March 28,2001

IRS Says Grassley is Right About Taxpayer Review, Audit Rates

WASHINGTON – Sen. Chuck Grassley, chairman of the Committee on Finance, hasreceived an acknowledgment from the Internal Revenue Service that the agency does indeed subjectmillions more taxpayers to review than it has maintained. The text of the letter from IRSCommissioner Charles Rossotti, Grassley’s response and the enclosures to the Rossotti letter follow.

March 26, 2001
The Honorable Charles E. Grassley
Chairman, Committee on Finance
United States Senate
Washington, D.C. 20510

Dear Mr. Chairman:

Thank you for your letter concerning how the IRS reports on audits and other forms ofverification of tax returns. I believe this is an important question and I will do my best to clarify theissues you raise as well as provide the specific data you request.

While the audit rate alone is not an adequate indicator of IRS’s compliance activities, auditsremain extremely important for ensuring the fairness of the tax system to the average taxpayer. Ourbudget proposal was carefully designed to meet this minimum goal.

I understand that the basic thrust of your letter is to question whether the "audit rate" aspublicly reported by the IRS understates the ability of the IRS to verify the accuracy of individualtax returns. My view is that your question is well founded. Simply focusing on the audit rate doessubstantially understate the IRS's capacity to find errors in returns, especially in certain kinds ofreturns. In my many press interviews in the past few years in which this topic has come up, I haveconsistently made this point, often citing our computer matching program as an example of atechnique that the IRS uses in addition to traditional audits. Recently I gave an interview to the NewYork Times on this subject, which resulted in a "quote of the day" (Friday, February 16, 2001) asfollows:

With the use of document matching as well as other return verification techniques that willeventually be enabled by new technology, it is my view that there is no need to return to the levelsof individual audit coverage that existed even five years ago, which was three times the FY 2000level. The IRS strategic plan and budget proposals as presented to the IRS Oversight Board do notcall for this approach. However, our strategic plan sets forth an approach in the short run to stabilizeour level of traditional compliance activities, such as individual audits, at or slightly above currentlevels and to focus them on the areas where they are most required. In the long run, we will rely onour business systems modernization program to increase the effectiveness and efficiency of theseactivities.

As noted in your letter, the IRS has for many years relied on a range of techniques to verifycertain items on tax returns. Each of these techniques is appropriate for particular returns or typesof potential errors. Enclosure 1 provides a list of techniques used by the IRS for verifying tax returnsand the type of return/issue for which they are most useful. As you requested, we have also shownhistorical data on the use of each such technique (Enclosure 2).

With respect to Information Returns Processing, or document matching as it has often beencalled, this technique is very effective for verifying certain income items reported on individualreturns against that reported by third parties, including wages, interest, dividends and miscellaneouspayments. It can also be used to verify gross sales of assets, such as securities, but cannot be usedto verify the gain or loss on such sales since we have no third-party reporting on the cost basis ofassets. It is also of limited value in verifying some deductions, such as mortgage interest.

Document matching is not useful for verifying business income, gain or loss on asset sales,or most itemized deductions. We estimate that the total personal income that cannot be verified bydocument matching represented about $1.2 trillion in FY 1998, or 19.7% of total reported personalincome. An important role of audits is to verify these major categories of income and deductions.

The significance of verifying income and deduction items through audits is illustrated by thefact that the average in-person audit of an individual return results in an assessment of approximately$9,540, while the average assessment from a document matching case is $1,506. In FY 2000, theIRS closed 247,212 in-person audits of individual returns and assessed $2.4 billion from thisprogram; in the document matching program in FY 2000, the IRS closed 1,353,545 cases andassessed $2.1 billion.

One of my real concerns about the decline in audits is fairness to the majority of taxpayerswhose income is reported and can be readily verified. It is relatively easy for the IRS to verify thereturns and reported income of taxpayers whose income results from wages, interest and dividendsand who take the standard deduction, who comprise the majority of taxpayers. It is harder, and oftenrequires audits, to verify the income of taxpayers with other forms of income and deductions or morecomplex returns, who are often higher income taxpayers. The proportion of income that cannot beverified through document matching is 10% for taxpayers with income under $100,000, as comparedwith 35% for taxpayers over $100,000. Also, 91% of returns reporting income over $100,000itemize deductions, compared to 26% of those below $100,000, and most itemized deductionscannot be verified through document matching. To the extent that the IRS uses more and moredocument matching and less and less auditing, the effect may be perceived as, and will in fact beunfair because higher income taxpayers will not have their returns verified to the same degree asmiddle income taxpayers.

Another concern with respect to both tax revenues and fairness is the income reported bycorporations and partnerships. While the IRS continues to audit the 1,100 largest corporations everyyear, the audit rate for all other corporations has declined from 3.0% in 1992 to 1.1% today. Aparticular source of concern is the growing number of entities, such as partnerships, trusts and Scorporations,which pay no income tax at the business level but pass their net income on to theirshareholders or partners. As shown in Enclosure 3, in 2000, these "passthrough" entities filed 7.4million returns, reported $5.0 trillion of gross revenues and $680 billion of income. However, theIRS audited only 29,057 of these entities, or only 1 of every 256 returns - .39% (Enclosure 4). Wedo plan to begin a program to match income reported on K-1 forms from these entities to individualtax returns. However, this technique will not provide any verification of the income reported by thebusiness entity itself, which requires an audit.

Our strategic plan attempts to reconcile all these factors with the objective of increasing theIRS’ s ability to achieve our second strategic goal, which is service to all taxpayers through fair anduniform application of the tax law. If our modernization program is successful, we believe we cando this while continuing to shrink the of the IRS in relation to the economy.

With respect to the question of why document matching cases are not considered audits, thetechnical reason is that Section 7605(b) of the tax code generally limits the ability of the IRS torequire a taxpayer to submit books and records for inspection by the IRS more than once per return.Since document matching cases do not require the taxpayer to submit books and records to the IRS,a document matching case does not preclude a subsequent audit. Revenue Procedure 94-68specifically defines IRS taxpayer contacts, including document matching, which are not consideredaudits for the purpose of Section 7605(c). More generally, it is my understanding that some yearsago the IRS proposed to change the definition of an audit to permit inclusion of the documentmatching cases in the overall reported number of audits, and this proposal was criticized in someparts of Congress as an attempt to inflate IRS's statistics.

Notwithstanding these previous issues, all of these IRS statistics, including the number ofdocument matching cases, are publicly reported and it is our goal to make these reports asinformative and meaningful as possible. I would be pleased to discuss with you how we couldimprove this reporting.

Enclosure 5 provides data in answer to your specific questions.


Charles O. Rossotti


March 28, 2001
Mr. Charles Rossotti
Internal Revenue Service
Washington, D.C. 20224

Dear Mr. Commissioner:

I am writing in response to your letter of March 26, 2001, regarding audit rates and overallenforcement efforts at the Internal Revenue Service (IRS). I believe our exchange of letters has beenextremely useful in highlighting for the Committee, and more importantly, the American taxpayer,that the IRS is quite effective at reviewing and verifying most information on the vast majority oftax returns.

The General Accounting Office (GAO) confirmed for the Committee today that almost allindividual tax returns are subject to IRS computer review for third party information under theinformation reporting program (IRP) as well as checked for math errors. The and scope of thiscomputer review are reflected in your Enclosure 5, which shows that over 14 million tax returns in1998 (the most recent data available) were flagged for potential discrepant income or deductions.There were approximately 120 million returns filed in the previous year, thus well over 10% oftaxpayer returns were identified by IRS computers for possible further review.

As we approach the end of filing season, it is critical that the American taxpayer hears fromthe IRS that both they and their neighbors’ tax returns will be subject to these vigorous reviews.Thus, I appreciate your statement in your March 26th letter that focusing solely on the face-tofaceaudit rate “does substantially understate the IRS’s capacity to find errors in returns, especiallyin certain kinds of returns” and encourage your office to widely publicize the effectiveness of thecomputer matching program.

That the IRS is on the job is an important message the public must hear, particularly giventhe explosive growth we have recently seen in Internet tax schemes, scams and cons – the main topicof the April 5th Finance Committee hearing at which you will be testifying. Many web sitespromoting these tax scams cite the low audit rate as an inducement to buy their schemes.Your letter is also extremely useful in beginning the discussion of where we need to be interms of compliance efforts. I appreciate your recognition that:

With the use of document matching as well as other return verification techniquesthat will eventually be enabled by new technology, it is my view that there is noneed to return to the levels of individual audit coverage that existed even five years ago,which was three times the FY 2000 level.

This statement is very useful and begins to bring the discussion of audit coverage into betterfocus and, as important, dismisses histrionic claims that we must return to earlier high levels of faceto-face audit levels. Further, this will allow a more objective and rational review of the IRS’ budgetproposal.

The focus of my letter was on individual tax returns not pass-through entities. However, Iwelcome your discussing this important aspect of enforcement. I wish to work with you to ensurethat, at a minimum, greater resources within the IRS are made available to address these enforcementissues. I would encourage you to make a priority of having a computer program in place to matchincome reported on K-1 forms. It is my understanding that the State of California already has acomputer program in place that provides this very useful function. Would it be possible for the IRSto gain from California’s experience in this area? We are seeing an extensive use of partnerships andother pass-through entities in the area of tax schemes that will be discussed at the April hearing andit is important that would technology be brought to bear on this matter.

In addition, it is important that the IRS have updated data on taxpayer compliance so that itcan better target the audits it does perform. More effective and targeted auditing can go far inlimiting the number of audits that need to be performed. I believe it is possible to obtain suchtaxpayer compliance information without subjecting individual taxpayers to an accountant’s versionof root canal. The IRS has stated for years that it is developing an alternative program to measuretaxpayer compliance. I would ask that in your testimony on April 5th, you state when we cananticipate seeing results in this area.

I particularly urge you to proceed in obtaining taxpayer compliance data in the areas of trustsand partnerships. The data in the field of trusts and partnerships, I understand, is very old and giventax schemes like “pure trusts,” it is particularly imperative that obtaining new taxpayer compliancedata in these fields be a priority.

As the Committee prepares for the April 5th hearing, I will have additional comments aboutyour March 26th letter. For now I want to thank you for your candor in responding to my letter andagain encourage the IRS to widely broadcast the news that tax cheats should not rest easy – the IRSis on the job.

Cordially yours,

Charles E. Grassley

Enclosures to Commissioner Rossotti Letter
Math Error ?? An addition, subtraction, multiplication or division error on any return.?? Incorrect use or selection from tax tables, schedules, etc. provided by the IRS.?? Entry on a return inconsistent with an entry on a schedule, form, statement, or list filed with the return.?? An omission of information required on the return to substantiate an entry on the return.?? An entry on a return of a deduction or credit in an amount which exceeds a statutory monetary limit, a percentage, a ratio, or a fraction. The items that are considered in applying the limitation have to appear on the return.?? Missing or incorrect taxpayer identification numbers (TINs) for personal exemptions for the primary and secondary taxpayers, dependents, Child andDependent Care Credit, EITC, Child Tax Credit, Lifetime Learning Credit, or Hope Scholarship Credit.?? EITC is claimed based on self-employment income but self-employment tax has not been paid.?? Primary and secondary taxpayers do not meet the age requirements for claiming EITC when there is no qualifying child.?? EITC is claimed based on a qualifying child but the child’s date of birth is missing or incorrect.?? EITC is claimed and EITC claimed in a previous year was disallowed through an examination and Form 8862 (Earned Income Credit Eligibility) is not filedwith the return.?? Child Tax Credit or Child and Dependent Care Credit is claimed but the dependent does not meet the age criteria.

Information Returns/DocumentIncome items reported by a third-party payer are matched to the income items reported on the individual’s income tax return. The types of income reported by payers are listed below. Contact is made when there is a discrepancy (income reported on the tax return is less (underreported) than theincome reported by the payer).

Matching - Wages- Interest- Dividends- State and Local Income Tax Refunds- Rents and Royalties- Patronage Dividends- Crop Insurance Proceeds- Unemployment Compensation- Taxable Grants- Proceeds from Securities Sales- Bartering- Refund of Overpaid Mortgage Interest- Social Security/Railroad Retirement Benefits- Gambling Income- Cancellation of Debt- Proceeds from the Sale or Exchange of Real Estate- Nonemployee Compensation (Fees, commissions, or any other compensation paid by a business to an individual who is not an employee.)- Retirement Plans (Pensions, annuities, IRA distributions, lump-sum distributions, employee savings plans, stock bonus plans, profit-sharing plans, etc.)- Substitute Payments in Lieu of Dividends or Interest- Medical Payments (Compensation paid to doctors, dentists and others in the medical profession.)- Fishing Income (Payments to fishing boat crew members.)- Distributive Shares of Income from a partnership, Subchapter S corporation or estates or trusts.- Agricultural Subsidies/Commodity Credit Corporation Loans ForfeitedEnclosure 1: RETURN VERIFICATION TECHNIQUES, CONTINUEDTECHNIQUE TYPE OF RETURN/ISSUEMatching, cont. ?? Only one deduction (mortgage interest and points) is subject to reporting by third-party payers. Mortgage interest reported as paid by lenders is matchedto the mortgage interest deducted on the individual’s income tax return. Contact is made when there is a discrepancy.?? If a taxpayer has not filed a tax return after repeated letters requesting a return and information documents indicate income amounts would require areturn, the income amounts reported by third-party payers will be used to compute a tax liability for the taxpayer. This is in accordance with the authoritygranted the IRS under IRC 6020(b).Examination ofBooks/Records?? Income, deductions, credits, or other issues on partnership, corporation, Subchapter S corporation, and trust returns.?? Itemized deductions reported by an individual. These include medical and dental expenses, taxes, interest other than home mortgage interest and points,gifts to charity, casualty and theft losses, unreimbursed employee expenses, business use of home or car, educational expenses, etc.?? Income and expenses from a sole proprietorship (Schedule C) business.?? Basis and acquisition and sold dates for long-term or short-term capital assets.?? Rental and royalty expenses.?? Income and expenses from farming (Schedule F).?? Credits (EITC, Child Tax Credit, Child and Dependent Care Credit, Adoption Credit, Education Credits, etc.) and Dependency Exemptions. Anyrequirements other than TINs and date of birth that are required to claim the credit or the dependent as an exemption. Criteria such as, child/dependentmeets certain criteria, such as relationship, residency and citizenship.?? Foreign tax credit?? Foreign earned income exclusion?? Deductions such as IRA, student loan interest, moving expenses, self-employed health insurance, alimony paid, etc. to arrive at adjusted gross income.?? Household employment taxesEnclosure 2: TOTAL NUMBER OF RETURNS CLOSED IN FISCAL YEARFISCAL YEAR MATH ERRORS UNDERREPORTER(Document Matching)CORRESPONDENCE AUDIT IN-PERSON AUDIT1971 N/A N/A N/A 1,346,0001972 N/A N/A N/A 1,343,0001973 N/A N/A 482,432 1,409,0001974 N/A N/A 714,000 1,686,6861975 N/A N/A 1,329,305 1,838,5581976 N/A N/A 1,882,178 2,043,5951977 N/A N/A 913,460 1,742,0561978 N/A N/A 663,173 1,675,8521979 N/A N/A 696,341 1,645,0791980 N/A N/A 728,119 1,638,7851981 N/A 1,200,000 975,541 1,482,5861982 N/A 2,900,000 819,366 1,352,0831983 N/A 2,900,000 1,078,065 1,279,8101984 N/A 3,900,000 761,126 1,135,5331985 N/A 3,600,000 680,951 1,143,5171986 N/A 3,200,000 643,578 1,031,4221987 N/A 2,242,000 684,560 927,9641988 N/A 3,800,000 735,534(2) 885,1341989 N/A 3,650,000 599,348 785,6471990 N/A 2,950,000 425,641 719,3191991 N/A 4,840,000 612,547 700,6211992 4,985,000 3,771,509 458,727 661,7141993 4,088,000 2,723,830 301,160 727,2231994 4,059,000 2,645,075 358,829 775,1421995 6,102,000 2,711,086 1,147,296 687,4961996 4,751,000 1,930,326 1,154,562 648,7031997 5,984,000(1) 931,354 769,181 686,1671998 5,669,000 1,726,098 628,484 560,7281999 6,552,000 1,770,695 708,886 384,1782000 5,751,000 1,353,545 362,830 247,212Sources:Math Error: Individual Master File Control Reports, Math Error Master File extract, Workload Inventory Tracking System ReportUnder-reporter: Management Information System for Top Level Executives; Work Plan and Control ReportsCorrespondence and in-person audits: FY1992-FY2000 is ERIS database; FY1988-FY1991 is Commissioner's Databook; FY1971-1987 is the Commissioner's Annual ReportNotes:(1) Expanded Math Error authority was implemented in CY1997 for TY1996, data prior to FY1998 does not include EITC or TIN math errors.(2) Prior to FY 1988, service center data from the Commissioner's Annual Report included all types of returns corrected. However, the majority of the work performed in the service center programdealt with individual returns.Enclosure 3: PASSTHROUGH ENTITIES REPORTINGSOURCE NUMBER OF RETURNS GROSS RECEIPTS TOTAL INCOMETotal number of returns filed, includesreturns reporting positive, negative orzero quantitiesAs reported on of Form 1065-line 1c,Form 1120S-line 1c, Form 1041-line 9Income distributed to partners,shareholders and beneficiariesPartnerships1 2,045,000 $1.57 trillion $408 billionSubchapter-SCorporations2,928,000 $3.32 trillion $234 billionTrusts2 2,470,500 $.12 trillion $38 billionTotal 7,443,500 $5.01 trillion $680 billionSource: Preliminary analysis of TY1999 data, Processing Year 2000 Business Return Transaction fileNotes: (1) Of the 2 million partnership forms filed, 1.3 million claimed zero or negative receipts.(2) Does not include approximately 1 million grantor trusts, which have no passthrough income.Enclosure 4: TOTAL PASSTHROUGH FILINGS & COVERAGEEnclosure 5: RESPONSE TO DATA REQUESTSNOTE ON DATA AVAILABILITY: The IRS does not have accurate data on enforcement revenue collections prior to FY1992. At thattime, the Enforcement Revenue Information System (ERIS) became fully operational - for the first time linking the Service's variousenforcement databases (primarily the masterfiles and the Audit Information Management System - AIMS). The service does have somedata available prior to that time, however that data is not comparable in format and substance to the data we have submitted in thisdocument. For example: some of the functions' revenue totals were estimated prior to FY1992 often resulting in double countingbetween functional areas (e.g. revenue resulting from an Examination assessment that was actually collected in the Collection functionwas counted by both functions). We believe the data provided in this enclosure provides accurate, timely information and provides thebest insight into enforcement actions at the IRS today.Request 1. The number of IRP contacts for 2000 as well as for the last twenty years the number ofreturns subject to computer-based review that did not result in the sending of a letter to the taxpayer.AUR DEFINITIONSTERM DEFINITIONTotal AUR Inventory Total number of tax returns identified by the computer-matching criteria for potentialdiscrepant income/deduction(s)Screened Number of tax returns that are physically reviewed for potential discrepantincome/deduction(s)Screened Out Number of tax returns where the income was identified as reported somewhere else onthe return, an explanation included on the return explained the discrepancy, etc.TaxpayersContactedNumber of taxpayers who received a notice/letter requesting an explanation for thediscrepancyChange Cases Number of taxpayer's returns that were changed due to the identified discrepancyAUR HISTORY BY TAX YEARINVENTORY 1990 1991 1992 1993 1994 1995 1996 1997 1998Total AUR Inventory 11,328,231 9,111,462 11,086,181 11,549,341 12,237,481 11,562,667 11,873,979 13,350,638 14,121,015Screened 4,175,651 4,168,986 3,744,373 2,283,197 1,625,714 3,219,329 3,112,509 2,992,551 2,448,389Screened Out 1,548,308 1,480,455 1,152,865 927,071 713,095 956,074 918,670 1,193,047 1,019,477Taxpayers Contacted 2,627,343 2,688,531 2,591,508 1,356,126 912,619 2,263,255 2,193,839 1,799,504 1,428,912Change Cases 2,001,096 2,020,041 2,010,606 988,987 659,153 1,718,345 1,760,017 1,309,327 1,096,788Sources: Management Information System for Top Level Executives; Work Plan and Control ReportsRequest 2. The number of dollars assessed and collected through the IRP program for the last twenty years.TOTAL ENFORCEMENT DOLLARS ASSESSED TO DATEFY 1992 FY 1993 FY 1994 FY 1995 FY 1996 FY 1997 FY 1998 FY 1999 FY 2000 TotalDistrict Audits (Inperson)$4,267,628,325 $4,265,055,282 $5,357,315,037 $5,782,643,649 $5,260,263,537 $5,462,733,565 $4,665,971,725 $3,276,169,647 $2,358,427,054 $40,696,207,821Service Center Audits(Correspondence)$2,107,600,679 $1,613,464,390 $1,252,687,782 $2,358,658,954 $2,784,065,045 $3,561,231,369 $2,787,166,772 $1,969,693,717 $1,362,326,701 $19,796,895,409Underreporter(Document Matching)Data not available $1,006,629,694 $1,624,331,331 $1,449,165,285 $1,478,888,181 $1,698,438,255 $2,062,718,637 $2,037,937,452 $11,358,108,835Source: ERIS dataNotes: Results above are related to cases closed in Exam or IRP Underreporter in that fiscal yearAssessed dollars include tax, penalty, and interest. Some of the cases closed in these fiscal years are in the Appeals process and therefore have no assessments.TOTAL ENFORCEMENT REVENUE ASSESSED DOLLARS COLLECTED TO DATEFY 1992 FY 1993 FY 1994 FY 1995 FY 1996 FY 1997 FY 1998 FY 1999 FY 2000 TotalDistrict Audits (Inperson)$1,726,691,290 $1,780,689,403 $1,828,776,163 $1,726,477,044 $1,813,788,376 $2,017,646,371 $1,774,277,141 $1,165,385,199 $671,430,303 $14,505,161,290Service Center Audits(Correspondence)$1,016,491,617 $687,557,704 $526,609,451 $666,618,496 $777,337,295 $815,523,030 $658,498,312 $483,559,878 $344,592,608 $5,976,788,391Underreporter(Document Matching)Data not available $869,075,960 $1,238,308,612 $1,025,978,898 $1,062,864,926 $1,171,395,284 $1,307,984,870 $984,887,701 $7,660,496,250Source: ERIS dataNotes: Results above are related to cases closed in Exam or IRP Underreporter in that fiscal yearCollected dollars include tax, penalty, and interestRequest 3. The cost per IRP review and contact as well as the cost perservice center audit and face-to-face audit for the past twenty years.Cost per contact/audit calculations are estimated using closed case data forthe year in which the case is closed.. Due to the long case processing timefor in-person audits and some service center audits –cases may take a yearor longer – case closures and new cases initiated can vary substantially eachyear with little change in actual costs. For example, a high number ofclosures toward the end of a year would appear to have low costs per casewhile the following year would show more cases initiated but not closedappearing to have high costs per case.The following table provides estimated costs per contact/case for fiscal years1992 through 2000 for the Automated Underreporter Program (AUR) –formerly known as IRP – and for both service center and face-to-face auditsfor individual income tax returns. These estimates were derived for AUR andservice center exams using the total labor costs (direct salary and benefitcosts) and the number of contacts completed/cases closed for each programfor each year. Historical staff year rates were applied to create estimates ofthe costs associated with individual exams only. Financial data wereavailable only for all cases combined, including corporate audits. Costestimates for face-to-face audits may be slightly overstated becausecalculations were made using an average labor cost. Revenue Agentsworking corporate cases are higher graded, resulting in a slightly higheraverage labor cost. All costs are expressed in actual dollars. More detail isavailable but will require additional analysis.ESTIMATED COSTS FOR CONTACT/CASEFY TYPE OF CONTACT/AUDIT – INDIVIDUAL RETURNSAUR* Service Center Audits Face-to-Face Audits1992 $ 26.68 $ 210 $ 6671993 $ 36.25 $ 332 $ 7301994 $ 31.60 $ 268 $ 6971995 $ 32.53 $ 116 $ 6921996 $ 36.16 $ 104 $ 7041997 $ 24.33 $ 174 $ 7281998 $ 25.89 $ 193 $ 9041999 $ 29.04 $ 171 $1,0542000 $ 28.13 $ 336 $1,535Source: IRS AnalysisNote: * AUR data consists of cost estimates per Tax YearRequest 4. The IRS has significant correspondence with taxpayers regarding math error, which realizes millions inadditional revenue. Please explain if these dollars are included in enforcement revenue and also how many math errorcontacts are done each year for the past twenty years and how much revenue has resulted from this effort for the pasttwenty years.Enforcement revenue can result from a taxpayer failing to file a return, a taxpayer filing a return with a balance due that is not paid until enforcement action is begun (a collectionnotice is issued) or IRS collects additional taxes assessed from an examination or document matching contact. Not all revenue resulting from math error notices is considered enforcementrevenue by the Service. Math error adjustments are made during the processing of a tax return. A math error adjustment will result in either the taxpayer receiving less of a refund than shownon the return or the taxpayer will have a balance owing. A taxpayer is sent a notice of the math error adjustment, including the amount of the “reduced refund” or the balance that is now owed.If the taxpayer, having a balance due as a result of a math error adjustment, pays before a collection notice is issued, the revenue is not enforcement revenue. This is the same asif the taxpayer had filed the return with a balance due and paid the balance due before a collection notice was issued. Revenue collected before or after the collection notice is issued isadditional dollars in the Treasury and included in IRS total receipts. Much of the benefit of math error notices is in the area of revenue protected rather than enforcement revenue. Revenueprotected is a situation where monies are prevented from leaving the Treasury, while enforcement revenue is usually new dollars coming into the Treasury. Data on revenue protected byEITC and Dependent Taxpayer Identification Number math error notices is tracked by the Service, but revenue protected for all other math error notices is not tracked.MATH ERROR ANALYSIS FY1992 – 2000 ($ in 000s)1992 1993 1994 1995 1996 1997 1998 1999 2000Math Errors in Earned Income TaxCredit (EITC)Data not available(2) 903 623 475 400Net Revenue Protected $930,244 $548,072 $382,963 $291,538Net Protected per Notice $1,029 $880 $807 $730Math Errors in Dependent/Primary/ Secondary TIN1,300 1,095 958 1,270Net Revenue Protected $468,000 $245,435 $244,796 $301,710Net Protected per Notice $360 $224 $256 $237EITC/TIN Math Error (#) 2,203 1,718 1,433 1,670Revenue Protected by EITC/TINMath Error ($000s)$1,398,244 $793,507 $627,760 $593,248Average revenue protected perEITC/TIN Notice$635 $462 $438 $355All Other Math Error(1) 4,985 4,088 4,059 6,102 4,751 3,781 3,951 5,119 4,081Total Math Error (#) 4,985 4,088 4,059 6,102 4,751 5,984 5,669 6,552 5,751Source: Individual Master File Control Reports, Math Error Master File extract, Workload Inventory Tracking System ReportNotes: (All) Counts (shown in bold) are number of returns with errors, not number of errors. One taxpayer return could have multiple math errors but wouldreceive one notice informing the taxpayer of all errors in the return.(1) Revenue protected for all other math errors is not tracked.(2) The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 expanded Math Error Authority for EITC and Dependent TINs. Thisexpanded authority was implemented in Calendar Year (CY) 1997 for Tax Year (TY) 1996. The only data available prior to TY 1997 is aggregatedata on all other math errors.Request 5. Please explain why the 715,000 Service Center audits, whichare essentially correspondence to the taxpayer, are considered an auditfor IRS statistical purposes, yet the 3.6 million IRP notices from theservice centers are not.Internal Revenue Code (IRC) section 7605(b) restricts the IRS to onlyone examination of a taxpayer’s books and records for each taxable year.The definition of an examination/audit, in IRS’ MasterFile and otherrecords, is a contact that required the taxpayer to submit books andrecords for IRS review. IRS uses this same definition of an audit forstatistical purposes. The 715,000 Service Center audits requiredtaxpayers to submit books and records to substantiate a deduction,credit, etc. The 3.6 million IRP notices result from 2 different types ofcontact neither of which meet the criteria of an audit. These are:?? Underreported Income – There is a discrepancy between theincome reported on the individual income tax return and the incomereported by a third-party payer. The taxpayer is asked to explain thediscrepancy but does not provide books and records for IRS review.?? Nonfiling – The taxpayer has not filed a tax return after repeatedletters requesting a return and income reported on informationdocuments indicate a return is required to be filed. Under theauthority in IRC 6020(b), IRS computes a tax liability based on theinformation documents and the taxpayer is sent a notice of theproposed tax due. The taxpayer does not provide any books andrecords for IRS