Interest Paid to Large Corporations Could Significantly
Increase Under a Proposed New Revenue Procedure
August 2003
Reference
Number: 2003-30-176
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.
August
27, 2003
MEMORANDUM FOR COMMISSIONER, LARGE AND MID-SIZE BUSINESS DIVISION
FROM: for Gordon C. Milbourn III /s/ Daniel R. Devlin
Assistant Inspector General for Audit (Small Business and
Corporate Programs)
SUBJECT: Final Audit Report - Interest Paid to Large Corporations Could
Significantly Increase Under a Proposed New Revenue Procedure (Audit #
200230044)
This
report presents the results of our review of claims filed during examinations
of Coordinated Industry Cases (CIC) in the Large and Mid-Size
Business (LMSB) Division. The overall
objective of this review was to determine the impact claims are having on the
effectiveness and efficiency of tax administration.
In summary, claims filed
during examinations in the LMSB Division have been identified as a risk that
could hamper efforts to meet performance goals because, among other things,
they are believed to be a significant factor in extending the length of and
time spent on examinations. Since 2000,
the LMSB Division has chartered numerous design teams that have revamped
decades-old business processes and practices to meet expectations for a more
efficient examination process in the Division’s CIC and Industry Case
programs. One recent design team was chartered in 2002 to design a new process
for dealing with the risk that is believed to be associated with claims filed
during CIC examinations.
Based on the design team’s proposal,
the LMSB Division is considering issuing a new revenue procedure. The procedure would provide special
treatment for CIC taxpayers that file claims during examinations by allowing
examiners to delay the claim evaluation process until after the ongoing
examination is completed. A second
study is nearing completion in the LMSB Division’s Office of Strategy,
Research, and Program Planning (SRPP); LMSB Division management anticipates the
results will provide the basis for an overall strategy to address claims issues
in the Division’s examinations. However, both studies contained technical and
procedural shortcomings that diminish their usefulness. More significantly, we found that claims had
relatively little effect on the length of CIC examinations and that the
proposed new revenue procedure could significantly increase the interest the
Internal Revenue Service (IRS) must pay on claims.
When a claim is filed on a CIC return
that is under examination, the IRS incorporates the claim into the ongoing
examination, and any refund that may be due the taxpayer is issued at the end
of the examination. While this practice
guards against taxpayers receiving refunds for more than they are entitled to,
the interest that may be owed taxpayers continues to increase during
examinations.
Between
Fiscal Years 1998 and 2002, the IRS processed, on average, 382 CIC examinations
containing claims activity per year, with examiners allowing an average of $10
million in claims per examination. We
estimate that during this 5-year period approximately $637.6 million of
interest accrued each year. However, if
the claims were not considered until after the examinations were completed, as
proposed by the new revenue procedure, we estimate that during this 5-year
period slightly more than $1 billion of interest would have accrued per
year. This would have increased
interest costs by over 57 percent ($367.7 million) each year, or $1.839 billion
over 5 years.
In
view of this substantial cost, we recommended that the Commissioner, LMSB
Division, (1) direct that the design team’s study no longer be represented as
the basis for changing the process to manage claims that are filed during CIC
examinations and (2) conduct a pilot program to gather pertinent information
concerning the effect the proposed procedure will have on reducing the length
of examinations and interest costs.
Also, survey results from the second study that is nearing completion
should not be represented as being statistically valid in determining the
characteristics of formal claims filed in examinations or the amount of revenue
protected from examining claims.
In
our discussions with LMSB Division management on the issues contained in this
report, the managers stated our analysis did not consider that the proposed new
revenue procedure would enhance their ability to plan and control
examinations. According to the
managers, this will make it easier for them to shorten the length of
examinations and thereby minimize additional interest that may accrue by not
addressing claims filed during ongoing examinations. However, the evidence needed to show that the proposed new
revenue procedure would shorten examinations and minimize additional interest
costs has yet to be developed. We
clarified this point in footnotes to our analysis in Appendix IV.
Management’s
Response: The
Commissioner, LMSB Division, agreed with all our recommendations and responded
that the Claims Implementation Team has been asked to no longer use the claims
study as the basis for changing the process to manage claims filed during CIC
examinations and to conduct a pilot program to collect pertinent information
about the effects of the proposed revenue procedure on the length of
examinations and interest costs. In
addition, when the LMSB Division’s Office of SRPP
completes its research project, a cautionary note against using the data from
the survey to reach conclusions about the LMSB Division population will be
included in references to the claims survey and references to “Revenue
Protected” in the survey data will be corrected to “Amount of Claims
Disallowed.”
Even though actions are
being taken to implement all our recommendations, the Commissioner, LMSB
Division, believes that several conclusions in the report were not fully
supported by actual facts and requested that we reconsider some of the
statements in the report. Management’s
complete response to the draft report is included as Appendix V.
Office
of Audit Comment: We are encouraged that the Commissioner is
implementing all of our recommendations.
We understand that the purpose of delaying the examination of certain
claims is to assist in managing the examination process. However, we respectfully disagree that our
conclusions are not supported by actual facts.
We made appropriate revisions to a draft of this report to clarify the
basis for our conclusions and recommendations but believe further clarification
and explanation may be helpful for three areas that remain a particular source
of concern in IRS management’s response.
First,
the LMSB Division does not concur with our statement that, as proposed, the new
revenue procedure would increase interest costs by over 57 percent ($367.7
million) each year, or $1.839 billion over 5 years, and did not agree with the
use of data from formal claims in our computation. We agree that we used the average time spent on examinations that
were initiated by a formal claim. Our
report explained that we had to use the average time spent on these examinations
because the IRS does not track the time spent examining claims that are filed
during ongoing examinations of originally filed returns. In computing the interest costs, our intent
was, and still is, to show the risk posed by the claims process. Our computation relied on IRS data that
recognized the actual number and average amount, allowed by examiners, of
formal and informal claims that were filed during CIC examinations.
Second,
we were asked to reconsider the statement “…we found that claims had relatively
little effect on the length of CIC examinations…” because the LMSB Division
considers 3.7 months significant when considering the length of
examinations. As noted in the report,
we considered and recognized the 3.7 months (111 days) in computing the
potential increase in interest costs.
Further, the report notes that revisions to planned examination
completion dates were made nearly as often in examinations not containing a
claim as in those with one, which suggests that factors other than claims are
affecting the length of examinations.
Third, the LMSB Division disagreed
with the statement “By not accurately identifying the formal claims universe,
all statistical uses of the data were problematic” and the related discussion
in the report. The Commissioner, LMSB
Division, believes the sample was valid.
We do not agree with the LMSB Division’s position. As explained in the report, when
appropriately selected, statistically valid samples allow researchers to
determine the characteristics of some larger group (universe) by studying the
characteristics in a much smaller portion (sample) of the universe. To allow the results of statistically valid
sample measurements to be extended or generalized to the universe, researchers
need to ensure the sample selected accurately represents the universe to be
studied. In this instance, the sample
did not accurately represent the universe of accounts with formal claims
because accounts that did not have formal claims were included in the
sample. As a result, we continue to
believe that statistical uses of the data are problematic since it be will very
difficult, if not impossible, to allow the results of the sample measurements
to be extended or generalized to the universe of formal claims.
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 Parker F. Pearson, Director
(Small Business Compliance), at (410) 962-9637.
Sufficient and Competent Evidence Is Needed to Support the Proposed New Revenue Procedure
Appendix I – Detailed Objective, Scope, and Methodology
Appendix II – Major Contributors to This Report
Appendix III – Report Distribution List
Appendix IV – Outcome Measures
Appendix V – Management’s Response to the Draft Report
Since 2000, the Large and Mid-Size Business (LMSB) Division has chartered numerous design teams that have revamped decades-old business processes and practices to meet expectations for a more efficient examination process in the Coordinated Industry Case (CIC) and Industry Case (IC) programs. Both the Division and large corporate taxpayers believe that examinations take too long, particularly those in the CIC program.
In response to concerns over claims filed during CIC examinations, a design team was chartered in 2002 to develop a new process for dealing with the risk that is believed to be associated with claims filed during CIC examinations. The team surveyed LMSB Division field managers and validated that field managers believed claims extend the length of examinations because they increase the number of tax issues that need to be examined. Additionally, these managers believed that claims disrupted their planned examination coverage and hampered their ability to meet completion date goals.
A second study on claims is nearing completion in the LMSB Division’s Office of Strategy, Research, and Program Planning (SRPP). Among other responsibilities, the Research Unit in the Office of SRPP provides the LMSB Division with research expertise for assessing emerging issues and trends. To assess the issues and trends in claims, the unit has surveyed team managers and analyzed historical Internal Revenue Service (IRS) tax data. Although at the time we completed our audit work the Research Unit had not issued a final report, LMSB Division management anticipates the results will provide the basis for an overall strategy to address claims issues in the Division’s examinations.
When a claim is filed on a CIC return that is under examination, the IRS incorporates the claim into the ongoing examination, and any refund that may be due the taxpayer is issued at the end of the examination. While this practice guards against taxpayers receiving refunds for more than they are entitled to, the interest that may be owed taxpayers continues to increase during examinations. This is because under the Internal Revenue Code (I.R.C.), the IRS must generally pay interest on overpaid taxes from the later of the due date of the return or the date the return is filed. Since taxpayers have up to 3 years to file a claim and interest must be paid back to the overpayment date of the original return, the IRS may ultimately pay several years of interest on any refund due the taxpayer.
This
review was performed in accordance with Government Auditing Standards
between August 2002 and February 2003.
To meet our objective, we relied on the IRS’ internal management
databases and reports. We did not
establish the reliability of these data because extensive data validation tests
were outside the scope of this audit and would have required extensive resources
and time. Onsite work was performed in
the IRS Submission Processing Site in Ogden, Utah, and the LMSB Division
offices in Oakland, California, and Chicago, Illinois. Detailed information on our audit objective,
scope, and methodology is presented in Appendix I. Major contributors to the report are listed in Appendix II.
The IRS Restructuring and Reform Act of 1998 (RRA 98) required the agency to significantly improve the way it provides service and ensures compliance. To assist the IRS in meeting its mandate, the LMSB Division has chartered numerous design and research teams that are revamping decades-old business practices. According to LMSB Division customer satisfaction surveys, taxpayers reported high levels of satisfaction in their experiences with these changed business practices.
One of the focuses to increase LMSB Division taxpayer satisfaction was to reduce the length and contentious nature of the examination process. In response, a design team successfully developed, tested, and implemented the Industry Issue Resolution (IIR) Program to address contentious issues in a more “preemptive” and less intrusive manner than in an examination. Under the IIR Program, taxpayers can submit frequently disputed tax issues to the IRS for guidance. The benefit is that broad groups of taxpayers rely on the guidance and avoid spending time and resources resolving the issues in an examination.
To reduce the length of examinations, a Limited Issue Focused Examination process, which is an alternative to the traditional broad-based examination process, has also been introduced to examiners. The process encourages examiners to use risk analysis and materiality considerations to limit examinations to a few critical issues on a tax return. As shown in Table 1, numerous other strategies, initiatives, and research projects are underway to replace outdated practices. If these efforts are successful, both taxpayers and the IRS should benefit.
|
Title |
Description |
|---|---|
|
Issue Management Strategy |
A group of initiatives focused on early issue resolution. |
|
Compliance Risk Strategy |
A group of initiatives aimed at ensuring the most high-risk returns
are examined. |
|
Information Document Request Management Process |
A process for obtaining complete, timely information during
examinations. |
|
Fast Track Dispute Resolution Program |
An opportunity for taxpayers to appeal issues earlier in the dispute resolution process. |
|
Risk Analysis Process |
A process focused on shortening examinations by targeting high-risk
issues. |
|
Issue-Based Scoring |
A research project that is studying dollar values and other criteria
to identify issues in need of examination. |
|
Trends in Book-Tax Reporting and |
A research project studying potential noncompliance in gaps between
financial statement income and taxable income. |
Source: The LMSB Division.
Building on efforts to improve the examination process, a new revenue procedure is being considered as a result of the design team’s study. The procedure would provide special treatment for CIC taxpayers that file claims during examinations by allowing examiners to delay the claim evaluation process until after the ongoing examination is completed. A second study to determine if additional changes may be warranted to better manage claims filed during LMSB Division examinations is nearing completion.
However, both studies contained technical and procedural shortcomings that diminish their usefulness. More substantially, we found that claims had relatively little effect on the length of CIC examinations and that the proposed new revenue procedure could significantly increase the interest the IRS must pay on claims.
The General Accounting Office’s (GAO) Business Process Reengineering Assessment Guide provides key concepts that can be applied to designing new business processes in the LMSB Division. The concepts include (1) justifying in sufficient detail the need for proposed changes, (2) securing sufficient stakeholder support, and (3) performing adequate cost-benefit analysis.
We found that, due to time and resource constraints, the design team’s approach for revamping the process for handling claims during examinations did not include some key concepts recommended in the GAO Guide. As a result, we identified enough technical and procedural shortcomings in the design team’s study for us to conclude that it should not be used as the basis for the proposed new revenue procedure.
In justifying the need for changing the way claims are handled, the design team did not adequately consider important data on the claims filed during CIC examinations from Fiscal Years (FY) 1993 through 2002. The need for the new proposed revenue procedure was not supported since there was no correlation analysis of the relationship that a claim filed during an examination has with the length of the examination.
We analyzed the IRS’ Coordinated Examination Management Information System (CEMIS), which contained results from 4,971 CIC examinations closed from FYs 1993 through 2002, and found that claims had relatively little impact on the length of examinations. For example, we found that revisions to planned completion dates were made nearly as often in the examinations not containing a claim as in those with one. Also, over the 10-year period, the CIC examinations that contained claims extended the length of examinations, on average, only about 111 days (3.7 months). Taking these two facts into consideration suggests that factors other than claims are affecting the length of CIC examinations.
In securing stakeholders’ support for the new process, tax professionals in 3 leading organizations and approximately 5,000 LMSB Division examiners and managers were surveyed. However, only 195 of those targeted for the survey responded. According to the IRS’ own internal criteria, this relatively small number of respondents (less than 5 percent) would be of little or no value in measuring the level of support for the new process.
In analyzing the costs and benefits of the new process, the design team did not sufficiently consider interest the IRS must pay on refunds. When a claim is filed on a CIC return that is under examination, the IRS currently incorporates the claim into the ongoing examination, and any refund that may be due the taxpayer is issued at the end of the examination.
The
proposed new revenue procedure changes this long-standing practice by allowing
examiners to delay the CIC examination of the claim until after the examination
of the original return is completed.
With FY 2002 CIC examinations taking, on average, 1,182 days (38.9
months) to complete, the new revenue procedure would significantly risk
increasing the interest the IRS must pay on any claims that are allowed by
examiners.
To determine the potential additional cost, we estimated the amount of additional interest that would be paid under the proposed new revenue procedure. Our analysis covered a 5-year period beginning in FY 1998 and is based on the number and dollar amounts of claims filed in CIC examinations that were allowed by examiners.
Table 2 shows that, between FYs 1998 and 2002, the IRS processed, on average, 382 CIC examinations containing claims activity per year, with examiners allowing an average of $10 million in claims per examination. We estimate that during this 5-year period approximately $637.6 million of interest accrued each year.
|
Fiscal Year |
Number of CIC Exams With
Claims |
Average Dollar Amount of
Claims Allowed |
Interest Accrued on Claims
During Exams |
Average Days Spent to
Examine Claims |
Interest Accrued on Claims
If Addressed After Exams Are Completed |
|---|---|---|---|---|---|
|
1998 |
438 |
$9.1 |
$700.6 |
575 |
$1,153.1 |
|
1999 |
405 |
$6.0 |
$407.4 |
499 |
$620.3 |
|
2000 |
322 |
$10.1 |
$519.7 |
530 |
$864.3 |
|
2001 |
349 |
$11.5 |
$606.5 |
585 |
$1,024.2 |
|
2002 |
394 |
$13.5 |
$953.6 |
769 |
$1,364.2 |
|
Averages |
382 |
$10.0 |
$637.6 |
592 |
$1,005.2 |
(~) Due to rounding, numbers may not add or subtract
precisely.
Source: The Treasury Inspector General for Tax
Administration’s combined analysis of the IRS’ CEMIS, Audit Information
Management System (AIMS), and Integrated Data Retrieval System (IDRS).
The analysis in Table 2 also shows that if the claims were not considered until after the examinations were completed, as proposed by the new revenue procedure, during this 5-year period slightly more than $1 billion of interest would have accrued each year. This would increase interest costs by over 57 percent ($367.7 million) each year, or $1.839 billion over 5 years. Thus, the cost may far exceed the benefits that would be achieved under the new proposed revenue procedure. Additional detail on the analysis is provided in Appendix IV.
In our discussions with LMSB Division management on the issues contained in this report, the managers stated our analysis did not consider that the proposed new revenue procedure would enhance their ability to plan and control examinations. According to the managers, this will make it easier for them to shorten the length of examinations and thereby minimize additional interest costs that may accrue by not addressing claims filed during ongoing examinations. However, the evidence needed to show that the proposed new revenue procedure would shorten examinations and minimize interest costs has yet to be developed. We clarified this point in footnotes to our analysis in Appendix IV.
Because of the study’s shortcomings and the uncertainties about whether the proposed new revenue procedure would shorten the length of CIC examinations and not increase interest costs, we recommend that the Commissioner, LMSB Division:
1.
Direct that the study no longer be represented as the basis for
changing the process to manage claims that are filed during CIC examinations.
Management’s Response: The Commissioner, LMSB Division, asked the Claims Implementation Team to no longer use the claims study as the basis for changing the process to manage claims filed during CIC examinations.
2. Gather pertinent information concerning the effect the proposed procedure will have on reducing the length of examinations and interest costs by conducting a pilot program to demonstrate the actual benefits that could be achieved.
Management’s Response: The Commissioner, LMSB Division, responded that the Claims Implementation Team will conduct a pilot program to collect pertinent information about the effects of the proposed revenue procedure on the length of examinations and interest costs. The team will also evaluate other cost and benefit aspects of the process.
Even though actions are being taken to implement the two recommendations discussed above, the Commissioner, LMSB Division, believes that several conclusions in the report related to these two recommendations were not fully support by actual facts. In addition, the Commissioner, LMSB Division, requested that we reconsider some of the statements in the report related to these two recommendations.
Office of Audit Comment: We are encouraged that the Commissioner is implementing all of our recommendations. We understand that the purpose of delaying the examination of certain claims is to assist in managing the examination process. However, we respectfully disagree that our conclusions are not supported by actual facts. We made appropriate revisions to a draft of this report to clarify the basis for our conclusions and recommendations but believe further clarification and explanation may be helpful for the following two areas that remain a particular source of concern in IRS management’s response.
First, the LMSB Division does not concur with our statement that, as proposed, the new revenue procedure would increase interest costs by over 57 percent ($367.7 million) each year, or $1.839 billion over 5 years, and did not agree with the use of data from formal claims in our computation. We agree that we used the average time spent on examinations that were initiated by a formal claim. Our report explained that we had to use the average time spent on these examinations because the IRS does not track the time spent examining claims that are filed during ongoing examinations of originally filed returns. In computing the interest costs, our intent was, and still is, to show the risk posed by the claims process. Our computation relied on IRS data that recognized the actual number and average amount, allowed by examiners, of formal and informal claims that were filed during CIC examinations.
Second, we were asked to reconsider the statement “…we found that claims had relatively little effect on the length of CIC examinations…” because the LMSB Division considers 3.7 months significant when considering the length of examinations. As noted in the report, we considered and recognized the 3.7 months (111 days) in computing the potential increase in interest costs by delaying the examination of claims as proposed by the LMSB Division. Further, the report notes that revisions to planned examination completion dates were made nearly as often in examinations not containing a claim as in those with one, which suggests that factors other than claims are affecting the length of examinations.
Despite a good faith effort, the combination of data limitations and technical and procedural problems diminish the usefulness of a second claims study that is underway in the LMSB Division’s Office of SRPP. In summary, the results from a survey on formal claims conducted during the study are not reliable since the Office of SRPP did not have the accurate universe of formal claims to allow statistically reliable inferences about the characteristics of formal claims filed in open examinations. In addition, the amount of revenue protected as measured by the survey is not accurate.
An initial technical error in identifying the universe of claims has caused problems. The IRS does not have readily available data on the number of claims filed by taxpayers served by the LMSB Division. To overcome this data limitation, the research team matched two databases intending to produce a universe of formal claims filed in open examinations. However, we found that because the match contained incorrect criteria, the team was not able to accurately identify the universe of formal claims in open examinations. To validate this, we reviewed information from 68 taxpayer accounts included in the research team’s universe of formal claims and found that 19 of the accounts contained no evidence that a formal claim was filed during an open examination.
By not accurately identifying the formal claims universe, all statistical uses of the data were problematic. For example, the team decided to survey a statistically valid sample of all LMSB Division managers with formal claims in open examinations. The incorrect criteria led to surveying 33 managers who should not have been selected for the survey because they were not involved with formal claims in open examinations. Because the 33 managers were included in the sample, conclusions determined from the sample cannot be statistically projected or generalized to the universe of managers involved with formal claims in examinations.
A conventional requirement for data
collection instruments, such as surveys, is that questions be clear and
understandable so that complete and accurate responses are received. To find out what may be confusing about
survey questions, researchers will often “try out” the survey on a number of
likely respondents and then clarify the questions or instructions as
needed.
Although efforts were made to test
the survey on likely respondents, our discussions with 13 managers that
responded to the survey suggested that more could have been done to ensure
questions were answered accurately.
The survey contained at least 42 items and as
many as 522 items for special cases, and several of the managers told us they
had difficulty understanding the instructions and some of the survey
questions. This helps explain why we
found numerous inaccuracies in their responses, including a claim that was
misstated by $107 million.
From a procedural standpoint, the
research team also encountered problems.
An important output measure that the survey attempted to quantify is the
amount of revenue protected as a result of examinations. In examinations, revenue protection measures
examiner efforts in preventing taxpayers from receiving refunds larger than
they are entitled to on the claims they file.
According to the IRS Internal
Revenue Manual (IRM), at least eight factors need to be considered in
calculating the measurement. However,
most of factors needed for the calculation were not collected in the survey,
which will make it impossible to accurately measure the revenue protected from
the survey.
Due to the shortcomings in the
survey, we recommend that the Director, SRPP, ensure:
3.
The survey results are not represented as being
statistically valid in determining the population or the characteristics of
formal claims filed in open examinations.
This qualification needs to be considered in any decision-making process
concerning the results.
Management’s Response: The Commissioner, LMSB Division, responded that when the Office of SRPP completes the Claims Research Project a cautionary note against using the data from the survey to reach conclusions about the LMSB Division’s population would be included in references to the claims survey. The LMSB Division will also revise all existing material about the survey to provide a similar caution.
4. The amount of revenue protected from examining claims as
determined by the survey is not represented as an accurate measurement.
Management’s Response: The Commissioner,
LMSB Division, responded that references to “Revenue Protected” in the survey
data will be corrected to “Amount of Claims Disallowed” when the Claims
Research Project is completed.
Even though actions are being taken to implement the two recommendations discussed above, the LMSB Division disagreed with the statement “By not accurately identifying the formal claims universe, all statistical uses of the data were problematic” and the related discussion in the report. The Commissioner, LMSB Division, believes the sample was valid.
Office of Audit Comment: We are encouraged that the Commissioner is implementing all of our recommendations. However, we do not agree with the Commissioner’s position and believe that further clarification and explanation may be helpful for this third area that remains a particular source of concern in IRS management’s response. As explained in the report, when appropriately selected, statistically valid samples allow researchers to determine the characteristics of some larger group (universe) by studying the characteristics in a much smaller portion (sample) of the universe. To allow the results of statistically valid sample measurements to be extended or generalized to the universe, researchers need to ensure the sample selected accurately represents the universe to be studied.
In this instance, the sample did
not accurately represent the universe of accounts with formal claims because
accounts that did not have formal claims were included in the sample. As a result, we continue to believe that
statistical uses of the data are problematic since it will be very difficult,
if not impossible, to allow the results of the sample measurements to be
extended or generalized to the universe of formal claims.
Appendix I
Detailed Objective, Scope,
and Methodology
The overall objective of this review was to determine the impact claims are having on the effectiveness and efficiency of tax administration. To meet our objective, we relied on the Internal Revenue Service’s (IRS) internal management databases and reports. We did not establish the reliability of these data because extensive data validation tests were outside the scope of this audit and would have required extensive resources and time. The specific audit tests included the following:
I. Performed a walk-through of the appropriate operations in the Ogden Submission Processing Site to determine how formal claims are received, controlled, and posted to accounts.
II. Reviewed work plans, status reports, and, if available, final reports of the LMSB Division design and research team to determine its goals, scope, methodology, and recommendations.
III. Interviewed officials from the LMSB Division design and research team and reviewed pertinent documents to learn about the various LMSB Division strategies, initiatives, and research projects that are changing work processes to improve service to taxpayers and ensure compliance.
IV. Used the General Accounting Office’s (GAO) Business Process Reengineering Assessment Guide in assessing how well the LMSB Division design team applied “best practices” outlined in the GAO Guide to its study.
V. Analyzed the Coordinated Examination Management Information System (CEMIS) and the Audit Information Management System (AIMS) in assessing the length of Coordinated Industry Case (CIC) examinations, number and amount of claims filed in CIC examinations, how claims were resolved in CIC examinations, and the impact claims had on timely completing CIC examinations.
VI. Used the IRS’ Integrated Data Retrieval System (IDRS) to compute the applicable rate of Federal interest under Internal Revenue Code Section 6621 on claims analyzed during the review.
VII. Evaluated the criteria used by the LMSB Division research team in determining the population of examinations containing claims and assessed whether the criteria would yield intended results.
VIII. Used the American Statistical Association’s Guide on Judging the Quality of a Survey in assessing how well the LMSB Division research team applied “best practices” outlined in the Guide to its survey.
IX. Used the IRS Internal Revenue Manual to determine whether the LMSB Division’s research team considered required factors in computing revenue protection amounts.
Appendix II
Major Contributors to This Report
Parker Pearson, Director (Small Business Compliance)
Philip
Shropshire, Director
Frank
Dunleavy, Audit Manager
Stanley
Pinkston, Senior Auditor
Lisa Stoy, Senior Auditor
William Tran, Senior Auditor
Debra Mason, Auditor
Appendix III
Commissioner N:C
Deputy Commissioner for Services and Enforcement N:SE
Acting Deputy Commissioner, Large and Mid-Size Business Division LM
Director, Retailers, Food, Pharmaceuticals, and Healthcare Industry, Large and Mid-Size Business Division LM:RFP
Director, Strategy, Research, and Program Planning, Large and Mid-Size Business Division LM:SR
Chief
Counsel CC
National
Taxpayer Advocate TA
Director,
Office of Legislative Affairs CL:LA
Director,
Office of Program Evaluation and Risk Analysis
N:ADC:R:O
Office
of Management Controls N:CFO:AR:M
Audit
Liaison: Commissioner, Large and
Mid-Size Business Division LM:CL
Appendix IV
This appendix presents detailed information on the measurable impact that our recommended corrective actions will have on tax administration. This benefit will be incorporated into our Semiannual Report to the Congress.
Type and Value of Outcome Measure:
· Funds Put to Better Use – Potential; an average of $367.7 million per year; $1.839 billion over 5 years. This represents the interest that could be avoided by continuing to incorporate claims into ongoing Coordinated Industry Case (CIC) examinations rather than delaying the examination of claim issues until after ongoing examinations are completed (see page 5).
Methodology Used to Measure the Reported Benefit:
To estimate the potential funds that could be put to better use, we analyzed three of the Internal Revenue Service’s (IRS) automated information systems. The first system, the Coordinated Examination Management Information System (CEMIS), contains information on the length of CIC examinations, the number and amount of claims filed by taxpayers during CIC examinations, and the number and amount of claims that examiners allowed during CIC examinations. The second system, the Audit Information Management System (AIMS), contains information on the time span of CIC examinations that involved only claims for refunds on amended returns. The third system, the Integrated Data Retrieval System (IDRS), contains a computer program that calculates the applicable amount of Federal interest on tax overpayments and underpayments.
Based on the analysis in Table 1, between Fiscal Years 1998 and 2002, the IRS processed, on average, 382 CIC examinations with claims activity per year, with examiners allowing an average of $10 million in claims per examination. We used the IDRS and determined that during this 5-year period, approximately $637.6 million of interest accrued each year. The analysis in Table 1 also shows that delaying the examination of the claims as currently proposed would risk increasing interest costs from $637.6 million to slightly over $1 billion each year. This is an increase of approximately $367.7 million each year, or a $1.839 billion increase over 5 years.
|
A |
B |
C |
D |
E |
F |
G |
H |
|---|---|---|---|---|---|---|---|
|
Fiscal Year |
Number of CIC Exams With Claims |
Average Dollar Amount of Claims Allowed |
Average Length of Exams in Days |
Interest Accrued on Claims During Exams |
Average Days Spent to Examine Claims |
Interest Accrued on Claims If Addressed After Exams Are Completed |
Additional Accrued Interest (G less E) |
|
1998 |
438 |
$9.1 |
914 |
$700.6 |
575 |
$1,153.1 |
$452.5 |
|
1999 |
405 |
$6.0 |
906 |
$407.4 |
499 |
$620.3 |
$212.9 |
|
2000 |
322 |
$10.1 |
918 |
$519.7 |
530 |
$864.3 |
$344.6 |
|
2001 |
349 |
$11.5 |
919 |
$606.5 |
585 |
$1,024.2 |
$417.7 |
|
2002 |
394 |
$13.5 |
1,071 |
$953.6 |
769 |
$1,364.2 |
$410.6 |
|
Averages |
382 |
$10.0 |
946.9 |
$637.6 |
592 |
$1,005.2 |
$367.7 |
(…) We found that CIC
examinations containing claims over a 10-year period may have been a factor in
extending the length of examinations by an average of 111 days. Consequently, we reduced the average length
of the examinations as shown in the table by the 111 days.
(à) We did not consider the impact that the proposed new
revenue procedure could have on reducing the length of examinations or on the
number of claims that would be addressed after examinations are completed
because the data were not available.
According to Large and Mid-Size Business Division managers, the proposed
revenue procedure would enhance their ability to plan and control
examinations. This, according to the
managers, will make it easier for them to shorten the length of examinations
and thereby minimize additional interest that may accrue by not addressing
claims during ongoing examinations.
(P) We used the average time spent on examinations that were
initiated by a claim for refund on an amended return because the IRS does not
track the time spent on claims that are filed during ongoing examinations of
originally filed returns. The time
spent on claims filed during ongoing examinations of originally filed returns
could be less because, among other things, taxpayer records are readily
available to examine.
(O) We excluded the time
spent examining claims that are subject to review by the Joint Committee on
Taxation. Generally, these claims
involve requests for refunds of $2 million or more and can take additional time
to process because of a special review process.
(~) Due to rounding, numbers may not add or subtract
precisely.
Source: The Treasury Inspector General for Tax
Administration’s combined analysis of examination results from the IRS’ CEMIS,
AIMS, and IDRS.
Appendix V
Management’s Response to
the Draft Report
The response
was removed due to its size. To see the
complete response, please go to the Adobe PDF version of the report on the
TIGTA Public Web Page.