TREASURY INSPECTOR GENERAL FOR TAX ADMINISTRATION

 

 

 

Reliance on New Procedures Allowed Some False Economic Stimulus Payments to Be Issued and Pose a Continuing Risk of False Refunds Being Issued in the Future

 

 

 

March 25, 2009

 

Reference Number:  2009-10-049

 

 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.

 Redaction Legend:

1 = Tax Return/Return Information

 

 

Phone Number   |  202-622-6500

Email Address   |  inquiries@tigta.treas.gov

Web Site           |  http://www.tigta.gov

 

March 25, 2009

 

 

MEMORANDUM FOR DEPUTY COMMISSIONER FOR SERVICES AND ENFORCEMENT

 

FROM:                            Michael R. Phillips /s/ Michael r. Phillips

Deputy Inspector General for Audit

 

SUBJECT:                    Final Audit Report – Reliance on New Procedures Allowed Some False Economic Stimulus Payments to Be Issued and Pose a Continuing Risk of False Refunds Being Issued in the Future (Audit # 200810032)

 

This report presents the results of our review of the Economic Stimulus Program.  Our overall objective was to evaluate the effectiveness of the Criminal Investigation (CI) Division’s actions to prevent the issuance of economic stimulus payments for tax returns identified by the Internal Revenue Service (IRS) as claiming false refunds.  This audit was conducted as part of Treasury Inspector General for Tax Administration’s risk-based audit coverage of the CI Division addressing the major management challenge of Erroneous and Improper Payments.

Impact on the Taxpayer

As of December 2008, the IRS distributed 119.2 million stimulus payments totaling $96.3 billion.  However, we identified $1.2 million in false stimulus payments and $16.3 million in false refunds that were inappropriately issued due to reliance on new controls that did not work effectively.  Although the amount of erroneously issued stimulus payments we identified is small in relation to the total amount issued, we remain concerned that the existing controls pose a risk that false refunds and stimulus payments will be erroneously issued in the future.  Allowing false stimulus payments or false refunds to be issued reduces the dollars in the United States Treasury, which is especially critical in this current economic environment.

Synopsis

The Economic Stimulus Act of 2008, signed on February 13, 2008, was passed to energize the national economy.  The most significant part of the Act is the individual stimulus payment, which is a credit for Tax Year 2008.  Overall, the payments were estimated using information reported on Tax Year 2007 tax returns and were scheduled to be issued in 2008[1] so individuals could benefit from the payments as soon as possible.  A stimulus payment will not increase the amount of tax an individual owes or reduce an individual’s refund for Tax Year 2008.

In January 2008, the CI Division began looking into the impact of the Economic Stimulus Payment Act legislation on the Questionable Refund Program.[2]  CI Division personnel actively participated in the Service-Wide Executive Steering Committee meetings throughout the planning process.  In addition, CI Division management coordinated with officials in the Wage and Investment Division and Modernization and Information Technology Services organization to coordinate procedural changes and revise applicable programming to prevent the issuance of stimulus payments associated with false returns.  Further, the CI Division developed a plan for key activities to prevent stimulus payments on known false returns and developed ways to identify potential refund schemes associated with “stimulus-only” returns.

However, the CI Division’s Action Plan did not adequately ensure that stimulus payments were not issued on known false returns and generally relied on new procedures to prevent payments associated with false returns from being issued.  In Processing Year (PY) 2008, the CI Division changed the controls and implemented a temporary freeze to hold the issuance of potentially false refunds.  However, our limited testing identified that $1.2 million in stimulus payments and $16.3 million in false refunds were inappropriately issued because the temporary freeze was released due to processing delays or systemic deficiencies.

We previously raised concerns with the effectiveness of the controls designed to prevent false refunds from being erroneously issued.  In our first report evaluating the Questionable Refund Program, we raised concerns that the programmed expiration for the temporary freeze may not give the IRS sufficient time to verify refunds.[3]  In our most recent report, we reported that the process for freezing taxpayer accounts was complex and confusing, which resulted in the erroneous issuance of false refunds.[4]  If CI Division management does not take adequate steps to monitor the false refund inventory and ensure false refund returns are referred for resolution before the temporary freeze expires, we estimate about $117.6 million[5] in false refunds could be erroneously released over the next 5 years.

The amount of erroneously issued stimulus payments that we identified is relatively small in relation to the total stimulus payments issued.  However, we remain concerned that the existing controls pose a risk that false refunds, as well as any stimulus payments associated with potential future legislation, will be erroneously issued in PY 2009 and beyond.

Finally, we estimate that more than $138 million in stimulus payments on false returns that were not issued in 2008 may be at risk of being issued during PY 2009 if false returns for these taxpayers are filed during the 2009 Filing Season (PY 2009).  Currently, any stimulus payments not issued during 2008 that are claimed on potentially false returns filed during the 2009 Filing Season must be referred to the Examination function for resolution.  This could place an extreme burden on the limited resources and other noncompliance priorities of the Wage and Investment Division during PY 2009.  We believe this high volume of returns, coupled with limited resources and other competing priorities, could cause the IRS to increase the dollar tolerance of the potentially false stimulus payments referred to Examination function personnel.  As a result, there is an increased risk that some of these stimulus payments could be issued.

Recommendations

We recommended that the Chief, CI, 1) extend the time period for the release of the temporary freeze on taxpayer accounts to provide CI Division personnel more time to verify potentially false returns and prevent the issuance of false refunds, and 2) establish a quality review process that will identify false returns that are not automatically transferred to the Scheme Tracking and Referral System within the required time period and establish a process to identify false returns that are not timely referred for resolution.

We also recommended that the Deputy Commissioner for Services and Enforcement seek a Counsel opinion to determine whether a rebate recovery credit calculated by the IRS requires the issuance of a statutory notice of deficiency.

Response

IRS management agreed with one of our three recommendations, and disagreed with the remaining two.  IRS management acknowledged the seriousness of the threat posed by refund fraud and the need for corrective actions.  CI Division management stated they remain committed to identifying and stopping the issuance of false refunds, in support of the IRS’ efforts to ensure that appropriate systems are in place to not only protect the tax system from vulnerability, but also to protect taxpayer rights.  IRS management agreed with recommendation three to request a formal written opinion from Chief Counsel to determine whether a rebate recovery credit calculated by the IRS requires a statutory notice of deficiency.

IRS management did not agree with recommendations one and two regarding extending the time period for the release of the temporary freeze and establishing a quality review process.  The IRS stated that in response to the same recommendation in our prior report,[6] the prior manually intensive process was replaced for PY 2008 with an automated, streamlined process that negates the need for manual intervention in placing controls on a return once it has been verified as false.  IRS management stated other freeze conditions are used to hold the refund while the automated referral process is completed and that it continues to monitor and has taken proactive measures to ensure that procedures are in place to hold the refund until all issues are resolved.  In addition, IRS management disagreed with the need to establish a quality review process, stating that the Office of Refund Crimes has always taken and continues to take proactive steps to monitor and refine processes to resolve issues.  In FY 2008, system enhancements were implemented to automate the referral process, and a Business Processing Improvement Team was established to monitor workload transfer processes, identify issues, and implement solutions.

Finally, IRS management expressed concern about our estimate of the approximate value of false refunds that could be erroneously released over the next 5 years, stating it is biased in its assumptions and questioning its accuracy.  Management’s complete response to the draft report is included as Appendix VII.

Office of Audit Comment

We previously reported that the temporary freeze may not give the IRS sufficient time to verify the refunds and that the freeze process used by the CI Division was complex and confusing.  In response to these findings, the CI Division indicated it would review the data to obtain more current information pertaining to the wage verification process and implemented new processes in PY 2008 that would prevent false refunds from being issued until all issues were resolved by the IRS.  However, in spite of the IRS’ new automated process and other freeze conditions designed to hold the refund, we identified instances where false refunds were erroneously issued due to the expiration of the temporary freeze within the established time periods.  We also identified control weaknesses in the CI Division’s processes.  Regarding the quality review process, the CI Division outlined several improvements it has made to monitor the referral process and ensure issues are timely resolved.  While we believe some of the actions taken by the CI Division could be an improvement, these actions will not be implemented until PY 2009.  As a result, we cannot comment on their effectiveness.  Finally, because we found that the controls were not always working as the IRS believed they would, if the IRS does not take corrective actions, it will continue to be at risk of allowing false refunds to be issued.  As such, we remain concerned that the programmed expiration for the temporary freeze may not give the IRS sufficient time to verify wages and refer false refund returns for resolution.

We acknowledge that our estimate of false refunds that could be released over the next 5 years does not account for changes in volumes of false refund returns, average refund amounts, improvements in the IRS’ processing of false refunds, or future economic conditions.  However, both the number of false returns and the average amount of false refunds has increased 177 and 36 percent, respectively, over the last 3 years.  We have no reason to believe that these numbers will decrease in the near future.  As a result, we believe our estimate is valid, and could actually underestimate the amount of false refunds that could be erroneously issued during this time period if the trend in false returns and false refunds continues.

Copies of this report are also being sent to the IRS managers affected by the report recommendations.  Please contact me at (202) 622-6510 if you have questions or Nancy A. Nakamura, Assistant Inspector General for Audit (Management Services and Exempt Organizations), at (202) 622-8500.

 

 Table of Contents

Background

Results of Review

The Criminal Investigation Division’s Initial Action Plan Was Inadequate and Did Not Fully Prevent the Issuance of Stimulus Payments Associated With False Returns

The Implementation of New Procedures Resulted in the Issuance of Some Stimulus Payments on False Returns

Recommendation 1:

Recommendation 2:

Additional Stimulus Payments Could Be Issued on False Returns During the 2009 Filing Season

Recommendation 3:

The Criminal Investigation Division Took Action to Identify Potential Refund Schemes on Stimulus-Only Returns

 

Appendices

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 – Synopsis of Account Freezes

Appendix VI – Glossary of Terms

Appendix VII – Management’s Response to the Draft Report

 

Abbreviations

 

CI

Criminal Investigation

IRS

Internal Revenue Service

PY

Processing Year

STARS

Scheme Tracking and Referral System

TY

Tax Year

  

Background

 

The Economic Stimulus Act of 2008,[7] signed on February 13, 2008, was passed to energize the national economy.  The most significant part of the Act is the individual stimulus payment, which is a credit for Tax Year (TY) 2008.  Overall, the payments were estimated using information reported on TY 2007 tax returns and were scheduled to be issued in 2008[8] so individuals could benefit from the payments as soon as possible.  Individuals who qualify for a larger payment as a result of changes between their TY 2007 and TY 2008 returns will receive the additional payment when they file their TY 2008 returns (generally between January and April 2009).  If individuals received a larger stimulus payment in 2008 than they would have if the payment had been calculated using information from their TY 2008 returns, they will be allowed to keep the excess and will not be asked to pay it back.  The stimulus payment will not increase the amount of tax an individual owes or reduce an individual’s refund for TY 2008.

To receive a stimulus payment, individuals must have an income tax liability or at least $3,000 in qualifying income.  Income from wages, tips, and net self-employment earnings as well as nontaxable combat pay and some Social Security, Veterans disability, and Railroad Retirement benefits qualifies as eligible income.  Individuals must file a 2007 tax return with a valid Social Security Number to receive the payment.

As of the end of December 2008, the IRS distributed 119.2 million stimulus payments totaling $96.3 billion.

The legislation required that the Internal Revenue Service (IRS) issue payments to taxpayers before December 31, 2008.  As of the end of December 2008, the Department of the Treasury reported that a total of 119.2 million payments had been distributed totaling $96.3 billion.

Implementing the economic stimulus payment presented two significant challenges for the IRS:

·        The process for issuing payments had to be implemented at the same time the IRS was processing an estimated 140 million individual income tax returns as part of its annual filing season.[9]

·        All affected programs and computer systems had to be modified in a relatively short time period to satisfy the intent of Congress to issue the payments to individuals as soon as possible.

The Criminal Investigation (CI) Division, whose responsibility includes detecting and stopping false claims for refunds on income tax returns, began evaluating the impact of the economic stimulus payment legislation in January 2008.  While the CI Division reflected on best practices and lessons learned from a similar law passed in 2001, there were a number of differences that made preparing for this current initiative unique.  First, the IRS dealt with an accelerated time period for issuing the stimulus payments.  Second, millions of taxpayers who were usually exempt from filing a tax return would be required to file a stimulus-only tax return to receive the stimulus payment.

In 2008, we issued a report that evaluated the IRS’ efforts in planning for the issuance of the stimulus payments.[10]  We reported that although the IRS’ planning for the stimulus payments was generally sufficient, there were areas where improvements were needed.  In that report, we noted weaknesses in the CI Division’s Action Plan and reported that the CI Division’s Action Plan did not address the need to develop procedures for stopping stimulus payments relating to certain TY 2007 false returns for which a determination of fraud was made after the refunds were issued.

This review was performed at the CI Division’s Office of Refund Crimes in Washington, D.C., during the period May through December 2008.  We conducted this performance audit in accordance with generally accepted government auditing standards.  Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objective.  We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objective.  Detailed information on our audit objective, scope, and methodology is presented in Appendix I.  Major contributors to the report are listed in Appendix II.

  

Results of Review

 

We determined CI Division personnel actively participated in the Service-Wide Executive Steering Committee meetings throughout the planning process and developed an Action Plan for key activities within the CI Division’s Questionable Refund Program.  In addition, CI Division management coordinated with officials in the Wage and Investment Division and Modernization and Information Technology Services organization to coordinate procedural changes and revise applicable programming to prevent the issuance of stimulus payments associated with false returns.  Further, the CI Division initiated actions that resulted in the identification of potential refund schemes associated with “stimulus-only” returns filed to obtain the stimulus payment.

However, we determined the Action Plan submitted by the CI Division to the Executive Steering Committee was not adequate.  Specifically, we identified several concerns with the CI Division’s Action Plan that would not prevent false stimulus payments from being issued on known false returns.  After we discussed these concerns with IRS management, the IRS agreed to take additional steps to prevent the issuance of false stimulus payments.  As a result of these additional changes, we estimate that at least 12,000 false stimulus payments totaling $7.1 million were prevented from being issued.

We also determined that the CI Division’s Action Plan generally relied on new procedures to prevent payments associated with false returns from being issued.  In Processing Year (PY) 2008, the CI Division changed the controls and implemented a temporary freeze to hold the issuance of potentially false refunds.  However, our limited testing identified that $1.2 million in stimulus payments and $16.3 million in false refunds were inappropriately issued because the temporary freeze was released due to processing delays or systemic deficiencies.

We previously raised concerns with the effectiveness of the controls designed to prevent false refunds from being erroneously issued.  In our first report evaluating the Questionable Refund Program, we raised concerns that the programmed expiration for the temporary freeze may not give the IRS sufficient time to verify refunds.[11]  In our most recent report, we reported that the process for freezing taxpayer accounts was complex and confusing, which resulted in the erroneous issuance of fraudulent refunds.[12]

The amount of erroneously issued stimulus payments that we identified is relatively small in relation to the total stimulus payments issued.  However, we remain concerned that the existing controls pose a risk that false refunds, as well as any stimulus payments associated with potential future legislation, will be erroneously issued in the 2009 Filing Season (PY 2009) and beyond.  We estimate that approximately $117.6 million[13] in false refunds could be erroneously released over 5 years due to the expiration of the temporary freeze within the currently established time periods.  Allowing false stimulus payments or false refunds to be issued reduces the dollars in the United States Treasury, especially in this current economic environment.

Finally, we estimate that more than $138 million in stimulus payments on false returns that were not issued in 2008 may be at risk of being issued during PY 2009 if false returns for these taxpayers are filed during 2009.  Currently, any stimulus payments not issued in 2008 that are claimed on potentially false returns filed during 2009 must be referred to the Examination function for resolution.  This could place an extreme burden on the limited resources and other noncompliance priorities of the Wage and Investment Division during PY 2009.  We believe this high volume of returns, coupled with limited resources and other competing priorities, could cause the IRS to increase the dollar tolerance of the potentially false stimulus payments referred to Examination function personnel.  As a result, there is an increased risk that some of these stimulus payments could be issued.

The Criminal Investigation Division’s Initial Action Plan Was Inadequate and Did Not Fully Prevent the Issuance of Stimulus Payments Associated With False Returns

In January 2008, the CI Division began looking into the impact of the Economic Stimulus Payment Act legislation on the Questionable Refund Program.  The CI Division developed a plan for key activities to prevent stimulus payments on known false returns and develop ways to identify schemes.  However, the CI Division’s Action Plan did not adequately ensure that stimulus payments were not issued on known false returns.

For example, the CI Division’s plan did not have procedures to stop stimulus payments on returns determined to be false after the refund was issued.  These accounts would not have any freeze input to indicate a potentially fraudulent return.  In addition, the CI Division did not consider how to prevent stimulus payments associated with those false returns that had been resolved (and the false refund properly stopped) through the Accounts Management organization or Examination function processes.  For these returns, the accounts would reflect a zero balance, thereby causing the civil freeze control to systemically release.  Without any freeze controls present on these accounts, the stimulus payment would be systemically issued to the taxpayer.

After bringing these concerns to management’s attention, the IRS immediately took corrective action and implemented procedures to prevent the issuance of stimulus payments on these returns.  Specifically, Office of Refund Crimes management issued a memorandum to its staff advising them to freeze accounts that have been verified as false after the refunds were issued to prevent the stimulus payment from issuing.  In addition, the IRS developed a Master File program to identify and stop stimulus payments for those returns where the false refund was stopped and the account adjusted before the stimulus payment posted.  As a result of these changes, we conservatively estimate[14] that about 12,000 stimulus payments totaling $7.1 million were prevented from being issued.

The Implementation of New Procedures Resulted in the Issuance of Some Stimulus Payments on False Returns

In PY 2008, the CI Division implemented new controls to prevent the issuance of refunds associated with false returns.[15]  These new controls relied on a temporary freeze to hold the issuance of potentially false refunds.  We previously raised concerns with the effectiveness of these controls in two prior audit reports.  However, our review of operational documents determined that the CI Division believed that these new controls would prevent the issuance of any stimulus payments associated with false returns.

The temporary freeze is programmed to expire within a certain time period.  Thus, if a false refund account is not timely worked or referred for resolution, the account is at risk of having the refund erroneously issued because the temporary freeze expires.  This process significantly differs from PY 2007, when the CI Division used a criminal freeze to prevent the issuance of false refunds.  Generally, CI Division personnel must manually release the criminal freeze (i.e., the criminal freeze does not systemically release).

As previously discussed, we identified two situations in which the CI Division’s new control process would not have prevented the stimulus payments from issuing.  In addition, we identified other systemic deficiencies or errors that resulted in the erroneous issuance of stimulus payments due to the expiration of the temporary freeze.

We reviewed 35 accounts in which the CI Division verified the returns were false and determined that the temporary freeze automatically released on 13 accounts resulting in the issuance of $6,900 in false stimulus payments and $49,973 in false refunds.  The high error rate (37 percent) on this sample suggested that this problem might be more widespread than the CI Division originally thought.  Therefore, we obtained account information for approximately 173,000 returns that the CI Division determined were false (as of June 3, 2008) for which a stimulus payment had been generated.  Our analyses identified 3,286 accounts in which the temporary freeze expired resulting in the issuance of $16.3 million in false refunds and 2,071 false stimulus payments totaling $1.2 million.

We identified several factors that allowed these false payments to be issued.  Specifically:

  • The CI Division implemented an automated referral process in PY 2008 to systemically transfer verified false returns from the Electronic Fraud Detection System to the Scheme Tracking and Referral System (STARS) if they are not manually placed in STARS for 7 or more days after being verified as false.  This process is intended to speed up the movement of returns to the Examination function or Accounts Management organization before the temporary freeze holding the false refund expires.

However, as of June 3, 2008, we identified 1,482 false returns that were not timely transferred into the STARS.  This delay increases the risk that the temporary freeze designed to prevent issuance of the false stimulus payments will expire.  We determined these returns either were not entered into or were not in the STARS for an average of 109 days.  We identified 746 stimulus payments totaling $449,000 that were issued because the temporary freeze expired.

  • Once the false returns are entered into the STARS, the Electronic Fraud Detection System is programmed to identify returns in the STARS eligible for referral to the Examination function or the Accounts Management organization, transfer referral information to the Dependent Database for filtering and initiation of Examination function and Accounts Management organization computer processes, and automatically update the STARS Return Dispositions returns referred for resolution.  This process is dependent upon and begins when the false returns are entered into the STARS and assigned a disposition code.  However, we identified 14,015 false returns from the STARS that did not have a disposition code.  These returns had been entered in the STARS an average of 47 days and a maximum of 140 days at the time we reviewed them.  We identified 4,560 false returns in which almost $2.8 million false stimulus payments were issued.

We brought our concerns to the attention of CI Division management and asked for an explanation of why this occurred and what procedures or controls were developed to ensure timely updates to the STARS.  According to Office of Refund Crimes management, there are various points in the automated referral process that could impact the status of the false refund.  Management believed some of the more common breakdowns that occurred and caused the issuance of false refunds and stimulus payments included human error and inventory backlog.  In addition, we believe the lack of any quality review testing by CI Division management of the new controls and procedures implemented in PY 2008 impacted management’s ability to identify control weaknesses.

This is now the third processing year we raised concerns with the effectiveness of the controls designed to prevent false refunds from erroneously issuing.  In our first report evaluating the Questionable Refund Program, we raised concerns that the programmed expiration for the temporary freeze may not give the IRS sufficient time to verify refunds.[16]  In our most recent report, we reported that the process for freezing taxpayer accounts was complex and confusing, which resulted in the erroneous issuance of false refunds.[17]  In response to those recommendations, CI Division management indicated it would review 2007 data to obtain more current time periods for completing verification, and stated the new process implemented in PY 2008 would prevent false refunds from issuance until all issues were resolved by IRS.  However, we do not believe these actions have been effective.  If CI Division management does not take adequate steps to monitor the false refund inventory and ensure false refund returns are referred for resolution before the temporary freeze expires, we estimate about $117.6 million[18] in false refunds could be erroneously released over the next 5 years.

Office of Refund Crimes management recently advised that it has taken proactive steps to monitor and refine the automated referral process to resolve these issues, such as establishing regular and periodic calls to review problems and issues with workload-related automation and developing new monitoring reports to track referrals for PY 2009.  Additionally, CI Division management stated a business processing improvement team has been established to monitor workload transfer processes to identify issues and recommend solutions, but did not provide any specific details on how the automated referral process will be monitored to identify potential delays.  Further, the Office of Refund Crimes requested that the automated transfer process of returns from the Electronic Fraud Detection System to the STARS be extended to 14 days for PY 2009.  In addition, a requirement has been added to the Electronic Fraud Detection System to give the CI Division flexibility to change the 14 days during the filing season if evaluation of the process shows it is necessary.  However, we do not believe extending the time to transfer returns into the STARS will reduce the likelihood that the temporary freeze will expire prior to issuance of false refunds.  Because the Office of Refund Crimes has only a certain time period to verify and refer the cases, extending the automatic referral time only shortens the available time for completing the referral process before the temporary freeze expires.

We remain concerned that if these weaknesses do not get corrected for the next filing season (PY 2009), additional false refunds will be at risk of being erroneously issued.  Further, with the uncertainty surrounding our current economic conditions, several Congressmen have recently discussed the need for another stimulus package.  While legislation had not been enacted as of the completion of our fieldwork, it is a possibility.  Therefore, the IRS needs to address the concerns we identified in this report and ensure that effective controls are established to prevent the erroneous issuance of refunds and stimulus payments associated with false returns.

Recommendations

The Chief, CI Division, should:

Recommendation 1:  Extend the time period for the release of the temporary freeze on taxpayer accounts to provide CI Division personnel more time to verify potentially false returns and prevent the issuance of false refunds.

Management’s Response:  IRS management disagreed with this recommendation, stating they responded to this same recommendation in our prior report.[19]  The prior, manually intensive process was replaced for PY 2008 with an automated, streamlined process that negates the need for manual intervention in placing controls on a return once it has been verified as false.  IRS management stated several issues were identified early in the processing year and they took immediate action, making the necessary adjustments to ensure the workflow continued uninterrupted and to prevent the issuance of fraudulent refunds.  In addition, other freeze conditions are used to hold the refund while the automated referral process is completed.  The CI Division continues to monitor and has taken proactive measures to ensure that procedures are in place to hold the refund until all issues are resolved.

Office of Audit Comment:  We previously reported that the temporary freeze may not give the IRS sufficient time to verify the refunds, and the freeze process used by the CI Division was complex and confusing.  In response to these findings, the CI Division indicated it would review data to obtain more current information pertaining to the wage verification process and implemented new processes in PY 2008 that would prevent false refunds from issuance until all issues were resolved by the IRS.  However, in spite of the IRS’ new automated process and other freeze conditions designed to hold the refund, we identified instances where false refunds were erroneously issued due to the expiration of the temporary freeze within the established time periods as well as control weaknesses in the CI Division’s processes.  If the IRS does not take corrective actions, the IRS is still at risk of allowing false refunds to be issued.  As such, we remain concerned that the programmed expiration for the temporary freeze may not give the IRS sufficient time to verify wages and refer false refund returns for resolution.

Recommendation 2:  Establish a quality review process that will identify any false returns that are not automatically transferred to the STARS within the required time period.  In addition, the CI Division should establish a process to identify false returns within the STARS that are not timely referred to the Accounts Management organization or the Examination function for resolution.

Management’s Response:  The IRS disagreed with this recommendation, stating that the Office of Refund Crimes has always taken and continues to take proactive steps to monitor and refine processes to resolve issues.  In FY 2008, system enhancements were implemented to automate the referral process and a Business Processing Improvement Team was established to monitor workload transfer processes, identify issues, and implement solutions.  IRS management cited the following actions that have been taken:  1) established regular calls with representatives from all affected areas to review problems and issues; 2) developed new reports to track referrals for use by the CI Division, Accounts Management organization, and Examination function management; 3) used the STARS data perfection process to improve quality and accuracy of verified false returns prior to the referral; 4) extended the STARS auto-push delay; and 5) identified reasons for referral rejects, implemented revised conditions to reduce reject volumes, and developed streamlined procedures for reject resolution.

Office of Audit Comment:  While we believe some of the actions taken by the CI Division could be an improvement, these actions will not be implemented until PY 2009.  As a result, we cannot comment on their effectiveness.  If the IRS does not take the additional corrective actions, the IRS is still at risk of allowing false refunds to be issued.

Additional Stimulus Payments Could Be Issued on False Returns During the 2009 Filing Season

Congress has mandated that no advance stimulus payment can be issued after December 31, 2008.  However, taxpayers may be eligible to receive the payment when they file their TY 2008 tax return in 2009 if they have not received their stimulus payment by December 31, 2008.  If the tax return filed by the taxpayer is determined to be potentially false by the IRS and also has a potential stimulus payment that has not already been paid to the taxpayer, the return will be transferred to the IRS Examination function for resolution.  We believe the potential increase in workload for the Examination function could be unmanageable and increases the risk that an estimated 229,000 false returns claiming $138.8 million in stimulus payments could be erroneously issued during PY 2009.

Current plans call for the IRS to automatically calculate a stimulus payment on the 2008 Individual Income Tax Return if the taxpayer is eligible and does not claim a stimulus credit, or if the taxpayer requests that the IRS compute the credit.  The IRS will determine eligibility, calculate the proper credit, subtract any credit issued in 2008, and include the balance as a refundable credit with the refund.[20]  Under current procedures, all questionable returns with a refundable credit must be sent to the Examination function so an assessment can be made whether to increase the tax.  Since Congress has declared that the recovery rebate should be treated as a refundable credit, virtually every false return filed during 2009 could claim or “be entitled” to the recovery rebate credit and must be referred to the Examination function for resolution.

We identified stimulus payments totaling $94.3 million that were frozen or had not posted to the individual’s account as of June 3, 2008.  We estimate that more than 229,000 stimulus payments totaling $138.8 million could be paid if or when these returns are filed in 2009.  This could place an extreme burden on the limited resources and other noncompliance priorities of the Wage and Investment Division during PY 2009.  As of September 29, 2008, the CI Division referred almost 140,000 returns, with 81,000 returns referred to the Accounts Management organization and 59,000 returns referred to the Examination function.  Processing false tax returns with recovery rebate credits could almost quadruple the workload of the Examination function during PY 2009 because the false returns would have to be sent to the Examination function instead of the Accounts Management organization.

We believe this increased workload, coupled with limited resources and other competing priorities, could cause the IRS to increase the dollar threshold of the potentially false returns referred to Examination function personnel.  Prior history has shown that the IRS has raised tolerances to address workload-related concerns.[21]  We previously recommended that the IRS initiate a legislative proposal to address the statutory notice of deficiency issue as it relates to false refunds; however, little progress has been made.  Unless and until the IRS takes action on this issue, the CI Division and the IRS may continue to find it necessary to raise tolerances for workload management.  This could have the unfortunate consequence of allowing millions of dollars of false refunds and false stimulus payments to be issued in 2009 and result in the continued growth of potentially false refund returns and refunds that will go undetected.

Unfortunately, we believe it is too late for any legislative change to address the processing of false recovery rebate credits during PY 2009.  However, as mentioned, the IRS plans to calculate the rebate recovery credit on returns in which the taxpayer does not claim it.  For the returns in which the taxpayer did not initially claim the credit, we do not know if the IRS can subsequently deny this credit without having to issue a statutory notice of deficiency.  If this is possible, the CI Division could refer these returns directly to the Accounts Management organization for resolution (rather than the Examination function).  This would eliminate the requirement for the issuance of a statutory notice of deficiency, as the IRS could process the return as containing a math error.

Recommendation

Recommendation 3:  The Deputy Commissioner for Services and Enforcement should seek a Counsel opinion to determine whether a rebate recovery credit calculated by the IRS requires the issuance of a statutory notice of deficiency.

Management’s Response:  The IRS agreed with this recommendation.  IRS management stated Counsel has provided the IRS with an informal opinion; however, the CI Division will request a formal written opinion from Chief Counsel.

The Criminal Investigation Division Took Action to Identify Potential Refund Schemes on Stimulus-Only Returns

As previously discussed, taxpayers who are not required to file a tax return or do not owe tax are eligible to receive the stimulus payment if they have at least $3,000 in qualifying income.  This could include income from Social Security or Veteran’s Disability benefits.  These taxpayers were asked to file a return and write “Stimulus Return” on the tax return.

The CI Division prepared an Action Plan that called for them to explore alternatives to identify the release of potentially false payments associated with stimulus returns filed for the sole purpose of obtaining the payment.  As part of this plan, the CI Division developed a series of queries that were run against the Electronic Fraud Detection System.  As a result, the CI Division identified eight schemes; ****(1)**** and four subject criminal investigations were initiated.

We conducted our own analysis to identify potential schemes involving stimulus-only returns.  Specifically, we obtained an extract from the Master File as of April 9, 2008, and determined that 936,374 stimulus-only returns were filed.  Subsequent analysis determined that 135,045 taxpayers requested their stimulus payments to be deposited directly into a bank account.  We conducted further analysis of these accounts and identified a high-risk trend involving excessive numbers of payments ****(1)****.  We found 14 bank accounts that would each receive more than 20 stimulus payments, including ****(1)****.  We referred this information to Office of Refund Crimes management and they advised ****(1)****.  The other accounts appeared to be legitimate returns or did not have enough payments to warrant additional investigations.

 

Appendix I

 

Detailed Objective, Scope, and Methodology

 

The overall objective was to evaluate the effectiveness of the CI Division’s actions to prevent the issuance of economic stimulus payments for tax returns identified by the IRS as claiming false refunds.  Through assessment of the electronic data sources used in this audit, we concluded some of the data were of undetermined reliability.  However, answering the audit’s objective would not be possible if the data were not used, and it was our opinion that using the data would not weaken the analysis or lead to an incorrect or unintentional message.  Additional steps to determine data reliability prior to testing were not feasible.  For samples selected, the electronic data were validated to the Master File.  To accomplish our objective, we:

I.                        Evaluated the actions taken to identify and stop economic stimulus payments on tax returns determined false after the IRS issued the refund, but before the stimulus payment was scheduled for issuance.

A.     Interviewed CI Division management to determine any actions planned or taken to stop stimulus payments associated with tax returns determined false after the IRS issued the refund.

B.     Obtained the CI Division’s procedures for processing false returns during PY 2008, including procedures for stopping stimulus payments from being issued on false returns.

C.     Determined whether the CI Division conducted any analysis to identify the potential volume and amount of stimulus payments that might be issued based on false refund returns.

D.     Obtained statistics that show the number and refund amount of false refunds identified and stopped as of June 2, 2008, and September 29, 2008.

E.      Obtained the following data extracts to conduct various audit tests:

1.      Obtained an extract from the STARS as of June 3, 2008, that contained 229,238 false returns.  We identified 194,191 Tax Year 2007 returns with a valid Social Security Number.

2.      Obtained an extract from the Electronic Fraud Detection System as of June 7, 2008, and identified 248,565 returns determined false that should have been entered into the STARS.

3.      Obtained a Master File extract that contained about 104.6 million stimulus payments that the IRS authorized for issuance as of July 17, 2008.

4.      Compared a Master File extract from Step I.E.3. of stimulus payments the IRS determined the taxpayer would be eligible to receive to the extract of the 194,191 false returns in Step I.E.1. and identified 172,867 false returns with approximately $104.6 million in stimulus payments.

5.      Obtained a Master File extract that identified various account transactions, freeze conditions, and other data for the 172,867 false returns identified in Step I.E.4.

F.      Compared the Electronic Fraud Detection System file obtained in Step I.E.2. to the STARS data obtained in Step I.E.1. and identified 9,901 returns with a 2007 Tax Year in the Electronic Fraud Detection System that were not in the STARS.  We further determined that 1,482 returns were verified as false before May 1, 2008.

1.      Provided a list of the 1,482 returns to the CI Division to determine the current status of the returns and why they were not input to the STARS.

2.      Compared the 1,482 returns in Step I.F.1. to the file of 172,867 stimulus payments issued from Step I.E.4. to determine the number and amount of stimulus payments paid.

G.     Conducted analyses to determine the amount of stimulus payments that would have been paid on false returns because the IRS did not originally plan to freeze these accounts.

1.      Identified 24,980 false returns in which the CI Division indicated the refunds were issued from the STARS extract as of June 3, 2008, that contained 229,238 false returns.

2.      Analyzed Master File data for the 24,980 false returns where the refunds were issued and identified those returns where the stimulus payments were stopped due to the placement of criminal freezes.

II.                 Evaluated the effectiveness of plans to identify and stop unissued stimulus payments for tax returns in which the IRS stopped the false refund and referred the account to either the Accounts Management organization or Examination function.

A.     Interviewed CI Division management to determine whether any actions were planned or taken to stop stimulus payments associated with tax returns determined false with refunds stopped by the IRS.

B.     Obtained CI Division statistics that show the number and refund amount of false refunds identified and sent to the Accounts Management organization or Examination function for action as of June 2 and September 29, 2008.

C.     Obtained the latest version of the Master File programming documentation and determined the criteria used for stopping the stimulus payments on those accounts in which the IRS stopped the false refund and adjusted the account.

D.     Obtained an extract from the STARS using IRS criteria to identify false returns identified by the CI Division and sent to the Accounts Management organization or Examination function.

1.      Identified those accounts with adjustment action pending at the time of the extract.  We selected a judgmental sample of 35 cases out of a population of 17,809 that the CI Division referred for adjustment action and determined whether the IRS properly stopped the stimulus payment.  We used a judgmental sample because we did not intend to project our results to the universe.

2.      Determined the amount of stimulus payments stopped using this procedure.

III.               Determined whether the CI Division had plans to identify schemes on stimulus-only tax returns.

                       A.      Interviewed CI Division management to determine whether any actions were planned or taken to identify false stimulus payments associated with stimulus-only tax returns.

                       B.      Obtained information on the results of the CI Division’s identification of potential refund schemes involving stimulus-only returns.

                       C.      Obtained a Master File extract of 936,374 stimulus-only returns filed as of April 9, 2008, and conducted queries to identity common characteristics.

1.      Referred to the CI Division a list of 14 bank accounts in which more than 20 stimulus payments were scheduled to be deposited in the same account.

2.      Followed up with the CI Division to determine the status of the 14 accounts that we identified in Step III.C.1.

 

Appendix II

 

Major Contributors to This Report

 

Nancy A. Nakamura, Assistant Inspector General for Audit (Management Services and Exempt Organizations)

Jeffrey M. Jones, Director

Diana Tengesdal, Audit Manager

Michael J. Hillenbrand, Lead Auditor

Jeff K. Jones, Senior Auditor

Joe Smith, Senior Auditor

Robert J. Carpenter, Information Technology Specialist

 

Appendix III

 

Report Distribution List

 

Commissioner C

Office of the Commissioner – Attn: Chief of Staff C

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

Commissioner, Wage and Investment Division  SE:W

Chief, Criminal Investigation Division  SE:CI

Director, Campus Compliance Services, Small Business/Self-Employed Division  SE:S:CCS

Director, Compliance, Wage and Investment Division  SE:W:CP

Director, Operations Policy and Support, Criminal Investigation Division  SE:CI:OPS

Director, Refund Crimes, Criminal Investigation Division  SE:CI:RC

Chief Counsel  CC

National Taxpayer Advocate  TA

Director, Office of Legislative Affairs  CL:LA

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

Office of Internal Control  OS:CFO:CPIC:IC

Audit Liaisons:

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

            Commissioner, Wage and Investment Division  SE:W

Chief, Criminal Investigation Division  SE:CI:S:PS

 

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 Congress.

Type and Value of Outcome Measure:

·        Revenue Protection – Actual; $7.1 million in economic stimulus payments that were prevented from being issued on false returns as a result of the IRS making computer program changes (see page 4).

Methodology Used to Measure the Reported Benefit:

We obtained data from the Master File on 172,867 tax returns identified as false by the CI Division, as of June 3, 2008, and for which a stimulus payment was generated.  We analyzed this data to determine the amount of stimulus payments that were prevented from being issued as a result of additional programming procedures instituted by the CI Division based on our work.  Based on our earlier inquiries, the CI Division implemented procedures to address the following two situations that may have left the accounts vulnerable to erroneous issuance of stimulus payments:

·        A refund claim was not determined to be false until after that refund was issued.  In a majority of these types of cases, the CI Division would no longer place a freeze on the account.  Without a freeze, the stimulus payment would be issued.

·        A refund claim is verified as false and the tax return is referred to either the Accounts Management organization or the Examination function for contact and further verification with the taxpayer.  If these civil functions complete their verification and adjust the account before the stimulus payment is generated then the civil freeze will be released.  This allows the stimulus payment to be issued.

As a result of our analysis, we estimate that 6,600 stimulus payments totaling $4.1 million were prevented from issuance due to the addition of the criminal freeze after the false refund was issued.  In addition, we estimate that 5,440 stimulus payments totaling $3 million were prevented from issuance due to the IRS’ subsequent programming.

Type and Value of Outcome Measure:

·        Revenue Protection – Potential; $118.8 million in false refunds and economic stimulus payments issued because the temporary freeze expired (see page 5).

Methodology Used to Measure the Reported Benefit:

We obtained a Master File extract of 172,867 false Tax Year 2007 returns identified by the CI Division as of June 3, 2008, that contained a stimulus payment.  As of October 2008, we identified 3,286 accounts in which the temporary freeze expired as programmed resulting in the issuance of $16.3 million in false refunds and 2,071 stimulus payments totaling $1.2 million.[22]  Although the amount of erroneously issued false refunds and stimulus payments we identified is relatively small in relation to the total false refunds stopped ($1.5 billion) and legitimate stimulus payments issued ($96.3 billion), we remain concerned that the procedures pose a risk that false refunds will be issued in the future.

The CI Division indicated that it continues to make improvements to this process.  However, this is the third year we have reported this issue and the CI Division’s corrective actions have not been effective.  If the CI Division does not take effective action to implement changes to the way it processes and freezes false refunds, we estimate that over $117.6 million[23] in false refunds could be erroneously released over 5 years due to the expiration of the temporary freeze.

We calculated this amount as follows:  Our analysis of 172,867 false Tax Year 2007 returns identified 3,286 returns where the refund was issued when the freeze expired.  This was about 1.4 percent of all false returns in STARS as of June 3, 2008.  As of September 29, 2008, the CI Division reported that it had identified 337,809 false returns.  We applied the 1.4 percent to the total of 337,809 false returns and arrived at a potential of 4,729 returns in which the freeze may have expired during PY 2008.  We multiplied the 4,729 returns by the average refund ($4,974) erroneously released ($23,522,046).  We then multiplied that result by 5 years ($117,610,230).  We did not include a 5-year projection for the amount of stimulus payments issued, as this was a one-time event.  The outcome measure of $118.8 million includes $1.2 million in stimulus payments actually released and a 5-year estimate of the false refunds that could be released.

 

Appendix V

 

Synopsis of Account Freezes

 

Processing Year

Temporary Freeze

Permanent Criminal Freeze

Civil Freeze

2005 and Earlier

Used if additional time was needed to determine whether the refund was false.  Freeze generally had to be released manually.

Used when a refund was determined false.  Generally current and future years’ accounts were frozen.

Generally only used if the
CI Division referred certain returns during a limited verification process.

2006

Used if additional time was needed to verify a highly suspicious refund return.  This freeze had to be resolved within a certain time period; otherwise, the false refund would be automatically released. 

Used when a refund was determined false.  However, only the account with the false refund was frozen (not future years).  Freeze was generally released when the civil freeze was placed on the account.

Placed on an account when the return is referred to either Accounts Management or Examination for resolution.

2007

Used if additional time was needed to verify a highly suspicious refund return.  This freeze had to be resolved within a certain time period; otherwise, the false refund would be automatically released. 

Used when a refund was determined false.  However, only the account with the false refund was frozen (not future years).  Freeze was generally released when the civil freeze was placed on the account.

Placed on an account when the return is referred to either the Accounts Management organization or Examination function for resolution.

2008

The CI Division relied on the temporary freeze to prevent issuance of false refunds while returns were being referred to the Accounts Management organization or Examination function for resolution.  This freeze was programmed to release within a certain time period.

 

The use of this freeze was limited to special circumstances.

Placed on the account before the temporary freeze expires; otherwise, the false refund will be issued.

Source:  Treasury Inspector General for Tax Administration analysis of Office of Refund Crimes’ referral transfer procedures.

 

Appendix VI

 

Glossary of Terms

 

Accounts Management Organization – The organization within the Wage and Investment Division responsible for answering taxpayer tax law/account inquiries and adjusting tax accounts.  In addition, it is responsible for providing taxpayers with information on the status of their returns/refunds and for resolving the majority of issues and questions to settle their accounts.

Dependent Database – A system that identifies and selects for examination taxpayer returns with possible erroneous Earned Income Tax Credit claims.  During initial processing, the Dependent Database Scoring Program analyzes tax returns that have claimed at least one Earned Income Tax Credit qualifying child or dependent child.  Using data from several sources, it analyzes each tax return for criteria that indicate the taxpayer might not be eligible for the Earned Income Tax Credit and assigns a numeric value to each criterion.  The Dependent Database then produces an overall score for the return.  Based on available resources to conduct examinations, the IRS selects certain types and quantities of returns for pre-refund examinations to verify the taxpayers’ eligibility for the Earned Income Tax Credit.

Electronic Fraud Detection System – A computer system, developed in 1994 and implemented in 1996, that automates the identification output for potentially false electronically filed tax returns, increases data available for analysis, and assists in the development of information relating to paper and electronically filed schemes detected by the CI Division.

Filing Season – The period between January and mid-April when most individual income tax returns are filed.

Fraud Detection CenterThe function responsible for identifying and detecting refund fraud, preventing the issuance of false refunds, and providing support for the CI Division field offices.

Freeze – Under certain conditions, the IRS computer system will “freeze” the account.  Depending on the reason, different freezes may prevent certain actions, such as processing a refund, from taking place.  In this report, we are referring to three general types of freezes:  a temporary freeze, a permanent criminal freeze, and a civil freeze.  See Appendix V for a description of these freezes and how they are used on accounts with false refunds.

Master File – The IRS database that stores various types of taxpayer account information.  The database includes individual, business, and employee plans and exempt organizations data.

Processing Year – The year in which taxpayers file their returns with the IRS.  For example, most Tax Year 2006 returns were filed in Processing Year 2007.

Questionable Refund Program – A nationwide program established to detect and stop false claims for refunds on income tax returns.

Scheme – Can include only one return but generally includes numerous returns.  In addition, many small false refunds that do not have common characteristics may be placed in a “dump” scheme.

Scheme Tracking and Referral System – The system of records maintained at each Fraud Detection Center for Questionable Refund Program and Return Preparer Program schemes.  It was designed to store information, for multiple processing years, that is used for tracking and historical purposes.

 

Appendix VII

 

Management’s Response to the Draft Report

 

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


[1] Unless otherwise noted, all dates in this report are calendar year.

[2] See Appendix VI for a glossary of terms. 

[3] Actions Have Been Taken to Address Deficiencies in the Questionable Refund Program; However, Many Concerns Remain, With Millions of Dollars at Risk (Reference Number 2007-10-076, dated May 31, 2007). 

[4] An Estimated $1.6 Billion in Fraudulent Refunds Was Issued During the 2006 and 2007 Filing Seasons (Reference Number 2008-10-172, dated September 22, 2008).

[5] This estimate is based on current dollar values and does not account for changes in volumes of false refund returns, average refund amounts, improvements in the IRS’ processing of false refunds, or future economic conditions. 

[6] An Estimated $1.6 Billion in Fraudulent Refunds Was Issued During the 2006 and 2007 Filing Seasons (Reference Number 2008-10-172, dated September 22, 2008).

[7] Pub. L. 110-185, 122 Stat. 613.

[8] Unless otherwise noted, all dates in this report are calendar year.

[9] See Appendix VI for a glossary of terms.

[10] Evaluation of Planning Efforts for the Issuance of Economic Stimulus Payments (Reference Number 2008-40-149, dated July 31, 2008).

[11] Actions Have Been Taken to Address Deficiencies in the Questionable Refund Program; However, Many Concerns Remain, With Millions of Dollars at Risk (Reference Number 2007-10-076, dated May 31, 2007).

[12] An Estimated $1.6 Billion in Fraudulent Refunds Was Issued During the 2006 and 2007 Filing Season (Reference Number 2008-10-172, dated September 22, 2008).

[13] This estimate is based on current dollar values and does not account for changes in volumes of false refund returns, average refund amounts, improvements in the IRS’ processing of false refunds, or future economic conditions.

[14] We based our estimate on the data we received from the Office of Refund Crimes dated June 3, 2008, which contained 172,867 false refund returns.  Since this data file does not represent returns identified for the whole processing year, the estimated revenue protected amount could be higher.

[15] See Appendix V for a synopsis of account freezes.

[16] Actions Have Been Taken to Address Deficiencies in the Questionable Refund Program; However, Many Concerns Remain, With Millions of Dollars at Risk (Reference Number 2007-10-076, dated May 31, 2007).

[17] An Estimated $1.6 Billion in Fraudulent Refunds Was Issued During the 2006 and 2007 Filing Seasons (Reference Number 2008-10-172, dated September 22, 2008).

[18] This estimate is based on current dollar values and does not account for changes in volumes of false refund returns, average refund amounts, and future economic conditions.

[19] An Estimated $1.6 Billion in Fraudulent Refunds Was Issued During the 2006 and 2007 Filing Seasons (Reference Number 2008-10-172, dated September 22, 2008).

[20] A credit is refundable if the taxpayer will receive a refund of the credit even if they have no tax liability.

[21] An Estimated $1.6 Billion in Fraudulent Refunds Was Issued During the 2006 and 2007 Filing Seasons (Reference Number 2008-10-172, dated September 22, 2008).

[22] This is an estimate of the amount of refunds released by the IRS’ computer system.  Our limited testing showed that some of these refunds may be recovered because they were undeliverable or the bank was suspicious and notified the IRS who then cancelled the refund.  We are unable to provide an estimate of this amount.

[23] This estimate is based on current dollar values and does not account for changes in volumes of false refund returns, average refund amounts, improvements in the IRS’ processing of false refunds, or future economic conditions.