The Federal, State, and Local Governments Office Is Taking Action to Identify Its Customers, but Improvements Are Needed
Reference Number: 2004-10-104
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.
May 25, 2004
MEMORANDUM FOR COMMISSIONER, TAX EXEMPT AND GOVERNMENT ENTITIES DIVISION
FROM: Gordon C. Milbourn III /s/ Gordon C. Milbourn III
Acting Deputy Inspector General for Audit
SUBJECT: Final Audit Report - The Federal, State, and Local Governments Office Is Taking Action to Identify Its Customers, but Improvements Are Needed (Audit # 200310031)
This report presents the results of our review assessing the Tax Exempt and Government Entities Division’s efforts to accurately identify its Federal, state, and local government customers.
In summary, Federal, State, and Local Governments (FSLG) office management has begun the process to identify Federal, state, and local government customers. Government Entities (GE) function management recognized that the FSLG office encountered difficulty in identifying its customers that has limited progress to date. However, the FSLG office needs to take a more structured approach to ensure actions for identifying customers are successful. For example, FSLG office management did not identify a structured and effective approach to address limitations for matching data between the United States (U.S.) Census Bureau’s 2002 Governments Integrated Directory (GID) and Internal Revenue Service’s (IRS) computer systems. Also, action taken by the FSLG office to correct employment codes used to categorize employers as Federal, state, or local governments on IRS computer systems was not always successful. In addition, efforts to develop a database to capture information about FSLG office customers were not included in management’s action plan. Finally, corrective actions taken in response to a prior report to develop operational plans were not effective. Specifically, these plans did not always include the actions planned, individuals assigned, responsible management official(s), completion dates, expected results, and methods to monitor and report performance.
We repeated our prior recommendation to develop action plans that include all of the information stated above. In addition, we recommended the Director, GE, ensure that a survey is mailed to all government entities appearing on the U.S. Census database that are not on IRS computer systems, confirm their names and addresses, and request the entities’ Employer Identification Numbers.
Management’s Response: IRS management agreed with the recommendations contained in the report. The Director, FSLG office, is preparing a detailed action plan that will include the application of employment codes to categorize employers as Federal, state, or local governments on IRS computer systems. A project leader has been selected to complete the steps in the action plan. In addition, the Director, FSLG office, will continue to refine customer identification through matching data between the U.S. Census Bureau’s GID and the IRS’ computer systems and will do another survey to resolve identification errors with mismatched entities. Management’s complete response to the draft report is included as Appendix V.
Copies of this report are also being sent to the IRS managers affected by the report recommendations. Please contact me at (202) 622-6510 if you have questions or Daniel R. Devlin, Assistant Inspector General for Audit (Headquarters Operations and Exempt Organizations Programs), at (202) 622-8500.
The Government Entities (GE) function of the Tax Exempt and Government Entities (TE/GE) Division is divided into three distinct offices that are modeled after its three different sets of customers—Indian Tribal Governments (ITG); Tax Exempt Bonds; and Federal, State, and Local Governments (FSLG). The FSLG office is the largest of the three. Its mission is to provide Federal Government agencies, state governments, local governments, and quasi-governmental entities top-quality service by helping them understand and comply with the tax laws. FSLG office customers are generally not subject to Federal income tax; however, these customers are generally required to file information returns and file and pay employment taxes. The FSLG office’s interactions with its customer base affects compliance levels of the government employers as well as their 23 million employees, which is the largest single block of employees in the United States (U.S.).
The Internal Revenue Service (IRS) Business Master File (BMF) includes 78,627 Federal Government agencies, state governments, and local municipalities. However, this figure does not include potentially thousands of subunits that provide employment tax and information return reporting information for consolidation to a particular entity. Because these subunits may affect the compliance of another government entity, the FSLG office has decided to identify these customers as well. However, FSLG office management’s attempts to identify new customers have been hampered because the FSLG office has not yet been fully staffed. In addition, the database used by the IRS to identify government customers was not complete or accurate when the GE function was established.
In a prior audit, we reported that 12,878 Federal, state, and local entities were identified as delinquent during Calendar Years 1999, 2000, or 2001. Additional analysis of these entities showed that 2,697 entities had not submitted the required tax payments (resulting in balances due of $104 million), and 1,604 entities had not submitted the required tax returns to account for almost $7.4 billion in tax deposit credits. TE/GE Division management believed a significant portion of the delinquent dollar amount represented error conditions or misapplied payments that, when resolved, would result in a zero balance.
In February 2003, the FSLG office published a Research Plan as part of a major initiative to better identify, quantify, and monitor its customers. The major goals of the Plan are to:
· Identify all government entity customers.
· Develop major market segments and their significant submarket segments.
· Develop customer and industry profiles.
· Identify trends, issues, and problems of customers.
· Determine market segment compliance risk levels.
· Identify workload for various compliance treatments.
This review was performed from July 2003 through February 2004 at the TE/GE Division Headquarters Office in Washington, D.C. We also contacted personnel located in FSLG offices in Cheektowaga, New York, and Austin, Texas. The audit was conducted in accordance with Government Auditing Standards. Detailed information on our audit objective, scope, and methodology is presented in Appendix I. Major contributors to the report are listed in Appendix II.
FSLG office management has begun the process to identify Federal, state, and local government customers. For example, GE function management developed a Research Plan that addresses the identification of its customers; customer characteristics (profiling); assessment of risk; allocation of resources; and trends, issues, and problem identification. In addition, GE function management has added data to the Return Inventory Classification System (RICS) from other IRS computer systems and the U.S. Census Bureau to obtain more accurate information about its customer base.
GE management recognized that it encountered difficulty in identifying its customers that has limited progress to date. In June 2003, GE management selected a new Director of the FSLG office, who stated one of his main priorities was the identification of FSLG office customers. To facilitate this effort, action plans for identifying customers were updated to provide additional detail. In addition, the Director initiated an effort to develop a database to capture information on FSLG office customers. However, the FSLG office needs to take a more structured approach to ensure actions for identifying customers are successful.
Differences between data contained in the U.S. Census database and IRS records have delayed identification efforts
The FSLG office has obtained a download of the U.S. Census Bureau’s 2002 Governments Integrated Directory (GID) to use in identifying incomplete or inaccurate information on the IRS BMF. This is one of the FLSG office’s initiatives to update existing customer information and identify new customers. The GID includes 87,900 active governments, while the BMF includes 78,627 government entities.
However, FSLG office management has had only limited success with this initiative. FSLG office management had previously received 1997 GID information but did not use it for identifying customers because they felt it was outdated. FSLG officials were aware of potential problems with matching the two databases when they first obtained the U.S. Census data. They recognized that the BMF contained fewer government entities than the U.S. Census database. In addition, the two databases cannot be systemically matched because the GID does not contain a unique identifier, such as an Employer Identification Number (EIN), to use for matching purposes. Further, both databases use different format and naming conventions for entities’ names and addresses, which makes systemic matching of the entities more difficult.
Although FSLG office management consulted with the TE/GE Division’s Research office in an attempt to identify an effective process to match the databases, they did not identify a structured and effective approach to address this limitation. Instead, database matching efforts have generally consisted of ad hoc trial and error techniques. For example, the FSLG office was able to match only 57 percent of the databases for county governmental units within specific zip code ranges. Further, 40 percent of those entities had to be matched manually.
FSLG officials stated that a timeline for completing the matching of all entities on the GID and BMF databases was not established because FSLG office management had not identified an effective methodology of matching these two databases. In addition, the action plan for this initiative did not provide the specific actions or detailed methodologies to be used to match the GID and BMF databases and to correct erroneous information on IRS computer systems. Further, the action plan items did not include information on how the work related to the unmatched organizations for the remaining government segments would be divided among FSLG field offices, how the field offices would conduct the match, what the scheduled completion dates would be, or how the results of the matching would be monitored and reported to senior management.
The FSLG office initially planned to use field offices or the Office of Outreach, Planning, and Review to resolve all unmatched organizations, for each government segment. We discussed with FSLG office management the feasibility of mailing surveys to customers to facilitate the matching of the GID and BMF databases. This would reduce the amount of work required by field offices to resolve unmatched organizations, which should result in more effective use of their limited resources. FSLG office management agreed that this methodology might make more effective use of its resources.
Subsequent to these discussions, FSLG office management informed us they sent out a pilot sample of 500 survey letters to government organizations that were on the U.S. Census database but not on the BMF. The survey letter contained five boxes that recipients could check to explain why they were included on the GID but not on the BMF. FSLG office management stated that they received 296 responses to the surveys and, based on the response level, will mail surveys to the remaining unmatched organizations. This will be done after the FLSG office refines the number of unmatched entities using an improved matching technique.
The FSLG office took action to correct government employment codes on IRS computer systems but was not always successful
Another initiative undertaken by the FSLG office during 2003 was to identify and correct inaccurate employment codes used to categorize employers as Federal, state, or local governments on IRS computer systems. FSLG office personnel informed us they reviewed a RICS report containing the names, addresses, employment codes, and business operating codes of over 80,000 entities, looking for obvious coding errors based upon the name and employment code of the entity. In addition, FSLG office personnel queried the RICS database using key words (e.g., city, town, and police) and specific filing requirements to identify miscoded entities. FSLG office personnel then attempted to assess the accuracy of the entities’ employment codes based upon these reviews (looking at the organizations’ names and researching filing requirements).
FSLG office management provided us with documentation indicating they identified 1,617 errors based upon their review of the approximately 80,000 government entities from the RICS data. They stated that 1,095 of the errors included entities mistakenly coded as governmental, previously unidentified government entities, and government entities that had the incorrect type of employment code (e.g., Federal, state or local), but they could not provide us with documentation showing how these errors were corrected. However, the FSLG office did provide us with documentation for the remaining 522 cases, which showed:
· Nongovernment entities were mistakenly coded as government employers in 445 cases.
· Government entities were not coded as government employers in 15 cases.
· Government entities were mistakenly coded as the wrong type of government employer in 26 cases.
· Employment codes for 36 entities were changed, but the case files did not contain the entities’ original employment codes. As a result, we could not determine the original type of employer based on our review of the documentation.
We randomly selected 50 of these 522 cases and researched IRS computer systems and/or the Internet to determine if the revised employment codes were accurate. Based on our research, we determined that the revised employment codes were not accurate in 10 (20 percent) of 49 cases.
In 5 of these 10 cases, the FSLG office appropriately determined that the entity was not a government but did not correctly revise the entity’s employment code to reflect its nonprofit status. These five cases included:
· Four churches that should have been reclassified as nonprofit organizations exempt from unemployment tax.
· One exempt organization that should have been reclassified as a nonprofit organization exempt from unemployment tax.
The remaining five cases consisted of instances in which the FSLG office deleted the employment code because it mistakenly determined that the entity was not a government employer. These cases included:
· Two instances in which the words “company” or “corporation” appeared in the names of actual government entities.
· A pension plan for a failed employer administered under the auspices of the Federal Government.
· An entity managed by a trust board created under an intergovernmental agreement among the Federal, state, and local governments.
· A state branch of government.
The FSLG official responsible for initially identifying inaccurately coded entities stated that FSLG office personnel in Austin, Texas, would be responsible for verifying that the entities were not governmental and changing the employment codes. However, an FSLG official in Austin told us they changed the employment code based upon the initial review. Our review of the case files showed that personnel in Austin performed only limited research on each entity and generally did not verify the proper employment code.
The FSLG office prepared an action plan but did not have a documented methodology for performing research to confirm that organizations were not governmental or for changing the employment codes of entities that were non-profit organizations. In addition, the action plan did not include guidance on what research to conduct to verify that the entities were not governmental or provide instructions for ensuring that miscoded entities were accurately corrected. As a result, inaccurate information about the TE/GE Division customer base was input to the IRS computer system. An FSLG official informed us that a lack of sophisticated data matching software and a shortage of clerical support hampered their progress.
Efforts to develop a database to capture information about FSLG office customers were not included in the action plan
The FSLG office is also developing a supplemental database to meet its needs for capturing additional information about Federal, state, and local government customers. This database was adapted from the TE/GE Division’s Office of ITG Work Assignment Database and will capture FSLG office outreach efforts, work assignment data, and customer identification information when fully implemented. FSLG office management stated the database includes information related to FSLG office customers identified on the BMF and is structured so that it captures not only the primary government entities but also the related government subunits for each primary entity. In addition, data from the ongoing GID/BMF matching will be added to the database, when the matching is completed. Further, FSLG office management stated the database would include market segment results when they are available.
FSLG office management has not yet developed a formal time period for completing the database, documentation about the system requirements, or information on how the database will be kept current once it is online. FSLG office management informed us they did not include the development of this database in their Research Plan or associated documents because they only began working on the database in September 2003. However, this information was also not included in the January 2004 update to the action plan.
Corrective actions to our July 2002 report were not effective
In response to our prior report, the TE/GE Division agreed to develop GE function operational plans that included the actions planned, individuals assigned, responsible management official(s), completion dates, expected results, and methods to monitor and report performance. The IRS reported that these corrective actions were completed as of May 16, 2002. The FSLG office developed its Research Plan in February 2003 and related action plans in May 2003, both of which occurred after the date that the IRS reported it completed the prior report’s corrective actions. However, as noted above, we determined that action plans related to identifying Federal, state, and local government customers did not always include all of the necessary information. Based on the reasons cited in this report, we believe it is important to repeat the recommendations from our prior report to develop adequate action plans.
The identification of FSLG office customers is a critical part of the FSLG office’s efforts to provide quality customer service as well as ensure compliance with Federal tax laws. To address noncompliance issues, while maximizing its limited resources, the FSLG office needs to identify its at-risk customers most in need of educational assistance to effectively communicate tax law changes, educate them, and, if necessary, undertake compliance checks and examinations. In addition, more accurate identification of its customers on IRS computer systems will help the IRS to identify filing noncompliance and effectively apply tax payments to the proper taxpayer accounts. Further, the identification of its customers is a critical step in the completion of a series of initiatives discussed in the FSLG Research Plan. By ensuring that as many government entities as possible are identified, the FSLG office will be better able to achieve its stated goals.
To provide a more structured approach for initiatives to identify FSLG office customers, the Director, GE, should:
1. Ensure action plans for each initiative include the:
· Specific actions, methodologies, and procedures for completing the initiative.
· Individuals assigned to complete each action.
· Management official(s) providing oversight.
· Scheduled completion dates, including milestone dates for each key activity.
· Expected results.
· Methods to monitor performance and prepare subsequent updates to action plans if the actions or methodologies are no longer viable.
Management’s Response: The Director, FSLG office, is preparing a detailed action plan that will include the application of employment codes to categorize employers as Federal, state, or local governments on IRS computer systems. A project leader has been selected whose primary assignment will be the identification of FSLG taxpayers and the completion of the steps in the plan.
2. Ensure that a survey is mailed to each government entity included on the U.S. Census database that was not on the BMF, to confirm the name and address of the entity and request its EIN (if applicable). In addition, the survey should request the entity to identify other government units that its employment tax and information return data are consolidated with for information reporting purposes.
Management’s Response: The Director, FSLG office, will continue to refine customer identification through matching data between the U.S. Census Bureau’s GID and the IRS’ computer systems and will conduct another survey to resolve identification errors with mismatched entities.
The overall objective of this review was to assess the Tax Exempt and Government Entities (TE/GE) Division’s efforts to accurately identify Federal, state, and local government customers. To accomplish our objective, we:
I. Determined all planned efforts or those initiated since the July 2002 Treasury Inspector General for Tax Administration (TIGTA) report to identify Federal, state, and local government customers.
A. Interviewed Government Entities function and Federal, State, and Local Governments (FSLG) office management to obtain information on all planned identification efforts or those initiated since July 2002.
B. Obtained any documentation (e.g., planning documents, action plans, status reports) related to efforts to identify FSLG office customers.
C. Determined if identification efforts are on schedule.
II. Determined whether TE/GE Division management took adequate corrective action in response to recommendations made to improve the process used to identify customers (related to the FSLG office) contained in the July 2002 TIGTA report.
A. Verified whether FSLG office planning documents and action plans obtained in Step I identify the management official(s) responsible for providing oversight.
B. Reviewed FSLG office plans obtained in Step I to determine whether they included specific actions needed to identify customers, individuals assigned to the project, completion dates, expected results, and methods to monitor performance.
III. Evaluated the progress of FSLG office efforts to accurately identify its customers.
A. Assessed whether FSLG office efforts have resulted in the identification of additional customers.
1. Determined if additional customers have been identified.
2. Determined how the Internal Revenue Service (IRS) updates its computer system to include newly identified customers.
B. Determined if the methodologies used to identify customers address inaccuracies in IRS systems.
1. Determined whether the IRS is using more timely data from the United States Census Bureau to identify government customers.
2. Determined how the IRS is correcting inaccurate employment codes on IRS computer systems.
3. Selected a random sample of 50 cases from the universe of 522 cases in which the IRS determined that the employment codes were incorrect, corrected the coding, and had documentation supporting the changes. We limited our random sample to 50 cases to more efficiently use Office of Audit resources.
4. Performed research using the Internet and/or the Integrated Data Retrieval System to determine if the IRS correctly determined that the entities were not Federal, state, or local government entities. We also determined whether the IRS input the correct revisions to the employment, Business Operating Division, and Business Operating Division client codes.
Daniel R. Devlin, Assistant Inspector General for Audit (Headquarters Operations and Exempt Organizations Programs)
Nancy Nakamura, Director
Jeffrey M. Jones, Audit Manager
Thomas Seidell, Senior Auditor
Michael McGovern, Auditor
Marjorie Stephenson, Auditor
Office of the Commissioner – Attn: Chief of Staff C
Deputy Commissioner for Services and Enforcement SE
Director, Communications and Liaison, Tax Exempt and Government Entities Division SE:T:CL
Director, Government Entities, Tax Exempt and Government Entities Division SE:T:GE
Director, Office of Federal, State, and Local Governments, Tax Exempt and Government Entities Division SE:T:GE:FSL
Chief Counsel CC
National Taxpayer Advocate TA
Director, Office of Legislative Affairs CL:LA
Director, Office of Program Evaluation and Risk Assessment RAS:O
Office of Management Controls OS:CFO:AR:M
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:
· Reliability of Information – Actual; 10 tax accounts affected (see page 2).
Methodology Used to Measure the Reported Benefit:
We selected a random sample of 50 cases from a universe of 522 tax accounts in which the Federal, State, and Local Governments office determined that the entities’ employment codes were incorrect, took action to correct the employment codes, and documented its actions. For the tax accounts selected, we researched Internal Revenue Service computer systems and/or the Internet to determine if the revised employment codes were accurate. Based on our research, we determined that the revised employment codes were not accurate for 10 of the 49 tax accounts sampled.
The response was removed due to its size. To see the response, please go to the Adobe PDF version of the report on the TIGTA Public Web Page.