G A Ot United States General Accounting Office Washington, D.C. 20548 General Government Division B-240560 August31, 1990 The Honorable J. J. Pickle Chairman, Subcommittee on Oversight Committee on Ways and Means House of Representatives Dear Mr. Chairman: On May 9, 1990, you asked us to provide information about the timeli- ness of Internal Revenue Service (IRS) deposits of tax payments. As dis- cussed with the Subcommittee, this report deals specifically with the lost interest associated with delays in depositing individual income tax payments around April 15 each year, a time when iRs' 10 service centers receive a large number of such payments. Results in Brief Our analysis of deposit activity at the Cincinnati and Philadelphia Ser- vice Centers showed that the two centers averaged about 7 days to deposit about $6 billion in individual income tax payments received between April 15 and May 7, 1990. iRs officials said that they needed additional time to make deposits during this period because ms processes returns on a first in, first out basis, and it does not have the staff or equipment to more quickly handle the large volume of returns it receives in mid-April each year. One way IRs could reduce deposit times and increase government interest earnings would be to isolate and expedite the deposit of large tax payments. For example, our results for the two centers showed that tax payments of $5,000 or more accounted for about 70 percent of the potential interest earnings during this period, but less than 8 percent of the volume. We estimate the government could have earned additional interest of about $6.2 million if the two centers had deposited payments of $5,000 or more within 24 hours-the time ins officials believe is achievable if they focus on the large payments. Although our results cannot be generalized to ins' other eight centers, iRs officials said that similar results could be expected at any center. This report (1) discusses several approaches Rms might use to isolate and expedite the deposit of large tax payments and (2) provides some infor- mation on the advantages and disadvantages of each. ms needs to assess those approaches, and any others it might identify, and adopt one that strikes an appropriate balance between costs and benefits. Page 1 GAO/GGD904120 Depoits of Large Tax Paymnt B240560 When IRms receives tax returns it processes them on a first in, firu-st out basis. As returns enter a service center, employees open the returns and sort them according to the type of return (such as forms 1040, 1040A, or 1040EZ) and whether the returns involve a refund (called refund returns) or a tax payment (called remittance returns). Normally, ins requires service centers to deposit the payments with a bank within 24 hours of receipt, and the service centers generally meet. that require- nlent. During periods of heavier volume, however, the centers are allowed to take longer. The period of heaviest volume is at the end of the filing season because many taxpayers wait until then to file their returns. Between January 1 and May 4, 1990, IRs' 10 service centers received 106 million individual income tax returns, about 22 million of which were remittance returns. About 30 million of the 106 million total returns and about 12.7 million of the 22 million remittance returns arrived between April 13 and May 4. ins officials said that they do not have the staff and equipment to handle such a large volume of mail in a way that would allow the cen- ters to deposit the remittances within the typical 24-hour time frame. For that reason, the centers were given until May 7, 1990, to deposit all of the payments received from April 15 to May 4 that year. Objectives, Scope, and Our objectives were to (1) determine the lost interest earnings associatec with deposits of individual income tax payments sent to IRs between Methodology April 15 and May 7, 1990, and (2) identify options available to IRS to make faster deposits. To determine the lost interest earnings, we obtained ns computer files showing the deposit activity at the Cincinnati and Philadelphia Service Centers. We selected these two centers because we had staff working on a related assignment at those locations. Although our results are not sta- tistically representative of Is' other eight centers, IRs officials said that we could expect similar results at any of the others. In analyzing the files, we foc;seai on remittances that Iss received and deposited from April 15 to May 7, 1990. The remittances were associ- ated with returns filed on forms 1040, 1040A, and 1040EZ. For each remittance, we determined (1) how long it took ms to deposit the remit- tance and (2) how much lost interest income was associated with the deposit time. To calculate the amount of interest, we used the Treasury Tax and Loan Account rate-the rate at which the Department of the Treasury invests excess funds. During mid-April 1990, that rate was Page 2 GAO/GGD-0120 Depoat of Large Tax Payas B240560 about 8 percent. We also calculated the potential interest earnings if IRS shortened its deposit time to 24 hours and expedited the deposit of remittances over certain amounts. To identify steps IRs might take to expedite its deposits, we reviewed IRS' current procedures for depositing remittances, analyzed ms' manage- ment information reports on its deposit activity, and interviewed various officials involved in the deposit process. These officials worked in IRS' National Office in Washington, D.C.; IRs service centers in Cincin- nati and Philadelphia; and the Department of the Treasury's Financial Management Service, which is responsible for establishing monitoring agreements with financial institutions that act as depositaries for the government. We did our work from April through June 1990 in accordance with gen- erally accepted government auditing standards. Much of our analysis required us to work with data in IRs' computerized files of deposit activity. We did not determine if the data were accurate. However, we did review the data for reasonableness. Responsible IRS officials pro- vided comments on this report Those comments are presented on page 9. Deposit Times Computer files from the Cincinnati and Philadelphia Service Centers showed that the two centers received and deposited about 2.5 million Resulted in Lost remittances from April 15 to May 7, 1990. The deposits totaled about $6 Interest Earnings billion, with an average deposit time of about 7 days. Had ms been able to deposit the remittances within 24 hours-the time it takes to make deposits when receipts are at normal levels-we estimate that the gov- ernment would have earned an additional $8.8 million in interest. Table 1 shows the results for each service center. Table 1: Analysis of Deposit Activity at ] Two IRS Service Centers (April 15 - May 7, Number of Amount of Average 1990) Service center deposits deposits deposit time Lost interest Cincinnati 1.2 million $2.7 billion 6.9 days $3.8 million Philadelphia 1.3 million 3.3 billion 7.2 days 5.1 million Total 2.5 million $6.0 billion $8.8 mniliona aDoes not total due to rounding. ms National Office and service center officials acknowledged that they do not deposit remittances received around mid-April as fast as they do at other times of the year. They attributed the delays to a lack of staff Page 3 GAO/GGDdIZLW Depott of lrge Tax Pamws B-24e560 and equipment needed to open the mail, sort it, and process the remit- tances. They expressed doubt that it would be cost effective to purchase additional equipment and hire the additional staff necessary to meet a once-a-year surge in receipts. IRsplans to upgrade its remittance processing equipment in 1993. Although that upgrade may alleviate some of the deposit delays, it will not allow the centers to make end-of-filing-season deposits within 24 hours. According to a Cincinnati Service Center official, the planned upgrade might reduce the average deposit time from 7 to 5 days. Over the longer term, IRs is exploring the feasibility of receiving tax pay- ments electronically. Currently, is' electronic filing program is almost exclusively limited to tax returns with refunds. Although lRs plans to have electronic payment of taxes available to all taxpayers by 1992, it does not expect to have substantial taxpayer participation within the next 4 or 5 years. IRs is also exploring the feasibility of allowing tax- payers to make their tax payments with credit cards. That too, in the long term, may affect IRs' deposit activity. Until such major changes take effect, however, we believe that IRs could reduce deposit times in the short term without purchasing additional equipment and with little change to existing procedures. Separate Mailing Although it may not be feasible for IRs to process all of the remittances it receives in mid-April within 24 hours, it should be possible to expedite Address for Large Tax the processing of large remittances-those that result in the greatest Payments Could Help potential interest earnings. Our analysis of IRS' computer files showed, for example, that payments of $5,000 or more accounted for about 70 IRS Reduce Deposit percent of the potential interest earnings ($6.2 million of the $8.8 million Times total), but only about 8 percent of the total number of remittances processed. (App. I contains a more detailed breakout of the remittances received by amount of payment.) is officials in the National Office and in the Cincinnati Service Center told us that if they knew which returns contained large payments they would be able to deposit those payments within 24 hours. To expedite the deposit of large payments, ms has to be able to identify envelopes containing those payments as soon as possible after receipt. It can then segregate those envelopes from the many others waiting to be opened, sorted, and processed and give them and their contents priority handling. That segregation could be facilitated if taxpayers making Page 4 GAO/GGD4120i Deposits of Lare Tax Paymeo B-240560 large payments sent their returns to a different address than other returns. Currently, IRs instructs taxpayers to send their returns to the particular service center that services their geographic area. IRsalso includes an envelope with the service center's address in the tax package that it mails to taxpayers before the start of each tax filing season. If IRs instructed taxpayers making large tax payments to mail their returns to the same service center but to a different address, those returns would come into the center in a different mail pouch from the rest of the returns. Center officials would then know which envelopes contain large tax payments and could expedite the handling of those envelopes and the deposit of those payments. The success of this procedure would require the cooperation of taxpayers in using the special address. There would be little risk to IRS, however, because the failure of taxpayers to cooperate would merely result in payments being sent to IRs the same way they are now. From the taxpayer's standpoint, being instructed to mail large payments to a different address could cause some confusion. To avoid confusion, IRS could identify taxpayers who are more likely to make large end-of- filing-season payments and send them a tax package that includes a spe- cially-addressed return envelope in lieu of the normal envelope. There would then be no need to provide separate instructions related to the mailing of large payments. Taxpayers would fill out and mail their returns just as they do now, but the special envelope would direct the return and payment to a different address than other returns. To identify taxpayers who would receive the specially-addressed enve- lope, IRS could research its computerized taxpayer accounts for tax- payers who have made large tax payments around April 15 in past filing seasons. One potential problem with this approach is that a taxpayer's filing history may not be a good indicator of whether the taxpayer will owe a large payment in the future. If not, ms could end up giving pri- ority attention to many returns that are received in the special enve- lopes that involve small payments. To test the predictability of a taxpayer's filing history, the Cincinnati Service Center sampled 100 taxpayers who made tax payments of $10,000 or more to the center between April 15 and May 7. Of the 100 taxpayers, 16 had made payments of $10,000 or more in April 1989, and 16 had made payments between $5,000 and $10,000. Of the other 68 taxpayers, 63 either made smaller payments or received refunds. In the Page 5 GAO/GGD-O.120 Depoits of Large Tax Payment B-240560 remaining five cases, the service center could not determine the tax- payer's 1989 filing record. Although Cincinnati's test results tend to show that filing history is not a good predictor, the results are not conclusive given the small sample. IRS plans to make a more extensive test later in 1990. us might be able to enhance the feasibility of this approach if it identifies all taxpayers who have a history of making large payments, even if they do not always make them at the end of the filing season, and sends each of those tax- payers a tax package containing the specially-addressed envelope. Because there are points during the filing season other than the very end when the volume of receipts causes IRs to deviate somewhat from the normal 24-hour deposit time, it might be useful to send specially- addressed envelopes to all taxpayers who have a history of making large payments no matter when they make the payments. The approaches discussed above involve the use of a special service center address. Instead of having large payments sent to a different address at the service center, however, Ras could decide to make use of a lock box, ' as it now does with estimated tax payments. Under that arrangement, taxpayers would be instructed to mail large payments to the lock box along with a payment voucher. Taxpayers would still send their returns to their designated service center and receive credit for the payment upon the center's receipt of the payment voucher. Although the use of a lock box would have some advantages in terms of mini- mizing IRS' work load, requiring taxpayers to send their payments and returns to different addresses could cause some confusion and heighten taxpayer anxiety. Any of the approaches discussed above would entail some additional costs. The additional costs include (1) the printing and distribution costs if IRS chooses to develop a tax package with a specially-addressed enve- lope for taxpayers making large tax payments, (2) the costs if IRS chooses to revise its instructions and/or redesign its tax payment voucher, and (3) the fees paid to banks if Is chooses to use a lock box. IRS could curtail some of these costs if it limits the changes to the Form 1040. Our analysis of tax payments of $5,000 or more made to the Cin- cinnati and Philadelphia Service Centers in conjunction with forms IA lock box is a postal rental box serviced by a conmreiaI bank where persom who owe the govern ment money are instructed to send their payments. The bank then proceses the payments and trans- fers the funds to Treasury's Federal Reserve Account, usually by the next business day. Page 6 GAO/GGD90-120 Deposits of Large Tax Payments B-240580 1040, 1040A, and 1040EZ showed that about 99 percent were remitted with a Form 1040. The approach IRs chooses could also affect the amount of additional interest the government earns. First, according to officials in IRs' National Office Accounting Branch, lock box depositaries might need 48 hours to process the number of payments shown by our results because current contractual agreements with the depositaries provide for 48- hour processing during high volume periods. We estimate that the interest earnings associated with expediting the deposit of payments of $5,000 or more would have decreased from $6.2 million to $5.3 million at the Cincinnati and Philadelphia centers if those deposits took 48 hours instead of 24 hours. Second, if IRs chooses to have service centers (instead of lock box depositaries) expedite the processing and depositing of large payments, work load considerations night delay the deposit of smaller payments. These delays would offset the gain in interest earn- ings by a relatively small amount. For example, we estimate that delaying the deposit of payments less than $5,000 at the two centers would have decreased interest earnings by about $237,000.2 Although each of the approaches discussed above has drawbacks, we believe, and IRS officials agree, that the potential interest earnings war- rant some action to reduce the deposit times discussed in this report. The officials indicated, in fact, that the potential earnings might be even greater than our assessment indicated. They pointed out that additional interest earnings would be possible if large payments made by tax- payers who apply for extensions to file were included among the expe- dited payments. 3 To see how much additional interest might be associated with extensions to file, we extracted data from IRs' computer- ized deposit files for the Cincinnati Service Center. The data showed that the center received 96,937 extensions to file from April 15 to May 7, 1990, of which 26,104 involved payments of $5,000 or more. The data showed also that IRS took an average of 8.1 days to deposit those large payments. We determined that if the large payments had been deposited in 24 hours, the government would have earned additional interest of 2 To compute this $237,000, we assumed that for each day of interest saved by expediting deposits of S5,000 or more IRS would incur an additional day of interest expense for the average deposit of less than $5,000. In the Cincinnati and Philadelphia Service Centers, the average deposits of less than $5,000 were $812 and $846, respectively. 3 Taxpayers who apply for extensions must still pay their taxes by April 15 in order to avoid penalty charges. Page 7 GAO/GGD-90-120 Deposits of Large Tax Payments B-240660 about $1.3 million. If the deposits had been made in 48 hours, the addi- tional interest would have been about $1.1 million.4 IRS officials also expressed doubt that IRs could implement a new approach before the 1992 filing season because of the time needed to make the type of changes discussed above. ms receives a large volume of tax remittances around nmid-April each year as taxpayers file their individual income tax returns in time to meet the April 15 filing deadline. Because of the heavy volume, IRS is not able to deposit tax payments within the 24 hours it normally takes to make deposits during less busy times. Our analysis of the deposits made by two service centers in April 1990 showed that the government loses substantial interest earnings because of the delays. IRs is considering certain procedural changes, like allowing taxpayers to pay electronically or with credit cards, that could significantly change IRS' deposit activity in the long term. Until then, however, IRS needs to consider steps it can take to reduce the amount of lost interest earnings. One step IRS could take would be to isolate large remittances for priority handling by having them sent to a different address than other pay- ments. Our inquiries identified several possible ways IRs could do that. Although each of those ways has drawbacks, some more than others, none of those drawbacks would appear to offset the potential benefits. To reduce the time it takes to deposit large tax payments and increase Recormmendation to the government's interest earnings, IRS should assess various options, the Commissioner of including the ones discussed in this report, and adopt one that strikes an Internal Revenue appropriate balance between adnunistrative costs, taxpayer impact, and potential benefits. That assessment should be completed in time to allow IrS to implement the adopted approach in time for the filing season beginning January 1, 1992. 4IRS officials said they were planning changes to the extersion-to-file procedures that may, if imple- mented, predude the need for service centers to handle large payments associated with extensions. Upon implementation, extensions will be automatic, thus eliminating the need for taxpayers to file applications and reducing the volume of mail coming into service centers at the end of the filing season by 5 or 6 million pieces. Taxpayers getting extensions would still be required to pay their taxes by April 15, but IRS is considering having taxpayers send those payments to a lock box rather than to a service center. Page 8 GAO/GGD-9120 Deposits of Large Tax Payments B-240560 We discussed the results of our work with IRs' Assistant Commissioner for Returns Processing and members of his staff. They agreed with the facts in our report and agreed that the revenue implications warranted further IRsstudy of various options to expedite large deposits. They said that the most feasible approach might be to require that large payments be made electronically once that capability is available to taxpayers. As for the approaches discussed in this report, they noted that (1) the option that involves preidentifying taxpayers who can be expected to make a large payment at the end of a filing season might not be feasible, as indicated by the small sample done in Cincinnati, and (2) they were not confident that taxpayers or their representatives would comply with any instruction to send large payments to a different address or would use any specially-addressed envelope. They pointed out that most tax- payers with such a large tax payment would probably be using a tax practitioner and that IRS has historically had problems getting practi- tioners to use the pre-addressed labels and envelopes that IRS now includes in tax packages. Despite those potential problems, the Assistant Commissioner acknowledged that rRs should do the kind of cost-benefit analysis called for in our recommendation. As arranged with the Subcommittee, unless you publicly announce its contents earlier, we plan no further distribution of this report until 30 days from the date of issuance. At that time, we will send copies to the Commissioner of Internal Revenue and other interested parties. The major contributors to this report are listed in appendix II. Please contact me on 275-6407 if you or your staff have any questions about the report. Sincerely yours, Jennie S. Stathis Director, Tax Policy and Administration Issues Page 9 GAO/GGD-90-120 Deposits of Large Tax Payments Contents Letter 1 Appendix I 12 Analysis of Deposit Activity at Cincinnati and Philadelphia Service Centers Appendix II 13 Major Contributors to This Report Tables Table 1: Analysis of Deposit Activity at Two IRS Service 3 Centers (April 15 - May 7, 1990) Table 1.1: Analysis of Deposit Activity by Amount of 12 Deposit at Cincinnati Service Center (All 1040/ 1040A/1040EZ) Table 1.2: Analysis of Deposit Activity by Amount of 12 Deposit at Philadelphia Service Center (All 1040/ 1040A/1040EZ) Abbreviation IRs Internal Revenue Service Page 10 GAO/GGD-90-120 Deposits of Large Tax Paymento BLANK PAGE Pagqe 11 GAO/CG-JsW12 Depotib of Lare Tax lay Appendix I Analysis of Deposit Activity at Cincinnati and Philadelphia Service Centers Table 1.1: Analyis of Deposit Activity by Amount of Deposit at Cincinnaf Service Cumulative Center (All 1040/1040A;1040EZ) Percent of Percis Number of totea Lost interest I Deposit amount deposits depotits earnings Atra $10,000 or more 41,048 3.5 $2.089.971 9,000 or more 46,088 3.9 2.159.218 8,000 or more 52.545 4.4 2.239.243 7,000 or more 60,737 5.1 2.327.676 6,000 or more 71,826 6.1 2.432.734 5,000 or more 87,685 74 2.557.711 4,000 or more 110,952 9.3 2.705.444 3,000 or more 148,346 12.5 2.885.973 2,000 or more 219,186 18.5 3,125.405 1,000 or more 384,529 32.4 3.436.708 1or more 1.186.693 100.0 3.760.620 1 "Estimated interest earnings it IRS deposited each remittance within 24 hours and the interest rate - 8 percent Source: IRS computer hiles of deposit activity from April 15, 1990. to May 7. t990 Table 1.2: Analysis of Deposit Activity by l Amount of Deposit at Philadelphia CumUlative Service Center(All 1040/1040A/1040EZ) Percent of PeroaSl Number of total Lost interest I Deposit amount deposits deposits earnmings' sayv $10,000 or more 50,200 38 $3,074.850 9.000 or more 56,565 43 3,167.300 8,000 or more 64,448 4.9 3.269.664 7.000 or more 74.632 5.7 3.384.583 6.000 or more 88,141 6.7 3,516,213 5,000 or more 107,240 8.2 3.672,468 4,000 or more 135.044 10.3 3.856.572 3,000 or more 179,228 13.7 4,078.307 2.000 or more 261,185 20.0 4.360.486 1,000 or more 447,309 34.2 4.716.887 1 or more 1,306,183 100.0 5,073.104 1 aEstimated interest earnings if IRS deposited remittances within 24 hours and the interest rate was percent. Source: IRS computer tiles of deposit activity from April 15. 1990. to May 7. 1990 Page 12 GAO/GGD-90-120 Deposits of Large Tax Pay Appendix II Major Contributors to This Report General Goverunment General Government lDavid J. Attianese, Assistant Director, Tax Policy and Administration Lssues Division, Washington, Joseph T. Valonis, Evaluator D.C. Cincinnati Regional Robert I.Lidman, Evaluator-in-Charge Cincinnati Regional Michael J. Enriquez, Evaluator Office Barbara L.Centers, Evaluator Michael W. Hoffman, Advisor Philadelphia Regional Donald R.White, Evaluator Office San Francisco Arthur L. Davis, Evaluator Regional Office (268449) Page 13 GAO/GGD90-120 Deposits of large Tax Payr
Tax Administration: IRS Needs to Assess Options to Make Faster Deposits of Large Tax Payments
Published by the Government Accountability Office on 1990-08-31.
Below is a raw (and likely hideous) rendition of the original report. (PDF)