Executive Summary

Objectives and Scope

Background

Results

Conclusion

Recommendations

Appendix I – Detailed Objectives, Scope, and Methodology

Appendix II – Major Contributors to This Report

Appendix III – Report Distribution List

Appendix IV – Concerns Raised With Management During the Review

Appendix V – Eligibility Criteria for Relief

Appendix VI – Management’s Response to the Draft Report

Executive Summary

The Internal Revenue Service (IRS) considered the provisions for innocent spouse relief to be 1 of the 10 most significant areas of the IRS Restructuring and Reform Act of 1998 (RRA 98) that became law on July 22, 1998. RRA 98 § 3201 liberalized the factors that the IRS considers when evaluating requests for relief from income taxes under the innocent spouse provisions. This liberalization of the law significantly increased the number of claims that the IRS received. Between July 22, 1998, and December 31, 1999, the IRS received approximately 64,000 innocent spouse claims for relief. During this period, the IRS reviewed almost 18,000 claims, of which approximately 9,000 claims were eliminated because they did not qualify for innocent spouse consideration. At the end of December 1999, the IRS still had approximately 46,000 claims open in inventory.

Our review had two objectives. First, we determined how effectively the IRS managed the increased receipt of innocent spouse claims for relief as a result of the RRA 98 § 3201. Second, we determined how effectively the IRS set up a control structure to ensure the quality of determinations. Our review was conducted in two phases. The first phase was conducted between April and September 1999. It was during this time period that the IRS acknowledged there were significant problems with the Innocent Spouse Program. The second phase was conducted in February 2000 as a follow-up to identify and assess the actions taken by the IRS to correct problems identified during phase one.

The IRS was not prepared to effectively process the increased volume of innocent spouse claims for relief received after the RRA 98 was passed. IRS management was slow to react to predictions of increased innocent spouse claims and did not adequately staff the Program when the predictions materialized. This created significant inventory backlogs. Instead of setting up a fundamentally sound Program with long-term solutions, IRS management initiated short-term solutions designed to reduce the inventory backlog. However, these steps were not successful, and cases in open inventory increased to a Program high of 46,000 by December 31, 1999.

The IRS has taken several positive actions to address the problems of implementing the Innocent Spouse Program. These actions included changing the management structure, appointing a full-time national project manager with expanded authority, developing a decision tree job aid to assist in making determinations, redesigning the application for relief to help taxpayers better understand eligibility requirements, and establishing a centralized quality review of innocent spouse case decisions. However, IRS management needs to take additional actions to implement an effective Program to provide timely, accurate determinations on innocent spouse claims for relief.

IRS management did not react to predictions of significant increases in the number of claims that would be filed and was slow to increase staffing to work the claims when the predicted number of receipts actually materialized. Also, management did not set goals for the Innocent Spouse Program and did not have a management information system to measure Program accomplishments. As a result, the IRS did not have accurate or complete information on the volume, status, or location of innocent spouse cases during the first year after passage of the RRA 98 and currently has only limited information on the status of cases in inventory.

We believe that the management structure of the Innocent Spouse Program and the absence of a reliable management information system impacted the IRS’ decision to not react sooner. The IRS has taken steps to address both of these issues by changing the Program’s management structure and increasing its staffing.

However, the IRS expects to reduce the number of open claims from 46,000 to 30,000 by the end of Calendar Year 2000. At that closure rate it will take approximately three years for the IRS to eliminate the inventory backlog. Staffing estimates are also only as reliable as the information used to make the projections. The information currently available is a snapshot of the inventory at a given point in time. Standards have not been established for the overall processing of cases, nor are cases aged to identify how long they have been open.

IRS management did not set standards for the timely and accurate processing of innocent spouse claims and did not establish a comprehensive set of internal controls to measure the quantity, quality, cost, and timeliness of claim processing. As a result, the IRS experienced ongoing problems determining staff needs, making accurate determinations, maintaining accurate taxpayer accounts, effectively communicating with taxpayers, protecting the rights of non-petitioning spouses, and preventing enforced collection activity while claims were being evaluated.

Our findings are consistent with the unpublished results of a review conducted by a multi-functional IRS team. The IRS team included as its primary concerns for the Innocent Spouse Program the need for management controls and an effective management information system.

We recommend that the IRS design a fundamentally sound, long-term approach to making determinations on innocent spouse claims for relief that includes: setting goals and standards for the Innocent Spouse Program; designing a management information system that provides complete, accurate, and useful feedback; and installing a comprehensive system of internal controls that addresses the quantity, cost, and timeliness of making determinations.

Management’s Response: The IRS responded to only two of our three recommendations. It has agreed to formalize statistical goals and standards and to enhance the management information system. The IRS did not respond to our recommendation to design and implement a system of internal controls that addresses the quantity, cost, and timeliness of the Innocent Spouse Program. We continue to be concerned with the controls in place for this Program. We will follow up with the IRS to determine its position on this recommendation since it neither agreed nor disagreed with it. Management’s complete response to the draft report is included as Appendix VI.

Office of Audit Comments: In its response to our draft report, the IRS indicated that we did not address the principal challenge of implementing the innocent spouse provisions: learning to properly and accurately apply new and complex statutory provisions to very sensitive taxpayer situations. However, our report acknowledges that the IRS believes the complexity of the law contributed to problems in implementing the Program. The objectives of our review were to determine how effectively the IRS managed the increased number of claims and the quality of determinations, despite the obstacles.

Regarding the backlog, we were pleased to see in the response to our report that the IRS will reduce the inventory of claims in which the taxpayer has not received a determination letter to 14,000 by the end of the fiscal year. This statistic is considerably better than that in the testimony presented to the Congress less than 3 months ago, when the inventory of pending innocent spouse claims was expected to be less than 30,000 by the end of December 2000. We will continue to monitor the IRS’ progress in this area.

Objectives and Scope

Our review had two objectives. First, we determined how effectively the Internal Revenue Service (IRS) managed the increased receipt of innocent spouse claims for relief as a result of the IRS Restructuring and Reform Act of 1998 (RRA 98) § 3201. Second, we determined how effectively the IRS set up a control structure to ensure the quality of determinations.

The review was conducted in two phases. The first phase consisted of on-site testing at the Centralized Site in the Cincinnati Service Center (CSC) and the National Office. Phase one of the audit was conducted between April and September 1999. It was during this time period that the IRS admitted that there were problems with the Innocent Spouse Program. The second phase was conducted in February 2000 as a follow-up to identify and assess the actions taken by the IRS to correct problems identified during phase one.

The review was conducted in accordance with Government Auditing Standards. Details of our audit objectives, scope, and methodology are presented in Appendix I. Major contributors to this report are listed in Appendix II.

Background

Married taxpayers who file a joint income tax return are jointly and individually responsible for the tax and any interest and penalty due on the joint return. This liability continues even if they later divorce. One spouse can be held responsible for all the tax due even if the other spouse earned all the income.

The Congress was concerned about protecting the innocent spouse who, without knowledge of items on the income tax return, signed a joint return and became personally liable for the errors of the other spouse. This became more critical if the taxpayers divorced and the innocent spouse became the best source for collecting the unpaid tax.

Prior to enactment of the RRA 98, I.R.C. § 6013(e) provided relief to an innocent spouse. However, eligibility requirements under this section were somewhat restrictive. The RRA 98 § 3201 relaxed some of the requirements and added additional provisions to make it easier to qualify for relief. The IRS considered the innocent spouse provisions of the RRA 98 as 1 of the 10 most significant areas in the Act. The new law provides for:

• Innocent spouse relief (I.R.C. § 6015(b))
• Separation of liability (I.R.C. § 6015(c))
• Equitable relief (I.R.C. § 6015(f))

Appendix V lists the criteria to be eligible for the different types of relief.

The law also provides that, while the IRS is considering the request, collection actions and the collection statute are suspended. In addition, taxpayers requesting relief under I.R.C. §§ 6015 (b) or (c) can petition the Tax Court if they disagree with the IRS’ determination.

Results

Overall, the IRS has not established a process to effectively implement the innocent spouse provisions of the RRA 98. The IRS was not prepared to effectively process the increased volume of innocent spouse claims for relief received after the RRA 98 was passed. IRS management was slow to react to predictions of increased innocent spouse claims and failed to adequately staff the Program when the predictions materialized. This created significant inventory backlogs. Between July 22, 1998, and December 31, 1999, the IRS received approximately 64,000 innocent spouse claims for relief. During this period, the IRS reviewed almost 18,000 claims, of which approximately 9,000 claims were eliminated because they did not qualify for innocent spouse consideration. At the end of December 1999, the IRS still had approximately 46,000 claims open in inventory.

Instead of setting up a fundamentally sound Program with long-term solutions, IRS management initiated short-term solutions designed to reduce the inventory backlog. Our findings are consistent with the unpublished results of a review conducted by a multi-functional IRS team. The IRS team included as its primary concerns for the Innocent Spouse Program the need for management controls and an effective management information system.

The IRS has taken several positive actions to address the problems of implementing the Innocent Spouse Program. These actions included changing the management structure, appointing a full-time national project manager with expanded authority, developing a job aid to assist in making determinations, redesigning the application for relief to help taxpayers better understand eligibility requirements, increasing the staffing to work innocent spouse claims, and establishing a centralized quality review of innocent spouse case decisions. However, IRS management needs to take additional actions to implement an effective Program and provide timely, accurate determinations on innocent spouse claims for relief.

The Internal Revenue Service Did Not Effectively Manage the Increased Number of Innocent Spouse Claims for Relief Received as a Result of the Restructuring and Reform Act
of 1998

IRS management failed to heed the predictions of studies conducted by the General Accounting Office (GAO) and the IRS itself that there would be significant increases in the number of claims filed and was slow to increase staffing to work the claims when the predicted number of receipts actually materialized. As a result, inventory became backlogged and open cases grew to an all-time high of 46,000 claims as of December 31, 1999.

More than one source forecasted that innocent spouse claims for relief would skyrocket. As early as March 1997, the GAO issued a report that predicted claims for innocent spouse relief would significantly increase if the law was liberalized.

When the Innocent Spouse Program was centralized at the CSC in April 1998, staffing was authorized at nine positions. After the RRA 98 became law, the IRS expected the number of receipts to increase. To determine what additional staffing was necessary, the IRS conducted a staffing study in the fall of 1998 that concluded 48 positions were needed to process 400 cases per week. Another IRS staffing study also concluded that additional staffing was needed at the Centralized Site to work the growing number of cases.

However, IRS management did not assign additional permanent staffing to the Program. Instead, they waited to see if the number of requests would decrease. Temporary staff was added to the Centralized Site to process innocent spouse claims. However, the use of temporary staff was not effective because there were not enough temporary employees; the permanent staff had to train the temporaries instead of working cases, and temporaries spent the majority of their details (usually 90 days) learning how to work cases.

In April 1999, over 12,000 cases were transferred from the Centralized Site to field offices for Examination Division employees to work. This did not close the cases; it simply shifted the open inventory out of the Centralized Site. By September 30, 1999, the inventory at the Centralized Site was back to over 10,600 cases, while authorized staff remained at 9. During this time, the receipt of innocent spouse claims continued to increase and the inventory of claims continued to grow.

We cannot fully explain why the IRS did not react sooner to the studies that showed the need for increased staffing. However, we believe that there were two significant contributing factors: the management structure over the Innocent Spouse Program and the absence of a reliable management information system (MIS). The IRS attributes many of the problems of implementing the Innocent Spouse Program to the complexity of the law.

The IRS also points to the complexity of the law, the need for its interpretation, and the IRS’ lack of experience in applying it as primary factors contributing to problems in implementing the Program. The RRA 98 added three new innocent spouse provisions to one already in the law. (A provision that already granted relief in community property states was also modified by the RRA 98.) Each of these provisions defines a different set of conditions under which one of the spouses may be granted relief from all or part of the liability arising out of a joint return. Understanding these distinctions and training employees to administer the varying provisions, while also attempting to establish a process for appropriately handling the growing stockpile of cases, proved to be a challenge. The IRS was slow to meet this challenge and is still grappling with the results.

In addition to the challenges inherent in the complexity of the new law, the IRS faced some difficult management issues as the volume of innocent spouse requests increased significantly. This influx of cases, the IRS’s failure to quickly dispel the confusion among its employees as to the proper interpretation of the varying provisions, and management’s initial failure to make changes to an overwhelmed process not only culminated in a high number of pending cases but also resulted in unacceptable results in some cases that were processed.

In April 1998, the IRS centralized innocent spouse case processing under the Examination Branch at the CSC. This unit is part of the Customer Service organization. Then, with the enactment of the RRA 98, the IRS assigned responsibility for implementing the new innocent spouse provisions to the Assistant Commissioner (Examination). At that time, it was expected that the Centralized Site would work the majority of the innocent spouse cases.

Although the Assistant Commissioner (Examination) was responsible for implementing the new innocent spouse provisions, we found no indication that he was given direct or implicit authority over the Centralized Site. This violates the basic management principle that responsibility and authority must be equal.

Without the necessary authority, the Assistant Commissioner (Examination) could not direct the staff at the Centralized Site, adjust staffing levels to ensure that innocent spouse determinations were worked timely, or implement procedures without first obtaining Customer Service approval.

Prior to enactment of the RRA 98, the IRS did not have an MIS that would track innocent spouse cases. In December 1998, the Information Systems function was requested to assist in the development of the Innocent Spouse Tracking System (ISTS). The ISTS was designed to track cases and provide reports on items such as receipts and closures. While the IRS began using the ISTS in March 1999, statistical reports were not produced until September 1999.

Consequently, during the first year after enactment of the RRA 98, the IRS was unable to provide accurate information on the number of cases closed, how they were closed, or the status of the inventory. During testimony before the Congress, the Commissioner provided statistics on the Program that were based on the number of cases received at the Centralized Site and an estimate of the number of cases in the field offices.

Since the Program began, IRS management has not had complete, accurate information on innocent spouse claims. This lack of information, combined with a cumbersome management structure, made it difficult to effectively manage the inventory of innocent spouse claims. The overall result is that the resolution of thousands of innocent spouse claims for relief has been delayed for extended periods of time.

The IRS has taken steps to address both of these issues. During our review, the IRS made changes to the Innocent Spouse Program management structure. The most significant positive change has been the addition of a Program Manager who reports directly to the Chief Operations Officer and has direct control over the Centralized Site at the CSC. However, we have concerns that the Program Manager does not have direct control over Examination resources used to make determinations in the field.

The Program Manager has taken steps to increase resources applied to the Innocent Spouse Program. Staffing at the Centralized Site has been increased to 107 full-time employees who have been fully trained. Also, the IRS plans to assign 153 revenue agents to the Program during Fiscal Year 2000. Our concern is that this may not be enough.

IRS inventory reports show 46,000 innocent spouse claims in open inventory as of December 31, 1999. Based on the Commissioner’s latest testimony before the Congress, the IRS expects to reduce the number of open claims to 30,000 by the end of Calendar Year 2000. At that closure rate, it should take about three years for the IRS to eliminate its inventory backlog. We believe that three years is too long.

The problem is that the staffing estimates are only as reliable as the information used to make the projections. The Program Manager estimates that inventory information from the ISTS will be available to innocent spouse managers by the end of March 2000 in an electronic format that can be used to produce customized reports. However, the information currently available through the ISTS could be best characterized as snapshots of inventory at a given point in time. The following conditions lead us to question the System’s reliability and completeness:

• Standards have not been established for the overall processing of cases or for the various interim stages of processing.
• Cases are not aged to identify how long they have been open or how long they have been in various stages of processing.
• Key dates that could be used to track the timeliness of case processing are being overwritten as claims move from one part of the process to another.
• Inventory reports are not produced on a regular basis.
• Time on cases is not captured by the ISTS.

The Internal Revenue Service Has Not Established a Comprehensive System of Internal Controls to Ensure the Quality of Innocent Spouse Claims for Relief

IRS management is responsible for establishing a comprehensive system of internal controls to ensure quality within the Innocent Spouse Program. Several factors fall under the term quality as it applies to the Program: making accurate claim determinations, posting accurate assessments to taxpayers’ accounts, protecting taxpayers against unwarranted enforced collection activity, and ensuring that both spouses are informed of their rights.

Unfortunately, in its haste to try to reduce the backlog of claims in open inventory, the IRS did not install a comprehensive system of internal controls that addressed the critical factors of the quantity, quality, cost, and timeliness of making determinations. As a result, we identified the following problems within the Innocent Spouse Program:

During the first phase of our audit, we reviewed a sampling of 123 claim determinations made at the Centralized Site after January 1, 1999, and identified 29 (24 percent) where either the decision lacked support or the decision was wrong.

• Fourteen cases lacked documentation or evidence to show that the spouse applying for relief qualified for relief.
• Seven cases were incorrectly disallowed because the taxpayers had pending offers to compromise or settle their liabilities. Taxpayers are ineligible for consideration for innocent spouse relief only if their offers have been accepted.
• Eight cases contained errors of various types.

The IRS has taken steps to improve the quality of determinations. At the examination level, a "decision tree" job aid has been developed to help ensure that determinations are accurate and consistent. This decision tree is a chart that takes examiners through the determination process and lists the criteria to be applied to the case at each step. Pro forma workpapers have been developed for each code section, and formal training has been developed to cover innocent spouse provisions of the law.

A quality review program has been established in the field that currently reviews all determinations made by Examination field personnel. As of January 28, 2000, reviewers did not concur with 21 percent of the determinations. However, current quality review reports do not provide enough detail to identify the specific reason for the nonconcurrence. We must point out that the Program Manager does not believe that these results are representative of the overall quality of determinations because only 551 claims have been reviewed and many field offices have had few, if any, cases reviewed. Although quality review results at this point may only be anecdotal, they indicate that there may be concerns about the quality of determinations.

The Program Manager also informed us that reviewers return a case to the field for correction any time they identify an innocent spouse who was incorrectly denied relief. However, they do not return claims if the taxpayer was incorrectly given relief. We question whether this is an equitable solution to both parties involved in the claim. We believe that the IRS’ position on correcting determinations should be referred to the IRS Office of the Chief Counsel for a ruling.

We are also concerned that there is no proposal to take actions to correct determinations made prior to establishing the quality review process. If the preliminary quality review results are at all indicative of the quality of determinations, the IRS may need to re-evaluate determinations on claims that were closed earlier in the Program and not subjected to the review process.

If the innocent spouse was granted relief, either partially or totally, the change in tax liability cannot be handled routinely using the IRS system which stores taxpayer accounts (Masterfile). Instead, the change must be made using another system the IRS uses to process transactions which cannot be processed on the Masterfile (like these innocent spouse transactions). This system is the Automated Non-Masterfile.

The changes due to innocent spouse determinations must be made at the service center that handles accounts where the taxpayer resides. Unfortunately, procedures to update the accounts are not consistent among the 10 service centers. For example:

• Some service centers require the original case file to process the case while others will accept a copy.
• Some service centers will reissue the final notice of potential enforced collection actions while others will not.

As a result, there were delays in processing the tax liability changes to the Automated Non-Masterfile accounts.

Management has recognized the problems associated with the Automated Non-Masterfile and has requested assistance from IRS data processing specialists. The Information Systems function has been asked to develop a new system for reallocating tax liabilities resulting from innocent spouse claim determinations that will eliminate the need for use of the Automated Non-Masterfile accounting system.

Virtually all of the innocent spouse claims for relief require correspondence with the taxpayer and/or other related parties. Much of this is handled through form letters. One year after the RRA 98 became law, the IRS had not approved 11 different letters drafted for the Innocent Spouse Program. Although the drafts had been reviewed and revised several times, the notices and letters had not been completed as of March 1999 when we conducted our review.

In the interim, the Centralized Site and the field were using locally developed letters or drafts of the unapproved national letters. In some instances, these letters were not designed for use on innocent spouse cases. As a result, some taxpayers received letters that were confusing and some taxpayers received letters that were incomplete.

We also identified that the non-filing spouse was frequently not informed of the pending claim or the results of the determination.

Until late July 1999, the Centralized Site routinely failed to send letters inviting the non-requesting spouse to participate in the determination process. This put the IRS at risk of making incorrect determinations because it was making decisions based on one-sided information.

For example, in one case where the IRS did notify the non-requesting spouse of the determination, that spouse responded that the IRS was in error. In making its determination, the IRS had used a document provided by the requesting spouse that was allegedly signed by the non-requesting spouse. The non-requesting spouse stated the document was forged.

Similarly, the Centralized Site confirmed it did not routinely issue determination letters to the non-requesting spouse. Notifying the non-requesting spouse of the determination results is especially important when relief is allowed and the requesting spouse receives a refund of taxes paid. This will result in an increase in the tax liability of the non-requesting spouse. Notices sent to the non-requesting spouse just show the increased tax due. There is no explanation that the tax increase is based on a favorable determination on an innocent spouse claim for relief.

The IRS decision tree job aid should reduce future communication problems with taxpayers.

The IRS is prohibited from most collection actions while a taxpayer’s innocent spouse request for relief is pending. To prevent collection activity, IRS management chose to input information (via a transaction code (TC), specifically TC 470) to accounts where requests for relief were under consideration. The TC was used to establish a freeze to:

• Stop the issuance of routine collection notices.
• Prevent money from being transferred from other tax years to offset the outstanding liabilities under consideration for relief.
• Remove the account from the system used by the IRS to control certain collection actions (Automated Collection System).

We found that there was no control to ensure that the TC 470 remained in place for all taxpayers with pending claims. In our review of 123 cases closed at the Centralized Site, 23 cases had the freeze code incorrectly reversed by personnel working in locations outside the Centralized Site. As a result, the freeze to prevent the automatic transfer of money among accounts was removed and the IRS computer system automatically offset (used to reduce the tax liability under consideration for relief) $7,000 in refunds that should have been issued to 5 taxpayers.

In January 1999, the IRS decided to add another transaction to identify accounts having requests for relief under consideration. This new transaction was a TC 971. The TC 971 should alert any IRS employee that the case requires special consideration and minimize the risk of freeze codes being incorrectly reversed. However, the element of human error is present as long as the IRS relies on employees recognizing the presence of the TC 971 and reacting correctly.

We are also concerned that the freeze generated by a TC 470 may automatically lapse before the claim determination has been completed. It has taken extended periods of time to process innocent spouse claims and, for most types of cases, the TC 470 freeze to suspend collection activity expires 180 days after it is input.

As a result, the current process requires all, or most, of the thousands of taxpayers’ accounts with innocent spouse requests under consideration to be reviewed or manually monitored to prevent unwarranted collection action.

Conclusion

The IRS has not established a process to effectively implement the innocent spouse provisions of the RRA 98. IRS management failed to heed the predictions of increases in claims for relief and was slow to increase staffing when the predictions materialized. As a result, the inventory of open claims became backlogged. Although IRS management has attempted to take corrective actions, they have not been totally effective. Many of the actions have been specifically designed to reduce the inventory backlogs. This simply attempts to treat the symptoms instead of addressing the root causes of the problems.

We believe that the IRS has to take actions that will be in the best interests of the Innocent Spouse Program on a long-term basis. The Program does not have goals and standards. Even if there were goals and standards, the current management information system is not able to provide complete, accurate, useful feedback on actual performance. Unless the IRS installs a reliable MIS, management will not have accurate information on case receipts, case closures, age of inventory, and time required to close cases. Lacking this information, IRS management will not be able to accurately assess resource needs or determine whether the Program meets its goals.

Finally, IRS management has not installed a comprehensive system of internal controls. Effective internal controls are necessary to not only meet legal requirements but also to minimize the risks of making incorrect determinations, failing to inform taxpayers of their rights, and taking unwarranted enforced collection actions.

Recommendations

The IRS has deemed the Innocent Spouse Program to be 1 of the 10 most important issues resulting from the RRA 98. Accordingly, we recommend that the IRS take the following actions to make the Program fundamentally sound:

1. Set goals and standards for the Innocent Spouse Program.
2. Upgrade the current MIS and ensure that it provides complete, accurate, and useful feedback.
3. Design and implement a system of internal controls that addresses the quantity, cost, and timeliness of the Innocent Spouse Program, to complement the existing internal controls over the quality of the Program.

Management’s Response: The IRS responded to only two of our three recommendations. It has agreed to formalize statistical goals and standards and to enhance the management information system. The IRS did not respond to our recommendation to design and implement a system of internal controls that addresses the quantity, cost, and timeliness of the Innocent Spouse Program. We continue to be concerned with the controls in place on this Program. We will follow up with the IRS to determine its position on this recommendation since it neither agreed nor disagreed with it.

Office of Audit Comments: In its response to our draft report, the IRS indicates that we did not address the principal challenge of implementing the innocent spouse provisions: learning to properly and accurately apply new and complex statutory provisions to very sensitive taxpayer situations. However, our report acknowledges that the IRS believes the complexity of the law contributed to problems in implementing the Program. The objectives of our review were to determine how effectively the IRS managed the increased number of claims and the quality of determinations, despite the obstacles.

Regarding the backlog, we were pleased to see in the response to our report that the IRS will reduce the inventory of claims in which the taxpayer has not received a determination letter to 14,000 by the end of the fiscal year. This statistic is considerably better than that in the testimony presented to the Congress less than three months ago, when the inventory of pending innocent spouse claims was expected to be under 30,000 by the end of December 2000. We will continue to monitor the IRS’ progress in this area.

Detailed Objectives, Scope, and Methodology

Our review had two objectives. First, we wanted to determine how effectively the Internal Revenue Service (IRS) managed the increased receipt of innocent spouse claims for relief as a result of the IRS Restructuring and Reform Act of 1998 (RRA 98) § 3201. Second, we wanted to determine how effectively the IRS set up a control structure to ensure the quality of determinations. The review was conducted in two phases. The first phase consisted of on-site testing at the Centralized Site in the Cincinnati Service Center (CSC) and the National Office. The second phase was a follow-up to identify and assess the actions taken by the IRS to correct problems identified during phase one.

1. To determine how effectively the IRS managed the increased receipt of innocent spouse claims for relief as a result of enactment of the RRA 98, we:

1. Reviewed the National Resource Center’s website, Provision Report, and documents received from the National Office and held discussions with National Office personnel to determine if the IRS met the requirements spelled out in the legislation to issue regulations and revise documents.
2. Analyzed weekly inventory reports from the Centralized Site for July 24, 1998, through July 31, 1999, and discussed staffing needs with appropriate personnel to determine if the IRS budgeted sufficient resources to handle the expected increased volume of innocent spouse claims.
3. Reviewed the National Resource Center’s website, other internet sites, local service center procedures, and draft Internal Revenue Manuals and held discussions to determine if the IRS effectively and timely communicated the new innocent spouse provisions and procedures to employees who would be working the cases.
4. Identified and reviewed training packages developed for the RRA 98 § 3201.
5. Analyzed overall review results to determine if the IRS has experienced any unintentional/unexpected consequences as a result of implementing provisions of the RRA 98 § 3201, including additional costs associated with the new requirements.
6. Reviewed communication published in the media by checking "In the News" on the IRS headquarters homepage and held discussions with the National Innocent Spouse Coordinator and the current Innocent Spouse Program Manager (ISPM) to determine if the IRS effectively pursued communicating the options for innocent spouse relief and eligibility requirements for the various types of relief.
7. Obtained and reviewed 11 different draft notices from the National Office to determine if the national innocent spouse notices are clear and concise.
8. Reviewed Requests for Information Services (RIS) and suggested changes to draft Innocent Spouse Tracking System (ISTS) reports.
9. Obtained an Integrated Data Retrieval System (IDRS) extract of innocent spouse cases as of April 4, 1999. This extract contained 16,368 innocent spouse cases.
10. Determined if notices related to the Innocent Spouse Program were clear.

1. Reviewed examples of IDRS-generated local letters developed by the Centralized Site.
2. Reviewed letters identified in our case review.

1. Interviewed the ISPM to obtain updated information on actions the IRS had taken to correct problems identified in Phase I.
2. Reviewed ISTS reports as of December 31, 1999, to determine current inventory levels.
3. Determined the status of RIS requests for programming changes to the ISTS.

1. To determine how effectively the IRS set up a control structure to ensure the quality of determinations, we:

1. Identified and reviewed procedures and controls for innocent spouse cases.
2. Reviewed a sampling of 123 cases from 1,471 innocent spouse cases at the CSC that had received an innocent spouse determination after January 1, 1999. We reviewed these cases to determine if:

1. Transaction Code (TC) 470s were properly input.
2. Any adverse effects resulted if TC 470s were not timely or properly input.
3. The case determination was consistent with the Internal Revenue Code, Regulations, and IRS training and instructional guidelines.
4. The non-requesting spouse was treated fairly and consistently by looking for any indication in the case files that the non-requesting spouse was notified or participated in the proceedings.

3. Compared the results of our case review to information entered into the CSC Access database to assess its accuracy.
4. Reviewed 21 cases, from our sample in step II.B, where the IRS made a determination of accepting, or partially accepting, the innocent spouse claim and the case required a transfer to the Automated Non-Masterfile. Reviewed these Automated Non-Masterfile cases to determine the time it takes to complete the transfer.
5. Discussed non-requesting spouse participation with National Office and Centralized Site personnel.

1. Interviewed the ISPM to obtain updated information on actions the IRS had taken to correct problems identified in Phase I.
2. Reviewed results of the Centralized Site Innocent Spouse Review to determine trends in the quality of case determinations.

Major Contributors to This Report

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

Susan Boehmer, Director

Gary Lewis, Director

Ronald Koperniak, Audit Manager

Stephen Root, Audit Manager

David Brown, Senior Auditor

Tom Cypert, Senior Auditor

John Hawkins, Senior Auditor

Michael Hillenbrand, Senior Auditor

Donald McDonald, Senior Auditor

Joseph Snyder, Senior Auditor

Donald Evans, Auditor

Matthew Miller, Auditor

Janice Murphy, Auditor

Elizabeth Stout, Auditor

Ahmed Tobaa, Auditor

Report Distribution List

Deputy Commissioner Operations C:DO

Office of the Chief Counsel CC

Chief Operations Officer OP

Deputy Chief Operations Officer OP

Assistant Commissioner (Customer Service) OP:C

Assistant Commissioner (Examination) OP:EX

Assistant Commissioner (Forms and Submission Processing) OP:FS

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

Director, Taxpayer Treatment and Service Improvements C:DO:TSI

National Chief of Appeals C:AP

National Director for Legislative Affairs CL:LA

Office of Management Controls M:CFO:A:M

Office of the National Taxpayer Advocate C:TA

Director, Cincinnati Customer Service Center C:CSC

Innocent Spouse Project Manager OP:EX:IS

Concerns Raised With Management During the Review

At the opening conference on April 1, 1999, the Treasury Inspector General for Tax Administration and the Internal Revenue Service (IRS) executive responsible for implementing the Innocent Spouse Program agreed to use e-mail to communicate as quickly as possible any problems the audit staff identified. The following summarizes these communications during the audit.

Date issued Concern raised

April 21, 1999 Two separate staffing studies had already determined that staffing at the centralized processing site in the Cincinnati Service Center was not adequate for the workload. From January through March 1999, Examination and Customer Service discussed the growing inventory of unanswered requests for relief but took little action to resolve it.

April 26, 1999 Procedures were incorrect for suspending collection on accounts in the Automated Collection System at the time a taxpayer filed a request for innocent spouse relief. As a result, collection actions could resume before the IRS made the innocent spouse determination.

May 6, 1999 The Customer Service function’s recently issued Manual had inadequate instructions for:

• Taxpayer use of Request for Innocent Spouse Relief (Form 8857).
• Taxpayer inquiries about or references to previously filed Forms 8857.

May 28, 1999 In some cases, there was a systemic release of the suspension of collection action and the claimant’s refund was offset into the account the individual was requesting relief from. In one case, the refund offset fully paid the delinquent account and the IRS subsequently disallowed the relief request because there was no longer an outstanding liability.

June 9, 1999 Ten months into the new Program the IRS still had not approved the various letters that would be used to correspond with taxpayers about their innocent spouse requests. As a result, the IRS offices were corresponding with ad hoc letters.

June 22, 1999 From our review of cases, we noted instances where the procedures were unclear and provisions of the law were misapplied. For example, requests for relief were incorrectly disallowed when:

• Offers in Compromise were pending.
• The IRS cited taxpayer knowledge but did not prove it, as required under Internal Revenue Code (I.R.C.) § 6015(c).
• The IRS cited denial because I.R.C. § 6015(f) applied only to underpayment, not understatement.
• The IRS cited denial because I.R.C. § 6015 did not allow claims of less than $100.

August 11, 1999 The letter used to notify the non-petitioning spouse of the request did not adequately advise of the adverse consequences of not responding.

Eligibility Criteria for Relief

Innocent Spouse Relief 26 U.S.C. § 6015(b) (1986)

Generally, an individual will be relieved of liability for tax (including penalty, interest, and other amounts) to the extent the liability is attributable to the understatement, if the individual meets the following criteria:

• A joint return was filed for the taxable year.
• There is an understatement of tax attributable to erroneous items of one individual filing the joint return.
• The other individual filing the joint return establishes that in signing the return he or she did not know, and had no reason to know, that there was an understatement.
• Taking into account all the facts and circumstances, it is inequitable to hold the other individual liable for the tax.
• The other individual files a claim within two years after collection activity begins.

Separation of Liability 26 U.S.C. § 6015(c) (1986)

A taxpayer may elect to have the liability for any deficiency limited to the portion of the deficiency that is attributable to items allocable to that taxpayer. In general, the items are allocated as if the taxpayers had filed separate returns. To qualify, at the time of the election the taxpayer must be either no longer married to, legally separated from, or living apart for at least 12 months from the person with whom the taxpayer originally filed a joint return. If the IRS establishes that the individual had actual knowledge of the item resulting in the understatement, then relief is not available to the extent of the actual knowledge.

Equitable Relief 26 U.S.C § 6015(f) (1986)

A taxpayer may also request equitable relief. Primarily, this relief is available where the tax was shown on a joint return but not paid with the return. For example, the taxpayer did not know, or have reason to know, that funds intended for the payment of the tax were taken by the other spouse for that spouse’s benefit. Relief is also available where there is an understatement of tax for which relief under the above two situations is not available.

To qualify for equitable relief, the following conditions must be met:

• Taking into account all the facts and circumstances, it would be unfair to hold the individual liable for the understatement or underpayment of tax.
• The individual is not eligible for innocent spouse relief.
• The individual is not eligible for separation of liability.

Management’s Response to the Draft Report

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