1) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests October 2004 Reference Number: 2005-40-006 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. 

2) DEPARTMENT OF THE TREASURY WASHINGTON, D.C. 20220 INSPECTOR GENERAL for TAX ADMINISTRATION October 27, 2004 MEMORANDUM FOR COMMISSIONER, WAGE AND INVESTMENT DIVISION FROM: Gordon C. Milbourn III Acting Deputy Inspector General for Audit SUBJECT: Final Audit Report - Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests (Audit # 200440017) This report presents the results of our review of the processing of tax returns in the Internal Revenue Service’s (IRS) Earned Income Tax Credit (EITC) concept tests. The overall objective of this review was to determine whether tax returns included in the IRS’ EITC concept tests were processed accurately. This is the fourth in a series of audits we have conducted to monitor the IRS’ testing of its long-term EITC vision.1 This audit focused on the certification of the qualifying child residency and filing status tests. In summary, the IRS closely monitored the implementation of the concept tests. Steps were taken to assess whether policies and procedures were being timely, accurately, and consistently followed. These monitoring efforts allowed the IRS to identify potential problems and take corrective action before the results of the tests were adversely affected. The success of the IRS’ concept tests also depends in large part on its ability to properly identify the taxpayers in the tests, hold and/or release the proper amount of EITC-related refunds, and forward the tax returns to the proper IRS function for validation. These key processes were functioning properly. In addition, the IRS was consistently accepting taxpayer documentation and making accurate validation decisions. 1 Management Controls Over the Proof of Concept Test of Earned Income Tax Credit Certification Need to Be Improved (Reference Number 2004-40-032, dated December 2003), The Risk of Inaccurate Computer Changes Can Be Reduced in Future Tests of the Earned Income Tax Credit (Reference Number 2004-40-089, dated April 2004), and The Statistical Sampling Method Used in the Earned Income Tax Credit Proof of Concept Test Appears Valid (Reference Number 2004-40-100, dated May 2004). 

3) 2 While the IRS properly handled and processed the tax returns included in the concept tests, some improvements can be made to future tests to reduce unnecessary burden on taxpayers. Unnecessary burden is any burden a taxpayer in the test experiences that is not inherent to participating in the tests. A review of a statistical sample of 379 taxpayers in the qualifying child residency and filing status tests showed that the IRS did not timely respond to taxpayers’ replies to IRS letters for 47 (12 percent) of the 379 taxpayers. Information gathered during the tests about why the goal was not met will provide the IRS with valuable data about the feasibility of responding to taxpayers within 30 days and identify ways returns can be worked more timely to ensure taxpayer burden is reduced as much as possible. In addition, an analysis of a statistical sample of 242 taxpayers in the filing status test indicated the IRS has additional internal data that could be used to modify the selection of taxpayers for future filing status tests. Our analysis indicated IRS information supported the taxpayer’s filing status for 82 (34 percent) of the 242 taxpayers. Using internal IRS address and filing history data when selecting taxpayers for future filing status tests could significantly reduce burden for taxpayers claiming the EITC by helping to ensure only those taxpayers with the highest risk of noncompliance are subjected to the tests. The IRS designed the tests using available information in an effort to include only those taxpayers most likely to have an error, and we agree that the burden experienced by taxpayers in the current concept tests was not preventable. The IRS’ analysis of the Fiscal Year (FY) 2004 concept tests, along with the insights provided through our audit, will provide information the IRS can use to reduce taxpayer burden in future concept tests. We recommended the Commissioner, Wage and Investment Division, ensure the Director, EITC, incorporates information gathered during the FY 2004 qualifying child residency and filing status concept tests into the planning and design of future EITC concept tests. Information related to increased or unnecessary burden should be used to ensure burden on taxpayers included in those tests is reduced as much as possible. Managements’ Response: The IRS agreed with our recommendation and indicated it has always planned to use what it learned from the first tests to improve future EITC initiatives. The IRS has contracted with an independent research firm to conduct an in-depth analysis of the qualifying child residency certification test. It will use the results of this analysis, as well as its internal analyses, to improve the design, including case selection, of future EITC tests. Management’s complete response to the draft report is included as Appendix VII. Copies of this report are also being sent to the IRS managers affected by the report recommendation. Please contact me at (202) 622-6510 if you have questions or Michael R. Phillips, Assistant Inspector General for Audit (Wage and Investment Income Programs), at (202) 927-0597. 

4) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests Table of Contents Background ............................................................................................... Page 1 Returns Were Properly Identified for Processing and Claims Were Properly Validated ........................................................................... Page 4 Online Monitoring Helped Identify and Correct Problems Before They Affected the Outcome of the Tests ....................................... Page 4 Fiscal Year 2004 Test Results Can Be Used to Reduce Taxpayer Burden in Future Tests.............................................................. Page 5 Recommendation 1 ........................................................................Page 10 Appendix I – Detailed Objective, Scope, and Methodology....................... Page 11 Appendix II – Major Contributors to This Report ....................................... Page 14 Appendix III – Report Distribution List ....................................................... Page 15 Appendix IV – Outcome Measures............................................................ Page 16 Appendix V – Earned Income Tax Credit Rules for Tax Year 2003........................................................................................... Page 18 Appendix VI – Filing Requirements for Single and Head of Household Filing Statuses ........................................................... Page 19 Appendix VII – Management’s Response to the Draft Report ................... Page 20 

5) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests Background The Earned Income Tax Credit (EITC) is a refundable credit designed to help move low-income taxpayers above the poverty level. However, the Internal Revenue Service (IRS) has historically experienced compliance problems with this credit. The IRS estimated that roughly 27 to 32 percent of the $31 billion in EITC claimed on Tax Year (TY) 1999 returns should not have been paid. A joint Department of the Treasury and IRS task force, formed in February 2002 to study the administration of the EITC, offered suggestions for improving EITC compliance. The IRS compiled these suggestions into a Concept of Operations outlining its future vision for the EITC Program. The first step in implementing this long-term vision is to test several of the vision’s key concepts. Beginning in Fiscal Year (FY) 2004, the IRS tested three of the main concepts contained in its long-term vision. • Certification of Qualifying Child Residency Requirements. • Verification of Filing Status. • Verification of Income. This is the fourth in a series of audits we have conducted to monitor the IRS’ testing of its long-term EITC vision.1 This audit focused on tests of the certification of the qualifying child residency requirements and the verification of filing status. We did not conduct reviews of the IRS’ test to verify taxpayers’ income because the test did not involve special processing of TY 2003 tax returns. We plan to conduct a review of that test as part of our FY 2005 audit coverage. Certification of qualifying child residency requirements The IRS envisions a prefiling certification program in which certain taxpayers would be required to verify that the 1 Management Controls Over the Proof of Concept Test of Earned Income Tax Credit Certification Need to Be Improved (Reference Number 2004-40-032, dated December 2003), The Risk of Inaccurate Computer Changes Can Be Reduced in Future Tests of the Earned Income Tax Credit (Reference Number 2004-40-089, dated April 2004), and The Statistical Sampling Method Used in the Earned Income Tax Credit Proof of Concept Test Appears Valid (Reference Number 2004-40-100, dated May 2004). Page 1 

6) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests child(ren) they are using to claim the EITC meet the EITC qualifying child requirements. These requirements can be found in Appendix V. The first test of the IRS’ certification concept involved 25,000 taxpayers2 who were asked to validate that the children they claimed for the EITC had lived with them for over 6 months of Calendar Year 2003. Taxpayers selected for the test were sent a letter in December 2003 telling them they had been chosen to test a new procedure. The letter explained the test and asked these taxpayers to submit documentation to support the residency requirements for any child(ren) claimed for the EITC when they filed their TY 2003 returns. If a taxpayer submitted documentation before his or her return was filed and the IRS validated the residency requirements, the refund was issued at the time the return was processed. If the taxpayer either did not submit the requested documents or submitted them with the return, the EITC portion of the refund was held until the taxpayer’s claim could be validated. In these cases, the IRS sent the taxpayer a series of letters informing him or her that a portion of the refund was being held and asking for documentation showing that the child(ren) lived with the taxpayer for over 6 months. Verification of filing status One of the concerns raised in the IRS’ 1999 EITC compliance study was with married taxpayers that circumvented the EITC income or claim limitations by incorrectly using the Head of Household or Single filing status to increase the amount of the EITC they received. The study showed that a portion of these taxpayers should have filed using either the Married Filing Jointly or Married Filing Separately filing status. Taxpayers who file Married Filing Jointly are ineligible for the EITC if their combined incomes exceed the EITC income limitation. The EITC is not allowed for taxpayers who file Married Filing Separately. The requirements for claiming the EITC in 2 The IRS reduced the number of taxpayers from the original volume of 25,000 for the FY 2004 test to exclude 313 taxpayers that no longer met the IRS’ selection criteria. Page 2 

7) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests 2003, including qualifying children and filing status, can be found in Appendix V. The IRS’ test to validate the filing status asked 36,000 taxpayers filing as Head of Household or Single to provide documentation supporting their filing status. When a taxpayer in the test filed a return using the Head of Household or Single filing status, the IRS held the EITC portion of the refund and sent a letter to the taxpayer. The letter informed the taxpayer that the IRS might examine his or her return for filing status, the EITC, or dependents. As with the qualifying child residency test, the taxpayer was sent a series of letters asking for documentation to support the filing status used on the return. In both the certification and filing status tests, the series of letters sent to taxpayers was interrupted if the taxpayer responded to the IRS at any time during the validation process. If the documents provided were adequate to support the taxpayer’s claim, the rest of the taxpayer’s refund was released. If the taxpayer’s claim was not supported, the IRS would contact the taxpayer for additional information. If the taxpayer still could not support the claim, the IRS disallowed the EITC, adjusted the return accordingly, and notified the taxpayer of the change. The IRS has informed us that the FY 2004 tests are the first in a series of tests to refine its concepts for the EITC Program. As such, it is important to note that information gathered from these tests is valuable regardless of whether it supports the IRS’ concept. The tests are providing the IRS with information it currently does not have and will enhance its ability to make sound decisions about the future of the EITC Program. This review was performed at the EITC Offices in Washington, D.C., and Atlanta, Georgia, and the Kansas City Compliance Site in Kansas City, Missouri, during the period February through July 2004. 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. Page 3 

8) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests Returns Were Properly Identified for Processing and Claims Were Properly Validated The success of the IRS’ concept tests depends in large part on its ability to properly identify the taxpayers in the tests, hold and/or release the proper amount of the EITC-related refunds, and forward the tax returns to the proper IRS function for validation. These key processes were functioning properly in both the qualifying child(ren) and filing status tests. For example: • • Taxpayers were properly identified as their TY 2003 returns were filed, allowing issuance of the notice that a portion of their refunds were being held. • The portion of a taxpayer’s refund attributable to the EITC was being held properly, and the remaining refund was correctly being released to the taxpayer when the return was filed. In addition, the EITC portion of the refund was being properly released once the IRS determined the taxpayer’s EITC claim was supported. • Online Monitoring Helped Identify and Correct Problems Before They Affected the Outcome of the Tests The IRS properly notified taxpayers in December 2003 that they would be part of the qualifying child residency test. The IRS consistently accepted taxpayer supporting documentation, made accurate validation decisions, and accurately computed the proposed tax changes when necessary. To ensure proper implementation of the tests, the IRS EITC Office designed a plan to assess whether policies and procedures were being timely, accurately, and consistently followed. Samples of taxpayers for each of the three tests were selected and monitored as TY 2003 returns were filed and processed. In addition, the IRS Office of Research and EITC Certification Unit at the Kansas City Compliance Site closely monitored implementation to identify any potential problems. The IRS’ monitoring efforts allowed it to identify potential problems and take corrective action before the results of the tests were adversely affected. Examples of problems identified include: • Taxpayer returns filed on paper that were part of the qualifying child residency test were not being properly coded during processing. Therefore, the returns and any Page 4 

9) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests supporting documentation were not being sent to the EITC Certification Unit for review. The IRS took additional steps to identify and forward all paper returns to the Unit. • Taxpayers in the qualifying child residency test who filed Married Filing Jointly as the spouse or secondary taxpayer on the return were not being properly identified for processing. The IRS’ analysis showed that only a minimal number of taxpayers had been affected when it discovered the condition and took actions to resolve the problem. • The IRS identified discrepancies between the two main databases it is using to capture the results of the tests. It has taken steps to compare these two databases and correct any discrepancies before the final analysis of the test results begins. Overall, the returns in the certification of qualifying child residency requirements and the filing status tests were properly identified for processing. In addition, refunds were properly held and released. However, some improvements can be made to future tests to reduce unnecessary burden on taxpayers. Unnecessary burden is any burden a taxpayer in the tests experiences that is not inherent to participating in the tests. Fiscal Year 2004 Test Results Can Be Used to Reduce Taxpayer Burden in Future Tests The long-term goal of the EITC Program is to increase compliance without adversely affecting participation or unduly increasing burden to taxpayers who claim the credit. As such, testing ways to reduce taxpayer burden is an integral part of the testing process. The IRS took steps to reduce the burden taxpayers in the FY 2004 tests would experience. It tested a very small portion of the total number of taxpayers that claim the EITC and held only those portions of the taxpayers’ refunds that were related to the credit. The test results will provide additional information and insight on ways future tests can be modified to further reduce or eliminate taxpayer burden. We made the following observations about how the information from the FY 2004 concept tests can be valuable in planning future tests of the IRS’ long-term vision of the EITC Program. Page 5 

10) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests Information about why responses were not provided to taxpayers within the IRS’ goal for the FY 2004 tests can provide insight on how to improve timeliness in future tests Once a taxpayer in either the certification or filing status test filed his or her TY 2003 return, the taxpayer began receiving a series of letters. These letters notified the taxpayer that the EITC portion of his or her refund was being held, notified the taxpayer of the possible tax adjustment if the EITC was denied, and asked for documentation to support either the qualifying child residency requirement or the taxpayer’s filing status. We found the IRS generally issued these letters timely. However, the IRS did not always respond to the taxpayer timely once documentation had been provided. The IRS has an examination standard to respond to taxpayers within 30 days of the receipt of a response to an IRS letter or notice. We reviewed a statistical sample of 379 taxpayers from the certification and filing status tests. The IRS did not respond to 47 (12 percent) of these taxpayers within the 30-day standard. Our analysis indicates these delays were attributable to two conditions. First, 15 of the 47 taxpayers sent documentation to the IRS before its examination reporting system became functional for Calendar Year 2004. This system is used to monitor IRS examination inventories and generate correspondence to taxpayers. Second, the IRS did not use the summary management reports available from its examination reporting system to properly control its inventory of test returns. Prior to early May 2004, the IRS was using detailed inventory reports to monitor the receipt and working of taxpayer replies to the various letters being sent. This made it very difficult to identify trends, such as aging inventory, and take actions to minimize the impact on taxpayers. The IRS has since started using summary management reports to monitor key aspects of its inventory for these tests. Delays in responding to taxpayers create an unnecessary taxpayer burden. Taxpayers experience uncertainty not knowing if the IRS has received submitted documentation. In addition, delays in reviewing that documentation and Page 6 

11) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests providing responses to taxpayers increase the amount of time they must wait to receive the remainder of their refunds. As of April 30, 2004, the IRS had accepted the documentation and closed the examinations without modifying the taxpayers’ EITC claims in 18 of the 47 cases we identified. These taxpayers had $44,540 in EITC held, and the IRS took an average of 53 days to close these examinations. We discussed our concerns with the IRS. The IRS agreed that reducing taxpayer burden was an important aspect of the concept tests. However, the IRS stated that meeting the 30-day standard is not mandatory, and a case is not considered untimely until 61 days have elapsed from the date of the taxpayer’s reply. One of the IRS Commissioner’s five-point EITC initiatives is to minimize burden and enhance the quality of communications with taxpayers by improving the existing examination process. Information gathered during the tests about why the 30-day standard was not met will provide the IRS with valuable data about the feasibility of responding to taxpayers within 30 days and identify ways returns can be worked more timely to reduce taxpayer burden as much as possible during the examination process. This information should be used not only when making future decisions about the EITC Program but also in planning future concept tests of its long-term vision. Should the IRS use the information related to timeliness of responses to taxpayers in planning for its FY 2005 concept tests, we estimate the IRS could prevent untimely responses to 7,529 taxpayers in the tests. This estimate is based on the 12 percent exception rate in our analysis and assumes the sample sizes selected for the FY 2005 tests are comparable to those in the FY 2004 tests. Changes in sample sizes would have an impact on the estimated number of taxpayers affected. Interim filing status test results indicate criteria used to select taxpayers for future tests can be modified to reduce taxpayer burden The filing status test was designed to test the IRS’ concept for reducing noncompliance among taxpayers who Page 7 

12) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests incorrectly file a tax return using the Head of Household or Single filing status to increase their EITC. A 1999 IRS study of EITC compliance indicated that approximately $2 billion in EITC claims was incorrectly paid to Head of Household and Single filers. These claims were attributable to 1.1 million out of 15.2 million taxpayers who claimed the EITC and used these two filing statuses.3 A description of the filing status requirements for Head of Household and Single can be found in Appendix VI. The filing status test was designed to assist the IRS in developing ways to effectively identify these taxpayers. The IRS selected a small cross-section of taxpayers who had claimed the EITC using the Head of Household or Single filing status and for which IRS data indicated the taxpayers might actually be married. Part of the test includes determining if using third-party data would improve the IRS’ ability to detect incorrect usage of filing statuses. The test would also help the IRS validate the accuracy of its internal taxpayer data. While the filing status test was not complete at the time of our audit, our analysis of a statistical sample of 242 taxpayers indicates that some portions of the IRS’ internal taxpayer data are likely valid and can be used to modify the selection of taxpayers for future tests. Testing the use of third-party data intended to validate and supplement IRS information has yet to be started. As a result, as the test progresses, additional information that would further refine the IRS’ analysis and our observations may come to light. However, based on our analysis, the IRS can rely on its internal taxpayer address and filing history information when selecting taxpayers for its FY 2005 filing status test. While the IRS’ address and filing history information appears valid, analyzing only one piece of this information would not adequately predict whether taxpayers are claiming the correct filing status. Rather, an in-depth analysis using all available information, including third-party data and address and filing history information 3 EITC Taskforce, Final Executive Steering Committee Meeting, June 28, 2002. Page 8 

13) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests on former spouses, would be necessary. For example, information may show the former spouse has relocated to a new address or has remarried. Using internal IRS and third-party data when selecting taxpayers for future filing status tests could significantly reduce burden for taxpayers claiming the EITC by helping to ensure only those with the highest risk of noncompliance are subjected to the tests. Of the 242 taxpayers in our filing status test sample, 167 were being examined by the IRS as part of the test.4 We reviewed IRS data for the taxpayers in the filing status test and their former spouses. Our analysis indicated IRS information for 82 (34 percent) of the 242 taxpayers in our sample appeared to support the taxpayer’s filing status. IRS examination results seem to support our analysis. As of May 28, 2004, the IRS had closed examinations on 44 of the 82 taxpayers we identified. All 44 examinations were closed without an adjustment to the taxpayer’s EITC. These 44 taxpayers had $79,166 in EITC delayed for an average of 10 weeks. The remaining 38 taxpayers in our sample were still being examined at the time we concluded our review. Their EITC claims totaled $63,175. In addition to having their EITC delayed, many of these taxpayers received a letter from the IRS telling them the IRS records show they may still be married, they are not eligible for the EITC, and they owe additional taxes as a result. For example, 1 taxpayer had EITC of over $1,400 held and was told he or she would owe the IRS over $1,000 in additional taxes unless he or she could show eligibility for the Single or Head of Household status. Overall, 46 of the 82 taxpayers we identified were told they might owe additional taxes totaling $23,333. While taxpayers were allowed the opportunity to dispute the proposed increase in tax by providing documentation supporting their filing status, the anxiety these taxpayers may have experienced could certainly be considered a burden. We estimate that 12,240 taxpayers filing Head of Household or Single and claiming the EITC would not be included in the FY 2005 filing status test if the IRS chooses to use our observations 4 Returns filed and processed as of May 28, 2004. Page 9 

14) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests in planning and designing the test. This estimate is based on the 34 percent exception rate in our analysis and assumes the sample size for the FY 2005 test is similar to that in the FY 2004 test. Increasing or decreasing the sample size in FY 2005 would have an impact on the estimated number of taxpayers affected. The IRS plans to carefully evaluate the results of this year’s filing status test and use that information in designing the FY 2005 test. However, the IRS reiterated that, although it designed the sample selection criteria to reduce unnecessarily subjecting taxpayers to the tests, the need to validate its internal data may lead to some taxpayers experiencing unnecessary burden in the current concept tests. We understand and agree that the burden experienced by taxpayers in the current concept tests was not preventable given the IRS’ purpose for the tests. However, the IRS should ensure it uses the information it gains from the tests, along with our audit results, to prevent as much burden to taxpayers as possible in future concept tests. Recommendation To avoid unnecessarily burdening taxpayers during future EITC concept tests, the Commissioner, Wage and Investment Division, should ensure: 1. The Director, EITC, incorporates information gathered from the FY 2004 qualifying child residency and filing status concept tests into the planning and design of future EITC concept tests. Information related to increased or unnecessary burden should be used to ensure burden on taxpayers included in those tests is reduced as much as possible. Management’s Response: The IRS indicated it has always planned to use what it learned from the first tests to improve future EITC initiatives. The IRS has contracted with an independent research firm to conduct an in-depth analysis of the qualifying child residency certification test. It will use the results of this analysis, as well as its internal analyses, to improve the design, including case selection, of future EITC tests. Page 10 

15) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests Appendix I Detailed Objective, Scope, and Methodology The overall objective of this review was to determine whether tax returns included in the Internal Revenue Service’s (IRS) Earned Income Tax Credit (EITC) concept tests were processed accurately. Sampling Criteria We selected a stratified statistical sample of 379 taxpayers as follows: • We randomly selected 137 taxpayers from the 24,687 taxpayers in the qualifying child residency test (95 percent confidence level, 10 percent expected error rate, and + 5 percent precision).1 • We randomly selected 242 taxpayers from the 36,000 taxpayers in the filing status test (95 percent confidence level, 20 percent expected error rate, and + 5 percent precision). Our expected error rates were based upon historical quality review information from the IRS Office of Performance Excellence. These historical rates were adjusted upward to account for the changes in the IRS work processes necessary to implement the concept tests and the results of our planning for this review. The full statistical sample was not reviewed for all tests. We used judgmental samples when initial testing indicated a full review may not be necessary or when time and information constraints prevented us from reviewing all 379 taxpayers. Unless otherwise noted, we selected our judgmental samples from our statistical sample of 379 taxpayers. To accomplish our objective, we: I. Assessed whether the IRS’ efforts to monitor the EITC concept tests were sufficient to timely identify significant problems and take corrective actions when necessary. We reviewed the IRS EITC Office’s Functionality Assessment Plan to determine what areas would be monitored and the extent of those monitoring efforts. We also contacted the IRS EITC Certification Unit Manager and obtained results of the IRS’ internal quality reviews. II. Determined whether Tax Year (TY) 2003 returns filed by taxpayers in the qualifying child residency and filing status tests were processed properly. We reviewed the Integrated Data Retrieval System (IDRS),2 the Earned Income Credit Proof of Concept 1 The IRS reduced the number of taxpayers from the original volume of 25,000 for the Fiscal Year 2004 test to exclude 313 taxpayers that no longer met the IRS selection criteria. 2 The IRS computer system capable of retrieving or updating stored information; it works in conjunction with a taxpayer’s account records. Page 11 

16) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests (EICPC) database,3 and Report Generation Software (RGS) System4 data. We also obtained information from the IRS on the status of the tests on a regular basis. A. Determined whether taxpayer notices were issued correctly, returns were properly identified when filed, non-EITC portions of refunds were properly released, and EITC portions were refunded to taxpayers in those cases in which the IRS accepted the EITC claim for a judgmental sample of 50 taxpayers.5 B. Verified whether any portion of the refund was held for a judgmental sample of 10 taxpayers6 from the 313 taxpayers eliminated from the qualifying child residency test.7 III. Determined whether TY 2003 returns filed by taxpayers in the qualifying child residency and filing status tests were being worked in the EITC Certification Unit according to established procedures. We reviewed the IDRS, the EICPC database, RGS data, and taxpayer and IRS examination documentation. A. Determined whether taxpayers were receiving the correct letters/correspondence and examinations were being worked timely for a statistical sample of 250 taxpayers.8 B. Determined whether determinations regarding the residency of qualified children and filing status were consistent and proper for a judgmental sample of 105 taxpayers.9 C. Validated the accuracy of the data in the RGS and the EICPC database for a judgmental sample of 209 taxpayers.10 D. Determined whether credits or deductions other than the EITC claimed on taxpayers’ returns were properly recalculated based upon the proposed filing status change to Married Filing Separately for a judgmental sample of 23 taxpayers11 in the filing status test. 3 One of the databases the IRS is using to capture the results of its examinations for the qualifying child residency and filing status concept tests. 4 The RGS System is used by the IRS in the examination process to compute taxes, generate examination correspondence and reports, and monitor Examination function inventories. 5 The first 25 qualifying child residency taxpayers that had filed a return as of March 18, 2004. The first 25 filing status taxpayers in the filing status test that had filed a return as of March 23, 2004. 6 Selected starting with the first taxpayer and using an interval of 31. 7 The IRS reduced the number of taxpayers from the original volume of 25,000 for the FY 2004 test to exclude 313 taxpayers that no longer met the IRS selection criteria. 8 Taxpayers in the overall sample of 379 that had filed a return as of April 19, 2004, and claimed the EITC with a qualifying child(ren). 9 Taxpayers in the overall sample of 379 that had filed a return as of April 19, 2004, and claimed the EITC with a qualifying child(ren), and IRS data indicated it had reviewed taxpayer documentation as of April 30, 2004. 10 We randomly selected these taxpayers from our stratified statistical sample of 379 taxpayers. 11 Taxpayers that had filed a return as of April 19, 2004, claimed the EITC with a qualifying child(ren), and claimed at least one credit other than the EITC. We reviewed at least 10 cases with a particular credit or all cases if there were less than 10. Some cases reviewed contained more than one credit. Page 12 

17) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests IV. Determined whether taxpayers in the qualifying child residency and filing status tests experienced unnecessary burden as part of the tests. We reviewed the IDRS and the EICPC database. We also reviewed the Income Tax Examination Changes (Form 4549-EZ) that were mailed to taxpayers. A. Determined whether information on the IDRS supported the taxpayer’s filing status for a statistical sample of 167 taxpayers12 in the filing status test. B. Determined whether abnormal or lengthy delays occurred between the IRS’ acceptance of the taxpayer’s EITC claim and the release of the EITC portion of the refund for a statistical sample of 66 taxpayers.13 12 Taxpayers who had filed returns that were processed as of May 28, 2004, had claimed the EITC with a qualifying child(ren), and were being examined as part of the test. 13 Taxpayers in the statistical sample of 379 that had had examinations closed as of April 30, 2004, without an adjustment to the tax return. Page 13 

18) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests Appendix II Major Contributors to This Report Michael R. Phillips, Assistant Inspector General for Audit (Wage and Investment Income Programs) Scott A. Macfarlane, Director Deann L. Baiza, Audit Manager John L. Hawkins, Lead Auditor Kathleen A. Hughes, Lead Auditor Sharon A. Buford, Senior Auditor Karen C. Fulte, Senior Auditor Areta G. Heard, Senior Auditor Sharla J. Robinson, Senior Auditor Sandra L. Hinton, Auditor Page 14 

19) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests Appendix III Report Distribution List Commissioner C Office of the Commissioner – Attn: Chief of Staff C Deputy Commissioner for Operations Support OS Deputy Commissioner for Services and Enforcement SE Deputy Commissioner, Wage and Investment Division SE:W Director, Office of Research, Analysis, and Statistics RAS Director, Office of Program Evaluation and Risk Analysis RAS:O Director, Compliance, Wage and Investment Division SE:W:CP Director, Earned Income Tax Credit, Wage and Investment Division SE:W:EITC Director, Earned Income Tax Credit Strategic Operations, Wage and Investment Division SE:W:EITC Director, Strategy and Finance, Wage and Investment Division SE:W:S Chief, Performance Improvement, Wage and Investment Division SE:W:S:PI Director, Reporting Compliance, Wage and Investment Division SE:W:CP:RC Chief Counsel CC National Taxpayer Advocate TA Director, Office of Legislative Affairs CL:LA Office of Management Controls OS:CFO:AR:M Audit Liaison: Senior Operations Advisor, Wage and Investment Division SE:W:S Page 15 

20) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests Appendix IV Outcome Measures This appendix presents detailed information on the measurable impact that our recommended corrective action will have on tax administration. These benefits will be incorporated into our Semiannual Report to the Congress. Type and Value of Outcome Measure: • Taxpayer Burden – Potential; 7,529 taxpayers may not receive a timely reply from the Internal Revenue Service (IRS) when they respond to an IRS inquiry in the Fiscal Year (FY) 2005 Earned Income Tax Credit (EITC) concept tests (see page 5). Methodology Used to Measure the Reported Benefit: The IRS provided us with the total population of taxpayers it selected for the FY 2004 qualifying child residency test (24,687 taxpayers) and the verification of filing status test (36,000 taxpayers). Because we decided during our audit planning to project our audit conclusions to each of the test populations and to the overall population of taxpayers in both tests, we used a stratified random sampling method to select a statistical sample from each test population. We selected a statistical sample of 137 taxpayers from the total population of 24,687 taxpayers in the qualifying child residency test.1 We determined that 90 of the 137 taxpayers had filed a Tax Year (TY) 2003 return claiming the EITC with qualifying child(ren) as of April 19, 2004. We determined that, as of April 30, 2004, the IRS had not responded to 18 of the 137 taxpayers (a 13 percent actual error rate) within 30 days.2 When this error rate is applied to the expected population for the FY 2005 qualifying child residency test, it is estimated that 3,209 taxpayers, + 5.62 percent, will be affected. The decrease in the precision from our original statistical sample is because the actual error rate was higher than originally expected (13 percent versus 10 percent). We selected a statistical sample of 242 taxpayers from the total population of 36,000 in the filing status test. We determined that 160 of the 242 taxpayers had filed a TY 2003 return claiming the EITC with qualifying child(ren) as of April 19, 2004. We determined that, as of April 30, 2004, the IRS had not responded to 29 of the 242 taxpayers (a 12 percent actual error rate) within 30 days. When this error rate is applied to the expected population for the filing status test, it is estimated that 4,320 taxpayers, + 4.08 percent, will be affected. The increase in the precision from our original statistical sample is because the actual error rate was lower than originally expected (12 percent versus 20 percent). 1 The IRS reduced the number of taxpayers from the original volume of 25,000 for the FY 2004 test to exclude 313 taxpayers that no longer met the IRS selection criteria. 2 The IRS has an examination standard to respond to a taxpayer’s reply within 30 days of the receipt of that reply. Page 16 

21) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests Combining the results from both of our statistical samples, it is projected that 7,529 taxpayers may receive untimely responses from the IRS during the FY 2005 EITC concept tests. This estimate is based on an overall 12 percent error rate in our analysis and assumes that the sample sizes selected for the FY 2005 tests are comparable to those in the FY 2004 tests. Changes in sample sizes would have an impact on the estimated number of taxpayers affected.3 Type and Value of Outcome Measure: • Taxpayer Burden – Potential; 12,240 taxpayers should not be subjected to the FY 2005 filing status test (see page 5). Methodology Used to Measure the Reported Benefit: First, we selected a stratified statistical sample (described above) from data provided by the IRS. From our statistical sample of 242 taxpayers in the filing status concept test, we determined that 167 of the 242 taxpayers who had filed a TY 2003 return claiming the EITC with qualifying child(ren) as of May 28, 2004, were being examined as part of the test. We determined that the IRS’ internal information would have likely supported the filing status claimed on the return for 82 of the 242 taxpayers (a 34 percent actual error rate). We estimate that 12,240 taxpayers would not be included in the FY 2005 filing status test if the IRS uses this information when selecting its test sample. This estimate is based on the overall error rate of 34 percent in our analysis and assumes that the sample size selected for the FY 2005 test is comparable to that in the FY 2004 test. Changes in sample size would have an impact on the estimated number of taxpayers affected.4 The precision for this estimate is + 5.95 percent. The decrease in the precision from our original statistical sample is due to the fact that our actual error rate was much higher than expected (34 versus 20 percent). 3 The IRS has indicated that it plans to test roughly the same number of taxpayers in FY 2005 as it tested in FY 2004. Therefore, our outcome is based on the assumption that the IRS will select the same size taxpayer population for both the qualifying child residency test (24,687) and the filing status test (36,000) that it selected for the FY 2004 tests. 4 This assumes that the IRS will select the same size taxpayer population (36,000) that it selected for the FY 2004 test. Page 17 

22) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests Appendix V Earned Income Tax Credit Rules for Tax Year 2003 Below is a general description of the qualifications taxpayers must meet to be eligible for the Earned Income Tax Credit (EITC). A detailed description of these rules can be found in the Internal Revenue Service publication Earned Income Credit (EIC) (Publication 596). Everyone must meet all of the following rules: • You must have a valid Social Security Number. • Your filing status cannot be “Married Filing Separately.” • You must be a United States citizen or resident alien all year. • You cannot file Foreign Earned Income (Form 2555) or Foreign Earned Income Exclusion (Form 2555-EZ). • Your investment income must be $2,600 or less. • You must have earned income. Rules to meet if you have a qualifying child (must meet all): • Your child must meet the relationship, age, and residency tests. • Your qualifying child cannot be used by more than one person to claim the EITC. • You cannot be a qualifying child of another person. Rules to meet if you do not have a qualifying child (must meet all): • You must be at least age 25 but under age 65. • You cannot be the dependent of another person. • You cannot be a qualifying child of another person. • You must have lived in the United States more than one-half of the year. Figuring and claiming the EITC (must meet both rules): • Your adjusted gross income (AGI)1 must be less than: o $33,692 ($34,692 for Married Filing Jointly) if you have more than 1 qualifying child. o $29,666 ($30,666 for Married Filing Jointly) if you have 1 qualifying child. o $11,230 ($12,230 for Married Filing Jointly) if you do not have a qualifying child. • Your earned income must be less than: o $33,692 ($34,692 for Married Filing Jointly) if you have more than 1 qualifying child. o $29,666 ($30,666 for Married Filing Jointly) if you have 1 qualifying child. o $11,230 ($12,230 for Married Filing Jointly) if you do not have a qualifying child. 1 AGI is a taxpayer’s gross income for the year adjusted for certain exclusions from income provided for in the Internal Revenue Code. Page 18 

23) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests Appendix VI Filing Requirements for Single and Head of Household Filing Statuses Below is a general description of the filing requirements for taxpayers to file as either Single or Head of Household. A detailed description of these requirements can be found in the Internal Revenue Service publication Exemptions, Standard Deduction, and Filing Information (Publication 501). Taxpayers can file as Single if on the last day of the tax year they are either: • Unmarried. • Legally separated from their spouse under a divorce or separate maintenance decree. Some single taxpayers may also qualify for another filing status, such as Head of Household, that would provide them with a lower tax. Taxpayers can file as Head of Household if they meet all of the following: • Unmarried or “considered unmarried” on the last day of the tax year. • Paid more than one-half the cost of keeping up a home for the tax year. • A “qualifying person” lived with you in your home for more than one-half of the year, with exceptions provided for temporary absences, or if the qualifying person is your dependent parent. To be “considered unmarried” for Head of Household, you must meet all of the following: • File a separate return from your spouse. • Pay more than one-half the cost of keeping up your home for the tax year. • Spouse did not live in your home during the last 6 months of the tax year. • Your home was the main home of your child, stepchild, or adopted child for more than one-half of the tax year or was the main home of your foster child for the entire year. • You are able to claim an exemption for the child except in certain situations involving custody agreements in which you have agreed not to claim the exemption. Page 19 

24) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests Appendix VII Management’s Response to the Draft Report Page 20 

25) Initial Results of the Fiscal Year 2004 Earned Income Tax Credit Concept Tests Provide Insight on Ways Taxpayer Burden Can Be Reduced in Future Tests Page 21