Electronic Returns Were Processed Effectively

November 2000

Reference Number: 2001-40-008

 

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.

November 14, 2000

MEMORANDUM FOR COMMISSIONER ROSSOTTI

FROM: Pamela J. Gardiner /s/ Pamela J. Gardiner

Deputy Inspector General for Audit

SUBJECT: Final Audit Report - Electronic Returns Were Processed Effectively

This report presents the results of our review to evaluate the effectiveness of the Internal Revenue Service’s (IRS) controls for processing individual electronic return transactions during the 2000 Filing Season.

In summary, we found that the individual electronic returns were processed effectively. However, we identified an opportunity for the IRS to improve the customer service provided to electronic transmitters who called with questions about their transmissions. As a result of our review, IRS management identified steps to improve the consistency of its customer service operations. We recommended the IRS carry out its plans to develop and implement National Headquarters standards and guidance to correct the operational weaknesses we identified.

Management agreed to the recommendations we presented. Management’s comments have been incorporated into the report where appropriate, and the full text of their comments is included as an appendix.

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 your staff may call Walter Arrison, Associate Inspector General for Audit (Wage and Investment Income Programs), at (770) 936-4590.

Table of Contents

Executive Summary

Objective and Scope

Background

Results

Individual Electronic Returns Were Processed Effectively

Service to E-File Providers Could Be Improved

Conclusion

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 – Overview of the Internal Revenue Service’s Processing Subsystems

Appendix VI – Processing Problems that Occurred Early in the 2000 Filing Season

Appendix VII – Management’s Response to the Draft Report

Executive Summary

Broadening the use of electronic tax administration is a key component of the Internal Revenue Service’s (IRS) modernization effort. Electronically filed (E-file) returns improve service for taxpayers and boost production by reducing errors, speeding refunds, and reducing labor costs. Section 2001(a) of the IRS Restructuring and Reform Act of 1998 (RRA 98) indicated it should be the goal of the IRS to have at least 80 percent of all tax and information returns filed electronically by 2007.

The IRS estimated that 126.9 million individual returns would be filed during Calendar Year 2000 and that 33.6 million of these would be filed electronically. As of June 15, 2000, the IRS had processed 35.2 million individual electronic returns. This volume is 105 percent of the 33.6 million individual electronic returns that the IRS projected it would receive during the 2000 Filing Season.

The overall objective of this audit was to evaluate the effectiveness of the IRS’ controls for processing individual electronic return transactions during the 2000 Filing Season.

Results

Overall, the IRS’ system for accepting and processing individual electronic returns was effective. While some problems occurred early in the filing season, they were quickly identified and corrected.

However, we identified an opportunity for the IRS to improve the service it provides to Authorized IRS E-file Providers through its Help Desk. This is important because one aspect of the IRS reorganization is the promotion of nationwide consistency in its business dealings with taxpayers and in its internal operations.

Individual Electronic Returns Were Processed Effectively

The IRS’ system for accepting and processing individual electronic returns was effective. For example, returns accepted for IRS processing were virtually error-free. In addition, the controls for ensuring that individual electronic returns are accepted only from Authorized IRS E-file Providers were working properly, and the IRS properly notified unauthorized E-file providers when their transmissions were rejected. While some problems occurred early in the 2000 Filing Season, they were quickly identified and corrected (see Appendix VI for more information).

The IRS effectively implemented the enhancements added to the individual E-file Program for the 2000 Filing Season. For example, after some initial problems with the Debt Indicator Pilot were identified and resolved early in the filing season, the IRS provided accurate debt indicators for taxpayers who entered into Refund Anticipation Loan agreements. We also determined that return data containing the tax forms or schedule added for the 2000 Filing Season were properly processed to taxpayers’ accounts.

Service to E-File Providers Could Be Improved

The IRS designed a Help Desk to provide customer service to Authorized IRS E-file Providers who experienced problems with their transmissions. We reviewed the Help Desk operations at 3 of the 5 processing sites and found that 1 site provided ineffective customer service to the approximately 37,000 Authorized IRS E-file Providers it serviced. This occurred because the IRS had not established national standards and guidelines to ensure that the Help Desk operations at the processing sites were consistent. This report discusses the weaknesses we identified with the one Help Desk and the proposed actions that IRS management plans to take to correct those weaknesses.

Summary of Recommendations

As a result of our review, IRS management identified steps to improve the consistency of its Help Desk operations. We recommend the IRS carry out its plans to develop and implement National Headquarters standards and guidance to correct the operational weaknesses we identified.

Management’s Response: IRS management acknowledged the need to improve the customer service provided to Authorized IRS E-file Providers and identified several steps to address this need. For example, the Internal Revenue Manual is being updated to provide guidance to the processing sites. Help Desk employees will be expected to answer calls from Authorized IRS E-file Providers, and recording devices should be used only when all telephone lines are busy. In addition, the hours of service provided at each site may be extended to manage the call volume during peak periods.

Additional training will be given to Help Desk employees at each site prior to the 2001 Filing Season, and the number of Help Desk employees trained will be increased.

Processing sites will be required to develop quality review procedures to evaluate the effectiveness of customer service provided by their Help Desks operations. Evaluations of the Help Desk operations against the national standards guidelines will be conducted during periodic on-site reviews at the processing sites. Finally, a team will be established to analyze and recommend solutions to long-term aspects of Help Desk service; for example, recommended telephone system, percentage of calls answered on-line, etc. This team will consist of representatives from the processing sites, Electronic Tax Administration National Headquarters, National Treasury Employees Union, and contractor support.

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

Objective and Scope

The overall objective of this review was to evaluate the effectiveness of the Internal Revenue Service’s (IRS) controls for processing individual electronic return transactions during the 2000 Filing Season. Since all individual electronic returns are processed similarly after they are received by the IRS, our review focused on testing key processing controls that included:

We conducted this review as part of the Treasury Inspector General for Tax Administration’s Fiscal Year 2000 audit plan. Our audit tests covered individual electronic returns processed by the Andover, Austin, and Ogden processing sites during January and February 2000. We also performed selective audit tests on individual electronic returns processed by the Cincinnati and Memphis processing sites during the same period. This review was conducted in accordance with Government Auditing Standards.

Details of our audit objective, scope, and methodology are presented in Appendix I. Major contributors to this report are listed in Appendix II.

Background

Broadening the use of electronic tax administration is a key component of the IRS’ modernization effort. E-file returns improve service for taxpayers and boost production by reducing errors, speeding refunds, and reducing labor costs.

The IRS offers individuals different options for electronically filing their tax returns and for paying their taxes. For example, individual taxpayers can:

The IRS Restructuring and Reform Act of 1998 (RRA 98) § 2001(a) established a goal for the IRS that 80 percent of all tax and information returns should be filed electronically by 2007. For Calendar Year 2000, the IRS estimated that 126.9 million individual returns would be filed and that 33.6 million of these (26.5 percent) would be filed electronically. Through June 15, 2000, the IRS had processed 35.2 million individual electronic returns. Following is a breakdown of these 35.2 million returns by filing option (the 3 filing options are described above):

Changes for the 2000 Filing Season

For the 2000 Filing Season, the IRS added four more forms and one more schedule (listed below) for acceptance in the Tax Year (TY) 1999 IRS E-file Program (The total number of forms and schedules accepted electronically is 63.):

The IRS also implemented the Debt Indicator Pilot. Under this pilot, the IRS provided Authorized IRS E-file Providers who met certain requirements to participate in the pilot with a debt indicator for each taxpayer who entered into a Refund Anticipation Loan (RAL) agreement with a financial institution. The debt indicator showed whether the taxpayer owed prior year taxes (IRS debt) or owed past-due child support or other Federal agency debt, such as student loans.

Overview of processing of electronic returns

The system for processing electronic returns includes four subsystems (listed below). See Appendix V for general descriptions of the subsystems.

The first day of the 2000 E-file season was January 14, 2000. This was the first day the IRS started accepting TY 1999 individual electronic returns from Authorized IRS E-file Providers. During the week preceding this date, the IRS conducted "start-up" tests at the Austin and Memphis processing sites. The purpose was to identify problems and correct them prior to January 14, 2000. Daily conference calls were held the week of start-up testing, and weekly conference calls were conducted thereafter. Representatives from several IRS offices and geographic locations participated in these conference calls.

Results

Overall, the IRS’ system for accepting and processing individual electronic returns was effective. While some problems occurred early in the 2000 Filing Season, they were quickly identified and corrected. (In Appendix VI, we present additional information about two situations that occurred in January 2000.) Although the IRS’ system for processing individual electronic returns was effective, the IRS has an opportunity to improve the service it provided to Authorized IRS E-file Providers through its Help Desk. Promoting nationwide consistency in its business dealings with taxpayers and in its internal operations is an important aspect of the IRS reorganization.

Individual Electronic Returns Were Processed Effectively

For the audit period January 14 through February 29, 2000, the IRS processed 22.9 million individual electronic returns, which was 68 percent of the 33.6 million returns that the IRS projected for the 2000 Filing Season.

Overall, the IRS’ system for accepting and processing individual electronic returns was effective. The returns that were accepted for processing were virtually error-free. For example, IRS processing reports showed less than 1 percent of the returns were subsequently rejected during processing.

We interviewed IRS personnel and examined IRS computer reports to determine how the IRS monitors the processing of individual electronic returns. We also attended nine filing season conference calls between January 12 and March 1, 2000, to determine how the IRS monitors filing season progress and addresses processing problems. We found that the IRS properly resolved problems that were identified.

We tested the IRS’ controls for ensuring that individual electronic returns were accepted only from Authorized IRS E-file Providers and found that the controls were working properly. To accomplish this, we obtained the computer listings of Authorized IRS E-file Providers for January 19 and February 2, 2000. We compared these listings to the return transmissions made on the same dates and verified that the IRS rejected returns transmitted by unauthorized E-file providers.

We verified that the IRS provided customer service to the unauthorized E-file providers by properly notifying them when their transmissions were rejected. We also verified that the rejected returns were included in one of the IRS’ computer reports used to monitor processing results.

The IRS also effectively implemented the individual E-file Program enhancements. After the IRS resolved some initial problems with the Debt Indicator Pilot that affected approximately 20,000 returns (see Appendix VI), we reviewed a statistically valid sample of 143 returns from a population of over 4 million individual E-file returns that contained taxpayer requests for RAL agreements. The positive test results verified that, after the initial problems with the Debt Indicator Pilot were corrected, the IRS provided accurate debt indicators for taxpayers who entered into RAL agreements.

We selected another statistically valid sample of 155 returns from a population of 657 individual E-file returns that contained 1 or more of the forms or schedule added to the individual E-file Program for the 2000 Filing Season. We traced the return data to the taxpayers’ accounts and verified that the IRS properly processed the returns.

Further, we reviewed the activity for 58 Error Reject Codes (ERC) that were added for the 2000 Filing Season. We determined that the IRS’ computer programs were working properly to reject returns that did not meet the IRS’ processing criteria. We also performed a limited review of the remaining 355 ERCs to identify reject trends that might indicate problems with the IRS’ computer systems. We found only 1 instance in which approximately 40,000 returns claiming the child care credit were erroneously rejected because of improper IRS computer programming. The error was identified and corrected early in the filing season (see Appendix VI).

Based on the results above, we concluded that the IRS’ system for processing individual electronic returns was effective.

Service to E-File Providers Could Be Improved

The Help Desk was designed to provide customer service to Authorized IRS E-file Providers who experience problems with their transmissions. The telephone numbers for the Help Desks are listed in the IRS E-file Filing Season Supplement for Electronic Return Originators (TY 1999), (Pub. 1345A (Rev. 12-99)).

When Authorized IRS E-file Providers call a processing site Help Desk, IRS employees should use the specially designed Help Desk computer application to research the problems to assist the providers. This application gives employees the capability to research transmission activity between Authorized IRS E-file Providers and the IRS.

We reviewed the Help Desk operations at the Andover, Austin, and Ogden processing sites and found that the Andover Help Desk did not provide effective customer service to its Authorized IRS E-file Providers. The Andover Help Desk is responsible for providing customer service to approximately 37,000 Authorized IRS E-file Providers. The Austin and Ogden Help Desks are responsible for providing service to approximately 63,000 Authorized IRS E-file Providers.

We found that the Help Desk operation at the one processing site was not adequately staffed with employees who were trained to work the Help Desk. In addition, the published telephone number for the Help Desk was for a telephone that was not in the same physical location as the employees who were responsible for working the Help Desk. When Authorized IRS E-file Providers dialed the number, they had to leave a message on a voice recorder. Later, employees manually transcribed the messages onto paper forms. The paper forms were evaluated and, depending on the nature of the problem, were forwarded to the Help Desk or other employee groups for resolution. We were told that the recorded messages were "retrieved once per hour or more frequently" and were "answered within 2 to 5 hours of receipt."

For the period January 14 through February 29, 2000, this processing site forwarded 1,795 transcribed messages (paper forms) to the Help Desk for resolution. We analyzed the information for 158 of these calls (9 percent) and determined that the recorded messages were not always retrieved within 1 hour. Our results showed:

We also found that the recorded messages were not always answered within 2 to 5 hours of receipt. Our results showed:

The other two Help Desk operations we reviewed used automated telecommunications systems that captured the volume of calls made to the Help Desks. For the period January 14 through February 29, 2000, these Help Desks received a total of 57,543 calls. This volume is significantly greater than the 1,795 calls received by the processing site that required Authorized IRS E-file Providers to leave a message. The disparity in call volumes suggests that the one Help Desk might not be providing effective customer service.

Table 1 shows the number of Authorized IRS E-file Providers served by the five Help Desks. It illustrates the potential impact that ineffective Help Desk service could have on Authorized IRS E-file Providers.

Table 1: Number of Authorized IRS E-file Providers for Each Processing Site, as of April 9, 2000

Processing
Site Help Desk

Authorized IRS
E-File Providers

1

37,331

2

32,327

3

27,349

4

25,999

5

30,543

Source: The IRS’ Applicants Database

Authorized IRS E-file Providers nationwide should expect the same level of customer service regardless of which processing site Help Desk they call. Authorized IRS E-file Providers who do not receive timely customer service from one Help Desk may ultimately call another processing site Help Desk.

During the audit, we discussed our concerns regarding the Help Desk operations with IRS executives. They shared our concerns and recognized the need to develop national guidelines to ensure that all processing site Help Desks provide consistent customer service to Authorized IRS E-file Providers. The executives also identified the following proposed areas that need additional guidance:

Recommendations

The actions that the IRS plans to take are positive steps towards improving the Help Desk operations. We believe that thorough evaluations of each proposed area would identify issues that, if properly addressed, could improve the overall consistency of customer service provided by the Help Desks. We recommend the IRS carry out the planned actions and:

  1. Develop national standards, goals, and procedural guidance, for each of the proposed areas mentioned above, to ensure consistent and effective customer service is provided by all processing site Help Desks.
  2. Management’s Response: Management plans to update the Internal Revenue Manual (IRM) to provide guidance to the processing sites. More specifically, the IRM will indicate that Help Desk employees should answer calls, and that recording devices should be used only when all telephone lines are busy. Management also plans to share its expectations for the 2001 Filing Season with the processing sites.

    The Andover processing site will increase the number of Help Desk employees to 25 full-time tax examiners. Additional training will be given to Help Desk employees at each site, and the number of Help Desk employees trained will range from 13 to 45 per site. In comparison, the IRS conducted "train the trainer" sessions for only two employees per site for the 2000 Filing Season.

    Finally, management plans to establish a team to analyze and recommend solutions to the long-term aspects of Help Desk operations. This team will consist of representatives from the processing sites, Electronic Tax Administration National Headquarters, National Treasury Employees Union, and contractor support.

  3. Implement the necessary changes at each Help Desk to bring them into compliance with the established national standards and guidelines.
  4. Management’s Response: Each Help Desk will be instructed to adhere to the updated IRM and program expectations for the 2001 Filing Season. In addition, management will monitor the delivery of training and evaluate the results.

  5. Design and implement a control process to:(a) evaluate each Help Desk operation against the national standards, goals, and procedures, and (b) implement corrective actions for unacceptable deviations from the standards, goals, or procedures.

Management’s Response: Management will evaluate the Help Desk operations during periodic on-site reviews at the processing sites. In addition, the processing sites will be required to develop quality review procedures to evaluate the effectiveness of telephone assistance and implement corrective actions if standards are not met.

Conclusion

The importance of electronic filing is evidenced by the goals established by the Congress in RRA 98 § 2001(a). Overall, we determined that the IRS’ controls for accepting and processing individual electronic returns were effective and that the program enhancements for the 2000 Filing Season were properly implemented. Some processing problems occurred early in the 2000 Filing Season, but they were quickly identified and corrected. Help Desk operations at one processing site could be improved, and we believe that implementation of the recommendations included in this report should improve the overall consistency and customer service provided by the Help Desks nationwide.

Appendix I

Detailed Objective, Scope, and Methodology

The overall objective of this review was to evaluate the effectiveness of the Internal Revenue Service’s (IRS) controls for processing individual electronic (E-file) return transactions during the 2000 Filing Season. To accomplish this objective, we evaluated the IRS’ controls for accepting and processing individual electronic returns. We also determined if the processing changes and tax forms and schedule added for the 2000 Filing Season were effectively implemented. Lastly, we evaluated the service provided to Authorized IRS E-file Providers who call the IRS for assistance. Specifically, we performed the following audit tests:

I. To identify and evaluate the controls over the acceptance and processing of individual electronic returns, we:

    1. Interviewed IRS personnel and reviewed local and national Electronic Filing (ELF) System reports or data to identify and evaluate performance measures, processing problems, and actions taken by the IRS personnel to correct processing problems. More specifically, we:
      1. Interviewed IRS personnel for background information about the pre-filing season tests conducted at the Austin processing site during the week beginning January 10, 2000, and attended the initial "kick-off" meeting on January 11, 2000.
      2. Attended nine filing season conference calls between January 12 and March 1, 2000, to identify processing problems and determine if the IRS took appropriate action.
      3. Obtained the IRS’ projections for the national volume of electronically filed individual returns for 2000 and the same projections for the Andover, Austin, and Ogden processing sites.
      4. Interviewed the Andover, Austin, and Ogden Reports Analysts for the ELF Units to determine what reports they use to monitor processing and identify problems.
      5. Reviewed the IRS’ Startup Meeting Volume Reports between January 17 and February 29, 2000. This is a daily report that shows the cumulative E-file transmissions for the nation and for each of the five IRS processing sites. We computed the acceptance rate (number of electronic returns accepted divided by the number of returns transmitted) for 2000 and compared it with the acceptance rate for the 1999 Filing Season. We also monitored Andover, Austin, and Ogden processing site volumes to identify any unusual activity or trends.
    2. Determined whether the controls were working to ensure that individual electronic returns were accepted only from Authorized IRS E-file Providers and whether Authorized IRS E-file Providers were properly notified of rejected returns. To accomplish this, we performed the following tests for the Andover, Austin, and Ogden processing sites on January 19, 2000, and February 2, 2000:
      1. Obtained the list of Authorized IRS E-file Providers and compared it against the list of providers who transmitted returns to the IRS and determined whether the IRS rejected all returns transmitted by unauthorized E-file providers. For the 2 days that we tested, the IRS rejected 793 returns associated with 191 unauthorized E-file providers. In other words, the IRS rejected all returns received from all unauthorized E-file providers.
      2. For all 191 unauthorized E-file providers identified in B.1., determined whether the IRS notified the E-file providers that their transmissions were rejected.
      3. Determined whether the 793 rejected returns identified in B.1. were included in the volume of returns reported on the Daily Error Reject Code Reports under Error Reject Code (ERC) 029. This ERC is used to capture the volume of returns transmitted from unauthorized E-file providers.
      4. Interviewed Information Systems personnel at the Andover, Austin, and Ogden processing sites to determine if the daily Applicants Database (ADB) was received timely for that day’s processing and if they had experienced any problems with the ADB.
    3. Reviewed a statistically valid sample of 143 returns and determined if the debt indicator was valid for each return. (Note: The scope of this test included all five processing sites that process individual electronic returns: Andover, Austin, Cincinnati, Memphis, and Ogden.) To accomplish this, we:
      1. Performed a computer analysis that identified a population of 4,364,938 individual electronic returns processed from January 30 through February 9, 2000, that requested Refund Anticipation Loan (RAL) agreements. We used a 90 percent confidence level, 5 percent expected error rate, and +/- 3 percent reliability to arrive at a sample size of 143. We used SAS Institute software to randomly select the 143 returns from the population. The breakdown of the 143 returns by processing site was: Andover - 28; Austin - 37; Cincinnati - 32; Memphis - 26; and Ogden - 20.
      2. Determined whether all 143 returns in our sample contained a debt indicator in the acknowledgement (ACK) file. (Note: Every return that requests a RAL agreement should have a corresponding debt indicator. The absence of the debt indicator would indicate processing problems.)
      3. Verified the accuracy of the debt indicator in the ACK file by comparing it to taxpayer data on the IRS’ Individual Masterfile (IMF).
      4. Recomputed an actual sample reliability of 1.6 percent because we had no errors in our sample results.
    4. Monitored the GMF 2740 Runs reports for the Andover, Austin, and Ogden processing sites from January 18 through February 29, 2000, to identify the volume of electronic returns sent to the Error Resolution function. Because the volume of electronic returns that went to Error Resolution was less than 1 percent for each processing site, we limited further testing to Austin. We reviewed Austin’s ERS 93-46 report dated February 25, 2000, for U.S. Individual Income Tax Returns (Forms 1040, 1040A, and 1040EZ) and determined the nature of the ERCs with 500 or more errors. We also reviewed Austin’s ERS 13-41 report dated February 25, 2000, to identify any aged inventory of returns in the Error Resolution function.
    5. Monitored the GUF 0749 reports for the Andover, Austin, and Ogden processing sites from January 20 through February 24, 2000, to identify the volume of electronic returns sent to the Unpostables function. We also interviewed the GUF Unit Manager, E-file Examiner, and GUF Analyst at the Austin processing site to identify the types of errors associated with electronic returns sent to Unpostables.
  1. To determine whether the new changes for the 2000 E-file season were effectively implemented, we:
    1. Monitored the Daily Error Reject Code Reports for the Andover, Austin, and Ogden processing sites for the period January 14 through February 29, 2000, to identify reject trends that would indicate systemic or software problems.
      1. Monitored the volumes of 24 ERCs associated with Schedule J and Forms 6781, 8271, 8582-CR, and 8586 that were added for the 2000 Filing Season (see page 3). We investigated unusual volume trends and determined if appropriate corrective action was taken, if warranted.
      2. Monitored the volumes of 34 additional ERCs that were added for the 2000 Filing Season. We investigated unusual volume trends and determined if appropriate corrective action was taken, if warranted.
    2. Performed a cursory review of the national volumes for all other ERCs on the Daily Error Reject Code Reports for the period January 14 through February 29, 2000, to identify reject trends that would indicate systemic or software problems. We investigated unusual volume trends and determined if appropriate corrective action was taken, if warranted. (Note: Although Objective II focused on the "new" changes added for the 2000 Filing Season, we decided to perform a cursory review of all ERCs to identify unusual volumes or trends that might indicate systemic or software problems.)
    3. Determined if the "new" forms and schedule added for the 2000 Filing Season were properly processed and posted to the IRS’ IMF. To accomplish this, we:
      1. Performed a computer analysis that identified a total of 657 "new" forms and Schedule J submitted with electronic returns processed by all 5 processing sites from January 14 through February 9, 2000. We used a 95 percent confidence level, 5 percent expected error rate, and +/- 3 percent reliability to arrive at a sample size of 155. We used an Access database statistical sampling application to randomly select a sample of 155 returns containing the new forms and schedule.
      2. For each of the 155 returns, we used command code TRDBV to verify that 1 or more of the "new" forms or schedule was filed with the return.
      3. For each of the 155 returns, we used command code TRDBV to determine whether the tax and the refund or amount due on the tax return agreed with the corresponding amounts on the taxpayer’s IMF account.
      4. Recomputed an actual sample reliability of 1.8 percent because we had no errors in our sample results.
    4. Determined if adequate corrective actions (e.g., processing changes, controls, new reject codes, etc.) were taken to ensure that processing problems that occurred during the 1999 Filing Season did not affect individual electronic returns processed during the 2000 Filing Season. To accomplish this, we:
      1. Reviewed documentation that described 1999 Filing Season problems and identified the problems that affected individual electronic returns. We interviewed IRS personnel to find out what corrective actions had been taken to address the problems. Based on the above, we estimated whether there was a low, medium, or high risk that the same problem would occur during the 2000 Filing Season.
      2. For each problem with a medium or high risk, either included tests in this review or recommended the problem be tested in a follow-up to this review.
  2. Determined if the IRS provided effective customer service to Authorized IRS E-file Providers who called the Andover, Austin, and Ogden Help Desks for assistance when they had problems with their transmissions. To accomplish this, we:
    1. Contacted IRS managers of the Help Desks at the Andover, Austin, and Ogden processing sites and obtained the daily volumes of calls received from Authorized IRS E-file Providers who called the Help Desk between January 14 and February 29, 2000. We analyzed the daily volumes for large or unusual trends that might indicate possible systemic problems that could affect the tax examiners’ (TE) abilities to provide effective customer service.
    2. Interviewed two Help Desk TEs at the Andover, Austin, and Ogden processing sites (for a total of six TEs) to determine if they had any problems or concerns that affected their abilities to provide effective customer service to Authorized IRS E-file Providers who called the Help Desk.

Appendix II

Major Contributors to This Report

Walter E. Arrison, Associate Inspector General for Audit (Wage and Investment Income Programs)

M. Susan Boehmer, Director

Gary L. Young, Audit Manager

Sharon A. Buford, Senior Auditor

Edith R. Lemire, Senior Auditor

Carl H. Parmer, Senior Auditor

John P. Ojeda, Auditor

Bonnie G. Shanks, Auditor

Appendix III

Report Distribution List

Deputy Commissioner Operations C:DO

Chief Counsel CC

Chief Operations Officer OP

Commissioner, Wage and Investment Division W

National Taxpayer Advocate TA

Director, Customer Account Services W:CAS

Director, Legislative Affairs CL:LA

Director, Office of Program Evaluation and Risk Analysis (OPERA) M:O

Office of Management Controls CFO:A:M

Audit Liaison:

Commissioner, Wage and Investment Division W

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.

Finding and recommendations:

The Internal Revenue Service (IRS) uses Help Desks at its five electronic filing (E-file) processing sites to research tax return processing problems experienced by Authorized IRS E-file Providers. We reviewed the Help Desks at the Andover, Austin, and Ogden processing sites and found that the Andover Help Desk did not provide effective customer service to its Authorized IRS E-file Providers. For the period January 14, through February 29, 2000, this site received only 1,795 calls, while the Help Desks at the other 2 processing sites we reviewed received a total of 57,543 calls. The Austin and Ogden Help Desks used automated telecommunications systems that captured the volume of calls made to the Help Desks. However, when Authorized IRS E-file Providers called the Andover Help Desk they had to leave a message on a voice recorder.

We analyzed 158 (9 percent) of the 1,795 calls to the Andover Help Desk and determined that the recorded messages were not always retrieved with 1 hour. Our results showed:

We also found that the recorded messages were not always answered within 2 to 5 hours of receipt. Our results showed:

Authorized IRS E-file Providers who do not receive timely customer service from one site’s Help Desk may ultimately call another processing site’s Help Desk, causing workload imbalances among the processing sites. These 3 Help Desks provide customer service to over 100,000 E-file Providers within their geographical boundaries.

To ensure that all Authorized IRS E-file Providers receive timely resolution to their inquires, we recommended that the IRS:

Type of Outcome Measure:

Taxpayer Burden - Potential

Value of the Benefit:

When Authorized IRS E-file Providers do not receive timely help in resolving their processing problems, the IRS is at risk of damaging relations with both the Authorized IRS E-file Providers and the taxpayers they represent. In addition, there is the potential effect of increased burden on Authorized IRS E-file Providers and taxpayers if additional contacts with the IRS are necessary to resolve inquires. We determined that the Help Desk at the Andover processing site services over 37,000 Authorized IRS E-file Providers. As a result, ineffective Help Desk operations could negatively affect many of the taxpayers represented by the 37,000 Authorized IRS E-file Providers.

Methodology Used to Measure the Reported Benefit:

We determined the number of Authorized IRS E-file Providers serviced by each Help Desk at the five processing sites. In addition, we used call volume data provided by IRS personnel to identify customer service inquires received for the period January 14 through February 29, 2000.

Appendix V

Overview of the Internal Revenue Service’s Processing Subsystems

The Internal Revenue Service’s (IRS) system for processing individual electronic returns includes the following four subsystems:

Front-End Processing Subsystem (FEPS)

This subsystem accepts returns from Authorized IRS E-file Providers, acknowledges the receipt of the information, prepares the information for mainframe computer processing, and returns acknowledgements to the Providers.

The IRS electronically acknowledges to the Providers the receipt of all transmissions. Individual returns in each transmission are either accepted or rejected for specific reasons. The IRS uses Error Reject Codes (ERC) to define the specific reasons why returns are rejected for processing and uses form field numbers (sequence numbers) to identify which fields of the electronic return data are involved. ERCs and sequence numbers are included in the acknowledgement files provided to Providers. Authorized IRS Electronic Filing (E-file) Providers use this information to correct the rejected data and retransmit the electronic returns. The Filing Season Supplement for Electronic Return Originators (Publication 1345A), which is issued each year, contains the ERC explanations and sequence numbers.

Authorized IRS E-file Providers who have transmission problems can call one of five IRS Help Desks for assistance. IRS employees at these Help Desks research electronic return transmission data between the FEPS and the processing sites to resolve questions.

State Retrieval Subsystem (SRS)

This subsystem transmits state tax return data to state revenue agencies participating in the Federal/State Electronic Filing Program. The FEPS receives the state return attached to a Federal return. Once the Federal return has been validated, checked, and accepted by the IRS, the state return is separated from the Federal return, copied to tape, and transferred to the SRS. The state returns are then available for retrieval, via telecommunication, by the participating state revenue agencies. (We did not review this subsystem during the audit.)

Returns Processing Subsystem (RPS)

This subsystem converts the electronic return data into a format for processing and generates the acknowledgements provided to Providers.

Tax Return Data Base (TRDB)

The law requires that tax returns be retained for seven years. This subsystem is the official repository of individual and business tax return data. It contains all of the tax return data for electronic returns and any corrections that are subsequently made to the returns during processing.

Appendix VI

Processing Problems That Occurred Early in the 2000 Filing Season

Child Care Credit
In late January 2000, approximately 40,000 individual returns were erroneously rejected because of programming errors in the Internal Revenue Service’s (IRS) computer programs used to process the returns. The processing problems involved the Child and Dependent Care Expenses (Form 2441) and the dependent care benefits reported on the Wage and Tax Statement (Form W-2). Both the IRS and the filing community identified the problem with the Form 2441. A software developer initially identified the problem with the W-2 and notified the IRS. By February 10, 2000, IRS personnel had investigated the problem and corrected the programming errors that caused the rejects. Returns that had been previously rejected could be re-transmitted beginning February 10, 2000.

Debt Indicator
For the 2000 Filing Season, the IRS also implemented the Debt Indicator Pilot for individual electronic returns. Under this pilot, the IRS provided Authorized IRS Electronic Filing (E-file) Providers, who met certain mandatory requirements to participate in the pilot, a debt indicator for each taxpayer who entered into a Refund Anticipation Loan (RAL) agreement with a financial institution. The debt indicator shows whether the taxpayer owes prior year taxes (IRS debt), owes past-due child support or other Federal agency debt such as student loans (Financial Management Service (FMS) debt), or has no debt.

A RAL agreement is money borrowed by a taxpayer from a lender based on the taxpayer’s anticipated refund amount. The IRS offsets taxpayers’ refunds to pay off prior year taxes that are owed, and the FMS offsets taxpayers’ refunds to pay off past-due child support or Federal agency debts that are owed. Offsets to non-tax debts occur after the IRS has certified the refunds to the FMS for payment but before the FMS issues the refund checks or makes the Direct Deposits. Refund offsets reduce the amount of the expected Direct Deposits or a paper check but do not delay the issuance of the remaining refund (if any).

In late January 2000, participants in the Debt Indicator Pilot initially identified and notified the IRS of three separate problems. The IRS responded immediately and initiated corrective actions that resolved the problems. First, financial institutions informed participating Authorized IRS E-file Providers that they had used the wrong RAL code in their transmissions of electronic returns to the IRS. Participants who expected the debt indicator code but did not receive it informed the IRS. Approximately 20,000 returns were affected. Second, the IRS received inquiries from taxpayers’ lenders who questioned why taxpayers’ refunds had been offset when they had been informed the taxpayers had no debt. A joint investigation by the IRS and the FMS determined that the FMS had failed to report to the IRS approximately $1.7 million in non-tax debt related to child support. Although the IRS did not cause either problem, both problems affected the accuracy of the debt indicators generated by the IRS. The third problem was caused by an IRS error. Participants notified the IRS that a debt indicator code did not appear in the appropriate field within the electronic acknowledgement file. The IRS sends an acknowledgement file to an Authorized IRS E-file Provider within 48 hours of receiving a transmission. It informs the Authorized IRS E-file Provider if the transmission was accepted or rejected for IRS processing.

Appendix VII

Management’s Response to the Draft Report

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