Opportunity for Improvement Exists for Compliance With the
Freedom of Information Act and Related Procedures
August 2003
Reference Number: 2003-10-164
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
22, 2003
MEMORANDUM FOR
CHIEF, COMMUNICATIONS AND LIAISON
FROM: Gordon C. Milbourn III /s/ Gordon C.
Milbourn III
Assistant Inspector General
for Audit (Small Business and
Corporate Programs)
SUBJECT: Final Audit Report - Opportunity for Improvement Exists for
Compliance With the Freedom of Information Act and Related Procedures
(Audit # 200210040)
This
report presents the results of our Fiscal Year (FY) 2003 Freedom of Information
Act (FOIA) review. The overall
objective of this audit was to determine if the Internal Revenue Service (IRS)
improperly withheld information requested by taxpayers in writing, based on
FOIA exemption (b)(3), in conjunction with Internal Revenue Code (I.R.C.) §
6103, and/or FOIA exemption (b)(7), or by replying that responsive records did
not exist. Under § 1102 (d)(3)(A) of
the IRS Restructuring and Reform Act of 1998 (RRA 98), the Treasury Inspector
General for Tax Administration (TIGTA) is required to conduct periodic audits
of a statistically valid sample of the total number of determinations made by
the IRS to deny written requests to disclose information to taxpayers on the
basis of I.R.C. § 6103 and/or FOIA exemption (b)(7).
In
summary, in the cases we sampled, the IRS improperly withheld information from
requestors in 7.4 percent of the denied, partially denied, and no responsive
record FOIA and Privacy Act (PA) requests. In addition, the IRS improperly withheld
information from requestors in 9.3 percent of the I.R.C. § 6103 cases sampled
where information was denied or requestors were told that the records did not
exist. These represent smaller
percentages of improper withholdings than those reported in our FY 2002 report,
when the IRS improperly withheld information in 10.6 percent of the FOIA and PA
requests and in 12.2 percent of the I.R.C. § 6103 requests. However, the percentage of untimely
responses to FOIA and PA requestors increased to 43.5 percent of the cases in
this year’s sample, as compared to 22.7 percent in the FY 2002 sample.
Management’s Response: IRS
management agreed with the observations in our discussion draft report. Since the recommendations made in previous
TIGTA audit reports are still valid for the issues reported, we made no
additional recommendations. Therefore,
we did not ask the IRS for a formal response.
Copies of this report are also being sent to the IRS managers who are affected by the report. 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 Office of Disclosure Did Not Provide Complete Responses to
Requestors
Appendix I – Detailed Objective, Scope, and Methodology
Appendix II – Major Contributors to This Report
Appendix III – Report Distribution List
Appendix IV – Outcome Measures
Section (§) 1102 (d)(3)(A) of the Internal Revenue Service (IRS) Restructuring and Reform Act of 1998 (RRA 98) requires the Treasury Inspector General for Tax Administration (TIGTA) to conduct periodic audits of a statistically valid sample of the total number of IRS determinations to deny written requests to disclose taxpayer information on the basis of Internal Revenue Code (I.R.C.) § 6103 and/or Freedom of Information Act (FOIA) exemption (b)(7).
The FOIA requires records of the Federal Government to be made available to the public upon request, unless specifically exempt. FOIA exemption (b)(3) restricts the release of records specifically exempt from disclosure by statute (e.g., under I.R.C. § 6103). In turn, I.R.C. § 6103 controls the release of tax returns and return information and provides a mechanism for taxpayers to request tax returns and return information or request that it be disclosed to their designee. FOIA exemption (b)(7) restricts the release of records or information compiled for law enforcement purposes.
The Privacy Act (PA) contains a provision that prevents Federal Government agencies from relying on any exemption in the PA to withhold records that are otherwise available to an individual under the FOIA. We included PA cases in the sampled population, but only those partially or fully denied under FOIA exemption (b)(3), in conjunction with I.R.C. § 6103, and/or FOIA exemption (b)(7), to determine if the IRS properly withheld the requested information.
Within the IRS, the Office of Governmental Liaison and Disclosure, through its Office of Disclosure, is responsible for ensuring that the IRS complies with the FOIA, PA, and I.R.C. § 6103 requirements and responds within the statutory time periods for FOIA and PA requests. As such, the Office of Disclosure provides to field offices national oversight and guidance on procedural and policy matters and related training.
The Disclosure field offices process almost all the FOIA and PA requests received by the IRS. Written requests for information under I.R.C § 6103 may be processed by either the individual Disclosure offices or other IRS offices having custody of the requested records.
The IRS’ FOIA Annual Report for Fiscal Year (FY) 2002 disclosed that out of 34,069 FOIA and PA requests processed through the Disclosure offices, the IRS denied or partially denied 2,667 requests (7.8 percent) and replied that responsive records did not exist for 7,200 requests (21.1 percent). The remaining requests were either granted in full, referred to the agency where the records originated, withdrawn by the requestor, or not provided because the requestor did not meet the criteria in Treas. Reg. § 601.702 or for some other miscellaneous reason.
The FOIA, PA, and I.R.C. § 6103 requests received by the Disclosure offices are controlled on the Electronic Disclosure Information Management System (E-DIMS). The IRS is not required to track I.R.C. § 6103 requests for return or return information from taxpayers or individuals with a material interest (such as a spouse, child, estate, etc.). Nevertheless, the IRS has elected to track all I.R.C. § 6103 requests received by the Disclosure offices, but requests received by other IRS offices are not controlled on the E-DIMS or otherwise inventoried. Therefore, the volume of I.R.C. § 6103 requests received outside of the Office of Disclosure is unknown. As a result, the TIGTA can statistically sample only the I.R.C. § 6103 requests processed directly by the Disclosure offices and tracked on the E-DIMS. The universe of I.R.C. § 6103 requests closed by other IRS offices cannot be determined or statistically sampled.
Except for the limitations described in the previous paragraph, this audit was conducted in accordance with Government Auditing Standards from October 2002 through July 2003. The audit was performed at the Office of Governmental Liaison and Disclosure in Washington, D.C.
Detailed information on our audit objective, scope, and methodology is presented in Appendix I. Major contributors to the report are listed in Appendix II.
In 8 (7.4 percent) of the 108 FOIA and PA cases sampled and 9 (9.3 percent) of the 97 I.R.C. § 6103 cases sampled, the IRS did not provide complete responses and improperly withheld requested information. Projected to the population of 4,323 FOIA and PA cases and 5,822 I.R.C. § 6103 cases closed by the Disclosure offices from January 1, 2002, to June 30, 2002, we estimate there were 320 FOIA and PA cases and 540 I.R.C. § 6103 cases where the IRS did not provide available tax records when requested by taxpayers. As a result, while the extent of nondisclosure varied, the taxpayers’ rights were potentially violated and the IRS could risk incurring costs associated with administrative appeals and civil litigation initiated by those requestors improperly denied information.
Chart 1 shows the percentage of improper withholdings for FOIA/PA and I.R.C. § 6103 requests over the last four audit periods.
Chart 1 was removed due to its size.
To see Chart 1, please go to the Adobe PDF version of the report on the
TIGTA Public Web Page.
The IRS had a smaller percentage of improper FOIA and PA withholdings during the period from which our sample cases were drawn (January 1, 2002, through June 30, 2002) than in any prior period audited. The rate of improper withholdings for I.R.C. § 6103 requests also improved over the level reported in FY 2002, but it was still three percentage points above the level identified in our first audit in September 2000.
We analyzed the eight FOIA and PA cases, and the nine I.R.C. § 6103 cases, with improper withholdings from our samples and determined that the types of information being improperly withheld have remained relatively constant over the past four audit periods. As in prior years, these errors occurred principally because the Disclosure employees overlooked or did not address one or more items in a long list of requested items or did not completely research accessible resources to identify the information available. In general, the most common types of information withheld were miscellaneous IRS forms and documentation, tax transcript information, and employee notes and/or case history notes.
In response to prior TIGTA reports, the IRS initiated actions to improve the quality of responses. For example, the IRS developed the E-DIMS as its case processing system that electronically records case history and status, established an internal peer review process for evaluating the Office of Disclosure’s performance, and required Disclosure managers to quality review each denied or partially denied case. The IRS has also developed an Integrated Data Retrieval System (IDRS) training course that is currently scheduled for FY 2004. The IDRS is the computer system the caseworkers principally use to identify available taxpayer records.
Although the IRS has taken several actions intended to strengthen the program, it has yet to initiate the skills assessment proposed in response to our September 2000 report. Without a skills assessment, management may be hindered in identifying employee needs.
The IRS should ensure the skills assessment and the IDRS training are completed. By doing so, the IRS will continue to improve the quality of its responses to FOIA, PA, and I.R.C. § 6103 requests and avoid the potential costs associated with appeals and litigation.
For 47 (43.5 percent) of 108 cases in the FOIA and PA request random sample, the Office of Disclosure did not respond to the requestors in the time allowed by law. This represents a large increase in the number of untimely cases over prior years. We estimate, from the population of 4,323 cases closed from January 1, 2002, through June 30, 2002, there were 1,881 FOIA and PA requests not processed in accordance with the statutory provisions. As a result, taxpayers did not receive the level of service envisioned under the statute.
As shown in Chart 2, while there was a slight increase in
the number of untimely cases for the FY 2002 audit to 22.7 percent, the cases
reviewed for the FY 2003 audit showed a much larger percentage of untimely
cases at 43.5 percent.
Chart
2 was removed due to its size. To see
Chart 2, please go to the Adobe PDF version of the report on the TIGTA Public
Web Page.
The FOIA requires Federal Government agencies to determine whether to comply with a request within 20 days (excluding Saturdays, Sundays, and legal public holidays) after the receipt of a FOIA request by the Disclosure office. Federal Government agencies are to immediately notify the person making the request of the resulting determination, the reasons for the determination, and the right to appeal the determination. For the PA cases, the IRS must respond within 30 days of the taxpayer’s request.
The IRS may request an automatic 10-day extension and/or 1 voluntary extension in excess of the initial statutory 20-day period for FOIA requests. However, the requestor must agree to the voluntary extension and the IRS must notify the requestor of its determination by the end of the extension period. The FOIA does not allow for multiple extension requests.
For 17 of the 47 untimely FOIA and PA cases, the Office of Disclosure took over 60 days to respond to the taxpayer. From a review of the case files, we identified two primary reasons these cases were untimely. In seven of the cases, Disclosure personnel needed information from other IRS functions (e.g., Collection, Examination, Criminal Investigation, etc.) before the request could be provided. However, due to the slow response from the other IRS function(s), Disclosure personnel were not able to provide a timely response to the requestor. In four cases, the Disclosure personnel needed additional time to copy and review a large volume of documents or other higher priority requests were processed first. In the six remaining cases, the reasons for the delays were not documented.
If the IRS had requested the available 10-day extension and/or the voluntary extension, a large number of the 47 untimely cases could have complied with the statutory response time requirements. Obtaining the proper extensions would certainly have improved compliance for the 30 cases closed within 60 days of the taxpayer’s request, because in 21 of the 30 cases Disclosure personnel did not request an extension. In the other nine cases, when extensions were requested, Disclosure personnel either requested the extensions late (two cases) or did not provide a response within the extension period (seven cases). If Disclosure personnel had taken advantage of and properly applied the extensions available under the statute in just these 30 cases, the percentage of untimely responses in our sample would have improved substantially, from 43.5 percent to 15.7 percent.
The IRS should take action to ensure that taxpayer responses are provided within the time periods established under the statute. One effective way of improving timeliness, which is under the control of Disclosure management, would be to timely execute and request extensions provided for under the law. Disclosure management should emphasize the importance of using authorized extensions to assure timely responses and continue to monitor case timeliness. Responding to taxpayers’ requests for information within reasonable time periods would further support the IRS’ efforts to provide America’s taxpayers with top-quality service.
In response to previous TIGTA audit reports, the IRS took actions to improve response timeliness. The Director, Office of Governmental Liaison and Disclosure, initiated bi-weekly conference calls with Disclosure Area Managers to address inventory imbalances and over-age cases. Also, the new internal peer review process included an evaluation of case timeliness. These efforts, coupled with the Office of Disclosure’s emphasis on obtaining proper extensions, should help to improve case timeliness.
Recommendations made in previous TIGTA audit reports are still valid for the issues contained in this report. As a result, we are making no additional recommendations.
Appendix I
Detailed Objective, Scope,
and Methodology
The
overall objective of this audit was to determine if the Internal Revenue
Service (IRS) improperly withheld information requested by taxpayers in
writing, based on Freedom of Information Act (FOIA) exemption (b)(3), in conjunction with
Internal Revenue Code (I.R.C.) § 6103, and/or FOIA exemption (b)(7), or by
replying that responsive records did not exist. The following tests were performed to accomplish this objective:
I.
Determined
if the IRS properly adhered to statutory FOIA and Privacy Act (PA)
requirements, as well as procedural requirements.
A.
Identified
4,323 national FOIA and PA cases that were closed as partially or fully denied
based on FOIA exemption (b)(3), in conjunction with I.R.C. § 6103, and/or FOIA
exemption (b)(7), or where the IRS replied that responsive records did not
exist during the period January 1, 2002, through June 30, 2002.
1.
Obtained an
extract from the Electronic Disclosure Information Management System (E-DIMS)
for our audit period and identified all FOIA and PA cases closed as partially
or fully denied based on FOIA exemption (b)(3), in conjunction with I.R.C. § 6103,
and/or FOIA exemption (b)(7), or where the IRS replied responsive records did
not exist.
2.
Performed
limited tests to verify whether the E-DIMS extract included the specified time
period and information requirements and found only a minor problem with two
records from one office that did not affect the sampling process.
B.
Selected a
statistical sample of 108 accounts from the population of 4,323 FOIA and PA
cases that were partially or fully denied based on FOIA exemption (b)(3), in
conjunction with I.R.C. § 6103, and/or FOIA exemption (b)(7), or where the IRS
replied that responsive records did not exist.
Our sample size was determined based on a 90 percent confidence level,
an expected error rate of 10.33 percent, and precision of +/-5 percent. We selected this sampling methodology to
project the number of cases with improper withholdings to the universe of cases
that were partially or fully denied or where the IRS replied responsive records
did not exist.
C.
Reviewed
the sampled cases meeting our criteria to determine if the determination to
withhold information was appropriate, the record search was adequate, and/or
the determination was made timely. We
reviewed legal case precedents, as appropriate.
D.
Analyzed
audit results from the last 4 years to determine any common causes for the
improper withholding of FOIA and PA request information.
E.
Projected
the number of improper withholdings in the range of 142 to 499 (3.3 percent to
11.6 percent) FOIA and PA cases that were partially or fully denied based on
FOIA exemption (b)(3), in conjunction with I.R.C. § 6103, and/or FOIA exemption
(b)(7), or where the IRS replied that responsive records did not exist. The projection was made using attribute sampling,
with a 90 percent confidence level and an error rate of 7.4 percent. As a result, the actual precision factor was
4.2 percent.
F.
Projected
the number of untimely determinations in the range of 1,544 to 2,219 (36
percent to 51.8 percent) FOIA and PA cases that were partially or fully denied
based on FOIA exemption (b)(3), in conjunction with I.R.C. § 6103, and/or FOIA
exemption (b)(7), or where the IRS replied that responsive records did not
exist. The projection was made using
attribute sampling, with a 90 percent confidence level and an error rate of
43.5 percent. As a result, the actual
precision factor was 7.9 percent.
II.
Determined
if Disclosure offices adhered to legal requirements when denying written
requests received from taxpayers under I.R.C. § 6103 (as opposed to the FOIA or
PA).
A.
Obtained a
national extract from the E-DIMS and identified 17,277 I.R.C. §§ 6103 (c) and
(e) requests received in Disclosure offices from taxpayers or their designees
that were closed during the period January 1, 2002, through June 30, 2002. We performed limited tests to determine if
the E-DIMS extract included the specified time period and information
requirements.
B.
Designed a
statistical sample based on a 90 percent confidence level, an expected error
rate of 9.5 percent, and precision of +/-5 percent. A statistical sample was taken because it would allow us to
project the number of cases with improper withholdings to the universe of
closed I.R.C. §§ 6103 (c) and (e) requests where information was partially or
fully denied or requestors were told that records did not exist.
C.
Randomly
sampled 300 of the 17,277 I.R.C. §§ 6103 (c) and (e) closed cases. Reviewed the first 288 cases sampled to
obtain the required number of cases to statistically project our results.
1.
Determined
that 97 (33.7 percent) of the 288 randomly sampled and reviewed cases included
instances where information was partially or fully denied or requestors were
told that records did not exist.
2.
Based on an
initial analysis of the 288 randomly sampled cases, estimated that the
population of 17,277 closed I.R.C. §§ 6103 (c) and (e) requests contained 5,822
requests where information was partially or fully denied or requestors were
told that records did not exist.
D.
Reviewed
the 97 cases where information was partially or fully denied, or requestors
were told that records did not exist, to determine if the decision to withhold
the information based on I.R.C. § 6103 was appropriate.
E.
Analyzed
Treasury Inspector General for Tax Administration audit results from the last 4
years to determine any common causes for the improper withholding of I.R.C. §
6103 request information.
F.
Projected
the number of improper withholdings in the range of 258 to 822 (4.4 percent to
14.1 percent) for the I.R.C. §§ 6103 (c) and (e) requests where information was
partially or fully denied or requestors were told that records did not
exist. The projection was made using
attribute sampling, with a 90 percent confidence level and an error rate of 9.3
percent (rounded). As a result, the
actual precision factor was 4.8 percent.
Appendix II
Major Contributors to This Report
Daniel R. Devlin, Assistant Inspector General
for Audit (Headquarters Operations and Exempt Organizations Programs)
Mary V. Baker, Director
James D. O’Hara, Audit Manager
Kenneth L. Carlson, Jr., Senior Auditor
Yasmin B. Ryan, Senior Auditor
Tracy K. Harper, Auditor
Lynn M. Ross, Auditor
Appendix
III
Commissioner N:C
Director, Office of Governmental Liaison and Disclosure CL:GLD
Director, Office of Disclosure CL:GLD:D
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:
Chief, Communications and Liaison
CL
Appendix
IV
This appendix presents detailed information on the measurable impact that the recommended corrective actions made in a prior Treasury Inspector General for Tax Administration report will have on tax administration. These benefits will be incorporated into our Semiannual Report to the Congress.
Type and Value of Outcome Measure:
· Taxpayer Rights – Potential; 320 responses to Freedom of Information Act (FOIA) or Privacy Act (PA) requests where information was improperly withheld (see page 3).
Methodology Used to Measure the Reported Benefit:
The attribute sample was randomly selected from the universe of FOIA and PA requests that were closed nationally during the period January 1, 2002, through June 30, 2002, as 1) a full or partial denial with either FOIA exemption (b)(3), in conjunction with Internal Revenue Code (I.R.C.) § 6103, and/or FOIA exemption (b)(7) cited as one of the reasons for withholding information; or 2) a no responsive record case.
We arrived at the estimate by:
· Multiplying the number of requests closed as partially or fully denied based on FOIA exemption (b)(3), in conjunction with I.R.C. § 6103, and/or FOIA exemption (b)(7), or where the Internal Revenue Service (IRS) replied that responsive records did not exist by the error rate of cases reviewed. A case was considered an “error” if the IRS improperly withheld information that was available and could have been released under the FOIA and PA.
4,323 * 7.4 percent = 320 cases.
·
Taxpayer Rights – Potential; 540 responses to I.R.C §
6103 requests where information was improperly withheld (see page 3).
Methodology Used to Measure the Reported Benefit:
The attribute sample was randomly selected from the universe of I.R.C. §§ 6103 (c) and (e) requests that were closed nationally by the Disclosure offices during the period January 1, 2002, through June 30, 2002. The Disclosure offices are not required to input a disposition code showing how I.R.C. § 6103 cases are closed (granted, denied, etc.). Therefore, the majority of these cases did not include a disposition code.
We estimated the size of the universe by:
·
Identifying 17,277 closed I.R.C. §§ 6103 (c) and (e)
requests.
· Randomly selecting for review 288 of these cases to estimate the universe of denied, partially denied, or no responsive record cases closed during the period January 1, 2002, through June 30, 2002.
· Multiplying the total number of closed requests in the audit universe by the percentage of cases in the sample where information was partially or fully denied or taxpayers were told that records did not exist (97 of 288 or 33.7 percent).
17,277 cases * 33.7 percent = 5,822 cases.
· Multiplying the estimated universe of cases where information was partially or fully denied, or taxpayers were told that records did not exist, by the error rate for the cases reviewed (9.28 percent). A case was considered an “error” if the IRS improperly withheld information from the requestor.
5,822 cases * 9.28 percent = 540 cases.
Type and Value of Outcome Measure:
Methodology Used to Measure the Reported Benefit:
The attribute sample was randomly selected from the universe of FOIA and PA requests that were closed nationally during the period January 1, 2002, through June 30, 2002, as 1) a full or partial denial with either FOIA exemption (b)(3), in conjunction with I.R.C. § 6103, and/or FOIA exemption (b)(7) cited as one of the reasons for withholding information; or 2) a no responsive record case.
We arrived at the estimate by:
·
Multiplying the number of requests closed as partially
or fully denied based on FOIA exemption (b)(3), in conjunction with I.R.C. §
6103, and/or FOIA exemption (b)(7), or where the IRS replied that responsive
records did not exist, by the percentage of untimely responses. The following calculation was made to arrive
at the estimate:
4,323 * 43.5 percent = 1,881 cases.