                   Using Limited Settlement Authority

                                                     In millions
                                                ----------------------
                                        No. of       Total     Average
                                    corporatio      amount      amount
Type of credit                              ns     claimed     claimed
----------------------------------  ----------  ----------  ----------
Foreign tax credit                          13        $215         $17
General business credit                      9          47           5
Alternative minimum tax credit               8          82          10
Total credits                               30         382          17
Net operating loss deduction                10         148          15
----------------------------------------------------------------------
Source:  GAO analysis using IRS data. 



                              Table III.5
                
                   Tax Credits and Net Operating Loss
                  Deduction Claimed by 47 Corporations
                      Using Accelerated Agreements

                                                     In millions
                                                ----------------------
                                        No. of       Total     Average
                                    corporatio      amount      amount
Type of credit                              ns     claimed     claimed
----------------------------------  ----------  ----------  ----------
Foreign tax credit                          24      $5,800        $242
General business credit                     26         335      12,900
Alternative minimum tax credit               7         302          43
Total credits                               35       6,500         186
Net operating loss deduction                16         292          18
----------------------------------------------------------------------
Source:  GAO analysis using IRS data. 



                              Table III.6
                
                   Tax Credits and Net Operating Loss
                  Deduction Claimed by 22 Corporations
                  Using Early Referrals and Mediation

                                                     In millions
                                                ----------------------
                                        No. of       Total     Average
                                    corporatio      amount      amount
Type of credit                              ns     claimed     claimed
----------------------------------  ----------  ----------  ----------
Foreign tax credit                          13      $5,200        $400
General business credit                     12         406          34
Alternative minimum tax credit               6         280          47
Total credits                               14       5,900         421
Net operating loss deduction                11       3,100         282
----------------------------------------------------------------------
Source:  GAO analysis using IRS data. 



                              Table III.7
                
                   Tax Credits and Net Operating Loss
                 Deduction Claimed by 110 Corporations
                    Having or Seeking APA Agreements

                                                     In millions
                                                ----------------------
                                        No. of       Total     Average
                                    corporatio      amount      amount
Type of credit                              ns     claimed     claimed
----------------------------------  ----------  ----------  ----------
Foreign tax credit                          38      $3,500         $92
General business credit                     28         126           5
Alternative minimum tax credit              10         116          12
Total credits                               45       3,700          82
Net operating loss deduction                20         624          31
----------------------------------------------------------------------
Source:  GAO analysis using IRS data. 


--------------------
\1 Does not include taxpayers that requested advance valuation of
works of art and limited settlement authority for coordinated issues. 
As of the date of our request, only one or two taxpayers had used
these methods. 


SUMMARY OF ACUS GUIDANCE ON
PROGRAM EVALUATION
========================================================== Appendix IV

ADR program evaluation is simply a way to determine whether an ADR
program is meeting its goals and objectives.  ACUS recommended
systematic planning for program evaluation at the time an ADR program
was set up because this allows agencies to establish data collection
mechanisms early in the program. 

ACUS guidance outlined three phases to program evaluation:  (1)
planning; (2) design and implementation; and (3) presentation,
dissemination, and use of results.  This summary addresses the
planning and design/implementation phases. 

For the planning phase, ACUS guidance noted four planning steps:  (1)
determining the evaluation's goals, (2) identifying the audience(s)
and their needs, 3) considering issues of timing and expense, and (4)
selecting an evaluator. 

ACUS guidance stated that the evaluation's goals should be tied
closely to the goals of the program being evaluated.  Ideally, the
agency will have established the ADR program's goals in the program
design phase.  Although terminology differs, evaluations are commonly
characterized as either "program effectiveness" or "program design
and administration" evaluations.  Program effectiveness evaluations,
also known as impact or outcome evaluations, focus on whether a
program is meeting its goals and/or having the desired impact. 
Program design and administration evaluations, also known as process
evaluations, examine how well a program is operating. 

Usually, a variety of people have an interest in the results of a
program evaluation.  These audiences may be interested in different
issues and seek different types of information.  Potential audiences
should be identified early and kept in mind in planning the
evaluations so that their questions will be addressed.  For example,
program (such as Appeals) officials and legal staff may be interested
in the ADR program's impact on case inventory and the nature of
settlements, how long it takes to resolve cases, and whether ADR
promotes long-term compliance; budget officers may be interested in
whether ADR has achieved cost savings; and members of Congress may be
interested in how the use of ADR affects budgets. 

Evaluation can be undertaken at different times during the life of an
ADR program.  Among the factors to consider are whether the program
has been in operation long enough to have sufficient cases available
for analysis, and whether the program has resolved early
implementation problems.  The agency also needs to (1) identify
budget and resource constraints, organizational opposition, and
operational difficulties; and (2) develop strategies for dealing with
them.  The agency should also consider the qualifications of
objectivity, experience, technical expertise, and understanding the
context in which the ADR program operates in selecting the
evaluator(s). 

For the design and implementation phase, ACUS guidance laid out broad
steps for an effective ADR program evaluation.  These included

  -- understanding ADR program design and operation,

  -- translating evaluation goals into measurable performance
     indicators,

  -- determining data needs and availability,

  -- selecting an appropriate design strategy,

  -- deciding how to collect the data,

  -- collecting the data, and

  -- analyzing and interpreting the data. 

Once the evaluation's goals have been established, evaluators can
select appropriate performance indicators.  Performance indicators
represent the questions being asked in the evaluation and serve as
the basis for data collection and analysis.  According to ACUS,
measuring all aspects of an ADR program is neither easy nor
necessary.  Some types of data are harder to obtain, and some areas
of a program are more important than others to examine. 

ACUS' evaluation guidance included a list of performance indicators,
which it intended as a "sampling" from which agencies could select as
they formulated ADR program goals and identified measures of program
effectiveness.  For program effectiveness evaluations, the list
presented indicators under three categories--efficiency,
effectiveness, and customer satisfaction. 

Indicators of efficiency included measures of cost and time. 

Indicators of effectiveness included measures of dispute outcomes,
durability of outcomes, and impact on the dispute environment. 
Suggested measures of dispute outcomes include (1) the number of
settlements achieved through ADR; (2) the number of cases going
beyond ADR to litigation; (3) the nature of monetary and other
outcomes; and (4) the relationship between outcomes and other
factors, such as complexity, number of issues, or number of
disputants.  Suggested measures of durability included (1) rate of
compliance with negotiated agreements, (2) rate of dispute
recurrence, and (3) impact on organizational environment.  Suggested
measures of impact on dispute environment included (1) size of case
inventory, (2) types of disputes, (3) timing or level of dispute
resolution, (4) management or public perceptions, and (5) negative
impacts. 

Indicators of customer satisfaction included measures of
participants' satisfaction with the process, impact on relationships,
and participants' satisfaction with outcomes.  Measures of
participants' satisfaction with the process included perceptions of
fairness, appropriateness, usefulness, and control. 

A variety of research designs are available for use in evaluating ADR
programs.  Possible design strategies include case studies, time
series analysis, and group comparisons.  Evaluation planners should
consult persons with research methodology expertise when selecting a
design strategy. 

Once the evaluators determine what to measure, the next consideration
is what data to collect.  In a program effectiveness evaluation,
evaluators need to be able to describe the situation without the ADR
program (sometimes referred to as the baseline) to serve as a basis
for comparison.  Comparison information is not always easy to obtain,
especially if the agency does not keep ongoing records reflecting the
relevant information.  Ideally, evaluators collect comparison data by
using a control group.  If a control group is not available,
alternatives must be used, although they may not provide as good a
comparison.  For example, comparison data can be collected by looking
at the historical period before the ADR program began. 


--------------------
\1 Evaluating ADR Programs:  A Handbook for Federal Agencies,
Administrative Conference of the United States, Dispute Systems
Design Working Group (Washington, D.  C.:  1995). 


MAJOR CONTRIBUTORS TO THIS REPORT
=========================================================== Appendix V


   GENERAL GOVERNMENT DIVISION
--------------------------------------------------------- Appendix V:1

Tom Short, Assistant Director, Tax Policy and Administration Issues


   KANSAS CITY REGIONAL OFFICE
--------------------------------------------------------- Appendix V:2

Royce L.  Baker, Tax Issue Area Coordinator
Patricia M.  Crown, Evaluator-in-Charge
James Slaterbeck, Evaluator
Thomas N.  Bloom, Computer Specialist

*** End of document. ***