The Child Tax Credit Advance Payment Was Effectively Planned and Implemented, but a Programming Discrepancy Caused Some Overpayments

January 2004

Reference Number: 2004-40-042

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.

January 28, 2004

MEMORANDUM FOR COMMISSIONER, WAGE AND INVESTMENT DIVISION

FROM: Gordon C. Milbourn III /s/ Gordon C. Milbourn III

Acting Deputy Inspector General for Audit

SUBJECT: Final Audit Report - The Child Tax Credit Advance Payment Was Effectively Planned and Implemented, but a Programming Discrepancy Caused Some Overpayments (Audit # 200340046)

This report presents the results of our review of the Child Tax Credit (CTC) advance payment. The overall objective of this review was to determine if the Internal Revenue Service (IRS) effectively prepared for and implemented the CTC advance payment provision of the Jobs and Growth Tax Relief Reconciliation Act of 2003.

In May 2003, the Congress passed legislation that provided for substantial tax cuts. This legislation also included provisions to increase the CTC from $600 per child to a maximum of $1,000 per child, beginning with Tax Year 2003. The Congress also wanted the increased credit issued to qualifying taxpayers as an advance payment. This would enable taxpayers to receive the benefit of the increased credit during 2003, rather than having to wait to claim it when filing their tax returns in 2004. The advance payments were based on information reported on 2002 tax returns.

Overall, the IRS effectively planned for and implemented the CTC advance payment provisions. Planning was started in January 2003, and the computer programming requirements were developed prior to the legislation being signed in May 2003. In addition, the IRS developed a notice to be issued concurrently with the advance payments and created a toll-free telephone number that taxpayers could use to inquire about the status of their advance payments.

While the IRS did a commendable job of getting the advance CTC payments to taxpayers, we identified a computer programming discrepancy that erroneously calculated the CTC advance payment for some taxpayers who had claimed losses on their 2002 tax returns from either Profit or Loss From Business (Schedule C) or Profit or Loss From Farming (Schedule F). The programming logic did not include these losses when computing the earned income amount used to calculate the CTC advance payments. By miscalculating earned income in some cases, the IRS computed a larger advance payment than was appropriate. We estimate at least 91,000 taxpayers received overstated advance payments totaling more than $39 million.

To ensure the IRS was aware of the scope and causes of the programming problem, we issued a memorandum to the Commissioner, Wage and Investment (W&I) Division, on August 29, 2003. In the memorandum, we outlined the information we developed on the CTC overpayments to taxpayers filing a return with Schedule C losses. Our memorandum is included as Appendix V.

Management’s Response to Memorandum #1: The Commissioner, W&I Division, agreed that some taxpayers received an overpayment of the advance CTC payment. He stated the IRS was aware that the computation did not consider some self-employment losses, and the Department of the Treasury’s Office of Tax Policy approved the programming requirements prior to the advance payment being implemented. The complete response to the memorandum is included as Appendix VI.

Although the IRS is not taking corrective action to recover the overpayments, we are not taking issue with this decision because some of the overpayments may be recovered when taxpayers file their 2003 tax returns and net out the advance payment amounts from the CTCs claimed on those returns. However, we do believe that the IRS can strengthen its controls to help prevent these problems in the future.

To reduce programming discrepancies for future advance payments, we recommended that the Commissioner, W&I Division, formalize the process used to develop programming requirements for future advance payments. This process should ensure that significant business decisions made because of data limitations or computer processing limitations are analyzed to determine their potential impact.

Management’s Response to the Draft Report: The IRS appreciated our recognition of the numerous complex issues that it overcame to successfully implement the CTC advance payment provisions. However, management did not agree with our recommendation or the reported outcome associated with it. The IRS has a process that considers programming requirements for implementing new tax legislation. Management believes that the IRS process for implementing new legislation is sound and that further requirements are not necessary. In addition, the IRS anticipates that the majority of overpayments will be reconciled when the taxpayers file their 2003 tax returns. Management’s complete response to the draft report is included as Appendix VII.

Office of Audit Comment: We recognize, as described in the report, that overall the IRS did a commendable job with implementing the CTC advance payment process. However, we still believe that had the IRS fully considered losses from self-employment income when computing the advance CTC payments, the number and amount of overpayments would have been significantly reduced. Further analysis of the impact of omitting losses from self-employment from the advance payment calculation should have revealed the potential for substantial overpayments. We believe these overpayments could have been reduced without significant additional use of resources.

More importantly, because decisions had already been made on the CTC payments at the time of our review, our recommendation was to formalize the process used to develop programming requirements for future advance payments, and this was not meant to apply to all new tax legislation. Because advance payments represent unique, high-profile, and potentially high-risk undertakings, we believe that a formal process specifically addressed to their implementation is appropriate. The conditions identified in this report serve to support the recommendation that improvements can be made for future events.

Regarding the potential value of the overpayments, we recognize that the IRS will recover a portion of these erroneous payments as taxpayers file their tax returns, but no one is certain as to the extent of the recovered payments. As stated in the report, taxpayers are not required to repay excess advance payments if they qualify for a lesser, or no, CTC amount for 2003. Because of the uncertainty of the amount of overpayments that will be collected, we classified the entire $39 million in potential overpayments as Inefficient Use of Resources because funds were sent to taxpayers who did not qualify for the payments. We will perform additional testing during the 2004 Filing Season to identify the amount of the payments that were recovered and to calculate the actual amount of improper payments.

The IRS response categorizes the potential outcome as Funds Put To Better Use rather than Inefficient Use of Resources. We contacted the IRS and discussed the reclassification of the outcome measure, but the IRS does not agree with either outcome measure.

While we still believe our recommendation and potential outcome measure are valid and worthwhile, we do not plan to elevate our disagreement to the Department of the Treasury for resolution.

Copies of this report are also being sent to the IRS managers who are 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.

Table of Contents

Background

The Child Tax Credit Advance Payment Provisions Were Effectively Planned for and Implemented

A Programming Discrepancy Resulted in Some Taxpayers Receiving an Overstated Advance Payment

Recommendation 1:

Appendix I – Detailed Objective, Scope, and Methodology

Appendix II – Major Contributors to This Report

Appendix III – Report Distribution List

Appendix IV – Outcome Measures

Appendix V – Memorandum #1: Results of Profit or Loss From Business (Schedule C) Sample

Appendix VI – Management’s Response to Memorandum #1

Appendix VII – Management’s Response to the Draft Report

Background

In May 2003, the Congress passed legislation that provided for substantial tax cuts. This legislation also included provisions to increase the Child Tax Credit (CTC) and to have the increase issued to qualifying taxpayers as an advance payment. This would enable taxpayers to receive the benefit of the increased credit during 2003, rather than having to wait to claim it when filing their tax returns in 2004.

The CTC began in 1998 as a credit for taxpayers with qualifying children. To qualify, the taxpayer’s child must be under age 17, be a citizen or resident of the United States, be claimed as the taxpayer’s dependent, and have a qualified relationship to the taxpayer. The May 2003 legislation increased the amount of the CTC from $600 per child to a maximum of $1,000 per child, beginning with Tax Year 2003. The legislation also required that the Internal Revenue Service (IRS) issue advance payments of the increased CTC “as rapidly as possible and, to the extent practicable, before October 1, 2003.”

To determine if a taxpayer was eligible for an advance payment, the IRS used the information reported on 2002 tax returns. For each eligible taxpayer, the IRS credited the taxpayer’s account with the advance payment amount and sent the taxpayer a notice advising him or her of the amount to be refunded. If the taxpayer had no outstanding debts, a refund check for the credit amount was issued to him or her.

By law, the IRS can issue the advance CTC payments only through the end of 2003. After that, eligible taxpayers who have not received the advance payment may be able to claim the increased credit when filing their 2003 tax returns in 2004. Also, taxpayers who received an advance payment will be required to deduct the advance payment amount from any CTC claimed on their 2003 tax returns.

This review was performed at the IRS National Headquarters and the New Carrollton, Maryland, office during the period June through October 2003. 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.

The Child Tax Credit Advance Payment Provisions Were Effectively Planned for and Implemented

The IRS effectively planned for and implemented the CTC advance payment provisions. As of September 30, 2003, the IRS had issued credits totaling $15 billion to approximately 25 million taxpayers, with the majority of these being issued in the first 3 weeks of the program.

The IRS began preparations in January 2003 by establishing a multifunctional working group to begin planning for the possibility that proposed tax legislation would include an advance payment provision. This working group included representatives from the major IRS functions affected by the advance payment. The IRS also coordinated with external stakeholders, such as the Financial Management Service (which actually issued the refund checks) and the Department of the Treasury.

When the IRS began developing the computer programming requirements, it faced a difficult task. Because the legislation had not been passed when the planning was started, the requirements were developed before the final form of the legislation was known. Accordingly, the IRS had to be prepared to make various changes as time progressed and passage of the legislation came closer. After the legislation was signed, the programming was properly tested to verify that it matched the programming requirements. However, the scope of the testing did not include verifying that the requirements matched the intent of the legislation, only that the programming output matched the requirements. The IRS implemented the advance payment provision in a little more than a month’s time after the legislation was signed.

While this advance payment was similar to the one issued during 2001, the fact that it was an advance payment of a credit rather than a tax reduction created increased complexity. The computations were very involved and required in-depth analyses of many items from the tax return. The requirements for the computer programming were revised several times prior to being finalized. Apart from the issue discussed in the next section, we determined that the advance payments were accurately computed and that all eligible taxpayers received an advance payment.

To inform taxpayers about their advance payments, the IRS developed an informational notice to be issued concurrently with the payments and contracted with an outside vendor to print and mail the notices to taxpayers. This notice included the reason the taxpayer was receiving a refund, the amount of the refund, the number of qualifying children used to calculate the amount, and a web site address and toll-free telephone number to call if the taxpayer had questions. The IRS also established an automated toll-free telephone application to provide the amount of a taxpayer’s advance payment and the date it was scheduled to be issued. We verified that the notices were accurate and informative and that the toll-free telephone application provided accurate information to taxpayers.

A Programming Discrepancy Resulted in Some Taxpayers Receiving an Overstated Advance Payment

While the IRS did a commendable job preparing and implementing the CTC advance payment process, a computer programming discrepancy caused some taxpayers to receive a larger advance payment amount than they were entitled to receive. We estimate that at least 91,000 taxpayers received over $39 million in excess advance payments.

An omission in the programming requirements caused the overpayments. The advance payment should have been limited if the earned income on the tax return was below a certain amount. Earned income is generally defined as income from wages and profit or losses from self-employment. When calculating the advance payment amount based on the 2002 tax return information, the programming logic did not always include self-employment losses claimed on either Profit or Loss From Business (Schedule C) or Profit or Loss from Farming (Schedule F), as required by law when computing earned income. By erroneously overstating earned income in some cases, the programming computed a larger advance payment than was appropriate.

The President’s Management Agenda requires agencies to track and report the amount of erroneous payments made as part of their programs, with the goal of reducing erroneous payments. In accordance with this goal, the IRS should have ensured that the advance payment amounts were as accurate as possible, given the information available to compute the amount.

We reported this programming discrepancy related to taxpayers claiming losses on Schedule C to the Commissioner, Wage and Investment (W&I) Division, via a memorandum issued August 29, 2003 (see Appendix V). We subsequently notified the Commissioner, W&I Division, regarding the related overpayments involving taxpayers claiming Schedule F losses.

The Commissioner’s response to our memorandum (see Appendix VI) agreed that some taxpayers received an overpayment of the advance CTC payment. The response also indicated that the IRS intentionally chose to omit some self-employment losses from the CTC advance payment calculations and that the computer programming requirements were approved by the Department of the Treasury’s Office of Tax Policy. The reason for omitting the losses was that the information needed to compute the true earned income was not available from the return information.

We do not agree that the information needed to more accurately compute earned income was not available from the 2002 tax returns. The information from the Schedules C and F was readily available and could have been used for the earned income computation. In fact, the IRS used this information during return processing to verify the amount of the CTC claimed.

The IRS has chosen not to take corrective action to recover the overpayments resulting from the programming discrepancy. As the Commissioner indicated in the response to our memorandum, some of the overpayments may be recovered when taxpayers file their 2003 tax returns and net out the advance payments received from the CTCs claimed on the returns. However, the law does not require taxpayers to repay excess amounts if they qualify for a lesser CTC amount (or no CTC amount) on their 2003 tax returns. For example, if a taxpayer received an erroneous advance payment of $400 based on his or her 2002 tax return, he or she would not be required to repay this amount if he or she did not have any qualifying children on his or her 2003 tax return. Accordingly, we cannot determine at this time how much of the overpayments may be recovered, although we do plan to perform additional work during our 2004 Filing Season review to quantify this amount.

Because the IRS intentionally decided to omit self-employment losses in the advance payment computation, and this omission caused a substantial overpayment, it appears that an evaluation of the impact of this decision was not made. Accordingly, we believe that the decision-making process related to developing the advance payment programming requirements could be improved. Because an undetermined amount of the overpayments may be recovered when taxpayers file their 2003 tax returns, we are not taking issue with the IRS’ decision to not take corrective action. However, we believe that the IRS can strengthen its controls to help prevent similar problems in the future.

Recommendation

To reduce programming discrepancies for future advance payments, the Commissioner, W&I Division, should:

1. Formalize the process used to develop programming requirements for future advance payments. This process should ensure that significant business decisions made because of data limitations or computer processing limitations are analyzed to determine the potential impact of these decisions. This analysis should include a documented evaluation of potential outcomes, including the numbers of taxpayers that may be affected.

Management’s Response: The IRS did not agree with this recommendation or the reported outcome associated with it. The IRS has a process that considers programming requirements for implementing new tax legislation. Programming requirements are driven by both data availability and the practicality of retrieving and interpreting the data. The level of complexity, the overall desired outcome, and timeliness of actions are also factors that influence the IRS’ decision-making process.

IRS management believes that the IRS process for implementing new legislation is sound and that further requirements are not necessary. In addition, the IRS anticipates that the majority of overpayments will be reconciled when the taxpayers file their 2003 tax returns.

Appendix I

Detailed Objective, Scope, and Methodology

The overall objective of our review was to determine if the Internal Revenue Service (IRS) effectively prepared for and implemented the Child Tax Credit (CTC) advance payment provision of the Jobs and Growth Tax Relief Reconciliation Act of 2003.

To accomplish our objective, we:

I. Evaluated the programming requirements and testing of the programming for computing and posting the advance payments to determine if it was sufficient to ensure that all eligible taxpayers received an advance payment for the correct amount.

A. Determined if the relevant programming documentation had any logic problems or discrepancies, if any adverse account impacts would occur due to the advance payments, and if the IRS programmers had any concerns or issues related to the programming.

B. Determined if the testing of the programming for the advance payment was sufficient to ensure that material problems had been identified and corrected. This included determining if the proper testing process was followed and if problems identified during testing were analyzed, corrected, and retested.

II. Evaluated the IRS’ planned methods to inform the public about the advance payments. This was limited to the initial notices and the automated toll-free telephone application.

A. Determined if the notices issued concurrently with the advance payment checks were accurate and informative. We obtained copies of the initial notices and reviewed them for accuracy and readability. During our accuracy review, we identified a problem with the calculation of the advance payments related to self-employment losses. Based on this information, we expanded the scope of our audit to include
subobjective III.

B. Determined if the automated toll-free telephone application accurately and effectively informed taxpayers about their advance payments. We obtained copies of the script used by the automated telephone system and analyzed it for effectiveness. We also placed test calls to verify the accuracy of the information being given.

III. Evaluated the advance payment amounts posted to taxpayer accounts for accuracy and determined if all eligible taxpayers received advance payments as required by law. For all tests, we used data obtained from the IRS Individual Return Transaction File or the original data files the IRS generated to post the advance payments to taxpayer accounts. We took appropriate data validation steps for all data extracts, including reviewing log files, matching record counts, and matching the data back against the source file.

A. Reviewed statistically valid samples of 322 returns with a loss on Profit or Loss from Business (Schedule C) and 379 returns with a loss on Profit or Loss from Farming (Schedule F) to quantify the effect of omitting self-employment losses from the advance payment computation. We had first identified this problem in subobjective II.A. We manipulated our data to isolate the returns most likely to contain an overpayment and identified populations of 89,909 returns with Schedule C losses and 27,862 with Schedule F losses. We used a confidence level of 95 percent and a precision of + 5 percent for both samples. The expected error rate for the Schedule C sample was 70 percent, with the actual error rate being 84 percent. For the Schedule F sample, the expected and actual error rates were both 59 percent. The difference in the error rates between the two samples was due to different return characteristics.

B. Reviewed a statistically valid sample of 249 returns to verify the accuracy of the posted advance payment amounts. We selected this sample from an overall population of approximately 24 million advance payment posting records obtained from the IRS. We used a confidence level of 95 percent, a precision of + 3 percent, and an expected error rate of 2 percent for this sample.

C. Selected and reviewed a statistically valid sample of 200 returns to determine if all eligible taxpayers received an advance payment as required by law. We selected this sample from the population of approximately 5 million accounts with a CTC claimed on the return, but with no advance payment issued by the IRS. We used a confidence level of 95 percent, a precision of + 3 percent, and an expected error rate of 5 percent for this sample.

Appendix II

Major Contributors to This Report

Michael R. Phillips, Assistant Inspector General for Audit (Wage and Investment Income Programs)

Scott Macfarlane, Director

Gary L. Young, Acting Director

Richard Calderon, Audit Manager

Linda Bryant, Senior Auditor

Carola Gaylord, Senior Auditor

Steven Stephens, Senior Auditor

Appendix III

Report Distribution List

Commissioner C

Office of the Commissioner – Attn: Chief of Staff C

Deputy Commissioner for Services and Enforcement SE

Senior Advisor to the Deputy Commissioner for Services and Enforcement SE

Deputy Commissioner, Wage and Investment Division SE:W

Chief Counsel CC

National Taxpayer Advocate TA

Director, Office of Legislative Affairs CL:LA

Director, Office of Program Evaluation and Risk Analysis RAS:O

Office of Management Controls OS:CFO:AR:M

Audit Liaison: GAO/TIGTA Liaison, Wage and Investment Division SE:W:S:PA

Appendix IV

Outcome Measures

This appendix presents detailed information on the measurable impact that our recommended corrective action will have on tax administration. This benefit will be incorporated into our Semiannual Report to the Congress.

Type and Value of Outcome Measure:

· Inefficient Use of Resources – Potential; 91,000 taxpayer accounts received erroneous credit overpayments totaling over $39 million (see page 3).

Methodology Used to Measure the Reported Benefit:

We obtained the current year (Tax Year 2002) Internal Revenue Service (IRS) Individual Return Transaction File (RTF) records and developed specific criteria to select those returns on the file with the most chance of being affected by the programming discrepancy. The overall population of returns on the RTF was approximately 121 million. The specific criteria included all returns with a Profit or Loss From Business (Schedule C) or Profit or Loss From Farming (Schedule F) loss greater than $1,000 and an Additional Child Tax Credit claimed. We selected and reviewed two samples, one focused on returns with a loss on Schedule C and the other on returns with a loss on Schedule F. The population for the Schedule C sample was 89,909, and the population for the Schedule F sample was 27,862.

We used a 95 percent confidence level and a + 5 percent precision for both samples, although the expected error rate differed between the 2 samples due to the characteristics of the returns. The expected error rate for the Schedule C loss sample was 70 percent, and for the Schedule F loss sample the expected error rate was 59 percent. We reviewed 322 returns with Schedule C losses and 379 returns with Schedule F losses.

We determined the correct advance payment amount that should have been allowed if the self-employment loss (from Schedules C or F) on the return had been included in the computation. We subtracted this amount from the amount actually allowed by the IRS. The difference was the overpayment amount. We then projected the average overpayment amount from the sample to the overall population.

For the Schedule C sample, we identified 269 out of 322 returns (84 percent actual error rate) that had overpayments averaging $414. From this we projected 75,000 taxpayers with a Schedule C loss were overpaid approximately $31 million. For the Schedule F sample, we identified 222 out of 379 returns (59 percent actual error rate) had overpayments averaging $490. From this we projected 16,000 taxpayers with a Schedule F loss were overpaid approximately$8 million. When we combined the results of the 2 samples, we determined that 91,000 taxpayers received erroneous CTC advance payments totaling over $39 million.

Appendix V

Memorandum #1: Results of Profit or Loss From Business (Schedule C) Sample

Memorandum #1 was removed due to its size. To see the Memorandum #1, please go to the Adobe PDF version of the report on the TIGTA Public Web Page.

Appendix VI

Management’s Response to Memorandum #1

The response to Memorandum #1was removed due to its size. To see the response to Memorandum #1, please go to the Adobe PDF version of the report on the TIGTA Public Web Page.

Appendix VII

Management’s Response to the Draft Report

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.