Executive Summary

Objective and Scope

Background

Results

Conclusion

Appendix I – Detailed Objective, Scope, and Methodology

Appendix II – Major Contributors to This Report

Appendix III – Report Distribution List

Appendix IV – Volume of Notices Reviewed—Week of April 16, 2000

Executive Summary

The Internal Revenue Service (IRS) has identified improving customer service as one of its major initiatives. As a result, it placed a priority on improving the quality of notices issued to taxpayers. From January through September 2000, the IRS mailed approximately 12.4 million notices to individual taxpayers. These notices were designed to inform them of taxes, interest, and penalties due; errors they made on their tax returns; or adjustments to their tax accounts.

The IRS established a process to identify and analyze potentially incorrect notices. This review was the last point in the IRS’ process where notices could be corrected before they were mailed to taxpayers. From January through September 2000, the IRS identified 4.1 million potentially erroneous notices for review.

The overall objective of this audit was to evaluate the IRS’ process for reviewing potentially inaccurate notices before they were mailed to taxpayers. We focused our review on notices that were prepared for individual taxpayers from January 1, 2000, through September 30, 2000, as a result of processing their tax returns.

The IRS could have done more to ensure taxpayers did not receive erroneous refunds and notices. The IRS did not review all potentially erroneous notices or prioritize its work so that notices with the highest potential for error were reviewed first. In our opinion, limited national oversight of the program contributed to these conditions.

From January through September 2000, the IRS reported that it did not review 539,852 (13 percent) of the 4.1 million notices identified as having a high potential for error. If the error rate for these notices is consistent with that found on notices that were reviewed, the IRS may have incorrectly notified 80,702 taxpayers about an additional tax liability, an error on their return, or an adjustment to their account. Wage and Investment taxpayers usually contact the IRS only once a year, when they file their tax returns. However, taxpayers who receive an incorrect notice will, most likely, be forced to make additional IRS contacts.

IRS procedures did not ensure that notices with the highest potential for error were worked first. From January through September 2000, the IRS reviewed approximately 1.5 million notices with error rates of 10 percent or less, while it did not review approximately 77,000 notices with anticipated error rates of 30 percent or higher. Not establishing a priority to review notices with the highest potential for error increased the risk that taxpayers received erroneous notices and refunds.

Prior to October 1, 2000, responsibility for the oversight of the Notice Review Program was assigned to two different IRS operating functions. We identified issues that indicated the national oversight of the Program could have been more effective. For example, the data compiled to monitor and evaluate the Program were not reviewed.

Objective and Scope

The objective of this review was to evaluate the Internal Revenue Service’s (IRS) process for reviewing potentially inaccurate notices before they were mailed to taxpayers. We focused our review on notices that the IRS prepared for individual taxpayers from January 1, 2000, through September 30, 2000, as a result of processing their tax returns.

We performed our work in the National Headquarters and the Atlanta, Austin, and Kansas City Submission Processing Centers. Our review was conducted from October through November 2000 and was performed in accordance with the President’s Council on Integrity and Efficiency’s Quality Standards for Inspections.

The Treasury Inspector General for Tax Administration is planning additional audit coverage of the Notice Review Program during Fiscal Year 2001. This review will focus on the methods the IRS uses to identify and select potentially erroneous individual notices for review.

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

Background

The IRS has identified improving customer service as one of its major initiatives. As a result, it placed a priority on improving notices issued to taxpayers. From January through September 2000, the IRS mailed approximately 12.4 million notices to individuals. These notices were designed to inform them of taxes, interest, and penalties due; errors they made on their tax returns; or adjustments to their tax accounts.

The IRS developed a computer program to identify potentially incorrect notices. Employees in the review function had 1 week or less to analyze these notices before they were sent to taxpayers. The review was also designed to identify and correct incorrect tax refunds. This review was the last point in the IRS’ process where notices could be corrected before they were mailed to taxpayers. From January through September 2000, the IRS identified 4.1 million potentially erroneous notices for review.

Results

The IRS could have done more to ensure taxpayers did not receive erroneous refunds and notices. It did not review all potentially erroneous notices or ensure that notices with the highest potential for error were reviewed first. In our opinion, limited national oversight of the Program contributed to these conditions.

The Internal Revenue Service Did Not Review All Potentially Erroneous Notices

From January through September 2000, the IRS reported that it did not review 539,852 (13 percent) of the 4.1 million notices identified as having a high potential for error. If the error rate for these notices is consistent with that found on notices that were reviewed, the IRS may have incorrectly notified 80,702 individual taxpayers about an additional tax liability, an error on their return, or an adjustment to their account.

Wage and Investment taxpayers usually contact the IRS only once a year, when they file their tax returns. However, taxpayers who receive an incorrect notice will, most likely, be forced to make additional IRS contacts.

The chart below shows the inventory volumes from January through September 2000 for three IRS processing centers. The largest inventories of notices identified for review occurred during March, April, May, and June 2000. This was the same time period that most individual taxpayers’ 1999 tax returns were expected to be filed with the IRS. Inventory for review in the 3 processing centers reached its highest point of 260,295 notices during May. Of these notices, 53,036 (20 percent) were not reviewed.

IRS guidelines required that all of the notices selected for review be reviewed. While we did not do a staffing analysis, we were advised by IRS executives that they did not have enough staff to work the entire inventory of potentially erroneous notices. Not reviewing the entire weekly notice inventory increased the probability that erroneous notices and refunds were mailed to taxpayers.

The Submission Processing Centers Did Not Place Priority on Reviewing Notices With the Highest Potential for Error

The IRS did not prioritize the notice inventory to ensure that notices with the highest potential for error were analyzed first. IRS procedures required that notices involving a refund be worked before those associated with a balance due. Guidelines indicated that when less than 100 percent of the refund notices could be reviewed, refund notices with the highest anticipated error rate should have been worked first. However, no such priority was established for notices that did not involve a refund.

From January through September 2000, the IRS reviewed approximately 1.5 million notices with error rates of 10 percent or less, while it did not review approximately 77,000 notices with anticipated error rates of 30 percent or higher. (Appendix IV shows the volume of notices reviewed and their corresponding error rates for the week of April 16, 2000.)

Not establishing a priority to review notices with the highest potential for error increased the risk that taxpayers received erroneous notices and refunds.

National Oversight of the Notice Review Program Could Be Improved

We observed the following conditions that indicated the national oversight of the Program could have been more effective:

• The data compiled to monitor and evaluate the Notice Review Program were not reviewed. Instead, an employee who was not responsible for reviewing this Program maintained the data.
• The expectation for the processing centers to work all identified notices within the established time frame had not been re-evaluated to assess its viability.

We did not analyze these conditions to determine why they occurred. However, prior to the IRS’ reorganization (October 1, 2000), oversight of the Notice Review Program was assigned to two different IRS operating functions. This could have contributed to these conditions.

Conclusion

It is likely the IRS’ Notice Review Program did not prevent some taxpayers from receiving incorrect notices and refunds. The IRS’ current reorganization presents an opportunity for it to redesign this Program so that it meets the objective of ensuring the accuracy of taxpayer notices.

Detailed Objective, Scope, and Methodology

This review was part of the Treasury Inspector General for Tax Administration’s discretionary audit coverage. The overall objective of the review was to evaluate the Internal Revenue Service’s (IRS) process for reviewing potentially inaccurate notices before they were mailed to taxpayers. We focused our review on notices that were prepared for individual taxpayers from January 1, 2000, through September 30, 2000, as a result of the processing of their tax returns.

To determine whether the IRS reviewed all potentially erroneous notices before they were mailed to taxpayers, we performed the following audit work:

1. Determined if emphasis was placed on the case selection hierarchy when the inventory was worked in the submission processing centers.
1. Analyzed the Notice Disposition Reports nationwide between April 17, 2000, and May 12, 2000, to determine if the processing centers placed a priority on reviewing the notices selected for review with specific selection criteria.
2. Interviewed managers and employees responsible for reviewing notices in the Atlanta, Austin, and Kansas City Submission Processing Centers to determine if there was a process to ensure that the inventory was worked based on the priority of selection keys.
3. Reviewed a written description of the notice review computer program and interviewed the notice review computer programmer to determine if there was a method for the processing centers to ensure they were reviewing the inventory in a priority order.
4. Compared the written explanation of the notice selection criteria to a notice review sign-out sheet for the week of October 20, 2000, to determine if the batch sheets used by the processing centers were printed to facilitate the prioritization of work by selection criteria.
5. Requested copies of written guidance from the National Headquarters to the processing centers indicating the priority of the notice review inventory.
6. Reviewed written IRS guidance documents to identify procedures involving the prioritization of the notice review inventory.
2. Determined the disposition of notices that were selected, but not reviewed, by the IRS.
1. Requested a copy of written guidance provided by the National Headquarters to the processing centers indicating procedures for handling notices that are selected for review but not reviewed.
2. Interviewed managers and employees responsible for reviewing notices in the Atlanta, Austin, and Kansas City Submission Processing Centers to determine if notices that were selected but not reviewed were:
1. Held for future review.
2. Mailed as is.
3. Held using the "Label" disposition code so they would be brought back for review the following week. If so, we:
1. Determined the criteria for electing to use the "Label" disposition code to hold a notice.
2. Identified the process for notices that were closed using the "Label" disposition code.
3. Determined if notices that were closed using the "Label" disposition code were appropriately reflected in the Notice Disposition Report for January 1, 2000, through September 30, 2000.
3. Obtained national notice disposition data from the IRS and compared the number of notices selected for review nationwide from January 1, 2000, through September 30, 2000, to the number of notices actually reviewed during the same period. We did not validate the notice disposition data provided by the IRS.
1. Analyzed the notices selected but not reviewed to determine if notices with higher potential error rates were not reviewed while notices with lower potential error rates were reviewed.
1. Identified 12.4 million selectable notices for the period January 1, 2000, through September 30, 2000, by reviewing the IRS Individual Masterfile Notice Volume Reports for Fiscal Year 2000.
1. Analyzed the national notice disposition data to determine the number of notices reviewed that were identified as inaccurate notices.
2. Analyzed the national notice disposition data for the notices selected but not reviewed and stratified by error rate to determine the number and/or type of potentially erroneous notices that were sent to taxpayers.

Major Contributors to This Report

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

M. Susan Boehmer, Director

Deann L. Baiza, Audit Manager

Linda L. Bryant, Senior Auditor

Kathleen A. Hughes, Auditor

Sharla Robinson, Auditor

Bonnie G. Shanks, Auditor

Report Distribution List

Commissioner N:C

Director, Customer Account Services W:CAS

Director, Submission Processing W:CAS:SP

Director, Legislative Affairs CL:LA

Chief Counsel CC

National Taxpayer Advocate TA

Office of Management Controls N:CFO:F:M

Director, Office of Program Evaluation and Risk Analysis N:ADC:R:O

Audit Liaisons:

Commissioner, Wage and Investment Division W

Director, Customer Account Services W:CAS

Director, Submission Processing W:CAS:SP

Volume of Notices Reviewed - Week of April 16, 2000

This Appendix provides a summary of the volume of notices reviewed and the associated error rate by notice selection category for the week of April 16, 2000. The Internal Revenue Service (IRS) uses a computer program to identify potentially erroneous notices for review. These notices are selected based on pre-established selection criteria. Each selection criterion uses a composite of more finite conditions called selection keys to identify and select notices for review.

Notice Selection Category A: Selects notices identified during local processing for manual editing, such as notices requiring an uncommon explanation or notices requiring more explanations than the system was designed to use. Also, in limited situations, selects notices found to have national program design errors.

Number of Notices Selected: 2,384

Number of Notices Reviewed: 2,325

Number of Notices Not Reviewed: 59

Error Rate of Notices Reviewed: 92%

Notice Selection Category B: Enables submission processing centers to select specific notices identified by the taxpayer account number or document locator number during local processing for deletion or to select notices with specific situations for additional review, such as by balance due amount, type of return, penalty code, or adjustment reason code.

Number of Notices Selected: 389

Number of Notices Reviewed: 385

Number of Notices Not Reviewed: 4

Error Rate of Notices Reviewed: 31%

Notice Selection Category B/J: The Treasury Inspector General for Tax Administration cannot determine whether these notices were selected in Category B or J because the IRS did not include the Category in the Notice Disposition Report data, and some keys can be selected under either category.

Number of Notices Selected: 2,818

Number of Notices Reviewed: 2,816

Number of Notices Not Reviewed: 2

Error Rate of Notices Reviewed: 78%

Notice Selection Category C: Selects notices with situations where the computer has identified issues requiring additional review. Situations include large refund or balance due amounts or a bill and refund issued at the same time from different returns for the same taxpayer.

Number of Notices Selected: 1,001

Number of Notices Reviewed: 923

Number of Notices Not Reviewed: 78

Error Rate of Notices Reviewed: 28%

Notice Selection Category D: Selects notices involving situations where the IRS payment records differ from the taxpayer’s records and the computer has identified new information that may eliminate the need for the notice.

Number of Notices Selected: 44,872

Number of Notices Reviewed: 41,505

Number of Notices Not Reviewed: 3,367

Error Rate of Notices Reviewed: 9%

Notice Selection Category E: Selects additional notices in which the computer has identified new information that may eliminate the need for the notice.

Number of Notices Selected: 8,609

Number of Notices Reviewed: 6,879

Number of Notices Not Reviewed: 1,730

Error Rate of Notices Reviewed: 41%

Notice Selection Category F: Enables each submission processing center to identify additional notices by type of taxpayer error. Centers can adjust the volume of each type depending on the available resources and quality of the work performed at the center.

Number of Notices Selected: 107,443

Number of Notices Reviewed: 97,899

Number of Notices Not Reviewed: 9,544

Error Rate of Notices Reviewed: 15%

Notice Selection Category G: Selects refund or balance due notices resulting from taxpayer requests after the taxpayer has filed his/her return where the computer has identified issues requiring additional review.

Number of Notices Selected: 2

Number of Notices Reviewed: 1

Number of Notices Not Reviewed: 1

Error Rate of Notices Reviewed: 0%

Notice Selection Category H: Selects refund or balance due notices in which the computer has determined the accounts have similar situations that caused errors on past notices.

Number of Notices Selected: 3,701

Number of Notices Reviewed: 3,394

Number of Notices Not Reviewed: 307

Error Rate of Notices Reviewed: 20%

Notice Selection Category I: Selects refunds in which the computer has found duplicate credit transactions or other indicators that the refund may be incorrect.

Number of Notices Selected: 6,376

Number of Notices Reviewed: 6,326

Number of Notices Not Reviewed: 50

Error Rate of Notices Reviewed: 1%

Notice Selection Category J: Enables each submission processing center to identify additional notices for review by type of return or type of notice. Centers can adjust the volume of each type depending on the available resources and quality of the work performed at the center.

Number of Notices Selected: 6,718

Number of Notices Reviewed: 6,476

Number of Notices Not Reviewed: 242

Error Rate of Notices Reviewed: 39%

Notice Selection Category K: Selects situations in which a taxpayer has not filed an individual tax return but has reported employment tax information for an employee who worked in the taxpayer’s home.

Number of Notices Selected: 95

Number of Notices Reviewed: 54

Number of Notices Not Reviewed: 41

Error Rate of Notices Reviewed: 11%

Associated Notices: Notices that did not meet selection criterion but may be affected by the review of another notice that was selected for the same taxpayer.

Number of Associated Notices Selected: 7,952

Number of Associated Notices Reviewed: 7,418

Number of Associated Notices Not Reviewed: 534

Error Rate of Associated Notices Reviewed: 11%