TREASURY INSPECTOR GENERAL FOR TAX ADMINISTRATION

 

 

Automated Underreporter Program Tax Assessments Have Increased
Significantly; However, Accuracy-Related Penalties Were Not
Always Assessed When Warranted

 

 

 

May 8, 2015

 

Reference Number:  2015-30-037

 

 

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

E-mail Address /  TIGTACommunications@tigta.treas.gov

Website           /  http://www.treasury.gov/tigta

 

 

HIGHLIGHTS

AUTOMATED UNDERREPORTER TAX ASSESSMENTS HAVE INCREASED
SIGNIFICANTLY; HOWEVER, ACCURACY-RELATED PENALTIES WERE NOT
ALWAYS ASSESSED WHEN WARRANTED

Highlights

Final Report issued on May 8, 2015

Highlights of Reference Number:  2015-30-037 to the Internal Revenue Service Commissioner for the Small Business/Self-Employed Division.

IMPACT ON TAXPAYERS

The Automated Underreporter Program matches taxpayer income and deductions submitted on information returns by third parties (e.g., employers, banks, brokerage firms) against amounts reported by taxpayers on their individual income tax returns to identify discrepancies.  The Automated Underreporter Program routinely identifies more than 20 million individual tax returns with discrepancies each year and, when warranted, assesses additional taxes as well as interest and penalties.

WHY TIGTA DID THE AUDIT

The overall objectives of this audit were to determine the type and extent of noncompliance addressed by the Automated Underreporter Program and whether additional steps could be taken to enhance program accomplishments.  This review is part of our Fiscal Year 2015 Annual Audit Plan and addresses the major management challenge of Tax Compliance Initiatives.

WHAT TIGTA FOUND

TIGTA found that Automated Underreporter Program tax assessments increased significantly in recent years, from $4.24 billion in Fiscal Year 2006 to $7.84 billion in Fiscal Year 2013, an increase of 85 percent.  During Fiscal Year 2013, the Automated Underreporter Program also assessed approximately $708 million in accuracy-related penalties; however, TIGTA found that such penalties were not always assessed when warranted.  For instance, the Automated Underreporter Program’s system does not apply the negligence penalty provided for by law unless the taxpayer has repeated the same type of income omission within four consecutive tax years.  Additionally, TIGTA’s review of Fiscal Year 2012 closed cases found that examiners were incorrectly waiving accuracy‑related penalties, which resulted in about $3.25 million in lost penalty revenue.  TIGTA also found that, due to an inaccurate programming condition, approximately $2.66 million in accuracy-related penalties were not assessed.

Also, after an in-depth study, a number of revisions were made to the taxpayer notice, which alerts taxpayers that additional taxes may be owed as a result of underreporting.  Although the revised taxpayer notice was implemented in Fiscal Year 2013, the IRS has not evaluated, or established plans to evaluate, the effectiveness of the revised notice on reducing taxpayer underreporting.

WHAT TIGTA RECOMMENDED

TIGTA recommended that the IRS: 1) implement controls to ensure that accuracy-related penalties are assessed when warranted and only waived in accordance with applicable policies and procedures, 2) address system issues to ensure that accuracy-related penalties are accurately assessed when warranted, 3) continue to research and take action on cases TIGTA identified as having potentially inaccurate accuracy-related penalty amounts, 4) address negligence as it occurs rather than when a taxpayer repeats noncompliance, and 5) evaluate the effectiveness of the revised taxpayer notice.

In response to the report, IRS management agreed with four of our five recommendations.  They disagreed that they should address negligence when it occurs and stated that they will continue within their current policy of waiting until taxpayers repeat noncompliance.  TIGTA contends this policy may contribute to further noncompliance and that addressing negligence when it occurs would promote fair and consistent treatment of all taxpayers.

May 8, 2015

 

 

MEMORANDUM FOR COMMISSIONER, SMALL BUSINESS/SELF-EMPLOYED DIVISION  

 

FROM:                 for Michael E. McKenney /s/ Matthew Weir

Deputy Inspector General for Audit

 

SUBJECT:                  Final Audit Report – Automated Underreporter Program Tax Assessments Have Increased Significantly; However, Accuracy‑Related Penalties Were Not Always Assessed When Warranted (Audit # 201230026)

 

This report presents the results of our review to determine the type and extent of noncompliance addressed by the Automated Underreporter Program and if additional steps could be taken to enhance program accomplishments.  This audit is included in the Treasury Inspector General for Tax Administration’s Fiscal Year 2015 Annual Audit Plan and addresses the major management challenge of Tax Compliance Initiatives.  The audit was conducted at the suggestion of the Internal Revenue Service Oversight Board.

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

Copies of this report are also being sent to the Internal Revenue Service managers affected by the report recommendations.  If you have any questions, please contact me or Matthew A. Weir, Assistant Inspector General for Audit (Compliance and Enforcement Operations).

 

 

Table of Contents

 

Background

Results of Review

Improvements to the Automated Underreporter Program Have Contributed to Significant Increases in Tax Assessments

Additional Improvements Are Needed to Ensure That Accuracy-Related Penalties Are Assessed When Warranted

Recommendations 1 through 3:

Recommendation 4:

Revisions to the Computer Paragraph 2000 Notice May Not Have Produced the Anticipated Results

Recommendation 5:

Appendices

Appendix I – Detailed Objectives, Scope, and Methodology

Appendix II – Major Contributors to This Report

Appendix III – Report Distribution List

Appendix IV – Outcome Measures

Appendix V – Example of a Reasonable Cause Lead Sheet

Appendix VI – Glossary of Terms

Appendix VII – Management’s Response to the Draft Report

 

Abbreviations

 

AUR

Automated Underreporter

CP

FTE

Computer Paragraph

Full-Time Equivalent

FY

Fiscal Year

GAO

Government Accountability Office

IRM

Internal Revenue Manual

IRS

Internal Revenue Service

SB/SE

Small Business/Self-Employed

TIGTA

Treasury Inspector General for Tax Administration

W&I

Wage and Investment

 

Background

 

The Internal Revenue Service’s (IRS) Automated Underreporter (AUR) Program[1] matches taxpayer income and deductions submitted on information returns by third parties (e.g., employers, banks, brokerage firms) against amounts reported by taxpayers on their individual income tax returns to identify discrepancies.  For Tax Year 2013,[2] the IRS received approximately 2.1 billion information returns.  As a result of matching third‑party information returns to individual tax returns,[3] the AUR Program has routinely identified more than 20 million individual tax returns with discrepancies since Tax Year 2009.

Due to resource constraints, the AUR Program cannot review every discrepancy it identifies.  As a result, additional data analyses are performed on the population of individual tax returns with discrepancies to select the inventory that will be reviewed by AUR Program examiners.  For the past several tax years, the AUR Program selected approximately five million returns to review each tax year, which represents more than 20 percent of the discrepancies identified each year.  The most frequently identifed discrepancies among the selected inventory involved underreported securities sales, wages, taxable pensions, unemployment compensation, and nonemployee compensation.

Once a return is selected for inventory, AUR Program examiners manually review the associated return along with other information received by the IRS to determine whether the discrepancy can be resolved internally (i.e., without contacting the taxpayer).  When the discrepancy cannot be resolved internally, AUR Program examiners will contact the taxpayers by sending them a notice (Computer Paragraph (CP) 2000 Notice, Request for Verification of Unreported Income, Payments, or Credits)[4] which alerts the taxpayer that a discrepancy was identified and that additional taxes may be owed as a result of the underreported income.  In response to receiving a CP 2000 Notice, the taxpayer may provide supporting documentation that fully resolves the discrepancy identified by the AUR Program.  In this situation, the AUR Program examiner would close the AUR case with no changes to the taxpayer’s account.  However, if the taxpayer is unable to provide any support for the discrepancy, the IRS will, after allowing the taxpayer an opportunity to internally appeal the proposed deficiency, issue a Notice of Deficiency proposing to assess additional taxes and penalties allowing the taxpayer to protest the matter in the U.S. Tax Court.[5] 

This review was performed at the AUR Program campus function in Philadelphia, Pennsylvania, and with information obtained from the Wage and Investment (W&I) Division AUR Program Office in Atlanta, Georgia, and the Small Business/Self-Employed (SB/SE) Division Compliance Office in New Carrollton, Maryland, during the period February 2013 through July 2014.  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 objectives.  We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives.  Detailed information on our audit objectives, scope, and methodology is presented in Appendix I.  Major contributors to the report are listed in Appendix II.

 

Results of Review

 

The AUR Program’s success can be attributed, in part, to recent enhancements to its inventory selection process, which has resulted in the selection of more productive returns for review.  Specifically, the percentage of AUR Program inventory that involved taxpayers being notified of a discrepancy and that additional taxes may be owed increased from 75 percent in FY 2006 to 86 percent in FY 2013.  As a result, the AUR Program’s additional tax assessments increased from $4.24 billion to $7.84 billion during the same time period.  The AUR Program also assessed approximately $708 million[6] in accuracy-related penalties during FY 2013.  However, our review found that accuracy-related penalties were not always assessed when warranted.  As a result, the IRS may be missing opportunities to further promote voluntary compliance and increase revenue.  In addition, the AUR Program needs to ensure that the revised CP 2000 Notices are meeting the needs of the taxpayers as well as the IRS.

Improvements to the Automated Underreporter Program Have Contributed to Significant Increases in Tax Assessments

Although resource limitations have resulted in the IRS reviewing fewer tax returns in FY 2013, AUR Program improvements have resulted in increased tax assessments.  Despite fewer resources, the AUR Program reviewed approximately 4.88 million (71 percent) of the 6.88 million individual tax returns reviewed by the IRS during FY 2013.  The remaining two million individual tax returns were reviewed by other IRS enforcement programs, such as the Correspondence and Field examination functions.  The AUR Program assessed approximately $7.84 billion (31 percent) of the $25.1 billion in total assessments made by the IRS during FY 2013.  The $7.84 billion in FY 2013 assessments is an increase of 85 percent from the $4.24 billion in assessments during FY 2006.  As shown in Figure 1, AUR case closures increased from 4.4 million in FY 2006 to 4.8 million in FY 2013.  In addition, the average assessment per AUR closed case increased from $960 per closure in FY 2006 to $1,605 in FY 2013. 

Figure 1:  Automated Underreporter Program Case Closures
During Fiscal Years 2006
Through 2013

Fiscal Year

Number of Closed Cases

Additional Tax
Assessment Amounts
[7]

Average Additional Tax Assessment Amount
Per Closed Case

2006

4,425,855

$4,247,050,494

$960

2007

4,531,748

$5,218,846,869

$1,152

2008

4,458,564

$6,540,214,177

$1,467

2009

4,536,778

$6,409,804,156

$1,413

2010

5,243,119

$7,342,352,375

$1,400

2011

5,623,423

$6,544,744,527

$1,164

2012

5,623,284

$7,241,642,521

$1,288

2013

4,887,957

$7,845,162,021

$1,605

Source:  IRS Information Return Program Case Analysis System, Table U28, March 2014. 
We did not perform any audit tests to assess the validity of this data.

The improvements in overall productivity and performance were accomplished despite declining budgets and the corresponding reductions in staffing.  Specifically, the IRS budget declined every year for three straight fiscal years, from $12.14 billion in FY 2010 to $11.2 billion in FY 2013, a decrease of approximately $1 billion (almost 8 percent).  During this time, as shown in Figure 2, the AUR Program experienced an approximate 11 percent reduction in full-time equivalents (FTE), from 2,253 in FY 2010 to 2,014 in FY 2013.  While this may have contributed to a reduction in the number of case closures in recent years, improvements to the AUR Program’s inventory selection process resulted in an overall increase in additional tax assessments.  Specifically, although the number of closed cases decreased by approximately 7 percent, from 5.24 million in FY 2010 to 4.88 million in FY 2013, additional tax assessments increased by about 7 percent, from $7.34 billion to $7.84 billion.

Figure 2:  Full-Time Equivalents for Fiscal Year 2006 Through Fiscal Year 2013

Fiscal Year

Additional Tax Assessment Amounts

Number of FTEs

Additional Tax Assessment Per FTE

Number of
Closed Cases

Closures
Per FTE

2006

$4,247,050,494

1,752.8

$2,423,009

4,425,855

2,525

2007

$5,218,846,869

1,742.1

$2,995,722

4,531,748

2,601

2008

$6,540,214,177

1,781.0

$3,672,215

4,458,564

2,503

2009

$6,409,804,156

1,828.0

$3,506,457

4,536,778

2,482

2010

$7,342,352,375

2,253.4

$3,258,344

5,243,119

2,327

2011

$6,544,744,527

2,343.9

$2,792,246

5,623,423

2,399

2012

$7,241,642,521

2,197.0

$3,296,150

5,623,284

2,560

2013

$7,845,162,021

2,013.6

$3,896,088

4,887,957

2,427

Source:  IRS Information Return Program Case Analysis System, Table U28, March 2014. 
We did not perform any audit tests to assess the validity of this data.

The AUR Program inventory selection process was enhanced to identify discrepancies on late-filed individual tax returns

Historically, the AUR Program would match information returns to individual tax returns twice a year to select its inventory.  The first match identifies discrepancies among individual tax returns filed on time (by April 15).  The second match identifies discrepancies among individual tax returns filed after April 15 but on or before December 31.  This generally includes taxpayers who received extensions to file or who filed their individual tax returns late.[8]  Starting in FY 2011, the AUR Program began testing a third match to identify additional discrepancies for tax returns filed significantly past the due date.  Specifically, this third match identifies tax returns filed between January and March of the calendar year subsequent to that in which the tax return was due. 

Prior to the implementation of this third match, tax returns filed between January and March (i.e., between nine months to a year after the filing due date) would not have been identified.  For instance, the implementation of the third match identified 127,330 Tax Year 2013 returns with discrepancies that otherwise would not have been identified during the first or second matches.  Of the 127,330 returns identified, the AUR Program selected 30,667 (24 percent) to review.  The percentage of cases selected for review during the third match is in line with the overall selection rate for all AUR Program inventory.  This is due to the fact that the process for all three matches includes the same filters to identify discrepancies and the inventory selected for review is based on available resources. 

Internal and external studies have resulted in more productive case inventory  

As noted earlier, due to resource constraints, the AUR Program only reviews a fraction of the total number of tax returns it identifies as having discrepancies.  The AUR Program has recently conducted studies to improve the inventory selection process to better ensure that more productive tax returns are selected for review by AUR Program examiners.  These studies identified new business rules and modifications of existing rules that were employed to refine the inventory selection process.  As a result, the AUR Program has significantly reduced its screen‑out rate (as shown in Figure 3), which represents the percentage of AUR inventory for which the discrepancies are resolved without taxpayer contact.  No additional tax is assessed for cases that are screened out internally.  A lower screen-out rate indicates that the AUR Program is potentially selecting more productive returns that require taxpayer contact for resolution (rather than being resolvable internally).  The screen-out rate decreased significantly from about 34 percent in FY 2004 to 14 percent in FY 2013.  The reduction in the screen-out rate resulted in the AUR Program reaching more noncompliant taxpayers.  Specifically, the percentage of AUR Program inventory that results in the taxpayer being alerted of a discrepancy and that additional tax may be due increased by 20 percent, from 66 percent in FY 2004 to 86 percent in FY 2013.

Figure 3:  Screen-Out Rate for Selected Inventory

Figure 1 was removed due to its size.  To see Figure 1, please go to the Adobe PDF version of the report on the TIGTA Public Web Page.

Source:  IRS Information Returns Processing Case Analysis System, Table U28, March 2013. 
We did not perform any audit tests to assess the validity of this data.

Additional Improvements Are Needed to Ensure That Accuracy-Related Penalties Are Assessed When Warranted

I.R.C. § 6662 authorizes the IRS to assess an accuracy-related penalty when there is an underpayment of tax due to taxpayer negligence (or disregard of the rules or regulations)[9] or a substantial understatement of income tax liabilities.  The I.R.C.[10] defines a substantial understatement of income tax as an understatement exceeding the greater of 10 percent of the tax required to be shown on the return for the taxable year or $5,000.  The accuracy-related penalty is designed to promote the preparation and submission of complete and correct information on tax returns as well as impose an economic cost on taxpayers who choose not to comply with the tax law.[11]  If the IRS imposes the accuracy-related penalty, the taxpayer can seek to have the penalty waived or abated for reasonable cause.[12] 

It is important to note that although the IRS may find that a taxpayer substantially understated his or her tax liabilities and there was negligence on the part of the taxpayer, only one accuracy‑related penalty (either for negligence or substantial understatement of tax liabilities) can be assessed on the underpayment.  The accuracy-related penalty for negligence and substantial understatement of tax liabilities is 20 percent of the resulting underpayment.[13]  Typically, an IRS examiner would determine whether an accuracy-related penalty should be assessed – either for negligence or substantially understated tax liabilities.  However, in the AUR Program, determining which underreporting cases should be assessed an accuracy-related penalty is an automated process.  

In accordance with the Government Accountability Office’s (GAO) Standards for Internal Control in the Federal Government,[14] the IRS has developed controls to ensure that penalties are assessed when warranted.  The standard for assessing the substantial understatement penalty is essentially a mathematical test, and the business rules in the AUR system will automatically include the accuracy-related penalty on the CP 2000 Notices for those cases that involve a substantial understatement of tax liabilities.  However, the AUR Program only assesses the negligence penalty after a taxpayer has multiple instances of understating income in the AUR Program.  In other words, the only taxpayers in the AUR Program to be assessed the negligence penalty are those taxpayers who have a prior history of being subject to an additional tax assessment as a result of an underreporting discrepancy in the same AUR Program discrepancy category[15] for at least twice in four consecutive tax years and did not meet the substantial understatement criteria.  These taxpayers are referred to in the program as “repeaters” because they have a prior history of the same type of income discrepancy.[16]  Under this definition of “repeaters,” a taxpayer under the AUR Program’s business rules could omit income once every fourth year and still be deemed a “first-timer” for purposes of assessing the negligence penalty.

During FY 2013, the AUR Program assessed approximately $708 million[17] in accuracy-related penalties.  However, our review of AUR Program data, as well as a statistical sample of AUR cases, found that such penalties are not always assessed when warranted.  Therefore, the IRS may be missing opportunities to further promote voluntary compliance and increase tax revenue. 

The AUR Program failed to assess accuracy-related penalties because examiners inappropriately waived them without properly substantiating reasonable cause

The AUR system is designed to identify AUR cases meeting the accuracy-related penalty criteria for substantial understatement of tax liabilities and automatically include the penalty on the CP 2000 Notice.  To verify that the system controls were working as intended, we performed an analysis on AUR cases involving Tax Year 2010 individual tax returns and found that 90 percent (46,583) of the 51,581 AUR cases involving substantial understatement of tax liabilities were assessed the applicable accuracy-related penalty.  For the remaining 4,998 AUR cases that were not assessed a substantial understatement penalty, we evaluated a statistical[18] sample of 51 cases to determine whether an accuracy-related penalty was originally included on the CP 2000 Notice and, if so, the reasons for the waiver of the penalty.

Although all 51 sample cases met the criteria for the accuracy-related penalty for substantial understatement of tax liabilities, we found that only 44 cases included the penalty on the CP 2000 Notice.  The remaining seven (14 percent) cases lacked the applicable accuracy-related penalty.  For the 44 cases that did include the penalty, we found that AUR Program examiners inappropriately waived those penalties for 21 cases.  Specifically, we found that the taxpayers associated with these cases did not provide sufficient documentation to support that their underreporting of tax liabilities was due to reasonable cause.[19]  The accuracy-related penalties for these 21 cases totaled $37,980.  When we project the sample results for these 21 cases to the universe of 9,339 closed AUR cases,[20] we estimate that taxpayers may have avoided being assessed approximately $3.25 million in accuracy-related penalties.  When forecasted over a five-year period, we estimate that the IRS may miss the opportunity to increase revenue by $16.25 million.[21]

Although the AUR system systematically includes the accuracy-related penalties on the CP 2000 Notices when certain criteria are met, the AUR Program relies on its examiners to determine if the penalty should be sustained once the taxpayer responds to the CP 2000 Notice.  Specifically, the accuracy-related penalty may be waived if the AUR Program examiner evaluates the facts and circumstances presented by the taxpayer and determines that the taxpayer’s justification for underreporting meets one of the reasonable causes as outlined in the IRM.[22]  According to the IRM,[23] generally reasons such as forgetfulness, ignorance of the law, or mistakes (e.g., the taxpayer/preparer forgot to include income or thought income was nontaxable) do not qualify for penalty relief because these actions do not demonstrate ordinary business care and prudence.  

According to the IRS, AUR Program examiners should be allowed to determine whether to waive a penalty based on their experience and judgment of the facts and circumstances of each case and apart from the specific criteria in the IRM.  While we agree that the examiner should evaluate all the facts and circumstances in each case when determining whether the taxpayer’s justification met one of the reasonable causes noted in the IRM, the development of additional guidance (e.g., lead sheet or desk guide) would help better ensure that existing IRM requirements are followed as well as ensure that penalties are waived appropriately and consistently.

IRS procedures require the accuracy-related penalty to be assessed unless the taxpayer provides sufficient documentation to support that the underreporting was due to reasonable cause.  In all 21 exception cases, we determined that the taxpayers did not provide sufficient documentation to support waiving the substantial understatement accuracy-related penalty.  We conducted a thorough review of all available AUR documentation, both hardcopy and electronic AUR case file information, for all 21 exception cases.  We compared the information the AUR Program examiners documented to justify the waiver of the penalty and found that in all 21 cases, the examiners’ justifications did not meet one of the reasonable causes as outlined in the IRM.  For instance, in five (24 percent) of the 21 cases, taxpayers did not provide any information explaining why they omitted income from their tax returns.  The examiner should not have waived accuracy-related penalties ranging from ***1*** to ***1*** because these taxpayers did not provide any information as to why income ranging from ***1*** to ***1*** was not included on their original Tax Year 2010 returns.  For three of these five cases, we did not find any evidence that the taxpayer requested a waiver of the accuracy-related penalty.  In the remaining 16 (76 percent) of the 21 cases, the taxpayer provided an explanation for the discrepancy AUR identified; however, we believe the explanation did not warrant penalty relief because the explanation did not meet one of the reasonable causes as outlined in the IRM.

It is important to note that the AUR Program analysts did not agree with the conclusions we reached in 10 (48 percent) of the 21 AUR substantial understatement penalty cases.  For these 10 cases, the taxpayers requested a waiver of the penalty and provided an explanation for the understatement of income; however, we found that that the taxpayers’ explanations did not warrant penalty relief because their explanations did not meet the reasonable cause criteria.  In the IRS’s written responses to our case reviews, the AUR Program analysts did not provide any relevant IRM criteria or other guidance with which to support their positions.  Instead, the AUR Program analysts stated that their positions were based on their 20 years of experience.

The AUR Program analysts also stated that they considered the following when reviewing our exception cases:  1) whether the income omission was a new source of income, 2) whether the taxpayer was compliant in the previous three tax years, and 3) how quickly the taxpayer responded to the CP 2000 Notice.  These additional factors were not listed in the IRM, or any other AUR Program guidance or training materials, as factors to consider when determining reasonable cause that would warrant penalty relief.   

We also reviewed the AUR Program’s training materials from FYs 2010 through 2013 and found that the materials did not provide sufficient instructions to the AUR Program examiners regarding when to waive penalties.  Although the training materials provide the AUR Program examiners with the most common reasonable cause issues, they did not provide guidance for interpreting the information provided by taxpayers in their responses to the CP 2000 Notices.  Given that the AUR Program analysts who reviewed our cases were considering a number of undocumented criteria, development and implementation of training related to these additional criteria could better ensure that examiners understand how to interpret the reasonable cause criteria when making a determination to waive a penalty.  These undocumented factors and lack of sufficient training may have contributed to the fact that AUR Program officials did not always agree when responding to our exception cases (for example, one AUR reviewer said they would have removed a penalty, while others said they would not have removed it). 

Since AUR Program examiners and managers are spread throughout the country, it is important that the AUR Program follow existing IRM criteria to ensure that the accuracy-related penalties are consistently assessed when warranted.  Without such consistency, fair and equitable treatment of taxpayers may be compromised.

Managerial reviews, use of examiner lead sheets, and improved notices to taxpayers could also help better ensure consistency with waiving accuracy‑related penalties

Although this review was focused on the AUR Program, our findings are consistent with prior TIGTA reviews[24] that looked at the IRS’s Examination functions and the consistency with which penalties were applied to taxpayers.  For example, in FY 2010, we reported that IRS field examiners were either too lenient and did not recommend penalties that were warranted or had not documented in the case files that applicable penalties were considered.  To improve how penalties are considered and assessed during audits, TIGTA made a number of recommendations that included establishing a requirement for Examination first-line managers to approve examiner decisions not to assess the substantial understatement penalty.  IRS executives agreed with the recommendation and, in July 2008, updated the IRM[25] to require managerial approval of decisions not to assess the penalty.  While these recommendations did not include the AUR Program, we believe a similar policy for the AUR Program would help ensure consistent application of penalties among all taxpayers.

AUR Program officials stated that they are not subject to the IRM requirement that outlines that the nonassertion of a penalty requires managerial approval.  Specifically, according to AUR Program officials, because the AUR Program is an automated function, whereby penalties are determined automatically by the AUR system, AUR Program examiners are excluded from needing to obtain managerial approval when waiving penalties.[26]  The lack of the managerial approval requirement may have contributed to the findings we identified during our case reviews.  For example, in 11 (52 percent) of the 21 exception cases we identified, AUR Program officials stated they agreed with our conclusions that the removal of the penalty was not appropriate.

In addition, our review found that AUR Program examiners are not documenting their justification for waiving the accuracy-related penalty.  This was evident in our case reviews as we found that examiners in 14 (67 percent) of the 21 exception cases did not document the reason for granting penalty relief in the AUR system or in the hard copy case files.  According to the IRM, “all penalty determinations involving a reasonable cause exception will be documented in the workpapers.  This will be done by identification on the AUR system and will remain with the case file.”  Although the AUR Program implemented training during FY 2015 to help train the examiners on addressing penalty waivers, we believe the development and use of some type of lead sheet[27] or desk guide could assist AUR Program examiners in making proper and consistent determinations to waive a penalty.  The use of a lead sheet may also provide the examiners one central source from which to obtain guidance in making a decision on waiving penalties.  Currently, the examiners are advised to review not only their own IRM for guidance but also two additional IRM sections on penalties.  With limited resources and time frames in which to work cases, a check sheet could provide a much needed resource from which to seek guidance quickly and accurately.

We believe that the results of our review indicate that the use of some type of a lead sheet could help AUR Program examiners determine whether a penalty should be waived as well as help AUR Program first-line managers evaluate whether their examiners are properly waiving accuracy-related penalties.  For example, we evaluated the results of the quality reviews of closed AUR cases performed by AUR first-line managers and found that the reviews performed during FYs 2010 through 2013 did not include a step to evaluate whether examiners waived the accuracy-related penalties in accordance with applicable procedures.  Since AUR Program examiners are waiving penalties that should not have been waived, it may be beneficial to revise the quality reviews to evaluate the examiners’ compliance with procedures on waiving penalties to identify areas that need improvement.

Additionally, the CP 2000 Notice does not provide taxpayers with any instructions regarding the documentation that must be submitted when requesting a waiver of penalties.  A revision to the CP 2000 Notice may help better ensure that sufficient documentation is provided to assist the AUR Program examiners in determining whether the key reasonable cause criteria, as outlined in the IRM, is met.  For instance, in future updates to the CP 2000 Notice, the AUR Program could revise the notice to direct the taxpayer to an IRS website that explains what documentation needs to be provided to the IRS when requesting that a penalty be waived.  Since the AUR Program examiners are required to complete their assigned inventory in a limited amount of time, alerting the taxpayer of the documentation requirements for penalty relief on the CP 2000 Notice may reduce the need for the taxpayer to contact the IRS with additional questions.

Accuracy-related penalties were not always included on the CP 2000 Notice when warranted due to an inaccurate programming condition in the AUR system  

To ensure the fair and equitable treatment of taxpayers, applicable penalties should be consistently considered and assessed when warranted.  As previously discussed, the AUR Program relies on electronic business rules that are established to screen each case for applicable penalties.  However, during our review, we found that these business rules did not always ensure that the accuracy-related penalty (for substantial understatement of tax liabilities or negligence) is consistently included on the CP 2000 Notice against taxpayers who meet the designated criteria.  For example, as discussed earlier in the report, we identified seven cases from our statistical sample of 51 AUR cases for which the accuracy-related penalty was not included on the CP 2000 Notice although it met the criteria for the substantial understatement of tax.  For these seven cases, the CP 2000 Notice included at least one paragraph explaining that the accuracy-related penalty for substantial understatement of tax liabilities was included, but no actual penalty was included on the summary of proposed changes on the first page of the notice.  The accuracy-related penalties for substantial understatement of tax liabilities for these seven cases would have totaled $7,964. 

According to the IRS, the issue occurred due to an adjustment of the business rules, which resulted in no penalties being noted on the CP 2000 Notices for these cases despite the fact that the accuracy-related penalty for substantial understatement of tax liabilities was warranted.  AUR Program officials determined that approximately 65,000 AUR cases were affected by this flawed business rule change.  However, due to system limitations, the IRS cannot determine the actual amount of penalty dollars the affected cases would have incurred.[28]  In July 2012, the AUR Program identified the systemic issue that caused the lack of an accuracy-related penalty and actions were taken at that time to revise the business rules to prevent similar issues on future CP 2000 Notices.

We also found issues with the AUR system business rules pertaining to the accuracy-related penalty for negligence.  Specifically, our analysis of AUR Program data found that 7,715 cases (about 48 percent) of 16,069 AUR cases[29] for Tax Year 2010[30] were not assessed the accuracy-related penalty for negligence totaling approximately $2.66 million.[31]  The potential accuracy-related penalties for negligence ranged from $100 to $1,000 for the 7,715 cases.  According to AUR Program officials, the primary reason these particular cases lacked the accuracy-related penalties for negligence was an inaccurate programming condition in the AUR system.  Specifically, the AUR system’s business rules, developed by the AUR Program officials, were inaccurate, thus resulting in the AUR system not being programmed to correctly screen for all cases that had been identified as “repeaters.”  Therefore, the corresponding CP 2000 Notices did not include the applicable accuracy-related penalty for negligence.  As a result of our findings, AUR Program officials stated that they began planning to make revisions to the AUR system’s business rules to ensure that future CP 2000 Notices include the accuracy-related penalty for negligence for all cases that involved a “repeater.”

Accuracy-related penalties were not always calculated correctly

Even when accuracy-related penalties for negligence or substantial understatement of tax liabilities were assessed, we found problems with their calculations.  Specifically, although we determined that the accuracy-related penalties were correctly calculated for the vast majority (99.5 percent) of the AUR cases,[32] we identified 7,393 cases that resulted in accuracy-related penalties for negligence or substantial understatement of tax liabilities that may have been improperly calculated because the amount of the penalty did not equal 20 percent of the underpayment that appeared on the CP 2000 Notice.  While each case requires further investigation to determine if the assessed accuracy-related penalty was correctly calculated, after bringing these cases to AUR Program officials’ attention, they confirmed several erroneous penalty calculations resulted from issues with the AUR system or employee error.  For these 7,393 cases, we determined that the there was an aggregate overassessment[33] of accuracy-related penalties of approximately $2.4 million and an aggregate underassessment[34] of accuracy-related penalties of approximately $1.1 million.

The accuracy-related penalty for negligence is only assessed when a taxpayer is deemed a “repeater

I.R.C. § 6662 states that the IRS shall impose a penalty for any portion of underpayment of tax required to be shown on the return that is attributable to negligence.[35]  Negligence includes any failure to make a reasonable attempt to comply with the provisions of the internal revenue laws or to exercise ordinary and reasonable care in the preparation of a tax return.  Regulations[36] expressly state that negligence is strongly indicated if a taxpayer fails to include on an income tax return an amount of income shown on an information return. 

Given that the AUR Program is based on matching the taxpayer return information to the information returns, each underreporting case should be considered for an accuracy-related penalty for negligence.  However, the AUR Program designed its system so that it does not assess the accuracy-related penalty for negligence until a taxpayer’s second substantiated underreporting incident.  For instance, if the AUR Program determines that the taxpayer underreports income in one year, he or she would not be assessed an accuracy-related penalty for negligence unless they previously had a substantiated underreporting incident in any of the last three tax years in the same AUR Program discrepancy category (i.e., the taxpayer is deemed a “repeater”).  In light of the fact that the purpose of assessing penalties is to improve taxpayer compliance with the tax laws, it is difficult to understand the IRS’s position to wait until a taxpayer becomes a “repeater” to attempt to encourage a taxpayer towards greater compliance,[37] especially considering that IRS Chief Counsel stated in a FY 2008 memorandum that it was appropriate to apply the accuracy-related penalty for negligence even when a taxpayer was not deemed a “repeater.”  Specifically, IRS Chief Counsel stated:

Absent any additional information submitted by the taxpayer to explain why that income was omitted, it is appropriate to apply the accuracy-related penalty for negligence to taxpayers who omit that income.  It is all the more appropriate to assess the penalty when the taxpayer has failed to include income shown on an information return not only once, but twice.”[38]

Therefore, we believe that the AUR Program should consider the potential increase in compliance and additional revenue that could be generated if the accuracy-related penalty for negligence was assessed for all[39] AUR cases when the accuracy-related penalty for substantial understatement of tax liabilities did not apply.[40]  This, in turn, would also help ensure that the AUR Program imposes penalties in a fair and consistent manner as limiting the accuracy-related penalty to “repeaters” seems to contradict one of the IRS’s stated policies, as follows:

Consistency:  The IRS should apply penalties equally in similar situations.  Taxpayers base their perceptions about the fairness of the system on their own experience and the information they receive from the media and others.  If the IRS does not administer penalties uniformly (guided by the applicable statutes, regulations, and procedures), overall confidence in the tax system is jeopardized.[41]

It is difficult to conclude that the IRS has a fair and consistent policy with respect to assessing the accuracy-related penalty when the AUR system includes a penalty on every CP 2000 Notice that involves understatements large enough to warrant an accuracy-related penalty for a substantial underpayment of tax liabilities, whereas understatements beneath the substantial understatement thresholds are not assessed an accuracy-related penalty for negligence until the taxpayer becomes a “repeater.”

The IRS’s basis for this treatment of the negligence penalty comes from a long-standing AUR Program policy.[42]  Although this policy has been in existence for several decades, no rational basis has been provided for the policy, and there is no legal authority that prohibits the AUR Program from imposing the negligence penalty against taxpayers on their first offence.  Furthermore, this policy allows a significant amount of noncompliance to go unaddressed.  Specifically, we estimate that on an annual basis there are approximately 1.9 million taxpayers who have understatements of tax liabilities beneath the substantial understatement threshold that are not assessed an accuracy-related penalty for negligence.  If the AUR Program addressed negligence as it occurred rather than waiting until a taxpayer becomes a “repeater,” approximately $657 million in accuracy-related penalties for negligence could have been assessed for Tax Year 2010 alone.[43]

Recommendations

The Director, Exam/AUR Policy, Small Business/Self-Employed Division, should:

Recommendation 1:  Ensure consistency in waiving accuracy-related penalties in the AUR Program.  This may include, but is not limited to, requiring AUR Program examiners to document on a lead sheet the specific reasons for waiving penalties, managerial review of the nonassertion of penalties, developing and implementing training for first-line managers and their examiners, and revising quality reviews to evaluate whether AUR Program examiners are only waiving accuracy-related penalties in accordance with applicable policies and procedures.

Management’s Response:  IRS management agreed with this recommendation and developed a penalties lesson that was incorporated into the AUR examiner and manager training for FY 2015 and is planning a subsequent penalties lesson to be delivered beginning in FY 2016.  In addition, the current quality review processes include all aspects of a case, including whether accuracy-related penalties are waived in accordance with applicable law, policies, and procedures.  However, IRS management stated that they did not agree with our outcome measures.  The use of penalties to raise revenue, as suggested by the outcome measures, is inconsistent with the position of many stakeholders.

Office of Audit Comment:  We maintain that our outcome measure is reasonable.  Penalties encourage voluntary compliance by demonstrating the fairness and effectiveness of the tax system to compliant taxpayers by imposing an economic cost to those taxpayers who do not voluntarily comply with the tax laws.  TIGTA agrees that increasing revenue should not be the motivating factor in creating or imposing penalties.  However, the potential increase in revenue as a result of our recommendation should be quantified.  

Recommendation 2:  Ensure that the AUR system business rules are appropriately identifying cases that meet the accuracy-related penalty criteria and including the appropriate penalty on the taxpayer notice. 

Management’s Response:  IRS management agreed with this recommendation and implemented corrective action in November 2014 by reprogramming the AUR system for Tax Year 2013 inventory to identify all cases that are subject to the accuracy-related penalty.

Recommendation 3:  Continue to research the closed cases that we identified as having potentially inaccurate accuracy-related penalty amounts (for negligence and substantial understatement of tax liabilities) and, as needed, take the necessary corrective actions if the statute of limitations date has not yet expired.

Management’s Response:  IRS management agreed with this recommendation and will review AUR Tax Year 2010 cases identified by TIGTA and will issue refunds or correct assessments on open cases as authorized by statute.

Recommendation 4:  Ensure that the AUR Program addresses negligence as it occurs rather than when a taxpayer becomes a “repeater.”  Perform a study to determine to what level of understatement it would be beneficial to assess the negligence penalty in instances other than when taxpayers become “repeaters.” 

Management’s Response:  The IRS disagreed with this recommendation.  IRS management believes the current AUR policy provides fair and consistent treatment for all taxpayers, complies with applicable laws, and promotes voluntary compliance.  The IRS will continue to apply penalties in conformity with the law to promote compliance and effective tax administration.

Office of Audit Comment:  Negligence entails the failure to exercise ordinary and reasonable care in the preparation of a tax return.  Unreported or understated income, combined with the taxpayer’s failure to offer a reasonable explanation, is an indicator of negligence.  By proposing the negligence penalty on the initial CP 2000 Notice, the taxpayer would be given the opportunity to provide a reasonable explanation as to why that income item was omitted.  Under the IRS’s policy, only a “repeater” is subject to the negligence penalty in the AUR Program.  Under the IRS’s definition, a taxpayer is only a “repeater” if he or she omits income at least twice within a consecutive four-year period.  Therefore, a taxpayer could omit income once every four years and be a “first-timer” in the eyes of the IRS, possibly contributing to further noncompliance.  Addressing negligence as it occurs rather than when a taxpayer becomes a “repeater” would promote fair and consistent treatment of all taxpayers, helping them to voluntarily comply with the law and avoid future penalties. 

It is important to clarify that TIGTA is not proposing that the AUR Program apply the negligence penalty to all taxpayers identified in its inventory.  Instead, AUR management should establish a level of understatement at which it would be beneficial to assess the negligence penalty on taxpayers, other than repeaters, in order to promote compliance and to help better ensure fair and consistent treatment for all taxpayers.  A more consistent application of the penalty could provide incentive to taxpayers to avoid inaccurate tax reporting positions. 

Revisions to the Computer Paragraph 2000 Notice May Not Have Produced the Anticipated Results

The AUR Program’s CP 2000 Notice is the primary notice that the IRS issues to a taxpayer as a result of underreporting income.  It is a complex, multipage document that compares income, credit, and deduction information reported to the IRS by third parties to the information provided by the taxpayer on his or her tax return.  In a FY 2008 TIGTA report,[44] we reported that the complexity of the CP 2000 Notice contributed to taxpayers agreeing to incorrect assessments of additional tax liability.  In addition, taxpayer satisfaction surveys conducted by the SB/SE and W&I Divisions found that a significant portion of the taxpayers surveyed were dissatisfied with the CP 2000 Notices.  One of the main reasons taxpayers were dissatisfied with the CP 2000 Notice was the difficulty in understanding the discrepancy identified and what actions to take in response to the notice.

In FY 2008, the IRS Commissioner created the Taxpayer Communications Taskgroup in an effort to provide strategic guidance and direction in improving IRS written communications with taxpayers.  The Taxpayer Communications Taskgroup used an independent contractor to conduct a study and create new processes for the IRS to develop and deliver simplified taxpayer communications that are clear, accurate, and effective.  The study provided a detailed analysis on the most effective ways to communicate with taxpayers about their accounts via notices and letters throughout several IRS communication delivery systems, including paper products and electronic media.  The goal of the study was to encourage compliance while minimizing taxpayer confusion. 

The study resulted in revisions to the CP 2000 Notice to better ensure that the recipients of the notice were properly notified of the discrepancies and what actions they needed to take in response to the notice.  The revised CP 2000 Notice was implemented in December 2012, for Tax Year 2011 AUR cases.  The goal of the revisions to the CP 2000 Notice was to clearly explain the actions required of the taxpayer by clarifying the options available to them and simplifying the tax calculations.  Although the revised CP 2000 Notice was implemented approximately two years ago, the IRS has not evaluated, or established plans to evaluate, the effectiveness of the revised notice on reducing taxpayer underreporting.

IRM and GAO guidance[45] require establishment of performance measures to assist in determining how well a new process is working compared to past performance.  Measuring performance objectively is important for making effective business decisions by determining whether desired results are being obtained and whether further improvements may be needed.  Additionally, the GAO provides further guidance, stating that a sound evaluation plan should be developed early and should include such details as:  1) clear and measureable objectives, 2) standards for determining performance and measuring success against stated objectives, and 3) details about the type and amount of data needed for program evaluation and how it will be collected.  Such information is critical to properly evaluate the revised CP 2000 Notice and determine whether the revisions achieved the desired results or if further improvements may be needed.

TIGTA used AUR customer satisfaction surveys to measure the impact the revised CP 2000 Notice may have had on overall taxpayer satisfaction as well as determine whether the revisions helped clarify the tax issues and what actions the taxpayer needed to take in response to the notice.  Since the revised notice was implemented in December 2012, we compared the surveys for FY 2012 and FY 2013 to identify any noticeable increases or decreases in taxpayer responses before and after implementation.  The overall customer satisfaction of taxpayers decreased slightly in both the SB/SE and W&I Divisions.  In addition, although the SB/SE Division found that there was some improvement in customer satisfaction regarding the ability to understand the reasons for the tax discrepancies noted on the CP 2000 Notice, the W&I Division’s customer satisfaction in this category declined.  While these results do not in themselves indicate the overall impact of the revised CP 2000 Notice, the results indicate that the revised notice may not be providing the intended benefits.

At the time of our review, the AUR Program, W&I Division’s Research and Analysis Office, and the Office of Taxpayer Correspondence had not yet established any plans for determining the effectiveness of the revised CP 2000 Notice.  However, AUR Program officials stated that they plan to work with the W&I Division’s Research and Analysis Office to identify the reason for the decline in customer satisfaction as well as evaluate the effectiveness of the revised CP 2000 Notice.

Recommendation

Recommendation 5:  The Director, Exam/AUR Policy, Small Business/Self-Employed Division, should evaluate the effectiveness of the revisions to the CP 2000 Notice and the potential impact it has had on taxpayer satisfaction and compliance.

Management’s Response:  IRS management agreed with our recommendation and will partner with the Office of Taxpayer Correspondence to review the revised CP 2000 Notice to determine the effectiveness of the changes on compliance.

 

Appendix I

 

Detailed Objectives, Scope, and Methodology

 

The objectives of this review were to determine the type and extent of noncompliance addressed by the AUR Program and if additional steps could be taken to enhance program accomplishments.  To accomplish these objectives, we:

I.                 Analyzed the type and extent of noncompliance addressed by the AUR Program[46] by interviewing AUR Program management, obtaining documentation that establishes program policy and practices regarding AUR notices sent to taxpayers, and reviewing documentation that establishes the goals and measures for the program.

II.               Evaluated the IRS’s efforts to correct noncompliance by reviewing studies and initiatives to improve the case selection process in order to select more productive cases and reduce the screen-out rates.[47]

III.             Determined whether the accuracy-related penalty for substantial understatement of tax liabilities was assessed when warranted.

A.    Used the IRS Individual Master File to identify the population of closed AUR cases involving a Tax Year 2010 individual tax return for which a CP 2000 Notice, Request for Verification of Unreported Income, Payments, or Credits, was issued and the understatement of tax liabilities exceeded 10 percent of the tax required to be shown on the return, with tax assessments of greater than or equal to $5,000, and that were not assessed the accuracy-related penalty.  This analysis identified 9,339 records. 

B.    Validated the data obtained in Step III.A by comparing the data to the Integrated Data Retrieval System.  Specifically, we reconciled a judgmental sample of 10 records from our population of closed Tax Year 2010 individual tax returns to verify that the records were worked by the AUR Program and were in fact closed during FY 2012.  Based on our sample, we determined that the data were sufficiently reliable for the purposes of this report.

C.    Selected and reviewed a statistically valid (simple random) sample of 51[48] closed AUR case files, as well as information contained on the AUR system, from a population of 9,339 closed AUR cases involving a Tax Year 2010 individual tax return to determine whether the accuracy-related penalty was included on the CP 2000 Notice.  We also determined whether the accuracy-related penalty was removed and, if so, whether the removal of the penalty was in compliance with applicable IRM procedures.  Our statistically valid sample of 51 closed AUR cases was selected using a 95 percent confidence level, ± 13.64 percent precision rate, and 50 percent occurrence rate.  A statistically valid sample was taken because we wanted to estimate the number of cases and the amount of additional accuracy-related penalties that were improperly waived for the population of 9,339 AUR cases.

D.    Provided exception cases identified during our sample review to appropriate AUR Program personnel to obtain their agreement.

E.     Collaborated with TIGTA’s contracted statistician in developing the sampling plans and forecasts.

IV.            Determined whether the accuracy-related penalty for negligence was included on the CP 2000 Notices when warranted.

A.    Requested data extracts from the AUR Program of all “repeaters” for Tax Year 2010 who had repeated the same category in at least one of the last three tax years.  Tax Year 2010 was selected for this analysis because it is the most recent and complete tax year available (i.e., the AUR Program completed the overwhelming majority of Tax Year 2010 cases as of the end of Calendar Year 2013).  Due to data limitations, the IRS was only able to provide TIGTA with a subset of the “repeater” dataset.  This subset included the most egregious repeaters (i.e., those that repeated noncompliance in the same category at least two times in the prior three tax years).

B.    Obtained Master File data for the population identified in Step IV.A.

C.    Validated the data obtained in Step IV.B by comparing it to the Integrated Data Retrieval System.  Specifically, we reconciled a judgmental sample of 10 records from the subset of the Tax Year 2010 “repeaters” dataset to verify that the records were worked by the AUR Program.  Based on our sample, we determined that the data were sufficiently reliable for the purposes of this report.

D.    Performed an analysis using the combined datasets from Steps IV.A and IV.B to determine the population of cases that involved unabated additional AUR tax assessments that were/were not assessed an accuracy-related penalty.

E.     Reviewed a judgmental sample[49] of cases that were not assessed an accuracy-related penalty to identify the reasons the penalties were not assessed.

F.     Obtained agreement from AUR Program officials on our data methodology and the results of our data analyses.

V.              Determined the number of taxpayers and amount of accuracy-related penalty for negligence that could be assessed if such penalty was consistently applied for all taxpayers.

VI.            Determined whether assessed accuracy-related penalties were computed correctly among AUR cases currently on the Master File.  Tax years associated with these cases ranged from Tax Years 1978 to 2012.

A.    Used the IRS Individual Master File to identify all available AUR cases for which the assessed and unabated accuracy-related penalty that posted to a taxpayer account was either less than or greater than 20 percent of the calculated underpayment of tax.  This analysis identified 7,393 records.

B.    Validated the data obtained in Step V.A by comparing the data to the Integrated Data Retrieval System.  Specifically, we reconciled a judgmental sample of 10 records to verify that the records were worked by the AUR Program.  Based on our sample, we determined that the data were sufficiently reliable for the purposes of this report.

C.    Selected a judgmental sample of cases from Step V.A that were assessed an accuracy-related penalty greater than 20 percent of the net tax increase (overassessment) or less than 20 percent of the net tax increase (underassessment) to determine whether the penalty was computed correctly.

D.    Obtained agreement from AUR Program officials on our data methodology and the results of our data analyses.

VII.          Determined whether a methodology had been developed to measure the effectiveness of the revised CP 2000 Notice by interviewing IRS personnel from the AUR Program and the Office of Taxpayer Correspondence.

Internal controls methodology

Internal controls relate to management’s plans, methods, and procedures used to meet their mission, goals, and objectives.  Internal controls include the processes and procedures for planning, organizing, directing, and controlling program operations.  They include the systems for measuring, reporting, and monitoring program performance.  We determined the following internal controls were relevant to our audit objectives:  IRS policies, procedures, and practices for:  

·       Identifying discrepancies between information returns and taxpayers’ returns and the notices issued to taxpayers as a result of such discrepancies.

·       Assessing accuracy-related penalties for negligence and substantial understatement of tax liabilities.

·       Waiving accuracy-related penalties for substantial understatement of tax liabilities and negligence. 

We evaluated these controls by reviewing source materials, interviewing management, reviewing a statistical sample of 51 closed AUR cases to determine if an accuracy-related penalty for substantial understatement of tax liabilities was assessed when warranted and waived in accordance with policies and procedures, and performing aggregate-level data analyses to determine whether accuracy-related penalties for negligence and substantial understatement of tax liabilities were correctly assessed when warranted.

 

Appendix II

 

Major Contributors to This Report

 

Nancy Nakamura, Assistant Inspector General for Audit (Compliance and Enforcement Operations)

Matthew Weir, Assistant Inspector General for Audit (Compliance and Enforcement Operations)

Randee Cook, Acting Assistant Inspector General for Audit (Compliance and Enforcement Operations)

Frank Dunleavy, Acting Assistant Inspector General for Audit (Compliance and Enforcement Operations)

Bryce Kisler, Acting Assistant Inspector General for Audit (Compliance and Enforcement Operations)

Glen Rhoades, Director

Michelle Philpott, Audit Manager

Curtis Kirschner, Acting Audit Manager

Donna Saranchak, Lead Auditor

Alberto Garza, Senior Auditor

William Tran, Senior Auditor

 

Appendix III

 

Report Distribution List

 

Commissioner  C

Office of the Commissioner – Attn:  Chief of Staff  C

Deputy Commissioner for Services and Enforcement  SE

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

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

Deputy Commissioner, Wage and Investment Division  SE:W

Director, Strategy and Finance, Wage and Investment Division  SE:W:S

Director, Headquarters Examination SE:S:E:HQ

Director, Examination/AUR Policy  SE:S:E:HQ:EP

Acting Director, Exam Performance Planning & Analysis SE:S:E:PP&AE

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:S

Commissioner, Wage and Investment Division  SE:W

 

Appendix IV

 

Outcome Measures

 

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

Type and Value of Outcome Measure:

·       Increased Revenue – Potential; $3.25 million in additional accuracy-related penalties for substantial understatement of tax liabilities; $16.25 million forecasted over five years (see page 7).[50]

Methodology Used to Measure the Reported Benefit:

To estimate the potential additional revenue associated with the AUR Program’s inappropriate waiver of accuracy-related penalties for substantial understatement of tax liabilities, we originally requested 144 closed Tax Year 2010 AUR cases in order to meet our sample criteria of 100 cases.  We received 109 AUR cases, of which 51 AUR cases met the criteria for the accuracy-related penalty for substantial understatement of tax liabilities, i.e., the understatement exceeded 10 percent of the tax required to be shown on the return and the tax was greater than or equal to $5,000.  The remaining 58 AUR cases did not meet the criteria for an accuracy-related penalty for substantial understatement of tax liabilities and therefore were not reviewed. 

From the sample of 51 closed AUR cases, we identified 21 cases for which the accuracy-related penalty for substantial understatement of tax liabilities was incorrectly waived by the AUR Program examiner.  Because we determined that a percentage of our population from which we selected our sample included closed AUR cases that did not meet our criteria, we had to account for this when projecting our results to the population.  Therefore, our rate to project to the population was based on the 51 AUR closed cases we reviewed as well as the 58 AUR closed cases that did not meet the criteria for the accuracy-related penalty (51 + 58 = 109 closed AUR cases).  As a result, our error rate for projection was calculated as 19.27 percent (21/109).

To estimate the total amount of additional accuracy-related penalties that could have been assessed for the 9,339 closed AUR cases for which taxpayers substantially understated their taxes, we computed the accuracy-related penalties that were warranted for the 21 closed AUR cases we identified as exception cases.  For these 21 closed AUR cases, we determined that $37,980 in accuracy-related penalties could have been assessed.  We calculated the average accuracy-related penalty by dividing the accuracy-related penalties that should have been assessed for the 21 exception cases by the number of cases received for our original sample[51] and determined the average was $348 ($37,980/109).[52]  We multiplied the number of cases in the total population by the average penalty to estimate the total amount of additional accuracy‑related penalties that could have been assessed for the closed AUR cases involving a Tax Year 2010 individual tax return [9,339 x $348 = $3,249,972].

We then estimated the amount of additional penalties owed that could be assessed over
five
years if the AUR Program revised its procedures to ensure that penalties were not inappropriately waived.  We multiplied the total amount of additional penalties we estimated could have been assessed for the closed AUR cases involving a Tax Year 2010 individual tax return by five [$3.25 million x 5 = $16.25 million].

Type and Value of Outcome Measure:

·          Increased Revenue – Potential; $2.66 million in additional accuracy-related penalties for negligence in 7,715 closed AUR cases for Tax Year 2010 (see page 7).

Methodology Used to Measure the Reported Benefit:

To estimate the potential additional revenue associated with Tax Year 2010 AUR closed cases that had not been assessed accuracy-related penalties for negligence in accordance with the AUR Program’s policies and procedures, we reviewed a subset[53] of the AUR Program’s Tax Year 2010 “repeater”[54] dataset, which included 31,139 cases that were screened by the AUR Program.  Of these, we identified 16,069 (52 percent) AUR cases that resulted in a CP 2000 Notice, Request for Verification of Unreported Income, Payments, or Credits, had unabated tax assessments, and resulted in underpayments ranging from $501 to $5,000.

Of the 16,069 cases, we identified 8,748 cases (54 percent) that did not have an accuracy-related penalty assessed although all 8,748 cases met the criteria for the accuracy-related for negligence.  These 8,748 cases had total underpayments of $14,910,106.  Had all of these cases resulted in accuracy-related penalties, the total estimated penalties would have equaled $2.98 million. 

According to AUR Program officials, these 8,748 cases should have been identified by the AUR system and the CP 2000 Notices should have automatically included an accuracy-related penalty for negligence.  However, 7,715 (88 percent) of the 8,748 cases were not classified as negligent cases due to an inaccurate AUR system programming condition.  As a result, these 7,715 cases avoided the automatic inclusion of the accuracy-related penalty for negligence.[55]  We determined that these 7,715 cases resulted in $13.3 million in total underpayments and, therefore, approximately $2.66 million in accuracy-related penalties for negligence[56] should have been assessed but were not.  This was calculated by multiplying the $13.3 million in total underpayments by the 20 percent negligence penalty.  Our calculation assumes that the penalty revenue resulting from the tax deficiency would have been included on the CP 2000 Notice, assessed, and paid by the taxpayer.

Type and Value of Outcome Measure:

·          Increased Revenue – Potential; $657 million in accuracy-related penalties for negligence in 1.9 million closed AUR cases for Tax Year 2010 (see page 7).

Methodology Used to Measure the Reported Benefit:

To estimate the potential additional revenue associated with Tax Year 2010 AUR closed cases that had not been assessed accuracy-related penalties for negligence but could have been if the AUR Program’s policy did not limit this penalty to taxpayers that repeated noncompliance, we reviewed the complete population of the AUR Program’s Tax Year 2010 records available on the IRS Master File, which included 5,366,304 AUR cases.  Of these, we identified 2,869,313 (53 percent) AUR cases that resulted in a CP 2000 Notice and unabated tax assessments. 

Of the 2,869,313 AUR cases that had unabated tax assessments, we identified 2,099,935 (73 percent) that had underpayments ranging from $501 to $5,000.[57]

Of the 2,099,935 cases that had underpayments ranging from $501 to $5,000, we identified 1,966,805 (94 percent) cases that were not assessed an accuracy-related penalty.  These 1,966,805 cases had total underpayments totaling $3,301,422,764.  Therefore, approximately $660,284,553 in accuracy-related penalties for negligence could have been assessed but were not.  This was calculated by multiplying the $3,301,422,764 in total underpayments by the 20 percent negligence penalty.  However, since we previously determined (see analysis in prior outcome measure) that 8,748 Tax Year 2010 cases met the AUR Program’s existing criteria for the accuracy-related for negligence (i.e., they had to repeat the same type of noncompliance) we removed these cases from our population so the same cases would not be included in both outcome measures.  Specifically, we subtracted the 8,748 cases and the associated $2.98 million in accuracy-related penalties from this outcome measure, which resulted in 1,958,057 cases and $657,302,532 in accuracy-related penalties for negligence that could have been assessed an accuracy-related penalty for negligence if the AUR Program’s policy did not limit such penalty to those taxpayers with repeated noncompliance.

Our calculation assumes that the penalty revenue resulting from the tax deficiency would have been included on the CP 2000 Notice, assessed, and paid by the taxpayer.  In addition, our calculation assumes that these cases did not have an accuracy-related penalty included on the summary of proposed changes on the CP 2000 Notice because, due to data limitations, we are unable to determine which cases, if any, had the accuracy-related penalty waived due to reasonable cause.  Furthermore, we assumed that the underpayment is equal to the understatement of tax.[58]

 

Appendix V

 

Example of a Reasonable Cause Lead Sheet

 

This appendix includes an example of a Reasonable Cause Lead Sheet used by Examination revenue agents and tax compliance officers to determine whether a penalty should be waived.

 

Reasonable Cause Lead Sheet

Conclusion: (Reflects the final determination on the issue.)

Audit Steps: (Document the audit staffs taken or to be taken.)

Workpaper Reference

Following is a list of the more common reasons offered by taxpayers as to why a penalty should not be applied.  Indicate which reason(s) have been advanced by the taxpayer.  Regardless of the reasons advanced, the taxpayer must have exercised ordinary business care and prudence.  In order to properly consider any reason, refer to the Penalty Handbook, IRM 20.1.1.3, along with other IRM 20.1 chapters for specific penalties.

 

1.       IGNORANCE OF THE LAW

         What is taxpayer’s education?

         Has the taxpayer be subject to tax previously?

         Has the taxpayer been previously penalized?

         Were there any recent tax law changes?

         Did the taxpayer comply within a reasonable period of time once they became aware of their obligation?

 

2.       DEATH, ILLNESS OR UNAVOIDABLE ABSENCE

         What was the relationship of the taxpayer to other parties involved?

         Date of death or illness?

         How the event prevented compliance?

         Were other business obligations impaired?

         Did the taxpayer comply within a reasonable period of time after the death, illness or unavoidable absence?

 

3.       WRITTEN AND ORAL ADVICE FROM IRS

         Was the advice in response to a specific request and was the advice received related to the facts contained in that request?

         Did the taxpayer reasonably rely on that advice?

         Did that taxpayer provide the IRS with adequate and accurate information?

         Is there supporting evidence that advice was received?

         Did the taxpayer comply within a reasonable period of time once they learned the advice was incorrect?

 

4.  Fire, Casualty or Natural Disaster

         What was the timing of the event in relation to the noncompliance?

         What effect did it have on the taxpayer’s business?

         What steps did the taxpayer take to comply?

         Was the taxpayer located in an official disaster area?

         Did the taxpayer comply within a reasonable period of time after the casualty or disaster?

 

5.  Mistake was Made and Forgetfulness

         Generally, this is not in keeping with the ordinary business care and prudence standard.

         When and how did the taxpayer become aware of the mistake?

         Did the taxpayer take timely steps to correct the mistake after it was discovered?

 

6.  Unable to Obtain Records

         Why were the records needed to comply?

         Why were the records unavailable and what steps were taken to secure the records?

         Did the taxpayer take other steps to comply?

         Does taxpayer have documentation to verify their effort to secure the needed information?

         Did the taxpayer comply within a reasonable period of time once they obtained their records?

 

 

Reasonable Cause Lead Sheet

7.  Advice from Tax Advisor

         Generally limited to technical and complicated issues.

         Was it reasonable for the taxpayer to reply on the advice of their CPA or tax attorney?

         Is there supporting evidence that written advice was given or that advice was received?

         Did the advice make sense compared to the amount taxpayer would invest, or be at risk, if the IRS later disallowed the item?

         Taxpayer’s responsibility to file, pay or deposit taxes cannot be excused by reliance on the advice of a tax adviser.

         What steps did taxpayer take to ensure that the advice was reasonable?

 

8.   Other

Facts: (Document the relevant facts.)

Law:  (Tax Law, Regulations, court cases, and other authorities. (If Unagreed, add Argument)

Taxpayer Position:  (If applicable}

Source:  Office of Servicewide Penalties Intranet Site.

 

Appendix VI

 

Glossary of Terms

 

Assessment – An assessment is the statutorily required recording of the tax liability.  This generally happens when the IRS determines that the taxpayer owed more taxes than what was reported on the tax return.

Automated Underreporter System – The case processing and inventory control system of the AUR Program.

Business Master File – The IRS database that consists of Federal tax-related transactions and accounts for businesses.  These include employment taxes, income taxes on businesses, and excise taxes.

Business Rules – Criteria established by the AUR Program to select case inventory.  These rules are designed to validate line items on a tax return, such as income limits for tax credits and deductions and the need to attach a specific schedule if certain conditions are present on the tax return.  A tax return may be included in AUR inventory if one or more of the business rules are met.

Campus – The data processing arm of the IRS.  The campuses process paper and electronic submissions, correct errors, and forward data to the IRS’s Computing Centers for analysis and posting to taxpayer accounts.

Computer Paragraph – A computer-generated notice resulting from an analysis of the taxpayer’s account.  It is used to notify the taxpayers of a balance due, refund, or no balance status. 

Examiner – IRS employees who examine tax returns to determine whether taxpayers accurately reported their tax liabilities.

Fiscal Year – Any yearly accounting period, regardless of its relationship to a calendar year.  The Federal Government’s fiscal year begins on October 1 and ends on September 30.

Full-Time Equivalent – A measure of labor hours in which one FTE is equal to eight hours multiplied by the number of compensable days in a particular fiscal year.  For FYs 2013 and 2014, one FTE was equal to 2,088 staff hours.

Individual Master File – The IRS database that maintains transactions or records of individual tax accounts. 

Integrated Data Retrieval System – The IRS computer system capable of retrieving or updating stored information.  It works in conjunction with a taxpayer’s account records.

Internal Revenue Manual – The primary, official source of instructions to staff that relate to the administration and operation of the IRS.  It details the policies, delegations of authorities, procedures, instructions, and guidelines for daily operations for all IRS organizations.

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

Nonemployee Compensation – Refers to fees, commissions, or any other compensation paid by a business to an individual who is not an employee.  Nonemployee compensation is reported on Form 1099-MISC, Miscellaneous Income.

Repeaters – Taxpayer who underreported the same type of income in at least one of the last three prior tax years and was also assessed taxes.

Revenue Procedure An official statement of a procedure published that either affects the rights or duties of taxpayers or other members of the public under the I.R.C. and related statutes, treaties, and regulations or, although not necessarily affecting the rights and duties of the public, should be a matter of public knowledge.

Screen-Out Rate – The percentage of AUR inventory for which the discrepancies are resolved internally and without taxpayer contact.

Tax Year – The 12-month period for which tax is calculated.  For most individual taxpayers, the tax year is synonymous with the calendar year.

Wages – Compensation for employee services that are not subject to self-employment tax and are reported to the IRS via a Form W-2, Wage and Tax Statement.

 

Appendix VII

 

Management’s Response to the Draft Report

 

DEPARTMENT OF THE TREASURY

INTERNAL REVENUE SERVICE

WASHINGTON, D.C. 20224

 

                            COMMISSIONER

SMALL BUSINESS/SELF·EMPLOYED DIVISION

 

April 16, 2015

 

 

 

 

MEMORANDUM FOR MICHAEL E. MCKENNEY

                                          DEPUTY INSPECTOR GENERAL FOR AUDIT

 

 FROM:                   for    Karen Schiller /s/ Mary Beth Murphy

                                          Commissioner, Small Business/Self-Employed Division

 

SUBJECT:                       Draft Audit Report - Automated Underreporter Program Tax Assessments Have Increased Significantly; However, Accuracy­ Related Penalties Were Not Always Assessed When Warranted (Audit# 201230026)

 

Thank you for the opportunity to review the subject report which examined the type and extent of noncompliance addressed by the Automated Underreporter Program (AUR).  We appreciate your acknowledgment that the AUR Program contributed positively to closing the Tax Gap and has improved overall productivity and performance despite declining budgets and the corresponding reductions in staffing.  We continually look for ways to improve compliance and provide benefits to taxpayers trying to comply with their filing and reporting obligations.  The purpose of the AUR Program is to address non-compliance by matching taxpayer income and deduction information submitted by third parties to amounts reported on individual income tax returns and to attempt to touch as many categories or segments of non-compliant taxpayers as resources allow.  With this program, we achieve the dual results of identifying amounts due to the Government and ensuring fairness to all taxpayers through uniform application of the tax code.

 

We have enhanced the AUR inventory selection process and have increased annual case closures by 400,000, resulting in the AUR Program increasing tax assessments from $4.24 billion in FY 2006 to $7.84 billion in FY 2013.  We are encouraged by your finding that AUR correctly calculated accuracy-related penalties 99.5 percent of the time.

 

Your audit found that improvements were needed to ensure accuracy related penalties are assessed as warranted,  Based on that finding and the resulting recommendations, TIGTA determined an outcome measure of "additional revenue" in the amount of $3.25 million, $16.25 million projected over five years.  We disagree with your presentation of the outcome measures in terms of "additional revenue."  The use of penalties to raise revenue, as suggested by the outcome measures, is inconsistent with the position of many stakeholders,  Treasury's Report to Congress on Penalties and Interest Provisions of the Internal Revenue Code (October 1999) (p. 36) states:  ''Penalties may raise revenue collaterally but this should not be a deliberate objective of penalty design.”'  Likewise, the American Bar Association, Section on Taxation has stated in both its 1988 Civil Penalty Tax Force Section Report and its 2009 Statement of Policy Favoring Reform of Federal Civil Tax Penalties states that using penalties to raise revenue or offset costs is detrimental to tax administration and discourages voluntary compliance.  In addition to disagreeing with the presentation of the outcome measures, we do not agree with your computation because it fails to account for any penalties which may be abated due to reasonable cause.  More importantly, your outcome measures include a computation assuming a policy change for which we do not agree specifically due to our assumption these taxpayers may be more likely to have reasonable cause.

 

We agree that penalties should be applied and waived on a consistent basis.  The Internal Revenue Manual (IRM) states that "consistency" means that "[t]he IRS should apply penalties equally in similar situations."  In accordance with the IRM, AUR policy provides uniform rules for applying the negligence penalty and likewise provides uniform rules for applying the substantial understatement of tax penalty.  We disagree with your position that "consistency" requires AUR to use the same set of rules to applying both penalty provisions as these are distinct provisions that apply to significantly different factual situations.  Negligence includes a taxpayer's failure to reasonably comply with the tax laws, failure to exercise reasonable care in the preparation of a tax return, failure to keep adequate books and records, or failure to properly substantiate items properly.  In contrast, for individuals, an understatement of income tax is substantial if the understatement for the year exceeds the greater of 10 percent of the tax required to be shown on the return or $5,000.  By definition, an understatement that is "substantial" is not the mere failure to report income or overstate expenses but, rather, the failure to correctly report items on the return that is significant and results in a substantial understatement of tax.

 

In addition, the application of a substantial understatement of tax penalty is a mathematical calculation that can be supported in the first instance in the AUR context.  In contrast, the application of the negligence penalty involves a determination of whether the error occurred despite the taxpayer's reasonable attempt to comply with the tax law.  Accordingly, when we identify taxpayers who have failed to report income for the first time and the failure does not result in a substantial understatement of tax, we notify them of their errors and allow them to pay the additional tax and interest owed.  Our approach conforms to the regulations that state that reasonable cause can exist even when a taxpayer commits an isolated computational or transcriptional error, such as a first-time failure to report income.  We propose the negligence penalty to repeaters because their mistakes are no longer isolated and the conclusion that they are negligent is more strongly supported.  The Office of Chief Counsel has stated that the AUR program "appropriately assesses" the negligence penalty.  The approach proposed by TIGTA is contrary to the position of many stakeholders, including Congress.  The Committee Report on the Omnibus Reconciliation Act of 1989 (p. 661) recommends that "in the application of penalties, the IRS should make a correct substantive decision in the first instance rather than mechanically assert penalties with the idea that they will be corrected later."

 

To promote consistency in application of the reasonable cause provision, the IRM provides examples of factors that may support reasonable cause but also states that reasonable cause relief is not limited to those factors and may be based on other acceptable explanations.  Additional training and guidance on penalties was delivered to employees in Fiscal Year (FY) 2015 and training is also planned for FY 2016.  We further support consistency through our quality control processes such as Notice Reviews, Manager Reviews and Program Analysis System (PAS) reviews.

 

Attached is a detailed response outlining our corrective actions to address your recommendations.

 

If you have any questions, please contact me, or a member of your staff may contact Shenita Hicks, Director, Examination, Small Business/Self-Employed Division at 859·669-5526.

 

Attachment

 

Attachment

 

The Director, Exam/AUR Policy, Small Business/Self-Employed Division, should:

 

RECOMMENDATION 1:

Ensure consistency in waiving accuracy-related penalties in the AUR Program.  This may include, but is not limited to, requiring AUR Program examiners to document on a lead sheet the specific reasons for waiving penalties, managerial review of the nonassertion of penalties, developing and implementing training for first-line managers and their examiners, and revising quality reviews to evaluate whether AUR Program examiners are only waiving accuracy-related penalties in accordance with applicable policies and procedures.

 

CORRECTIVE ACTION:

To address TIGTA's concerns with waiving accuracy-related penalties, we have already developed a penalties CPE lesson that was incorporated into the AUR examiner and manager training for FY 2015 and are planning a subsequent penalties CPE lesson to be delivered beginning in FY 2016.  In addition, the current quality review processes include all aspects of a case, including whether accuracy-related penalties are waived in accordance with applicable law, policies, and procedures.

 

IMPLEMENTATION DATE:

Implemented

 

RESPONSIBLE OFFICIAL:

Director, Examination/AUR Policy, Small Business/Self-Employed Division

 

CORRECTIVE ACTION MONITORING PLAN:

N/A

 

RECOMMENDATION 2:

Ensure that the AUR system business rules are appropriately identifying cases that meet the accuracy-related penalty criteria and including the appropriate penalty on the taxpayer notice.

 

CORRECTIVE ACTION:

We implemented this recommendation in November 2014 by reprogramming the AUR system for the tax year 2013 inventory to identify all cases that are subject to the accuracy-related penalty.

 

IMPLEMENTATION DATE:

Implemented

 

RESPONSIBLE OFFICIAL:

Director, Examination/AUR Policy, Small Business/Self-Employed Division

 

CORRECTIVE ACTION MONITORING PLAN:

N/A

 

RECOMMENDATION 3:

Continue to research the closed cases that we identified as having potentially inaccurate accuracy-related penalty amounts (for negligence and substantial understatement of tax liabilities) and, as needed, take the necessary corrective actions where the statute of limitations date has not yet expired.

 

CORRECTIVE ACTION:

We will review AUR Tax Year 2010 cases identified by TIGTA and will issue refunds or correct assessments on open cases, as authorized by statute.

 

IMPLEMENTATION DATE:

March 15, 2016

 

RESPONSIBLE OFFICIAL:

Director, Examination/AUR Policy, Small Business/Self-Employed Division

 

CORRECTIVE ACTION MONITORING PLAN:

IRS will monitor this corrective action as part of our internal management system of control.

 

RECOMMENDATION 4:

Ensure that the AUR Program addresses negligence as it occurs rather than when a taxpayer becomes a "repeater,'' Perform a study to determine to what level of understatement it would be beneficial to assess the negligence penalty in instances other than when taxpayers become "repeaters,''

 

CORRECTIVE ACTION:

Our current AUR policy provides fair and consistent treatment for all taxpayers, complies with applicable law, and promotes voluntary compliance.  We will continue to apply penalties in conformity with the law to promote compliance and effective tax administration,

 

IMPLEMENTATION DATE:

N/A

 

RESPONSIBLE OFFICIAL:

N/A

 

CORRECTIVE ACTION MONITORING PLAN:

N/A

 

RECOMMENDATION 5:

Evaluate the effectiveness of the revisions to the CP 2000 Notice and the potential impact it has had on taxpayer satisfaction and compliance.

 

CORRECTIVE ACTION:

We will partner with the Office of Taxpayer Correspondence (OTC) to review the revised CP 2000 to determine the effectiveness of the changes on compliance.

 

IMPLEMENTATION DATE:

March 15, 2016

 

RESPONSIBLE OFFICIAL:

Director, Examination/AUR Policy, Small Business/Self-Employed Division

 

CORRECTIVE ACTION MONITORING PLAN:

IRS will monitor this corrective action as part of our internal management system of controls.



[1] The AUR Program is subject to Revenue Procedure 2005-32 Section 4.03 (June 2005).  The AUR Program has the right to verify a discrepancy between the taxpayer’s tax return and an information return or between a tax return and information otherwise in the IRS’s possession. 

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

[3] In Fiscal Year (FY) 2013, the AUR Program expanded its matching program to identify discrepancies from matches between corporate taxpayer returns (e.g., Form 1120, U.S. Corporation Income Tax Return) and information returns submitted by third parties to verify that all income is reported. 

[4] In some cases, a CP 2501 Notice, Initial Contact to Resolve Discrepancy Between Income, Credits, and/or Deductions Claimed on Return and Those Reported by Payer, will be issued prior to issuance of a CP 2000 Notice.  The CP 2501 Notice is sent prior to the CP 2000 Notice when the taxpayer is owed a refund or when the discrepancy identified by the AUR Program will result in the taxpayer owing $10,000 or more in additional taxes.

[5] Internal Revenue Code (I.R.C.) section (§) 6212.

[6] This includes accuracy-related penalties for negligence and substantial understatement of tax liabilities. 

[7] The additional tax assessment amounts do not include assessed interest or penalties.

[8] Generally, the IRS cannot extend the due date of an individual tax return for more than six months (to October 15 for most calendar year taxpayers).  However, there may be an exception for individuals living out of the country.  See IRS Publication 54, Tax Guide for U.S. Citizens and Resident Aliens Abroad, for more information.

[9] I.R.C. § 6662 (c) defines “negligence” as including any failure to make a reasonable attempt to comply with the provisions of the law and “disregard” as including any careless, reckless, or intentional disregard.

[10] I.R.C. § 6662 (d)(1)(A). 

[11] According to I.R.C. § 6662(d)(2)(B), there are exceptions to the substantial understatement penalty, which include instances in which there is either substantial authority for an unreported item or an item for which the facts were adequately disclosed and there was a reasonable basis for the understatement.

[12] I.R.C. § 6664(c)(1) provides that the accuracy-related penalty will not apply if the taxpayer had reasonable cause and attempted to comply with the tax laws and acted in good faith.

[13] I.R.C. § 6662(a).

[14] GAO (formerly known as the General Accounting Office), GAO/AIMD-00-21.3.1, Standards for Internal Control in the Federal Government (Nov. 1999).

[15] The discrepancy categories include, but are not limited to:  securities sales, wages, taxable pensions, nonemployee compensation, and unemployment compensation.

[16] Internal Revenue Manual (IRM) 4.19.3.16.6(1) (Sept. 30, 2014).

[17] This amount represents accuracy-related penalties for negligence and substantial understatement of tax liabilities and was provided by the AUR Program.  The Treasury Inspector General for Tax Administration (TIGTA) did not perform any audit tests to assess the validity of these data.  Due to IRS system limitations, neither the IRS nor TIGTA know with any degree of certainty what percentage of the $708 million was due to negligence or substantial understatement of tax liabilities.   

[18] Our statistical sample was a simple random sample, which is a probability sample (i.e., statistical sample) in which each sampled item of a given population has an equal chance of being selected.  The sample results can be projected to the population.

[19] According to IRM 20.1.1.3.2(1), reasonable cause penalty relief is generally granted when the taxpayer has exercised ordinary business care and prudence in determining their tax obligations but nevertheless failed to comply with those obligations. 

[20] See Appendices I and IV for additional details regarding the methodology for the sample selection and projections.

[21] See Appendix IV.  The five-year forecast for the potential increased revenue is based on multiplying the base year result by five and assumes, among other considerations, that economic conditions and tax laws do not change. 

[22] IRM 20.1.1.3.2 (Nov. 25, 2011).

[23] IRM 4.19.3.16.6 (Sept. 30, 2014).

[24] TIGTA, Ref. No. 2010-30-059, Accuracy-Related Penalties Are Seldom Considered Properly During Correspondence Audits (June 2010) and TIGTA, Ref. No. 2005-30-123, The Strategy to Reemphasize Penalties in Corporate Examinations Could Be Enhanced (Aug. 2005).  Although correspondence and corporate audits are significantly different from the cases worked by the AUR Program, TIGTA believes the findings regarding the inconsistent use of penalties are relevant.

[25] IRM 20.1.5.1.6 (01-24-2012), Managerial Approval of Penalties.

[26] In the AUR Program, managerial approval is only required when an examiner proceeds with assessing a penalty after a taxpayer requests that it be removed.

[27] See Appendix V for an example of an administrative Reasonable Cause Lead Sheet used by the IRS’s Examination function.

[28] In addition to our statistical sample of 51 cases, we also performed aggregate-level data analyses and identified additional instances of missing accuracy-related penalties for substantial understatement of tax liabilities.  ***1** ********************************************1************************************************
********************************************1************************************************
************************1********************** 

[29] The 16,069 AUR cases represents the total population of cases for TY 2010 that repeated the same issue category in at least two of the last three tax years, resulted in an underpayment between $501 and $5,000, and involved a taxpayer receiving a CP 2000 Notice.  This population is a subset of the entire “repeater” population since the IRS considers a taxpayer a “repeater” if the taxpayer repeated the same issue category in at least one of the last three tax years.

[30] Generally, the AUR Program completes the majority of its review of a given tax year’s selected inventory within approximately two years from when the tax return is due, which is typically April 15.  For example, for Tax Year 2010, the returns were due by April 15, 2011.  Therefore, the AUR Program did not complete most of its review of identified discrepancies until approximately September 30, 2013.  Therefore, at the time of our review, Tax Year 2010 was the most complete tax year available. 

[31] We determined that these 7,715 cases resulted in $13.3 million in total underpayments and, therefore, approximately $2.66 million in accuracy-related penalties for negligence should have been assessed.  For the purpose of this estimate, we assumed that the underpayment is equal to the understatement of tax.  The $2.66 million was calculated by multiplying the $13.3 million in total underpayments by the 20 percent negligence penalty.  For this estimate, we also assumed that none of the 7,715 cases met the criteria for the accuracy-related penalty for substantial understatement of tax liabilities as both penalties cannot be used on the same underpayment.

[32] We analyzed a total of 1,476,532 closed AUR cases, currently on the Individual Master File, with unabated assessed accuracy-related penalties as of Cycle 201352.  These cases represent Tax Years 1984 to 2011. 

[33] The $2.4 million potential overassessments involved 3,880 of the 7,393 cases, and the potential overassessments ranged from **1**  to ***1***.  The tax years for these cases ranged from Tax Years 1984 to 2011.  As of the end of our review, AUR Program officials confirmed six overassessments that totaled $768,096.

[34] The $1.1 million potential underassessments involved 3,513 of the 7,393 cases and the potential underassessments ranged from **1** to ***1***.  As of the end of our review, AUR Program officials confirmed three underassessments that totaled $108,654.

[35] I.R.C. § 6662(a) and 6662(b)(1).

[36] Treas. Reg. § 1.6662-3(b)1 and 1.6662-3(b)(1)(i).

[37] IRM 20.1.1.2 (Feb. 22, 2008) constitutes the IRS’s policy statement for assessment of penalties and provides that penalties promote voluntary compliance by demonstrating the fairness of the system to compliant taxpayers and increasing the cost of noncompliance.

[38] IRS Chief Counsel Memorandum, POSTN-146385-07, Accuracy-Related Penalties and the Automated Underreporter Program (Oct. 22, 2007).

[39] Except for those under the current IRS threshold. 

[40] This is important to note since both penalties cannot be used on the same underpayment.

[41] IRM 20.1.1.2.2 (Nov. 25, 2011).

[42] IRS officials informed us that this policy has been in place for at least 27 years.

[43] This estimate is based on the results of Tax Year 2010 as it is the most recent and complete tax year available (i.e., the AUR Program completed the overwhelming majority of Tax Year 2010 cases as of the end of Calendar Year 2013).  The estimate includes the number of AUR cases with underpayments ranging from $501 to $5,000 that were not assessed accuracy-related penalties.  Our estimate assumes that the overwhelming majority of these cases did not have an accuracy-related penalty included on the summary of proposed changes on the CP 2000 Notice as they likely were not deemed a “repeater” or subject to the substantial understatement penalty.  For the purposes of this estimate, we assumed that the underpayment is equal to the understatement of tax. 

[44] TIGTA, Ref. No. 2008-40-180, Most Automated Underreporter Program Notices Are Correct; However, Additional Oversight Is Needed (Sept. 2008).

[45] See GAO (formerly known as the General Accounting Office), GAO/AIMD-10.1.15, Business Process Reengineering Assessment Guide (May 1997), and GAO-04-287, Tax Administration:  Planning for IRS’s Enforcement Process Changes Included Many Key Steps but Can Be Improved, (Jan. 2004).

[46] This review was limited to the AUR’s Individual Master File program and the discrepancies identified on individual tax returns, such as the Form 1040, U.S. Individual Income Tax Return.

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

[48] We originally requested 144 closed AUR cases in order to meet our sample criteria of 100 cases.  We received 109 AUR cases, of which 51 AUR cases met the criteria for the accuracy-related penalty for substantial understatement of tax liabilities, i.e., the understatement exceeded 10 percent of the tax required to be shown on the return and the tax was greater than or equal to $5,000.  The remaining 58 AUR cases did not meet the criteria for an accuracy-related penalty for substantial understatement of tax liabilities and, therefore, were not reviewed.

[49] A judgmental sample is a nonstatistical sample, the results of which cannot be used to project to the population.  Judgmental samples were used for this step, as well as Step IV.C, because we did not intend to project the results of these tests to the entire population.

[50] The five-year forecast for the potential for increased revenue is based on multiplying the base year result by five and assumes, among other considerations, that economic conditions and tax laws do not change.

[51] Based on the recommendation of TIGTA’s contracted statistician, we used the projection based upon the T distribution method.  The T distribution is a probability distribution that is used to estimate population parameters when the sample size is small and/or when the population variances are unknown.

[52] To correct for skewness, the penalty dollars were averaged based on all cases received.

[53] We requested that the IRS provide us with the dataset of AUR cases that the IRS had identified as “repeaters.”  Due to data limitations, the IRS stated that it could only provide a subset of this dataset that included the most egregious repeaters (i.e., those that repeated noncompliance in the same category at least two times in the prior three tax years).  Therefore, we reviewed the subset of the egregious data provided to us to determine if the accuracy‑related penalties for negligence were assessed according to the AUR Program’s current policy.  Given that this data were only a subset, the amount of potential revenue associated with the noncompliance with the AUR Program’s policy and procedures would be greater than the amount shown here; however, we have another outcome measure regarding the AUR Program’s use of the accuracy-related penalty for negligence that likely covers all the additional cases that were not included in the subset provided to us.  

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

[55] We assume that the remaining 1,033 cases (8,748-7,715) had the negligence penalty included on the notice and was later waived by an AUR employee.

[56] The potential accuracy-related penalties for negligence ranged from $100 to $1,000.  For the purpose of this estimate, we assumed that the underpayment is equal to the understatement of tax.  We also assumed that none of the 7,715 cases met the criteria for the accuracy-related penalty for substantial understatement of tax liabilities as both penalties cannot be used on the same underpayment.

[57] We did not include cases with underpayments over $5,000 as those cases may have warranted the accuracy-related penalty for substantial understatement of tax liabilities (i.e., if the understatement exceeded 10 percent of the tax required to be shown on the return) and, thus, would not have been eligible for the accuracy-related penalty for negligence as both penalties cannot be used on the same underpayment.  Although the estimate for potential increased revenue would have been greater had we factored in cases that were above $5,000 but the understatement did not exceed 10 percent of the tax required to be shown on the return, we used the $5,000 underpayment threshold for simplicity.  

[58] There are instances in which the understatements may exceed or be less than the underpayment; however, due to data limitations, we assumed that these two are equal in order to complete this analysis.