This is the accessible text file for GAO report number GAO-03-314 
entitled 'IRS's 2002 Tax Filing Season: Returns and Refunds Processed 
Smoothly; Quality of Assistance Improved' which was released on 
December 20, 2002.



This text file was formatted by the U.S. General Accounting 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.



Report to the Chairman, Subcommittee on Oversight, Committee on Ways 

and Means, House of Representatives:



United States General Accounting Office:



GAO:



December 2002:



IRS’s 2002 Tax Filing Season:



Returns and Refunds Processed Smoothly; Quality of Assistance Improved:



2002 Tax Filing Season:



GAO-03-314:



GAO Highlights:



Highlights of GAO-03-314, a report to the Chairman, Subcommittee on 

Oversight, Committee on Ways and Means, House of Representatives. 



Why GAO Did This Study:



The tax filing season is when millions of taxpayers file their 

returns and seek assistance by calling or visiting IRS’s offices 

or Web site. Because of the large number of returns and critical 

nature of IRS’s filing season activities, GAO was asked to assess 

IRS’s 2002 filing season performance in processing tax returns and 

refunds and providing timely and accurate assistance to taxpayers. 



What GAO Found:



The Internal Revenue Service (IRS) processed returns and issued 

refunds smoothly and the quality of assistance provided to taxpayers 

improved in the 2002 filing season. In light of this, IRS should be 

commended for the various efforts it took to prepare for the 2002 

filing season and improve performance. Still, opportunities exist 

for IRS to further improve aspects of its performance and some of 

its performance measures. In 2002, IRS’s performance included 

issuing almost all refunds on time, providing more accurate telephone 

service than in 2001, and meeting many of its 2002 performance goals 

in all areas. IRS also began measuring the accuracy of assistance at 

its walk-in sites to obtain better performance data. In addition, 

IRS’s redesigned Web site was easier to access and more user friendly.



IRS’s improved performance can be traced to its use of performance 

measures, which are part of its strategy to improve returns processing 

and taxpayer assistance as shown below. However, GAO identified 

opportunities for IRS to make further improvements. For example, 

IRS’s suite of telephone measures lacks an indicator of how long 

callers wait to speak to an assistor-a key aspect of assistance that 

provides useful information for decision making by external 

stakeholders.



Although not a primary focus of this report, GAO also found that 

IRS lacked comprehensive plans related to the extent and timing of 

anticipated workload and staffing changes at its walk-in sites. 



[See PDF for image]



[End of figure]



What GAO Recommends:



GAO recommends that IRS (1) reinstate a telephone assistance caller 

wait-time measure, (2) revise its bi-weekly wait-time reports for 

walk-in sites, (3) explore the feasibility of revising its sampling 

plan for assessing the accuracy of walk-in account assistance, and 

(4) comprehensively evaluate walk-in site resource requirements after 

clarifying the extent and timing of workload and other changes at the 

sites.



In commenting on a draft of this report, IRS cited actions it has 

taken or plans to take that appear responsive to three of GAO’s four 

recommendations. IRS’s comments did not specifically address the 

recommendation that bi-weekly  wait-time reports be revised.



www.gao.gov/cgi-bin/getrpt?GAO-03-314.



To view the full report, including the scope and methodology, 

click on the link above. For more information, contact James R. 

White at 202-512-9110 or whitej@gao.gov.



Contents:



Letter:



Results in Brief:



Scope and Methodology:



Background:



IRS Processed Individual Income Tax Returns and Refunds Smoothly in 

2002 Despite Challenges:



IRS Exceeded Its 2002 Electronic Filing Performance Goal and Is Taking 

Action to Help Ensure Future Growth:



Telephone Service Improved Compared with Last Year:



Quality of Walk-in Assistance Showed Some Improvement, While the Number 

of Taxpayers Assisted Continued to Decline:



Performance of IRS’s Web Site Was Generally Better Than Last Year, but 

Improvements Still Needed:



Conclusions:



Recommendations for Executive Action:



Agency Comments and Our Evaluation:



Appendix I: Data on IRS’s Processing Performance Relative

to Fiscal Year 2001 Performance and Fiscal Year

2002 Goals:



Appendix II: Challenges Affecting the 2002 Filing Season and IRS’s 

Response:



Advance Tax Refunds and the Rate Reduction Credit Led to Many Errors on 

Filed Returns:



Terrorist Actions Required Tax Relief for Affected Taxpayers and 

Security Enhancements:



IRS’s Reorganization Required a Shifting of Computer Files and Affected 

Processing Performance:



Appendix III: Changes to Third-Party Authorizations in 2002:



Authority to Designate a Third Party Expanded:



Oral Authorizations Offered:



Appendix IV: Results of Surveys of Taxpayers and Tax Practitioners Who 

Prepared Returns on a Computer but Filed on Paper:



Appendix V: IRS Received Fewer E-Mail Questions Via Its Web Site and 
Did 

a Worse Job Responding to the Questions:



Appendix VI: Comments from the Internal Revenue Service:



Appendix VII: GAO Contacts and Staff Acknowledgments:



GAO Contacts:



Acknowledgments:



Tables:



Table 1: Number of Individual Income Tax Returns Received, by Filing 

Type:



Table 2: IRS’s Performance in Providing Filing Season Telephone 

Assistance:



Table 3: IRS’s Processing Performance Relative to Fiscal Year 2001 

Performance and Fiscal Year 2002 Goals as of September 2002:



Table 4: Processed Returns with Rate Reduction Credit Errors as of 

September 27, 2002:



Table 5: Third-Party Authorizations:



Table 6: Factors That Influence Preparers of Returns on a Computer Not 

to File Electronically and Incentives That Could Influence Them to File 

Electronically:



Figures:



Figure 1: IRS’s Primary Filing Season Activities Are Processing Returns 

and Assisting Taxpayers:



Figure 2: Highlights of IRS’s Filing Season Preparations, Challenges, 

and Responses:



Figure 3: Number of Individual Income Tax Returns Filed Since 1995:



Figure 4: Of 104.1 Million Calls to IRS, 30.4 Million Were Covered by 

Accessibility and Accuracy Performance Measures:



Figure 5: IRS Walk-in Assistance, 1999-2002 Filing Seasons:



Abbreviations:



CSR: customer service representative:



ETAAC: Electronic Tax Administration Advisory Committee:



IRS: Internal Revenue Service:



NAP: National Account Profile:



ODC: Oral Disclosure Consent:



OTIA: Oral Tax Information Authorization:



PIN: personal identification number:



Q-Matic: Queuing Management System:



SSN: Social Security number:



TIGTA: Treasury Inspector General for Tax Administration:



W&I: Wage and Investment Operating Division:



United States General Accounting Office:



Washington, DC 20548:



December 20, 2002:



The Honorable Amo Houghton

Chairman, Subcommittee on Oversight

Committee on Ways and Means

House of Representatives:



Dear Mr. Houghton:



In response to your request, this report discusses various aspects of 

the Internal Revenue Service’s (IRS) performance during the 2002 tax 

filing season.[Footnote 1] It is during the filing season that most 

taxpayers have their only contacts with IRS, generally to either seek 

assistance or resolve problems, such as computational errors or missing 

Social Security numbers (SSN), on their returns. The normal challenges 

associated with any filing season were exacerbated in 2002 primarily by 

a new tax credit on the individual income tax forms and efforts IRS 

took in response to terrorist activities in 2001. Because of the large 

number of returns and the essential nature of IRS’s filing season 

activities, IRS’s filing season performance is important to taxpayers 

and the Congress.



This report discusses IRS’s performance in five key filing season 

activities: (1) processing individual income tax returns, refunds, and 

remittances; 

(2) increasing the extent to which individual income tax returns are 

filed electronically; (3) assisting taxpayers over the telephone; (4) 

providing face-to-face assistance at its walk-in locations; and (5) 

providing services via its Web site on the Internet. For each of those 

five areas, we discuss 

(1) IRS’s performance compared with its performance in 2001 and its 

performance goals for 2002, as well as any other evidence relating to 

IRS’s performance and (2) any factors or initiatives that substantively 

enhanced or impeded IRS’s performance and its ability to meet 

performance goals.



We testified before the Subcommittee on Oversight on the interim 

results of our assessment in April 2002.[Footnote 2] Our assessment was 

based on analyses of IRS data and data obtained from sources outside 

IRS, interviews of IRS officials and persons in the private sector, and 

observations of IRS operations.



Results in Brief:



Available data for each of the five filing season activities we 

reviewed indicated that IRS’s performance in 2002 was generally better 

than or comparable to its performance in 2001 and that IRS met many of 

its 

2002 performance goals. Overall, individual income tax returns and 

refunds were processed smoothly (i.e., without any significant 

disruption); the use of electronic filing continued to grow; and the 

quality of assistance provided over the telephone, at walk-in sites, 

and through the Web site improved.



Processing--Available evidence indicates that IRS was able to smoothly 

process about 131 million individual income tax returns and about 99 

million refunds in 2002. This assessment is based on IRS’s performance 

in 2002 compared with its performance in 2001 and its goals for 2002 as 

well as other evidence, such as testimony from the tax practitioner 

community. IRS’s smooth processing was accomplished despite several 

challenges that were unique to the 2002 filing season, including a new 

rate reduction credit that led to a significant number of errors that 

IRS had to correct. IRS attributed its ability to successfully meet 

these challenges, in part, to extensive preparation in advance of the 

filing season.



Electronic Filing--The number of individual income tax returns filed 

electronically grew from about 40.2 million in 2001 to about 46.9 

million in 2002, an increase of about 16.5 percent, and the percentage 

of individual tax returns filed electronically reached 35.9 percent. 

This 16.5-percent increase over the number of returns received 

electronically in 2001 was more than IRS’s goal of 15 percent and 

continued the upward trend in the number of returns filed 

electronically since 1995. IRS took some positive steps, including an 

increased focus on taxpayers and tax practitioners who prepare returns 

on computer but file on paper, that helped it achieve that increase and 

that could lead to further increases in the future.



Telephone Service--IRS’s performance measures showed that IRS provided 

more accessible and accurate telephone service in the 2002 filing 

season than in 2001 and met most of its telephone service performance 

goals. IRS achieved these results despite an increase in calls due to 

the rate reduction credit. Although it is difficult to determine 

specific cause/effect relationships, several initiatives that IRS 

implemented for the 2002 filing season, including the implementation of 

an accuracy improvement strategy and enhancements to the telephone call 

routing system, likely contributed to the improved service. IRS’s 

accessibility and accuracy measures reflect the experiences of those 

callers who attempted to reach an IRS assistor (30.4 million calls), 

but not the experiences of other callers who received automated service 

or either hung up or were disconnected without receiving service (73.7 

million calls).



Walk-in Assistance--Despite data limitations, there was enough evidence 

to suggest that the accuracy of tax law assistance provided by IRS’s 

walk-in sites improved in 2002 compared with 2001, although the level 

of accuracy (about 50 percent) was still low. Some of that improvement 

might be attributed to better training of field assistance employees. 

Although field assistance has made and anticipates making more workload 

and staffing changes at its walk-in sites, it has no comprehensive 

plans for doing so. When we discussed this with field assistance 

officials, they told us that they put the changes on hold until they 

are able to do more comprehensive planning.



Internet Web Site--Our assessment and external assessments indicated 

that IRS’s redesigned Web site, which became operational in January 

2002, was easier to access and more user friendly, although there are 

certain aspects of the site that need attention. For example, someone 

using the site’s search engine to find information on the earned income 

credit would receive a list of over 1,500 items and would have to 

scroll down several pages of items before finding reference to the 

primary publication on that credit. IRS is aware of this problem and is 

working to identify solutions. IRS data also indicated that the Web 

site was used more often in 2002, although IRS’s primary measure of 

usage--number of hits--overstates usage.



We are making recommendations to the Commissioner of Internal Revenue 

to improve certain performance measures and to ensure that changes to 

walk-in sites and staffing are thoroughly analyzed and planned. In 

commenting on a draft of this report (see app. VI), the Acting 

Commissioner of Internal Revenue cited actions that IRS has taken or 

plans to take that appear responsive to three of our four 

recommendations. The Acting Commissioner’s comments on our fourth 

recommendation did not speak to the specific issue addressed by the 

recommendation.



Scope and Methodology:



To assess IRS’s performance in the five key filing season activities 

covered by this report, we:



* reviewed and analyzed IRS documents and data, including workload and 

performance data;



* interviewed IRS officials about current operations, performance 

relative to 2001 performance and 2002 goals, and significant factors 

and initiatives that affected performance;



* observed operations at three of the eight processing centers operated 

by IRS’s Wage and Investment Operating Division (W&I) and four of IRS’s 

approximately 470 walk-in locations;[Footnote 3]



* observed a leadership conference and program review at one processing 

center;



* analyzed information posted to IRS’s Web site, specifically assessing 

the ease of finding information on the site (i.e., navigation) and the 

accuracy and currency of data on the site (i.e., content);



* reviewed information from private firms that assessed various aspects 

of IRS’s Web site;



* reviewed information from and interviewed representatives of various 

private organizations that prepare tax returns and participated in IRS 

and other conferences about their views on IRS’s 2002 operations and 

performance; and:



* reviewed related congressional testimony and work performed by the 

Treasury Inspector General for Tax Administration (TIGTA).



This report discusses measures that reflect the continuing interest of 

the Subcommittee--including the quality, accessibility, and timeliness 

of IRS’s performance during the filing season. We did not independently 

verify the data reported by IRS or, in most cases, assess the 

methodologies used for computing performance measures or the 

appropriateness of IRS’s goals. However, we recently issued a report 

that discussed whether the performance measures that IRS and we use to 

assess various aspects of IRS’s filing season performance had certain 

key desirable attributes, such as objectivity, reliability, and 

clarity.[Footnote 4] Although we reported that some measures had 

weaknesses in these and other areas and we made recommendations to 

address the weaknesses we identified, IRS and we still consider the 

measures useful in providing data to assess performance.



We did our work at IRS headquarters; W&I headquarters and Joint 

Operations Center in Atlanta;[Footnote 5] and W&I processing centers in 

Atlanta, Memphis, and Philadelphia and walk-in locations in Georgia. We 

selected these offices for a variety of reasons, including the 

proximity of our audit teams and the location of IRS managers. We 

performed our work from January through November 2002 in accordance 

with generally accepted government auditing standards.



Background:



IRS’s filing season activities encompass two critical areas--returns 

processing and taxpayer assistance. Figure 1 provides information on 

the number of returns processed and the extent of assistance provided.



Figure 1: IRS’s Primary Filing Season Activities Are Processing Returns 

and Assisting Taxpayers:



[See PDF for image]



Note: Workload data in this figure are for the following time periods: 

returns processing--1/1/02 to 10/25/02; toll-free telephone--1/1/02 to 

7/13/02; walk-in--1/1/02 to 4/20/02; and Internet downloads--1/1/02 to 

8/31/02.



[End of table]



The Economic Growth and Tax Relief Reconciliation Act of 2001[Footnote 

6] directed the Secretary of the Treasury to issue advance tax refunds 

to eligible taxpayers. Accordingly, about 86 million taxpayers received 

checks of up to $600 between July and December 2001. Taxpayers who did 

not receive an advance tax refund as part of that process or who 

received less than the maximum allowed by law may have been entitled to 

a rate reduction credit when filing their tax year 2001 returns in 

2002.[Footnote 7] Accordingly, IRS added a line to the individual 

income tax forms for eligible taxpayers to enter a credit amount and 

provided a worksheet for taxpayers to use in determining if they were 

eligible.



In accordance with the IRS Restructuring and Reform Act of 

1998[Footnote 8] and the Government Performance and Results 

Act,[Footnote 9] IRS established a balanced performance measurement 

system to emphasize accountability for achieving specific results and 

to reflect its mission and its three strategic goals--top quality 

service to all taxpayers through fair and uniform application of the 

law, top quality service to each taxpayer in every interaction, and 

productivity through a quality work environment. IRS has defined three 

elements of a balanced performance measurement system--customer 

satisfaction, employee satisfaction, and business results (quality and 

quantity measures)--to ensure balance. IRS intends to use balanced 

measures to hold managers and frontline staff more accountable for 

improving filing season performance. Although IRS had measures of 

performance prior to its establishment of balanced measures beginning 

in fiscal year 2001, IRS managers have spent much effort revising 

measures since that time.



IRS Processed Individual Income Tax Returns and Refunds Smoothly in 

2002 Despite Challenges:



Qualitative and quantitative information indicates that IRS processed 

about 131 million individual income tax returns and about 99 million 

refunds smoothly (i.e., without any significant disruption, such as a 

computer breakdown that could cause major delays in issuing 

refunds).[Footnote 10] Part of that information resulted from our 

comparison of IRS’s processing performance in 2002 with its performance 

in 2001 and its goals for 2002.



The smooth processing in 2002 was accomplished despite:



* the new rate reduction credit, which led to a significant number of 

errors by taxpayers and tax return practitioners;



* the terrorist attacks of September 11th, which led to IRS reissuing 

notices as part of its effort to provide relief to affected taxpayers, 

and the subsequent anthrax attacks perpetrated through the mail, which 

led to significant security enhancements; and:



* IRS’s reorganization, which caused IRS to shift computer files to 

match its new organizational units.



IRS attributed its success in meeting those challenges, in part, to 

extensive preparation in advance of the filing season. In addition, IRS 

continued initiatives begun last year to improve processing 

performance.



IRS’s Processing Went Smoothly:



Regarding IRS’s processing in 2002,



* IRS’s performance in 2002 improved relative to 2001 for five measures 

(deposit error rate, deposit timeliness-paper, refund timeliness-

individual (paper), refund error rate-individual (paper), and refund 

interest paid). As discussed later, we could not compare IRS’s 

performance for three other measures (letter error rate, notice error 

rate, and productivity). Table 3 in appendix I describes these eight 

measures and shows the results IRS reported for each of the measures as 

of September 2002.



* IRS met its fiscal year 2002 performance goal for four measures 

(deposit timeliness-paper, refund timeliness-individual (paper), 

refund error rate-individual (paper) and refund interest paid) and 

missed its goals for two measures (deposit error rate and 

productivity). As discussed later, we could not compare IRS’s 

performance against goals for the other two measures (letter error rate 

and notice error rate).



* Production data that IRS uses to monitor operations and identify and 

resolve issues that could disrupt operations showed IRS met processing 

deadlines and did not experience significant disruptions.



* In testimony given April 2002 before the Subcommittee, the 

Commissioner of Internal Revenue testified that, among other things, 

the filing season was smooth, with returns being processed on time.



* At that same hearing, (1) H&R Block, the largest tax return 

preparation firm, testified that the filing season had gone smoothly 

with some exceptions, such as the rate reduction credit, and commented 

on how well IRS and the tax preparation industry were working together 

and (2) a representative of the National Association of Enrolled 

Agents[Footnote 11] characterized the filing season, on the whole, as 

smooth.



* Representatives of other tax return preparation firms told us that, 

from their perspective, processing went relatively smoothly.



* Directors, managers, and staff at processing centers we visited in 

Memphis, Atlanta, and Philadelphia voiced similarly positive views 

about the filing season and processing based, in part, on their 

monitoring of production and inventory levels and their ability to meet 

processing deadlines despite the additional error-correction workload 

resulting from the rate reduction credit.



* W&I’s Director of Submission Processing characterized processing 

activities as smooth because of the lack of major disruptions and IRS’s 

ability to handle the extensive error-correction workload and still 

meet projected production rates and processing timeframes. He and other 

IRS officials attributed IRS’s success to several factors, including 

extensive planning and close monitoring of tax return receipts and 

inventory levels at the eight W&I processing centers through such means 

as weekly production meetings and a well-qualified workforce.



* According to TIGTA’s report on the 2002 filing season, IRS had a 

successful filing season despite implementing numerous tax law changes 

related to the Economic Growth and Tax Relief Reconciliation Act of 

2001, having to correct millions of errors related to the rate 

reduction credit, and experiencing problems with its error-correction 

system.[Footnote 12] TIGTA reported that IRS was able to complete the 

additional workload without any material slowdowns in processing and 

issue refunds in a timely manner by working extra hours when 

necessary.[Footnote 13]



Revisions to Two Measures Precluded a Complete Quantitative Comparison:



Because IRS revised its letter error rate and notice error rate 

performance measures since last filing season, we could not compare its 

performance in 2002 with its performance in 2001 or against its goals 

for 2002.[Footnote 14] In computing these error rates, IRS eliminated 

certain errors, such as capitalization mistakes, that it referred to as 

“professionalism errors.” As a result, IRS established new baselines 

for these two measures in 2002.[Footnote 15] According to IRS 

officials, the decision to eliminate professionalism errors in 

computing these measures was based on the fact that although those 

errors may make documents, such as letters and notices, less 

professional in appearance, they do not make the documents inaccurate.



According to IRS officials, changes to measures used to assess 

processing activities were necessary and not unexpected because the 

measures are relatively new and still under development and because 

2002 processing goals were based on limited 2001 performance data. 

Furthermore, although comparisons cannot be made between 2001 and 2002, 

IRS officials believe that the measures provide important information 

to assess current year performance.



IRS Faced Several Challenges in Processing Returns in the 2002 Filing 

Season and Took Steps to Meet Those Challenges:



IRS’s ability to process returns and refunds in 2002 without any major 

disruptions was enhanced by extensive early preparation and 

coordination. That preparation and coordination helped IRS meet the 

following significant challenges that were unique to the 2002 filing 

season:



* Passage of the Economic Growth and Tax Relief Reconciliation Act of 

2001, specifically the provisions relating to advance tax refunds and 

rate reduction credits.



* The September 11, 2001, attacks, which led to actions by IRS to 

provide tax relief to affected taxpayers, and the subsequent anthrax 

attacks perpetrated through the mail, which led to significant security 

enhancements.



* IRS’s shifting of computer files in conjunction with its 

reorganization.



Details about these challenges can be found in appendix II.



Preparation and Coordination Helped IRS Meet Challenges:



Preparation and coordination helped ensure smooth processing during the 

2002 filing season. IRS begins extensive preparation well in advance of 

each filing season to help ensure smooth processing. Figure 2 reflects 

some of IRS’s key activities and associated deadlines in preparation 

for each filing season. It also reflects the challenges that IRS faced 

in planning and delivering the 2002 filing season and IRS’s responses 

to those challenges.



Figure 2: Highlights of IRS’s Filing Season Preparations, Challenges, 

and Responses:



[See PDF for image]



[A] The deadline for filing tax year 2000 individual income tax returns 

was April 16, 2001, because April 15 was a Sunday.



[See PDF for image]



IRS began preparing for the 2002 filing season by translating tax law 

requirements into computer system changes and testing those system 

changes in early 2001. As IRS continued preparing for the filing 

season, it had to make changes to the computer systems to accommodate 

the tax law changes related to the rate reduction credit and test those 

changes concurrently with regular testing and September 11 work. In 

March 

2002, TIGTA concluded that IRS appropriately prepared for the effective 

processing of tax returns by modifying its computer systems, updating 

procedural and training manuals, and performing staffing 

analyses.[Footnote 16] In responding to that report, IRS officials 

indicated that preparatory efforts were successful and were 

instrumental in ensuring smooth processing.



IRS also established a new office--the Release and Filing Season 

Readiness Office--in October 2001 to support the coordination of 

computer system development, testing, and operation in the actual 

processing environment. Among other things, this office helped IRS 

respond to changing priorities and unanticipated challenges by 

monitoring the status of computer system changes, facilitating 

discussions between information technology and other processing units 

on how to mitigate problems, and helping support and coordinate testing 

and resources.



IRS Implemented or Expanded Initiatives to Improve Processing 

Performance:



IRS implemented or expanded several initiatives to improve processing 

performance for both the 2002 and 2003 filing seasons. For example:



* IRS established site-level performance goals in 2002 to allow better, 

more individualized comparison of each processing center’s performance 

against its previous performance and toward corporate goals. According 

to IRS officials, these site-level comparisons should help identify 

best practices.



* In October 2001, IRS conducted a 2-day “leadership conference” at all 

eight W&I processing centers to communicate site goals and train 

managers and employees on the importance of performance measures and 

individual employees’ role in helping achieve goals. IRS officials 

credit the conferences, in part, with helping improve performance in 

2002.



* Given the perceived success of last year’s leadership conferences, 

IRS conducted similar conferences in 2002. At the same time, IRS 

officials assessed whether the processing centers and employees were 

following procedures when sampling returns for review and reporting 

performance data for balanced measures. As a result, IRS identified 

issues that can be corrected by updating processing guidance, which 

will likely improve performance. In addition, IRS identified local 

procedures to help reduce the interest paid on refunds that officials 

said are being communicated to the other centers as a best practice.



* During the 2002 filing season, IRS expanded the third-party checkbox 

authorization, to allow taxpayers to designate friends, family members, 

or paid practitioners to help resolve problems IRS might encounter in 

processing their returns, such as a missing SSN, that could result in 

incorrectly computing a refund, for example. See appendix III for more 

information on the third-party checkbox authority.



IRS Exceeded Its 2002 Electronic Filing Performance Goal and Is Taking 

Action to Help Ensure Future Growth:



IRS received about 47 million individual income tax returns 

electronically, or about 36 percent of all individual income tax 

returns filed, as of October 25, 2002. The growth rate in 2002 

projected through 2007 would bring IRS very close to its long-term goal 

of 80 percent of all individual income tax returns filed electronically 

by 2007.[Footnote 17] However, the Electronic Tax Administration 

Advisory Committee (ETAAC)[Footnote 18] does not believe that the 

current growth rate can be sustained and, instead, expects the growth 

rate to decline. According to ETAAC, if IRS is to continue or increase 

its current growth rate, it needs to overcome various barriers to 

electronic filing. IRS took several steps that were aimed at overcoming 

existing barriers and that may have helped it exceed its electronic 

filing goal for 2002. Those steps could lead to more growth in the 

future. In addition, two initiatives planned for 2003 could further 

encourage many taxpayers and practitioners to convert from paper filing 

to electronic filing.



Electronic Filing Continued to Grow:



As shown in table 1, the number of individual tax returns filed 

electronically grew from about 40.2 million in 2001 to about 46.9 

million in 2002--an increase of about 16.5 percent--and the percentage 

of individual income tax returns filed electronically reached 35.9 

percent. This 

16.5-percent increase over the number of returns received 

electronically in 2001 was more than IRS’s goal of 15 percent and 

continued the upward trend in the number of returns filed 

electronically since 1995.



Table 1: Number of Individual Income Tax Returns Received, by Filing 

Type:



Number of returns in thousands.



Paper; 1/1/00 to 10/27/00: 92,322; 1/1/01 to 10/26/01: 89,834; 
Percentage

change 2000 to 2001: -2.69; 1/1/02 to 10/25/02: 83,816; Percentage

change 2001 to 2002: -6.70.



Electronic; 1/1/00 to

10/27/00: [Empty]; 1/1/01 to

10/26/01: [Empty]; Percentage

change 2000

to 2001: [Empty]; 1/1/02 to

10/25/02: [Empty]; Percentage

change 2001

to 2002: [Empty].



Traditional[A]; 1/1/00 to

10/27/00: 25,211; 1/1/01 to

10/26/01: 28,988; Percentage

change 2000

to 2001: 14.98; 1/1/02 to

10/25/02: 33,286; Percentage

change 2001

to 2002: 14.83.



On-line[B]; 1/1/00 to

10/27/00: 5,022; 1/1/01 to

10/26/01: 6,838; Percentage

change 2000

to 2001: 36.16; 1/1/02 to

10/25/02: 9,428; Percentage

change 2001

to 2002: 37.88.



TeleFile[C]; 1/1/00 to

10/27/00: 5,161; 1/1/01 to

10/26/01: 4,419; Percentage

change 2000

to 2001: -14.38; 1/1/02 to

10/25/02: 4,176; Percentage

change 2001

to 2002: -5.50.



Subtotal; 1/1/00 to

10/27/00: 35,394; 1/1/01 to

10/26/01: 40,245; Percentage

change 2000

to 2001: 13.71; 1/1/02 to

10/25/02: 46,891; Percentage

change 2001

to 2002: 16.51.



Total; 1/1/00 to

10/27/00: 127,716; 1/1/01 to

10/26/01: 130,079; Percentage

change 2000

to 2001: 1.85; 1/1/02 to

10/25/02: 130,708; Percentage

change 2001

to 2002: 0.48.



Percentage filed electronically; 1/1/00 

to

10/27/00: 27.7; 1/1/01 to

10/26/01: 30.9; Percentage

change 2000

to 2001: [Empty]; 1/1/02 to

10/25/02: 35.9; Percentage

change 2001

to 2002: [Empty].



Source: IRS data.



Note: Totals may not add due to rounding.



[A] Traditional electronic filing involves the transmission of returns 

over communication lines through a third party, such as a tax return 

practitioner, to an IRS processing center.



[B] On-line returns are prepared and transmitted by taxpayers through 

on-line intermediaries using personal computers and commercial 

software.



[C] Under TeleFile, certain taxpayers who are eligible to file a Form 

1040EZ are allowed to file using a toll-free number on touch-tone 

telephones.



[End of table]



Figure 3 shows that the growth in electronic filing in 2002 continued 

the upward trend started in 1995 and the rate of growth in 2002 was 

larger than the rate of growth in 2001.



Figure 3: Number of Individual Income Tax Returns Filed Since 1995:



[See PDF for image]



[End of figure]



Assuming continued annual growth rates of 16.51 percent for individual 

returns filed electronically and 0.48 percent for the total number of 

individual tax returns filed, about 75.2 percent of all tax returns 

will be filed electronically by 2007, slightly below the long-term goal 

of 80 percent. However, in its 2002 report to the Congress, ETAAC said 

that although the electronic filing goal for 2002 was exceeded, the 

trend of electronic filing is clearly towards lower annual growth 

rates, primarily for the following reasons:



* The taxpayers most easily attracted to electronic filing have already 

made the switch to electronic filing while the remaining taxpayers will 

be more difficult to convert.



* As the base of total filers and electronic filers grows, the number 

of new electronic filers needed to sustain the same growth rate 

increases.



Taxpayers and Tax Practitioners Report Being Satisfied with Electronic 

Filing:



Each year, IRS surveys taxpayers and tax practitioners who 

electronically file to determine their satisfaction level with 

electronic filing. The survey for 2002 showed that about 98 percent of 

the responding taxpayers and 

85 percent of the responding tax practitioners were either satisfied or 

very satisfied with electronic filing, which compares favorably with 

IRS’s goal of 85 percent. Although the percentage of responding 

practitioners who expressed overall satisfaction was relatively high 

(85 percent), it was lower than in 2001 (90 percent), which, in turn, 

was lower than in 

2000 (93 percent). It is not clear why the level of satisfaction among 

practitioners has been decreasing because the customer satisfaction 

survey does not probe for reasons behind a respondent’s choices.



IRS Initiatives May Have Helped It Exceed Its 2002 Electronic Filing 

Performance Goal:



Over the years, IRS has identified many impediments to electronic 

filing, including privacy and security concerns, cost, a lack of 

awareness of or interest in electronic filing, an inability to file all 

forms and schedules electronically, and a need to send certain paper 

documents to IRS even if the return was filed electronically. In 2002, 

IRS took numerous steps to alleviate these impediments and thus 

encourage more electronic filing that included the following:



* Increasing the electronic filing marketing budget from $9 million in 

2001 to about $15 million in 2002 and focusing the marketing campaign 

on taxpayers and practitioners who filed computer-prepared returns on 

paper.



* Mailing a postcard to about 23 million taxpayers, including about 

8 million taxpayers who had prepared their returns on a computer but 

filed on paper, informing them of the benefits of electronic filing and 

the self-select personal identification number (PIN) program. The 

postcard also informed them that a professional tax practitioner could 

be used to file their returns electronically and explained how the 

self-select PIN can be used to sign returns filed through a tax 

practitioner.



* Mailing two different letters to tax practitioners. The first letter 

was sent to practitioners who were already participating in the 

electronic filing program to thank them for participating and encourage 

them to continue to support the program. The second letter was sent to 

practitioners who were not participating in the electronic filing 

program and told them about the benefits of electronic filing, changes 

that had been made to the program, and pending changes. IRS enclosed a 

document with both letters that provided more information on the 

benefits of electronic filing and the self-select PIN program. In 

total, about 250,000 letters were sent to practitioners.



* Reinstating the practitioner PIN program. In 2001, practitioners 

informed IRS that they liked the practitioner PIN program, which IRS 

had terminated in favor of the self-select PIN program, and would like 

to see it reinstated. As of October 25, 2002, 83.9 percent of the tax 

returns prepared by practitioners and signed by a PIN were signed 

through the practitioner PIN program. The other 16.1 percent were 

signed using a self-select PIN.



* Changing the self-select PIN program at the request of tax 

practitioners. In 2002, taxpayers entered only their adjusted gross 

income from the prior tax year to verify their identity. In 

2001, taxpayers had to enter not only their adjusted gross income but 

also their total tax to verify their identity. Tax practitioners wanted 

the number of items used to verify identity--known as “shared secrets”-

-reduced from two to one, because errors related to shared secrets 

resulted in numerous electronic tax returns being rejected. Consistent 

with the reduction in the number of shared secrets from two to one, the 

number of reject conditions on electronic tax returns as a result of 

shared secrets went down from about 2.1 million in 2001 to about 

1.3 million in 2002.



* Making 30 additional forms and schedules eligible to be filed 

electronically, thereby enabling, according to IRS, 99 percent of all 

individual forms and schedules to be filed electronically.



Although no information is available to confirm the extent to which 

these actions, either individually or collectively, contributed to the 

growth of electronic filing, it seems reasonable to assume that they 

had a positive effect since they address most of the identified 

impediments noted earlier.



IRS also took steps to obtain additional information on impediments to 

electronic filing. As in the past, for example, it held electronic 

filing forums and practitioner meetings and conducted attitudinal 

surveys and market research. One significant step was the surveying of 

taxpayers and tax practitioners who prepared returns on a computer but 

filed them on paper. We had recommended such surveys in our report on 

the 2001 filing season in response to the fact that about 40 million 

computer-prepared individual income tax returns were filed on paper in 

2001.[Footnote 19]



The surveys revealed that taxpayer and tax practitioner costs, security 

and technology concerns, and a perceived lack of interest were among 

the primary reasons why taxpayers and tax practitioners are not filing 

electronically. With respect to the perceived lack of interest, (1) 87 

percent of taxpayers who used paid practitioners indicated on the 

survey that their practitioner never discussed electronic filing with 

them and 4 percent said that their practitioner recommended against 

electronic filing and 

(2) 66 percent of tax practitioners said that they did not offer 

electronic filing because their clients did not ask for it. See 

appendix IV for more details on what survey respondents cited as 

factors that strongly influenced them not to file electronically and 

incentives that could encourage them to do so.



Planned Initiatives for 2003 Could Lead to More Electronic Filing:



IRS has two initiatives planned for the 2003 filing season--making 

electronic filing free for millions of taxpayers and offering services 

to practitioners who file a certain number of returns electronically. 

IRS intends for these initiatives to address two major impediments--the 

cost to taxpayers and the lack of interest on the part of some 

practitioners.



In an attempt to remove cost as an impediment to electronic 

filing,[Footnote 20] IRS entered into an agreement with a consortium of 

companies in the electronic tax preparation and filing industry that 

will offer free on-line tax filing services via IRS’s Web site. IRS 

estimates that 78 million taxpayers will be eligible to file their tax 

returns for free through this consortium. However, in its 2002 report 

to the Congress, ETAAC stated that lack of access to a computer and/or 

the Internet may prevent many of the 

78 million eligible taxpayers from taking advantage of the free-filing 

service.



To address the lack of interest on the part of some practitioners, IRS 

plans to offer electronic services (such as taxpayer identification 

number matching and account resolution) to tax practitioners who file a 

certain number of returns electronically. The aim of this initiative is 

to provide practitioners who file electronically with valuable tools to 

improve their service to their customers and reduce costs. In its 2002 

report to the Congress, ETAAC stated that electronic services are 

likely to provide a major incentive for practitioners to file their 

clients’ returns electronically.



Telephone Service Improved Compared with Last Year:



IRS’s performance measures showed that (1) telephone service was more 

accessible and accurate during the 2002 filing season than it was in 

2001and (2) IRS met most of its 2002 performance goals. IRS implemented 

several initiatives and strategies to improve performance, particularly 

in the area of accuracy where improvement was greatest.



IRS serves telephone callers either through assistors, who are also 

known as customer service representatives (CSR), or automation, 

depending on how callers respond to menu prompts. As shown in figure 4, 

of the about 104.1 million calls that IRS received in the 2002 filing 

season,



* 21.2 million of the 30.4 million calls IRS estimates were from 

callers attempting to reach an assistor[Footnote 21] actually reached 

one;



* 42.9 million calls resulted in the caller receiving automated 

service; and:



* 40 million calls resulted in the caller hanging up or being 

disconnected without receiving service, including callers who received 

a busy signal.



Figure 4: Of 104.1 Million Calls to IRS, 30.4 Million Were Covered by 

Accessibility and Accuracy Performance Measures:



Note: Data are for calls made to IRS’s three main live-and-automated 

assistance numbers; its special number for victims of the September 11, 

2001, terrorist actions; and its automated-service-only (TeleTax) 

number from January 1, 2002, through July 13, 2002.



The measures that IRS and we used to assess the accessibility and 

accuracy of IRS telephone assistance during the 2002 filing season, 

which are discussed in the next section, reflect only the calls that 

IRS estimates were from callers attempting to reach an 

assistor.[Footnote 22] Specifically, the CSR level of service measure 

reflects the experience of the 30.4 million callers who IRS estimates 

were attempting to reach an assistor, and other measures reflect the 

experiences of the 21.2 million callers who actually reached an 

assistor. The measures do not reflect the experiences of the 73.7 

million calls from callers who either completed an automated service or 

hung up or were disconnected without receiving service.[Footnote 23] As 

a result, for example, the measures do not reflect an increased number 

of calls in 2002 from callers who hung up without completing menu 

responses or an automated service. According to IRS officials, the 

biggest part of this increase was probably due to problems with IRS’s 

telephone service menu, particularly during the early weeks of the 

filing season. When IRS revised the menu in mid-February, it noted a 

decline in the hang-up rate, which may indicate that taxpayers had been 

frustrated or confused by the menu.



Telephone Service Was More Accessible and Accurate Than Last Year, and 

Most 2002 Goals Were Met:



Table 2 provides information on IRS’s performance in providing 

telephone assistance to taxpayers who were attempting to reach or who 

received service from an IRS assistor in the 2002 filing 

season.[Footnote 24] In summary, the table shows the following:



* A greater percentage of calls got through and received service, and 

IRS met two of its three accessibility goals for 2002. IRS likely would 

have performed better in the other measure, CSR level of service, if 

not for the significant demand for telephone assistance related to the 

rate reduction credit.



* The accuracy of assistance improved considerably in 2002 compared 

with 2001, and IRS met most of its goals. The quality and correct 

response rates for both tax law and account calls were better in 

2002, indicating that assistors more closely adhered to IRS procedural 

guidance and more often provided callers accurate information and 

service.



* Customer satisfaction was comparable to 2001. Detailed data show that 

customers who were successful in reaching an assistor were least 

satisfied with the time they had to wait to speak with an assistor in 

both 2001 and 2002.



Table 2: IRS’s Performance in Providing Filing Season Telephone 

Assistance:



Accessibility measures[A]: CSR level of service[B]; 2001: Actual: 68%; 

2002: Actual: 71%; Goal: 72%.



Accessibility measures[A]: Assistor response level[C]; 2001: Actual: 

39%; 2002: Actual: 51%; Goal: 44%.



Accessibility measures[A]: Average speed of answer[D]; 2001: Actual: 

337 seconds; 2002: Actual: 268 seconds; Goal: 289 seconds.



Accessibility measures[A]: Accuracy measures[E]; 2001: Actual: 

[Empty]; 2002: Actual: [Empty]; Goal: [Empty].



Accessibility measures[A]: Tax law quality rate[F]; 2001: Actual: 75% 

+/-1%; 2002: Actual: 82% +/-1%; Goal: 78%.



Accessibility measures[A]: Accounts quality rate[F]; 2001: Actual: 69% 

+/-1%; 2002: Actual: 76% +/-1%; Goal: 72%.



Accessibility measures[A]: Tax law correct response rate[G]; 2001: 

Actual: 79% +/-1%; 2002: Actual: 85% +/-1%; Goal: 85%.



Accessibility measures[A]: Accounts correct response rate[G]; 2001: 

Actual: 88% +/-1%; 2002: Actual: 91% +/-1%; Goal: 93%.



Accessibility measures[A]: Customer satisfaction[H]; 2001: Actual: 

[Empty]; 2002: Actual: [Empty]; Goal: [Empty].



Accessibility measures[A]: Wage and Investment; January through March; 

April through June; 2001: Actual: ; 3.45 +/-.03; 3.46 +/-.03; 2002: 

Actual: ; 3.51 +/-.03; 3.43 +/-.03; Goal: ; 3.49.



Accessibility measures[A]: Small Business/Self-Employed; January 

through March; April through June; 2001: Actual: ; 3.49 +/-.03; 3.43 +/

-.03; 2002: Actual: ; 3.52 +/-.03; 3.47 +/-.03; Goal: ; 3.58.



Source: IRS data.



[A] Accessibility data are based on actual counts from January 1 

through July 13. The actual data shown here for the 2001 filing season 

differ from data we cited in an earlier report. See U.S. General 

Accounting Office, IRS Telephone Assistance: Limited Progress and 

Missed Opportunities to Analyze Performance in the 2001 Filing Season, 

GAO-02-212 (Washington, D.C.: Dec. 7, 2001). According to IRS, the 

difference in CSR level of service is due to an error in data it 

provided and a change in the way it calculates the measure, and 

differences in the other accessibility measures are due to enhancements 

to the telephone database that resulted in more accurate data.



[B] The percentage of callers that IRS estimates attempted to reach an 

assistor that received service, including some callers that received 

automated service.



[C] The percentage of callers that waited 30 seconds or less before 

speaking to an assistor.



[D] The average number of seconds callers waited before speaking to an 

assistor.



[E] Accuracy measures are based on representative samples and are 

estimated at the 90-percent confidence level from January through June.



[F] The percentage of calls in which assistors followed all IRS 

procedures for the call type and provided correct answers.



[G] The percentage of calls in which assistors provided correct answers 

for the call type, discounting procedural errors.



[H] The overall satisfaction rating on a 4-point scale based on a 

survey of callers who reached an assistor, estimated at the 95-percent 

confidence level. IRS compiles and reports these data quarterly by the 

two operating divisions that provide toll-free telephone assistance.



[End of table]



CSR Level of Service Improved Despite Increased Volume Due to the Rate 

Reduction Credit:



The volume of calls is a key factor that can affect CSR level of 

service. As we previously reported, IRS data suggest that demand for 

telephone assistance related to the rate reduction credit was 

significant during the 2002 filing season and negatively affected 

telephone performance, especially in mid-to late-February, when the 

greatest number of taxpayers called with questions about the 

credit.[Footnote 25] Although performance declined significantly 

during this period of peak demand for information on the rate reduction 

credit, IRS performed well enough during the rest of the 

2002 filing season so that, as shown in table 2, the CSR level of 

service in 2002 improved compared with 2001, and IRS almost met its 

goal. According to IRS officials, although taxpayers had greater 

difficulty getting access to IRS’s telephone system during certain 

weeks--when there was great demand for rate reduction credit assistance 

and, as discussed earlier, there were problems with IRS’s telephone 

service menu--taxpayers, overall, had a considerably better telephone 

service experience in the 2002 filing season than in 2001.



IRS Made Changes to Its Measures and Plans to Make More:



Effective October 2001, IRS stopped using the assistor response level 

and average speed of answer measures to assess its performance in 

providing telephone assistance. According to IRS officials, IRS adopted 

these measures in fiscal year 2001 to focus its efforts on improving 

access and allow it to gauge its performance against world-class 

telephone service organizations. However, officials said that 

subsequent experience showed that IRS’s telephone call sites do not 

have control over, nor are they directly accountable for, key factors 

that affect these measures, such as the volume of calls routed to them.



Although the sites’ control over these measures may be limited, they 

are an important component of any assessment of IRS’s telephone 

performance because they measure a key aspect of the customer 

experience--how long callers wait to speak to an assistor. In that 

regard, selected noteworthy public and private call centers we studied 

often used caller wait-time measures to set service-level goals and 

benchmarked their performance against the best practices of leading 

telephone service providers.[Footnote 26] Without a caller wait-time 

measure similar to those used by others, IRS’s suite of telephone 

measures is missing an important indicator of the ease with which 

taxpayers reach IRS.



According to IRS officials, the current accuracy measures, while useful 

in assessing the filing season, do not adequately reflect customer 

needs and employee performance. With the assistance of a contractor, 

IRS developed new measures that it believes better reflect what matters 

most to its customers--accuracy, timeliness, and professionalism. IRS 

plans to baseline the new measures in 2003 and use them in 2004 to 

replace the current accuracy measures.



IRS Implemented Various Initiatives to Improve Telephone Service:



IRS implemented several initiatives to improve telephone service in the 

2002 filing season. For example, IRS established accessibility 

performance measures and goals for its call sites and enhanced its 

call-routing capabilities to provide taxpayers better access to 

assistors. IRS also implemented several initiatives, including earlier 

assistor training, as part of a strategy to improve accuracy and 

piloted a new hiring program to get candidates with higher aptitudes 

and skill levels.



IRS Established Accessibility Performance Measures and Goals for Call 

Sites:



In 2002, IRS established two new accessibility performance measures--

services provided and total handle time[Footnote 27]--and related goals 

for its 

26 telephone call sites. These measures replace those discontinued in 

1998. IRS officials said that the measures and goals were intended to 

provide the sites with incentives for handling calls efficiently and 

answering their share of total calls, thereby enhancing their 

contributions to meeting IRS’s overall accessibility goals. According 

to IRS officials, the new measures led to improved performance by 

giving the call sites a clearer understanding of what they were 

expected to achieve and how their performance would help IRS achieve 

its goals. IRS took action to get employees to understand the measures 

and contribute to achieving goals, by, for example, holding “leadership 

conferences” for telephone operations managers similar to those 

conducted in the processing centers.



IRS Enhanced Call Routing:



IRS enhanced its call routing system by implementing a feature, called 

“network call screening.” Before network call screening, IRS’s contract 

long-distance service provider routed calls to one of IRS’s call sites 

where the caller heard and responded to a menu of options. The number 

of calls IRS received often exceeded its telephone system capacity, 

resulting in many calls receiving busy signals. Also, because calls 

were routed before the callers had indicated, through use of a menu, 

the subject matter of their calls, callers were sometimes given access 

to only automated service, even though assistors were available to 

handle their calls. With network call screening, callers hear and 

respond to the menu options before the call is routed to a call site. 

According to IRS officials, this reduced the number of calls that 

receive busy signals, increased the chance that a caller would reach an 

assistor, and decreased wait times. In that regard, (1) IRS data show 

that the number of busy signals callers received decreased from about 

1.5 million in the 2001 filing season (or about 

3 percent of total call attempts) to about 0.7 million in the 2002 

filing season (or about 1 percent of total call attempts) and (2) an 

IRS analysis showed that the percentage of calls sent to automated-only 

service decreased from 11 percent in the 2001 filing season to 8 

percent in the 2002 filing season.



IRS Implemented an Accuracy Improvement Strategy:



According to IRS officials, improving the accuracy of telephone 

assistance was a major emphasis in planning for the 2002 filing season. 

With the assistance of a contactor, IRS analyzed its accuracy data to 

identify problems and developed initiatives to improve accuracy. IRS’s 

accuracy rates in January were historically the lowest of any month for 

tax law inquiries and, according to the analysis, a classic learning 

curve was evident from January through April. IRS implemented several 

initiatives to combat this January slump, including training assistors 

in their specialized topics in November and December and requiring 

their managers to certify them, by the start of the filing season, as 

capable of providing the correct response to taxpayer questions. Other 

initiatives included reviewing and correcting the guidance assistors 

use to respond to taxpayer questions and assigning call site managers 

ownership of selected tax law topics, thus making them accountable for 

assistor training and improving assistors’ performance in their 

assigned topics.



IRS Tested a New Process

for Hiring Better Qualified Assistors:



For the 2002 filing season, IRS piloted a new hiring method at some 

call sites to improve service to taxpayers by hiring better-qualified 

assistors. The new method used live role-playing to identify candidates 

with the required competencies for successful job performance, free up 

front-line manager’s time previously spent interviewing possible 

candidates, and reduce assistor turnover. As we previously reported, 

IRS officials have said that attrition has negatively affected IRS’s 

performance,[Footnote 28] and the lack of time to interview applicants 

has led to concerns about the suitability of new hires.[Footnote 29] 

With respect to attrition, IRS data showed that the turnover rate at 

sites that hired under the new process was 9.3 percentage points lower 

than the turnover rate at comparable call sites that used the 

traditional hiring method. IRS plans to expand this initiative in 2003.



Quality of Walk-in Assistance Showed Some Improvement, While the Number 

of Taxpayers Assisted Continued to Decline:



Quality of walk-in assistance improved and wait-time was about as good 

as last year, based on available information, although fewer taxpayers 

were assisted. The accuracy of walk-in tax law assistance--the only 

type of walk-in assistance[Footnote 30] for which quality was measured 

in recent filing seasons--appears to have improved compared with the 

2001 filing season. IRS took several steps in 2002 to improve walk-in 

service--some may have helped improve the accuracy of tax law 

assistance in 2002, the effect of others should be realized later. IRS 

also anticipates changes in the workload to be handled by walk-in 

staff. Thus, although the number of taxpayers assisted at the walk-in 

sites declined about 20 percent over the past 3 filing seasons, IRS has 

been increasing the number of field assistance staff years used to 

provide that assistance. However, the extent and timing of the workload 

changes are not clear, and field assistance does not have comprehensive 

plans that clearly relate that workload to resource needs.



Accuracy of Tax Law Assistance by Walk-in Sites Improved Since Last 

Year but Remains Low:



The Congress directed TIGTA to review the accuracy of assistance 

provided by all of IRS’s more than 400 walk-in sites. TIGTA expects to 

complete its review in fiscal year 2003. From January through April 

2002, TIGTA reviewers made 157 anonymous visits to 77 walk-in sites, 

asking two tax law questions on each visit. TIGTA found that only about 

50 percent of its questions were answered correctly.[Footnote 31]



The reviews TIGTA did in 2002 were not comparable to the reviews it did 

in 2001 because TIGTA asked questions from a broader range of tax law 

topics in 2002 (i.e., 22 topics in 2002 compared with 4 topics in 

2001). It seems reasonable to assume, and TIGTA agrees, that covering 

22 topics versus 4 required a broader knowledge of the tax law and thus 

increased the possibility for error. Yet, despite the increased 

possibility for error, reported accuracy was about double the 24-

percent accuracy TIGTA reported for 2001.



In commenting on a draft of this report, the Acting Commissioner of 

Internal Revenue said that our use of TIGTA’s accuracy rates is 

inappropriate because TIGTA included referrals to IRS publications as 

errors. While noting that IRS does not want employees to improperly 

refer questions, the Acting Commissioner said that those referrals 

should not be counted as incorrect answers. According to the Acting 

Commissioner, the “true cumulative accuracy rate” through June 2002 was 

66 percent instead of 50 percent. We disagree. As TIGTA explained in 

its report, referrals were counted as incorrect when the IRS employee 

merely provided the publication, without walking the customer through 

it to identify the answer, as required by established field assistance 

procedures. In our view, the complexity of tax laws and varying 

education levels among taxpayers seeking assistance suggest that 

requiring field assistance employees to walk the taxpayer through a 

publication to identify the correct response is a necessary procedure 

that should be followed in practice for the response to be considered 

correct.



Most Taxpayers Waited 30 Minutes or Less for Walk-in Assistance in 

2002, but Many Others Waited Much Longer:



IRS’s data for January 1 through April 20, 2002, showed that about 

85 percent of the 2.6 million walk-in customers for whom wait-time was 

tracked waited 30 minutes or less to obtain assistance between January 

1 and April 20, 2002.[Footnote 32] However, IRS made the following two 

changes to the way it reported wait-time in 2002 that precluded any 

comparison to 2001.



* IRS reduced the number of sites for which wait-time data was tracked. 

In 2001, all walk-in sites tracked wait-time either automatically, if 

they were equipped with the Queuing Management System (Q-Matic), or 

manually, if they did not have Q-Matic.[Footnote 33] As of March 2002, 

however, the approximately 297 sites not equipped with Q-Matic were no 

longer required to track or report wait-time--a decision we believe was 

prudent because manual tracking was not practical at some sites with 

limited staff and because manual tracking of wait-times is an error-

prone process.



* IRS doubled the threshold used for wait-time tracking from 15 minutes 

or less in 2001 to 30 minutes or less in 2002, thus increasing the 

reported percentage of taxpayers receiving timely assistance.



The bi-weekly wait-time reports submitted by the sites and derivative 

summary data provided to us by IRS did not provide any details on the 

wait-times for the 15 percent of customers (about 384,000 taxpayers) 

who waited longer than 30 minutes for assistance, even though this 

information was available through Q-Matic,[Footnote 34] thus limiting 

management’s ability to identify and minimize excessive wait-time.



IRS Took Steps in 2002 to Improve Walk-in Service:



IRS took steps to improve its walk-in service, some of which may have 

helped improve the accuracy of tax law assistance in 2002. Other 

efforts that involve implementing two new accuracy measures and testing 

new 

Q-Matic requirements have the potential for improved service in the 

future.



IRS Took Steps to Improve Accuracy in 2002:



Field assistance officials said that they had taken a number of steps 

to improve the accuracy of assistance in 2002. For example, they said 

that IRS had completed all actions listed in response to our report on 

the 

2001 filing season[Footnote 35]--which included providing more 

consistent and standardized services, better training, and improved 

access to taxpayer account information. IRS also refined the process 

for referring complex tax law questions that are beyond the scope of 

normal walk-in staff training to expert field assistance, toll-free 

telephone, or compliance staff. The impact of some of those improvement 

efforts seemed to be reflected in the results of IRS’s annual employee 

satisfaction survey. For example, 49 percent of field assistance 

employees reported being satisfied with their training in 2002 compared 

with the 37 percent who reported being satisfied in 2001.



IRS Implemented Two New Accuracy Measures:



IRS implemented two new accuracy measures for account and return-

preparation assistance. We are concerned about one of those measures--

account accuracy--because the results of that measure are not 

representative of all walk-in sites. Instead of using statistical 

sampling methods to select the sites to be reviewed from among the 

about 420 sites that provide account assistance, IRS selected for 

review the largest walk-in sites from among the 123 that were equipped 

with Q-Matic.



IRS had also planned to measure return-preparation accuracy through 

visits to the largest walk-in sites. IRS subsequently decided to use 

existing return-accuracy data generated by the submission processing 

centers. This will allow IRS to measure the accuracy of all returns 

prepared at its walk-in sites, rather than basing the measure on sample 

observations.



IRS Is Testing New Q-Matic Requirements:



In May 2002, IRS changed its Q-Matic requirements to include networking 

and summary reporting capabilities and decided to test these changes at 

28 walk-in sites in California. In a stand-alone configuration, Q-Matic 

provides site-level, real-time monitoring and tracking of assistance, 

including customer wait-times, assistance provided, and staffing used. 

Networking the systems is expected to provide the same real-time 

monitoring and tracking at organizational levels above the site--group, 

territory, area, and nationwide. Summary reporting capability is 

expected to eliminate the requirement for manually recording and 

summarizing assistance provided and staff resources used, which should 

enhance site efficiency by increasing employee availability to provide 

more, or more timely, assistance.



Number of Taxpayers Assisted Declined, but Field Assistance Staffing 

Has Increased:



As shown in figure 5, the number of taxpayers assisted at IRS walk-in 

sites[Footnote 36] declined for the third consecutive filing season--

from 6.1 million in 1999 to 4.9 million in 2002--a drop of about 20 

percent. Included in that overall decline was a decrease in the number 

of contacts involving return preparation, a very time-consuming 

service.[Footnote 37]



Figure 5: IRS Walk-in Assistance, 1999-2002 Filing Seasons:



[See PDF for image]



Notes: “Total Walk-in” includes all face-to-face assistance. It does 

not include the number of taxpayers assisted by walk-in employees via 

telephone or correspondence, which ranged from over 100,000 taxpayers 

in 1999 to over 125,000 taxpayers in 2002.



The time periods covered by this figure each began on January 1 and 

ended on April 24, 1999; April 22, 2000; April 21, 2001; and April 20, 

2002.



[End of figure]



Field assistance officials attributed these declines to the following 

two factors:



* Taxpayers made greater use of the assistance available from 

volunteers and from IRS’s Web site and toll-free telephone network. 

Data provided by the two major volunteer organizations--Volunteer 

Income Tax Assistance and Tax Counseling for the Elderly--confirmed 

that volunteer assistance increased substantially. Specifically, the 

number of taxpayers assisted at volunteer sites increased about 45 

percent from nearly 2.5 million to 3.6 million between the 2001 and 

2002 filing seasons. Data on IRS’s Web site and toll-free telephone 

network are discussed elsewhere in this report.



* IRS has been reducing the income ceiling for return preparation 

assistance, thus limiting the number of taxpayers eligible to receive 

that kind of assistance. For the 2001 filing season, IRS had a ceiling 

of $41,000. For the 2002 filing season, IRS reduced that ceiling to 

$33,000--a level that approximated the $32,121 income ceiling on 

taxpayer eligibility for the earned income credit.



Traditionally, IRS has detailed staff from its compliance functions 

(such as Examination and Collection) to help provide assistance at 

walk-in sites. Over the last 2 filing seasons, IRS has reduced its 

reliance on compliance staff to provide walk-in assistance by about 53 

percent, thus freeing those staff to perform their normal examination 

and collection duties. At the same time, and despite the continued 

decline in the number of taxpayers assisted, IRS increased the number 

of field assistance staff years by about 31 percent, in anticipation of 

field assistance staff eventually assuming some work currently done by 

IRS’s compliance functions, such as office audits.



Field Assistance Has No Comprehensive Workload and Staffing Plans:



Field assistance does not have clear, comprehensive plans and 

timetables for making the workload changes just discussed as well as 

other changes. When we discussed this issue with field assistance 

officials, they informed us that they have put their plans on hold 

until anticipated compliance workloads are clarified to enable more 

comprehensive planning. Field assistance officials anticipated that 

some compliance work would migrate to walk-in staff over the next 

several years, although the extent and timing of that work was not yet 

clear. As of October 4, 2002, a team was studying the technological and 

other issues associated with compliance work anticipated to be done at 

walk-in sites. Field assistance officials also planned to phase out 

sites staffed with only one employee, because of difficulties providing 

back-up staff,[Footnote 38] and add many new walk-in sites. The planned 

phase-outs were to be done through a combination of methods, such as 

staff attrition and/or site consolidation. Instead of closing some one-

employee sites, IRS may add staff where warranted by demand. However, 

field assistance officials told us that these site changes also have 

been put on hold until the aforementioned comprehensive planning is 

completed.



Performance of IRS’s Web Site Was Generally Better Than Last Year, but 

Improvements Still Needed:



Our assessment of IRS’s Web site on the Internet, external assessments 

of the site, and IRS data indicate that, compared with last year, the 

site was:



* more user friendly, although there are certain aspects of the site 

that need attention;



* generally used more often, although IRS’s primary measure of usage--

number of hits--is flawed; and:



* easier to access.



IRS’s Web site provides a vehicle whereby taxpayers can receive 

assistance without having to call or visit an IRS office. Among other 

things, the site provides the potential to download hundreds of tax 

forms and publications, contains current information on tax issues and 

electronic filing, and gives taxpayers the opportunity to ask IRS tax 

law and procedural questions via E-mail (see app. V for information on 

IRS’s performance in answering E-mail questions.):



Web Site More User Friendly Than Last Year:



Our observations of IRS’s Web site indicated that the site was more 

user friendly than it was in 2001, although we still encountered some 

navigation problems and identified some problems with the site’s 

content.



Navigation of Web Site:



The Web site had many changes that improved navigation over 2001. For 

example, the home page was formatted more logically, with like items 

being grouped together, and we observed only a couple of broken links 

compared with last year when we found many more.[Footnote 39] However, 

there were still some navigational problems. The most significant 

problem in navigating the site is a search engine that often generates 

a very large number of items relating to the subject of the search, 

which makes it extremely difficult to locate the most useful/

appropriate item. For example, when we entered “earned income credit” 

into the general search engine, we were provided a list of over 1,500 

items. The publication that would provide someone interested in the 

earned income credit with the most thorough information (Publication 

596--”Earned Income Credit”) was the 64th item on the list. This would 

require users to scroll through 

6 pages of items before finding the basic publication dealing with the 

subject of the search.



We also noted the following navigation problems.



* The E-mail and comments sections of the Web site are difficult to 

find. The links to pages for users to leave E-mail questions or to 

provide IRS with comments on the site were not accessible through the 

“Contact Us” part of the site, where one might logically expect to go 

to send an E-mail or comment to IRS, but rather through the “Help” part 

of the site. In that regard, IRS officials responsible for answering E-

mail questions informed us that they believe the number of E-mail 

questions received this year declined, as discussed in appendix V, 

because individuals may have had trouble locating the E-mail section of 

the Web site. According to IRS officials responsible for the Web site, 

the 

E-mail feature was not made easier to locate because IRS is still 

considering whether it is best to answer taxpayer questions via the Web 

site or the telephone.



* We had difficulty in May 2002 finding telephone numbers and other 

information for local field assistance sites. For example, the page 

entitled “How to contact us in Virginia” said to use local numbers if 

they are not long distance, otherwise use the toll-free 800 number. 

However, to obtain local numbers, we had to scroll through 11 other 

topics, including information on where to file a return, TeleTax, 

Braille materials, and volunteer and education programs. Since then, 

IRS has taken steps to make this process less cumbersome.



* We found linkage problems in addition to the missing links noted 

earlier. For example, when we accessed IRS’s Web page for electronic 

filing options and paying taxes electronically, we found several 

references to more data that we could obtain. However, the link to 

those data was www.irs.gov, where we already were, rather than the 

appropriate areas of the site.



Content of Web Site:



We identified some problems with the content of IRS’s Web site. First, 

the site used document numbers as primary identifiers for publications, 

instructions, and forms. Many taxpayers do not know the numbers of 

forms and publications, which requires them to examine voluminous data 

to find the specific information they are seeking. For example, if 

taxpayers were looking for data on the earned income credit and did not 

know that Publication 596 is the primary publication on that topic, 

typing “earned income credit” in the form and publication search engine 

would produce a list of about 1,600 publications that mention the 

credit. Here again, as we noted earlier with respect to the general 

search engine, a taxpayer would have to scroll through several pages of 

the list to find Publication 596. IRS is aware of this problem with its 

search engines and is working with contractors to identify solutions.



Second, we continued to find some incorrect and inconsistent data, 

outdated material, and missing information, but not as much as last 

year. For example, during our review of the Web site in 2002, we found 

that the site had incorrect office hours and addresses for walk-in 

assistance sites and obsolete data that pertained to 2001. In our 

report on the 2000 filing season, we made recommendations to help 

ensure that the Web site contained accurate and consistent 

data.[Footnote 40] In response to our recommendations, according to IRS 

officials, IRS awarded a site redesign contract, instituted an interim 

process that required executives to approve all new data and 

significant changes to existing data, and began implementing a content 

management program that is expected to be fully operational around May 

2003. Although these steps appear to have reduced the volume of 

inaccurate and inconsistent data, a cognizant IRS official opined that 

it would take about 2 years before all data on the Web site are made 

current and consistent.



Two independent assessments tended to confirm our observations on IRS’s 

Web site. In that regard:



* Arthur Andersen’s Office of Government Services Experience Design 

Group concluded that the current version of the site is an improvement 

over the previous version but still requires some fine-tuning to be 

truly useful. They too noted that the search engines often result in a 

very large number of items relating to the subject of the search making 

it extremely difficult to locate the most useful/appropriate item.



* Brown University’s Center for Public Policy evaluated how well 

50 federal government Web sites would aid an average citizen logging 

onto a public sector Web site. IRS’s site received a score of 76 out of 

100, making it the eighth highest rated Web site in the study.



Available Data Show Increase in Web Site Use in 2002:



IRS’s two key measures for gauging the use of its Web site are “hits” 

and “downloads.” Both of those measures showed increased usage in 

2002 compared with 2001. As of May 31, 2002, the site had recorded 

about 2.2 billion hits compared with about 1.7 billion hits at the same 

point in time in 2001, a 27-percent increase. As of August 31, 2002, 

about 

379 million forms, publications, and instructions had been downloaded 

compared with about 279 million in 2001, a 36-percent 

increase.[Footnote 41]



However, as we have reported before, hits is a flawed measure of use 

because (1) every time a user accesses IRS’s home page it counts for 

about 16 hits and (2) a hit is counted every time the user moves to 

another page of the site.[Footnote 42] Thus, an increase in the number 

of hits may not necessarily be due to increased usage but could be 

attributed to changes in the structure of the home page or problems 

users are experiencing in finding what they want on the site. IRS 

officials had informed us that the problem would be corrected by an 

improved method for counting hits when a more sophisticated 

comprehensive Web analytical program became operational in January 

2002. Although the new analytical program is operational, the problem 

remained. However, our concern about the hits measure could be 

mitigated by the development of another Web site performance measure--

unique visitors. According to a cognizant IRS official, IRS is now able 

to track unique visitors to the Web site and has begun the process to 

formalize that measure.[Footnote 43]



Web Site’s Availability

and Average Delivery

Time Improved in 2002:



Keynote, an independent Web site rater and a recognized authority on 

Internet performance, reviewed the availability and average delivery 

time of IRS’s Web site and reported improved performance during the 

2002 filing season. Keynote reported that its measure of IRS’s Web 

site’s availability averaged close to 100 percent (99.8 to 100 percent) 

during the entire 2002 filing season. This is a significant improvement 

over the 

2001 filing season when Keynote identified periods during which 

availability was as low as 67 percent before rebounding to about 

97 percent for the first 2 weeks of April. Keynote reported an average 

delivery time in 2002 of 1 second or less for the entire filing season 

compared with 2 seconds or more for most of the 2001 filing season.



Conclusions:



Generally, during the 2002 filing season, IRS processed returns and 

issued refunds smoothly, and the quality of assistance provided to 

taxpayers improved. In light of this performance, IRS should be 

commended for the various efforts it took to prepare for and improve 

performance during the filing season. IRS’s filing season performance 

matters because it impacts well over 100 million taxpayers.



IRS established performance measures to emphasize accountability and 

improve service. IRS’s improved performance can be traced to its use of 

and emphasis on performance measures, illustrating the importance of 

good measures that effectively assess performance and enable 

improvements to be quantified. IRS has established some new measures 

and has revised some existing measures, which should serve to further 

enhance its performance assessments. However, we believe that some 

decisions, specifically (1) the decision to stop using the “assistor 

response level” and “average speed of answer” measures to assess 

telephone service; (2) not reporting readily available Q-Matic 

information showing the extent of customer wait-times; and (3) limiting 

reviews of account accuracy to only some walk-in sites, will reduce 

IRS’s ability to gauge the level of service being provided to 

customers. With respect to the latter, IRS’s decision to use existing 

data to measure return-preparation accuracy in lieu of having staff 

visit walk-in sites to observe return preparation could free up the 

resources needed to expand the reviews of account accuracy. IRS also 

needs to ensure that changes to walk-in sites and staffing are 

thoroughly analyzed and planned.



Recommendations for Executive Action:



We recommend that the Commissioner of Internal Revenue direct the 

appropriate officials to do the following:



* Reinstate a telephone assistance caller wait-time measure, such as 

assistor response level or average speed of answer, to assess this 

important aspect of the customer experience.



* Revise bi-weekly wait-time reports to show the numbers and 

percentages of customers in each of the wait-time intervals tracked by 

Q-Matic.



* Explore the feasibility of including all walk-in sites, rather than 

just sites equipped with Q-Matic, in the universe for the purpose of 

conducting sample visits/reviews of the accuracy of account assistance.



* Ensure that field assistance officials do a comprehensive evaluation 

of resource requirements (sites and staffing) after clarifying the 

extent and timing of (1) expected field assistance and compliance 

workloads and (2) expected walk-in site changes (additions and 

closures).



Agency Comments and Our Evaluation:



We requested comments on a draft of this report from IRS. The Acting 

Commissioner of Internal Revenue provided written comments in a 

December 16, 2002, letter (see app. VI). The Acting Commissioner 

appreciated our recognition of the many challenges that IRS 

successfully faced during the planning and execution of the 2002 tax 

filing season and noted that IRS’s filing season readiness process 

“served us well, especially in handling unanticipated challenges.”:



In response to our specific recommendations, the Acting Commissioner 

said the following:



² IRS agrees that customer wait-time is an important aspect of toll-

free telephone service and will “begin refining an appropriate wait 

time measure” for implementation in fiscal year 2004.



² In response to our recommendation that IRS revise its bi-weekly wait-

time reports, IRS said that it currently does not use wait-time as an 

official measure of field assistance performance because past 

experience has shown that employee reaction to timeliness measures 

tends to increase the likelihood of inaccurate or incomplete answers. 

We are aware of IRS’s position on having an official wait-time measure 

and discussed our disagreement with that position in our report on 

IRS’s filing season performance measures.[Footnote 44] Our 

recommendation in this report is not directed at the existence or non-

existence of a wait-time performance measure. It is directed, instead, 

at enhancing the information available to field assistance managers in 

the bi-weekly wait-time reports. IRS did not speak to that issue in its 

comments.



² Working with staff from IRS’s Statistics of Income group, field 

assistance has developed a statistically reliable sample of accounts 

work. For fiscal year 2003, IRS said it plans to sample small, medium, 

and large walk-in sites. These plans should provide for a more 

representative assessment of the level of account assistance being 

provided to walk-in customers and, as such, appear to be responsive to 

our recommendation.



² In response to our recommendation that field assistance do a 

comprehensive evaluation of resource requirements, IRS said that it is 

doing a workload and staffing study that will provide detailed 

information on the location of field assistance work and where IRS 

should provide service and staffing. According to IRS, although it had 

previously projected the need to add new sites, it is currently 

projecting shortages in field assistance staffing for fiscal year 2004 

that may cause it to reassess the number and location of walk-in sites. 

The study described by IRS appears responsive to our recommendation.



The Acting Commissioner also commented on our use of TIGTA’s reported 

accuracy rate for tax law assistance provided by walk-in sites. We 

addressed those comments earlier in the report where TIGTA’s reported 

accuracy rate is discussed.



We are sending copies of this report to the Chairmen and Ranking 

Minority Members of the Senate Committee on Finance and the House 

Committee on Ways and Means and the Ranking Minority Member, 

Subcommittee on Oversight, House Committee on Ways and Means. We are 

also sending copies to the Secretary of the Treasury; the Commissioner 

of Internal Revenue; the Director, Office of Management and Budget; and 

other interested parties. We will also make copies available to others 

on request. In addition, the report will be available at no charge on 

the GAO Web site at http://www.gao.gov.



This report was prepared under the direction of David J. Attianese, 

Assistant Director. Other major contributors are acknowledged in 

appendix VII. If you have any questions about this report, contact me 

on (202) 512-9110.



Sincerely yours,



James R. White

Director, Tax Issues:



[End of section]



Appendix I: Data on IRS’s Processing Performance Relative to Fiscal 
Year 

2001 Performance and Fiscal Year 2002 Goals:



As shown in table 3, (1) the Internal Revenue Service’s (IRS) 

performance in 2002 improved relative to 2001 for five measures and (2) 

IRS met its 2002 performance goals for four measures and missed its 

goals for two measures. IRS’s performance in 2002 could not be compared 

with its performance in 2001 for three measures or with its goals for 

2002 for two measures because IRS revised the way those measures were 

computed.



Table 3: IRS’s Processing Performance Relative to Fiscal Year 2001 

Performance and Fiscal Year 2002 Goals as of September 2002:



Measure name: Deposit error rate[B]; Description: Percentage of 

payments applied in error by, for example, reimbursing a taxpayer who 

overpaid when the taxpayer wanted any overpayment credited to next 

year’s tax bill; Fiscal year 2001 actual[A]: 5.0%; Fiscal year 2002 

goal: 3.4%; Fiscal

year

2002 actual: 4.8%; +/-0.3%; [Empty]; Revisions to measure: None known; 

Comments: Performance improved, but performance goal not met; 

According to IRS officials, the goal was too challenging and 

performance was consistent with 2001.



Measure name: Deposit timeliness-paper; Description: Interest value of 

money not deposited by the business day after receipt per $1 million in 

deposits. Measure assumes an 8 percent interest rate; Fiscal year 2001 

actual[A]: $748; Fiscal year 2002 goal: $751; Fiscal

year

2002 actual: $578; [Empty]; Revisions to measure: None known; 

Comments: Performance improved, and performance goal met.



Measure name: Letter error rate[B]; Description: Percentage of letters 

issued to taxpayers with errors, (includes systemic errors).[C]; Fiscal 

year 2001 actual[A]: Not comparable because of revisions to the 

measure; Fiscal year 2002 goal: 10.9%; Fiscal

year

2002 actual: 7.4%; +/-0.6%; [Empty]; Revisions to measure: Revised 

measure to exclude errors that cause a letter to look unprofessional 

but do not affect its accuracy; Comments: We cannot assess if 2002 

performance goal was met due to the exclusion of professionalism 

errors.



Measure name: Notice error rate[B]; Description: Percentage of 

incorrect notices issued to taxpayers (includes systemic errors).[C]; 

Fiscal year 2001 actual[A]: Not comparable because of revisions to the 

measure; Fiscal year 2002 goal: 15.2%; Fiscal

year

2002 actual: 18.7%; +/-2.7%; [Empty]; Revisions to measure: Revised 

measure to (1) exclude errors that cause a notice to look 

unprofessional but do not affect its accuracy and (2) include all 

processing notices, some of which we reported were inadvertently 

omitted in 2001 (see GAO-02-144); Comments: We cannot assess if 2002 

performance goal was met due to the exclusion of professionalism 

errors; According to IRS officials, employee confusion over new error 

codes associated with the rate reduction credit led to more erroneous 

notices, thereby increasing the notice error rate.



Measure name: Refund error rate - individual (paper)[B]; Description: 

The percentage of refunds with IRS-caused errors in the entity 

information (e.g., name or Social Security number) or refund amount; 

includes systemic errors.[C]; Fiscal year 2001 actual[A]: 9.8%; Fiscal 

year 2002 goal: 9.3%; Fiscal

year

2002 actual: 8.0%; +/-3.8%; [Empty]; Revisions to measure: Although 

this measure was revised to exclude errors that cause a refund check to 

look unprofessional, the revision did not require; establishment of a 

new baseline; Comments: Performance improved, and performance goal 

met.



Measure name: Refund interest paid; Description: Amount of refund 

interest IRS paid per $1 million of refunds issued; Fiscal year 2001 

actual[A]: $128.63; Fiscal year 2002 goal: $94.14; Fiscal

year

2002 actual: $62.55; [Empty]; Revisions to measure: None known; 

Comments: Performance improved, and performance goal met; According to 

officials,; IRS implemented strategies to address refund interest, 

resulting in lower refund interest paid in 2002.



Measure name: Refund timeliness-individual (paper)[B]; Description: 

Percentage of refunds issued in 40 days or less; Fiscal year 2001 

actual[A]: 95.2%; Fiscal year 2002 goal: 98.4%; Fiscal

year

2002 actual: 98.2%; +/ - 2.3%; [Empty]; Revisions to measure: Measure 

revised to reflect new start date procedures; 2001 performance data 

converted to allow for comparison; Comments: Performance improved; We 

consider IRS to have met the 2002 performance goal because a certain 

degree of imprecision exists due to sampling and the reported result 

differed from the goal by a statistically negligible amount.



Measure name: Productivity[D]; Description: Weighted volume of 

documents processed per staff year expended at the processing centers; 

Fiscal year 2001 actual[A]: Data being revised; Fiscal year 2002 goal: 

29,836; Fiscal

year

2002 actual: 28,182; [Empty]; Revisions to measure: Measure revised to 

reflect additional workload; 2001 performance data being converted to 

allow for comparison; Comments: Performance goal not met; According 

to IRS officials, productivity may have been lowered by (1) the 

increased error-correction workload resulting from the rate reduction 

credit and (2) changes associated with IRS’s reorganization.



Source: IRS data.



[End of table]



[A] These figures may be different than what IRS previously provided 

and we reported in GAO-03-143. IRS converted 2001 data to provide 

comparability between fiscal years 2001 and 2002.



[B] We previously reported on potential reliability weaknesses 

associated with these measures because data are collected manually and 

evaluations of data are based on judgment. See U.S. General Accounting 

Office, Tax Administration: IRS Needs to Further Refine Its Tax Filing 

Season Performance Measures, GAO-03-143 (Washington D.C.: Nov. 22, 

2002).



[C] Systemic errors are computer generated errors over which a 

particular processing center would have no control.



[D] We reported concerns about the understandability and 

appropriateness of this measure in 

GAO-03-143 because the measure’s definition is not clearly stated, unit 

managers do not understand their unit’s contribution to the formula, 

and unit managers do not use the measure to assess performance.



[End of section]



Appendix II: Challenges Affecting the 2002 Filing Season and IRS’s 

Response:



The following significant challenges were unique to the 2002 filing 

season:



* Passage of the Economic Growth and Tax Relief Reconciliation Act of 

2001, specifically the provisions relating to advance tax refunds and 

rate reduction credits.



* The September 11, 2001, attacks, which led to actions by IRS to 

provide tax relief to affected taxpayers, and the subsequent anthrax 

attacks perpetrated through the mail, which led to significant security 

enhancements.



* IRS’s shifting of computer files in conjunction with its 

reorganization.



Advance Tax Refunds and the Rate Reduction Credit Led to Many Errors on 

Filed Returns:



For tax years beginning after 2000, the Economic Growth and Tax Relief 

Reconciliation Act of 2001, signed into law on June 7, 2001, applied a 

new 10-percent income tax rate to a portion of an individual’s income 

that was previously taxed at 15 percent. To stimulate the economy more 

rapidly than would be achieved if taxpayers had to wait until they 

filed their tax year 2001 return to realize the full impact of this 

rate reduction, the act provided for eligible taxpayers to receive an 

advance tax refund in 

2001. The amount of the refund was to be based on the filing status and 

amount of taxable income on the taxpayer’s 2000 return.



Taxpayers who were eligible to receive an advance tax refund in 2001 

but who (1) did not receive a check because IRS did not have their 

current address or (2) did not have enough taxable income in 2000 to 

qualify for the maximum amount allowable, may have been entitled to a 

rate reduction credit when filing their tax year 2001 returns.[Footnote 

45] In addition, taxpayers who were not eligible for an advance tax 

refund, such as those who did not have taxable income in 2000, may have 

been entitled to a rate reduction credit provided they had taxable 

income in 2001.



According to IRS, passage of the Economic Growth and Tax Relief 

Reconciliation Act of 2001 required computer system programming changes 

that were complicated by frequent revisions as the Congress and the 

U.S. Treasury worked together to understand the legislative 

requirements for the rate reduction credit. This resulted in concurrent 

testing of those and other computer system changes. After the filing 

season began and large numbers of returns were received with errors 

related to the rate reduction credit, IRS made additional programming 

changes until March 2002 to help manage the high number of errors and 

assist tax examiners in correcting them.



As shown in table 4, as of September 27, 2002, IRS had identified over

8 million individual returns with rate reduction credit errors, which 

represented 57 percent of the returns identified with errors and 

6.5 percent of the total number of returns processed at that time.



Table 4: Processed Returns with Rate Reduction Credit Errors as of 

September 27, 2002:



Returns prepared by: Taxpayers; Number of returns processed: 

49,939,318; Number of returns with errors: 8,369,035; Percentage of 

returns with errors: 16.8%; Number of returns with rate reduction 

credit errors: 4,208,991; Percentage of returns with rate reduction 

credit errors: 8.4%; Rate reduction credit errors as a percentage of 

returns with errors: 50.3%.



Returns prepared by: Tax return practitioners; Number of returns 

processed: 73,013,833; Number of returns with errors: 5,691,483; 

Percentage of returns with errors: 7.8%; Number of returns with rate 

reduction credit errors: 3,816,860; Percentage of returns with rate 

reduction credit errors: 5.2%; Rate reduction credit errors as a 

percentage of returns with errors: 67.1%.



Returns prepared by: Total; Number of returns processed: 122,953,151; 

Number of returns with errors: 14,060,518; Percentage of returns with 

errors: 11.4%; Number of returns with rate reduction credit errors: 

8,025,851; Percentage of returns with rate reduction credit errors: 

6.5%; Rate reduction credit errors as a percentage of returns with 

errors: 57.1%.



Source: GAO’s analysis of IRS’s data.



Note: The number of returns processed does not include about 4.2 

million returns that were filed via TeleFile.



[End of table]



Taxpayers and tax return practitioners made various types of errors 

related to the rate reduction credit during the 2002 tax filing season. 

For example,



* taxpayers and practitioners filed over 5 million returns in which 

they failed to claim a credit even though the taxpayers were entitled 

to one;



* about 2 million returns were filed in which a credit was claimed even 

though the taxpayers had received the maximum advance tax refund in 

2001 and thus were not entitled to a credit; and:



* almost 1 million returns were filed in which a credit was 

appropriately claimed, but the credit amount was incorrectly computed.



Problems Related to the Rate Reduction Credit Were Identified Early in 

the 2002 Filing Season:



The Treasury Inspector General for Tax Administration (TIGTA) 

identified and IRS corrected two problems related to the rate reduction 

credit. One problem involved the lack of advance tax refund data in the 

National Accounts Profile (NAP) for certain taxpayers.[Footnote 46] 

According to TIGTA, IRS corrected this problem by January 11, 2002, 

thus preventing about 

217,000 taxpayers from receiving up to $50 million in rate reduction 

credits to which they were not entitled. The other problem, which TIGTA 

attributed to a misinterpretation of programming requirements, involved 

IRS’s failure to add information in the NAP for some taxpayers who did 

not receive an advance tax refund. According to TIGTA, it notified IRS 

of this problem, which could have affected as many as 35 million 

taxpayers, on January 8, 2002, and IRS made the necessary corrections 

by January 

15, 2002.



IRS field staff identified another problem in which IRS’s computer 

system was generating rate reduction credits for some taxpayers who had 

already received the maximum advance tax refund. According to IRS 

officials, this problem would have affected no more than about 15,000 

taxpayers, who in a worst-case scenario may have received an additional 

$300 credit, for a total of $4.5 million in potentially erroneous 

credits. According to the officials, IRS will not attempt to recover 

any erroneous payments resulting from this problem because recovery 

would not be cost-effective.



IRS Handled the Extra Workload from the Rate Reduction Credit:



To help handle the additional error correction workload resulting from 

the rate reduction credit workload, IRS, among other things, briefly 

detailed individuals from other projects, hired retirees, and postponed 

activities, such as technical training and performance reviews, that 

were eventually completed later in the year.



The Rate Reduction Credit Affected Performance Relative to Two 

Measures:



The rate reduction credit directly and adversely affected IRS’s 

performance relative to its notice error rate and productivity 

performance measures. According to IRS officials, (1) employee 

confusion over new error codes associated with the rate reduction 

credit led to more erroneous notices, thereby increasing the notice 

error rate, and (2) the new codes associated with the rate reduction 

credit initially reduced processing efficiency and caused lower 

productivity rates.



Many Taxpayers Received Corrected Rate Reduction Credits:



According to IRS officials, IRS sent corrected rate reduction credits 

to about 1.6 million taxpayers. When IRS originally reviewed these 

taxpayers’ returns, it was determined that the taxpayers had 

underclaimed the amount of their rate reduction credits. However, in 

accordance with IRS policy about making corrections under certain 

monetary thresholds, IRS did not initially correct the errors or give 

the taxpayers the additional amounts. During its review of the 2002 

filing season, TIGTA pointed out that the policy seemed inequitable 

because the same kind of threshold was not used in determining the 

amount of a taxpayer’s advance tax refund--thus the amount taxpayers 

received could vary depending on whether they received their money in 

the form of an advance tax refund or a rate reduction credit. According 

to IRS officials, these taxpayers’ returns did not go to error 

correction and, thus, are not included in the number of returns with 

rate reduction credit errors in table 4--which IRS officials believe 

contributed to their ability to handle the substantial error-correction 

workload.



Terrorist Actions Required Tax Relief for Affected Taxpayers and 

Security Enhancements:



The terrorist actions in 2001 created additional challenges to which 

IRS responded. The September 11, 2001, terrorist actions created 

challenges by causing additional workload and requiring concurrent 

testing of the programming changes related to that workload with other 

testing. For example, according to IRS officials, programmers modified 

computer systems in order to suppress certain correspondences, such as 

balance due notices, to taxpayers affected by the terrorist actions. 

IRS officials also reported that where the correspondence was already 

printed, processing centers held that correspondence until November. 

IRS then generated new correspondence with revised balance due dates, 

for example, and included leaflets with the correspondence that advised 

taxpayers to contact IRS if they were unable to meet their tax 

obligations due to the events of September 11.



Also, as reported by TIGTA in March 2002, IRS, in response to anthrax 

attacks perpetrated through the mail, “quickly implemented actions to 

increase security and safety precautions during the filing 

season.”[Footnote 47] Those actions included providing additional 

security in areas where mail is handled, changing mail handling 

procedures, and providing additional training and protective equipment 

for employees.



IRS’s Reorganization Required a Shifting of Computer Files and Affected 

Processing Performance:



As part of its reorganization, IRS shifted computer files to match 

taxpayer accounts with the operating division responsible for serving 

the taxpayers. For example, computer files for individual taxpayers 

were shifted to the computer systems for the Wage and Investment 

Operating Division (W& I) processing centers and small business 

accounts were shifted to the computer systems for the centers that now 

specialize in small business processing and compliance activities. IRS 

completed this work before the filing season began in spite of the 

unanticipated challenges associated with the rate reduction credit and 

terrorist actions.



According to processing officials, IRS’s ongoing reorganization 

affected submission processing’s performance in other ways. It 

contributed to lower overall productivity, in part by changing the 

predictability of filing patterns. Also, the eight W&I centers no 

longer process some of the easier business returns and are left with 

processing the more complicated and time-consuming individual income 

tax returns, such as those with Schedule Ds (Capital Gains and Losses). 

Combined with the trend toward increased use of electronic filing by 

those taxpayers who file simpler tax forms, such as the Form 1040 EZ, 

IRS officials believe that these changes contributed to slower 

processing and a corresponding lower productivity rate.



[End of section]



Appendix III: Changes to Third-Party Authorizations in 2002:



IRS offers taxpayers a variety of authorizations that allow IRS to 

disclose or discuss tax return information with a third party. These 

authorizations range from allowing IRS to discuss the processing of a 

return, including the status of tax refunds (hereafter referred to as 

the third-party checkbox authority), to authorizing a third party to 

receive confidential tax return information. For the 2002 filing 

season, IRS expanded the third-party checkbox authority to allow the 

taxpayer to designate any third party, including friends or family 

members (in 2001, only paid practitioners could be designated). In 

response to a recommendation in our report on the 

2001 filing season,[Footnote 48] IRS now has a system in place for 

tracking how use of the third-party checkbox authority facilitated the 

resolution of issues related to a return’s processing. IRS also 

introduced the Oral Disclosure Consent (ODC) and Oral Tax Information 

Authorization (OTIA) to enable taxpayers to orally authorize IRS to 

disclose tax return information to a third party.



Authority to Designate a Third Party Expanded:



For the 2002 filing season, IRS expanded the third-party checkbox 

authority on individual income tax returns (Forms 1040, 1040A, and 

1040EZ) to allow taxpayers to designate anyone, including a friend or 

family member, to talk with IRS to resolve problems (such as math 

errors and missing information) that come to light during IRS’s 

processing of a return and obtain answers to questions regarding 

refunds or payments. When this authority was first used during the 2001 

filing season, taxpayers were only allowed to authorize paid 

practitioners to serve as their designees in resolving problems or 

answering questions. According to IRS, the expansion of this authority 

to any third party was intended to reach taxpayers who wanted to have a 

family member or close associate assist with simple tax inquiries.



According to IRS, the third-party checkbox authority is part of IRS’s 

Taxpayer Treatment and Service Improvement Initiative on third-party 

authorizations. IRS considers this initiative a way to reduce taxpayer 

burden by reducing the number of notices sent to taxpayers and allowing 

a designee to talk directly with IRS to resolve issues during a 

return’s processing. During a typical year, IRS sends out about 8 

million notices regarding tax return processing issues. According to 

IRS officials, the Commissioner of Internal Revenue forecasted that, 

with this authority and other innovations, such as oral consent 

authorizations, 90 percent of those situations might be resolved 

through telephone discussions with designees.



According to IRS, taxpayers checked the box on about 32.4 million 

returns as of October 7, 2002. When IRS first implemented this 

authority in 2001, it lacked a system for tracking how use of the 

authorization facilitated the resolution of return processing issues. 

In response to our recommendation that IRS develop a plan for 

evaluating the implementation and effectiveness of processing 

initiatives, IRS implemented computer programming changes that now 

enable it to know not only how many people used the checkbox option but 

also how many inquiries were received and responded to by type of 

return and notice.



IRS officials estimated, based on their tracking, that the 32.4 million 

returns with marked checkboxes translated to about 4 million fewer 

pieces of correspondence. In addition, IRS responded to about 4 million 

inquiries from persons whom taxpayers had identified as their designees 

using the checkbox authority.



Oral Authorizations Offered:



A taxpayer can now orally authorize IRS to disclose tax return 

information to a third party. Both ODC and OTIA allow a taxpayer to 

contact IRS and establish disclosure authority for all types of tax 

accounts. The oral authorizations, however, differ in the range of 

authorities given to the third party.



The oral authorizations were made possible by the Taxpayer Bill of 

Rights II,[Footnote 49] which removed the requirement that all requests 

of or disclosures to a third party be in writing. The new disclosure 

regulations[Footnote 50] authorize IRS employees to accept a taxpayer’s 

verbal request or consent to disclose return information to third 

parties to assist the taxpayer in resolving account issues. When oral 

authorizations are used, IRS must verify the taxpayer’s identity and 

the identity of the third party as well as confirm the specific tax 

matter.



To establish an ODC, the taxpayer must have received a notice from IRS 

and have open account issues. The taxpayer may call IRS and designate a 

third party to resolve issues relative to the information indicated on 

the notice. Once an ODC has been established, the taxpayer does not 

have to be physically present or on the telephone when IRS discloses 

the tax information to the third party nor does the third party need an 

authorization number. The ODC does not authorize a third party to 

receive notices or other written taxpayer information. The authority is 

valid until the tax related issues are resolved. If the taxpayer 

receives subsequent notices from IRS, he or she must call or visit IRS 

and establish another ODC even if it is for the same third party and 

tax matter.



An OTIA allows taxpayers to designate third parties to represent them 

by telephone or in person and to receive notices or account transcripts 

on open account issues. The taxpayer does not have to receive a notice 

from IRS before calling to establish an OTIA. Similar to the ODC, the 

taxpayer does not have to be physically present or on the telephone 

when IRS discloses tax information to the third party. However, to 

confirm their identify, the third party must have an authorization 

number given by IRS, which must be on file, and the third party must 

provide this number to the taxpayer. This authority expires upon 

written request by either the taxpayer or representative.



Table 5 provides comparative information on the various authorizations 

discussed in this appendix.



Table 5: Third-Party Authorizations:



Effective date; Third-party checkbox (2001): Available for the 2001 

filing season; Third-party checkbox (2002): Available for the 2002 

filing season; Oral disclosure consent: Available for the 2002 filing 

season; Oral tax information authorization: Available for the 2002 

filing season.



Purpose; Third-party checkbox (2001): Allows a paid practitioner to 

discuss return processing and refund issues; Third-party checkbox 

(2002): Allows a practitioner, friend, or family member to discuss 

return processing and refund issues; Oral disclosure consent: Allows 

IRS to disclose information orally, but not written information, to a 

third party on issues related to a notice; Oral tax information 

authorization: Allows a third party who has an authorization number 

given by IRS to receive information orally and some written taxpayer 

information.



Who can exercise this authority; Third-party checkbox (2001): 

Taxpayer; Third-party checkbox (2002): Taxpayer; Oral disclosure 

consent: Taxpayer; Oral tax information authorization: Taxpayer.



How is authority granted; Third-party checkbox (2001): Taxpayers check 

a box on individual income tax forms; Third-party checkbox (2002): 

Taxpayers check a box on individual income tax forms and some business 

forms; Oral disclosure consent: Orally; Oral tax information 

authorization: Orally.



Expiration of authority; Third-party checkbox (2001): April 15 of the 

year following the year the checkbox election was made, through the 

life of the first notice, or until the account is resolved; Third-

party checkbox (2002): April 15 of the year following the year the 

checkbox election was made, through the life of the first notice, or 

until the account is resolved; Oral disclosure consent: Until the tax 

related issue is resolved or when a new oral consent is entered on the 

account; Oral tax information authorization: By written request of 

taxpayer or representative.



Processing timeframes; Third-party checkbox (2001): 4 to 6 weeks; 

Third-party checkbox (2002): 4 to 6 weeks; Oral disclosure consent: 

Within minutes; Oral tax information authorization: Within minutes.



Must taxpayer have received a notice; Third-party checkbox (2001): No; 

Third-party checkbox (2002): No; Oral disclosure consent: Yes; Oral 

tax information authorization: No.



Source: IRS data.



[End of table]



[End of section]



Appendix IV: Results of Surveys of Taxpayers and Tax Practitioners Who 

Prepared Returns on a Computer but Filed on Paper:



In 2002, IRS surveyed taxpayers and tax practitioners who prepared 

their returns on computer but filed on paper. We had recommended such 

surveys in our report on the 2001filing season in response to the fact 

that about 40 million computer-prepared individual income tax returns 

were filed on paper in 2001.[Footnote 51] The surveys involved (1) 

taxpayers who prepared their own tax return on a computer but filed a 

paper return, (2) taxpayers who used paid practitioners who prepared 

their returns on computer but filed on paper, and (3) practitioners who 

prepared returns on a computer but filed on paper. Table 6 shows, for 

each category of respondent, 

(1) factors that influenced them not to file electronically and (2) 

incentives that could encourage them to file electronically.



Table 6: Factors That Influence Preparers of Returns on a Computer Not 

to File Electronically and Incentives That Could Influence Them to File 

Electronically:



Category of respondent: Taxpayers who prepare their own returns; 

Factors that influenced them not to file electronically: * Cost (51%); 

* Security and technology concerns (32%); * Like seeing return, 

signing it, and putting it in the mail (30%); Incentives that could 

influence them to file electronically: * Free electronic filing (71%); 

* Tax credit to cover cost of electronic filing (46%); * Electronic 

refund in 3 to 5 days (36%); * Ability to file electronically on IRS’s 

Web site (34%).



Category of respondent: Taxpayers who used paid practitioners; Factors 

that influenced them not to file electronically: * Taxpayer always 

filed on paper (34%); * Practitioner did not mention or recommend it 

(33%); * Taxpayer did not want to pay cost to file electronically 

(28%); Incentives that could influence them to file electronically: * 

Free electronic filing (60%); * Tax credit to cover cost of electronic 

filing (42%); * Electronic refund in 3 to 5 days (40%).



Category of respondent: Paid practitioners; Factors that influenced 

them not to file electronically: * Taxpayers do not request it (66%); 

* Practitioners do not like the need for a paper signature document 

(51%); * Software cost is too high (51%); * Electronic filing is too 

time-consuming (51%); Incentives that could influence them to file 

electronically: * IRS paying $3 for each electronically filed return 

(63%); * IRS offering free transmission of return (52%); * Electronic 

refund in 3 to 5 days (51%); * Free software (48%); * Lower 

transmission costs (47%).



Source: IRS reports on the results of its surveys.



[End of table]



[End of section]



Note: The percentages in this table represent the percent of 

respondents who cited the factor or incentive as having a strong 

influence.



[End of section]



Appendix V: IRS Received Fewer E-Mail Questions Via Its Web Site and 
Did 

a Worse Job Responding to the Questions:



One aspect of IRS’s Web site where available data indicated worse 

performance in 2002 than in 2001 was the feature that allows taxpayers 

to ask IRS tax law and procedural questions via E-mail. For the 2002 

filing season, IRS received fewer E-mail questions and did a worse job 

responding to those questions than in 2001. Also, the E-mail function 

of the site received mixed reviews by users who responded to the 

customer satisfaction survey.



As of May 31, 2002, IRS received about 109,000 E-mail questions 

compared with about 160,000 for the same time period in 2001--a 32-

percent decrease. IRS officials we talked with did not know why the 

number of E-mail questions had decreased but speculated that redesign 

of the Web site might have made it difficult for taxpayers to locate 

the link for submitting questions. As noted earlier, the link for 

submitting questions was accessible not through the “Contact Us” part 

of the site but through the “Help” part of the site.



IRS’s goal is to respond to E-mail questions within 2 business days. 

However, according to IRS data, IRS took 2.4 business days on average 

to respond to taxpayer’s E-mail questions in 2002, compared with 

0.8 business days in 2001. The IRS data indicated that timeliness was a 

problem in January and February (4.2 and 2.4 business days, 

respectively) but improved substantially in March and April (1.2 and 

0.8 business days, respectively). According to responsible IRS 

officials, the increase in time to respond to E-mail questions was 

caused by several factors:



* The most significant factor was a decision to concentrate on E-mail 

questions received in the work period just before the start of the 

2002 filing season.



* IRS consolidated the responsibility for responding to all E-mail 

questions in one site, and there was an inadequate alignment of 

employees’ skills with the new workload.



* IRS believed that far more complex questions were received in 

2002 than in 2001 because the expanded “frequently asked questions” 

section of the Web site was being used by taxpayers to get answers to 

more common, less complex questions.



IRS informed us that, for the 2003 filing season, it will be conducting 

more training and adding another site to answer E-mail questions. The 

site being added answered E-mail questions in 2001.



In 2002, IRS did not meet its 78-percent accuracy goal and, on a 

percentage basis, answered fewer questions accurately, despite 

receiving 32 percent fewer E-mail questions than in 2001. IRS data for 

January 1 to April 

20, 2002, showed that IRS responded accurately to 76 percent of the E-

mail questions in 2002. That compares with a 78-percent accuracy rate 

for a similar time frame in 2001. Some of the same factors cited as 

possible reasons for the timeliness problems (i.e., an inadequate match 

of employee skills with workload and the receipt of more complex 

questions) could also help explain the decrease in accuracy.



IRS provides all E-mail customers an opportunity to respond to a 

customer satisfaction survey, and about 2,200 did so between January 1 

and April 

30, 2002. The customers who responded to the survey in 2002 provided 

mixed reviews of the E-mail service.



* 91 percent of the respondents said that they were satisfied with the 

time it took to get a response, compared with 96 percent in 2001--a 

decline that may reflect the increase in average response times 

discussed earlier.



* 92 percent of the respondents said that they would use the E-mail 

service again, compared with 91 percent in 2001.



* 79 percent of the respondents said that IRS’s response answered their 

question, compared with 75 percent in 2001.



[End of section]



Appendix VI: Comments from the Internal Revenue Service:



An IRS official told us that the figures cited for Web site hits and 

downloads are for the full fiscal year (October 1 through September 

30).



DEPARTMENT OF THE TREASURY INTERNAL REVENUE SERVICE WASHINGTON. D.C. 

20224:



COMMISSIONER:



December 16, 2002:



Mr. James R. White, Director, Tax Issues, U.S. General Accounting 
Office 

441 G Street, N.W. Washington, D.C. 20548:



Dear Mr. White:



I appreciate your recognition of the many challenges we successfully 

faced during the planning and execution of the 2002 tax filing season. 

Our Filing Season Readiness process served us well, especially in 

handling unanticipated challenges. As you reported, we have greatly 

expanded electronic services to taxpayers and the quality and level of 

service our field assistance centers provide.



Your report referenced the Treasury Inspector General for Tax 

Administration (TIGTA) reviews on Field Assistance Quality. Using the 

TIGTA accuracy rates is inappropriate. In the Fiscal Year (FY) 2001 and 

2002 results, the TIGTA included referrals to publications in computing 

the accuracy rate. I do not agree that referring taxpayers to IRS tax 

publications is the same as providing incorrect answers to questions. 

While we do not want employees to improperly refer questions, those 

referrals should not be counted as incorrect answers and should be 

excluded from the calculation. The use of TIGTA’s calculation of 50 

percent is not a true reflection of our performance. By excluding these 

referrals as errors, the true cumulative accuracy rate is 66 percent 

through June 2002. We do agree, however, that we still must improve.



Some of our other significant improvements this filing season include:



Customer Accounts Management:



Accuracy of toll-free telephone service tax law and tax account answers 

improved substantially. Through October 31, 2002, the quality of tax 

law answers improved 8 percentage points to 81 percent. The quality of 

tax account answers improved 7 percentage points from the prior year to 

74 percent. These quality measures include adherence to procedures as 

well as the accuracy of the response. When measured for accuracy alone, 

the tax law correct response rate was over 84 percent and the account 

correct response rate was over 90 percent.



We exceeded our toll-free telephone Customer Service Representative 

(CSR) Level of Service (LOS) goal of 71.5 percent for 9 of the 15 weeks 

in the filing season. We were less than 2.7 percentage points below our 

goal in three of the missed weeks. As of October 31, 2002, cumulative 

CSR level of service was over 85 percent, a 106 percent increase over 

the prior year.



*Our Submission Processing Centers processed almost 131 million returns 

and generated almost 196 billion in tax refunds, while successfully 

dealing with the additional workload generated by the Rate Reduction 

Credit.



Taxpayer Face to Face Services:



*We have taken a number of steps to improve the accuracy of assistance 

in our Taxpayer Assistance Centers (TAC), such as providing 

standardized services and better training, and refining the process for 

referring complex tax law questions that are outside the scope of 

services provided by Field Assistance employees.



*To address employee training needs, we require all TAC technical 

employees to complete 1 hour of uninterrupted self-directed learning 

each week.



*Beginning in July 2002, managers interviewed taxpayers as they left 

the TAC to determine the level of service they received. Managers also 

conducted anonymous “shopping visits” at TACs other than their own to 

determine the scope of the accuracy problem. We used the results of 

these visits and similar reviews conducted by TIGTA to identify error 

trends and training needs to improve quality.



During FY 2003 Continuing Professional Education (CPE) classes, 

employees will receive job aids to guide them through the publication 

method. The publication method is a technique employees use to “walk” a 

taxpayer through a publication as they respond to a tax law question. 

It helps employees ask appropriate probing questions so they can 

correctly answer customers’ questions.



*Field Assistance is installing a networked queuing management system 

(QMatic) in 28 TACs in California. This will enable IRS to test Q-Matic 

networking and summary reporting capabilities. These sites are 

scheduled to be operational in January 2003.



Electronic Filing and Internet Services:



*As of October 31, 2002, nearly 46.9 million taxpayers filed 

electronically, surpassing the 40.2 million for the prior year. Nearly 

105,000 electronic return originators, 15,000 more than last year, 

participated in e-file resulting in more accurate and faster refunds.



*Home computer e-filing increased as taxpayers filed more than 9.4 

million returns from their home computers, a 38 percent increase over 

last year.



*Electronic payments are up. Taxpayers made over 768,000 e-payments, an 

increase of 8.5 percent over the same time last year. This includes 

313,385 credit card payments.



*More than 24.8 million taxpayers signed their returns electronically 

by using one of the two Personal Identification Number options.



*The IRS website, www.irs.gov, is one of the most used government 

sites. During the tax season we had 3.1 billion hits, a 15 percent 

increase from last year. In addition, the number of files downloaded 

last year topped 437 million, an increase from 317 million the year 

before.



For the FY 2003 filing season, we are taking steps to improve service 

by encouraging Internet filing with free tax preparation software, 

adding seven new e-file formats, and increasing services to electronic 

return preparers.



My responses to your recommendations for Executive Action follow:



Recommendation:



Reinstate a telephone assistance caller wait time measure, such as 

assistor response level or average speed of answer, to assess this 

important aspect of the customer experience.



Response:



We agree that measuring customer wait time is an important indicator of 

the service we are providing to our customers, and we have consistently 

tracked and used this information to manage telephone operations. 

Although we removed this as a Balanced Measure in FY 2002, we recognize 

that speed of service is important to our customers. We continue to 

closely monitor and report our wait time performance to all of our 

stakeholders. Wait time is a primary factor for real time call 

management and call:



routing actions taken by the Joint Operation Center. Minimizing 

customer wait time is also a key consideration when we plan workload 

and schedule staffing. The fact that we eliminated wait time measures 

from the Balanced Measures has not diminished its importance to us. We 

agree with your recommendation and will begin refining an appropriate 

wait time measure for our FY 2004 -2005 Strategy and Program Plan.



Recommendation:



Revise biweekly wait time reports to show the numbers and percentages 

of customers in each of the wait time intervals tracked by Q-Matic.



Response:



Wait time at TACs is not currently a measure because we concluded that 

this measure tended to adversely affect the thoroughness and quality of 

answers. Unlike toll-free telephone operations, where wait time is a 

function of call management and systemic call routing, our past 

experiences in the walk-in environment have shown that employee 

reaction to timeliness measures tend to increase the likelihood of 

inaccurate or incomplete answers. Our current focus is devoted to 

improving the accuracy of answers to tax law questions and the quality 

of other work.



Recommendation:



Explore the feasibility of including all walk-in sites, rather than 

just sites equipped with Q-Matic, for the purpose of conducting sample 

visits/reviews of the accuracy of account assistance.



Response:



We worked with Statistics of Income (SOI) to develop a national 

statistically reliable sample of accounts work within the TACs. We then 

reviewed account accuracy in the largest TACs equipped with Q-Matic as 

a means of developing goals. These baseline goals will be used to 

measure account accuracy in all TACs.



For FY 2003, we worked with SOI and developed an Area statistically 

reliable sample of accounts work. Large, medium and small TACs will be 

sampled with every workgroup within Field Assistance being sampled 

during the year. We will assess accounts accuracy in conjunction with 

the tax law accuracy review. Our Quality Assurance Team will review a 

representative sample of accounts work in the TAC being reviewed, and 

account accuracy will be based on whether all of the taxpayer’s 

concerns were resolved.



Recommendation:



Ensure that Field Assistance officials do a comprehensive evaluation of 

resource requirements (sites and staffing) after clarifying the extent 

and timing of (1) expected field assistance and compliance workloads, 

and (2) expected walk-in site changes (additions and closures).



Response:



We are doing a workload and staffing study that will provide detailed 

information on the location of work and where we should provide service 

and staffing. This study will include inventory driven work, such as 

office examinations and collection queue work, and the more traditional 

demand driven work such as accounts and notices. During the design 

stage for this program, we included new assistance sites in our 

projections. However, we now project shortages in Field Assistance 

staffing for FY 2004. As a result, we may have to reassess the number 

and location of TAC sites.



Again, I appreciate your observations and recommendations. If you have 

questions or comments, please call Floyd Williams, Director, 

Legislative Affairs, at (202) 622-3720.



Sincerely,



Bob Wenzel

Acting Commissioner



Signed by Bob Wenzel



[End of section]



Appendix VII: GAO Contacts and Staff Acknowledgments:



GAO Contacts:



James White (202) 512-9110

Dave Attianese (202) 512-9110:



Acknowledgments:



In addition to those named above, Bob Arcenia, Heather Bothwell, 

Grace Coleman, Ron Heisterkamp, Ronald W. Jones, John Lesser, 

Tina Smith, and Joanna Stamatiades made key contributions to this 

report.



FOOTNOTES



[1] Most taxpayers file their returns between January 1 and April 15, 

which is the deadline for filing individual income tax returns. 

However, millions of taxpayers get extensions from IRS that allow them 

to delay filing until as late as October 15.



[2] U.S. General Accounting Office, Internal Revenue Service: 

Assessment of Budget Request for Fiscal Year 2003 and Interim Results 

of 2002 Tax Filing Season, GAO-02-580T (Washington, D.C.: Apr. 9, 

2002).



[3] As part of a reorganization that took effect in October 2000, IRS 

established four operating divisions that serve specific groups of 

taxpayers. The four divisions are (1) W&I, (2) Small Business and Self-

Employed, (3) Large and Mid-Size Businesses, and (4) Tax Exempt and 

Government Entities. Our review focused on W&I, which generally serves 

taxpayers whose income is from wages and investments and which, as part 

of that service, processes individual income tax returns and provides 

assistance to taxpayers who call on the telephone or walk into an IRS 

office.



[4] U.S. General Accounting Office, Tax Administration: IRS Needs to 

Further Refine Its Tax Filing Season Performance Measures, GAO-03-143 

(Washington, D.C.: Nov. 22, 2002). 



[5] The Joint Operations Center is the organization responsible for 

managing IRS’s telephone operations. 



[6] P.L. 107-16.



[7] The terms “advance tax refund” and “rate reduction credit” are not 

synonymous. Taxpayers who were eligible to receive an advance tax 

refund in 2001 but who did not receive a check may have been entitled 

to a rate reduction credit when filing their tax year 2001 returns. A 

rate reduction credit would reduce a taxpayer’s 2001 tax liability and 

could have resulted in a tax refund once the 2001 return was filed. 



[8] P.L. 105-206.



[9] The Government Performance and Results Act (P.L. 103-62) was 

enacted to hold federal agencies accountable for achieving program 

results. IRS’s balanced measurement system is consistent with the 

intent of the act.



[10] Of the 131 million individual income tax returns, about 84 million 

were filed on paper and about 47 million were filed electronically. 

This section of the report focuses on IRS’s processing of paper; 

electronic filing is discussed later. 



[11] The National Association of Enrolled Agents is a national 

association of over 10,000 independent, licensed tax professionals 

called enrolled agents. Enrolled agents are licensed by the federal 

government and are authorized to appear in place of the taxpayer at 

IRS. 



[12] Although the 2002 Filing Season Was Completed Timely, Customer 

Service Can Be Improved During Error Processing, TIGTA, Reference No. 

2002-40-200, Sept. 26, 2002. 



[13] According to TIGTA, although employees worked overtime, the amount 

of overtime paid employees in the eight W&I processing centers in 2002 

did not increase compared with the amount paid in 2001.



[14] We also could not compare IRS’s performance in 2002 and 2001 

relative to the productivity measure because IRS was revising its data 

for 2001 at the time we completed our audit work.



[15] Although this change also applied to the refund error rate-

individual (paper) measure, a new baseline for refund error rate-

individual (paper) was not warranted because, according to IRS 

officials and supporting data provided by IRS, the policy change did 

not affect this measure to the extent it affected the other two. 



[16] The Internal Revenue Service Certifies Its Readiness to Process 

Individual Income Tax Returns in 2002, TIGTA, Reference No. 2002-40-

080, Mar. 29, 2002.



[17] This section discusses the measures that IRS uses to evaluate 

electronic filing performance compared to past performance, 2002 goals, 

and IRS’s long-term goal for individual income tax returns. The report 

we issued on IRS’s filing season performance measures (GAO-03-143) 

includes many performance measures related to electronic filing that 

are not discussed in this report because those measures relate to other 

returns, such as those from businesses.



[18] The IRS Restructuring and Reform Act of 1998 mandated that the 

Secretary of the Treasury convene an electronic commerce advisory group 

to ensure that the Secretary receives input from the private sector on 

IRS’s plan to increase electronic filing. ETAAC was created in 1998 in 

response to that mandate and, among other things, is required to report 

to the Congress annually on IRS’s progress towards meeting the 

electronic filing goals set in the act. 



[19] U.S. General Accounting Office, Tax Administration: Assessment of 

IRS’ 2001 Tax Filing Season, GAO-02-144 (Washington D.C.: Dec. 21, 

2001).



[20] To electronically file, taxpayers must generally either pay a tax 

practitioner to prepare and submit their tax return, pay a transmitter 

to transmit their tax return, or purchase a software package and file 

their tax return on-line through an on-line intermediary using a 

personal computer. 



[21] IRS determines which callers were attempting to reach an assistor 

based on how they responded to menu prompts. For callers who received 

busy signals and, therefore, did not get a chance to respond to the 

menu, IRS estimates how many were attempting to reach an assistor based 

on the percentage of calls that were served by assistors that week. 

IRS’s estimate of the callers who attempted to reach an assistor does 

not reflect callers who did not get a busy signal but yet hung up 

before completing the menu responses necessary to determine the type of 

assistance (assistor or automation) that would have handled their 

calls.



[22] IRS has other measures that we are not reporting on because they 

do not provide adequate information to make conclusions about the ease 

with which taxpayers reached IRS or the accuracy of the responses they 

received after reaching IRS. 



[23] In GAO-03-143, we recommended that IRS establish measures for its 

automated telephone services. 



[24] According to IRS officials, the filing season for telephone 

assistance generally begins January 1 and ends around July 15 each 

year. Although the tax return filing deadline is April 15, the demand 

for telephone assistance continues past the deadline as taxpayers call 

to, among other things, inquire about the status of their refunds or 

respond to notices they received from IRS about their filed returns. 



[25] U.S. General Accounting Office, Advance Tax Refund Program Was a 

Major Accomplishment, but Not Problem Free, GAO-02-827 (Washington, 

D.C.: Aug. 2, 2002).



[26] U.S. General Accounting Office, Customer Service: Human Capital 

Management at Selected Public and Private Call Centers, GAO/GGD-00-161 

(Washington, D.C.: Aug. 22, 2000).



[27] “Services provided” counts the number of times callers reached 

assistors. “Handle time” measures the total time an assistor spent 

talking to a taxpayer, keeping the taxpayer on hold, and finishing the 

call--up to the time the assistor indicated his or her readiness to 

receive another call.



[28] GAO-02-212.



[29] U.S. General Accounting Office, IRS Telephone Assistance: 

Opportunities to Improve Human Capital Management, GAO-01-144 

(Washington, D.C.: Jan. 30, 2001).



[30] In general, taxpayers who visit IRS walk-in locations can obtain 

(1) tax forms and publications; (2) answers to questions about the tax 

law; (3) assistance with their tax accounts; (4) limited return-

preparation assistance; and (5) various other types of assistance, such 

as help getting a taxpayer identification number. 



[31] The 50-percent accuracy rate represents the cumulative rate from 

visits done from January through April 2002; the true accuracy rate 

will not be known until TIGTA completes its reviews in fiscal year 

2003. TIGTA reviewers considered responses correct when they were 

answered accurately and in enough detail to avert filing errors. 

Conversely, they considered responses incorrect when walk-in staff did 

not answer accurately or in enough detail, referred reviewers to a 

publication without walking them through it to get the right answer as 

required by field assistance procedures, or denied assistance by 

failing to either answer their question or refer them to someone who 

could.



[32] Beginning with fiscal year 2001, IRS discontinued wait-time as an 

official measure of field assistance performance because, according to 

field assistance officials, employees felt pressured to hurry 

assistance to customers, which could diminish other aspects of quality, 

such as accuracy. Nevertheless, realizing that timeliness is an 

important aspect of quality, IRS has continued to track and report 

wait-time. 



[33] Q-Matic is an automated system that keeps track of such things as 

customer wait-times, assistance provided, and staff time used. 



[34] In general, Q-Matic records the number of customers whose wait-

times fell into each 

15-minute interval, ranging from less than 15 minutes to less than 2.5 

hours, as well as the number that waited longer than 2.5 hours.



[35] GAO-02-144.



[36] Although the numbers may vary during the year, in 2002 IRS 

provided walk-in assistance through approximately 420 walk-in sites 

that generally provide the full range of assistance on a year-round 

basis, 50 alternate sites (such as at libraries and shopping malls) and 

7 vans or mobile units that are set up mainly to provide return-

preparation assistance only during the filing season, and 15 self-

service computer/printer stations (i.e., kiosks) that print tax forms 

and provide answers to frequently asked tax law questions.



[37] Field assistance officials estimated that return-preparation 

assistance accounted for only about 10 percent of their walk-in 

contacts but consumed about 25 percent of walk-in staff years. 



[38] IRS reported having 83 one-person sites during the 2002 filing 

season. 



[39] According to a cognizant IRS official, a tool that helps eliminate 

broken links was not added to the site in 2002 as IRS had originally 

intended; implementation is now planned for 2003.



[40] U.S. General Accounting Office, Tax Administration: Assessment of 

IRS’ 2000 Tax Filing Season, GAO-01-158 (Washington, D.C.: Dec. 22, 

2000).



[41] IRS updated this information as of September 30, 2002, in its 

written comments on a draft of this report (see app. VI).



[42] GAO-02-144 and GAO-03-143. 



[43] If someone accesses IRS’s site and moves around the site, the 

unique visitor measure would count that as one visit, no matter how 

many hits the person accumulates while navigating the site. If the 

person leaves the site and returns again after at least 1 hour has 

elapsed since exiting the site, the person will be counted as a new 

unique visitor.



[44] GAO-03-143.



[45] A rate reduction credit would reduce a taxpayer’s 2001 tax 

liability and could result in a tax refund once the 2001 return was 

filed.



[46] NAP is an IRS database with information on taxpayers’ names, 

addresses, and SSNs. To help determine whether taxpayers were correctly 

claiming the rate reduction credit on their 2001 tax returns, IRS 

included information on advance tax refunds in NAP.



[47] The Internal Revenue Service Certifies Its Readiness to Process 

Individual Income Tax Returns in 2002, TIGTA, Reference No. 2002-40-

080, Mar. 29, 2002.



[48] U.S. General Accounting Office, Tax Administration: Assessment of 

IRS’ 2001 Tax Filing Season, GAO-02-144 (Washington D.C.: Dec. 21, 

2001). 



[49] P.L. 104-168 



[50] Temp. Treas. Reg. § 301.6103(c)-IT (2001). 



[51] U.S. General Accounting Office, Tax Administration: Assessment of 

IRS’ 2001 Tax Filing Season, GAO-02-144 (Washington D.C.: Dec. 21, 

2001).



GAO’s Mission:



The General Accounting Office, the investigative arm of Congress, 

exists to support Congress in meeting its constitutional 

responsibilities and to help improve the performance and accountability 

of the federal government for the American people. GAO examines the use 

of public funds; evaluates federal programs and policies; and provides 

analyses, recommendations, and other assistance to help Congress make 

informed oversight, policy, and funding decisions. GAO’s commitment to 

good government is reflected in its core values of accountability, 

integrity, and reliability.



Obtaining Copies of GAO Reports and Testimony:



The fastest and easiest way to obtain copies of GAO documents at no 

cost is through the Internet. GAO’s Web site ( www.gao.gov ) contains 

abstracts and full-text files of current reports and testimony and an 

expanding archive of older products. The Web site features a search 

engine to help you locate documents using key words and phrases. You 

can print these documents in their entirety, including charts and other 

graphics.



Each day, GAO issues a list of newly released reports, testimony, and 

correspondence. GAO posts this list, known as “Today’s Reports,” on its 

Web site daily. The list contains links to the full-text document 

files. To have GAO e-mail this list to you every afternoon, go to 

www.gao.gov and select “Subscribe to daily E-mail alert for newly 

released products” under the GAO Reports heading.



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. General Accounting 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: www.gao.gov/fraudnet/fraudnet.htm E-mail: fraudnet@gao.gov



Automated answering system: (800) 424-5454 or (202) 512-7470:



Public Affairs:



Jeff Nelligan, managing director, NelliganJ@gao.gov (202) 512-4800 U.S.



General Accounting Office, 441 G Street NW, Room 7149 Washington, D.C.



20548: