This is the accessible text file for GAO report number GAO-08-38 
entitled 'Tax Administration: 2007 Filing Season Continues Trend of 
Improvement, but Opportunities to Reduce Costs and Increase Tax 
Compliance Should be Evaluated' which was released on December 17, 
2007. 

This text file was formatted by the U.S. Government Accountability 
Office (GAO) to be accessible to users with visual impairments, as part 
of a longer term project to improve GAO products' accessibility. Every 
attempt has been made to maintain the structural and data integrity of 
the original printed product. Accessibility features, such as text 
descriptions of tables, consecutively numbered footnotes placed at the 
end of the file, and the text of agency comment letters, are provided 
but may not exactly duplicate the presentation or format of the printed 
version. The portable document format (PDF) file is an exact electronic 
replica of the printed version. We welcome your feedback. Please E-mail 
your comments regarding the contents or accessibility features of this 
document to Webmaster@gao.gov. 

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately. 

United States Government Accountability Office: 
GAO: 

Report to Congressional Requesters: 

November 2007: 

Tax Administration: 

2007 Filing Season Continues Trend of Improvement, but Opportunities to 
Reduce Costs and Increase Tax Compliance Should be Evaluated: 

GAO-08-38: 

GAO Highlights: 

Highlights of GAO-08-38, a report to congressional requesters. 

Why GAO Did This Study: 

In 2007, the Internal Revenue Service (IRS) will spend over 
$3 billion to process returns and provide taxpayer service. Effective 
service can reduce taxpayersí burden of complying with tax laws and, 
many tax experts believe, may improve compliance. GAO was asked to 
assess IRSís performance relative to 2007 goals and prior yearsí 
performance including identifying actions that might generate 
efficiencies and increase compliance. GAO analyzed IRS performance 
data, reviewed IRS operations at various locations, and interviewed IRS 
and paid preparer representatives.

What GAO Found: 

IRS improved most filing season services during 2007, continuing a 
longer-term trend. Tax return processing exceeded last yearís 
performance by most measures. Electronic filing grew at a faster rate 
than IRS anticipated and continued to generate savings. Access to IRS 
telephone assistors was comparable to last year, and the accuracy of 
responses to questions remained at about 90 percent. The performance of 
IRSís Web site improved in several measures, such as customer 
satisfaction. Continuing past trends, more taxpayers used volunteer 
sites, which are less costly than IRSís walk-in sites. 

Despite these improvements, IRS could reduce the number of paper tax 
returns processed and also transcribe all lines from the residual paper 
returns, making that data more available for enforcement. Two options 
for reducing paper processing are electronic filing mandates, 
previously suggested by GAO, and bar coding, which could be required 
for paper returns prepared on computers and reduces processing costs. 
Currently, because of the cost, IRS does not transcribe all lines from 
paper returns. Further, IRS policy is to post the same lines from 
electronic and paper returns to its enforcement databases. As a result, 
IRS does not use all tax return information in its automated compliance 
checking programs. However, IRS does not know the actions needed to 
require software vendors to include bar codes on printed tax returns; 
the benefits, in terms of processing savings and improved enforcement, 
of having all return data available electronically; or how much 
electronic filing would have to increase, either through mandates or 
bar coding, for the benefits of transcribing all residual paper returns 
to exceed the costs. 

Despite more reliance on its volunteer program, IRS has not evaluated 
its effectiveness at reaching underserved taxpayers. Further, IRS may 
be missing an opportunity to assess the quality of return preparation 
assistance at its walk-in sites through contact recording, a system IRS 
uses to record and assess the quality of other interactions between its 
employees and taxpayers.

Figure: IRSís 2007 Filing Season Activities: 

What GAO Recommends: 

GAO recommends that IRS determine the actions needed to require for 
software vendors to include bar codes on printed returns, the benefits 
of having all return information available electronically, and how much 
electronic filing would have to increase for the benefits of 
transcribing all paper returns to exceed the costs. GAO also recommends 
that IRS develop estimates of the effectiveness of its volunteer 
program, and determine the feasibility of using contact recording to 
monitor return assistance quality at walk-in sites.

In response, the Acting IRS Commissioner agreed with our 
recommendations and outlined the actions that IRS would take. 

To view the full product, including the scope and methodology, click on 
[hyperlink, http://www.GAO-08-38]. For more information, contact James 
White at (202) 512-9110, or WhiteJ@gao.gov. 

[End of section] 

Contents: 

Letter: 

Results in Brief: 

Scope and Methodology: 

Background: 

Electronic Filing Continued to Generate Savings, Processing Performance 
Improved, but CADE Was Delayed, and Opportunities Exist to Reduce Costs 
and Improve Compliance: 

IRS Has Limited Information on Paid Preparers Who Prepare the Majority 
of Tax Returns, but Has Efforts Underway to Acquire More: 

Access to IRSís Telephone Assistors Was Comparable to Last Year, 
Accuracy Remains High, and IRS Is Reviewing Options for Reducing Excess 
Capacity at Call Sites: 

Web Site Performance Remains High, and Initiatives Are On-going to 
Enhance Web Capabilities and User Experience: 

Taxpayer Assistance Continued to Grow at Volunteer Sites and Decline at 
IRSís Walk-in Sites; While Quality has Improved in Some Areas, More 
Reliable Data Are Needed: 

The Value of TAB Depends on Funding and Ability to Link Taxpayer 
Service to Compliance: 

Conclusions: 

Recommendations for Executive Action: 

Agency Comments and Our Evaluation: 

Appendix I: IRSís Processing Performance Relative to 2001-2006 
Performance and 2007 Goals: 

Appendix II: Comments from the Internal Revenue Service: 

Tables: 

Table 1: IRS Telephone Assistors Performance, 2001ó2007 Filing Seasons: 

Table 2: IRS Telephone Assistor Accuracy Performance, 2001ó2007 Filing 
Seasons: 

Table 3: IRS Web Site Use, 2006 and 2007 (data are in thousands): 

Table 4: IRSís Fiscal Year 2005 Estimated Unit Costs: 

Table 5: IRSís Processing Performance, Fiscal Years 2001ó2007: 

Figures: 

Figure 1: IRSís 2007 Filing Season Activities: 

Figure 2: Percentage of Returns with a Paid Preparer Signature: 

Figure 3: How IRS Handled Calls During 2007: 

Figure 4: Telephone Assistance and Paper Inventory Resources by 
Workload: 

Figure 5: Trends in Face-to-face Return Preparation, 2001 through 2007: 

Figure 6: Contact Recording at Walk-in Sites: 

Abbreviations: 

AMS: Account Management System: 

BSM: Business Systems Modernization: 

CADE: Customer Account Data Engine: 

IRS: Internal Revenue Service: 

RRA 98: IRS Restructuring and Reform Act of 1998: 

SSN: Social Security numbers: 

TAB: Taxpayer Assistance Blueprint: 

TETR: Telephone Excise Tax Refund: 

TIGTA: Treasury Inspector General for Tax Administration: 

[End of section] 

United States Government Accountability Office: 
Washington, DC 20548: 

November 15, 2007: 

The Honorable Max Baucus: 
Chairman: 
The Honorable Charles E. Grassley: 
Ranking Member: 
Committee on Finance: 
United States Senate: 

The Honorable Jim Ramstad: 
Ranking Member: 
Subcommittee on Oversight: 
Committee on Ways and Means: 
House of Representatives: 

The quality of the Internal Revenue Serviceís (IRS) filing season 
service is important for several reasons. Effective service can reduce 
the burden for taxpayers, in terms of time and money, of complying with 
tax laws. Further, many tax experts believe that quality service has a 
positive effect on compliance, although it is difficult to demonstrate 
empirically. 

Taxpayer service also matters because IRS allocated over 30 percent of 
its $11 billion budget to provide such service for fiscal year 2007. 
IRS spends most of this during the filing season when most taxpayers 
interact with the agency.[Footnote 1] Filing season service includes 
individual income tax return processing as well as telephone, Web site, 
and face-to-face assistance. In past reports and testimonies we said 
that IRS has made significant progress improving taxpayer service since 
the passage of the IRS Restructuring and Reform Act of 1998 (RRA 98). 
[Footnote 2] In an effort to continue improving service, IRS recently 
released, as mandated by the Congress, its Taxpayer Assistance 
Blueprint (TAB). TAB is intended to outline a 5-year comprehensive 
strategy for improving taxpayer service while minimizing cost and 
determining the link between service and compliance.[Footnote 3] 

In addition to IRS, paid tax return preparers are another important 
provider of service to taxpayers during the filing season, preparing 
over half of individual income tax returns. In previous reports, we 
have expressed concern about the quality of paid preparer service and 
lack of information IRS has about the extent of problems.[Footnote 4] 

In light of the tax filing seasonís cost and importance to taxpayers, 
you asked us to assess IRSís 2007 filing season performance. Our 
objectives, for each of IRSís main filing season services, were to 
assess IRSís performance relative to 2007 goals and prior yearsí 
performance, including identifying actions that might generate 
efficiencies and increase compliance, and the extent to which IRS 
monitors paid preparer performance. Additionally, you asked us to 
describe the challenges associated with TAB. We testified on IRSís 
interim 2007 performance before the Subcommittee on Financial Services 
and General Government, Committee on Appropriations, U.S. Senate, on 
May 9, 2007, and before the Committee on Finance, U.S. Senate, on April 
12, 2007; we also issued a report to the Subcommittee on Oversight, 
Committee on Ways and Means, House of Representatives on April 3, 
2007.[Footnote 5] 

We based our assessment on the results and analyses of IRS data and 
reports, workload and performance information, observations of IRSís 
operations, interviews with IRS officials, information from 
representatives of some of the larger tax preparation companies and 
professional associations that represent preparers providing tax 
assistance, and reports by the Treasury Inspector General for Tax 
Administration (TIGTA). We have used IRS data in past filing season 
reports and continue to find them sufficiently reliable for the 
purposes of this report. Further details on our scope and methodology 
are provided later in this report. We performed our work from January 
through October 2007 in accordance with generally accepted government 
auditing standards. 

Results in Brief: 

IRS improved most of its filing season servicesópaper and electronic 
processing and telephone, Web site, and face-to-face assistanceóbut 
opportunities to realize additional savings and increase tax compliance 
exist, but have not been evaluated. 

Processing: IRS processed approximately 135 million individual income 
tax returns during 2007, by most measures exceeding last yearís 
performance and meeting this yearís goals. However, further 
improvements in processing are possible. First, IRS delivered the 
latest release of its new tax return processing system, the Customer 
Account Data Engine (CADE), late, which prevented millions of taxpayers 
from getting faster refunds in 2007. The new release was fully 
operational in May, which should benefit taxpayers in the 2008 filing 
season. Second, despite almost 60 percent of returns being filed 
electronically, taxpayers filed over 50 million returns on paper. 
Because of the cost, IRS does not transcribe all lines from tax 
returns, precluding the use of these data in IRSís computerized 
compliance checks. Further, to avoid disadvantaging taxpayers who file 
electronically, IRS does not post those lines from electronic returns 
to its enforcement data bases. At least two options exist for 
transcribing and posting all lines from tax returns. One, which we 
previously suggested, is to mandate electronic filing for paid 
preparers. Another option would be to require paper returns prepared 
using commercial software to be bar coded, allowing for scanning and 
some of the efficiencies of electronic filing. Either option might 
reduce the number of residual paper returns enough to allow for full 
transcription. Although IRS has done some research on bar coding and 
full transcription, it does not know the actions needed to require 
commercial software vendors to include bar codes on printed returns, 
the benefits in terms of processing cost savings and increased 
enforcement revenue of having all tax return data available 
electronically, or how much electronic filing would have to increase 
(either through mandates or bar coding) for the benefits of fully 
transcribing all residual paper returns to exceed the costs. 

Paid preparers: In 2006, paid preparers accounted for 61 percent of 
returns filed. IRS gathers basic information on preparers, such as 
names and Social Security numbers (SSN), but does not compile other 
information about each preparer, such as the number of returns filed and
type of preparer.[Footnote 6] IRS has research and other initiatives 
under way to gather more information on paid preparers, which should 
give the Congress and IRS a better basis for making decisions about the 
future oversight of paid preparers. 

Telephone assistance: Access to IRS telephone assistors was comparable 
to last year and the accuracy of responses to telephone tax law and 
account questions is at least 90 percent. IRS has several improvement 
initiatives under way and is reviewing options for reducing excess 
capacity at call sites, addressing our previous recommendation to 
timely develop and implement a plan to consolidate excess capacity at 
call sites. 

Web site: IRSís Web site performed well according to independent rating 
services. IRSís Web site is the least expensive way to provide service 
and is consistent with IRSís strategy of increasing self-assistance. 

Face-to-face assistance: Fewer taxpayers used IRSís walk-in sites, 
while more used return preparation assistance at volunteer sites. IRS 
continues to miss opportunities to assess the quality of face-to-face 
assistance. Despite IRSís increased reliance on community-based 
volunteers to target underserved populations, such as the elderly, IRS 
does not know the effectiveness of such targeting. At its walk-in 
sites, IRS uses direct observation to assess the accuracy of return 
preparation. However, because this method is likely to yield biased 
results, IRS lacks valid performance information. In contrast, contact 
recordingóa system capturing the audio portion of the assistor/taxpayer 
interactions and assistorsí computer screen activityóshows promise in 
monitoring the accuracy of tax law and account assistance. 

Taxpayer Assistance Blueprint (TAB): IRS has completed TAB, a data-
driven strategy for improving services to taxpayers, lowering costs, and
understanding the link between taxpayer service and compliance. TAB has 
potential value. In the past we have stressed the importance of 
reconsidering the level and types of services IRS provides to 
taxpayers, such as whether to maintain the same number of walk-in 
sites. Realizing the value of TAB will depend on several factors, such 
as future funding and progress quantifying the link between service and 
compliance. 

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

* determine actions needed to require software vendors to include bar 
codes on printed individual income tax returns and the cost of those 
actions; 

* determine the benefits, in terms of processing costs and improved 
enforcement, of having all return information available electronically; 

* determine how much electronic filing would have to increase, either 
through electronic filing mandates or bar coding, for the benefits of 
transcribing all remaining paper returns to exceed the costs; 

* develop estimates of the effectiveness of IRSís volunteer program at 
targeting underserved populations; and: 

* determine the feasibility of using contact recording as a method of 
monitoring and improving the quality of return preparation assistance 
at IRSís walk-in sites. 

In commenting on a draft of this report (see app. II), the IRS 
Commissioner agreed with our recommendations and outlined IRSís actions 
to address those recommendations. 

Scope and Methodology: 

To assess IRSís 2007 filing season performance in the key filing season 
activities compared to goals and past performance, and identify and 
assess potential improvements or efficiencies in filing season 
operations, we: 

* reviewed and analyzed IRS reports, testimonies, budget submissions, 
and other documents and data, including workload data, data related to 
IRSís performance measures and goals, and data on taxpayer usage and 
other statistics such as the number of paid preparers; 

* observed operations at IRSís Atlanta, Ga. and Andover, Ma. paper 
processing centers, the Atlanta and Andover call sites, the Joint 
Operations Center (which manages IRSís telephone services) in Atlanta, 
and several of IRSís walk-in locations in Atlanta and Baltimore, Md. 
and selected these offices for a variety of reasons, including the 
location of key IRS managers, such as those responsible for telephone 
and walk-in site services; 

* tested for statistically significant differences between annual 
performance measures based on IRS sample data; 

* analyzed staffing data for paper and electronic filing, telephone 
assistance, and walk-in assistance; 

* reviewed information from organizations, such as Keynote Systems, 
that evaluate Internet performance; 

* interviewed IRS officials about current operations, performance 
relative to 2007 performance goals and prior filing season performance, 
trends, significant factors and initiatives that affected or were 
intended to improve performance, monitoring and oversight of paid tax 
preparers, and research conducted or planned to examine links between 
taxpayer service and compliance; 

* interviewed representatives of some of the larger private and 
nonprofit organizations that prepare tax returns, including H&R Block 
and trade organizations that represent both individual paid preparers, 
tax preparation companies, and professional associations, including the 
American Institute of Certified Public Accountants; 

* reviewed TIGTA reports and interviewed TIGTA officials about IRSís 
performance and initiatives; and: 

* reviewed prior GAO reports and followed up on our recommendations 
made in prior filing season and related reports. 

This report discusses numerous filing season performance measures and 
data that cover the quality, accessibility, and timeliness of IRSís 
services that, based on our prior work, we consider sufficiently 
objective and reliable for purposes of this report. To the extent 
possible, we corroborated information from interviews with 
documentation and data and where not possible, we report the 
information as attributable to IRS officials. We reviewed IRS 
documentation, interviewed IRS officials about computer systems and 
data limitations, and compared those results to GAO standards of data 
reliability.[Footnote 7] Data limitations are discussed where 
appropriate. Finally, we conducted our work primarily at IRS 
headquarters in Washington, D.C., the Wage and Investment Division 
headquarters in Atlanta, Ga., as well as the other sites mentioned 
earlier. 

Background: 

IRSís filing season is an enormous and critical undertaking that 
consists of two primary activities, processing individual income tax 
returns and providing taxpayer assistance. It consumes more than $3 
billion annually. 

Figure 1: IRSís 2007 Filing Season Activities: 

[See PDF for image] 

This figure includes a graphic illustration of tax forms, as well as 
depicting the following data: 

Returns processing: 
* 56 million paper returns at 5 paper processing centers; 
* 79 million electronic returns at 5 electronic processing centers. 

Taxpayer assistance: 
* 69 million toll-free calls at 24 call sites and automation; 
* 5 million walk-in contacts at 401 walk-in sites and 12,000 volunteer 
sites; 
* 121 million Internet downloads at IRS website. 

Source: GAO analysis of IRS data, IRS (images). 

Notes: The number of paper and electronic returns are estimated for the 
time period January 1, 2007, to September 14, 2007; toll-free calls for 
the time period January 1, 2007 to June 30, 2007; walk-in contacts, 
which includes returns prepared at IRS walk-in sites from December 31, 
2006, through April 21, 2007, and volunteer sites from January 1 
through April 30, 2007; and Internet downloads for the time period 
January through July 2007. We used different dates for the various 
areas that best reflect IRSís filing season workload in that area. 

[End of figure] 

Taxpayers can file their returns by mailing paper returns to one of 
IRSís five paper processing centers or submitting the returns 
electronically.[Footnote 8] 

Electronic filing has provided IRS with significant cost savings. 
Electronic filing is faster, which allows taxpayers to receive refunds 
faster, and is less error proneóIRS does not have to transcribe 
electronic tax return information and built-in checks eliminate many 
errors that IRS has to address when processing paper tax returns, such 
as correcting computational mistakes and SSNs. In addition, taxpayers 
below an income ceiling can access the Free File program offered 
through IRSís Web site by a consortium of 19 tax preparation companies 
that offer free online tax preparation and filing services. 

IRSís Business Systems Modernization (BSM) program, which began eight 
years ago, is a high-risk, highly complex effort critical to supporting 
IRSís taxpayer service goals, such as increasing electronic filing and 
providing more accurate and timely responses to taxpayer inquiries. To 
date, IRS has deployed releases of modernized systems that have 
delivered substantial benefits to taxpayers and the agency, including e-
Services (a new Web portal and services for paid preparers) and parts 
of CADE, which is the new taxpayer information database that 
facilitates faster refund processing and provides IRS with more up-to-
date account information. CADE is intended to eventually replace IRSís 
antiquated Individual Master File legacy processing system, which is 
the agencyís repository of taxpayer account information, and therefore, 
is the BSM programís cornerstone and highest priority project. 
[Footnote 9] 

IRS also provides a variety of taxpayer services, including tax law 
assistance, account resolution, limited return preparation, tax forms 
and publications distribution, outreach, and education, mainly through 
its telephone operations, Web site, and face-to-face assistance sites. 
Taxpayers can call IRSís toll-free assistance telephone lines with 
questions about tax law or their refunds. Depending on how taxpayers 
respond to menu choices, questions are answered by an automated message 
or calls are routed to telephone assistors located at 24 call sites 
around the country. Calls are routed to the next available assistor 
assigned to answer specific questions, e.g., tax law, account, or 
refund questions. 

IRS provides many services through its Web site, including ďWhereís My 
Refund,Ē which enables taxpayers to use the Web site to determine 
whether the agency received their tax returns and processed their 
refunds. Taxpayers can also download forms, instructions, and 
publications, research their own tax law issues through Frequently 
Asked Questions or Tax Topics, and receive help with specific tax law 
questions and procedural questions via e-mail. 

IRSís volunteer program has become an increasingly important part of 
IRSís face-to-face assistance. It expanded to 66 national partners that 
include financial, social service, corporate, educational, and 
government organizations and helps IRS serve traditionally underserved 
taxpayer segments, including the elderly, low-income, disabled 
taxpayers, taxpayers with limited-English proficiency, Native 
Americans, and rural taxpayers. Additionally, IRS has 401 walk-in sites 
where taxpayers can ask basic tax law questions, get account 
information, receive assistance with their accounts, and have returns 
prepared if their annual income is $39,000 or less.[Footnote 10] 

In April 2007, IRS released its final TAB report, a 5-year plan 
designed to assist the agency in providing, evaluating, and improving 
taxpayer services at lower costs. As part of its research into the 
costs of taxpayer services, IRS developed estimates of the cost-per-
service contact for providing different types of taxpayer services and 
conducted preliminary research about the effect of taxpayer service on 
compliance.[Footnote 11] We have reported that IRS lacks quantitative 
estimates of the impact of taxpayer service and its enforcement efforts 
on voluntary compliance; TAB should help IRS obtain better estimates. 
We also have long supported IRSís research efforts to measure the tax 
gap, which is the difference between what is paid voluntarily and on 
time and what is owed, and help reduce it.[Footnote 12] 

As we previously reported, despite the 2007 filing season being 
characterized as high-risk, in large part because of the Telephone 
Excise Tax Refund (TETR), the impact of TETR on taxpayer services has 
been much less than IRS anticipated.[Footnote 13] 

In addition to IRS, paid preparers are a critical part of the nationís 
tax administration system because of the wide variety of services they 
offer and their unique relationship with taxpayers. Paid preparers may 
combine several taxpayer services, including help understanding tax 
obligations, answering tax law questions, and providing tax forms and 
publications, return preparation, and electronic filing. In previous 
reports and testimony, we have reported that paid preparers have made 
serious errors, although undoubtedly many do their best to provide 
their clients with quality service.[Footnote 14]

Electronic Filing Continued to Generate Savings, Processing Performance 
Improved, but CADE Was Delayed, and Opportunities Exist to Reduce Costs 
and Improve Compliance: 

Electronic filing and CADE both have the potential to further reduce 
IRSís costs and improve processing speed and accuracy. Many tax returns 
are still filed on paper, which increases processing cost, and limits 
information readily available for IRSís enforcement programs.[Footnote 
15] 

Electronic Filing Continues to Increase and Generate Savings: 

As of September 14, 2007, IRS processed approximately 135 million 
individual income tax returns. Of those returns, approximately 79 
million (58 percent) were filed electronically and the remaining 56 
million were filed on paper. The number of returns filed electronically 
increased 9 percent over last year, which is greater than last yearís 
increase and more than IRS projected.[Footnote ]16 One factor 
contributing to the growth in electronic filing is mandates. Thirteen 
states now mandate electronic filing of state returns, which has 
increased electronic filing of both state and federal tax returns. 
Despite the overall increase in electronic filing, taxpayerís use of 
the Free File program continued to decline. For example, as of June 21, 
IRS processed about 3.8 million returns filed through the Free File 
program, 2 percent less than last year. IRS officials attributed this 
decline in part to companies offering free electronic online filing 
separate from the Free File program. 

Growth in electronic filing generates savings for IRS by reducing staff 
needed for labor-intensive paper processing.[Footnote 17] In 2006, IRS 
used almost 1,700 (36 percent) fewer staff for processing tax returns 
than in 1999. IRS estimates that this saved the agency $78 million in 
salary, benefits, and overtime in 2006.[Footnote 18] In addition, IRS 
has achieved ancillary space cost savings from the closing of 
submission processing centers. IRS estimated the cost savings as a 
result of closing the Brookhaven and Memphis paper processing centers 
to be $24.9 million in 2007.[Footnote 19] IRS closed its Philadelphia 
paper processing center at the end of September 2007, and plans to 
close its Andover paper processing center in 2009. 

IRSís Processing Performance Continues to Improve: 

IRS issued over 103 million refunds, totaling almost $233 billion, as 
of September 14, 2007. Nearly 60 percent of the refunds were directly 
deposited, almost 7 percent over the same period last year, which is 
important because direct deposit is faster, more convenient for 
taxpayers, and less expensive for IRS than mailing paper checks. 

IRS met or exceeded seven out of nine processing performance goals and 
continues a trend of improvement (see app. I for details). For example, 
IRS met or exceeded its goals for the percentage of errors included in 
letters, notices, and refunds. For eight of its nine measures, IRS met 
or exceeded its previous yearís performance. IRS did not meet one goal 
and paid more interest on delayed refunds than expected, because of 
problems with a computer system that processes some taxpayer 
identification numbers and the inexperience of new staff. IRS did not 
meet its efficiency goal because it received fewer easy to process TETR 
returns than expected. 

Groups and organizations we spoke with, including the National 
Association of Enrolled Agents, the American Institute of Certified 
Public Accountants, and large tax preparation and tax preparation 
software companies, corroborated what the processing performance data 
showed, saying the filing season went smoothly. Similarly, TIGTA 
recently reported that IRS had a successful filing season.[Footnote 20]

Delays in CADE Prevented Millions of Taxpayers from Benefiting from 
Faster Refunds: 

IRS was late in delivering its latest version of CADE, its new tax 
return database, intended to replace the legacy Individual Master File. 
CADE Release 2.2 started processing returns in early March but did not 
become fully operational until late May, nearly 5 months behind 
schedule, because of problems identified during testing. IRS originally 
intended CADE to post 33 million taxpayer returns during 2007, more 
than four times the 7.4 million posted by CADE last year. With the 
delay IRS revised its estimate down to approximately 17 to 19 million 
returns, then down to 11 to 12 million returns. As of early August, 
CADE had posted 11 million returns, or about one-third of what IRS 
expected, and disbursed 11 million refunds totaling over $11 billion. 

A major benefit of CADE is that it is faster than the legacy Individual 
Master File system. According to IRS officials, direct deposit refunds 
are issued by CADE 1 to 5 business days faster than the current legacy 
system, and paper check refunds are issued 4 to 8 business days faster. 
Because of the delays in CADE millions of taxpayer refunds were not 
issued as fast as IRS planned. 

Even with the latest version of CADE deployed, last year IRS estimated 
that over $500 million more would be required to fully implement the 
individual income tax processing part of the system.[Footnote 21] 
Moreover, the CADE setback adversely affected IRSís ability to deliver 
the functionality planned for the next release, such as processing the 
earned income tax credit schedule, and has caused IRS to reconsider the 
functionality of other future CADE releases. IRS expects future 
releases of CADE to be linked with its Account Management System (AMS) 
releases, requiring increased attention to issues of synchronization 
and coordination between the two systems. According to IRS officials, 
CADE and AMS working together will enable IRS to process tax returns 
and address many taxpayer issues, such as address changes, in a near 
real-time manner, which benefits both IRS and taxpayers. 

Further complicating matters, IRS has additional plans for CADE, such 
as processing tax returns claiming the earned income tax credit, during 
the 2008 filing season. IRS is aware of these challenges and is working 
on a revised Release Content Master Plan, which it plans to complete by 
November 2007, detailing the capabilities and interdependencies 
associated with future releases of CADE and AMS and the plans for 
replacing the legacy Individual Master File system.[Footnote 22]

IRS May Be Missing Opportunities to Further Reduce Costs and Increase 
Compliance by Not Doing More to Reduce the Number of Paper Returns 
Processed and Transcribe all Tax Return Data: 

Paper returns cost IRS tens of millions of dollars to process. In all 
of 2006, IRS received about 63 million paper returns.[Footnote 23] Of 
these, 60 percent (38 million) were prepared on a computer, but then 
printed and mailed to IRS. IRS codes such returns with a ďV.Ē IRS 
estimates that a paper return costs $2.36 more to process than an 
electronically filed return. This difference implies that, if IRS could 
convert all v-coders to electronic filing, IRS could save roughly $90 
million annually. Furthermore, if all v-coders could be converted to 
electronic filing, the number of paper returns remaining and needing to 
be transcribed would be greatly reduced. 

Paper returns also limit the effectiveness of IRSís enforcement 
programs. To control costs, IRS does not transcribe all the lines on 
paper tax returns into its computer databases, such as taxpayersí 
telephone numbers, limiting the amount of information available 
electronically for enforcement purposes. As we previously reported, 
even small changes in the amount of information IRS transcribes can 
consume substantial resources that might offset some potential savings 
from electronic filing.[Footnote 24] Further, to avoid disadvantaging 
taxpayers who file electronically, IRS has a policy of posting the same 
information from electronic and paper returns to its databases. 
Consequently, if a line is not transcribed from paper returns, it is 
not posted from electronic returns either. Only information posted to 
computer databases is readily available for use in IRSís automated 
compliance checking programs. These programs include matching tax 
return entries with information returns from third parties, such as 
Form W-2s from employers or Form 1099s from financial institutions, and 
selecting suspicious returns for audit. 

According to IRS officials, transcribing and posting more comprehensive 
information from individual income tax returns could facilitate the 
audit process, expedite contacts for faster resolution, reduce handling 
costs, allow for improved case selection, and potentially better define 
specific tax gap issues. Although we have not independently verified 
IRSís methodology, for one of its main enforcement programsóthe 
Automated Underreporter ProgramóIRS officials estimate that having all 
tax return information available electronically would result in a $175 
million increase in tax revenue annually, while at the same time, 
reduce its ďno-changeĒ rate, making it less likely that IRS would 
select taxpayers for further contact where no additional tax was 
assessed, thus lowering taxpayer burden.[Footnote 25] 

State experiences provided some corroboration of IRSís views, as 
officials we spoke with reported that greater electronic access to 
return information allowed them to verify information on state returns 
and improved their enforcement programs while reducing the number of 
compliant taxpayers who were contacted. 

At least two options exist to increase electronic filingómandates and 
bar coding. Last year we suggested that the Congress mandate electronic 
filing by paid preparers meeting certain criteria, such as a threshold 
number of returns.[Footnote 26] Mandating electronic filing would not 
convert all v-coders, because it would not apply to taxpayers who 
prepare their own returns on a computer. Still, most v-coders use a 
paid preparer (68 percent in 2006). 

Another option to increase electronic filing is bar coding of all 
printed returns prepared using commercial software. IRS has done 
preliminary research showing that 100 percent of the information on a 
return could be condensed into a bar code, which could be read by hand-
held scanners. Consequently, bar coding produces some of the 
efficiencies of electronic filing by replacing the labor-intensive 
transcription process and eliminating transcription errors. However, 
bar coding would still require some processing of paper such as 
receiving and opening paper mail. 

Several states already require bar coding on their printed returns and 
there is an industry standard among developers and vendors for creating 
bar codes.[Footnote 27] Moreover, many of the largest software 
providers already support bar coding. 

IRSís Wage and Investment Operating Division staff developed a proposal 
for bar coding. However, Wage and Investment officials rejected that 
proposal in June 2007, in part because it involved upgrading a legacy 
system. Wage and Investment officials told us they are developing a new 
funding proposal for no earlier than 2010, which would likely include 
bar coding. 

IRS, however, lacks several pieces of information that would help it 
make a decision about whether the benefits exceed the costs of bar 
coding and transcribing the residual paper returns. 

* Actions required: IRS does not know the actions needed to require 
commercial software vendors to include bar codes on printed returns. 
IRS officials told us that they have not conferred with the software 
vendors about the actions that would be necessary. Nor have they 
conferred with the states about the actions they took to mandate bar 
coding. Without knowing the necessary actions, IRS cannot estimate the 
costs and time frames for a bar code program. 

* Benefits: While IRS has estimates of the savings from converting from 
paper to electronic filing, it does not have estimates of the savings 
from converting paper to bar coded returns. Although it estimated the 
increased revenue possible from the Automated Underreporter Program, 
IRS does not have a complete estimate of the revenue gains across all 
its enforcement programs that would be realized if all tax return 
information were transcribed and posted. 

* Breakeven: IRS does not know how much electronic filing would have to 
increase, through either mandated electronic filing or bar coding, 
before transcribing the residual paper returns would pass a 
benefit/cost test.[Footnote 28] 

Gathering the above information would have some costs to IRS. However, 
the cost savings to IRS and increased revenue could be in the range of 
hundreds of millions of dollars. Additionally, taxpayer compliance 
burden might be reduced. 

IRS Has Limited Information on Paid Preparers Who Prepare the Majority 
of Tax Returns, but Has Efforts Underway to Acquire More: 

Although they are an important part of the tax administration system, 
IRS has limited information about paid preparers. However, IRS has 
research and other initiatives under way intended to gain more 
information. 

As shown in figure 2, paid preparers have filed an increasing majority 
of all individual income tax returns over the last decade.[Footnote 29] 
During the 2006 filing season, paid preparers prepared nearly 78 
million (61 percent) of the 127 million individual income tax returns. 
Of the preparer returns, nearly 70 percent were filed electronically. 

Figure 2: Percentage of Returns with a Paid Preparer Signature: 

[See PDF for image] 

This figure is a line graph representing the percentage of returns with 
a paid preparer signature. The vertical axis of the graph represents 
percentage of returns from 0 to 70. The horizontal axis of the graph 
represents years from 1997 to 2006. The line depicts a gradual increase 
in the percentage of returns with a paid preparer signature from about 
52% in 1997 to 61% in 2006. 

Source: GAO analysis of IRS data. 

[End of figure] 

As we said in past reports, paid preparers are a critical quality-
control checkpoint for the tax system and, as with IRS, the quality of 
the service they provide is important.[Footnote 30] However, we also 
said that some paid preparers made serious errors. For example, in 2006 
we found that all 19 of the commercial preparers we visited made errors 
completing returns. Some of the most serious errors involved not 
reporting business income and failing to itemize deductions. Our 
limited work did not permit generalizing about the quality of all paid 
preparersí work and undoubtedly, many preparers do their best to 
prepare returns that are compliant with tax laws. We also previously 
reported that, while many taxpayers believe they benefited from using a 
paid preparer, millions may have been poorly served.[Footnote 31] 

IRS has some information on paid preparers. For example, IRS requires 
paid preparers to identify themselves on all income tax returns they 
prepare by entering their SSN or PTIN. However, IRS does little to 
monitor or track basic information about individual paid preparers. For 
example, IRS does not collect information on the type of preparers, 
such as whether the preparer is an enrolled agent or part of a 
commercial chain, or the number or types of returns filed by preparers. 
Having such information could allow IRS to better identify filing 
errors and target its outreach to specific preparers or preparer 
groups. 

IRS currently has efforts under way intended to begin remedying the 
lack of information about paid preparers. One important step was the 
Return Preparer Summit, which took place in late September 2007. 
According to IRS, the Summit included officials from multiple IRS 
offices who discussed how to better coordinate and communicate between 
offices with paid preparer responsibilities. IRS officials expect to 
present their findings and potential recommendations in March 2008 when 
IRS updates its recent report on voluntary compliance and the tax 
gap.[Footnote 32] 

IRS is exploring the development of a database that will serve as a 
centralized repository of paid preparer information. IRS envisions that 
such a database could facilitate more robust compliance efforts and 
allow taxpayers to access information on the status of an enrolled 
agent. Additionally, the database could potentially be expanded to 
operate as a tracking system if proposed legislation mandating 
universal enrollment and testing of preparers was to be passed by the 
Congress.[Footnote 33] In the meantime, IRS is working to better 
utilize and potentially improve existing databases to share preparer 
information across its offices with paid preparer responsibilities. 

Further, according to IRS, the agency plans to conduct research on paid 
preparers as outlined in TAB. Paid preparer initiatives include: 

* studying how taxpayers select a paid preparer; 

* evaluating the scope and nature of paid preparer errors; 

* assessing IRSís outreach strategy to paid preparers; and: 

* observing paid preparersí business models in order to better 
understand how to provide effective services to reduce errors. 

Paid preparers can support IRSís efforts to improve compliance. They 
assist the majority of individual income tax filers, and the assistance 
they offer can potentially help taxpayers avoid errors. Having 
additional information on paid preparers could help the agency identify 
and possibly prevent systematic errors and noncompliance, and could be 
important in guiding IRSís outreach and education strategies for paid 
preparers. TAB and other initiatives IRS has planned are steps that 
should give the Congress and IRS a better basis for making decisions 
about the future oversight of paid preparers. 

Access to IRSís Telephone Assistors Was Comparable to Last Year, 
Accuracy Remains High, and IRS Is Reviewing Options for Reducing Excess 
Capacity at Call Sites: 

Access and accuracy are key measures of IRSís telephone performance, 
because IRS received 69 million calls during the filing season. At the 
same time, reducing excess capacity at call sites is important, because 
of the magnitude of IRSís telephone operations. 

Telephone Access Was Comparable to Last Year and Accuracy of Responses 
to Telephone Inquiries Remains High: 

Taxpayersí access to IRSís telephone assistors was comparable to last 
year and IRSís goal, as measured by level of service which is the 
percentage of taxpayers who wanted to talk with an assistor and 
actually got through and received service (see table 1).[Footnote 34] 

For two other access measures, average speed of answeróthe length of 
time taxpayers wait to get their calls answeredóand taxpayer 
disconnectsóthe rate at which taxpayers waiting to speak with an 
assistor abandoned their calls to IRSóIRSís performance weakened 
somewhat compared to the same period last year, but the comparison is 
complicated by a policy change. As of June 30, average speed of answer 
increased about 15.4 percent and was slightly higher than IRSís goal, 
and the taxpayer disconnect rate also increased. According to IRS 
officials, both increases were due, in part, to a decision by IRS to 
disconnect fewer taxpayers waiting to speak with an assistor. This 
decision resulted in taxpayers having the option of either waiting to 
speak with an assistor or hanging up rather than IRS deciding for them 
by disconnecting the call, resulting in increased taxpayer choice along 
with increases to both the average speed of answer and the taxpayer 
disconnect rate. Finally, IRS officials noted that Customer Voice 
Portal, a new call queuing process that routes calls directly to 
available assistors instead of to call sites, resulted in a more 
equalized wait-time for taxpayers calling to speak with an assistor 
this year. 

Table 1: IRS Telephone Assistors Performance, 2001ó2007 Filing Seasons: 

Telephone assistance[A]: Assistor level of service[B]; 
2001 Actual: 68.6%; 
2002 Actual: 68.6%; 
2003 Actual: 83.7%; 
2004 Actual: 85.9%; 
2005 Actual: 82.2%; 
2006 Actual: 81.0%; 
2007 Actual: 81.6%; 
Fiscal year 2007 goals: 82%. 

Telephone assistance[A]: Average speed of answer[C]; 
2001 Actual: 5.5; 
2002 Actual: 4.6; 
2003 Actual: 2.8; 
2004 Actual: 2.8; 
2005 Actual: 4.3; 
2006 Actual: 3.9; 
2007 Actual: 4.5[E]; 
Fiscal year 2007 goals: 4.3. 

Telephone assistance[A]: Taxpayer disconnect rate[D]; 
2001 Actual: 16.0%; 
2002 Actual: 14.6%; 
2003 Actual: 8.3%; 
2004 Actual: 8.6%; 
2005 Actual: 12.0%; 
2006 Actual: 10.6%; 
2007 Actual: 13.3%[E]; 
Fiscal year 2007 goals: NA. 

Source: GAO analysis of IRS data. 

[A] Telephone assistance data are based on the date range of January 1 
through June 30. 

[B] The percentage of callers who want to speak to an assistor who get 
through and receive service. 

[C] Average number of minutes a taxpayer waits in queue for an 
assistor. 

[D] The percentage of callers who hang up while waiting in queue for an 
assistor. 

[E] IRS officials attributed increases in average speed of answer and 
taxpayer disconnects to a conscious decision by IRS to decrease the 
times IRS disconnects taxpayers, which resulted in taxpayers being able 
to decide whether or not to hang up (taxpayer disconnect) or wait to 
speak with an assistor (average speed of answer). 

[End of table] 

IRS estimates that the accuracy of telephone assistorsí responses to 
tax law and account questions was comparable to the same time period 
last year and met IRSís goals. Moreover, as noted in table 2, since 
2005, accuracy has been about 90 percent or more and was significantly 
better than in 2001. 

Table 2: IRS Telephone Assistor Accuracy Performance, 2001ó2007 Filing 
Seasons: 

Accuracy measures[A]: Tax law accuracy rate (in percent)[B]; 
2001 Actual: 79.1 +/-0.6; 
2002 Actual: 84.9 +/-0.5; 
2003 Actual: 81.3 +/-0.7; 
2004 Actual: 79.5 +/-0.8; 
2005 Actual: 89.5 +/-0.6; 
2006 Actual: 90.6 +/-0.6; 
2007 Actual: 90.7 +/-0.9; 
Fiscal year 2007 goals: 91.0. 

Accuracy measures[A]: Accounts accuracy rate (in percent)[B]; 
2001 Actual: 88.1 +/-0.6; 
2002 Actual: 90.5 +/-0.4; 
2003 Actual: 88.6 +/-0.4; 
2004 Actual: 89.0 +/-0.5; 
2005 Actual: 91.3 +/-0.4; 
2006 Actual: 93.3 +/-0.3; 
2007 Actual: 93.2 +/-0.5; 
Fiscal year 2007 goals: 93.3. 

Source: GAO analysis of IRS data. 

[A] Based on representative samples from January through June for 2001, 
2002, 2003, 2004, 2005, 2006, and 2007. 

[B] The percentage of calls in which telephone assistors provided 
accurate answers for the call type and took the appropriate action, 
with a 90 percent confidence interval.

[End of table] 

Call Demand Was Comparable to Last Year, and Telephone Service Remains 
Important for Taxpayers: 

IRS received about 69 million calls on its toll-free telephone lines 
through June 2007, which was less than expected, but slightly more than 
in 2006. IRSís automated service handled roughly the same number of 
calls as those handled by IRS assistors, although the number of 
automated calls received was fewer than last year (22.1 million), while 
the calls answered by assistors was equivalent to last year.[Footnote 
35] 

Figure 3: How IRS Handled Calls During 2007: 

This figure is a pie-chart representing how the IRS Handled Calls 
During 2007. The following data is depicted: 

Total calls to IRS: 69 million; 
* Received automated service: 20.8 million (30.1%); 
* Reached an assistor: 21.2 million (30.7%); 
* Hung up or disconnected without receiving service: 14.2 million 
(20.6%); 
* Called after business hours: 6.7 million (9.7%); 
* Transferred out to another IRS operation, such as Examination: 6.1 
million (8.8%). 

Source: GAO analysis of IRS data and Art Explosion (images). 

Note: Data cover the period January 1 through June 30, 2007. Due to 
rounding, percentages do not add to 100. 

Of the calls received by IRS, 28.8 million calls were from taxpayers 
trying to obtain information on the status of their tax refunds, 
despite increasing use of IRSís Web site feature, ďWhereís My Refund.Ē 
Another 22.2 million were about account questions, and 13.1 million 
were about tax law questions. 

Because millions of taxpayers continue to depend on IRSís telephone 
assistance, it remains a critical part of taxpayer services. Moreover, 
IRS and our prior reports have noted that assisting taxpayers with tax 
questions reduces burdensome notices and inadvertent 
noncompliance.[Footnote 36] As part of TAB and other initiatives, IRS 
plans to conduct more research into the effect of telephone assistance 
on compliance. 

IRSís telephone operations consume most of IRSís taxpayer service 
budget. According to TAB, assisted telephone services accounted for 65 
percent and automated assistance accounted for 2 percent of the nearly 
$1 billion total costs to the agency for providing taxpayer assistance 
in 2005. At this time IRS does not have plans to automate more calls, 
since the agency has already shifted easier calls to automation and Web 
assistance, leaving the more difficult and time-consuming account and 
tax law calls that require the attention of assistors. 

In addition to answering telephones, IRS devotes resources to 
processing paper inventory, such as amended returns and taxpayer 
correspondence at some of its call sites. Because of limited TETR call 
demand, IRS closed more paper inventory than anticipated, meaning that 
taxpayer issues were resolved more quickly. As seen in figure 4, call 
volume has not declined and accordingly, resources devoted to answering 
telephones have not declined noticeably since 2001, while resources 
devoted to paper inventory have declined almost 14 percent since 2001.

Figure 4: Telephone Assistance and Paper Inventory Resources by 
Workload: 

[See PDF for image] 

This figure is a combination vertical bar and line graph depicting 
telephone assistance and paper inventory resources by workload. The 
left vertical axis of the graph represents staff years from 0 to 9,000. 
The right vertical axis represents volume from 0 to 40 million. The 
horizontal axis of the graph represents fiscal years 2001 to 2007 
(projected). For each fiscal year there are vertical bars depicting 
staff years used for assistor calls and staff years used for paper 
inventory. Also, for the entire time period represented in the graph 
there are lines depicting volume of assistor calls answered and volume 
of paper inventory. 

Source: GAO analysis of IRS data. 

Note: The figure includes data for entire fiscal years, not just for 
the filing season. 

[End of figure] 

IRS has initiatives to expand and improve on telephone services 
including: 

* Centralized Customer Contact Forecasting and Scheduling to modernize 
the current system of workload forecasting, scheduling, and tracking 
workload and training requirements; 

* Working groups looking into ways of improving taxpayer services, such 
as developing ways to synchronize telephone message information with 
Web site services to better link different service channels. IRSís goal 
is to revise the messages heard by taxpayers calling to speak with 
assistors after business hours, during holidays, or when telephone 
lines are too crowded to accommodate them, so that taxpayers are told 
not only when to try back, but are also referred to IRSís Web site for 
further information. 

IRS Is Reviewing Options for Reducing Excess Capacity at Call Sites: 

Last year we recommended that IRS timely develop, validate, and 
implement a plan to consolidate call sites because, at the filing 
seasonís peak, IRS determined that it had 850 unused assistor 
workstations spread across its 24 call sites.[Footnote 37] According to 
IRS officials, the number of unused workstations has not changed. IRS 
completed a workload analysis designed to evaluate changes in telephone 
staff resources and call demand and used business criteria, such as the 
types of calls answered at call sites and facility and employee costs, 
to identify call sites for potential closure. IRS plans to conduct an 
extensive cost-analysis review of different options for reducing excess 
capacity prior to potentially closing any sites, has secured funding 
for the analysis, and is in the process of obtaining a contractor to 
assist with the review. 

Web Site Performance Remains High, and Initiatives Are On-going to 
Enhance Web Capabilities and User Experience: 

Web site performance remains high, which is important because IRSís Web 
site is used by millions of taxpayers, is available to taxpayers 24 
hours a day, is much less costly than other types of assistance, and 
will remain important as more taxpayers use self-assistance. 

Use of IRSís Web site increased this filing season compared to prior 
yearsí except for downloads of files such as tax forms and 
publications, as shown in table 3 below. According to IRS officials, 
one reason for the decrease in downloads is the conversion of IRS 
publications and instructions from file format to another format, which 
IRS does not count as part of downloads. Additionally, the increased 
use of electronic filing and tax preparation software reduces the need 
for taxpayers to download tax forms and publications. However, because 
of the decline, IRS will continue to monitor the number of downloads 
and identify possible issues associated with downloading files. 

Table 3: IRS Web Site Use, 2006 and 2007 (data are in thousands): 

Uses: Visits[A]; 
2006: 152,345; 
2007: 168,332; 
Percentage change: 110.5. 

Uses: Downloads[A]; 
2006: 160,477; 
2007: 120,596; 
Percentage change: -24.9. 

Uses: Searches[A]; 
2006: 88,250; 
2007: 101,072; 
Percentage change: 14.5. 

Uses: Whereís My Refund[B]; 
2006: 30,663; 
2007: 40,325; 
Percentage change: 31.5. 

Uses: Number of TETR-related visits[C]; 
2006: N/A; 
2007: 4,570; 
Percentage change: N/A. 

Source: GAO analysis of IRS data. 

Note: N/A means not applicable. 

[A] Web site visits, searches, and downloads are from January through 
July 2006 and 2007. A visit begins when a visitor views their first 
page on IRS.gov, and ends when the visitor leaves the site. A visit is 
not a count of the number of unique individuals accessing the site. 

[B] For January 1 through July 29, 2006, and July 28, 2007. 

[C] Visits to a Web page specific to TETR, which was not operational in 
2006. Data are for October 1, 2006, through June 9, 2007. 

[End of table] 

Evidence that IRSís Web site is performing well includes the following: 

* According to the American Customer Satisfaction Index, for the 2007 
filing season, January 1 through April 17, 2007, IRSís Web site scored 
above other government agencies, nonprofits, and private sector firms 
for customer satisfaction (74 for IRS versus 72 for all government 
agencies versus 71 for all business surveyed), and better than during 
the 2006 filing season.[Footnote 38] 

* An independent weekly study by Keynote Systems, Inc., a company that 
evaluates Web sites, reported that, as of July 16, 2007, IRSís Web site 
repeatedly ranked in the top 6 out of 40 government agency Web sites 
evaluated in terms of average download time. 

* According to IRS, the number of e-mails and telephone calls to the 
Web site HelpDesk decreased compared to the 2006 filing season, and 
average page views per visit decreased, indicating that visitors to the 
Web site are finding information more quickly. 

Additionally, this year IRS implemented a new sales tax deduction 
calculator, which IRS wants to use as a standard for developing other 
online calculators, and updates to the Free File, Alternative Minimum 
Tax, and Earned Income Tax Credit applications. [Footnote 39] Also, 
this year the ďWhereís My RefundĒ feature allowed taxpayers to check on 
the status of split refunds, and told the taxpayer if one or more 
deposits were returned from the bank because of an incorrect routing or 
account number. 

IRS has initiatives to expand and improve on its Web site services. 

* Tax Law Support Tools will provide Web site self-assistance tools to 
taxpayers and other external stakeholders, allowing users to conduct 
keyword and natural language queries. 

* The first release of Internet Customer Account Services will allow 
individual taxpayers to view their account information online and is 
expected to be deployed in July 2008. 

* A Spanish version of ďWhereís My RefundĒ will enable Spanish speaking 
taxpayers to check refund status and receive account information. 

Taxpayer Assistance Continued to Grow at Volunteer Sites and Decline at 
IRSís Walk-in Sites; While Quality has Improved in Some Areas, More 
Reliable Data Are Needed: 

Volunteer sites play an increasingly important role in the tax system 
and are part of IRSís and the Department of Treasuryís integrated 
multiyear strategy to reduce the tax gap by better identifying and 
recruiting partners that are more skilled than IRS at reaching 
underserved taxpayers, including the elderly, disabled, low-income, 
Native American, and rural taxpayers. Volunteer sites prepared 2.4 
million returns, an increase of 17 percent over last year and higher 
than projected through April 21, 2007. As illustrated in figure 5, this 
is consistent with previous trends for volunteer and walk-in sites. 
Return preparation at volunteer sites has more than doubled since 2001, 
while return preparation at IRSís walk-in sites has declined by almost 
74 percent during the same period. 

Figure 5: Trends in Face-to-face Return Preparation, 2001 through 2007: 

[See PDF for image] 

This figure is a multiple line graph depicting trends in face-to-face 
return preparation, 2001 through 2007. The vertical axis of the graph 
represents number in millions from 0.0 to 2.5. The horizontal axis of 
the graph represents years from 2001 to 2007. The line depicting 
volunteer sites shows a gradual increase, while the line depicting IRS 
walk-in sites shows a gradual decrease. 

Source: GAO analysis of IRS data. 

Note: For walk-in sites, data covers the time periods of December 31, 
2006 through April 21, 2007. According to IRS, most taxpayers having 
their returns prepared at volunteer sites do so during the filing 
season, which is from January 1 through April 30. 

[End of figure] 

Increased return preparation assistance at volunteer sites, rather than 
at walk-in sites, allows IRS to devote more resources to services such 
as account assistance, which can only be provided by IRS staff. 
Further, TAB data show return preparation assistance provided at walk-
in sites is four times more expensive per return than at volunteer 
sites. Moreover, TAB data show that walk-in sites are used by a small 
percentage of taxpayers, yet account for over 20 percent or, $201 
million, of the almost $1 billion spent on taxpayer assistance in 2005. 

Despite an increase in the number of returns prepared at volunteer 
sites, the quality of volunteer-prepared returns remains largely 
unknown and what is known is not statistically reliable. IRSís current 
initiative to monitor the quality of returns at volunteer sites is in 
its second year of implementation and includes mystery shopping, site 
and return reviews.[Footnote 40] However, according to IRS officials, 
volunteer sites had advanced notice of site and return reviews, but not 
mystery shopping. Consequently, IRS acknowledged that the advance 
notice of these reviews was likely to bias the results. 

While IRS plans to continue to perform site and return reviews in 2008, 
officials said that following a sampling plan that would yield results 
that could be generalized to the universe of volunteer sites would be 
too costly. Instead, IRS plans to conduct unannounced return reviews, 
continue site reviews, and increase the number of mystery shopping 
visits from the 39 done in 2007 to 100. According to IRS officials, 
mystery shopping yields the best evaluative result of the three reviews 
IRS uses to assess the quality of volunteer-prepared returns. We agree 
that having unannounced visits and increasing the number of mystery 
shopping visits is likely to yield a better evaluative result than the 
potentially biased methods previously used for monitoring return 
preparation. Furthermore, we and TIGTA acknowledge the difficulties in 
collecting statistically valid information about return preparation 
assistance at volunteer sites.[Footnote 41] 

In addition to quality, IRS uses some other measures to monitor the 
performance of its volunteer program, such as the number of returns 
prepared, percentage prepared electronically, and the satisfaction of 
its partners, such as financial, social service, corporate, and 
government organizations. While useful, these measures do not reflect 
the extent to which the volunteer program improves taxpayer compliance 
and participation in programs such as the earned income tax credit. IRS 
has increased its reliance on the volunteer program and plans to commit 
more resources. However, as we testified earlier this year, IRSís 
budget submission did not include basic performance information about 
the program.[Footnote 42] More specifically, IRS does not know the 
extent to which the program influences the taxpaying behavior of its 
target populations. 

Regarding IRSís walk-in sites, IRS recently started implementing 
contact recording intended to better measure the accuracy of tax law 
and account assistance.[Footnote 43] As shown in figure 6, contact 
recording automatically captures the audio portion of a taxpayer 
contact and synchronizes it with an IRS walk-in site assistorís 
computer screen activity. With the exception of a microphone that sits 
atop the desk of the walk-in site assistor, contact recording is 
similar to the system IRS uses to monitor and assess the quality of 
telephone interaction between taxpayers and telephone assistors. IRS 
has deployed contact recording at 127 of its 401 walk-in sites, which 
represents 46 percent of walk-in site contacts. As of May 26, 2007, IRS 
data show accuracy rates of 85 percent for accounts assistance and 80 
percent for tax law assistance, and IRS officials expect accuracy rates 
to improve as IRS deploys contact recording at more sites.[Footnote 44] 

Figure 6: Contact Recording at Walk-in Sites: 

[See PDF for image] 

This figure is an illustration of the following way in which contact is 
recorded at walk-in sites: 

* Walk-in staff initiate contact recording, which captures taxpayer 
contacts and screen displays. 

* Data are stored in a database for subsequent quality review. 

* Managers and quality review staff hear and ďseeĒóthrough screen 
displaysóthe taxpayer and walk-in staff interaction. 

* Walk-in staff reexperience both sides of the interaction, to improve 
performance. 

Source: GAO analysis of IRS data and Art Explosion (images). 

[End of figure] 

Although IRS currently has plans to fully deploy the contact recording 
systems to all walk-in sites by 2009, contact recording is not included 
in its plans to assess the quality of return preparation 
assistance.[Footnote 45] IRS intends on relying on managerial 
observation of employee/taxpayer interactions while providing return 
assistance. We have previously reported that employeesí performance 
could be influenced by the knowledge that they are being observed by 
managers, biasing the results. Additionally, managers were not 
consistently coding employee performance.[Footnote 46] Consequently, 
without an unbiased method, IRS lacks reliable information to assess 
performance, establish obtainable goals, and measure progress toward 
goals.[Footnote 47] 

According to IRS officials, contact recording used to measure the 
accuracy of tax law and account assistance could be extended to include 
return preparation assistance. For example, the equipment is already at 
many of the sitesóit is used to measure the accuracy of tax law and 
account assistance, but turned off for return assistance. IRS officials 
said that IRS had not yet determined the feasibility of using contact 
recording for return assistance. IRS officials were concerned that 
contact recording, as it is currently configured, might miss some 
aspects of return preparation assistance, such as verifying taxpayersí 
documents. Further, IRS officials noted that because it takes longer to 
prepare a return than answer a tax law or account question, managerial 
review of return preparation takes longer than for tax law or account 
assistance. However, officials also noted that managers would be likely 
to spend the same amount of time using the manager observation method 
as with contact recording but manager observation would not yield 
reliable results. Without fully determining the feasibility of using 
contact recording for return assistance, IRS will not know if the 
concerns could be addressed and the existing equipment could be more 
fully utilized.[Footnote 48] 

The Value of TAB Depends on Funding and Ability to Link Taxpayer 
Service to Compliance: 

TAB is a data-driven strategy that includes the results of extensive 
research designed to assist IRS in better understanding taxpayersí 
service needs.[Footnote 49] For example, TAB research included several 
surveys that provided IRS with information about the services that 
taxpayers use or would use. TAB also provides plans to improve taxpayer 
services and a multiyear portfolio of research designed to provide 
information on the value of IRS services to taxpayers and partners and 
the effect of taxpayer service on compliance. According to IRS 
officials, with TAB, they now have more information about taxpayers 
than ever before. 

IRS developed estimates of the cost-per-service contact for providing 
different types of taxpayer services, although there were 
qualifications to those estimates.[Footnote 50] According to TAB, these 
estimates will provide a baseline methodology for computing and 
comparing the costs of services in the future. Having reliable cost 
information, together with a better understanding of taxpayersí 
preferences, needs, and expectations, will assist IRS in determining 
whether it can provide taxpayers with the same services and benefits at 
lower cost through alternative methods. As table 4 shows, these costs 
vary widely depending on the type of service provided. 

Table 4: IRSís Fiscal Year 2005 Estimated Unit Costs: 

Service: Answering tax law questions via e-mail; 
Estimated cost-per-contact: $52.51. 

Service: Providing assistance at walk-in sites; 
Estimated cost-per-contact: $28.73. 

Service: Answering correspondence; 
Estimated cost-per-contact: $24.97. 

Service: Providing assistance by assistors via toll-free telephones; 
Estimated cost-per-contact: $19.46. 

Service: Providing assistance through VITA sites; 
Estimated cost-per-contact: $12.01. 

Service: Providing assistance by automation via toll-free telephones; 
Estimated cost-per-contact: $0.71. 

Service: Providing assistance such as downloads and searches on IRSís 
Web site; 
Estimated cost-per-contact: $0.13. 

Source: GAO analysis of IRS data. 

TAB has potential value. In the past, we have stressed the importance 
of reconsidering the level and types of services IRS provides to 
taxpayers, such as whether to maintain the same number of walk-in 
sites.[Footnote 51] While we support IRSís efforts to develop a data-
driven strategy, the value of TAB will depend on several factors, such 
as: 

* Quantifying the linkage between taxpayer service and voluntary 
compliance. As noted above, TAB outlines a research portfolio that 
includes additional research on the impact of taxpayer service on 
voluntary compliance, which is consistent with the Department of 
Treasury tax gap report. According to TAB, a better understanding of 
the causes of noncompliance and the impact of specific taxpayer 
services on compliance would allow IRS to focus investments in taxpayer 
service to positively impact compliance. However, TAB recognizes that 
previous research has shown that quantifying this linkage is difficult. 

* Coordinating and managing oversight responsibilities for TAB 
initiatives. IRS recently established a TAB Program Management Office 
responsible for facilitating, coordinating, and integrating TAB 
activities. The Program Management Office currently does not have 
authority to allocate funding for initiatives, and TAB officials said 
that it could be difficult to develop initiatives that require funding 
or coordination between different organizations without such authority. 

* Obtaining funding. While some of the initiatives under TAB are funded 
mostly as extensions of existing efforts, most are not yet funded, and 
TAB does not contain any information related to how much, as a whole, 
it will cost. According to IRS, unfunded and future initiatives will 
have to compete with other IRS projects for funding within approved 
annual budgets. 

* Including submission processing services. IRS officials acknowledge 
that submission processing is a significant component of service 
delivery and must be integrated into TAB. The timely processing of 
returns supports tax compliance and possibly reduces the need for 
taxpayer services. 

Conclusions: 

Over the last decade, IRS has improved its performance processing tax 
returns and providing taxpayer assistance. We see value in two IRS 
initiatives intended to further improve this performance. TAB, besides 
being a data-driven strategy for further improving service, may provide 
IRS with a better understanding of the effect of taxpayer service on 
compliance and thus of the return on IRSís large investment in service. 
Similarly, obtaining more information about paid preparers is valuable 
because doing so should give the Congress and IRS a better basis for 
making decisions about the future oversight of paid preparers. 

In addition to these initiatives, we have identified other potential 
means for improving service. One opportunity is to increase 
electronically filed returns, either through mandates or bar coding, 
and transcribing the remaining lines on residual paper returns. 
However, IRS does not have the benefit/cost estimates that would be 
useful to make decisions about how to proceed. In particular, IRS does 
not know the actions needed to require software vendors to include bar 
codes on printed individual income tax returns and the cost of those 
actions; the benefits, in terms of processing costs and improved 
enforcement, of having all information available electronically; and 
how much electronic filing would have to increase for the benefits of 
transcribing all remaining paper returns to exceed the costs. Such 
information would allow IRS to compare the costs of additional 
transcription to the benefits of processing costs savings, additional 
enforcement revenues and reduced taxpayer burden. 

Additionally, despite increasing reliance on volunteer organizations to 
target underserved taxpayer groups, IRS lacks information on the 
effectiveness of its efforts. Finally, while IRS has been implementing 
contact recording at walk-in sites to monitor the accuracy of tax law 
and account assistance, it has not determined the feasibility of 
extending contact recording to its low-income return preparation 
assistance program. In both cases, the missing information impedes 
IRSís ability to make data-based decisions about how to improve 
assistance to underserved taxpayers. 

Recommendations for Executive Action: 

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

* determine actions needed to require software vendors to include bar 
codes on printed individual income tax returns and the cost of those 
actions; 

* determine the benefits, in terms of processing costs and improved 
enforcement, of having all return information available electronically; 

* determine how much electronic filing would have to increase, either 
through electronic filing mandates or bar coding, for the benefits of 
transcribing all remaining paper returns to exceed the costs; 

* develop estimates of the effectiveness of IRSís volunteer program at 
targeting underserved populations; and: 

* determine the feasibility of using contact recording as a method of 
monitoring and improving the quality of return preparation assistance 
at IRSís walk-in sites. 

Agency Comments and Our Evaluation: 

The Acting Commissioner of Internal Revenue provided written comments 
in a November 7, 2007 letter in which she agreed with all our 
recommendations and outlined IRSís actions to address those 
recommendations. With respect to reducing processing costs and 
improving enforcement, IRS plans on specifically addressing these 
recommendations in a report to the Congress. With respect to assessing 
the effectiveness and quality of IRSís face-to-face services, IRS 
reported it plans to develop methods of estimating how it can provide 
better face-to-face assistance. 

As agreed with your offices, unless you publicly announce its contents 
earlier, we plan no further distribution of this report until 30 days 
after its date. At that time, we will send copies of this report to the 
Secretary of the Treasury; the Acting Commissioner of Internal Revenue; 
the Director, Office of Management and Budget; relevant congressional 
committees; and other interested parties. This report is available at 
no charge on GAOís web site at [hyperlink, http://www.gao.gov]. 

For further information regarding this testimony, please contact James 
R. White, Director, Strategic Issues, on 202-512-9110 or 
whitej@gao.gov. Contacts for our Offices of Congressional Relations and 
Public Affairs may be found on the last page of this statement. 
Individuals making key contributions to this testimony include Joanna 
Stamatiades, Assistant Director; Amy Dingler; Evan Gilman; Timothy D. 
Hopkins; Jennifer McDonald; Paul B. Middleton; Emily Norman; Karen 
OíConor; Cheryl Peterson; Neil Pinney; and Lerone Reid. 

If you or your staffs have any questions, please contact me at (202) 
512-9110 or WhiteJ@gao.gov. 

Signed by: 

James R. White: 
Director: 
Tax Issues: 
Strategic Issues: 

[End of section] 

Appendix I: IRSís Processing Performance Relative to 2001-2006 
Performance and 2007 Goals: 

As of June 30, the Internal Revenue Service met or exceeded seven out 
of nine processing performance goals and continues a general trend of 
improvement. As shown in table 6 below, IRS met or exceeded its goals 
for the percentage of errors included in letters, notices, refunds, and 
deposits; deposit timeliness (i.e., interest foregone by untimely 
deposits); and productivity. IRS met its goal for refund timeliness but 
did not meet its goals for refund interest paid and Individual Master 
File efficiency. IRS officials said that the lighter than expected TETR 
volume meant that IRSís goal for Individual Master File efficiency 
could not be achieved because TETR returns were the easiest to process. 
Additionally, problems with a computer system that processes some 
taxpayer identification numbers and the inexperience of new staff 
adversely affected performance on refund interest paid. Table 5 also 
shows that IRSís processing performance has significantly improved in 
some areas, such as the deposit and refund error rates, since 2001. 

Table 5: IRSís Processing Performance, Fiscal Years 2001ó2007: 

Measure name: Deposit error rate[C]; 
Definition: Percentage of payments applied in error by, for example, 
reimbursing a taxpayer who overpaid when the taxpayer wanted the 
overpayment credited to next yearís tax bill;
Fiscal year 2002 actual: 4.8%(+/-.3%); 
Fiscal year 2003 actual: 4.2%(+/-.3%); 
Fiscal year 2004 actual: 3.5%(+/-.31%); 
Fiscal year 2005 actual: 2.2% (+/-.26%); 
Fiscal year 2006 actual: 1.6%(+/-.24%); 
Fiscal year 2007 actual (through June)[B]: 1.3%(+/-.25%); 
Fiscal year 2007 goal: 1.5%. 

Measure name: Deposit timeliness Ė paper; 
Definition: Interest foregone by not depositing monies the business day 
after receipt, per $1 million in deposits. Measure assumes an 8 percent 
interest rate; 
Fiscal year 2001 actual[A]: Not comparable because of revisions to the 
measure; 
Fiscal year 2002 actual: Not comparable because of revisions to the 
measure; 
Fiscal year 2003 actual: Not comparable because of revisions to the 
measure; 
Fiscal year 2004 actual: $407; 
Fiscal year 2005 actual: $390; 
Fiscal year 2006 actual: $354; 
Fiscal year 2007 actual (through June)[B]: $343; 
Fiscal year 2007 goal: $380. 

Measure name: Letter error rate[C]; 
Definition: Percentage of letters issued to taxpayers with errors 
(includes systemic errors)[D]; 
Fiscal year 2001 actual[A]: Not comparable because of revisions to the 
measure; 
Fiscal year 2002 actual: 7.4%(+/-.6%); 
Fiscal year 2003 actual: 7.1%(+/-.5%); 
Fiscal year 2004 actual: 6.6%(+/-.38%); 
Fiscal year 2005 actual: 3.1% (+/-.30%); 
Fiscal year 2006 actual: 3.6%(+/-.34%); 
Fiscal year 2007 actual (through June)[B]: 2.3%(+/-.37%); 
Fiscal year 2007 goal: 3.0%. 

Measure name: Notice error rate[C]; 
Definition: Percentage of incorrect notices issued to taxpayers 
(includes systemic errors)[D]; 
Fiscal year 2001 actual[A]: Not comparable because of revisions to the 
measure; 
Fiscal year 2002 actual: 18.7%(+/-2.4%); 
Fiscal year 2003 actual: 9.4%(+/-1.2%); 
Fiscal year 2004 actual: 9.5%(+/-1.3%); 
Fiscal year 2005 actual: 9.2% (+/-1.6%); 
Fiscal year 2006 actual: 5.4%(+/-1.1%); 
Fiscal year 2007 actual (through June)[B]: 4.3%(+/-1.0%); 
Fiscal year 2007 goal: 7.3%. 

Measure name: Refund error rate - individual (paper)[C]; 
Definition: Percentage of refunds with IRS-caused errors in the entity 
information (e.g., incorrect name, Social Security number, or refund 
amount); includes systemic errors.[D] 
Fiscal year 2001 actual[A]: 9.8%; 
Fiscal year 2002 actual: 8.0%(+/-.46%); 
Fiscal year 2003 actual: 5.3%(+/-.41%); 
Fiscal year 2004 actual: 4.9%(+/-.44%); 
Fiscal year 2005 actual: 5.0% (+/-.48%); 
Fiscal year 2006 actual: 4.5%(+/-.46%); 
Fiscal year 2007 actual (through June)[B]: 2.8%(+/-.41%); 
Fiscal year 2007 goal: 4.6%. 

Measure name: Refund interest paid; 
Definition: Amount of refund interest IRS paid per $1 million of 
refunds issued; 
Fiscal year 2001 actual[A]: Not comparable because of revisions to the 
measure; 
Fiscal year 2002 actual: Not comparable because of revisions to the 
measure; 
Fiscal year 2003 actual: $36.29; 
Fiscal year 2004 actual: $20.55; 
Fiscal year 2005 actual: $29.21; 
Fiscal year 2006 actual: $30.12; 
Fiscal year 2007 actual (through June)[B]: $36.09; 
Fiscal year 2007 goal: $34.00. 

Measure name: Refund timeliness - individual (paper)[C]; 
Definition: Percentage of refunds issued within 40 days or less; 
Fiscal year 2001 actual[A]: 95.2%; 
Fiscal year 2002 actual: 98.2%(+/-.32%); 
Fiscal year 2003 actual: 98.8%(+/-.26%); 
Fiscal year 2004 actual: 98.3%(+/-.17%); 
Fiscal year 2005 actual: 99.2% (+/-.18%); 
Fiscal year 2006 actual: 99.3%(+/-.13%); 
Fiscal year 2007 actual (through June)[B]: 99.3%(+/-.18%); 
Fiscal year 2007 goal: 99.2%. 

Measure name: Productivity; 
Definition: Weighted volume of documents processed per staff year 
expended at the processing centers; 
Fiscal year 2001 actual[A]: 30,133; 
Fiscal year 2002 actual: 28,389; 
Fiscal year 2003 actual: 30,179; 
Fiscal year 2004 actual: 30,405; 
Fiscal year 2005 actual: 31,444; 
Fiscal year 2006 actual: 33,237; 
Fiscal year 2007 actual (through June)[B]: 33,535; 
Fiscal year 2007 goal: 32,640. 

Measure name: Master File efficiency; 
Definition: Measure of Individual Master File returns processed per 
staff year expended; 
Fiscal year 2001 actual[A]: Measure not in existence; 
Fiscal year 2002 actual: Measure not in existence; 
Fiscal year 2003 actual: Measure not in existence; 
Fiscal year 2004 actual: Measure not in existence; 
Fiscal year 2005 actual: 14,965; 
Fiscal year 2006 actual: 16,124; 
Fiscal year 2007 actual (through June)[B]: 23,532; 
Fiscal year 2007 goal: 24,650. 

Source: GAO. 

Note: GAO analysis of IRS data. 

[A] According to IRS officials, they did not compute a margin of error 
for these measures in 2001. 

[B] The measures for fiscal year 2006 are through July 31, which were 
the latest data available at the time we ended our audit work. 
According to IRS officials, the 2006 results through July 31 are 
reflective of IRSís performance during the filing season. In addition, 
IRS officials told us that the results for the measures should not 
change significantly through September 30. 

[C] IRS estimates these measures to have a 90 percent confidence 
interval. 

[D] Systemic errors are computer-generated errors over which a 
particular processing center would have no control. 

[End of table] 

[End of section] 

Appendix II: Comments from the Internal Revenue Service: 

Commissioner: 
Department Of The Treasury: 
Internal Revenue Service: 
Washington, DC 20224: 

November 7, 2007: 

Mr. James R. White: 
Director, Tax Issues: 
U.S. Government Accountability Office: 
441 G Street, NW: 
Washington, DC 20548: 

Dear Mr. White: 

I have reviewed your draft report entitled Tax Administration: 2007 
Filing Season Continues Trend of Improvement, but Opportunities to 
Reduce Costs and Increase Tax Compliance Should be Evaluated. I 
appreciate your recognition of our significant achievements in 
successfully delivering the 2007 filing season, which as you note 
continues a long-term trend of continually improving IRS services to 
taxpayers. 

Processing Ė We had an outstanding filing season, successfully 
implementing significant and sometimes extremely late tax law changes. 
Through September 28, 2007, we processed over 134 million individual 
income tax returns and issued over 103 million refunds, totaling nearly 
$234 billion. 

Electronic filing grew again this year with 78.7 million, or 58.5 
percent, of individual taxpayers filing electronically. The remaining 
55.9 million were filed on paper. The number of returns filed 
electronically increased 9 percent over last year, which exceeds the 6 
percent increase we achieved in the prior year. So far, this year's e-
file rate has also exceeded our projection for an 8 percent increase 
for the year. 

The most significant increase in e-file occurred in the number of 
taxpayers who filed from their home computer. Over 22.4 million returns 
were filed from home, nearly an 11 percent increase from the prior 
year. Over 3.8 million taxpayers took advantage of the free online 
filing services offered by the Free File Alliance. Also notable is the 
fact that over 66.6 million taxpayers electronically signed their 
electronic returns, further increasing the efficiency of e-file. 
Additionally, more than 65,000 individuals have registered for e-
services and over 53 percent of applications to electronically file 
were submitted online.
Through May 2007, the IRS received 84.5 million individual tax returns 
with Telephone Excise Tax Refund (TETR) claims totaling $3.6 billion. 
The IRS processed over 600,000 paper, and 100,000 e-file, Forms 1040EZ-
T for taxpayers without a return filing requirement that claimed the 
TETR. The IRS also received 400,000 business tax returns with TETR 
claims totaling $204.5 million. 

For the first time, the IRS offered taxpayers the opportunity to split 
their refunds and direct deposit them into multiple accounts. We 
successfully processed approximately 80,000 split refund requests, 
which was significantly lower than the anticipated 1.4 million. 
However, several major software vendors did not provide the ability to 
split refunds in their tax preparation software and may have 
contributed to the lower than expected volume. 

Late passage of the Tax Relief and Health Care Act of 2006 included an 
extension of the tuition and fees deduction, state and local sales tax 
deduction, and the deduction for educator expenses. The late passage of 
this legislation caused IRS delays in processing some returns until mid-
February. Nonetheless, the IRS successfully managed taxpayer, 
practitioner, and tax software-provider expectations through extensive 
information and outreach. As a result, there were minimal negative 
consequences for taxpayers from this delay. 

Successful Implementation of the Customer Account Data Engine (CADE) 
continued during 2007. Release 2.2, although delayed, successfully 
deployed in March 2007 followed by the successful deployment of Release 
3.1 into production in August 2007. This portion of Release 3.1 
introduced Disaster Area Designation Processing into CADE and has 
resulted in the ability to maintain over 616,000 accounts within CADE 
that would have returned to the Individual Master File without this 
functionality. The second portion of Release 3.1 introduces online 
address changes, which enables CADE to process and post address changes 
for the first time. In addition to faster refunds, these new 
functionalities working together set the framework to enable IRS to 
post additional account changes in a near real-time manner in the 
future. As of September 28, 2007, CADE posted over 11.2 million returns 
and disbursed over 10.9 million refunds for over $11.6 billion. 

Telephone Service - We appreciate your recognition that customer access 
was better this year than last year. This increase in access was made 
possible through proper planning and the IRS ability to quickly respond 
to actual demand by redeploying resources. Although we overestimated 
the demand for telephone assistance related to TETR, the demand for 
other telephone assistance services was slightly higher than last year. 
We achieved an 81.6 percent Level of Service for the filing season, 
exceeding the planned Level of Service of 81.1 percent and ended the 
fiscal year at 82.1 percent, exceeding the goal of 82 percent. We also 
successfully managed our staffing to ensure non-telephone 
correspondence inventories were maintained at levels below Fiscal Year 
2006. 

In October 2006, we transitioned to Customer Voice Portal, which is a 
key telephone technology implemented to improve customer experience by 
equalizing wait times within the various topics. This revealed 
approximately nine to twelve seconds that was not previously accounted 
for when calculating the length of time a caller waits before 
connecting to an agent. We report the average of this wait time as 
average speed of answer (ASA). This additional time increased our ASA. 
Despite this, we achieved an ASA of 273 seconds, four seconds higher 
than the planned filing season ASA of 269 seconds, and delivered a 
fiscal year ASA of 266 seconds, six seconds above the goal of 260 
seconds. 

During peak periods, we will sometimes advise the taxpayer to call back 
at a less busy time and disconnect the call. In an effort to improve 
the customer experience this year, we administered fewer courtesy 
disconnects up front allowing customers to determine how long they were 
willing to wait to receive assistance. This decision contributed to a 
10 percent decrease when combining taxpayer and courtesy disconnects 
for the filing season. 

We appreciate your continued recognition of our improvement and high 
performance in the accuracy of our tax law and account answers on the 
toll-free telephone lines. The 2007 accuracy rates for toll-free tax 
law and accounts surpassed FY 2006 achievements. Through September 
2007, we achieved a Toll-Free Tax Law accuracy rate of 91.2 percent. We 
continually focus on identifying defects and provide the information to 
call sites to share with telephone assistors. Through September 2007, 
the Toll-Free Accounts accuracy rate was 93.4 percent. Accuracy gains 
can be attributed to the improved quality review process, conference 
calls held to address specific tax law or accounts topics, and Contact 
Recording that allows managers to listen to and review recorded 
customer contacts as a part of performance feedback to employees. 
Customer satisfaction for toll-free through July was 94 percent, with a 
2 percent dissatisfaction rate demonstrating the excellent service that 
we continue to provide our customers. 

Website Ė The IRS website, IRS.gov, is consistently one of the most 
heavily used government sites. So far in 2007, our website has been 
visited more than 214 million times, a 10.3 percent increase over 2006. 
These visits resulted in more than 1.3 billion page views, an almost 4 
percent increase over 2006. Visits are the measurement of actions 
beginning when a visitor views their first web page on IRS.gov and ends 
when the visitor leaves our website. Taxpayers and practitioners also 
downloaded more than 128 million forms, publications, instructions, and 
other documents. 

Walk-In Assistance - During the 2007 filing season, the IRS continued 
to provide services at all of its 401 Taxpayer Assistance Centers. As 
of May 26, 2007, accuracy of the services we provided in these walk-in 
sites was 85 percent for accounts assistance and 80 percent for tax law 
assistance. To better measure the accuracy of tax law and account 
assistance, the IRS recently started implementing Contact Recording. As 
you note in your report, we have deployed Contact Recording in 127 of 
the 401 walk-in sites and plan to fully deploy this systems to all walk-
in sites by 2009. 

In addition, the IRS successfully responded to late legislation by 
hiring seasonal employees to lessen the impact of anticipated demands 
of the TETR legislation. The IRS also conducted additional training for 
all employees to address legislative changes enacted after our regular 
fall 2006 Continuing Professional Education classes. 

Volunteer Assistance - During the 2007 filing season, approximately 
76,619 IRS supported volunteers in 11,922 locations provided assistance 
to traditionally underserved populations. These include individuals 
with low incomes, the elderly, the disabled, rural, Native American, 
and those with Limited-English proficiency. Through September 2007, 
over 2.6 million returns were filed through our Volunteer Income Tax 
Assistance and Tax Counseling for the Elderly sites, an increase of 
15.8 percent over the prior year. Of these, 84.6 percent were e-filed, 
a 2.4 percent increase over 2006. To increase awareness of these 
services, as well as taxpayer education on basic income tax return 
filing and payment requirements, the availability of the Earned Income 
Tax Credit, e-file, and other tax related issues, more than 410 million 
outreach contacts were made during 2007. These contacts include 
multiple messages on varied topics to targeted taxpayer segments 
through both media and non-media channels. In order to serve more 
taxpayers through these programs, we also expanded our partnerships 
with community-based coalitions to approximately 326 in 2007. These 
strategic partnerships leverage IRS resources and allow us to reach far 
more taxpayers through trusted local community sources. 

Responses to your specific recommendations are enclosed. I appreciate 
your observations on the successful filing season for 2007, and if you 
have any questions, please contact me or Floyd Williams, Director, 
Legislative Affairs, at (202) 622-3720. 

Sincerely, 

Signed by: 

Linda E. Stiff: 
Acting Commissioner of Internal Revenue: 

Enclosure: 

[End of letter] 

Enclosure: 

Recommendation for the Commissioner: 
Determine actions needed to require software vendors to include bar 
codes on printed individual income tax returns and the cost of those 
actions. 

Response: 
We agree with this recommendation. Senate Report 110-129 of the 
Financial Services and General Government Appropriations Bill of 2008 
requires the IRS to provide to Congress a report addressing 2-D Bar 
Coding as stated, "Accordingly, the Committee directs the IRS, in 
consultation with stakeholders, such as the National Taxpayer Advocate, 
to develop a detailed strategic plan to meet the 80 percent e-File 
goal. This plan should address alternate electronic filing strategies, 
including Telefile and 2-D Bar Coding and methods of e-filing directly 
with the IRS for free." 

The report to Congress will include the information requested in your 
recommendation and specifically address: 

* The impact on commercial software providers to include bar codes on 
paper returns. 

* The IRS cost for processing bar coded returns. 

* Cost-benefit analysis of capturing all tax return information. 

* "Break-even" point paper processing versus electronic filing. 

Recommendation for the Commissioner: 
Determine the benefits, in terms of processing costs and improved 
enforcement, of having all return information available electronically. 

Response: 
We agree with this recommendation. Our planned actions in response to 
the first recommendation above will also address this recommendation. 

Recommendation for the Commissioner: 
Determine how much electronic filing would have to increase, either 
through electronic filing mandates or bar coding, for the benefits of 
transcribing all remaining paper returns to exceed the costs. 

Response:
We agree with this recommendation. Our planned actions in response to 
the first recommendation above will also address this recommendation. 

Recommendation for the Commissioner: 
Develop estimates of the effectiveness of IRS's volunteer program at 
targeting underserved populations. 

Response:
We agree with the recommendation. Our Stakeholder Partnerships, 
Education and Communication (SPEC) organization is currently working to 
develop appropriate measures to determine the success of targeted 
activities in specific underserved markets. Our territories throughout 
the country have targeted their local underserved markets (including 
low-income, disabled, rural, elderly, and Limited-English Proficient) 
since the October 2000 standup of the organization. The type and extent 
of activities vary, depending upon the local population, identified 
partners, partner needs and that of their customers, and the amount of 
available SPEC resources. The SPEC organization has recently taken 
steps to bring additional focus to these efforts by developing national 
initiatives to address these markets. National teams comprised of 
headquarters and field managers and employees are assessing current 
activities, identifying gaps, and developing action plans for 
nationwide direction. 

Recommendation for the Commissioner:
Determine the feasibility of using contact recording as a method of 
monitoring and improving the quality of return preparation assistance 
at IRS's walk-in sites. 

Response:
We agree with this recommendation. The IRS will determine the 
feasibility of testing the use of contact recording as a tool for 
monitoring the quality of return preparation assistance at walk-in 
sites. We will assess several methodologies using a Data Collection 
Instrument to monitor accuracy of return preparation. This is a new 
area of review with additional contingencies not experienced in 
accounts and tax law contacts. At the end of testing, we plan to 
recommend whether or not to adopt Centralized Quality Review Systems as 
a new review process. 

[End of section] 

Footnotes: 

[1] Most taxpayers file their tax returns between January 1 and April 
15, which is the deadline for filing individual income tax returns. 
However, millions of taxpayers receive extensions from IRS, which 
allows them to delay filing until as late as October 15. 

[2] See, for example, GAO, Tax Administration: Most Filing Season 
Services Continue to Improve, but Opportunities Exist for Additional 
Savings, GAO-07-27 (Washington, D.C.: Nov. 15, 2006). Pub. L. No. 105-
206 (1998). 

[3] Joint Explanatory Statement of the Committee of Conference: 
Internal Revenue Service, Processing Assistance, and Management 
(Including Rescission of Funds), H. R. Rep. No. 109-307 (2005). 

[4] See, for example, GAO, Paid Tax Return Preparers: In a Limited 
Study, Chain Preparers Made Serious Errors, GAO-06-563T (Washington, 
D.C.: Apr. 4, 2006). 

[5] GAO, Internal Revenue Service: Assessment of the 2008 Budget 
Request and an Update of 2007 Performance, GAO-07-719T (Washington, 
D.C: May 9, 2007), 2007 Tax Filing Season: Interim Results and Updates 
of Previous Assessments of Paid Preparers and IRSís Modernization and 
Compliance Research, GAO-07-720T (Washington, D.C: Apr. 12, 2007), and 
Internal Revenue Service: Interim Results of the 2007 Tax Filing Season 
and the Fiscal Year 2008 Budget Request, GAO-07-673 (Washington, D.C.: 
Apr. 3, 2007). 

[6] We use ďpaid preparerĒ to refer to any individual acting alone or 
as an employee of an entity who provides fee-based return preparation 
assistance. Within this broad classification, there are tax 
practitioners and commercial preparers. Tax practitioners, who are 
governed by the Department of the Treasury Circular 230, are authorized 
to represent taxpayers before IRS. They can include attorneys, 
certified public accountants, and enrolled agents. Commercial preparers 
include corporations and self-employed individuals who are not 
regulated by IRS or governed by Circular 230. While there are currently 
no federal competence standards or registration requirements for 
unenrolled commercial preparers, federal legislation has been proposed 
to mandate Treasury to test and certify tax return preparers. 

[7] GAO, Assessing the Reliability of Computer-Processed Data, GAO-02-
15G (Washington, D.C.: Sept.1, 2002). 

[8] While taxpayers and paid preparers can mail a paper return directly 
to IRS, only paid preparers and tax preparation software companies that 
IRS has designated as electronic return originators (ERO) can transmit 
tax returns electronically to IRS, which may involve a fee to the 
taxpayers. IRS conducts suitability checks of applicants who are 
applying to become EROs. These checks may include checking applicantís 
criminal background, credit history, and tax compliance. 

[9] IRS also has plans to replace its Business Master File legacy 
processing system. 

[10] According to IRS, this limitation approximates the amount set in 
the Internal Revenue Code for claiming the Earned Income Tax Credit. 
IRS has required appointments for taxpayers seeking this assistance 
since 2003. 

[11] The cost accounting module in IRSís Integrated Financial System is 
not yet capable of producing the detailed cost information needed to 
perform estimates. See, GAO, Financial Audit: IRSís Fiscal Years 2006 
and 2005 Financial Statements, GAO-07-136 (Washington, D.C.: Nov. 9, 
2006). 

[12] IRS estimated the gross tax gap to be $345 billion for tax year 
2001. After late payments by taxpayers and revenue brought in by IRSís 
enforcement efforts, the resulting net tax gap is estimated to be $290 
billion. See GAO-07-719T. 

[13] The IRS Commissioner characterized this yearís filing season as 
high-risk because of challenges related to implementing the new 
Telephone Excise Tax Refund, a split refund option that allows for 
direct deposits in up to three separate accounts, and tax law changes 
that were enacted in December 2006. We reported earlier this year that 
these challenges had little impact on the filing season performance, in 
large part because workload failed to materialize. See GAO 07-720T. 

[14] See, for example, GAO-07-720T. 

[15] IRS auditors can get the entire paper return. Consequently, all 
the information can be used during an audit, but not in the automated 
audit selection process. 

[16] The total number of returns processed during the 2007 filing 
season includes 717,000 1040EZ-T forms, of which 100,000 were filed 
electronically. This form was for individuals who requested a refund of 
federal telephone excise tax in 2007 and, therefore, did not apply to 
previous filing seasons. 

[17] See GAO-07-727. 

[18] TIGTA found that some technology and start-up costs were not 
reliable or properly tracked (see Treasury Inspector General for Tax 
Administration, Consolidation of Tax Return Processing Sites Is 
Progressing Effectively, But Improved Project Management Is Needed, 
Reference No. 2007-40-165 (Washington, D.C.: Aug. 31, 2007). 

[19] We have not independently verified this estimate. As we have 
previously reported, IRSís cost accounting system is not yet able to 
support preparation of such cost estimates. See, for example, GAO-07-
136. 

[20] TIGTA determined that IRSís 2007 filing season was successful 
because most key tax law changes were implemented correctly and IRS 
timely processed returns and issued refunds and electronically filed 
returns increased 9 percent over last year. See the Treasury Inspector 
General for Tax Administration, The 2007 Tax Filing Season was 
Generally Successful, and Most Returns were Timely and Accurately 
Processed, Reference No. 2007-40-187 (Washington, D.C.: Sept. 21, 
2007). 

[21] Because of the substantial costs associated with fully 
implementing CADE, the Congress could find information about the 
benefits of CADE, such as faster refund timeliness, useful as an 
indication of both the magnitude and likelihood of the future benefits 
from full implementation. Consequently, last year we recommended that 
IRS report to the Congress on refund timeliness for CADE compared to 
the Master File legacy system. According to IRS officials, IRS is in 
the process of developing this measure. 

[22] The most recent release of CADE, originally scheduled for delivery 
in July 2007, was deployed into production on August 8, 2007, but with 
reduced functionality. This release will also interface with AMS, 
currently scheduled to be deployed in October 2007 to allow address 
changes for CADE accounts. 

[23] Full data for 2007 is not available. 

[24] GAO, Tax Administration: Electronic Filingís Past and Future 
Impact on Processing Costs Dependent on Several Factors, GAO-02-205 
(Washington, D.C., Jan. 10, 2002). 

[25] IRS officials said that if all data from tax returns were 
transcribed and posted, the Automated Underreporting Program could 
eliminate human screeners who currently review return information that 
has not been transcribed or posted. IRS estimated that if the screeners 
could be reallocated to performing audits, they could bring an 
additional $175 million annually. 

[26] GAO-07-27. 

[27] According to a recent TIGTA report, some states estimate that bar 
codes reduce the time and effort involved in processing paper returns 
by 50 to 90 percent and use of bar codes is $.57 to $.70 cheaper to 
process per return than a conventional paper return. See Treasury 
Inspector General for Tax Administration, Additional Action is Needed 
to Expand the Use and Improve the Administration of the Free File 
Program, Reference No. 2007-40-105 (Washington, D.C.: June 27, 2007). 

[28] Such an analysis may have to rely on estimates, since IRSís cost 
accounting system is not yet capable of producing the detailed cost 
information. See GAO-07-136. 

[29] GAO previously reported statistics that support that this upward 
trend has existed since at least 1981. See GAO, Tax Administration: 
Most Taxpayers Believe They Benefit from Paid Tax Preparers, but 
Oversight for IRS Is a Challenge, GAO-04-70 (Washington, D.C.: Oct. 31, 
2003). 

[30] GAO-06-563T. 

[31] GAO-04-70. 

[32] Internal Revenue Service, U.S. Department of the Treasury, 
Reducing the Federal Tax Gap: A Report on Improving Voluntary 
Compliance (Washington, D.C.: Aug. 2, 2007). 

[33] Legislation has been introduced in the Congress to regulate paid 
preparers. See the Taxpayer Protection and Assistance Act of 2007, S. 
1219, 110th Cong. ß 1 (2007). 

[34] IRS does not have a specific productivity measure for telephone 
assistance, but instead uses two measures, average handle timeóthe 
average amount of time assistors spend assisting taxpayersóand customer 
contacts resolved per staff yearóthe number of contacts resolved in 
relation to time expended based on staff usage. As of last June, 
average handle time was 10 minutes and comparable to last year and the 
number of contacts resolved far surpassed IRSís goal. 

[35] Of the 42 million calls answered by automation or assistors as of 
June 30, only about 716,000 were related to the telephone excise tax 
refunds, a small portion of the 17.6 million expected calls. 

[36] GAO, IRS Modernization, Continued Progress Requires Addressing 
Resource Management Challenges, GAO-05-707T (Washington, D.C.: May 19, 
2005) and IRS Telephone Assistance: Opportunities to Improve Human 
Capital Management, GAO-01-144 (Washington, D.C.: Jan. 30, 2001). 
Internal Revenue Service, U.S. Department of the Treasury, Reducing the 
Federal Tax Gap: A Report on Improving Voluntary Compliance 
(Washington, D.C.: Aug. 2, 2007). 

[37] GAO-07-27. 

[38] The American Customer Satisfaction Index tracks trends in customer 
satisfaction and is considered to be an industry leader. 

[39] According to IRS, the sales tax deduction calculator allows 
taxpayers to figure out the amount of optional general sales tax they 
are eligible to claim. 

[40] Mystery shopping involves IRS officials visiting volunteer sites 
posing as taxpayers to assess the experience of the taxpayer. Site 
reviews assess whether volunteer preparers were adhering to minimum 
quality standards required by all sites. Return reviews determine 
whether the tax law was applied properly for critical items and 
resulted in an accurate return based on the taxpayer interview and 
supporting documentation. 

[41] Treasury Inspector General for Tax Administration, Internal 
Revenue Service FY 2008 Budget Request, (Washington, D.C.: 2007). 

[42] GAO-07-719T. 

[43] According to IRS officials, the benefits of contact recording 
include, for example, IRSís ability to monitor site activity, train 
employees, and improve quality of service without having to visit the 
site. Importantly, contact recording provides managers with a 
statistically valid sample. 

[44] The accuracy measures include precision margins of +/-2 percentage 
points. 

[45] There are challenges that may delay further deployment. First, 
according to IRS officials and a recent TIGTA report, contact recording 
will have to compete with other systems that may have priorityósee the 
Treasury Inspector General for Tax Administration, System Updates and 
Control Improvements Are Needed to Ensure Contact Recording Will 
Provide an Accurate Assessment of Taxpayer Assistance Quality, 
Reference No. 2007-40-115 (Washington, D.C.: July 9, 2007). Second, 
many walk-in sites will not be able to accommodate contact recording 
without additional modifications to the offices. Finally, IRS is in the 
process of evaluating the viability of its walk-in sites to determine 
which and how many of its sites should be consolidated, expanded or 
closed. 

[46] GAO-05-51. 

[47] GAO, Tax Administration: IRS Improved Performance in the 2004 
Filing Season, but Better Data on the Quality of Some Services are 
Needed, GAO-05-67 (Washington, D.C.: Nov. 10, 2004). 

[48] Additionally, by extending contact recording to return 
preparation, IRS would have an opportunity to learn more about the 
steps taxpayers and preparers need to follow to ensure filing accurate 
returns. 

[49] IRS issued TAB reports in two phases. Phase I outlined the results 
of preliminary research on taxpayer expectations and established five 
strategic themes for improving taxpayer services including optimizing 
use of partner services (e.g., paid preparers and community-based 
organizations) and elevating self-service options, such as on IRSís Web 
site. Phase II included research results on taxpayer service needs and 
preferences and 54 recommended initiatives for service improvements. 

[50] IRS qualified these estimates by noting that they do not fully 
allocate all indirect overhead and support costs and represent the 
average cost per contact rather than the marginal costs per contact, 
which could be significantly lower due to the unused capacity and fixed 
costs associated with many services. Moreover, we previously commented 
on long-standing limitations in IRSís cost accounting capabilities (see 
for example, GAO-07-310 and GAO-07-247). However, it is our 
determination that IRS is a competent source for these cost estimates, 
which seem to be reasonable; therefore, we consider these estimates to 
be sufficiently reliable for describing how IRS could potentially use 
the information to provide taxpayer services at a lower cost. 

[51] See GAO, IRSís 2002 Tax Filing Season: Returns and Refunds 
Processed Smoothly; Quality of Assistance Improved, GAO-03-314 
(Washington, D.C.: Dec. 20, 2002). 

[End of section] 

GAO's Mission: 

The Government Accountability Office, the audit, evaluation and 
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. 

Obtaining Copies of GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through GAO's Web site [hyperlink, http://www.gao.gov]. Each 
weekday, GAO posts newly released reports, testimony, and 
correspondence on its Web site. To have GAO e-mail you a list of newly 
posted products every afternoon, go to [hyperlink, http://www.gao.gov] 
and select "Subscribe to Updates." 

Order by Mail or Phone: 

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. Government Accountability Office: 
441 G Street NW, Room LM: 
Washington, D.C. 20548: 

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

To Report Fraud, Waste, and Abuse in Federal Programs: 

Contact: 

Web site: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]: 
E-mail: fraudnet@gao.gov: 
Automated answering system: (800) 424-5454 or (202) 512-7470: 

Congressional Relations: 

Gloria Jarmon, Managing Director, JarmonG@gao.gov: 
(202) 512-4400: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7125: 
Washington, D.C. 20548: 

Public Affairs: 

Chuck Young, Managing Director, youngc1@gao.gov: 
(202) 512-4800: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7149: 
Washington, D.C. 20548: