Tax Administration: Assessment of IRS' 2001 Tax Filing Season	 
(21-DEC-01, GAO-02-144).					 
								 
The tax filing season for most individual taxpayers extends from 
January 1 through April 15. It is during that time when most	 
taxpayers file their returns, call the Internal Revenue Service  
(IRS) with questions, and make other contracts with IRS related  
to filing. The challenges presented by the sheer scope of IRS'	 
filing season activities are exacerbated by the need for IRS to  
keep pace with changing technology, organizational structures,	 
and taxpayer demands. GAO assessed IRS' performance in the	 
following five key filing season activities: (1) processing	 
individual tax returns and refunds, (2) increasing the extent to 
which individual income tax returns are filed electronically, (3)
answering telephone calls and providing quality telephone	 
service, (4) providing accurate and timely face-to-face 	 
assistance at its Taxpayer Assistance Centers (TAC), and (5)	 
providing services via its Web site on the Internet. IRS'	 
performance during the 2001 filing season varied depending on the
specific filing season activity being discussed. First, although 
there was less information available on which to base a judgement
than in the past filing seasons, IRS' processing of about 130	 
million individual income tax returns and about 94 million	 
refunds in 2001 generally went smoothly. Problems that arose were
addressed quickly, with relatively minor impact on taxpayers.	 
Second, about 31 percent of all individual income tax returns	 
were filed electronically in 2001--an increas of 13.7 percent	 
compared to 2000. That rate of increase was below IRS' goal of 20
percent and the lowest percentage since 1996. Although IRS has	 
identified several such impediments, it does not have sufficient 
information to determine specifically why about 40 million	 
individual income tax returns were prepared on computer but filed
on paper in 2001. Third, IRS has several measures directed at at 
assessing the ability of taxpayers to reach an IRS assistor by	 
telephone and the accuracy of IRS' responses to callers. IRS	 
improved its performance to varying degrees in three of four	 
accessibility measures and met or exceeded two of its four goals 
for 2001, with respect to accuracy. Next, the number of taxpayers
assisted at TAC's declined in 2001, in part, IRS officials	 
surmised, because taxpayers made greater use of other forms of	 
assistance, including IRS' Web site. IRS did not measure the	 
quality of service provided by TACs during the filing season, but
a review by the Treasury Inspector General for Tax Administration
showed that the accuracy of tax law assistance was poor. Finally,
various indicators related to IRS' Web site show that (1) overall
usage increased compared to the 2000 filing season after a slow  
start. GAO's assessment of the Web site showed that (1) a key IRS
measure for tracking  site usage--number of ''hits'' --is	 
calculated in a way that calls into question the measure's	 
usefulness and (2) the site is useable, but could be more	 
user-friendly.							 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-02-144 					        
    ACCNO:   A02607						        
  TITLE:     Tax Administration: Assessment of IRS' 2001 Tax Filing   
Season								 
     DATE:   12/21/2001 
  SUBJECT:   Customer service					 
	     Electronic forms					 
	     Income taxes					 
	     Information resources management			 
	     Internet						 
	     Performance measures				 
	     Tax administration 				 
	     Tax administration systems 			 
	     Tax refunds					 
	     Tax returns					 
	     Taxpayers						 
	     Telephone						 
	     Web sites						 
	     IRS E-file Customer Number Program 		 
	     IRS Personal Indentification Number		 
	     Program						 
								 
	     IRS TeleFile Program				 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Testimony.                                               **
**                                                              **
** 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-02-144
     
Report to the Chairman, Subcommittee on Oversight, Committee on Ways and
Means, House of Representatives

United States General Accounting Office

GAO

December 2001 TAX ADMINISTRATION

Assessment of IRS' 2001 Tax Filing Season

GAO- 02- 144

Page i GAO- 02- 144 IRS' 2001 Tax Filing Season Letter 1

Results in Brief 2 Background 4 Scope and Methodology 5 With Two Significant
Exceptions, Processing of Returns and

Refunds Went Smoothly 7 Despite Various Efforts, IRS Did Not Achieve Its
Electronic Filing

Goal for 2001 and Is Not on Track to Achieve Its Long- Range Goal 13 Limited
Progress in Providing World- Class Telephone Service 19 Fewer Taxpayers
Visited Taxpayer Assistance Centers; Limited

Performance Data Indicate Continuing Problems With the Quality of Assistance
22 Indicators Related to IRS? Web Site Are Generally Positive, but the

Site Could Be More User- friendly 29 Conclusions 34 Recommendations for
Executive Action 35 Agency Comments and Our Evaluation 36

Appendix I Problems IRS Experienced in 2001 Related to Processing 39

TeleFile 39 Debt Indicator Database 39 Printing of Notices 40

Appendix II IRS Efforts to Encourage Electronic Filing and Enhance Returns
Processing in 2001 41

IRS Continued to Expand Electronic Filing Options, but Evaluations of
Effectiveness Have Been Limited 41 IRS Implemented Several Processing
Initiatives, But Problems

Arose and Evaluations Were Limited 43

Appendix III TAC Managers? Views on Their Sites? Filing Season Performance
50

Most TAC Managers Viewed Their Sites? Filing Season as Successful or Very
Successful 50 Most Managers Cited Factors That Hampered Their Sites? Success
54 Contents

Page ii GAO- 02- 144 IRS' 2001 Tax Filing Season Appendix IV Comments From
the Internal Revenue Service 58

Appendix V GAO Contacts and Staff Acknowledgments 66 GAO Contacts 66
Acknowledgments 66

Tables

Table 1: New and Revised Performance Measures for Processing Paper Returns,
Refunds, and Remittances 10 Table 2: Number of Individual Income Tax Returns
Received, by

Filing Type 14 Table 3: Number of Individual Income Tax Returns Filed

Electronically Since 1995 15 Table 4: IRS Filing Season Telephone Assistance
Performance 21 Table 5: Percentage Estimates of TAC Managers? Views of Their

Sites? Filing Season Success 51 Table 6: Percentage Estimates of TAC
Managers? Views on the

Extent to Which Certain Factors Hampered Their Sites? Filing Season Success
55

Figures

Figure 1: Taxpayers Assisted at TACs During the Filing Season 22 Figure 2:
Wait- Time Reported by Customer Satisfaction Survey

Respondents 27

Abbreviations

CERCA Council for Electronic Revenue Communication ECN E- file customer
number ERO electronic return originator IRS Internal Revenue Service NCOA
National Change of Address PIN personal identification number RRA 98 IRS
Restructuring and Reform Act of 1998 SSN Social Security Number TAC Taxpayer
Assistance Center TIGTA Treasury Inspector General for Tax Administration
TRR tax resolution representative

Page 1 GAO- 02- 144 IRS' 2001 Tax Filing Season

December 21, 2001 The Honorable Amo Houghton Chairman, Subcommittee on
Oversight Committee on Ways and Means House of Representatives

Dear Mr. Chairman: In response to your request, this report discusses
various aspects of the Internal Revenue Service?s (IRS) performance during
the 2001 tax filing season. The filing season for most individual taxpayers
extends from January 1 through April 15. It is during that time when most
taxpayers file their returns, call IRS with questions, and make other
contacts with IRS related to filing. Because IRS? activities during the
filing season affect every American taxpayer and because those activities
will often be the only contact most taxpayers have with IRS, breakdowns in
IRS? performance can have serious implications. The challenges presented by
the sheer scope of IRS? filing season activities are exacerbated by the need
for IRS to keep pace with changing technology, organizational structures,
and taxpayer demands.

In this report, we assess IRS? performance in five key filing season
activities:

 processing individual income tax returns and refunds,

 increasing the extent to which individual income tax returns are filed
electronically,

 answering telephone calls and providing quality telephone service,

 providing accurate and timely face- to- face assistance at its Taxpayer
Assistance Centers (TAC) 1 , and

 providing services via its Web site on the Internet. We testified before
this Subcommittee on the interim results of our assessment in April 2001. 2

1 IRS? Taxpayer Assistance Centers were previously referred to as ?walk- in
sites.? 2 Internal Revenue Service: 2001 Tax Filing Season, Systems
Modernization, and Security of Electronic Filing (GAO- 01- 595T, Apr. 3,
2001).

United States General Accounting Office Washington, DC 20548

Page 2 GAO- 02- 144 IRS' 2001 Tax Filing Season

Our assessment was based on interviews of IRS officials and persons in the
private sector, analyses of IRS data and data obtained from sources outside
IRS, observations of IRS operations, and a telephone survey of a random
sample of TAC managers.

Because of the importance of IRS? telephone operations and of performance
data and measures in assessing IRS? filing season, we conducted more
detailed reviews of those issues. We issued a separate report on telephone
operations on December 7, 2001, 3 and anticipate issuing a separate report
on IRS? performance measures in April 2002.

IRS? performance during the 2001 filing season varied depending on the
specific filing season activity being discussed.

Processing of Returns and Refunds- Although there was less information
available on which to base a judgment than in past filing seasons, IRS?
processing of about 130 million individual income tax returns and about 94
million refunds in 2001 generally went smoothly. With the following two
exceptions, problems that arose were addressed quickly, with relatively
minor impact on taxpayers.

 A bank that was under contract to collect tax returns and process tax
remittances for IRS destroyed or failed to process returns and about 71,000
payments amounting to about $1.2 billion.

 Problems that caused many electronic submissions to be rejected limited
the impact of an alternative signature program that IRS developed to
encourage more electronic filing.

Our assessment of IRS? performance in processing returns and refunds is
based on interviews of IRS officials, information from representatives of
the tax practitioner community, an analysis by the Treasury Inspector
General for Tax Administration (TIGTA), and reported results of some
performance measures and workload indicators. Unlike past years, we were
unable to compare IRS? processing performance against its performance in
prior years or its goals for the current year because IRS revised the suite
of measures it uses to judge processing performance and had not yet set
goals against which performance could be compared. IRS

3 IRS Telephone Assistance: Limited Progress and Missed Opportunities to
Analyze Performance in the 2001 Filing Season (GAO- 02- 212, Dec. 7, 2001).
Results in Brief

Page 3 GAO- 02- 144 IRS' 2001 Tax Filing Season

also implemented several processing initiatives for 2001. It experienced
some problems during implementation, and its evaluations of the initiatives
were limited. More thorough evaluations would provide a better basis for
deciding whether to continue, curtail, or revise the initiatives.

Electronic Filing- About 31 percent of all individual income tax returns
were filed electronically in 2001- an increase of 13.7 percent compared to
2000. That rate of increase was below IRS? goal of 20 percent and the lowest
percentage increase since 1996. This declining growth rate occurred despite
various IRS efforts to identify and eliminate impediments to electronic
filing and reduces the likelihood that IRS will achieve its long- range goal
of having 80 percent of individual income tax returns filed electronically
by 2007. Although IRS has taken various steps to identify impediments to
electronic filing and has identified several such impediments, it does not
have sufficient information to determine specifically why about 40 million
individual income tax returns were prepared on computer but filed on paper
in 2001.

Telephone Service- IRS made limited progress toward its long- term goal of
providing world- class telephone service. IRS has several measures directed
at assessing the ability of taxpayers to reach an IRS assistor by telephone
(accessibility) and the accuracy of IRS? responses to callers. IRS improved
its performance to varying degrees in three of the four accessibility
measures. However, taxpayers waited longer for service in 2001 compared to
2000, and IRS failed to meet any of its accessibility goals for 2001. With
respect to accuracy, IRS met or exceeded two of its four goals for 2001 and
did significantly better in one area (account quality) compared to 2000.

Face- to- Face Assistance- The number of taxpayers assisted at TACs declined
in 2001, in part, IRS officials surmised, because taxpayers made greater use
of other forms of assistance, including IRS? Web site. Most taxpayers who
visited a TAC during the 2001 filing season waited 15 minutes or less to get
assistance. IRS did not measure the quality of service provided by TACs
during the filing season, but a review by TIGTA showed that the accuracy of
tax law assistance was poor. IRS plans to begin measuring TAC service
quality in 2002. As a result, IRS may not realize improvements in quality
until the 2003 filing season or later, after it has had time to analyze the
results of the new measures and determine the causes of any poor service.
Contrary to TIGTA?s results and despite IRS? lack of relevant performance
data, managers responsible for most TACs, in response to our survey, rated
their sites? performance in providing accurate tax law assistance as
?successful? or ?very successful.? Many

Page 4 GAO- 02- 144 IRS' 2001 Tax Filing Season

managers, however, said problems with such things as facilities, staffing,
equipment, and training hampered performance in 2001.

Web Site- Various indicators related to IRS? Web site show that (1) overall
usage increased compared to the 2000 filing season; (2) IRS improved its
timeliness in answering taxpayers? E- mail questions; and (3) the Web site?s
availability and average delivery time, as measured by an independent Web
site rater, improved late in the filing season after a slow start. 4 Our
assessment of the Web site showed that (1) a key IRS measure for tracking
site usage- number of ?hits?- is calculated in a way that calls into
question the measure?s usefulness and (2) the site is useable, but could be
more user- friendly. For example, we found 10 instances where links within
the site forwarded us to pages indicating that the document or site could
not be found. IRS is in the process of revising the content and management
of its Web site- changes that should address the problems we identified.

We are making recommendations to the Commissioner of Internal Revenue
directed at (1) obtaining information on why computer- prepared returns are
filed on paper, (2) evaluating processing initiatives, (3) improving TAC
services, and (4) discontinuing or revising the Web site hits measure. In
commenting on a draft of this report, the Commissioner cited various IRS
plans that, in general, appear responsive to our recommendations. We will be
assessing IRS? implementation of those plans as part of our review of the
2002 tax filing season.

The IRS Restructuring and Reform Act of 1998 (RRA 98) 5 required, in part,
that the Commissioner of Internal Revenue develop and implement a plan to
reorganize IRS and that the plan, among other things, (1) eliminate or
substantially modify IRS? existing organization, which was based on a
national, regional, and district structure and (2) establish organizational
units serving particular groups of taxpayers with similar needs. Before the
passage of RRA 98, IRS? organizational structure had not significantly
changed for about 50 years. In accordance with RRA 98, IRS initiated a major
reorganization to streamline its management structure and create a

4 Availability is the percentage of time the Web site?s home page downloads
fully. Average delivery time is the time it takes for the home page to fully
download from the time the user hits the ?enter? key on the keyboard.

5 P. L. 105- 206. Background

Page 5 GAO- 02- 144 IRS' 2001 Tax Filing Season

more taxpayer- focused organization centered around four customerfocused
operating divisions. 6 The responsibilities of one of those divisions (the
Wage and Investment Division) include processing individual income tax
returns and providing assistance to taxpayers who call on the telephone or
walk into an IRS office.

The 2001 filing season was the first under the new IRS organization. In May
2001, the Commissioner of Internal Revenue testified that although IRS had
made substantial progress in implementing the new organizational structure,
many challenges remained as the organization was staffed and trained.
Although IRS reported that tens of thousands of IRS managers and employees
have new jobs, for example, in the field assistance area, it anticipated
that, in the short term, the reorganization would be invisible to taxpayers
and tax practitioners.

In response to the requirements of the Government Performance and Results
Act of 1993 7 and RRA 98, IRS established a system of balanced performance
measures that would allow IRS to determine its performance against program
plans and goals. 8 This is part of an agencywide effort to develop
performance measures to help IRS achieve its mission of providing America?s
taxpayers with top quality service. Throughout the 2001 filing season, IRS
made changes to the measures it uses to assess its performance in various
areas, particularly in processing returns and refunds.

In response to your request that we assess various aspects of IRS? tax
filing season performance, we

6 The four operating divisions are Wage and Investment, which serves
taxpayers whose only income is from wages and investments; Small Business
and Self- Employed, which serves fully or partially self- employed
individuals as well as businesses with assets of $5 million or less; Large
and Mid- Size Businesses, which serves businesses with assets over $5
million; and Tax Exempt and Government Entities, which serves pension plans,
exempt organizations, and governments.

7 P. L. 103- 62. 8 IRS? balanced measures system has three components-
customer satisfaction, employee satisfaction, and business results. Business
results performance measures consist primarily of quantity and quality
measures. IRS also uses workload indicators, such as number of returns
processed, for purposes of resource planning. Scope and

Methodology

Page 6 GAO- 02- 144 IRS' 2001 Tax Filing Season

 interviewed IRS officials about current operations, initiatives, and
performance;

 observed operations at 2 of IRS? 10 submission processing centers;

 reviewed and analyzed IRS documents and data, including information posted
to IRS? Web site and related material;

 interviewed representatives of various private organizations that prepare
tax returns and that participated in conferences and tax forums sponsored by
IRS and external stakeholders; and

 reviewed related congressional testimony and work performed by TIGTA. We
also surveyed, by telephone, a random sample of managers responsible for 84
of IRS? 413 TACs to solicit their views on the services provided at those
offices. We randomly selected a sample of 84 TACs from a listing of TACs
provided by IRS, which excluded nontraditional sites, such as mobile vans or
kiosks. We conducted separate interviews for each of the 84 TACs, even if
the same manager was responsible for more than 1 of the TACs. All percentage
estimates from our sample of managers have sampling errors of plus or minus
10 percentage points or less, with 95 percent confidence unless otherwise
noted. We also observed operations at three TACs.

We conducted our work at IRS? National Office; Wage and Investment Division
headquarters in Atlanta; submission processing centers in Atlanta and
Cincinnati; the Joint Operations Center in Atlanta; 9 and TACs in the
Atlanta area. We selected these offices for a variety of reasons, including
the proximity of our audit teams and the location of key IRS managers. While
we did not independently assess the accuracy of IRS? performance data, we
verified that IRS had procedures in place that were intended to ensure data
reliability. We did our work from January through October 2001, in
accordance with generally accepted government auditing standards.

9 The Joint Operations Center is the organization responsible for managing
IRS? telephone operations.

Page 7 GAO- 02- 144 IRS' 2001 Tax Filing Season

Generally speaking, the processing of individual income tax returns and
refunds in 2001 went smoothly. However, because IRS was replacing or
revising many performance measures, it is difficult to put IRS? performance
in 2001 into context. Specifically, for important areas of performance, we
could not make comparisons to IRS? accomplishments in 2000 or to its goals
for 2001.

Given the size and scope of IRS? processing operations, problems are
inevitable, and there were some problems in 2001. However, with two
exceptions, those problems were addressed quickly, with relatively minor
impact on taxpayers. IRS also implemented several initiatives related to the
processing of individual income tax returns in 2001. IRS experienced
problems in implementing some of the initiatives, and evaluations of the
initiatives? effectiveness were limited.

IRS processed about 130 million individual income tax returns and issued
about 94 million individual income tax refunds in 2001- about 2 million more
returns and 2. 4 million more refunds than in 2000. According to IRS
officials, TIGTA, and representatives of the tax practitioner community, IRS
did a good job processing those returns and refunds, especially considering
the simultaneous challenge of reorganization. For example:

 in a September 2001 report, TIGTA concluded that IRS had successfully
processed individual income tax returns during the 2001 filing season while
undergoing organizational restructuring and implementing key processing
changes; 10

 in testimony before this Subcommittee, one of the largest tax return
preparation firms characterized the filing season as successful and
positive; and

 representatives from other, smaller companies with whom we spoke voiced
similar views.

The one negative aspect mentioned by some practitioners was the alternative
signature program for electronic filers, which is discussed later.

10 The Internal Revenue Service Successfully Processed Individual Tax
Returns During the 2001 Filing Season, TIGTA, Reference No. 2001- 40- 192,
Sept. 28, 2001. With Two Significant

Exceptions, Processing of Returns and Refunds Went Smoothly

IRS? Processing During the 2001 Filing Season Generally Went Smoothly

Page 8 GAO- 02- 144 IRS' 2001 Tax Filing Season

Not unexpectedly, some processing problems did arise during the filing
season involving such things as incorrect addresses on and information in
certain tax packages, incorrect information given to providers of Refund
Anticipation Loans, 11 and untimely notices. As discussed in appendix I,
available information indicated that IRS handled these problems in a timely
manner to limit any adverse effect on taxpayers. Two problems that did
adversely affect many taxpayers involved a bank that was under contract to
collect tax returns and process tax remittances for IRS and an alternative
signature program that IRS had developed to encourage more electronic
filing. The bank problem is discussed next. The alternative signature
program is discussed in a later section of this report dealing with
electronic filing.

IRS uses lockboxes, operated by private banks, to process some tax payments
submitted with individual income tax returns. Lockboxes are to receive tax
returns and accompanying payments from taxpayers, process the payments, and
ship the returns to IRS for processing. During the 2001 filing season, IRS
learned that some returns and remittances that were to be processed by a
lockbox operated by a Pittsburgh bank were missing. According to information
available when we finalized this report, employees at that lockbox may have
concealed or destroyed about 71,000 payments amounting to about $1.2
billion. The bank?s contract was terminated in August 2001. At the time we
completed our audit work, TIGTA was investigating this matter. That
investigation should provide more specific information on the number of
returns and dollar amount of remittances affected.

According to IRS, during the first week of June 2001, after it became aware
of this situation, it released to the press and public, information about
the missing documents, which included a toll- free telephone number for
taxpayers to call for assistance. In that regard, a June 9, 2001, article in
the Pittsburg Post- Gazette noted that IRS had found 647 payments that had
never been processed in Pittsburgh. According to IRS, once it learned that a
larger number of taxpayers were affected, it posted an alert on its Web site
on August 31, 2001.

11 Private firms offer Refund Anticipation Loans to taxpayers who file
electronically. The loan amount is based on the amount of the taxpayer?s
expected refund, and the refund is sent directly to the loan provider to pay
off the loan. Returns and Remittances

Destroyed or Not Processed by Contractor

Page 9 GAO- 02- 144 IRS' 2001 Tax Filing Season

Because it was unclear whether IRS would locate the missing documents, IRS
advised affected taxpayers to send another copy of their return. It also
advised taxpayers who had sent remittances to Pittsburgh that had not yet
cleared their banks to stop payment and said that it would reimburse
taxpayers for any bank fees incurred for the stop- payment orders.

As part of its agencywide efforts to develop balanced performance measures,
IRS has revised some measures and developed new ones to judge its
performance in processing paper returns, refunds, and remittances, while
keeping comparable workload indicators, such as the numbers of returns
processed and the time it takes to process returns. 12 In past years, our
assessment of IRS? performance in processing paper returns and refunds
included a comparison of various performance measures against IRS? goals and
prior years? performance. We were unable to make such a comparison for
measures this year because IRS (1) revised some measures that it had been
using to assess processing performance and established some new measures and
(2) had not established goals for the new or revised measures in 2001.

Table 1 describes the performance measures for processing individual income
tax returns, refunds, and remittances that were in place during the 2001
filing season and shows the performance results IRS reported for each of
those measures through July 2001. IRS officials agreed with our assessment
that it is difficult to put the reported results into context because of the
absence of performance goals and trend data. However, according to IRS
officials, while they understand the importance of such information, they
also rely heavily on workload indicators and other management information
they have used for years to identify and correct problems that could affect
processing and other activities and help judge IRS? overall filing season
success.

12 This section deals only with paper processing performance measures. Any
relevant electronic filing measures will be discussed in the electronic
filing section of this report. New and Revised

Performance Measures for Paper Processing Limit Ability for Comparison

Page 10 GAO- 02- 144 IRS' 2001 Tax Filing Season

Table 1: New and Revised Performance Measures for Processing Paper Returns,
Refunds, and Remittances Measure Description New or revised Baseline year a
Performance through

July 2001

Letter error rate Percentage of incorrect letters issued by the submission
processing organization. New 2001 12.7% Notice error rate Percentage of
submission processing

notices issued incorrectly to taxpayers that are automatically generated for
various reasons, including advising taxpayers of missing Social Security
numbers (SSN). Not all notices are automatically generated; some for which a
standard notice would not apply are generated manually.

Revised to include systemic errors b and only submission processing notices.

2002 13.8% Deposit error rate Percentage of payments processed by IRS

centers that were misapplied (excludes lockboxes).

New 2001 5.4% Deposit timeliness The lost opportunity cost of interest on
the

money received but not deposited by the next business day, per $1 billion,
using a constant interest rate of 8 percent. (Excludes money received by
lockboxes.)

New 2001 $824,154 Refund timeliness Percentage of refunds issued in 40 days
or

less from the IRS- received date. Revised the start date for determining

calculation. 2002 96.6%

Refund error rate Percentage of refunds with IRS- caused errors in the name,
address, or refund amount, including systemic errors.

Revised to include systemic errors. 2001 9.8%

Refund interest paid Amount of refund interest paid for each $1

million in refunds issued. New 2001 $118.92 Productivity Weighted volume for
different types of

processing work per staff year expended. New 2001 27,807 a IRS intended to
use its performance in 2001 as the baseline against which to compare its

performance in future years. However, IRS made changes to two measures
(notice error rate and refund timeliness) that will result in it not
establishing some baselines until 2002. b Systemic errors are computer
generated errors over which a particular processing center would not have
control. Source: GAO summary of IRS data including the Wage and Investment
Division?s Strategy and Program Plan for 2001 to 2003, dated Sept. 21, 2001.

During the 2001 filing season, IRS made changes to many of the measures in
table 1. Some of the changes, such as renaming a measure and inverting its
calculation, should not affect IRS? ability to begin trending data from the
2001 filing season. 13 However, IRS made more substantial changes to

13 Where once IRS had a notice accuracy rate measure that was the percentage
of correct notices issued, for example, it now has a notice error rate
measure that is the percentage of incorrect notices issued.

Page 11 GAO- 02- 144 IRS' 2001 Tax Filing Season

two measures- notice error rate and refund timeliness- that limited IRS?
ability to baseline those measures until the 2002 filing season.

IRS mails millions of notices to advise taxpayers of such things as math
errors. IRS reviews samples of these notices before they are mailed to
ensure that they are accurate. For 2001, IRS redesigned six submission
processing notices in an attempt to make them clearer. Due to a programming
oversight, IRS inadvertently omitted the redesigned notices from the
sampling it does to measure the notice error rate. Therefore, the measure
was not representative of the errors for all submission processing notices.
Omitting these particular notices was significant because they are some of
the most commonly used notices. IRS officials said that necessary
programming changes were made in June 2001, so that the sampling and
resulting measure will be reflective of all submission processing notices
for the 2002 filing season.

About 72 percent of the individual income tax returns processed by IRS in
2001 involved refunds totaling about $161 billion. IRS changed the way it
measures the timeliness of issuing refunds to taxpayers who filed on paper.
Before 2001, IRS used the date the taxpayer signed the return as the start
date for computing refund timeliness and had a goal of processing a certain
percentage of those refunds within 40 days from that date. For the 2001
filing season, IRS used the IRS- received date as the start date for
computing timeliness because it had no control over the date the taxpayer
put on the return, but did have control over its own operations. As we
reported in our April 2001 testimony, while we agreed with the new start
date, keeping the goal at 40 days effectively gave IRS a better chance of
meeting the 40- day goal because the IRS- received date could be several
days later than the date the taxpayer signed the return.

For the 2002 filing season, IRS plans to change the computation by adding 2
days to the IRS- received date to allow for mailing time, but keep the
overall goal at 40 days. We believe that this represents a change in the
methodology used to calculate the measure that will make it impossible to
compare the results for 2001 with the results for 2002. IRS officials
disagreed. Because the overall goal of 40 days has not changed, they believe
the results will be comparable.

As noted previously, we anticipate issuing an assessment in April 2002 of
IRS? new performance measures, including measures discussed in this report.
Notice Error Rate

Refund Timeliness

Page 12 GAO- 02- 144 IRS' 2001 Tax Filing Season

IRS implemented several initiatives related to its processing of returns
that were intended to either improve processing operations or enhance
compliance. IRS validated secondary SSNs, 14 added a checkbox to the
individual income tax forms through which taxpayers could authorize IRS to
discuss their return with their paid practitioner instead of having to
submit a separate authorization form, transcribed information from Schedule
K- 1s, 15 and began using a U. S. Postal Service database to update taxpayer
addresses. These initiatives may have obvious or inherent benefits beyond
improving performance, such as increasing compliance, and it may take years
until the benefits are fully achieved. However, in some cases, IRS
experienced problems that limited the effectiveness of these initiatives.
For example, despite concerted efforts, IRS experienced problems in
processing Schedule K- 1s- problems that contributed to lower than
anticipated productivity rates.

Also, despite the demonstrated benefits of evaluating initiatives that are
aimed at improving performance 16 and the presence of IRS guidance requiring
that plans for improving performance include steps to monitor and evaluate
the implementation and effectiveness of actions, IRS? evaluations of such
initiatives were limited. IRS officials generally drew conclusions about the
effectiveness of initiatives based on broad numbers and trends. For example,
a responsible IRS official considered the thirdparty- authorization checkbox
to be very successful because about 28 million taxpayers checked the box.
IRS officials assumed that the number of people who checked the box directly
equated to a reduction in the number of separate authorization forms it
would have received from those taxpayers, but it lacked detailed analysis to
support this assumption. Also, IRS had no system in place for tracking how
use of the third- party

14 The Social Security Administration assigns SSNs that IRS uses to identify
individual taxpayers. In the case of a joint return, IRS considers the
person whose name appears first on the return to be the primary taxpayer and
the next person listed on the return to be the secondary taxpayer.

15 Schedule K- 1s are attached to returns filed by partnerships, trusts, and
subchapter S corporations. A K- 1 is prepared for each taxpayer who received
a distribution from the organization, and the receiving taxpayers are to
report that income on their individual income tax returns.

16 Our work on the Government Performance and Results Act of 1993 has
demonstrated the benefits of evaluating the effectiveness of actions taken
to improve performance. See

Program Evaluation: Studies Helped Agencies Measure or Explain Program
Performance (GAO/ GGD- 00- 204, Sept. 29, 2000) and Program Evaluation:
Agencies Challenged by New Demand for Information on Program Results (GAO/
GGD- 98- 53, Apr. 24, 1998). IRS Implemented

Processing Initiatives but Had Some Problems and Did Limited Evaluation

Page 13 GAO- 02- 144 IRS' 2001 Tax Filing Season

authorization checkbox facilitated the resolution of issues related to a
return?s processing.

IRS officials said that although initiatives may not be specifically
evaluated for effectiveness, they are evaluated for resource planning and
other purposes as part of filing season readiness reviews and the
development of action plans to prepare for the next filing season. While we
are aware of this fact, we are unconvinced that evaluations as part of
filing season readiness reviews would be sufficient for determining
effectiveness.

See appendix II for a more detailed discussion of these processing
initiatives.

For the first time since 1995, IRS did not achieve its anticipated rate of
growth for the electronic filing of individual income tax returns and, as a
result, fell short of its goal for the 2001 filing season. This reduces the
likelihood of IRS achieving its long- range electronic filing goal of 80
percent by 2007. 17 Because electronic filing is beneficial to IRS and
taxpayers, 18 IRS has taken numerous actions to identify impediments to
electronic filing. One group of potential electronic filers that IRS does
not know enough about are the large number of practitioners and taxpayers
who prepare tax returns on computer but file on paper. IRS took steps to
encourage more taxpayers and practitioners to file electronically in 2001.
However, the most significant of those actions, implementation of a
selfselect personal identification number (PIN) program for signing
electronic returns, encountered problems that limited its effectiveness.

17 RRA 98 provides, in part, that it should be the goal of IRS to have at
least 80 percent of all federal tax and information returns filed
electronically by 2007. IRS? goal for individual income tax returns is
consistent with this overall goal- 80 percent by 2007.

18 Electronic filing is beneficial to IRS and taxpayers because it ensures a
more accurate tax return and thus reduces the need for correspondence
between them. Accuracy is enhanced because (1) checks are built into the
electronic filing process that are designed to catch certain taxpayer
errors, such as incorrect SSNs, so that they can be corrected before IRS
takes possession of the tax return, and (2) returns filed electronically
bypass the more error- prone manual procedures that IRS uses to process
paper returns. Electronic filing has other benefits as well. For example,
taxpayers receive their refunds faster and get an acknowledgement from IRS
that their returns were received, and IRS has less paper to store. Despite
Various

Efforts, IRS Did Not Achieve Its Electronic Filing Goal for 2001 and Is Not
on Track to Achieve Its LongRange Goal

Page 14 GAO- 02- 144 IRS' 2001 Tax Filing Season

IRS has the following three types of electronic filing: (1) traditional,
whereby taxpayers transmit returns to IRS through a third party (such as a
tax return practitioner); (2) TeleFile, whereby taxpayers send returns
directly to IRS over telephone lines using a Touch- Tone telephone; and (3)
on- line, whereby taxpayers send returns to IRS through an on- line
intermediary using a personal computer and commercial software. As shown in
table 2, the use of traditional filing and on- line filing increased in
2001, but not as substantially as they had in 2000, while the use of
TeleFile continued to decrease (this was the third consecutive year). Table
2 also shows that about 30.9 percent of the individual income tax returns
filed in 2001 were filed electronically.

Table 2: Number of Individual Income Tax Returns Received, by Filing Type
Number of returns in thousands

Filing type 1/ 1/ 99 to 10/ 29/ 99 1/ 1/ 00 to 10/ 27/ 00 Percentage

change 1999 to 2000 1/ 1/ 01 to 10/ 26/ 01

Percentage change 2000

to 2001

Paper 96,178 92,322 -4.01 89,834 -2.69

Electronic

Traditional 21,227 25,211 18.77 28,988 14.98 TeleFile 5,665 5,161 -8.90 4,
419 -14.38 On- line 2, 457 5,022 104.40 6,838 36.16

Subtotal 29,349 35,394 20.60 40,245 13.71 Total 125,527 127,716 1. 74
130,079 1. 85

Percentage filed electronically 23.4 27.7 a 30.9 a

a Not applicable Source: IRS? Management Information System for Top Level
Executives.

Table 3 shows that the upward growth in electronic filing depicted in table
2 continues the upward trend started in 1995. However, table 3 also shows
that the percentage increase between 2000 and 2001 was smaller than at any
other time during this period. Electronic Filing Growth

Rate Has Slowed; IRS at Risk of Not Achieving Long- Range Goal

Page 15 GAO- 02- 144 IRS' 2001 Tax Filing Season

Table 3: Number of Individual Income Tax Returns Filed Electronically Since
1995 Number of returns in thousands

Year Number of electronic returns Increase Percentage

increase

1995 11,807 a a 1996 14,968 3, 161 26.8 1997 19,136 4, 168 27.8 1998 24,580
5, 444 28.4 1999 29,349 4, 769 19.4 2000 35,394 6, 045 20.6 2001 40,245 4,
851 13.7

Cumulative increase 28,438 240.9

a Not applicable. Source: IRS? Management Information System for Top Level
Executives and IRS Publication 3187 (Rev. 11- 1999).

As shown in table 3, the rate of growth for electronic filing slowed to 13.7
percent in 2001, which was about 6 percentage points less than the 20
percent growth IRS had projected. As a result, IRS did not meet its goal of
receiving 42 million returns electronically in 2001 and, as of September
2001, had lowered its projected rate of growth for 2002 to about 13 to 15
percent.

The decreasing growth rate puts at risk IRS? chances of reaching its long-
range goal of 80 percent electronic filing by 2007. Assuming continuing
annual growth rates of 13.7 percent for individual returns filed
electronically and 1. 85 percent for the total number of individual returns
filed, about 60 percent of all individual income tax returns will be filed
electronically in 2007.

IRS officials did not know why IRS fell short of the projected 20 percent
increase in electronic filing in 2001. However, in 2001, as in other years,
IRS took various steps to identify reasons why practitioners and taxpayers
do not file electronically. These steps include electronic filing forums,
meetings of the Council for Electronic Revenue Communication Advancement
(CERCA), taxpayer attitudinal surveys, customer satisfaction surveys, market
research, focus groups, and practitioner meetings.

Although IRS has taken numerous steps to identify impediments to electronic
filing, our review of recent IRS research efforts and our IRS Has Taken
Steps to

Identify Impediments to Electronic Filing, but It Has Inadequate Information
on Why Many Computer- Prepared Returns Are Filed on Paper

Page 16 GAO- 02- 144 IRS' 2001 Tax Filing Season

attendance at IRS tax forums indicated that IRS does not have sufficient
information to determine why about 40 million individual income tax returns
were prepared on computer but filed on paper in 2001. Of those returns,
about 32 million were filed by tax return practitioners and about 8 million
were filed by taxpayers. Converting these filers to electronic is important
if IRS is to achieve its 80 percent electronic filing goal. That importance
was underscored in an internal IRS research report dated May 16, 2001.
According to the report, without a successful marketing plan designed to
convert filers of computer- prepared paper returns to electronic, ?it is
difficult to see how the Service can attain its 80 percent electronic filing
mandate.?

IRS? various research efforts have resulted in limited direct contact with
the filers of computer- prepared paper returns. The tax forums that we
attended were generally attended by practitioners who were already involved
in electronic filing or already interested in becoming electronic filers.
Two recent research efforts involving focus groups involved about 90 filers
of computer- prepared paper returns. One research effort completed in May
2001 by a private marketing and research firm involved 18 in- depth
interviews with 3 tax preparation firms that electronically file, 4 large
accounting firms, and 11 organizations that represent tax practitioners. In
its report, the firm identified many knowledge gaps, including why computer-
prepared forms are filed on paper. The most recent research study was done
by the same firm and completed in June 2001. The firm contacted
practitioners who had filed at least 1 tax return electronically and a
sample of the general taxpayer population, which included 70 taxpayers who
filed computer- prepared paper returns in 2001. None of these research
efforts was specifically targeted at determining what specific factors were
most significant in explaining why so many computer- prepared returns are
filed on paper.

In past years, IRS took steps to encourage taxpayers and practitioners to
file electronically. For example, in response to concerns voiced by
practitioners that electronic filing was not entirely paperless, IRS, during
the 1999 and 2000 filing seasons, tested alternative signature options that
allowed certain electronic filers to sign their returns electronically and
thus avoid having to send any paper to IRS. Additionally, each year IRS has
expanded the list of forms and schedules that can be filed electronically.
For the 2001 filing season, IRS expanded on its prior years? efforts and
took some additional actions. A discussion of the most significant and most
problematic of those actions- the self- select PIN program- follows. Other
actions are discussed in appendix II. IRS Took Steps to

Encourage More Electronic Filing in 2001, but the Most Significant Step
Encountered Problems

Page 17 GAO- 02- 144 IRS' 2001 Tax Filing Season

For the 2001 filing season, IRS introduced the self- select PIN program to
replace two alternative signature programs that IRS had tested during the
1999 and 2000 filing seasons- the practitioner PIN program and the E- file
customer number (ECN) program. Taxpayers who decided to use the selfselect
PIN were to (1) input any five numbers as their electronic signature and (2)
verify their identity by providing IRS with certain personal information,
including two pieces of information (known as ?shared

secrets?) from their previous year?s tax return. The following are three
major differences between the self- select PIN program and the two
alternative signature programs tested in 1999 and 2000.

 The self- select PIN program significantly increased the number of
taxpayers who could file a totally paperless tax return. 19 Under this
program, all eligible taxpayers who filed on- line or through any of the
about 124,000 approved electronic return originators (EROs)- a total of
about 60 million taxpayers, according to IRS- could file a totally paperless
return. 20 That level of eligibility compares to the 2000 filing season when
the only persons who could file a totally paperless return were (1)
taxpayers who filed through 1 of the 18,000 EROs who participated in the
practitioner PIN program or (2) the 12 million potential on- line filers who
received an ECN.

 Practitioners using the self- select PIN did not have to fill out and
maintain an authentication worksheet, unlike with the practitioner PIN
program.

 The self- select PIN program allowed all taxpayers to select their own
PIN. In contrast, on- line filers had to use an ECN assigned by IRS.

In 2001, about 9 million taxpayers filed their tax returns electronically,
either through a practitioner or on- line, using a self- select PIN.
Although this number is more than the about 6.8 million filers who used
either the practitioner PIN or ECN programs in 2000, the increase is due
primarily to on- line filers. In that regard, (1) about 2.9 million more on-
line filers used a self- select PIN in 2001 than participated in the ECN
program in 2000 and

19 Taxpayers who file electronically and use the electronic signature option
are not required to submit a signature document or copies of their Wage and
Tax Statements (Form W- 2). 20 Individuals who filed Form 1040, 1040A, or
1040EZ or who used TeleFile the previous year, with a few exceptions, were
eligible to use a self- select PIN in 2001. Also eligible were individuals
who did not file a tax return in the previous year and would be 16 or older
by December 31, 2001. Self- select PIN Program

Intended to Encourage More Electronic Filing

Self- select PIN Program Encountered Problems

Page 18 GAO- 02- 144 IRS' 2001 Tax Filing Season

(2) about 4.8 million practitioner- prepared tax returns were filed in 2001
using a self- select PIN- about 600,000 fewer than the 5.4 million tax
returns that were submitted via the practitioner PIN program in 2000. The
decline in practitioner- prepared returns occurred in spite of the fact that
many more practitioners were eligible to participate in the self- select PIN
program than were eligible to participate in the practitioner PIN program.

One key factor that limited the number of participants in the self- select
PIN program was the fact that at least some tax practitioners became
disenchanted with the program because a large number of returns being
submitted with self- select PINs were being rejected by IRS. In that regard

 IRS records, as of October 16, 2001, showed about 2.1 million reject
conditions on returns filed electronically because some of the ?shared

secrets? that taxpayers and practitioners submitted to verify the taxpayer?s
identity did not match tax information in IRS? database; 21 and

 representatives of the country?s largest tax return preparation company
testified on April 3, 2001, that the company was experiencing a reject rate
of about 20 percent during the 2001 filing season compared to a normal
electronic filing reject rate of 12 to 13 percent, with almost all of the
difference attributed to rejected PINs.

According to IRS, the large number of rejects associated with the selfselect
PIN program were caused not by systemic problems with IRS? program but by
incorrect information provided by taxpayers or practitioners.

Because of the reject rates, the company recommended that members of its
organization who prepared returns and taxpayers who used the company?s
software package to prepare their own returns not use the selfselect PIN. In
their April 3 testimony, representatives of this company said the following:

?This year, thousands of PIN applications submitted with correct ?shared
secrets? information were rejected, costing tax professionals time and
income as they worked to correct the rejects, and angering many customers
who anticipated easier filing and faster refunds. Some preparers abandoned
the program as it became easier to use the paper

21 The number of reject conditions cannot be equated to the number of
rejected returns because one return can have more than one reject condition.

Page 19 GAO- 02- 144 IRS' 2001 Tax Filing Season

[signature document] instead of having to deal with a PIN reject and asking
the client to return to file [a paper signature document].?

?The IRS believes the majority of PIN rejects are caused by customer error.
But whether the problem is with the practitioner, the taxpayer, or the IRS,
one fact remains: If you [file electronically] using a PIN, your return is
more likely to be rejected.? 22

Practitioners who had problems with the self- select PIN testified before
the Congress and stated at tax forums and CERCA meetings that they still
filed electronically but, instead of using a PIN, sent IRS a paper signature
document. However, as noted earlier, at least one practitioner who produces
software for filing on- line advised users of its software not to use the
self- select PIN. Although neither we nor IRS know why about 20 percent
fewer taxpayers than the 8.5 million IRS projected to file on- line actually
did so, it seems reasonable to assume that this advice discouraged some
potential on- line filers.

During a CERCA meeting that we attended in May 2001, CERCA members told IRS
that they would like to see the practitioner PIN program brought back and
the number of ?shared secrets? for the self- select PIN program reduced from
two items to one, thus reducing the potential for rejection. IRS said that
it would make both of these changes for the 2002 filing season.

One of the most widely used ways that IRS provides service to taxpayers is
via its toll- free telephone operations. In April 2001, we reported that IRS
had received about 61 million telephone calls during the 2000 filing season
and that the quality of service provided was mixed and below IRS? longterm
goal of providing world- class telephone service. 23 During the 2001 filing
season, IRS received about 71 million calls and made limited progress
towards its long- term goal.

IRS had eight measures to assess its performance in providing telephone
service during the 2001 filing season, four relating to accessibility and
four relating to accuracy.

22 2001 Tax Return Filing Season, Hearings Before the Subcommittee on
Oversight of the Committee on Ways and Means, House of Representatives, 107
Cong. 80- 81 (2001). 23 IRS Telephone Assistance: Quality of Service Mixed
in the 2000 Filing Season and Below IRS? Long- Term Goal (GAO- 01- 189, Apr.
6, 2001). Limited Progress in

Providing World- Class Telephone Service

Page 20 GAO- 02- 144 IRS' 2001 Tax Filing Season

?Assistor level of service? is designed to show IRS? effectiveness in
providing callers access to an assistor.

?Assistor response level? reflects the percentage of taxpayers who waited 30
seconds or less to speak with an assistor.

?Abandon rate? shows the percentage of taxpayers who hang up while waiting
to speak with an assistor.

?Average speed of answer? shows the average number of seconds taxpayers wait
to speak to an assistor.

?Tax law quality rate? and ?account quality rate? show, for a representative
sample of calls, the percentage in which assistors follow all procedures and
provide correct answers.

?Tax law correct response rate? and ?account correct response rate? are
intended to more accurately reflect the taxpayers? experience in obtaining
correct information by discounting procedural errors and measuring the
percentage of calls in which IRS assistors provide correct responses to
inquiries.

As table 4 shows, IRS improved its performance to varying degrees in three
of the four accessibility measures, but fell short of its goals for all
four. The fourth measure showed a significant decrease in performance. As
for the accuracy of its telephone service, table 4 shows that IRS met or
exceeded two of its four goals for 2001 (tax law quality and account
quality) and did significantly better- 10 percent better- in account quality
compared to 2000. 24

24 Accessibility measures are based on comparable data for the weeks
beginning January 1 and ending July 14. Accuracy measures are based on data
for the months of January through June.

Page 21 GAO- 02- 144 IRS' 2001 Tax Filing Season

Table 4: IRS Filing Season Telephone Assistance Performance 2000 2001
Measure a Actual Actual Target

Accessibility Assistor level of service 60% 61% 69% Assistor response level
39 41 49 Abandon rate 22 18 12 Average speed of answer 237 seconds 273
seconds 132 seconds Accuracy Tax law quality rate 73% +/- 2% b 75% +/- 1% b
74% Account quality rate 59 +/- 2% b 69 +/- 1% b 63 Tax law correct response
rate

c 79 +/- 1% b 82 Account correct response rate

c 88 +/- 1% b 91 a Accessibility measures are based on actual counts.
Accuracy measures are based on representative samplings. b Actual values are
estimates, with 90 percent confidence.

c Comparable data do not exist. Source: IRS data.

We conducted a separate review of IRS? efforts to identify and address the
factors affecting its performance in providing toll- free telephone
assistance. We issued a report on the results of our review in December
2001. 25

25 GAO- 02- 212, Dec. 7, 2001.

Page 22 GAO- 02- 144 IRS' 2001 Tax Filing Season

The number of taxpayers who were assisted at TACs during the 2001 filing
season declined about 11 percent compared to the 2000 filing season, with a
variety of factors contributing to the decline. As a result of
reorganization, IRS changed the way TACs were organized and staffed and did
considerable hiring, mostly from within, but maintained the same network of
about 400 TACs where taxpayers could walk in to obtain forms, answers to tax
law questions, and assistance in preparing their returns. Also as a result
of the reorganization, IRS deferred measuring or improving quality in 2001.
However, a TIGTA review in early 2001 indicated that the accuracy of tax law
assistance provided by TACs- the only type of service measured- was poor.
Contrary to these findings, managers responsible for most TACs rated their
site?s performance positively in our telephone survey.

The number of taxpayers assisted at TACs declined about 11 percent- from 5.8
million contacts during the 2000 filing season to 5.1 million contacts
during the 2001 filing season, as shown in figure 1.

Figure 1: Taxpayers Assisted at TACs During the Filing Season

Note: The time periods covered by this figure each began on January 1 and
ended on April 24, 1999, April 22, 2000, and April 21, 2001.

Source: IRS data.

Our interviews of IRS officials identified the following factors that may
have contributed to the decline in the number of taxpayers assisted at TACs.
Fewer Taxpayers

Visited Taxpayer Assistance Centers; Limited Performance Data Indicate
Continuing Problems With the Quality of Assistance

Number of Taxpayers Assisted at TACs Declined

6.1 0.7

5.8 0.9

5.1 0.8

0 1 2 3 4 5 6 7

Millions of Taxpayers

1999 2000 2001

Filing Season

Total Walk-In Return Preparation sss

Page 23 GAO- 02- 144 IRS' 2001 Tax Filing Season

 With the intent of reducing the need to rely on compliance staff to help
meet the demand for assistance at TACs, IRS limited eligibility for
returnpreparation assistance- a very labor- intensive service- to taxpayers
with gross income of $41,000 or less 26 and tried to even out its workload
by scheduling appointments for return- preparation assistance. These changes
likely contributed to a decrease in the number of filing season contacts for
return- preparation assistance from 894,729 in 2000 to 786,103 in 2001.

 There were fewer tax law changes applicable to returns filed in 2001 than
in the recent past.

 As discussed in more detail later, more taxpayers were looking to IRS? Web
site for assistance. For example, the number of forms and publications
downloaded from IRS? Web site as of August 31, 2001, had more than doubled-
to about 279 million- compared to the same period in 2000.

 IRS had directed TACs to ?rigorously promote alternatives,? such as
volunteer sites that provide free return- preparation assistance. Although
these sites generally target their services toward certain groups of
taxpayers, such as low- income and elderly taxpayers, they generally do not
deny service based on income. According to data reported by volunteer sites,
their number increased from about 17, 600 in 2000 to about 18,400 in 2001.
The number of taxpayers who received assistance from volunteer sites
increased from about 3.3 million to about 3.6 million for the same periods.

The field assistance organization underwent considerable change in 2001
although the number of TACs remained about the same as in 2000. However,
except for establishment of an income limit on returnpreparation assistance
and the need to make appointments for this type of assistance, these changes
should have been largely transparent to taxpayers during the 2001 filing
season.

Before IRS? reorganization, the TACs reported to 33 district offices.
According to IRS officials, differences in the way the former walk- in sites
were organized and operated within each district caused inconsistencies in
the assistance provided to taxpayers. To provide more consistency in field
assistance, the TACs now report to the Wage and Investment Division?s Field
Assistance Director, through a network of 7 area and 34 territory offices.
According to field assistance officials, IRS began the year

26 According to IRS, this income limit will be lowered to $33,000 for 2002.
IRS Changed the Way

TACs Were Organized and Staffed in 2001

Page 24 GAO- 02- 144 IRS' 2001 Tax Filing Season

with about 1,000 technical field assistance employees and had hired another
504 as of March 16, 2001. Of those 1,504 technical employees, 1,041 were in
a new position- tax resolution representative (TRR)- that IRS established as
part of its reorganization. As of September 30, 2001, the number of
technical employees had increased to 1, 843, including 1,278 in the new TRR
position. IRS established the TRR position to help TACs meet the demand for
assistance during the filing season while reducing reliance on IRS?
compliance functions, such as Examination and Collection. 27

Although IRS filled the TRR positions primarily from qualified staff in
related job series, additional training was required. According to
officials, IRS surveyed the new staff to assess training gaps and
prioritized the delivery of abbreviated training to fill the gaps. However,
not all of the gaps were filled in time for the 2001 filing season. For
example, about 100 staff placed in TRR positions in January 2001, who needed
the full 6 weeks of required first- year training, received only 3 weeks of
that training.

Staffing the new field assistance management structure also required
considerable hiring and training. Managers of the former walk- in sites were
generally compliance staff who moved to the new Small Business and Self-
Employed Division. In addition to the new Director of Field Assistance, the
new management structure includes 7 area directors, 34 territory managers,
and 226 group managers who are the first- line managers responsible for the
day- to- day operations of one or more TACs. As of December 31, 2000, IRS
had filled all 7 area director positions, 29 of the territory manager
positions, and 154 of the group manager positions (3 more territory manager
position and 8 more group manager positions had been filled by September 30,
2001). According to IRS officials, about onehalf of the new managers had no
field assistance experience and some had no managerial experience. 28

27 Reducing the need for compliance staff to help provide assistance frees
up those staff to audit returns and collect tax debts. The number of
compliance staff years used to provide field assistance during the filing
season decreased from 386 staff years in 2000 to 162 staff years in 2001- a
decrease of 58 percent. According to field assistance officials, the 386
staff years for 2000 was understated by an unknown amount, because many
compliance staff reported their time under the compliance time reporting
system rather than the one used for field assistance.

28 According to information provided by IRS in November 2001, the managers
completed a new course entitled Managing a Taxpayer Assistance Center, which
was designed especially for them.

Page 25 GAO- 02- 144 IRS' 2001 Tax Filing Season

Important measures of IRS? performance in helping taxpayers who walk into
one of its TACs are the quality and timeliness of services provided.
Available quantitative information indicates that (1) the quality of tax law
assistance provided by TACs was poor, notwithstanding the more optimistic
views of TAC managers but (2) most taxpayers who visited a TAC received
timely assistance.

As discussed in our report on the 2000 filing season, 29 IRS? review of the
assistance provided to taxpayers who walked in with tax law questions showed
that accuracy was poor that year. 30 In that review, IRS employees posing as
taxpayers made 544 visits to walk- in sites. In a report on the results, IRS
noted, among other things, that 92 percent of the ?assistors

spoke to reviewers in a pleasant manner and tone of voice,? but 81 percent
of the reviewers? questions were not answered correctly. 31

Because of staffing and training challenges associated with the
reorganization, IRS did not measure or assess the quality of TAC service
during the 2001 filing season. However, a TIGTA review of tax law assistance
in early 2001 showed that accuracy was still poor. According to TIGTA, its
review involved 90 contacts at 47 TACs in which tax law questions were posed
to IRS representatives. In 7 of those 90 contacts, or 8 percent of the time,
service was denied. In 41 of the remaining 83 cases, or almost half the
time, TIGTA reviewers received inaccurate answers. Although TIGTA?s results
might indicate that the accuracy of tax law assistance provided by TACs,
while still poor, had improved compared to the results of IRS? reviews in
2000, such a comparison cannot be made because TIGTA used a different
methodology from the one used by IRS.

In contrast to TIGTA?s results, we estimate, based on our random sample,
that the vast majority of field assistance group managers would have spoken
positively about the accuracy of their sites? tax law assistance during the
2001 filing season. We estimate that managers responsible for

29 Tax Administration: Assessment of IRS? 2000 Tax Filing Season (GAO- 01-
158, Dec. 22, 2000). 30 Although TACs provide various services, IRS only
assessed the quality of tax law assistance. 31 About one fourth of the 81
percent (22 percent of the reviewers? questions) were instances in which
assistors were, in effect, denied service because their questions were not
answered. For example, reviewers were sometimes told to take a form or
publication and figure out the answer themselves. IRS and TIGTA Studies

Document Poor Tax Law Assistance, but Most Taxpayers Received Timely Help

TAC Quality

Page 26 GAO- 02- 144 IRS' 2001 Tax Filing Season

about 388 of the 413 TACs- or 94 percent- would have said that their sites
were successful or very successful in providing accurate answers to tax law
questions. However, managers we spoke with said that their opinions were
based on their knowledge of such considerations as staff, training, and
experience, since IRS currently lacks a system or measures for assessing tax
law accuracy. We discuss the group managers? views of their sites?
performance in more detail later.

IRS recognizes that the time that taxpayers wait to receive assistance is
also an important indicator of TAC performance. Accordingly, TAC managers
monitor daily average wait- times at each site in an attempt to minimize
average wait- times and maximize the percentage of taxpayers who receive
assistance within 15 minutes.

Most taxpayers waited 15 minutes or less to get assistance during the 2001
filing season, although the timeliness of TAC assistance declined slightly
from last year, based on various data. In 2000, IRS reviewers who posed as
taxpayers reported that wait- times exceeded 15 minutes for 9 percent of
their 544 contacts. For the 2001 filing season, however, TIGTA reviewers
reported waiting more than 15 minutes for 24 percent of their 90 contacts.
According to wait- time data reported by TACs, about 12 percent of the
nearly 3.5 million taxpayers whose wait- times were recorded waited more
than 15 minutes to get assistance during the 2001 filing season. 32
Similarly, IRS customer satisfaction surveys conducted during similar filing
season periods in 2000 and 2001 showed that most respondents waited less
than 15 minutes for assistance in both years, but a greater proportion of
respondents waited longer than 15 minutes in 2001 than in 2000, as
illustrated in figure 2. 33

32 Based on our calculations, the data reported by TACs did not account for
the wait- times of about 661,000 taxpayers. 33 Customer satisfaction surveys
were conducted every fifth week, and the results were summarized and
reported the following quarter by the survey contractor. According to IRS,
the survey methodology will be different for the 2002 filing season. Instead
of conducting a survey every fifth week, every customer who comes to the
screener?s desk in a TAC seeking assistance will be given a survey card. TAC
Timeliness

Page 27 GAO- 02- 144 IRS' 2001 Tax Filing Season

Figure 2: Wait- Time Reported by Customer Satisfaction Survey Respondents

Source: IRS Customer Satisfaction Survey, Field Assistance Intercept Survey,
June 2001.

Of the 10 areas that TAC customers were asked to rate as part of the
satisfaction survey, ?promptness of service,? was among the three areas
rated lowest (along with ?convenience of office location? and

?convenience of office hours?). 34 However, even those lowest- rated areas
received average scores above 6 on a 7- point rating scale, with 7 being the
highest rating. Overall, the combined survey results from January, February,
and March 2001 showed that 90 percent of the respondents rated their overall
satisfaction with assistance received as 6 or 7 on a 7- point scale, for an
average overall rating of 6.46.

The IRS and TIGTA teams that reviewed the quality of TAC tax law assistance
in 2000 and 2001, respectively, recommended several changes to improve
quality. For example, one recommendation resulting from IRS? reviews was
that IRS develop a comprehensive, year- round quality review program for
TACs. Field assistance officials informed us that IRS was studying how field
assistance quality should be measured and that IRS deferred making changes
to improve the accuracy of tax law assistance until the results of that
study were known.

34 The survey results also showed that the three areas rated highest on
average by respondents were, in order: ?showing the right attitude;?
?employee courtesy;? and ?treating

you fairly.? IRS Deferred Making

Quality Improvements During 2001

72 61

21 26 7 13

0 10 20 30 40 50 60 70 80

Percentage of respondents

<15 15-30 >30

Minutes

Jan-Mar 00 Jan-Mar 01

Page 28 GAO- 02- 144 IRS' 2001 Tax Filing Season

According to information provided by, or discussed with, field assistance
officials, IRS had successfully piloted a new method of measuring the
accuracy of tax law assistance and expected to begin using it in 2002. Under
this new method, contractor employees, posing as taxpayers in need of
assistance, are to visit TACs and ask pre- defined questions- similar to the
technique used by TIGTA and IRS quality reviewers during the last 2 filing
seasons. For 2002, contractor employees are to use questions selected from
those frequently asked by taxpayers who call IRS? toll- free telephone lines
until IRS develops a database of questions frequently asked by TAC
customers.

According to officials, IRS also plans to measure the quality of the account
and return- preparation assistance provided by TACs, using a cadre of seven
full- time IRS staff beginning in January 2002. IRS plans to fill these
positions and train selected staff by November 2001. These IRS reviewers
will evaluate the accuracy of prepared returns and adjustments made to
taxpayer accounts on a sample basis.

Because of the many changes related to TAC operations and the lack of
performance measures for assessing quality, we obtained the views of
managers responsible for a randomly selected sample of 84 TACs. 35 On the
basis of this sample, we estimate that managers would have been
overwhelmingly positive in rating their sites? performance in nearly every
category, in spite of a variety of obstacles that the managers said hampered
site performance to varying extents. As noted earlier, our estimate shows
that the managers were much more positive about the accuracy of their sites?
tax law assistance than would seem warranted by the results of TIGTA?s
study. However, the managers generally acknowledged that their ratings were
based more on their knowledge of such considerations as site staffing,
training, and experience than on any specific performance data.

The managers in our sample also cited several obstacles that hampered site
performance to a moderate or great extent.

 Managers cited problems with facilities at 38 percent of the sites. For
example, managers often mentioned that their sites? workspace was too

35 Many managers were responsible for more than one TAC. Most Managers Rated

Their Sites? Filing Season as Successful or Very Successful Despite
Obstacles

Page 29 GAO- 02- 144 IRS' 2001 Tax Filing Season

small to accommodate customers, ensure privacy in discussing taxpayer
information, or accommodate assigned staff or computers.

 Managers mentioned staffing problems at 32 percent of the sites. Many
managers said that their sites lacked sufficient field assistance staff,
which forced some sites to rely more heavily on compliance staff and
sometimes forced taxpayers to wait longer than an hour for assistance.

 Managers cited equipment problems at 31 percent of the sites. Managers
said that their sites lacked sufficient computers- especially those equipped
to access taxpayer accounts or file returns electronically. Some managers
said that staff virtually had to stand in line to use the only computer with
access to customer accounts.

 Managers mentioned training problems at 29 percent of the sites. Many
managers said that employees, especially those in the new TRR positions,
received only part- and in some cases, none- of their required first- year
training before the beginning of the filing season. Some said their
employees received training during the busy filing season, affecting
operations or requiring compliance backup. Still other managers said that
their employees had received the required training, but it did not
adequately cover complex tax law topics or topics that were unique to
certain geographic areas.

Although our telephone survey was designed to address only the filing
season, some managers informed us that IRS had resolved some problems after
the filing season. However, many responses implied that the sites? problems
remained unresolved. In that regard, field assistance officials told us in
October 2001 that IRS plans to resolve TAC computer hardware and software
shortages by 2004 and TAC facility design and space inadequacies by 2007.

The results of our telephone survey are highlighted in appendix III. IRS?
Web site on the Internet provides a vehicle whereby taxpayers can receive
assistance without having to contact IRS employees via more expensive modes
of communication, such as calling or visiting an IRS office. Among other
things, the Web site provides the potential to download hundreds of tax
forms and publications, contains current information on tax issues and
electronic filing, and gives taxpayers the opportunity to ask IRS tax law
and procedural questions via E- mail.

Various indicators related to IRS? Web site show that (1) overall usage
increased compared to the 2000 filing season; (2) IRS received fewer Email
questions from taxpayers but generally did a good job responding to
Indicators Related to

IRS? Web Site Are Generally Positive, but the Site Could Be More User-
friendly

Page 30 GAO- 02- 144 IRS' 2001 Tax Filing Season

the questions; and (3) the site?s availability and average delivery time as
measured by Keynote, an independent Web site rater, improved late in the
filing season after a slow start. 36 Our assessment of the Web site showed
that (1) a key IRS measure for tracking site usage- hits- is calculated in a
way that calls into question its usefulness and (2) the site is useable, but
could be more user- friendly. IRS is in the process of revising the content
and management of its Web site- changes that should address the problems we
identified.

IRS? Web site was used more in 2001 than in 2000 based on various measures.
A key measure IRS uses to track usage is hits. Although IRS data showed that
the number of hits, as of June 1, 2000 and 2001, had increased from 1.1
billion to 1.7 billion, the way hits are counted calls into question the
usefulness of that measure as an indicator of usage. For example, every time
a person accesses the home page of IRS? Web site, about 16 hits are counted
because the page and each graphic on it are counted as a single hit. A hit
is also counted every time a user moves to another page on the Web site.
Thus, an increase in the number of hits may not necessarily be due to
increased usage, but could be attributed to changes in the structure of the
home page or problems users are experiencing in finding what they want on
the site. Officials responsible for the Web site acknowledged that number of
hits is not a good measure of use and said that IRS is limited in the
measures it can use, in part due to privacy considerations.

The number of hits was the key measure that IRS used to assess the
performance of its Web site until 2001, when it added the number of forms
downloaded as a measure. That measure and another indicator compiled by a
non- IRS source showed the following growth in the use of IRS? Web site.

36 According to Keynote, availability is the percentage of time the Web
site?s home page downloads fully and average delivery time is the time it
takes for the home page to fully download from the time the user hits the
?enter? key on the keyboard. According to IRS, a home page is charged with
an error and therefore receives a lower percentage of availability even if a
graphic that has nothing to do with the user?s ability to get pertinent
content information from the page does not download properly. As such,
according to IRS, the measure does not reflect whether the Web site itself
is available or not. Also, according to IRS, errors can be the result of
heavy traffic on the Internet in general and may have nothing to do with the
home page. Overall Usage of IRS? Web

Site Increased

Page 31 GAO- 02- 144 IRS' 2001 Tax Filing Season

 About 279 million forms and publications had been downloaded from the site
as of August 31, 2001, an increase of 116 percent compared to the 129
million downloaded as of the same point in time in 2000.

 Nielsen// NetRatings, an Internet audience measurement service, reported
about 13 million unique visitors to IRS? site from February 4 through May 6,
2001- an increase of about 57 percent over the 8.3 million unique visitors
during the same time period in 2000. 37

Nielsen// NetRatings also reported that on the last day of the 2001 filing
season (April 16), 613,000 unique visitors accessed IRS? site. 38 According
to Nielsen// NetRatings, IRS? site has become the ?go to? online resource
for timely and useful tax information.

One area of IRS? Web site that was not used more in 2001 than in 2000 was
the feature that allows taxpayers to ask IRS tax law and procedural
questions via E- mail. As of June 1, 2001, IRS had received about 28 percent
fewer E- mail questions- about 159,000 compared to about 220,000 as of the
same time in 2000. IRS officials attributed this decrease to the absence of
major tax law changes affecting returns to be filed in 2001.

IRS uses three measures for assessing its performance in responding to Email
questions: (1) timeliness, (2) accuracy, and (3) customer satisfaction. IRS
data for the 2001 filing season show that IRS (1) improved its timeliness
compared to 2000, (2) maintained the same level of accuracy as in 2000, and
(3) received generally positive ratings from respondents to its customer
satisfaction survey.

IRS? goal is to respond to an E- mail question within 2 business days. IRS
data indicate that IRS almost always met its goal during the 2001 filing
season and that it did better than in 2000. For example, IRS took an average
of 0.8 business days to respond to taxpayers? E- mail questions in 2001,
compared to 1.3 business days in 2000.

37 ?Unique visitors? refers to the number of different persons who accessed
the Web site at least once during a day. 38 This information helps
illustrate how misleading the ?hits? measure can be. While Nielsen//
NetRatings reported 613, 000 unique visitors to IRS? site on April 16, 2001,
IRS reported about 60 million hits that day. Number of E- Mail

Questions Decreased; IRS Generally Did a Good Job Responding

Page 32 GAO- 02- 144 IRS' 2001 Tax Filing Season

To measure accuracy, a quality review group reviews a random sample of E-
mail responses. IRS data on the results of those reviews showed that 78
percent of E- mail questions were answered accurately during the 2001 filing
season. Given the statistical imprecision surrounding a sample, that rate is
consistent with the 76 percent accuracy rate in 2000 and IRS? 79 percent
goal for 2001.

IRS gives all E- mail customers an opportunity to respond to a customer
satisfaction survey, and about 1,400 did so between January 1 and April 16,
2001. The survey asked three questions about the E- mail service received.
In response to those questions, 96 percent of the respondents said that they
were satisfied with the time it took to get a response, 91 percent said that
they would use E- mail again in the future, and 75 percent said that the
response they received answered their question.

Keynote, an independent Web site rater and recognized authority on Internet
performance, reviewed the availability and average delivery time of IRS? Web
site and reported that IRS? performance was mixed during the 2001 filing
season.

Keynote reported that its measure of IRS? Web site?s availability ranged
from 65 to 75 percent between February 4 and March 27, 2001, and improved
dramatically after that- increasing from 67 percent on March 27 to 83
percent the next day. Availability continued to improve after that-
averaging 97.7 percent and 97.5 percent, respectively, for the weeks ending
April 8 and April 16, 2001, respectively. Average delivery time also
fluctuated considerably during the filing season. Keynote reported that
average delivery time ranged from 2.5 seconds to 3.5 seconds from March 1
through March 27, 2001, and improved to around 2 seconds thereafter through
April 15, 2001. Keynote reported that the improvements in availability and
average delivery time coincided with an upgrading of computer servers in the
latter part of March.

Our review of IRS? Web site disclosed certain aspects that, if improved,
would make the site more user- friendly.

Broken Links- We found 10 instances where links within the site forwarded us
to pages indicating that the document or sites could not be found. For
example, in the ?Earned Income Tax Credit? section, when clicking on the
question ?How Do I Get Advanced Earned Income Tax Credit Payments?, we were
sent to a page that had a link for getting a copy Web Site?s Availability
and

Average Delivery Time Improved After a Slow Start

Web Site Could Be More User- friendly

Page 33 GAO- 02- 144 IRS' 2001 Tax Filing Season

of Form W- 5 (Earned Income Credit Advance Payment Certificate). When we
clicked on that link, we were sent to the Fedworld. gov site where we got
the message ?Address Not Found?.

Missing Information- In the ?Around the Nation? section of the site, there
are one or more pages for all states but one. In our report on the 2000
filing season, we noted that many of those state pages did not include
information on the location of TACs- basic information that we thought
should be part of every state page. 39 When we checked on May 31, 2001, we
found that 14 state pages still had no data on TAC locations.

Obsolete Data- In our report on the 2000 filing season, we also noted that
certain data on the Web site were obsolete. We also found obsolete data in
2001, although not as much as in 2000. For example, in the ?IRS

Newsstand? part of the site, there is a section entitled ?Tax Calendar for
Small Business.? When we looked at that section in April 2001, we found a
calendar for 1999 but none for 2001. Our follow- up in May 2001 revealed
that IRS had since removed the 1999 calendar and added one for 2001.

Difficult to Navigate- We experienced several instances where we found the
Web site to be laborious and cumbersome to maneuver through. For example,
the site offers users the option to comment in a section called ?A Penny for
Your Thoughts.? When we clicked on the comment link, we were directed to a
page entitled ?Help, Comments, and Feedback.? We expected to be forwarded to
a page for comments since we clicked to reach a comment page. However, we
had to scroll down eight topics to get to the comment link.

Our overall assessment of IRS? Web site, on the basis of reviews we
performed throughout the year, is that the site contains a plethora of
information that provides taxpayers with options other than the traditional
methods of walking into an IRS site or calling on the telephone. Given its
size, we found it usable, but not user- friendly. In general, we found the
maze of links not well organized or straightforward. IRS officials
responsible for the Web site agreed with our assessment.

In our report on the 2000 filing season, we recommended that IRS (1) assign
clear responsibility in a central location for identifying and correcting
outdated and inconsistent Web site data and (2) develop

39 GAO- 01- 158.

Page 34 GAO- 02- 144 IRS' 2001 Tax Filing Season

minimum requirements for information to be included on the state pages.
According to IRS officials, IRS has taken several steps to address our
recommendation. Those steps include the following.

 The Electronic Tax Administration office within IRS instituted an interim
process that required executives to approve all new data and significant
changes to existing data.

 The Electronic Tax Administration office is implementing a content
management program to ensure that data on the Web site is current and
consistent.

 Beginning in 2002, the IRS Web site will have a central list that provides
the location of all TACs and their hours of operation.

 IRS awarded a 5- year, $33. 8 million contract to redesign and provide
support for IRS? Web site. Anticipated changes include incorporating
features for easier maneuverability.

We believe IRS? actions go a long way toward addressing our recommendation.
We plan to continue monitoring their implementation.

IRS? performance during the 2001 filing season varied depending on the
particular filing season activity being considered. For some activities-
processing and the Web site- available evidence indicated that IRS generally
did a good job and taxpayers generally received good service. Even so, there
were some exceptions involving processing by a lockbox, implementation of an
alternative signature program, and initial problems with the Web site?s
availability and delivery time. For other filing season activities-
telephone service and TAC assistance- there was evidence to indicate that
many taxpayers had problems reaching IRS by telephone and getting accurate
responses to their tax law questions from TAC personnel.

In each of these areas, IRS has taken steps to improve its performance-
steps that include initiatives to enhance the processing of paper returns
and encourage more electronic filing, a contract to redesign and provide
support for IRS? Web site, and a study on how best to measure the quality of
TAC assistance. However, there are opportunities for IRS to further enhance
its performance.

In some cases, there is available information from quality reviews done by
TIGTA and IRS and from the results of our survey of TAC managers that IRS
can use to immediately identify and begin implementing needed enhancements.
While we recognize that there are limits to what IRS can do to affect the
2002 filing season, there may be certain conditions Conclusions

Page 35 GAO- 02- 144 IRS' 2001 Tax Filing Season

identified by the quality reviews and our survey that lend themselves to
relatively easy resolution.

In other cases, needed enhancements may be going unidentified because IRS
has inadequate information on the effectiveness of processing initiatives
and the reasons why computer- prepared tax returns are filed on paper.
Without adequate information on the effectiveness of initiatives (those
undertaken in 2001 as well as any that may be undertaken in the future), IRS
does not have a sound basis for deciding whether to continue, curtail, or
revise the initiatives. Similarly, the absence of detailed research focused
on paper filers of computer- prepared returns leaves IRS without sufficient
information to identify specific impediments that must be overcome if that
large group of returns (40 million in 2001) is to be successfully converted
from paper to electronic.

In spite of our characterization of IRS? performance in processing returns
and refunds as smooth and recognizing that IRS does have comparable
historical data for workload indicators, we are concerned about our limited
ability to compare IRS? performance in 2001 against its goals for 2001 and
its performance in past years. While we support IRS? efforts to develop and
refine its performance measures to help assure that they are valid and
balanced, frequent or extensive changes deprive the various programs of
stability and comparability, thus hampering the ability to set or achieve
goals. We are currently assessing IRS? filing season performance measures
and expect to issue a report on our results in April 2002. However, we
obtained enough information on one filing season measure- Web site hits-
during this review to indicate that it is not a useful measure for assessing
the site?s performance.

We recommend that the Commissioner of Internal Revenue direct the
appropriate officials to do the following:

 To the extent practical, take action to correct tax law accuracy problems
reported by IRS and TIGTA in 2000 and 2001, respectively, in time to improve
the accuracy of tax law assistance provided by TACs in the 2002 filing
season rather than deferring quality improvements until data from the new
measurement system are available.

 Work with TAC managers to determine if there are any significant facility,
staffing, equipment, and training problems that are not addressed in
existing improvement plans, and develop appropriate plans for resolving
those problems. Recommendations for

Executive Action

Page 36 GAO- 02- 144 IRS' 2001 Tax Filing Season

 Directly survey tax practitioners and taxpayers who prepare returns on
computer but do not file electronically.

 Develop a plan for evaluating the implementation and effectiveness of
processing initiatives.

 Either discontinue the use of hits as a measure of the performance of IRS?
Web site or revise the way hits are calculated so that the measure more
accurately reflects usage.

Additional recommendations were included in our report on IRS? telephone
assistance during the 2001 filing season 40 and may be forthcoming as a
result of our ongoing review of IRS? filing season performance measures.

We requested comments on a draft of this report from IRS. We obtained
written comments in a December 13, 2001, letter from the Commissioner of
Internal Revenue (see app. V).

The Commissioner said that overall he concurred with our conclusion that the
processing of returns and the issuance of refunds went smoothly. He provided
some additional information related to IRS? handling of the problem with the
lockbox contractor in Pittsburgh, which we added to the report.

The Commissioner pointed out, as our report reflects, that developing and
revising performance measures is critical to IRS? ?Balanced Measures concept
as is aligning those measures with [IRS?] strategic plan.? At the same time,
he said that he understood our concerns that too frequent or extensive
changes to performance measures could invalidate subsequent comparison and
said that ?we factor that into changes necessary to accomplish our
transition.?

In response to our specific recommendations, the Commissioner said the
following:

 Improving tax law accuracy at TACs is ?a core competency of the Service
and we are not waiting for data to make incremental improvements.? To
improve quality during the 2002 filing season, according to the
Commissioner, IRS will focus on providing (1) consistent, standardized

40 GAO- 02- 212. Agency Comments

and Our Evaluation

Page 37 GAO- 02- 144 IRS' 2001 Tax Filing Season

services; (2) a larger and better trained workforce; and (3) improved
accessibility to taxpayer information for employees directly responsible for
case- specific assistance.

 IRS has plans to resolve all issues concerning TAC facilities, staffing,
equipment, and training identified in the report. To engage managers fully
in the improvement process, IRS will be implementing Manager Councils in
each of the seven field assistance areas.

 IRS agrees with the need for a survey of tax practitioners and taxpayers
who prepare their returns on computer but file on paper.

 As new processing initiatives are developed, IRS will incorporate a plan
in its project documentation to assess the implementation and effectiveness
of each initiative. While agreeing that IRS should review its initiatives to

?ensure their proper implementation and ensure they are targeted to achieve
the results intended,? the Commissioner said that the processing initiatives
discussed in the report are inherently beneficial because ?they

resolve or mitigate known problems or weaknesses.?

 IRS believes that it should continue to use hits as a measure of Web site
performance because the number of hits ?indicates site usage/ traffic? and
because IRS uses the number of hits ?to measure system performance and to
estimate system needs.? However, IRS said that it can improve its method for
counting hits when it implements a more sophisticated, comprehensive web
analytic program in January 2002. We agree that the current measure of hits
may have uses internally. However, unless computation of the measure is
appropriately revised, it should not be used as an indicator to describe
overall usage to external stakeholders, such as the press or the Congress.
In an April 26, 2001, news release, for example, IRS used the number of hits
to support its statements that the Web site

?attracts more taxpayers? and ?remains one of the most frequently visited
sites during tax season.? We believe that such uses of the measure, as
currently constructed, are inappropriate.

In further commenting on the Web site, IRS expressed concern that readers
would be mislead by our use of Keynote data on site availability unless we
more clearly explained what Keynote means by ?availability.?

We revised the report to do that.

Page 38 GAO- 02- 144 IRS' 2001 Tax Filing Season

In general, the various plans described in IRS? comments appear responsive
to our recommendations. We will be assessing IRS? implementation of those
plans during our review of the 2002 filing season.

We are sending copies of this report to the Chairmen and Ranking Minority
Members of the Senate Committee on Finance and the House Committee on Ways
and Means and the Ranking Minority Member of this Subcommittee. We are also
sending copies to the Secretary of the Treasury; the Commissioner of
Internal Revenue; the Director, Office of Management and Budget; and other
interested parties. We will make copies available to others on request.

This report was prepared under the direction of David J. Attianese,
Assistant Director. Other major contributors are acknowledged in appendix V.
If you have any questions about this report, contact Mr. Attianese or me on
(202) 512- 9110.

Sincerely yours, James R. White Director, Tax Issues

Appendix I: Problems IRS Experienced in 2001 Related to Processing

Page 39 GAO- 02- 144 IRS' 2001 Tax Filing Season

Given the size and complexity of IRS? processing operations, it is
inevitable that IRS may experience some problems during the filing season.
In addition to the more significant problems discussed in the body of the
report related to (1) returns and remittances destroyed or not processed by
the Pittsburgh lockbox and (2) rejected electronic submissions that limited
the impact of the self- select PIN program, IRS experienced some relatively
minor problems related to TeleFile, 1 the debt indicator database, 2 and the
printing of notices.

IRS experienced two problems with TeleFile. The first problem involved a
computer programming error that created some incorrect addresses in the
TeleFile database. According to IRS, the problem affected taxpayers who had
moved in 2000 and whose address included an apartment number. Because IRS
did not know if the problem would be resolved by April 16, 2001, it told
affected taxpayers, through an April 3 alert on its Web site, that they
should only use TeleFile if they expected a refund and would have the refund
directly deposited into their bank account. Otherwise, their refund or
balance due notice, if they owed money, would be mailed to the wrong
address. According to IRS, about 45,000 taxpayers could have been affected
by this problem, which, according to IRS, has been corrected for the 2002
filing season.

The second TeleFile problem involved incorrect customer numbers printed on
about 192,000 TeleFile packages by an IRS contractor. Customer numbers are
used by taxpayers to identify themselves when using TeleFile. The contractor
sent notices with corrected numbers to affected taxpayers to enable the
affected taxpayers to file via telephone. IRS notified taxpayers of the
problem via the Web site or over the telephone if they called IRS.

IRS reported a programming problem with the debt indicator database at the
start of the filing season that resulted in IRS? failure to identify some
individuals as having offsets before they were provided a Refund

1 TeleFile is the system through which taxpayers send returns directly to
IRS over telephone lines using a Touch- Tone telephone. 2 The debt indicator
database contains information on certain debts owed by taxpayers, such as
federal taxes, unpaid student loans, and child support. The indebted
taxpayer?s refund check is intercepted, and funds are redirected to pay the
debt. Appendix I: Problems IRS Experienced in

2001 Related to Processing TeleFile

Debt Indicator Database

Appendix I: Problems IRS Experienced in 2001 Related to Processing

Page 40 GAO- 02- 144 IRS' 2001 Tax Filing Season

Anticipation Loan. These are loans that private firms offer taxpayers who
file electronically. The loan amount is based on the amount of the
taxpayer?s expected refund, and the refund is sent directly to the loan
provider to pay off the loan. Firms that provide such loans rely on
information from IRS to determine whether there will be an offset against
the taxpayer?s refund. If there is going to be an offset, the firm will deny
the loan. IRS determines if an offset exists by checking against the debt
indicator database. Because of the programming problem, loan providers
received inaccurate information from IRS about the existence of an offset.
Some returns that were shown without offsets actually had offsets taken when
the refunds were processed. According to IRS and other sources, loan
providers were adversely affected by this problem in that they made loans
that were to be repaid by tax refunds that were not forthcoming. According
to IRS, the problem was first identified in January 2001 and corrected on
February 1, 2001.

IRS reported and corrected a computer- system related problem at the
beginning of the filing season related to the printing of notices. In one
case, about 8,000 payment due notices were not mailed in a timely manner. To
remedy the situation, IRS included Publication 3746 in the notice when it
was mailed to apologize for the delay and note that (1) the payment due date
was extended without impact on the amount due and (2) any additional penalty
and interest incurred due to the delayed mailing would be abated. In
addition, IRS informed employees of the issue, gave them procedures for
dealing with taxpayers who inquired about the delayed mailings, and
authorized them to abate any penalty or interest amounts when appropriate.
According to IRS, it identified and corrected this problem in less than 2
weeks- by mid- January 2001. Printing of Notices

Appendix II: IRS Efforts to Encourage Electronic Filing and Enhance Returns
Processing in 2001

Page 41 GAO- 02- 144 IRS' 2001 Tax Filing Season

In 2001, IRS took several steps to encourage electronic filing and
implemented various processing initiatives that were intended to either
improve its processing operations or enhance compliance. Our past work on
the Government Performance and Results Act of 1993 demonstrated the benefits
of evaluating initiatives that are aimed at improving performance. Likewise,
IRS guidance requires that plans for improving performance include steps to
monitor and evaluate the effectiveness of those actions. IRS? filing season
initiatives may have inherent benefits beyond improving performance, such as
increasing compliance, and it may take years until the benefits are fully
achieved. However, in some cases, IRS experienced problems that affected
implementation and limited the effectiveness of these initiatives. For
example, despite concerted efforts, IRS experienced problems in processing
Schedule K- 1s. We had a difficult time determining the effect of such
problems and the overall effectiveness of the initiatives. IRS? evaluations
of initiatives were limited, and IRS officials generally drew conclusions
about the effectiveness of initiatives based on broad numbers and trends.
IRS officials said that although initiatives may not be specifically
evaluated for effectiveness, they are evaluated for resource planning and
other purposes as part of filing season readiness reviews and the
development action plans to prepare for the next filing season. While we are
aware of this fact, we are unconvinced that evaluations as part of filing
season readiness reviews would be sufficient for determining effectiveness.

As it does every year, IRS undertook initiatives designed to encourage
electronic filing. The initiatives for 2001, in addition to the alternative
signature program discussed in the body of the report, included increasing
the number of forms and schedules to be electronically filed and allowing
tax practitioners to include notes to IRS on their electronic submissions.
As noted previously, IRS uses meetings with the tax practitioner community
to identify barriers and options for expanding electronic filing, but
because evaluations of effectiveness have been limited to date, IRS may be
missing key data needed to design an effective strategy for continuing to
increase electronic filing.

IRS added 23 more forms and schedules to the list of forms and schedules
that can be filed electronically and increased the number of certain forms
that could be filed electronically in 2001. IRS took these actions to enable
more taxpayers and practitioners to file more returns electronically and to
address an issue raised by some practitioners who say that they will file
all of their returns either electronically or on paper, but not both. IRS
estimated that with the 23 additional forms and schedules, 98 percent of
Appendix II: IRS Efforts to Encourage

Electronic Filing and Enhance Returns Processing in 2001

IRS Continued to Expand Electronic Filing Options, but Evaluations of
Effectiveness Have Been Limited

Appendix II: IRS Efforts to Encourage Electronic Filing and Enhance Returns
Processing in 2001

Page 42 GAO- 02- 144 IRS' 2001 Tax Filing Season

all individual taxpayers could file their income tax returns electronically
in 2001. IRS plans to add 38 more forms and schedules for 2002, which,
according to IRS, will enable over 99 percent of all individual taxpayers to
file their income tax returns electronically in 2002. 1

As of October 18, 2001, about 1,044,000 of the newly eligible forms and
schedules had been filed. 2 Of the 23 forms and schedules, only one, Form
2106EZ (Unreimbursed Employee Business Expenses), was filed electronically
to any great extent- about 930,000 times- and 18 were filed electronically
less than 3,000 times. As of October 2001, other than compile these
statistics, IRS had not done, nor did it plan to do, other evaluations of
this initiative. As a result, IRS does not know, for example, how many new
taxpayers filed electronically or how many practitioners decided to join the
electronic filing program because of the addition of these 23 forms and
schedules.

Besides allowing more forms and schedules to be filed electronically in
2001, IRS also increased the number of certain forms or schedules that can
be filed with each electronic return. For example, the number of Schedule Es
(Supplemental Income and Loss) that could be filed with each individual
electronic tax return was increased from 5 to 15.

Before 2001, there was no way for a practitioner to add a note to an
electronic submission to support or explain an entry on an electronic tax
return. Practitioners had suggested that this limitation provided a
disincentive to file electronically. In response to that input, IRS, for the
2001 filing season, allowed tax practitioners to write notes to support or
clarify an entry on an electronic tax return. Although practitioners
supported this initiative, it was not used extensively. IRS figures showed
that about 4,000 electronically filed tax returns contained such notes. What
IRS does not know is (1) how many, if any, of these 4,000 occurrences
involved a first time electronic filer, and (2) how many

1 This 99- percent figure relates to the electronic filing of individual
income tax returns and not business- related returns. Significant
restrictions for filing business returns electronically still exist. For
example, key business- related documents, including the basic corporate
income tax form (Form 1120), cannot be filed electronically. Some
practitioners have said that they will not participate in the electronic
filing program until they can file both individual and business- related
forms electronically.

2 This number cannot be equated to the number of returns because one return
could include more than one of the 23 forms and schedules.

Appendix II: IRS Efforts to Encourage Electronic Filing and Enhance Returns
Processing in 2001

Page 43 GAO- 02- 144 IRS' 2001 Tax Filing Season

practitioners decided to join the electronic filing program because of this
new feature.

IRS implemented various initiatives related to its processing of individual
tax returns. IRS experienced some problems that affected implementation, but
the extent to which these problems affected implementation and potential
benefits downstream was difficult to determine because of the limited data,
analysis, and evaluations of effectiveness.

To better ensure that persons are correctly determining their tax liability,
IRS verifies the SSNs on all individual income tax returns. Until the 2001
filing season, however, that verification did not include all secondary
SSNs. 3 IRS began verifying all secondary SSNs on paper and electronic
returns in the 2001 filing season and disallowing questionable exemptions
using its math- error authority. 4

IRS established an extensive, multistep process to determine whether to
accept a secondary SSN, which, according to IRS, met the requirements of the
Internal Revenue Code and minimized the burden placed on taxpayers. As a
first step in the process, IRS sent notices to taxpayers who it had
identified in 2000 as having invalid secondary SSNs to encourage them to
take steps to resolve the problem before filing their return in 2001. During
the 2001 filing season, all incoming joint returns were to be manually
reviewed to determine whether the taxpayer enclosed any information about a
name change, marriage, or divorce. IRS was to use this information to accept
the secondary SSN and process the return according to normal procedures. If
no such information was enclosed with the

3 IRS had attempted to verify secondary SSNs during the 2000 filing season
but reported significant problems that caused a high number of returns to
fall out as errors during processing. As a result, IRS discontinued the
verification process in 2000 and made changes for the 2001 filing season.

4 As IRS processes returns, it looks for computational errors made by
taxpayers or their representatives in preparing the returns. When IRS finds
such errors, it can automatically adjust the return through the use of its
math- error authority. In 1996, the Congress first authorized IRS to treat
invalid SSNs as math errors. IRS Implemented

Several Processing Initiatives, But Problems Arose and Evaluations Were
Limited

Secondary SSNs Were Validated

Appendix II: IRS Efforts to Encourage Electronic Filing and Enhance Returns
Processing in 2001

Page 44 GAO- 02- 144 IRS' 2001 Tax Filing Season

return, IRS was to check the name and secondary SSN against Social Security
Administration and IRS computer records. If a mismatch occurred, IRS was to
pull the return out of normal processing for manual research. According to
IRS officials, the notices that were issued prior to the filing season and
the research IRS performed on filed returns helped keep to a manageable
level the percentage of returns that had to be adjusted because of secondary
SSNs that did not match. Also, as part of this process, IRS did not alter
the joint filing status claimed on a return if it contained a mismatched
name and secondary SSN- this had been a concern at one point. However, using
its math- error authority, IRS would disallow the spousal exemption and any
earned income credit if there were a mismatch that IRS could not resolve
through research.

In September 2001, TIGTA reported that it found inconsistencies between IRS?
instructions for employees and instructions to taxpayers for secondary SSN
validation. 5 For example, TIGTA reported that the instructions to taxpayers
did not make clear that it was acceptable to send IRS documents, such as
marriage certificates, to support a name change. However, IRS? processing
instructions cited such documentation as valid support for accepting a
secondary SSN. TIGTA also found that IRS incorrectly denied about $1 million
in personal exemptions and earned income credits after erroneously
determining that about 1,800 secondary SSNs on returns that posted to IRS?
computer system during the week of April 29, 2001, were invalid. TIGTA?s
review showed that the secondary SSNs were valid and that the exemptions and
credits should have been allowed. According to TIGTA, these processing
errors occurred because IRS had not either (1) considered certain
documentation provided by the taxpayers to support the validity of the
secondary SSN or (2) followed its research and resolution procedures.

IRS officials reported few processing problems with this initiative, other
than those identified by TIGTA, and considered the initiative to be
effective based on broad trends, such as the number of returns that had
errors. In August 2001, IRS officials told us that although they had ideas
about what to include in a written plan to obtain data and perform more in-
depth analysis to help determine the effectiveness of the effort, they had
not yet written such a plan. Due to other priorities, IRS officials reported

5 The Internal Revenue Service Successfully Processed Individual Tax Returns
During the 2001 Filing Season, TIGTA, Reference No. 2001- 40- 192, Sept. 28,
2001.

Appendix II: IRS Efforts to Encourage Electronic Filing and Enhance Returns
Processing in 2001

Page 45 GAO- 02- 144 IRS' 2001 Tax Filing Season

that as of October 5, 2001, plans to design an evaluation and do more
indepth analysis of this initiative were postponed indefinitely.

For the 2001 filing season, IRS implemented the third- party- authorization
checkbox on individual income tax returns (excluding returns filed via
TeleFile) in response to input from external stakeholders. The checkbox
allowed taxpayers to authorize an individual return practitioner- but not
the company for which the practitioner worked- to serve as their
representative in resolving problems (such as math errors, inquiries about
missing tax return information, and questions regarding refunds or payments)
that arise during processing of the return. With the checkbox, taxpayers
could authorize this kind of limited representation without having to submit
a Form 2848 (Power of Attorney and Declaration of Representative) or Form
8821 (Tax Information Authorization). These other designations give
practitioners much more authority than the thirdparty- authorization
checkbox.

IRS has referred to the initiative as a ?burden reduction project,?
estimating that it would save taxpayers about 75,000 hours by not having to
prepare Form 8821, about 2 million hours by not having to prepare Form 2848,
and more than a million hours by IRS directly communicating with paid
practitioners to resolve issues rather than going through the taxpayer. In
addition, IRS asserted that the designation would allow processing issues to
be resolved immediately, thereby eliminating some postfiling contacts, such
as the mailing of notices.

From our perspective, this initiative was of limited success for three
reasons. First, representatives of the largest tax return preparation
company told us that the initiative was of little value to them, and it was
their official policy not to use the checkbox. They said that the company
blocked the checkbox field in the software used by its affiliates so that it
could not be used. The representatives explained that the authorization
required the name or designation of an individual and not a company to
discuss tax matters. Because this company relies heavily on temporary
employees who may or may not be working for the company after the filing
season, no one else in the company would be able to address issues related
to a particular return.

Second, according to IRS officials, IRS had limited capability to call
taxpayers or their practitioners to resolve issues during processing.
Therefore, according to IRS officials, in many cases, IRS still issued
notices instead of calling taxpayers or their practitioners directly during
Third- Party- Authorization

Checkbox Was Implemented

Appendix II: IRS Efforts to Encourage Electronic Filing and Enhance Returns
Processing in 2001

Page 46 GAO- 02- 144 IRS' 2001 Tax Filing Season

processing. IRS officials acknowledged that not all submission processing
centers are equipped to allow employees to call taxpayers or practitioners
to resolve issues that arise during processing. Of the two centers we
visited, one had a ?call- out? unit where IRS employees could make calls to
resolve issues that arose during processing. At that submission processing
center, IRS officials said that they received limited guidance and training
related to the third- party- authorization checkbox. While they did make
some calls out, they did not keep track whether they were calling a
practitioner, using the checkbox authorization, or calling a taxpayer, using
the taxpayer?s telephone number on the return. An IRS official told us that
some employees would rather call the taxpayer, even if the authorization box
were checked, because in all likelihood the taxpayer would be able to answer
the call and be more responsive than a practitioner. Employees at the other
center never had a call- out unit and, therefore, did not use the authority
to contact practitioners. Without the ability to call, IRS would have had to
send notices to taxpayers, negating the intended benefit of reducing such
contacts.

Third, because of discrepancies between instructions to taxpayers and IRS?
processing instructions, it was unclear if the primary intent of the
checkbox was to enable IRS employees to call practitioners or enable
practitioners to call IRS. Several practitioners we talked to said that they
were confused by the intent. The Director of Submission Processing
Operations told us that this initiative was driven by practitioners? desire
to call IRS on behalf of their clients versus IRS? desire to call the
practitioners. However, we believe that if some centers are calling
practitioners and others are not, this represents disparate treatment of
taxpayers and represents a failure on IRS? part to realize the full
potential of this authority. While practitioners we interviewed and ones
that we met with at IRS? tax forums expressed frustrations and had
complaints about IRS? implementation of this initiative, we heard strong
support of the initiative as well.

The Director of Submission Processing Operations told us that IRS has
delayed establishing additional call- out units as IRS continues to study
the cost, resources, and overall effectiveness of such units. According to
the Director, IRS is not sure if it is more cost effective to call taxpayers
or their practitioners during processing or to send a notice and have the
taxpayer or practitioner respond. As a result, IRS did not have the benefit
of this information prior to implementation of the initiative for comparison
purposes.

Appendix II: IRS Efforts to Encourage Electronic Filing and Enhance Returns
Processing in 2001

Page 47 GAO- 02- 144 IRS' 2001 Tax Filing Season

According to IRS, taxpayers checked the box on about 28 million returns
during the 2001 filing season, or less than half of the approximately 69
million returns prepared by practitioners. IRS used this broad number as the
basis for its assumption about the initiative's success. However, it lacked
supporting detailed analyses and a written plan for fully assessing
effectiveness to support claims of success. For example, IRS could tell us
the number of taxpayers who checked the authorization box, but did not have
in place a system for sampling or tracking how the use of the checkbox
authority facilitated the resolution of issues related to a return?s
processing. Nor could IRS officials tell us how many, if any, fewer notices
were issued as a result of the checkbox authority.

Trust, partnership, and subchapter S corporation filings have been steadily
increasing. For example, IRS reported that between 1997 and 2000,
partnership filings have increased almost 18 percent to over 2 million.
These filings are referred to as ?flow- through returns? because they pass
their income through the trust, partnership, or corporation to
beneficiaries, partners, and shareholders who are to claim the income on
their individual income tax returns. IRS estimates that billions of dollars
of such income (referred to as ?pass- through income?) are underreported
every year. IRS expected 8.5 million flow- through returns would be filed in
2001, passing through hundreds of billions of dollars to beneficiaries,
partners, and shareholders.

When the project to transcribe Schedule K- 1s was initiated, IRS expected
that 30 million schedules would be filed with these flow- through returns-
20 million on paper and 10 million electronically. IRS? goal was to
transcribe 100 percent of the schedules during the 2001 filing season. 6 The
transcribed data would later be matched against income shown on individuals?
income tax returns to help ensure that taxpayers accurately reported income
received from partnerships, trusts, and subchapter S corporations. The last
time IRS had transcribed Schedule K- 1 information was in 1995, when IRS
transcribed about 4.1 million schedules.

IRS officials took various actions to help ensure successful implementation
of this program. Specifically, they piloted the initiative at

6 IRS apparently overestimated the number of Schedule K- 1s it would receive
during the 2001 filing season. As of October 2001, IRS officials reported
receiving about 15 million Schedule K- 1s, or about one- half of the
anticipated volume. IRS officials are trying to determine the reasons behind
the discrepancies between anticipated and actual volumes. Schedule K- 1s
Were

Transcribed for Matching Later

Appendix II: IRS Efforts to Encourage Electronic Filing and Enhance Returns
Processing in 2001

Page 48 GAO- 02- 144 IRS' 2001 Tax Filing Season

four submission processing centers in the Summer of 2000, developed related
procedures, and tested computer programs at the beginning of the 2001 filing
season. Despite these and other efforts, IRS experienced significant
problems in processing the Schedule K- 1s. For example, in some cases, IRS
employees undercounted the Schedule K- 1s attached to a return, which
created problems later on when IRS tried to reconcile the number of Schedule
K- 1s in a particular batch of work. At that point, IRS would have to
renumber the Schedule K- 1s and ensure that all taxpayers were identified
and that information such as their names and addresses had been accurately
transcribed.

These processing problems contributed to (1) employee productivity levels
that were lower than IRS had estimated, and (2) the need for more processing
resources than anticipated. For example, the initial work plan estimated
that 136 schedules would be processed per hour, a projection that was later
reduced to 60 per hour. The actual productivity level realized was 50
documents per hour. IRS also estimated that it would require 585 staff years
to process the Schedule K- 1s. After taking into account the fewer staff
years needed because of lower than expected volumes and the larger number of
staff years needed because of reduced productivity, IRS officials estimated
a shortfall of about 50 staff years, with the shortfall being covered by
allocating resources from other program areas. IRS officials attributed the
failure to meet productivity levels to unrealistic expectations about
productivity.

Although IRS spent considerable effort working on implementation issues and
some evaluations were done, those evaluations have been limited. For
example, although IRS recalculated productivity to help determine the cost
of processing Schedule K- 1s, IRS has yet to compare this information
against original estimates of benefits. Although the true program benefits
are to occur later in the process, when IRS begins matching the Schedule K-
1 information to individuals? income tax returns and identifies potential
underreporting of income, IRS still could benefit from interim evaluations
of implementation issues and downstream benefits.

In conjunction with the United States Postal Service, IRS implemented an
initiative to use the Postal Service?s National Change of Address (NCOA)
database to update taxpayer addresses. IRS meant for the NCOA update to
provide quicker resolution of undelivered refunds. According to IRS, it
receives downloads of the NCOA database weekly and updates the addresses in
its computer systems. Where there was an exact name match, IRS was to update
its file with the taxpayer?s new address. IRS Began Using Postal

Service Database to Update Taxpayer Addresses

Appendix II: IRS Efforts to Encourage Electronic Filing and Enhance Returns
Processing in 2001

Page 49 GAO- 02- 144 IRS' 2001 Tax Filing Season

Before this initiative was implemented, there was disagreement among some
IRS officials over the usefulness of the NCOA database to IRS. The primary
concern of some officials was that the Postal Service could not supply IRS
with statistics regarding the accuracy of information contained in the
database. Despite these concerns, IRS never took action to assess the
database?s accuracy before using it for updates. Similarly, once
implemented, IRS never took action to evaluate the initiative?s
effectiveness.

Appendix III: TAC Managers? Views on Their Sites? Filing Season Performance

Page 50 GAO- 02- 144 IRS' 2001 Tax Filing Season

Because IRS did not measure the quality of TAC performance during the 2001
filing season, we obtained the views of group managers regarding (1) their
sites? success in meeting the demand for services and providing accurate or
timely assistance and (2) the extent that their sites? success was hampered
by factors related to such things as staffing, training, and operating
hours. Most managers rated their sites? performance overwhelmingly positive,
despite citing factors that hindered performance to a varying extent.

We randomly selected a sample of 84 TACs from a listing of 413 TACs that IRS
provided. IRS? listing did not include alternative or nontraditional sites,
such as mobile units and kiosks. We interviewed some managers more than
once, because they were responsible for more than one site in our sample.
However, we asked them about each site separately. All percentage estimates
from our sample of managers have sampling errors of plus or minus 10
percentage points or less, with 95 percent confidence.

For each TAC in our sample, we asked the responsible group manager to rate
the site?s success and briefly explain the basis for that rating, in each of
the following areas:

 meeting the total demand for assistance as well as the specific demands
for tax law assistance, return- preparation assistance, and tax forms or
publications;

 keeping the average wait- time for assistance below 15 minutes;

 providing accurate answers to tax law questions; and

 providing accurate return- preparation assistance. Table 5 presents our
nationwide percentage estimates for managers? views regarding success in
each of these areas for all 413 TACs. Appendix III: TAC Managers? Views on
Their

Sites? Filing Season Performance Most TAC Managers Viewed Their Sites?
Filing Season as Successful or Very Successful

Appendix III: TAC Managers? Views on Their Sites? Filing Season Performance

Page 51 GAO- 02- 144 IRS' 2001 Tax Filing Season

Table 5: Percentage Estimates of TAC Managers? Views of Their Sites? Filing
Season Success For the 2001 filing season, to what extent was your site
successful with respect to

Very successful Successful

Neither successful

nor unsuccessful Unsuccessful Very

unsuccessful No opinion

Meeting the total demand for all types of TAC assistance 42 48 8 2 0 0
Meeting the demand for tax law assistance 46 46 6 1 0 0 Meeting the demand
for returnpreparation assistance 51 36 11 2 0 0 Meeting the demand for tax
forms or publications 80 18 2 0 0 0 Keeping the average wait- time for
assistance below 15 minutes 68 25 5 1 1 0 Providing accurate answers to tax
law questions 48 46 6 0 0 0 Providing accurate returnpreparation assistance
64 31 5 0 0 0

Source: GAO data and analysis.

We also asked managers to explain their positive ratings (? successful? or

?very successful?) and their negative ratings (? unsuccessful? or ?very

unsuccessful?) for the various categories in table 5. The explanations that
follow are based on the actual responses from the managers of the 84 sites
in our sample. The numbers of sites shown in parentheses following each
reason do not add up to the number of sites in our sample because (1)
managers often cited more than one reason in explaining their rating and (2)
some managers provided no reason.

Managers cited the following reasons for rating 75 of the 84 sites ?very

successful? or ?successful.?  Employees were adequately trained, capable,
and/ or experienced in

handling all types of assistance. (51)

 Employees responded appropriately so that few referrals to other sources
of assistance were necessary. (8)

 The same- day appointment system for return preparation ensured
availability of all types of service. (5)

 Sites attained success despite insufficient staff, equipment, and space.
(5) Managers cited the following reason for rating 2 of the 84 sites

?unsuccessful.?

Meeting the total demand for all types of TAC assistance

Appendix III: TAC Managers? Views on Their Sites? Filing Season Performance

Page 52 GAO- 02- 144 IRS' 2001 Tax Filing Season

 Employees were insufficiently trained, and sites had insufficient
operating hours. (2)

Managers cited the following reasons for rating 78 of the 84 sites ?very

successful? or ?successful.?  Employees were adequately trained, capable,
and/ or experienced in

dealing with tax law questions. (77)

 Employees responded appropriately to technical questions without referrals
to other IRS areas, such as Compliance. (4)

 Low demand existed for tax law assistance. (2)

 Success was attained despite small work space. (1) One manager cited the
following reason for an ?unsuccessful? rating.

 Compliance withdrew back- up support, necessitating many referrals to
other sources of assistance. (1)

Managers cited the following reasons for rating 73 of the 84 sites ?very

successful? or ?successful.?  Employees were adequately trained, capable,
and/ or experienced in

providing return- preparation assistance. (38)

 The same- day appointment system facilitated the return- preparation
assistance process. (18)

 The income limitation lowered customer demand for return preparation. (10)

 Non- IRS sources of help were available to provide assistance when
taxpayer income exceeded TAC limitations. (10)

 Electronic filing helped ensure ease, speed, and accuracy of return
preparation. (8)

Managers rated two sites ?unsuccessful? because there were an insufficient
number of employees. One of the managers also cited a reduction of
Compliance back- up support as a hindrance (2)

Managers cited the following reasons for rating 82 of the 84 sites ?very

successful? or ?successful.?  Required quantity of tax forms and
publications were ordered and stored.

(70) Meeting the demand for

tax law assistance Meeting the demand for return- preparation assistance

Meeting the demand for tax forms or publications

Appendix III: TAC Managers? Views on Their Sites? Filing Season Performance

Page 53 GAO- 02- 144 IRS' 2001 Tax Filing Season

 Adequate display racks and shelving were available, making it easier for
customers to see and select forms. (25)

 Employees were adequately trained, capable, and/ or experienced in
handling forms and publications. (20)

 Ample storage space was available. (4)

 Demand for forms or publications was low, or free parking or drive up
service made it easier to obtain them. (3)

No sites were rated ?unsuccessful? or ?very unsuccessful? in this area.
Managers cited the following reasons for rating 78 of the 84 sites ?very

successful? or ?successful.?  Employees were adequately trained, capable,
and/ or experienced, which

helped minimize wait- time. (35)

 An automated system tracked wait- times, numbers of customers, and types
of queries, keeping managers alert to wait- times. (7)

 Low customer demand existed, and employees made immediate referrals to
other sources of assistance. (4)

Managers rated 2 of the 84 sites as ?unsuccessful? or ?very unsuccessful.?

 Appointment scheduling caused problems that affected wait- time in a
oneperson office, and the wait- time often exceeded 15 minutes. (2)

Managers cited the following reasons for rating 79 of the 84 sites ?very

successful? or ?successful.?  Employees were adequately trained, capable,
and/ or experienced. (66)

 Few referrals for assistance were made to other sources of assistance, and
employees used a frequently- asked- questions guide to help identify the
proper response. (3)

No sites were rated ?unsuccessful? or ?very unsuccessful? in this area.
Managers cited the following reasons for rating 80 of the 84 sites ?very

successful? or ?successful.?  Employees were adequately trained, capable,
and/ or experienced in

providing return- preparation assistance. (39) Keeping the average waittime

for assistance below 15 minutes

Providing accurate answers to tax law questions

Providing accurate returnpreparation assistance

Appendix III: TAC Managers? Views on Their Sites? Filing Season Performance

Page 54 GAO- 02- 144 IRS' 2001 Tax Filing Season

 Lower than expected rejection rates were experienced for electronically
filed returns. (25)

 Electronic filing and return- preparation software helped ensure ease,
speed, and accuracy in preparing returns. (22)

 Non- IRS sources of assistance were used to help meet demand. (3)

 The same- day appointment system was helpful, and demand for electronic
filing was light. (5)

No sites were rated ?unsuccessful? or ?very unsuccessful? in this category.
For each TAC in our sample, we asked managers for their views on the extent
to which each of the following factors hampered their sites? success during
the 2001 filing season.

 Facilities

 Staffing

 Training

 Days or hours of operation

 Equipment (e. g., computers)

 Supplies (e. g., ink cartridges, paper)

 Tax forms or publications

 Policies or procedures

 Any other item they wished to identify Although most TAC managers rated
their sites? performance during the 2001 filing season as ?successful? or
?very successful,? most also cited factors (mostly related to facilities,
staffing, equipment, and training) that hindered their sites? success, at
least to some extent.

Table 6 presents our nationwide percentage estimates for managers? views
regarding the extent that factors hampered success for all 413 TACs. Most
Managers Cited

Factors That Hampered Their Sites? Success

Appendix III: TAC Managers? Views on Their Sites? Filing Season Performance

Page 55 GAO- 02- 144 IRS' 2001 Tax Filing Season

Table 6: Percentage Estimates of TAC Managers? Views on the Extent to Which
Certain Factors Hampered Their Sites? Filing Season Success

To what extent was your site?s success for the 2001 filing season hampered
by factors related to To a great extent To a moderate

extent To some extent Not at all

Facilities 23 15 35 27 Staffing 20 12 26 42 Training 6 23 32 39 Days or
hours of operation 4 7 2564 Equipment (e. g., computers) 14 17 31 38
Supplies (e. g., ink cartridge, paper) 4 7 1773 Tax forms or publications 0
2 1483 Policies or procedures 6 15 38 40 Other (specify)- 3 managers
responded 67 33 0 0

Source: GAO data and analysis.

We also asked managers for explanations whenever they answered ?to a great
extent? or ?to a moderate extent.? The explanations that follow are based on
the actual responses from the managers of the 84 sites in our sample. The
numbers in parentheses after each reason are the number of sites for which
that reason was cited. As was the case with table 5, some managers cited
more than one reason in explaining their responses.

Managers for 32 of the 84 TACs selected ?moderate? or ?great extent.? The
following reasons were cited.

 TAC facilities were too small. (29)

 General characterization. (13)

 Customer waiting areas were not large enough. (14)

 Workspace for assigned employees was not sufficient. (11)

 Private space for discussing taxpayer information was not sufficient. (10)

 Storage space for forms and/ or publications was not sufficient. (3)

 Space to accommodate computer to facilitate electronic filing was not
sufficient. (2)

 TAC facilities were not adequately designed or laid out. (12)

 General characterization. (4)

 Available equipment, including computer, printer, or copier, were not
located close enough to work area. (5) Facilities

Appendix III: TAC Managers? Views on Their Sites? Filing Season Performance

Page 56 GAO- 02- 144 IRS' 2001 Tax Filing Season

 The service counter design was outdated or did not provide the privacy
needed to protect taxpayer information, or the workspace was not laid out
properly to accommodate individual workstations. (3)

 Facility- related issues other than size or layout. (5)

 Geographical location was remote and/ or inconvenient, free parking was
not available, facility doors were not accessible to disabled customers and
facility heating was not adequate. (5)

Managers for 27 of the 84 TACs selected ?moderate? or ?great extent?. The
following reasons were cited.

 General characterization based on insufficient number of field assistance
employees. (20)

 Employees were absent because of training, or positions were vacant due to
turnover. (3)

 Insufficient back- up from IRS compliance employees or other non- field
assistance employees. (7)

 Newly hired employees lacked experience. (2) Managers for 24 of the 84
TACs selected ?moderate? or ?great extent.? The following reasons were
cited.

 Employees needed more training; they received only part of required
training or needed training beyond that normally required. (18)

 Employees needed better training because either training was not effective
or subjects covered during the training were not adequately addressed. (4)

 Training was not timely, and required training was not provided
sufficiently in advance of the filing season. (7)

Managers for 9 of the 84 TACs selected ?moderate? or ?great extent.? The
following reasons were cited.

 More days or hours of operation were needed. (3)

 Some extended hours were underused. (3)

 Alternative hours were needed. (2)

 Too many hours were worked or hours were lost due to federal building
closing. (1)

Managers for 26 of the 84 TACs selected ?moderate? or ?great extent.? The
following reasons were cited. Staffing

Training Days or hours of operation Equipment

Appendix III: TAC Managers? Views on Their Sites? Filing Season Performance

Page 57 GAO- 02- 144 IRS' 2001 Tax Filing Season

 Inadequate computers were available. (23)

 Insufficient numbers of computers were available. (13)

 Insufficient computers with a capability to access taxpayer account
information were available. (7)

 An insufficient number of computers were capable of filing returns
electronically. (3)

 Computers were outdated or malfunctioned. (7)

 Printers were inadequate or insufficient in number. (4)

 Access to a printer for a specialized collection system was shared. (3)

 Lacked access to a printer capable of printing taxpayer account
information. (1)

 Other equipment- related problems were experienced. (11)

 Equipment/ technical support was lacking. (2)

 Facsimile machines were not available in sufficient numbers. (2)

 Equipment utilization limited by office design. (2)

 Other equipment deficiencies, such as the lack of return- preparation
software or a copier, were cited. (5)

Managers for 9 of the 84 TACs selected ?moderate? or ?great extent.? The
following reasons were cited.

 Overall shortage of printer ink cartridges and requisitioning problems
existed. (9)

Managers for 2 of the 84 TACs selected ?moderate? or ?great extent.? The
following reasons were cited.

 Storage space was insufficient. (1)

 Forms and publications were not available at the beginning of the filing
season. (1)

Managers for 18 of the 84 TACs selected ?moderate? or ?great extent.? The
following reasons were cited.

 Problems with, or negative reactions to, the new income limitation on
return- preparation assistance and/ or the new same- day appointment process
for that assistance. (14)

Managers for 3 of the 84 TACs cited other factors that hampered their site?s
success to a ?moderate? or ?great extent,? such as the need for security
guards. Supplies

Tax forms or publications Policies or procedures Other

Appendix IV: Comments From the Internal Revenue Service Page 58 GAO- 02- 144
IRS' 2001 Tax Filing Season

Appendix IV: Comments From the Internal Revenue Service

Appendix IV: Comments From the Internal Revenue Service Page 59 GAO- 02- 144
IRS' 2001 Tax Filing Season

Appendix IV: Comments From the Internal Revenue Service Page 60 GAO- 02- 144
IRS' 2001 Tax Filing Season

Appendix IV: Comments From the Internal Revenue Service Page 61 GAO- 02- 144
IRS' 2001 Tax Filing Season

Appendix IV: Comments From the Internal Revenue Service Page 62 GAO- 02- 144
IRS' 2001 Tax Filing Season

Appendix IV: Comments From the Internal Revenue Service Page 63 GAO- 02- 144
IRS' 2001 Tax Filing Season

Appendix IV: Comments From the Internal Revenue Service Page 64 GAO- 02- 144
IRS' 2001 Tax Filing Season

Appendix IV: Comments From the Internal Revenue Service Page 65 GAO- 02- 144
IRS' 2001 Tax Filing Season

Appendix V: GAO Contacts and Staff Acknowledgments

Page 66 GAO- 02- 144 IRS' 2001 Tax Filing Season

James White (202) 512- 9110 David Attianese (202) 512- 9110

In addition to those named above, Bob Arcenia, Pat Brewer, Rudy Chatlos, Ron
Heisterkamp, Janelle Hu, Brian James, Ron Jones, John Lesser, and Joanna
Stamatiades made key contributions to this report. Appendix V: GAO Contacts
and Staff

Acknowledgments GAO Contacts Acknowledgments

(440014)

The General Accounting Office, the investigative arm of Congress, exists to
support Congress in meeting its constitutional responsibilities and to help
improve the performance and accountability of the federal government for the
American people. GAO examines the use of public funds; evaluates federal
programs and policies; and provides analyses, recommendations, and other
assistance to help Congress make informed oversight, policy, and funding
decisions. GAO?s commitment to good government is reflected in its core
values of accountability, integrity, and reliability.

The fastest and easiest way to obtain copies of GAO documents is through the
Internet. GAO?s Web site (www. gao. gov) contains abstracts and full- text
files of current reports and testimony and an expanding archive of older
products. The Web site features a search engine to help you locate documents
using key words and phrases. You can print these documents in their
entirety, including charts and other graphics.

Each day, GAO issues a list of newly released reports, testimony, and
correspondence. GAO posts this list, known as ?Today?s Reports,? on its Web
site daily. The list contains links to the full- text document files. To
have GAO e- mail this list to you every afternoon, go to www. gao. gov and
select "Subscribe to daily e- mail alert for newly released products" under
the GAO Reports heading.

The first copy of each printed report is free. Additional copies are $2
each. A check or money order should be made out to the Superintendent of
Documents. GAO also accepts VISA and Mastercard. Orders for 100 or more
copies mailed to a single address are discounted 25 percent. Orders should
be sent to:

U. S. General Accounting Office P. O. Box 37050 Washington, D. C. 20013

To order by Phone: Voice: (202) 512- 6000 TDD: (202) 512- 2537 Fax: (202)
512- 6061

GAO Building Room 1100, 700 4th Street, NW (corner of 4th and G Streets, NW)
Washington, D. C. 20013

Contact: Web site: www. gao. gov/ fraudnet/ fraudnet. htm, E- mail:
fraudnet@ gao. gov, or 1- 800- 424- 5454 or (202) 512- 7470 (automated
answering system).

Jeff Nelligan, Managing Director, NelliganJ@ gao. gov (202) 512- 4800 U. S.
General Accounting Office, 441 G. Street NW, Room 7149, Washington, D. C.
20548 GAO?s Mission

Obtaining Copies of GAO Reports and Testimony

Order by Mail or Phone Visit GAO?s Document Distribution Center

To Report Fraud, Waste, and Abuse in Federal Programs

Public Affairs
*** End of document. ***