Access to the Toll-Free Telephone System Was Significantly Improved in 2003, but Additional Enhancements Are Needed

 

January 2004

 

Reference Number: 2004-30-038

 

 

This report has cleared the Treasury Inspector General for Tax Administration disclosure review process and information determined to be restricted from public release has been redacted from this document.

 

January 26, 2004

 

 

MEMORANDUM FOR COMMISSIONER, SMALL BUSINESS/SELF-EMPLOYED DIVISION

                                         COMMISSIONER, WAGE AND INVESTMENT DIVISION

 

FROM:     Gordon C. Milbourn III /s/ Gordon C. Milbourn III

                 Acting Deputy Inspector General for Audit

 

SUBJECT:     Final Audit Report - Access to the Toll-Free Telephone System Was Significantly Improved in 2003, but Additional Enhancements Are Needed  (Audit # 200330007)

 

This report presents the results of our review of the Internal Revenue Service’s (IRS) ability to provide taxpayers with effective access to its toll-free telephone system during the 2003 Filing Season.   We performed the audit in response to the continuing concerns that the Congress, General Accounting Office, National Taxpayer Advocate, and Treasury Inspector General for Tax Administration have raised about the IRS’ ability to effectively meet the significant annual taxpayer demand for its toll-free telephone services.

In summary, many aspects of the taxpayer experience in accessing the IRS toll-free telephone system were significantly improved during the 2003 Filing Season. This improvement was reflected in the measures that the IRS uses to gauge the performance of its toll-free telephone system.  In comparison to the prior filing season, for example, the Customer Service Representative (CSR) Level of Service (LOS) increased, CSRs answered more calls, and fewer taxpayers abandoned (i.e., hung up) their calls before receiving assistance. Further, taxpayers that called with account- or refund-related questions had shorter wait times to receive service, and taxpayers that called with account-related questions were more likely to receive assistance when they reached a CSR assigned to an account application.

We attributed this improved performance to several factors.  First, the IRS has made many technological, organizational, and process changes over the past several years to improve taxpayer access to its toll-free telephone system.  These changes included grouping the CSRs into specialized skill areas and enhancing the skill-based call routing system, establishing six new toll-free numbers to further assist taxpayers in reaching the appropriate IRS resource, realigning call sites by customer segment, and increasing reliance on call screeners to route tax law calls to the appropriate assistor.

Second, customer demand for toll-free telephone services was down significantly in comparison to the prior filing season.  The reduced demand was likely attributable to increased electronic filing options for taxpayers and other IRS-initiated efforts to shift customer demand for assistance to other communication channels such as the Internet. Another possible reason for the reduced demand was the absence of major tax law changes affecting returns filed in 2003.

Although taxpayer access to its toll-free telephone services improved, the IRS has opportunities to further enhance the taxpayer experience and reduce the costs of providing toll-free telephone services.  A major improvement opportunity involves the continuing difficulties with initially routing calls to where taxpayers can be helped in an effective and efficient manner.  A technical call routing solution, the Customer Communications Project Fiscal Year 2001 Release (CC 2001), implemented in August 2001, was expected to eliminate much of the need for call screeners by automatically routing the calls to the proper application.  During the 2002 Filing Season, however, taxpayers were unable to understand the complex menu scripts and make the proper selections.  As a result, their calls often ended up in the wrong applications and had to be transferred from one CSR to another. 

For the 2003 Filing Season, the IRS used call screeners to send tax law calls to the appropriate applications.  While this solution significantly reduced the number of taxpayers that had their tax law calls transferred from one CSR to another, it caused all calls for tax law assistance to be transferred at least once (i.e., from the screener to a CSR).  In addition, using screeners to manually route calls is a more costly alternative than using automation.  We estimate that the IRS used nearly 109 additional Full-Time Equivalents (FTE) for screeners, at a cost of almost $3.6 million in salaries and benefits, that would not have been needed if the CC 2001 call routing solution had worked as planned. 

In 2003, the IRS also experienced routing problems for the three other categories of calls (i.e., account, procedural, and refund). Usability tests, completed by an independent contractor after new routing scripts were implemented, showed that the menu scripts were confusing and caused taxpayers to make wrong selections.

Another improvement opportunity involved reducing the high Assistor Availability levels that have existed for at least the past two filing seasons.  The IRS had planned for an Assistor Availability of 5.5 percent in Fiscal Year (FY) 2003.  During the 2003 Filing Season, however, the Assistor Availability was 11.2 percent, as compared to 10.8 percent in the prior filing season.  Through the end of June 2003, the Assistor Availability had further increased to 12.15 percent, as compared to 9.24 percent for the same period last year.  When the Assistor Availability is too high, the CSRs’ time is not being productively used.  We estimate that the cost of this additional (i.e., actual versus planned) availability, in CSR salaries and benefits, was almost $6.4 million.

Finally, the IRS still does not have a financial system that will accurately track its cost-per-call of providing various toll-free telephone services.  While some efforts have been made in this direction, none have been finalized.  Without accurate cost-per-call measurements for each of the various toll-free telephone services the IRS provides, management and key stakeholders do not have sufficient information with which to evaluate the efficiency of toll-free telephone system operations, the return on investment in technology and human capital, and the costs and benefits of key program decisions. Because the IRS plans to pilot test the feasibility of outsourcing some tax law calls during the 2004 Filing Season, accurate cost-per-call data are critical for comparative purposes. 

To address the call routing problems, we recommended that the Directors, Customer Account Services (CAS), in the Wage and Investment (W&I) and Small Business/Self-Employed (SB/SE) Divisions develop a set of key numerical identifiers that taxpayers would select to link their calls to the correct tax law application, revise the menu scripts for the tax law lines and submit them for independent usability testing prior to implementation, and establish separate toll-free numbers for general account calls and tax law calls.

To address the continued high Assistor Availability, we recommended that the Directors, CAS, in the W&I and SB/SE Divisions reevaluate, as recommended in our audit report on the 2002 Filing Season, whether some applications would benefit by being combined into a more pooled environment, determine those applications for which availability is highest and retrain the CSRs so they can be transferred as needed to those applications for which availability is lowest, and reduce either staffing or hours of operation during those times of the day when call demand is extremely low.

To ensure that IRS management and key stakeholders can effectively monitor and evaluate the efficiency of the IRS toll-free telephone system operations, we recommended that the Directors, CAS, in the W&I and SB/SE Divisions develop an Activity-Based Costing system that reliably captures and reports both the total cost and the cost-per-call of providing services on each toll-free product line.

Management’s Response:  The Commissioner, W&I Division, provided a detailed response to our draft report.  He disagreed with our recommendations to improve the call routing process.  He stated there were no substantiated data or documentation associated with the recommendation to develop a set of key numerical identifiers that taxpayers would select to link their calls to the correct tax law application to indicate this approach would be beneficial.  He added the Intelligent Contact Manager system used to route taxpayer calls has systemic limitations that would not support this initiative without an extensive upgrade.  The Commissioner also disagreed with our recommendation to revise the menu scripts for the general tax law line and Business and Specialty Tax Line and submit them for independent usability testing prior to implementation.  He stated that changes to the menu scripts implemented during the 2003 Filing Season improved performance drastically and that, when future menu changes are made, the IRS will modify its process to allow time for independent usability testing. The Commissioner disagreed with our recommendation to establish separate toll-free numbers for general account calls and tax law/procedural calls.  He stated that general account and tax law/procedural calls are closely related and it would be difficult for customers to differentiate between the two services.

The Commissioner, W&I Division, disagreed with our estimate that the use of call screeners had resulted in additional costs.  He stated the IRS’ analysis showed that, when compared to the prior year, the screener strategy resulted in a savings of over $1.1 million from January through June 2003.

The Commissioner, W&I Division, disagreed with our recommendations to reduce Assistor Availability.  He stated the specific recommended actions are already standard practices that are performed as part of the IRS’ ongoing evaluation, assessment, and planning processes.  The Commissioner also disagreed with our estimate of the costs associated with Assistor Availability during the 2003 Filing Season.

In part, the Commissioner, W&I Division, agreed with our recommendation to develop an Activity-Based Costing system.  He stated that the IRS Chief Financial Officer is implementing a comprehensive IRS-wide solution to modernize the accounting system and that, when it is successfully implemented, the IRS will develop costing information at the toll-free product line level, if sufficient detailed information is available for analysis.

Management’s complete response to the draft report is included as Appendix V.

Office of Audit Comment:  While the current call routing system may not be able to support the use of key numerical identifiers without costly upgrades, we believe the IRS must continue to pursue a more efficient automated alternative to the process of using call screeners to manually route incoming tax law calls. The IRS’ calculation of the cost benefits of its screener strategy incorrectly compares FY 2003 performance with that of FY 2002, when the CC 2001 call routing technology failed to route calls as intended.  Our estimate takes into consideration the impact the call routing technology should have had if it had been effective and the current annual cost of screeners as a result of the routing technology not working as anticipated.

Independent usability studies showed that the main complaint about the IRS toll-free telephone system was the menu scripts. The IRS provided no compelling support to show that changes to the menu scripts had any effect on improved performance during the 2003 Filing Season.  However, even if it is assumed that they had a positive effect, it is clear that additional improvements are still needed.  In fact, the W&I Division’s Program Letter for 2004 lists reducing the call transfer rate as one of its challenges for 2004.

We continue to believe our recommendation to provide separate toll-free telephone numbers for general account calls and tax law/procedural calls has merit.  With the proper customer education on the separate numbers, we believe taxpayers would enjoy less cumbersome menu scripts and experience fewer routing problems. 

We are encouraged that the IRS assesses, on an on-going basis, whether applications would benefit by a more pooled environment.  Continued attention towards pooling work groups where possible can lead to greater efficiencies.  The IRS did not have readily accessible data on availability by application for the 2003 Filing Season, although it has requested that these data be added to its management information system in FY 2004.  Once these data become available, we will assess the effectiveness of the IRS’ actions in this area.  We continue to believe the IRS needs to address the large difference between its planned availability and actual availability. In our report on the 2002 Filing Season, we recommended that the IRS explore ways to make the forecast more accurate.  The IRS disagreed with that recommendation, stating its methodology is an accepted industry practice that has been used to successfully forecast demand.  However, the variance in planned and actual Assistor Availability in FY 2003 shows this is not always the case.

During the course of this audit, we were informed that call demand was extremely low between 6 and 7 a.m. Central Time (CT) and somewhat higher, but still low, between 7 and 8:30 a.m.  Demand was also low after 8 p.m. Pacific Time.  We still believe the IRS should consider staffing adjustments to these time periods to reduce availability.  Furthermore, despite the IRS’ disagreement with this recommendation, the W&I Division’s FY 2004 Customer Account Services Work Plan – Planning Assumptions states, “As we have increased service levels, customer demand shifted away from evening hours and weekends.  To the extent possible, staffing patterns will mirror demand, resulting in a shift of staffing to the Monday through Friday core hours of 9:00 to 20:00 CT.”

While we still believe that all of our recommendations are worthwhile, we do not intend to elevate our disagreement concerning these matters to the Department of the Treasury for resolution.

Copies of this report are also being sent to the IRS managers who are affected by the report recommendations.  Please contact me at (202) 622-6510 if you have questions or Philip Shropshire, Acting Assistant Inspector General for Audit (Small Business and Corporate Programs), at (215) 516-2341.

 

Table of Contents

Background

Taxpayers Experienced Improved Access to the Toll-Free Telephone System During the 2003 Filing Season

Call Routing Problems Continue to Add to Taxpayer Burden and the Costs of Providing Toll-Free Telephone Services

Recommendation 1:

Recommendations 2 and 3:

Continuing High Assistor Availability Levels Are Adding to the Costs of Providing Toll-Free Telephone Services

Recommendations 4 and 5:

Recommendation 6:

An Activity-Based Costing System Is Needed to Evaluate the Efficiency of the Toll-Free Telephone System Operations

Recommendation 7:

Appendix I – Detailed Objective, Scope, and Methodology

Appendix II – Major Contributors to This Report

Appendix III – Report Distribution List

Appendix IV – Outcome Measures

Appendix V – Management’s Response to the Draft Report

 

Background

The Internal Revenue Service’s (IRS) toll-free telephone system is the contact method most taxpayers choose when seeking answers to tax law questions or trying to resolve tax account-related issues.  Taxpayers called the IRS toll-free telephone system over 50 million times during the 2003 Filing Season.  The IRS’ strategy for handling this significant customer demand is to direct those taxpayers with less complicated issues to its automated services (i.e., recorded information and interactive applications)  and allow its Customer Service Representatives (CSR) to assist taxpayers with more difficult issues.

The Congress, General Accounting Office, National Taxpayer Advocate, and Treasury Inspector General for Tax Administration (TIGTA) have raised continuing concerns about the IRS’ ability to effectively meet the significant annual taxpayer demand for access to its toll-free telephone system.  Over the past several years, the IRS has made many technological changes, as well as organizational and process changes, to its toll-free telephone system in an effort to provide taxpayers with better access and improve the quality of its service.  In addition to substantial efforts to improve the performance of its toll-free telephone system operations, the IRS has taken other actions that should reduce taxpayer demand for this type of service.  These actions include implementing an Electronic Tax Law Assistance program in which taxpayers can ask the IRS tax law questions via the Internet, enhancing the IRS web site to provide many types of information to taxpayers including the ability to inquire about the status of their refunds, and increasing electronic filing options.

We performed this review because, in spite of these positive actions taken by the IRS, criticisms of its toll-free telephone services were raised regarding the 2002 Filing Season. For example, an October 2002 assessment of the IRS telephone services by the National Society of Accountants stated that the IRS, “must increase first-call resolutions, decrease waiting times, and lower the abandon rates.”  In addition, in its April 2003 Annual Report to the President and the Congress, the IRS Oversight Board stated that while access to the toll-free telephone system had improved, it was not at a level that was acceptable to the Board because only 7 out of 10 callers were assisted in Fiscal Year (FY) 2002.

To perform this audit, we monitored the performance of the IRS toll-free telephone system during the 2003 Filing Season.  The audit was conducted at the IRS Joint Operations Center (JOC) in Atlanta, Georgia, from January through June 2003 in accordance with Government Auditing Standards.

During the audit, we identified opportunities for the IRS to improve access to its toll-free TTY/TDD services for hearing- and speech-impaired taxpayers. These services are provided on a separate toll-free line that uses special equipment allowing users to communicate with the IRS through text messages that each party alternately sends via a keyboard.  Because the demand for these services is extremely small in comparison to the demand for services on other IRS toll-free lines, a separate audit report was issued to present the issues in proper perspective.

Detailed information on our audit objective, scope, and methodology is presented in Appendix I. Major contributors to the report are listed in Appendix II.

Taxpayers Experienced Improved Access to the Toll-Free Telephone System During the 2003 Filing Season

Many aspects of the taxpayer experience in accessing the IRS toll-free telephone system were significantly improved during the 2003 Filing Season.  This improvement is reflected in many of the measures the IRS uses to gauge the performance of its toll-free telephone system.  For example, Table 1 shows that CSRs answered more calls in FY 2003 and fewer taxpayers abandoned their calls before reaching a CSR.

Table 1:  More Taxpayers Were Helped by CSRs and Fewer Taxpayers Hung Up Before Receiving Assistance

Filing Season

Calls Answered
by CSRs

Primary Abandons

Secondary Abandons

2003

13,788,464

19.3%

 11.7%

2002

13,744,104

26.7%

 18.1%

Difference

       44,360

( 7.4%)

(  6.4%)

Source:  Weekly Enterprise Snapshot Reports for April 20, 2002, and April 19, 2003.

 

Taxpayers that called the IRS with account- or refund-related questions were also served more quickly in the 2003 Filing Season.  In comparison to the prior filing season, the Average Speed of Answer (ASA) for account calls decreased from 273 seconds to 142 seconds, and the ASA for refund calls decreased from 315 seconds to 79 seconds.  In concert with the ASA, the Assistor Response Level (ARL) increased from 50.7 percent to 56.6 percent for account calls and from 35.7 percent to 69.2 percent for refund calls.  Thus, for these last two categories, wait time was reduced for a greater percentage of callers.

According to data received from the IRS, the CSR Level of Service (LOS), the IRS’ primary measure of providing callers with access to a live assistor, increased from 69.6 percent during the 2002 Filing Season to 83.2 percent in the 2003 Filing Season. For the specific customer service lines that we reviewed, the CSR LOS for 2003 ranged from 80.4 percent to 91.9 percent.

Despite our continued concern with the IRS’ formula for computing the CSR LOS, we concur that access to the toll-free telephone system was significantly improved for the 2003 Filing Season.  For example, we computed an LOS measure by dividing the number of calls that were answered by the number of net call attempts.  The result showed an increase of 12.9 percent, from a 56.5 percent LOS in 2002 to a 69.4 percent LOS in 2003.

The IRS implemented new initiatives to improve access

For 2003, the IRS implemented a number of initiatives to improve the taxpayer experience when calling the toll-free telephone system.  Some of those initiatives identified by the IRS are:

·        Customer Contact Center Optimization (CCCO) ProjectThe CCCO Project resulted in changes, implemented in October 2002, to the toll-free telephone system that were designed to support increased specialization and ensure that callers reached an assistor with the appropriate skills to help them. This involved grouping the CSRs into specialized skill areas and implementing a skill-based routing process that matches each caller with a CSR who has the specialized skills needed to handle the call.

·        IRS Toll-Free Vision and Marketing Strategy – This strategy involved the implementation of six new toll-free numbers in December 2002.  The new numbers were intended to ensure that callers are directed to the appropriate resource through the use of scripts that are tailored to each number.  These changes are part of a new toll-free telephone system strategy to improve customer service by targeting customer segments and creating more accountability. Two of the new numbers (the Refund Hotline and the Business and Specialty Tax Line) were advertised in public news releases.  The other four are Customer Response Lines for the numbers that appear on taxpayer account notices sent to taxpayers when they owe taxes or have not filed returns.

·        Call Site and Workforce Specialization – Five call sites were realigned from the Small Business/Self-Employed (SB/SE) Division to the Wage and Investment (W&I) Division in October 2002.  The two Divisions assumed individual ownership of each of the six new toll-free numbers.  This initiative contributed to further site and workload specialization.

·        Enhancements to Automated Platforms – The Telephone Routing Interactive System was revised to reduce multiple call attempts by providing additional information, such as the anticipated refund delivery date, in lieu of referring the caller to another number.

·        Screener Strategy – This strategy was implemented in October 2002 to address the problems with the complex automated call routing menu scripts that had caused high levels of abandoned and transferred calls during the 2002 Filing Season.  Prior to the 2002 Filing Season, screeners were used to route many categories of calls, including tax law.  However, in 2002, the IRS attempted to automate the routing for tax law calls. This effort was unsuccessful, and the IRS reverted to using specially trained CSRs during the 2003 Filing Season to screen incoming tax law calls to the primary toll-free line (800-829-1040, or the 1040 line).  The screener who answered the call determined the taxpayer’s specific question or issue and then transferred the call to the appropriate application.

The Screener Strategy significantly decreased the number of tax law calls that were transferred once they reached an application. However, we have concerns with the efficiency of this practice that are addressed later in this report.

Decreased demand contributed to the improved access

In addition to those initiatives mentioned above, the IRS took several significant actions that likely contributed to the decreased call demand during the 2003 Filing Season.  The substantial decrease in call attempts was a factor in the IRS’ ability to provide taxpayers with improved access to the toll-free telephone system during the 2003 Filing Season.  Net call attempts, including calls to the automated TeleTax system, were down by more than 30.5 million (38 percent) in comparison to the 2002 Filing Season.  Excluding the TeleTax system calls, net call attempts to the 8 main toll-free customer service lines decreased by 12.9 million (34 percent).

Actions taken by the IRS included:

·        An Internet Refund Fact of Filing (IRFOF) service was implemented that allowed taxpayers to check the status of their tax refunds by accessing the IRS web site (www.irs.gov).  During the 2003 Filing Season, 12.4 million taxpayers accessed the IRFOF application, and 8.8 million taxpayers successfully completed and received refund status information. When taxpayers use the Internet to check the status of their refunds, it is anticipated that they will be less likely to also check the status by calling the IRS toll-free telephone system.

·        The IRS has taken a number of actions in recent years to encourage the electronic filing of tax returns.  In 2003, taxpayers could use the IRS Internet web site to access commercial tax preparation software and electronic filing services at no cost.  Table 2 shows that the number of practitioner-filed electronic returns increased by more than 10 percent during the 2003 Filing Season and that the number of tax returns filed by home computer increased by almost 27 percent. An internal study by the IRS has shown that practitioner-prepared returns result in far fewer calls to the IRS than self-prepared returns.  Because the proper handling of many tax law issues is programmed into tax preparation software packages, it is also probable that taxpayers using electronic filing methods would be less likely to call the IRS with tax law questions.

Table 2:  Electronic Filing Increased for the2003 Filing Season

Type

2002

2003

Increase

(number)

Change

(percent)

Practitioner

32,476,000

35,783,000

3,307,000

10.2%

Home Computer

 

 9,145,000

 

11,583,000

 

2,438,000

 

26.7%

Total

41,621,000

47,366,000

5,745,000

13.8%

Source:  IRS Cumulative Filing Season Statistics, April 18, 2003.

·        When taxpayers are able to successfully access the IRS toll-free telephone system on their first attempt, they are less likely to call back, thus reducing the total call attempts. Table 3 shows that the changes made to the toll-free telephone system resulted in about 2.3 million fewer “blocked” calls during the 2003 Filing Season. 

Table 3:  Fewer Calls Were Blocked During the 2003 Filing Season

 

2002

2003

Decrease

(number)

Change

(percent)

Total Blocked Calls

 

3,413,644

 

1,072,850

 

2,340,794

 

-68.7%

Source: IRS Snapshot Table.

While it is not possible to definitively identify all of the reasons for the significant decrease in customer demand for toll-free telephone services, another probable factor was the relatively few major tax law changes affecting returns filed in 2003.  Significant tax law changes can cause large increases in taxpayer demand for toll-free telephone services.  In the 2002 Filing Season, for example, the IRS attributed part of the problem that taxpayers experienced in getting through to the toll-free telephone system to the influx of calls received concerning the changes to the tax forms that were necessitated by the Rate Reduction Credit.

Access for Spanish-speaking taxpayers improved in 2003

The IRS successfully continued its efforts to improve the access for Spanish-speaking taxpayers to toll-free telephone services.  For the 2003 Filing Season, the IRS increased its hiring of bi-lingual (i.e., English- and Spanish-speaking) CSRs by approximately 10 percent.  The number of services provided to Spanish-speaking taxpayers increased from 725,188 in the 2002 Filing Season to 773,833 during the 2003 Filing Season – an increase of 6.7 percent.  An additional indication of the improved service to Spanish-speaking taxpayers is that the number of calls transferred decreased. When screener-transferred calls are omitted, the number of calls from Spanish-speaking taxpayers that were transferred from 1 CSR to another decreased by 18 percent, from 123,836 during the 2002 Filing Season to 101,581 during the 2003 Filing Season.

The IRS has already made plans to further improve access for Spanish-speaking taxpayers.  Of the approximately 400 new CSRs that the IRS plans to hire for FY 2004, 270 to 280 will be bi-lingual.  At those call sites that handle calls from Spanish-speaking taxpayers, all of the newly hired CSRs will be bi-lingual.

Customer satisfaction surveys showed positive results

The W&I and SB/SE Divisions conduct separate customer satisfaction surveys on a quarterly basis. The SB/SE Division uses an outside contractor to conduct the surveys and prepare the survey reports.  Beginning with the first quarter in FY 2003, the W&I Division began producing its own reports using the same data source.

On January 1, 2003, a new customer satisfaction survey was implemented for both Divisions.  In addition to containing different questions, the survey also had a change in the satisfaction rating scale, from a four-point to five-point scale.  Consequently, comparisons cannot be made between the customer satisfaction ratings for 2003 and the ratings for previous years.  The contractor will be converting prior surveys to a five-point scale so that comparisons can be made.

For January through March 2003, the vast majority of taxpayers that completed customer satisfaction surveys gave the IRS a rating of 4 (“mostly satisfied”) or 5 (“completely satisfied”) regarding their experience in using the IRS toll-free telephone system. Although these ratings cannot be compared with those from prior quarters, the current results are impressive. For example, 95 percent of the callers surveyed for the W&I Division gave the IRS a rating of 4 or 5. Similarly, 92 percent of the callers surveyed for the SB/SE Division gave those same ratings.

Although many continuing improvements have been made to the toll-free telephone system, the IRS still has opportunities to further enhance the taxpayer experience and reduce the costs of providing toll-free telephone services.

Call Routing Problems Continue to Add to Taxpayer Burden and the Costs of Providing Toll-Free Telephone Services

The Customer Communications Project Fiscal Year 2001 Release (CC 2001), which became operational in August 2001, was intended to increase the IRS’ service levels to those of similar customer service operations in the private sector.  One of the major system improvements included in the CC 2001 was the capability to more accurately route taxpayer calls to the most appropriate IRS resource.  With this improved capability, the IRS expected that taxpayers would have to spend less time speaking with screeners that help transfer calls to the right CSR and that, because of the improved menu system, callers could be directed to the right CSR in a shorter time. 

The CC 2001 call routing solution was not successful in ensuring that calls were routed to the correct applications.  During the 2002 Filing Season, for example, 29 percent of the calls to the IRS were transferred after they were routed to applications.  The highest transfer rate (63 percent) involved tax law calls.  By comparison, 17 percent of the calls to the IRS were transferred during the filing season prior to the deployment of the CC 2001 call routing solution. 

Although the CC 2001 call routing solution had been tested to assure that calls arrived at the correct application given the menu selection made by the caller, taxpayers were often unable to understand the complex menu scripts and make the proper selections.  As a result, their calls often ended up in the wrong applications.  These call routing problems created additional costs for the IRS.

While the IRS made changes to the menu scripts early in the 2002 Filing Season that helped to address the call transfer problem, the changes had limited success. Consequently, the IRS recognized the need for longer-term solutions and, prior to the 2003 Filing Season, implemented three specific actions to reduce the customer burden caused by unnecessary call transfers.  These actions included the introduction of six new toll-free numbers, enhanced skill-based routing, and the use of CSRs as screeners for tax law calls.

Even with these changes, call transfers remain a problem for the IRS.  During the 2003 Filing Season, calls routed to 27 of the 45 applications in the tax law, procedural, and account categories were transferred at rates ranging from 15.7 percent to 67.8 percent.  IRS management believes that a 15 percent transfer rate is an appropriate target for the agency to meet world-class standards. 

The use of call screeners reduced transfers in the majority of tax law applications

In lieu of having taxpayers navigate a complex menu script, the IRS used screeners for tax law calls received on the 1040 line during the 2003 Filing Season.  In comparison to the prior year, this change resulted in far fewer taxpayers having their calls transferred from one CSR to another after they reached a tax law application.  In 2003, for example, the call transfer rates decreased, sometimes significantly, for 18 of the 21 tax law applications for individual taxpayers. However, the transfer rates actually increased for the remaining 3 tax law applications and remained higher than 15 percent in many of the other tax law applications that showed a decrease from a year ago.  For example, 1 tax law application still had a transfer rate of more than 40 percent, and another had a transfer rate of nearly 35 percent.

While the use of call screeners for tax law calls during the 2003 Filing Season may have been the only viable short-term solution to the high transfer rates experienced during the prior filing season, it resulted in all calls for tax law assistance being transferred at least once (i.e., from a screener to a CSR) whereas, in the 2002 Filing Season, 37 percent of the taxpayers calling for tax law assistance experienced no transfers.  In addition, the use of screeners to route calls is a more costly alternative than the use of automation.  In the long term, automated routing is the only way to significantly reduce these costs.

Routing problems were experienced with other types of calls to the 1040 line

Even though screeners were used for tax law calls, all taxpayers that called the 1040 line had to initially respond to an automated menu script.  If the tax law callers responded correctly to the first two menu options, they were connected to a screener.  Other callers to this line (i.e., those with account, procedural, or refund questions) were prompted through an automated script that was designed to either provide them information through automation or send them to a CSR for assistance.

During the 2003 Filing Season, however, callers with account, procedural, or refund questions frequently selected menu options that sent them to screeners rather than directly to the appropriate applications.  Table 4 shows that nearly one-half of the English calls and more than 70 percent of the Spanish calls transferred by the screeners did not go to tax law applications.

Table 4:  Calls Transferred by Screeners to Applications Other Than Tax Law

Language

Account

Procedural

Refund

Total

English

34.7%

12.4%

0.4%

47.5%

Spanish

61.1%

  9.8%

2.4%

73.3%

Source:  IRS Agent Transfer Reports.

The IRS did not have an independent usability test conducted prior to implementing the new routing scripts for the 1040 line.  A subsequent usability test completed by a contractor identified problems with the routing scripts that could cause taxpayers to make wrong menu selections.  Unfortunately, taxpayers that realized they had made a wrong selection could not back up in the system.

Some business taxpayers also experienced routing problems

Taxpayers that called the Business and Specialty Tax Line (800-829-4933, or the 4933 line) during the 2003 Filing Season navigated through automated menu scripts rather than speaking with call screeners. All calls (i.e., tax law, procedural, and account) to this line were automatically routed to applications based on the menu selections made by the callers. 

While there is not a definitive relationship between this dialed number and the 8 business and specialty applications, the IRS advised us that this dialed number likely handled a high percentage of these applications.  Furthermore, it is clear that, in some instances, far too many taxpayers arrived at incorrect applications that required their calls to be transferred. For example, the transfer rates for 3 of the 8 business and specialty applications exceeded 15 percent, including 2 applications that had transfer rates of 54.2 percent and 45.6 percent. The number of calls handled in these 3 applications represented about 30 percent of the total calls handled in the 8 business and specialty applications; however, the 2 applications with the highest transfer rates handled only 11.2 percent of these total calls. 

In January 2003, the IRS used a contractor to conduct a usability test on the Business and Specialty Tax Line.  The contractor’s report included the following overall analysis:

Most participants had significant problems interacting with the system.  The majority of the problems stemmed from the difficulty of presenting, in a clear and intuitively organized manner, the large number of options corresponding to the tax issues that this system deals with.

Call routing problems add to the costs of providing toll-free telephone services and create taxpayer burden

A very important effect of the IRS’ decision to revert to the use of screeners to route tax law calls in 2003 is that any improvement in accurately routing the calls came at a significant cost.  Using the IRS’ planning data for the 2002 Filing Season, when the CC 2001 call routing solution was expected to eliminate much of the need for screeners by automatically routing calls, we calculated the percentage of planned call screener services to total planned services.  We then applied that percentage to the actual services for FY 2003, through June 2003, to determine the baseline screener services that should have occurred had the CC 2001 call routing solution worked as planned.  We estimate that the IRS used nearly 109 additional Full-Time Equivalents (FTE) for screeners in FY 2003 as a result of the CC 2001 call routing solution problems.  The minimum cost for these additional FTEs, through June 2003, was $3.6 million.

Beyond the costs of hiring screeners, call routing problems add other costs to providing toll-free telephone services.  When calls are misrouted and must be transferred, CSR talk time is increased because the taxpayers have to repeat their questions.  Thus, the IRS’ cost to handle a call is increased when the call is transferred, rather than completed, by the first CSR it was sent to. Similarly, call transfers also result in taxpayers having to spend more time on the telephone that, in turn, can lead to additional burden.  Additionally, using more screeners in 2003 reduced the number of CSRs that were available to directly take tax law calls.

The effect of the menu scripting problems on increased taxpayer burden was evident from the usability study report for the 1040 line.  The study, performed by a contractor in January 2003 using a group of participants, found that problems with the “menu options” were, by far, the main complaint about the system.  According to the study report, the participants found some of the verbiage unclear, confusing, ambiguous, and not descriptive enough; the menus seemed to lack options to satisfy some of the callers’ inquiries; and the organization of the options seemed to be unclear, resulting in the callers not being able to find what they needed and many times having to guess which option to choose. The study report added that, even when they had made the correct selection, most participants did not feel confident they were on the right track to get the help they needed.

A potential effect of complex menu scripts is that they can cause taxpayers to abandon (i.e., hang up) their calls before receiving assistance.  This can lead to taxpayers having to call back or possibly not receive assistance because they find the toll-free telephone system too difficult to use.  When taxpayers have these types of difficulties in attempting to contact the IRS, the IRS’ strategic goal of providing top-quality service to each taxpayer in each interaction is not being met. Consequently, the continued improvement of the menu scripts is necessary.

An effective solution to the routing problem is needed

Despite the many actions taken by the IRS to overcome the call routing problems caused by the ineffectiveness of the CC 2001 call routing solution, this remains a major area for improvement.  While improving the menu scripts has been necessary for solving a portion of the routing problems, the revisions have not solved all of the routing problems, and there continue to be high costs associated with excessive call transfers in some applications.  A more innovative approach is clearly called for since traditional solutions have not been effective in solving the root problem.

Recommendations

To address the call routing problems, the Directors, Customer Account Services (CAS), SB/SE and W&I Divisions, should:

1.      Develop a set of key numerical identifiers that taxpayers would select to link their calls to the correct tax law application.  One way of doing this would be to use topic numbers similar to those used in the IRS automated TeleTax system.

Management’s Response: The Commissioner, W&I Division, disagreed with this recommendation and stated there were no substantiated data or documentation associated with the recommendation to indicate this approach would be beneficial for tax law.  He added the Intelligent Contact Manager system used to route taxpayer calls has systemic limitations that would not support this initiative without an extensive upgrade.

The Commissioner also disagreed with our estimate that the use of call screeners during the 2003 Filing Season resulted in additional costs of $3.6 million.  He stated the IRS’ analysis showed that, when compared with the prior year, the screener strategy resulted in a savings of over $1.1 million from January through June 2003.

Office of Audit Comment:  While the current call routing system may not be able to support the use of key numerical identifiers without costly upgrades, we believe the IRS must continue to pursue a more efficient automated alternative to using call screeners to manually route incoming tax law calls.  The IRS has already established the effectiveness of a numerical identification system through its automated TeleTax system.  In more than one instance, the IRS has also identified the importance of automation and changing the way it conducts business.  For example, an IRS study stated that, “The most cost-effective activity is the one that is most automated and therefore cheapest because that would have the highest marginal indirect revenue to cost ratio.”  In another example, one of the operational priorities identified in the W&I Division’s FY 2005 Strategic Assessment is to “Identify less costly methods of service delivery to reduce customer dependence on traditional methods of delivery, improve quality, and utilize resources more efficiently.”  

In our judgment, the IRS’ calculation of the cost benefits of its screener strategy is inflated because it compares FY 2003 performance with that of FY 2002, when the CC 2001 call routing technology failed to route calls as intended.  As stated in our report on the 2002 Filing Season, CSRs answered 1.73 million fewer calls in 2002 than in 2001 and call transfers increased dramatically.  Our estimate of the additional costs associated with the call screening strategy takes into consideration the impact the call routing technology should have had if it had been effective and the current annual cost of screeners as a result of the routing technology not working as anticipated.

2.      Revise the menu scripts for the 1040 and 4933 lines and submit them for independent usability testing prior to implementation.

Management’s Response: The Commissioner, W&I Division, disagreed with this recommendation.  He stated that changes to the menu scripts implemented during the 2003 Filing Season improved performance drastically.  He stated that, when future menu changes are made, the IRS will modify its process to allow time for independent usability testing.

Office of Audit Comment:  Independent usability studies showed that the main complaint about the IRS toll-free telephone system was with the menu scripts.  The IRS provided no compelling support to show that changes to the menu scripts had any effect on improved performance during the 2003 Filing Season. However, even if it is assumed they had a positive effect, it is clear from the very high transfer rates in some applications that additional improvements are still needed.  In fact, the W&I Division’s Program Letter for 2004 lists reducing the call transfer rate as one of its challenges for 2004.

3.      Implement separate toll-free numbers for general account calls and tax law/procedural calls.

Management’s Response: The Commissioner, W&I Division, disagreed with this recommendation.  He stated that general account and tax law/procedural calls are closely related and it would be difficult for customers to differentiate between the two services.  He added that the IRS had considered a similar idea as part of the Toll-Free Marketing Vision and Strategy but decided not to adopt it.  He said the IRS believes retaining the name recognition associated with the 1040 number to be more customer-focused.

Office of Audit Comment:  We continue to believe this recommendation has merit.  The IRS implemented six new toll-free numbers in December 2002 to improve customer service by targeting customer segments.  With the proper customer education on the separate numbers for account and tax law/procedural calls, we believe taxpayers would be exposed to less cumbersome menu scripts and, as a result, would experience fewer routing problems.

Continuing High Assistor Availability Levels Are Adding to the Costs of Providing Toll-Free Telephone Services

Assistor Availability is the measure the IRS uses to calculate how long its CSRs are available to take calls when none are coming in for their specific applications. Achieving the optimum Assistor Availability level is critical for effective and efficient call site operations. When the Assistor Availability is too low, taxpayers calling for assistance may have to wait longer to speak with a CSR or, frustrated with the waiting time, may abandon their calls and call back at another time.  When the Assistor Availability is too high, the CSRs’ time is not being productively used. Thus, funds are being spent that could be better used to provide additional services to taxpayers.

In an audit of the 2002 Filing Season, we reported that the Assistor Availability was high and recommended IRS management review its decision to specialize the CSR workforce to determine whether some applications would benefit from a more pooled environment. We also recommended that IRS management find ways to make the forecasting of call demand more accurate and explore ways to achieve greater flexibility in scheduling the CSRs to meet the call demand. In response to these recommendations, management committed to a number of alternative actions that included establishing additional toll-free numbers, enhancing the skill-based routing, providing screeners to route tax law calls, and developing a training strategy for 2003 that would increase their flexibility to move employees.  For example, all CSRs would be trained in a primary specialty and a back-up application.

Despite these actions, the overall Assistor Availability rate increased to 11.2 percent during the 2003 Filing Season, from 10.8 percent in the prior filing season.  For FY 2003 through the end of June 2003, the overall Assistor Availability rate had further increased to 12.15 percent, as compared to 9.24 percent for the same period last year.  The IRS had planned for an Assistor Availability rate of only 5.5 percent in FY 2003.

If the IRS had achieved its planned Assistor Availability rate through the end of June 2003, CSRs would have spent 401,355 fewer hours waiting for calls to arrive.  We estimated that the cost of this additional (i.e., actual versus planned) availability in CSR salaries and benefits was almost $6.4 million.

Through June 2003, the Assistor Availability rate ranged from a low of 4.9 percent at 1 call site to a high of 18.4 percent at another site.  The difference in Assistor Availability among call sites is likely related to the types of applications that were handled at the various sites and to whether there was alternate work for the CSRs to perform when calls were not being received as expected.

The increased Assistor Availability rate in 2003 was likely caused, at least in part, by the decrease in call demand.  In particular, call demand was very low early in the morning and late in the evening. Assistor Availability rates are also affected by the accuracy of the forecasted call demand (which involves the difficult task of predicting the volume of each type of call by arrival time in half-hour intervals) and by the effectiveness of the scheduling process (i.e., assigning CSRs to specific applications by half-hour intervals to meet the incoming call demand).

The increased workforce specialization at the IRS call sites has made forecasting and scheduling accuracy especially critical. Throughout the call site industry, a characteristic of the best-managed centers is that they find the right balance between specialization and pooling and that they get as close as possible to the pooled end of the spectrum, to improve efficiency.  A potential benefit of pooling is a lower Assistor Availability rate.

Recommendations

To address the continued high Assistor Availability rate, the Directors, CAS, SB/SE and W&I Divisions, should:

4.      Reevaluate, as recommended in our audit report on the 2002 Filing Season, whether some applications would benefit by being combined into a more pooled environment.

Management’s Response:  The Commissioner, W&I Division, disagreed with this recommendation.  He stated that assessments of the effectiveness of applications and agent groups are conducted on an ongoing basis.

Office of Audit Comment:  We are encouraged that the IRS assesses, on an ongoing basis, whether applications would benefit from a more pooled environment.  Continued attention towards pooling work groups where possible can lead to greater efficiencies. 

5.      Determine those applications for which availability is highest and retrain the CSRs so they can be transferred as needed to those applications for which availability is lowest.

Management’s Response:  The Commissioner, W&I Division, disagreed and stated that the recommended process already occurs on an ongoing basis.  He added that, in FY 2003, the IRS implemented skill-based routing that enabled it to designate backup agent groups for each application.  He stated that variable business rules trigger the use of backup agent groups, minimizing their availability and improving overall service.  He added that the use of skill-based routing, coupled with the increased level of account-trained CSRs in FY 2003, contributed to improved service.

The Commissioner also disagreed with our estimate of the costs associated with Assistor Availability during the 2003 Filing Season.  He stated that our comparison of planned availability to actual availability was flawed because planned availability was not a goal or a measure but a planning assumption used with other assumptions to formulate staffing schedules.  Furthermore, he stated that the IRS’ research of industry benchmarks revealed an availability benchmark of 17 percent, while the IRS availability for the 2003 Filing Season was 11.2 percent.

Office of Audit Comment:  The IRS did not have readily accessible data on availability by application for the 2003 Filing Season, although it has requested that these data be added to its management information system in FY 2004.  Once these data become available, we will assess the effectiveness of the IRS actions in this area.

We continue to believe the IRS needs to address the large difference between its planned availability and actual availability.  In our report on the 2002 Filing Season, we recommended that the IRS explore ways to make the forecast more accurate.  In disagreeing with that recommendation, IRS management stated its methodology is an accepted industry practice that has been used to successfully forecast demand.  However, the variance in planned and actual Assistor Availability in FY 2003 shows this is not always the case.

The IRS was unable to provide us with a copy of the document from which it quoted a 17 percent availability benchmark.  Thus, we cannot comment on the applicability of this benchmark to IRS operations.  However, the IRS FY 2004 SB/SE CAS Accounts Management Program Letter and Measures states that, “While the Agent Availability Plan for the year is 8 percent, sites should strive to minimize the amount of agent availability.”  In addition, the W&I Division’s Strategy and Program Plan for 2004 states that, “Additional refinements in how applications are grouped will reduce the number of transfers, improve customer satisfaction, and reduce the amount of available time.”  These statements clearly indicate that availability is a concern of IRS management.

6.      Reduce either staffing or hours of operation during those times of the day when call demand is extremely low.

Management’s Response:  The Commissioner, W&I Division, disagreed with this recommendation and stated this action is already done on an as needed basis.  He also stated site hours, staffing and intra-day calling patterns are evaluated each year and appropriate changes are made based on the analysis.  In addition, daily real-time adjustments to staffing are made to deal with changes in call demand and Assistor Availability as needed.

Office of Audit Comment:  In meeting with IRS managers during the course of this review, we were informed that call demand was extremely low between 6 a.m. and 7 a.m. Central Time (CT) and somewhat higher, but still low, between 7 a.m. and 8:30 a.m.  It was also low after 8 p.m. Pacific Time.  We still believe the IRS should consider adjustments to these time periods to reduce availability.  Furthermore, despite the IRS’ disagreement with this recommendation, the W&I Division’s FY 2004 Customer Account Services Work Plan – Planning Assumptions states, “As we have increased service levels, customer demand shifted away from evening hours and weekends.  To the extent possible, staffing patterns will mirror demand, resulting in a shift of staffing to the Monday through Friday core hours of 9:00 to 20:00 CT.”

An Activity-Based Costing System Is Needed to Evaluate the Efficiency of the Toll-Free Telephone System Operations

While the IRS has done some cost analyses of its toll-free telephone system operations, it has yet to develop accurate cost-per-call data.  Developing cost-per-call data is a difficult process for the IRS because it does not have a financial system that will track program costs; however, management has initiated action to obtain a new financial system, scheduled for implementation in FY 2004, which will track program costs.

An August 2001 TIGTA report recommended that the IRS develop a cost-per-call indicator.  In responding to this report, the IRS provided information showing that the total cost-per-call answered for FY 2000 was $5.94 and that the projected cost-per-call for FY 2001 was $5.47. The IRS had based its cost-per-call computation for FY 2001 on total toll-free telephone system operating costs of $514.2 million.  However, another TIGTA report, also issued in August 2001, found that, based on information largely provided by the IRS, the total toll-free telephone system costs for FY 2001 would be at least $627 million.

In October 2002, the IRS expanded and updated the cost analysis contained in the second TIGTA report referred to in the preceding paragraph.  The IRS data, which had not been finalized at the time we completed our current audit, estimated the toll-free telephone system costs for FY 2002 at more than $690 million.  This estimate was further broken down into a cost-per-call of $22.09 for calls answered by CSRs and a combined cost-per-call of $13.13 for all calls answered (i.e., either by CSRs or automated services).  However, the IRS’ cost-per-call estimate for calls answered by CSRs included some costs that are applicable only to calls answered by automated services.

These examples show that further refinement is needed to provide stakeholders, such as the Congress and the IRS Oversight Board, with more accurate cost-per-call information. In addition to not having accurate or separate cost-per-call data for calls answered by CSRs and calls answered by automated services, the IRS has not attempted to determine the costs of providing toll-free telephone services on its various product lines (i.e., tax law versus account calls) or the costs of providing toll-free telephone services to its customer segments (i.e., wage earner/investor versus small business/self-employed).  Best practices in the call site industry suggest that cost-per-call data should differentiate each type of contact and each type of service provided.

Federal Government agencies are under increasing pressure to develop cost systems that link budget to performance.  Guidance from the Office of Management and Budget (OMB) states that, “Agencies should strive to include goals or indicators for unit cost, even if only approximate costs can be estimated.”  In addition, the Statement of Federal Financial Accounting Standards Number 4, “Managerial Cost Accounting Concepts and Standards for the Federal Government,” contains several standards related to the necessity for cost data.  One costing method that has been used by both private and public sector organizations is Activity-Based Costing.

Without accurate cost-per-call measurements for each of the various toll-free telephone services the IRS provides, management and key stakeholders do not have sufficient information with which to evaluate the efficiency of toll-free telephone system operations, the return on investment in technology and human capital, and the costs and benefits of key program decisions.  This concern will be even more critical for the 2004 Filing Season when the IRS plans a small pilot test to evaluate the feasibility of outsourcing some tax law calls.  Without accurate and specific data on its own cost to answer this type of call, IRS management will be unable to determine if outsourcing is a cost-effective alternative for providing this type of service.

Recommendation

To ensure that IRS management and key stakeholders can monitor and evaluate the efficiency of the IRS toll-free telephone system operations, the Directors, CAS, SB/SE and W&I Divisions, should:

7.      Develop an Activity-Based Costing system that reliably captures and reports both the total cost and the cost-per-call of providing services on each toll-free product line.

 

Management’s Response:  The Commissioner, W&I Division, stated that the IRS Chief Financial Officer is implementing a comprehensive IRS-wide solution to modernize the accounting system and that it is designed to meet the need for activity-based costing.  When the system is successfully implemented, the IRS will develop costing information, at the toll-free product line level, if sufficient detailed information is available for analysis.

 

Appendix I

 

Detailed Objective, Scope, and Methodology

 

The overall audit objective was to evaluate the Internal Revenue Service’s (IRS) ability to provide taxpayers with effective access to its toll-free telephone system during the 2003 Filing Season.

To accomplish this objective, we:

I.                    Reviewed the IRS’ plans for the toll-free telephone system for the 2003 Filing Season.

A.        Analyzed IRS planning documents.

B.         Analyzed workload planning for the 2003 Filing Season toll-free telephone system.

C.        Attended bi-weekly executive status meetings held by the IRS for the toll-free telephone system.

II.                 Reviewed the IRS processes for promptly responding to taxpayer access problems with the toll-free telephone system.

A.        Participated in a walk-through at the IRS Joint Operations Center (JOC) to obtain an understanding of current practices regarding toll-free telephone system operations.

B.         Determined how automated systems are monitored/tested to ensure that taxpayer calls are properly routed.

            C.        Reviewed procedures for addressing identified access problems.

III.       Evaluated new business processes implemented by the IRS to improve taxpayer access to the toll-free telephone system.

A.        Discussed the implementation of the Customer Communications Project Fiscal Year 2001 Release with IRS management and analyzed documentation on its effectiveness in improving taxpayer access to the toll-free telephone system.

B.         Reviewed the IRS’ rationale and plans to employ additional call screeners for the 2003 Filing Season and determined what effect the change had on the Customer Service Representative (CSR) Level of Service (LOS) based on Full-Time Equivalent (FTE) costs and the number of calls answered by CSRs.

C.        Discussed any changes to telephone applications with IRS management.

D.        Discussed the establishment of any additional toll-free telephone numbers with IRS management and determined the effectiveness of the additional numbers by evaluating whether there was sufficient public information on the purpose of the new numbers.

E.         Coordinated with the Treasury Inspector General for Tax Administration (TIGTA) audit team that is currently evaluating whether the Small Business/Self-Employed Division is effectively managing CSR turnover at its toll-free telephone system call sites to determine whether the IRS has taken any action to address the high turnover rate at some call sites.

F.         Discussed changes to routing scripts with IRS management and reviewed applicable documentation, evaluated whether the scripts were simplified, and determined whether the revised scripts were effective in reducing the number of callers opting out of the scripts (i.e., primary abandons).

G.        Identified and reviewed any additional changes that IRS management implemented to improve productivity.

H.        Reviewed the instructions on using the toll-free telephone system provided to taxpayers by the IRS.

IV.       Reviewed recent productivity analyses and related studies on the toll-free telephone system conducted by the IRS and its contractors.

V.        Reviewed the IRS performance measures for the toll-free telephone system.

A.        Analyzed IRS reports to identify current toll-free telephone system performance measures, indicators, etc.

B.         Evaluated the IRS performance measures for accuracy.

 

C.                 Determined whether the IRS has developed a methodology to determine the cost-per-call for toll-free telephone services.

 

D.                 Using the IRS’ performance measures (and any other measures deemed appropriate) for the toll-free telephone system, compared performance for the 2003 Filing Season with that of prior years.

VI.              Conducted an analysis of time usage by CSRs.

A.                 Determined the total number of FTEs planned for toll-free telephone services in the 2003 Filing Season and compared it to the number for the 2002 Filing Season.

B.                 Conducted an analysis of the availability of CSRs.

 

Appendix II

 

Major Contributors to This Report

 

Philip Shropshire, Acting Assistant Inspector General for Audit (Small Business and Corporate Programs)

Frank Dunleavy, Acting Director

William E. Stewart, Audit Manager

Karen J. Stafford, Senior Auditor

Denise M. Gladson, Auditor

Debra D. Mason, Auditor

 

Appendix III

 

Report Distribution List

 

Commissioner  C

Office of the Commissioner – Attn:  Chief of Staff  C

Deputy Commissioner for Operations Support  OS

Deputy Commissioner for Services and Enforcement SE

Chief Information Officer  OS:CIO

Acting Deputy Commissioner, Small Business/Self-Employed Division  SE:S

Deputy Commissioner, Wage and Investment Division SE:W

Associate Commissioner, Business Systems Modernization OS:CIO:B

Director, Customer Account Services, Small Business/Self-Employed Division  SE:S:CAS

Director, Customer Account Services, Wage and Investment Division  SE:W:CAS

Director, Accounts Management, Wage and Investment Division  SE:W:CAS:AM

Director, Joint Operations Center  SE:W:CAS:JOC

Deputy Director, Accounts Management, Small Business/Self-Employed Division

SE:S:CAS:AM

Chief Counsel  CC

National Taxpayer Advocate  TA

Director, Office of Legislative Affairs  CL:LA

Director, Office of Program Evaluation and Risk Analysis RAS:O

Office of Management Controls  OS:CFO:AR:M

Audit Liaisons:

            Commissioner, Small Business/Self-Employed Division  SE:S

            Commissioner, Wage and Investment Division  SE:W

            Director, Customer Account Services, Small Business/Self-Employed Division

SE:S:CAS

            Director, Customer Account Services, Wage and Investment Division  SE:W:CAS

 

Appendix IV

 

Outcome Measures

 

This appendix presents detailed information on the measurable impact that our recommended corrective actions will have on tax administration.  These benefits will be incorporated into our Semiannual Report to the Congress.

Type and Value of Outcome Measure:

·        Inefficient Use of Resources – Potential; $3,605,742 per year.  Over 5 years, $18,028,710 (see page 10).

Methodology Used to Measure the Reported Benefit:

Using the Internal Revenue Service’s (IRS) work plan for Fiscal Year (FY) 2002, we compared the percentage of default screener services planned (Applications 1 and 3 for English; Applications 2 and 4 for Spanish) with all services planned for the three main customer service toll-free lines.  The total services planned, including screener services, totaled 41,016,739.  We then used the number of English screener services planned divided by the total services to determine the percentage (3.86 percent) of English screener services to total services. We used the same process to calculate Spanish services (.0032 percent).  Next, we used the Site-Level Balanced Measure–Assistor Services Provided Report from the IRS Enterprise Telephone Data (ETD) Warehouse to calculate the actual services provided for the comparable lines for FY 2003 through June 2003.  We used the Customer Account Services subtotal of 32,067,684 and subtracted calls to the following product lines:  Criminal Investigation, National Taxpayer Advocate, Practitioner Priority Service, Volunteer Income Tax Assistance, Teletype/Telecommunications Device for the Deaf, and International. This gave a total of 30,814,710. We then obtained screener services from the Agent Transfer Report (also on the IRS ETD Warehouse) for the same period for Application 1 (7,795,644) and Application 2 (629,301).  We added these services to the prior subtotal to arrive at total services provided of 39,239,655.

Next, we applied the FY 2002 percentage of planned English screener services to the actual services for FY 2003 and determined that screener services should have been 1,514,651 (39,239,655 x .0386).  We used the same process for Spanish calls and found that screener services should have been 125,567 (39,239,655 x .0032).  We then subtracted these numbers from the actual screener calls for each category and calculated the total time spent on these calls based on the Average Handle Time (AHT) for each category listed on the Agent Transfer Report.

·        English: 7,795,644 – 1,514,651 = 6,280,993 services x AHT of 54.73 seconds =  343,758,747 seconds.

·        Spanish: 629,301 – 125,567 = 503,734 services x AHT of 58.56 seconds =  29,498,663 seconds.

Next, we added the above 2 figures to arrive at actual hours of 103,683 which, when divided by 2,088 hours, equaled 49.66 Full-Time Equivalents (FTE).  However, because this is all direct time and the IRS planned only 45.6 percent of toll-free telephone system direct time per FTE, we divided the 49.66 by .456 to arrive at 108.9 actual FTEs.  To determine the minimum Customer Service Representative (CSR) cost, we used the normal entry-level (i.e., Grade 5, Step 1) salary for a CSR, although availability can occur at any grade.  We also used the lowest locality pay rate (i.e., the “Rest of United States [U.S.]”) at that grade level ($25,697) and calculated the annual salary costs at $2,798,403 (108.9 FTEs x $25,697).  We then multiplied this figure by the Wage and Investment (W&I) Division’s CSR 2003 floor benefit rate of 28.85 percent and added that result ($807,339) to the salary figure for a total of $3,605,742.

Type and Value of Outcome Measure:

·        Inefficient Use of Resources – Potential; $6,365,749 per year.  Over 5 years, $31,828,745 (see page 17).

Methodology Used to Measure the Reported Benefit:

The IRS established a planning assumption of 5.5 percent Assistor Availability for FY 2003. Therefore, we used this percentage to calculate the number of available CSR hours that should have occurred. From the IRS ETD Warehouse, we reviewed the Site-Level Performance Indicator–Assistor Availability Reports for FY 2003 through June 2003.  These Reports provide data on the actual number of hours that CSRs spent handling calls (i.e., handle time) and the actual number of hours that CSRs spent waiting for calls to arrive (i.e., available time).  When these 2 figures are added together, they give the total telephone time that, in this case, was 6,035,012 hours.  Using the 5.5 percent planned availability rate, the number of available hours should have been 331,926.  The actual available hours as shown on the reports were 733,261, or a difference of 401,335.  To determine the minimum salary costs of these 401,335 hours, we applied the normal entry level (Grade 5, Step 1) salary ($25,697) for a CSR using the lowest locality pay rate (i.e., the “Rest of U.S.”) and calculated the hourly salary rate at $12.31 per hour ($25,697/2,088 hours).  This hourly rate multiplied by the number of unplanned available hours equaled $4,940,434. We then multiplied this figure by the W&I Division’s CSR 2003 floor benefit rate of 28.85 percent and added that result ($1,425,315) to the salary figure for a total of $6,365,749.

 

Appendix V

 

Management’s Response to the Draft Report

 

The response was removed due to its size.  To see the response, please go to the Adobe PDF version of the report on the TIGTA Public Web Page.