The New Suspension of Interest Provision Is Not Always
Calculated Correctly
September 2002
Reference
Number: 2002-10-187
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.
September
27, 2002
MEMORANDUM FOR
COMMISSIONER, SMALL BUSINESS/SELF-EMPLOYED DIVISION
DEPUTY COMMISSIONER FOR
MODERNIZATION/CHIEF INFORMATION OFFICER
FROM: Michael R. Phillips /s/ Michael R.
Phillips
Acting Deputy Inspector
General for Audit
SUBJECT: Final Audit Report – The New Suspension
of Interest Provision Is Not Always Calculated Correctly (Audit # 200110045)
This
report presents the results of our review to determine if the Internal Revenue
Service (IRS) was properly suspending
and calculating interest in accordance with the Internal Revenue Code (I.R.C.)
§ 6404(g)(2001) as revised by the IRS Restructuring and Reform Act of 1998 (RRA
98).
In
summary, we found the IRS needs to prevent potential taxpayer rights
violations and decrease taxpayer burden by changing current computer programming
and improving internal controls over the application of the suspension of
interest provision. For example, we
identified approximately 29,000 taxpayers that were under assessed interest
totaling $1.6 million due to a discrepancy in the programming. The IRS changed the program in July 2001 to
address this situation. However, the
programming change sometimes allowed the under assessed interest to be
reestablished on taxpayers’ accounts and subject to collection action. We identified approximately 2,100 taxpayers
in a disaster relief situation where this occurred. For 12 of these taxpayers, the IRS had already collected $3,275
by automatically transferring refunds or overpayments from other tax years,
creating an additional burden for these taxpayers.
In
addition, the IRS computer program had not correctly suspended interest for
adjustments to the Earned Income Tax Credit (EITC) when disallowing all or part
of the EITC. The IRS has submitted a
request for information services to address this issue. However, the IRS still needs to improve
current computer programming to properly calculate the suspension of interest
notification period based on the proper number of days, weekends, and/or
holidays. There were approximately
4,600 taxpayers that were over assessed an estimated $536,000 in interest and
approximately 4,100 taxpayers that were under assessed an estimated $121,000 in
interest because the IRS did not timely or correctly input a suspension of
interest notification date code to its computer systems.
We
recommended the IRS identify all suspension of interest taxpayers with under
assessed interest due to the original programming discrepancy and zero out any
under assessed amounts, regardless of whether or not the interest had been
reestablished on the accounts, and refund any amounts that had already been
collected. We also recommended the IRS submit a request for information
services and evaluate another request for information services that had a later
operational date to address the number of days, weekends, and/or holidays when
calculating the suspension of interest notification period. In addition, the IRS should adjust the
interest assessed on applicable 1998 and 1999 taxpayer accounts with disallowed EITC
amounts. Finally, the IRS should
provide a systemic reminder about
suspension of interest when employees close audits and also include this
provision in its quality review program.
Management’s Response:
Management’s response to the draft report was due on September 19,
2002. As of September 20, 2002,
management had not responded to the draft report.
Copies of this report are
also being sent to the IRS managers who are affected by the report
recommendations. Please contact me
at (202) 622-6510 if you have questions or Daniel R. Devlin, Assistant
Inspector General for Audit (Headquarters Operations and Exempt Organizations
Programs), at (202) 622-8500.
The Notification Date Is Not Always Correctly Input or Indicated on Taxpayer Accounts When Needed
Current Master File Computer Programming Incorrectly Calculates the Suspension of Interest
Appendix I – Detailed Objective, Scope, and Methodology
Appendix II – Major Contributors to This Report
Appendix III – Report Distribution List
Appendix IV – Outcome Measures
The Congress believes that interest and penalties can quickly increase a tax debt to the point that it becomes virtually impossible for the average taxpayer to satisfy the unpaid tax liabilities without affecting his or her lifestyle and/or financial goals. In order to avoid additional taxpayer burden, the Congress included a provision within the Internal Revenue Service (IRS) Restructuring and Reform Act of 1998 (RRA 98) to help limit taxpayer liabilities.
This RRA 98 provision updated the Internal Revenue Code (I.R.C.) § 6404(g)(2001) and requires the IRS to suspend the accumulation of further interest and penalties unless a taxpayer is notified of an additional tax assessment within 18 months following a timely filed tax return. A timely filed return is one received by the IRS on or before either the original due date or a valid extension of time to file.
The suspension of interest provision applies only for additional taxes proposed by the IRS’ Automated Underreporter Program (AUR), Examination function, or Appeals function. In addition, it applies only to individual income tax returns for tax years ending after July 22, 1998. Consequently, the first applicable tax returns subject to this suspension of interest would be for calendar year 1998. The suspension of interest does not apply to amended returns or adjustments initiated by taxpayers. It also does not apply to any tax due on the original return or to any return involving a fraud or criminal penalty.
The return filing date plus 18 months is known as the 18-month notification period. If the taxpayer is not properly notified of an additional tax assessment within this period, the IRS is required to suspend further interest on the additional tax assessment until after notification is given. The key date is when the taxpayer is notified of the additional assessment, not when the assessment is entered on the taxpayer’s account (which can be substantially later). The IRS can still charge up to 18 months of interest on a taxpayer’s account for the additional tax assessment but is restricted from charging further interest until after notification is given. Consequently, the suspension of interest provision may effectively reduce taxpayer burden by not penalizing taxpayers when IRS audits take longer than 18 months from the return filed date.
Our audit fieldwork was performed between July 2001 and July 2002. Interviews were conducted with IRS Small Business/Self-Employed (SB/SE) Division and Modernization, Information Technology, and Security (MITS) Services officials located in the New Carrollton, Ogden, and Denver IRS offices. The audit was conducted in accordance with Government Auditing Standards. Detailed information on our audit objective, scope, and methodology is presented in Appendix I. Major contributors to the report are listed in Appendix II.
IRS procedures state that, if a tax audit takes longer than 18 months to complete, employees should indicate on the IRS’ Master File the date they notified the taxpayer of the additional assessment. The notification date is recorded by means of a Master File transaction code. This date is essential because the Master File computer programming uses this date to determine if interest should be suspended and, if so, to conduct the calculations in accordance with the suspension of interest provision.
We reviewed two distinctive populations to determine if Tax Year (TY) 1998 individual taxpayer accounts with additional tax assessments entered on the Master File between January 1 and June 30, 2001, had the notification date properly recorded on the Master File. Based on our review of these two populations, we estimate 10,506 of the 43,220 taxpayer accounts in the combined populations included notification date inaccuracies or omissions. Of the 10,506 accounts, 8,651 resulted in an over or an under assessment of interest. Below are details for each of the samples reviewed.
· For the first population of 7,955 taxpayer accounts, we reviewed case history files of a statistically valid stratified sample of 339 taxpayer accounts that did have the notification date transaction code information on the Master File. We determined that 53 taxpayer accounts (16 percent) had incorrect notification dates recorded on the Master File, resulting in 16 taxpayers being under assessed interest of $3,120 and 37 taxpayers being over assessed interest of $5,536. From this review, we estimate that 1,266 taxpayer accounts did not have the correct notification date on the Master File. As a result, approximately 896 taxpayers were over assessed interest of $35,668 and 370 taxpayers were under assessed interest of $17,356.
· For the second population of 35,265 taxpayer accounts, we reviewed case history files of a statistically valid stratified sample of 342 taxpayer accounts that did not have the notification date transaction codes on the Master File. We determined that 101 taxpayer accounts (30 percent) should have had a notification date on the Master File, resulting in 41 taxpayers being under assessed interest of $22,965, 40 taxpayers being over assessed interest of $19,746, and 20 taxpayers having no effect on interest. From this review, we estimate that 9,239 taxpayer accounts did not contain a notification date on the Master File. As a result, 3,663 taxpayers were over assessed interest of $500,612, another 3,721 taxpayers were under assessed interest of $103,229, and 1,855 taxpayers had no effect on interest.
In the two combined populations reviewed, we estimate that 16,001 taxpayers should have had the notification date entered on their accounts. However, we estimate that only 6,762 accounts (42 percent) were correctly identified with the notification date. As stated earlier, without accurate notification date information, the Master File computer programming would not be able to recognize that the additional assessment qualified for the suspension of interest and would not correctly suspend interest as required by law.
Until the notification dates are correctly entered on the Master File, the IRS is at risk of over assessing interest on taxpayer accounts per the I.R.C. § 6404(g), which could result in potential taxpayer rights violations. In addition, under assessment of interest on taxpayers’ accounts may result in loss of revenue for the government.
The primary reasons for the incorrect or missing notification dates on the Master File included some IRS employees not receiving appropriate guidelines for the suspension of interest provision and some employees not recognizing or properly recording the notification transaction code on the Master File. In addition to the primary reasons, IRS management agreed that the following were contributing factors.
· National procedures for implementing the suspension of interest provision for tax audits were not adequately or timely distributed to the field offices.
· Employee training on the RRA 98 provisions did not adequately cover the suspension of interest provision for tax audits.
The IRS is in the process of addressing these deficiencies. In July 2001, national procedures were revised for the suspension of interest provision in the IRS’ Internal Revenue Manual (IRM). However, the procedures were not formally published until February 2002. In addition, the SB/SE Division is currently reviewing their lines of communication and they plan to include training on the suspension of interest issue in their next national continuing professional education interactive training video scheduled for July 2002.
Although we agree these actions will be beneficial, we believe the IRS should also provide some type of systemic reminder when closing cases to ensure taxpayer rights are protected. Currently, when tax audits are completed there is no systemic reminder for employees or managers to determine if suspension of interest applies. In addition, there is no quality review feedback specifically addressing suspension of interest. The SB/SE Division uses the Examination Quality Review Process system to track its quality reviews of closed tax audits but the review process does not cover suspension of interest.
1. The SB/SE Division should coordinate with MITS Services to create a systemic reminder when tax audits are completed that would prompt employees and managers to consider whether the suspension of interest provision applies. IRS management should consider this for two possible computer systems, the Report Generating System and the Examination Reporting Control System.
Management’s Response: Management’s response to the draft report was due on September 19, 2002. As of September 20, 2002, management had not responded to the draft report.
2. The SB/SE Division should incorporate review requirements for the suspension of interest provision in the Examination Quality Review Process system.
The original Master File computer programming for the suspension of interest provision was inappropriately based on the assumption that if there was no notification date recorded on the Master File at the time the 18-month notification period expired, the taxpayer had not been advised of the additional assessment and his or her account qualified for a suspension of interest. This assumption resulted in computer programming that automatically suspended interest on a number of taxpayer accounts that did not qualify for suspension of interest, resulting in an under assessment of interest.
We reviewed a statistically valid stratified sample of 842 accounts from a population of 87,595 taxpayers that had additional tax assessments entered on the Master File between January 1 and June 30, 2001. We determined that 251 taxpayer accounts (30 percent) were under assessed $160,479 in interest because of the original Master File computer programming discrepancy.
The Master File programming was changed in July 2001 to consider the suspension of interest provision only when the notification date information is recorded on taxpayer accounts. In addition, for those TY 1998 taxpayer accounts with under assessed interest, the revised computer programming will reestablish the under assessed interest when the account is updated with additional activity, such as additional assessments or payments.
However, the IRS did not anticipate some types of additional taxpayer activity, including disaster relief, when reprogramming the Master File in July 2001. We identified situations where taxpayers were under assessed interest because of the original Master File programming discrepancy, full paid their account balance before July 2001, and received disaster relief after July 2001. When the disaster code was placed on the taxpayers’ accounts, it triggered the Master File to recalculate interest on the additional assessment and to add the under assessed interest to the account.
In these situations, we found no indication that the IRS had informed the taxpayers that they now owed additional interest on their previously full-paid accounts. In some instances, the IRS automatically applied refunds or overpayments from other tax years to collect the additional interest and there was no indication the taxpayer was notified. This results in increased taxpayer burden in two ways. First, taxpayers may not have known that interest was recalculated and assessed on their accounts. Second, taxpayers who are facing difficult situations because of a disaster may not be informed until after the fact that the IRS has applied refunds or overpayments to pay off the additional interest.
Upon further review of the 251 taxpayers in our sample with under assessed interest, we identified 28 disaster relief taxpayers with reestablished interest totaling $31,228. Additional research of the 28 taxpayers determined that 12 taxpayers had $3,275 collected by the IRS through automatic transfers of available overpayments or refunds from other tax years to these accounts.
Based on our sample review, we estimate the IRS under assessed approximately 28,728 taxpayers $1.6 million in interest and we estimate that at least 2,123 of these taxpayers qualified for disaster relief at the time of our fieldwork. For the estimated 2,123 disaster relief taxpayers, approximately $324,842 of interest was reestablished on their tax accounts. Due to the size of our sample results, we cannot provide a statistically valid estimate of the number and dollar amount the IRS has already collected from these estimated 2,123 disaster relief taxpayers.
Because it is difficult to predict when or where a disaster will occur, the remaining 26,605 taxpayers could be affected by a disaster in the future and receive disaster relief. This would trigger the recalculation of interest and have under assessed interest reestablished on their accounts. In addition, the reestablished interest could be potentially collected from taxpayer overpayments or refunds at any time.
Discussions with IRS employees confirmed this situation. They indicated that other taxpayer types and situations, such as military personnel and combat zones, could also cause a recalculation of interest. Interest could be reestablished on taxpayers’ accounts and subsequently collected when the IRS automatically transfers available taxpayer overpayments or refunds to pay the reestablished interest.
3. The SB/SE Division should coordinate with MITS Services to identify all taxpayers with under assessed interest caused by the original Master File computer programming, regardless of whether or not the interest had been reestablished on the accounts, and zero out any under assessed amounts to prevent burden on future disaster relief taxpayers or other taxpayers in special military conditions.
4. The SB/SE Division should coordinate with MITS Services to identify disaster relief taxpayers and refund the amounts that were automatically collected on the under assessed interest.
We found that current Master File computer programming incorrectly calculates the suspension of interest based on three separate types of computer programming miscalculations. We identified this issue from our statistically valid stratified sample of 939 taxpayer accounts with additional tax assessments entered on the Master File between January 1 and June 30, 2001, that had a notification date recorded on the Master File.
One additional day of interest is being charged to taxpayers before suspension
Per the I.R.C. §6404(g), if the suspension of interest provision applies to an additional assessment, interest should be suspended beginning on the day following the end of the 18-month notification period. Current Master File computer programming incorrectly suspends interest on the second day following the end of the 18-month notification period. This inappropriately overcharges taxpayers 1 day of interest.
This Master File computer programming discrepancy affected very few TY 1998 taxpayers due to a different programming discrepancy involving weekends and holidays (discussed in the next section). We identified only 2 taxpayer accounts that were adversely affected by this 1-day computer programming discrepancy for TY 1998. However, for TY 1999 and other future tax years, this programming discrepancy could involve many taxpayers.
Our review identified 7.4 million taxpayers who filed TY 1998 returns that later had additional assessments on their accounts. The majority of those taxpayers, 6.5 million (88 percent), filed timely on or before April 15, 1999. For these taxpayers, the 18-month notification period ended on a Sunday. If the suspension of interest provision applies, these taxpayers would receive the correct amount of interest because the 1-day discrepancy would overcharge 1 day of interest but the weekend/holiday discrepancy (discussed in the next section) would undercharge by 1 day of interest.
However, this will not be true for future years. For example, if the majority of the TY 1999 returns were timely filed on or before April 15, 2000, the 18-month notification period will end on a Monday and the weekend/holiday computer programming discrepancy will not be involved. The 1-day discrepancy may involve many taxpayers if the suspension of interest provision applies. Although the dollar impact to each taxpayer may be minimal, the IRS will be at risk for potentially violating taxpayer rights per I.R.C. §6404(g) by not correctly suspending interest.
When we discussed this difference with IRS employees, they agreed that current Master File computer programming miscalculated the interest by incorrectly adding 1 day to the 18-month notification period. They attributed this to complexity of the tax law, resulting in miscommunication between the analysts and computer programmers.
The IRS was unable to provide a proposed or existing SB/SE Division request for information services that would change the programming discrepancy for this 1-day issue. As a result, the IRS remains at risk of calculating interest incorrectly according to I.R.C. §6404(g) and potentially violating taxpayer rights.
Weekends
and legal holidays are not being considered when determining the end of the
notification period
If the end of the 18-month notification period falls on a weekend or legal holiday, I.R.C. § 7503 (2001) states that the action (i.e., the notification period) continues until the next day that is not a weekend or legal holiday. Below are examples of the 18-month notification period when taking into account the weekend/holiday provisions.
· For a TY 1998 return filed by April 15, 1999, the computer calculates the 18-month notification period to end on October 15, 2000. However, because this is a Sunday, the notification period should be extended to October 16, 2000.
· For a TY 1998 return filed by June 25, 1999, with a valid extension of time to file, the computer calculates the 18-month notification period to end on Monday, December 25, 2000. Because this is a legal holiday (Christmas), the notification period should be extended to Tuesday, December 26, 2000.
The Master File computer program for the suspension of interest provision does not take into consideration the weekend/holiday provision. When the weekend/holiday computer programming discrepancy is combined with the 1-day computer programming discrepancy previously mentioned, the net taxpayer impact is eliminated for most TY 1998 returns. However, we identified approximately 1,598 taxpayer accounts for TY 1998 that incorrectly calculated the notification period because the weekend/holiday provision of the law was not considered. This resulted in a premature ending of the notification period. As a result, although the IRS notified these taxpayers timely, the Master File interpreted the notification date as untimely and suspended interest. We estimate these taxpayers were undercharged interest of approximately $17,947.
When we discussed this difference with IRS employees, they agreed that current Master File computer programming miscalculates the interest by not considering weekends or legal holidays for the end of the 18-month notification period. They attributed this to complexity of the tax law resulting in miscommunication between the analysts and computer programmers.
The IRS included this Master File programming change in a draft SB/SE Division request for information services. However, the expected request submission date is July 2002 and the anticipated operational date is January 2004. Until this Master File programming change is operational, the IRS remains at risk of miscalculating interest per the I.R.C. §7503, which could cause loss of revenue to the government.
Low-income taxpayers may be overcharged interest when
the IRS disallows all or part of the Earned Income Tax Credit amount
The Earned Income Tax Credit (EITC) is a refundable Federal income tax credit for low-income taxpayers. The Congress originally approved the tax credit legislation in 1975, in part to offset the burden of social security taxes and to provide an incentive to work. Taxpayers applying for the EITC must meet certain requirements or the IRS may reduce or disallow all or part of the EITC amount. This adjustment or disallowance of EITC causes additional tax to be assessed, which should be subject to the 18-month notification period for applying suspension of interest.
From a statistically valid stratified sample of 939 taxpayers filing 1998 returns, we identified 28 taxpayer accounts (3 percent) that had a full or partial EITC disallowance. Although suspension of interest should have applied for the EITC portion of the assessment on all 28 taxpayer accounts, none of the taxpayers received this benefit. As a result, these taxpayers were overcharged $786 in interest.
By not suspending interest when applicable on the disallowed EITC amounts, we estimate 1,997 taxpayers in the population were overcharged nearly $25,717 in interest. This could result in potential taxpayer rights violations.
This condition occurred because the request for the Master File computer program did not specify that the additional tax assessment from an EITC disallowance was eligible for the suspension of interest provision. The IRS was already taking steps to correct this problem by the time we identified it in our case reviews. In August 2001, the SB/SE Division submitted a request for this programming to be changed. However, this change is not yet operational. In addition, IRS management informed us that this programming change will apply to those future taxpayer accounts with disallowed EITC amounts and will not retroactively adjust current and prior EITC disallowances that qualify for suspension of interest. This leaves the IRS at risk of miscalculating interest per the I.R.C., which could cause potential taxpayer rights violations.
5. The SB/SE Division should submit to MITS Services a request for information services that will change the Master File computer programming for this 1-day discrepancy when calculating suspension of interest.
6. The SB/SE Division should coordinate with MITS Services and determine the feasibility of an earlier anticipated operational date of this Master File computer programming modification involving weekend and holidays when calculating suspension of interest.
7. The SB/SE Division should coordinate with the MITS Services to identify taxpayers’ accounts for TYs 1998 and 1999 with disallowed EITC amounts that qualified for suspension of interest and reduce the amount of over assessed interest on these accounts.
Appendix I
Detailed Objective, Scope, and Methodology
The overall objective of this audit was to determine if the Internal Revenue Service (IRS) is properly suspending and calculating interest in accordance with Internal Revenue Code (I.R.C.) § 6404(g) (2001). We conducted the following tests to accomplish this objective.
I. Determined if the IRS guidelines and procedures for suspension of interest were consistent with the requirements of the I.R.C. § 6404(g).
A. Researched national and local memoranda,
guidelines, and procedures.
B. Consulted with the Treasury Inspector
General for Tax Administration (TIGTA) Counsel.
C. Reviewed IRS Counsel opinions.
D. Interviewed national IRS employees.
II. Determined if any individual taxpayers were overcharged interest because the IRS did not properly suspend interest in accordance with I.R.C. § 6404(g).
III. Determined if the IRS is properly
suspending and calculating interest in accordance with I.R.C. § 6404(g) for
those taxpayer accounts identified with a suspension of interest.
A. Obtained a computer extract from the Individual Master File (IMF), of all taxpayer account modules with filed 1998 returns. The IMF extract contained 7,429,725 records and had taxpayer account information beginning February 1, 1999, and ending August 25, 2001.
B. Separated the Automated Underreporter Program (AUR) taxpayer accounts from the Examination (Field, Office, and Service Center) taxpayer accounts and established an AUR database and an Examination database with two sub-populations: 1) those accounts containing codes indicating that I.R.C. § 6404(g) suspension of interest has been considered (those having a Transaction Code 971 with an action code of 64 (TC 971-64)) and 2) those accounts that did not indicate that I.R.C. § 6404(g) suspension of interest has been considered (no TC 971-64).
C. Selected statistically valid stratified samples using a variable sampling technique recommended by an expert statistician. See Appendix IV for details of the sample selection and results.
D. Analyzed the taxpayer accounts and stratified by 3 levels of the amounts of interest assessed. For the largest strata, we selected at least 250 taxpayer accounts from the AUR database (interest under $500) and at least 250 taxpayer accounts from the Examination database (interest $1,000 and under). For the second strata, we selected at least 100 taxpayer accounts from the AUR database (interest between $500-$10,000) and at least 100 taxpayer accounts from the Examination database (interest between $1,001 and $20,000). For the third strata, we selected 50 taxpayer accounts from the AUR database and 50 taxpayer accounts from the Examination database with the highest interest assessed.
E. Ordered and evaluated closed tax audit files and taxpayer account transcripts for the samples to determine any over or under assessments of interest.
F. For both AUR and Examination taxpayer accounts, we projected the number of taxpayers and total amount of interest that the IRS over or under assessed taxpayers in the applicable population of Tax Year 1998 accounts.
G. Interviewed IRS Small Business/Self-Employed Division and Modernization, Information Technology, and Security Services personnel to obtain agreement and explanation of possible causes for inaccurate suspension of interest.
H. Evaluated and determined if current controls and ongoing corrective actions were adequate to properly suspend interest to prevent misapplication or to change those taxpayer accounts in future case processing in accordance with I.R.C. § 6404(g).
Appendix
II
Major Contributors to This Report
Daniel R. Devlin, Assistant Inspector General for Audit
(Headquarters Operations and Exempt Organizations Programs)
Nancy A. Nakamura, Director
Aaron R. Foote, Audit Manager
Theresa M. Berube, Senior Auditor
Mark A. Judson, Senior Auditor
Daniel M. Quinn, Senior Auditor
Janice M. Pryor, Auditor
Appendix III
Commissioner N:C
Deputy Commissioner N:DC
Chief, Information Technology Services M:I
Commissioner, Wage and Investment Division W
Director, Compliance, Small Business/Self-Employed Division S:C
Deputy Director,
Compliance Policy, Small Business/Self-Employed Division S:C:CP
Chief Counsel CC
National Taxpayer Advocate TA
Director, Legislative Affairs CL:LA
Director,
Office of Program Evaluation and Risk Analysis
N:ADC:R:O
Office of Management Controls N:CFO:F:M
Audit Liaisons:
Commissioner, Small Business/Self-Employed Division S
Commissioner, Wage and Investment Division W
Deputy Commissioner for Modernization/Chief Information Officer M
Appendix IV
This appendix presents detailed information on the measurable impact that our recommended corrective actions will have on tax administration. These benefits will be incorporated into our Semiannual Report to the Congress.
General Methodology Used to Measure the Reported Benefit:
We further evaluated the extract to identify 251,129 taxpayers who had automated assessments from the Automated Underreporter Program (AUR) and 43,220 taxpayers with assessments through examinations (Field or Office Audit or Service Center Correspondence) and met the following criteria: 1) timely filed a TY 1998 return, 2) had an additional assessment input to their accounts between January 1 and June 30, 2001. We then separated the AUR and Examination taxpayers into databases of those accounts with a suspension of interest provision notification date (TC 971-64) and those accounts without the notification date.
For the taxpayer accounts selected in our statistically valid stratified samples, we ordered tax returns (with the related case files) and Master File transcripts for review. Additionally, we conducted on-line research when needed to determine the most current taxpayer account information available.
Type and Value of Outcome Measure:
· Taxpayer Rights – Potential; 4,559 taxpayer accounts were over assessed approximately $536,280 in interest because the IRS did not correctly input a suspension of interest notification date to the Master File; (see page 2).
· Increased Revenue – Potential; 4,091 taxpayer accounts were under assessed approximately $120,585 in interest because the IRS input a suspension of interest notification date to the Master File, but it was untimely or incorrect; (see page 2).
Specific Methodology Used to Measure the Reported Benefit:
For the Notification issue, we selected two statistically valid stratified samples. We employed an attribute and variable sampling methodology suggested by an expert statistician. The separate samples were stratified by interest assessed into low, medium, and high strata. One statistically valid stratified sample contained 339 taxpayers from 7,955 Examination accounts with a notification date (TC 971-64) input on to the Master File and the other statistically valid stratified sample included 342 taxpayers from 35,265 Examination accounts where no notification date (TC 971-64) was input on the Master File. Of the 681 taxpayers (339 + 342) reviewed, we identified 53 taxpayers that had incorrect notification dates recorded on the Master File, resulting in 16 taxpayers being under assessed interest of $3,120 and 37 taxpayers being over assessed interest of $5,536. In addition, we identified another 101 taxpayers without any notification date recorded on the Master File, resulting in 41 taxpayers being under assessed interest of $22,965, 40 taxpayers being over assessed interest of $19,746, and 20 taxpayers having no effect on interest. We projected these results to the populations of 7,955 Examination accounts with a notification date and 35,265 Examination accounts without a notification date on the Master File.
· We estimate the IRS over assessed approximately 4,559 taxpayers $536,280 in interest.
1. Based on a 95 percent confidence interval, we estimate 3,663 taxpayers (plus or minus 3.4 percent precision) were over assessed $500,612 (plus or minus 1.2 percent precision) in interest.
2. Based on a 95 percent confidence interval, we estimate 896 taxpayers (plus or minus 3.5 percent precision) were over assessed $35,668 (plus or minus 0.8 percent precision) in interest.
· We estimate the IRS under assessed approximately 4,091 taxpayers $120,585 in interest.
1. Based on a 95 percent confidence interval, we estimate 3,721 taxpayers (plus or minus 3.5 percent precision) were under assessed $103,229 (plus or minus 0.2 percent precision) in interest.
2. Based on a 95 percent confidence interval, we estimate 370 taxpayers (plus or minus 2.3 percent precision) were under assessed $17,356 (plus or minus 0.3 percent precision) in interest.
Type and Value of Outcome Measure:
· Taxpayer Burden – Potential; 28,728 taxpayer accounts were under assessed approximately $1.6 million and could be subject to having this under assessed interest reestablished on their accounts because of a Master File programming inaccuracy. In addition, we estimate that 2,123 of these taxpayers currently qualify for disaster relief and have had $324,842 in under assessed interest reestablished on their accounts because of the revised Master File programming; (see page 6).
Specific Methodology Used to Measure the Reported Benefit:
For the Understated interest issue, we selected two statistically valid stratified samples. We employed an attribute and variable sampling methodology suggested by an expert statistician. The separate samples were stratified by interest assessed into low, medium, and high strata. One statistically valid stratified sample contained 500 taxpayers from 52,330 AUR accounts and the other statistically valid stratified sample included 342 taxpayers from 35,265 Examination accounts. None of these accounts contained a notification date (TC 971-64) on the Master File. Of the 842 taxpayers (500 + 342) reviewed, we identified 251 taxpayers under assessed $160,479 in interest because of the original Master File computer programming discrepancy. Upon further review of the 251 taxpayers, we identified that 28 disaster relief taxpayers had $31,228 in interest reestablished on their accounts. We projected these results to the populations of 35,265 Examination accounts and 52,330 AUR accounts without a notification date on the Master File.
· We estimate the IRS under assessed approximately 28,728 taxpayers $1.6 million in interest.
1. For the Examination sample, based on a 95 percent confidence interval, we estimate 9,548 taxpayers (plus or minus 5.1 percent precision) were under assessed $722,044 (plus or minus 1.1 percent precision) in interest.
2. For the AUR sample, based on a 95 percent confidence interval, we estimate 19,180 taxpayers (plus or minus 4.8 percent precision) were under assessed $927,420 (plus or minus 1.0 percent precision) in interest.
· We estimate the IRS under assessed 2,123 disaster relief taxpayers approximately $324,842 in interest and reestablished this interest on their accounts.
1. For the Examination sample, based on a 95 percent confidence interval, we estimate 427 taxpayers (plus or minus 1.1 percent precision) were under assessed $124,848 (plus or minus 0.6 percent precision) in interest.
2. For the AUR sample, based on a 95 percent confidence interval, we estimate 1,696 taxpayers (plus or minus 1.7 percent precision) were under assessed $199,994 (plus or minus 0.6 percent precision) in interest.
Type and Value of Outcome Measure:
· Increased Revenue – Potential; 1,598 taxpayers were under assessed interest of approximately $17,947 because the Master File programming was not designed to consider the I.R.C. § 7503 (2001) provision for weekends and holidays; (see page 8).
Specific Methodology Used to Measure the Reported Benefit:
For the weekend/holiday issue, we computer analyzed 198,838 AUR taxpayer records identified with a suspension of interest notification date (TC 971-64) on the Master File. We identified taxpayer accounts where the 18-month notification period ended on October 15, 2000, which is a Sunday. The notification period on these accounts should have been extended to Monday, October 16, 2000, per I.R.C. §7503. We then identified 1,598 taxpayer accounts that had interest incorrectly suspended because the notification date was October 16, 2000, which was a timely notification. We determined that 1,598 taxpayers were under assessed approximately $17,947 in interest because the Master File programming did not adjust for weekends and/or holidays when determining the end of the notification period.
Type and Value of Outcome Measure:
· Taxpayer Rights – Potential; 1,997 taxpayers were over assessed interest of approximately $25,717 because the Master File programming did not properly include earned income tax credit (EITC) amounts in the suspension of interest calculations; (see page 8).
Specific Methodology Used to Measure the Reported Benefit:
For the EITC issue, we selected two statistically valid stratified samples. We employed an attribute and variable sampling methodology suggested by an expert statistician. The separate samples were stratified by interest assessed into low, medium, and high strata. One statistically valid stratified sample contained 600 taxpayers from 198,799 AUR accounts and the other statistically valid stratified sample included 339 taxpayers from 7,955 Examination accounts. All contained a notification date (TC 971-64) on the Master File. Of the 939 taxpayers (600 + 339) reviewed, we identified 28 accounts where the EITC was not included in the suspension of interest calculation. As a result, these taxpayers were overcharged $786 in interest. We projected these results to the populations of 198,799 AUR accounts and 7,955 Examination accounts with a notification date on the Master File.
· We estimate 1,997 taxpayers were overcharged approximately $25,717 in interest because the Master File programming did not include the EITC portion of the additional assessment in the suspension of interest calculation.
1. For the Examination sample, based on a 95 percent confidence interval, we estimate 605 taxpayers (plus or minus 3.0 percent precision) were over assessed $20,055 (plus or minus 0.3 percent precision) in interest.
2. For the AUR sample, based on a 95 percent confidence interval, we estimate 1,392 taxpayers (plus or minus 0.7 percent precision) were over assessed $5,662 (plus or minus 0.1 percent precision) in interest.