Summary

Franchise Tax Board: Its Revenue From Audits Has Increased, but the Increase Did Not Result From Additional Time Spent Performing Audits March 1999 98118.2 The first copy of each California State Auditor report is free. Additional copies are $3 each. You can obtain reports by contacting the Bureau of State Audits at the following address: California State Auditor Bureau of State Audits 555 Capitol Mall, Suite 300 Sacramento, California 95814 (916) 445-0255 or TDD (916) 445-0255 x 248 OR This report may also be available on the World Wide Web http://www.bsa.ca.gov/bsa/ Permission is granted to reproduce reports. March 18, 1999 98118.2 The Governor of California President pro Tempore of the Senate Speaker of the Assembly State Capitol Sacramento, California 95814 Dear Governor and Legislative Leaders: As requested by the Joint Legislative Audit Committee, the Bureau of State Audits presents its audit report concerning the Franchise Tax Board’s (board) return on its investment in additional audit positions. This report concludes that although its revenue from audits has increased, the increase did not result from additional time performing audits. In fact, we found that revenues actually decreased from prior years in those areas where we would expect the board to assign new staff. One significant reason for this decrease is that the board did not spend additional hours performing revenue-generating audits. Respectfully submitted, KURT R. SJOBERG State Auditor CONTENTS Summary 1 Introduction 5 Chapter 1 An Increase in Audit Revenues Did Not Result From Additional Time Spent Performing Audits 9 Recommendations 18 Chapter 2 Internal Revenue Service Leads Contribute Significantly to Tax Assessments, but the Board Is Concerned That These Leads May Be Reduced in the Future 19 Recommendation 22 Chapter 3 The Franchise Tax Board’s Report on Audits Does Not Contain Sufficient Information to Assess the Return on Its Investments in Its Audits 23 Recommendation 25 Appendix 27 Response to the Audit Franchise Tax Board R-1 California State Auditor’s Comments on the Response From the Franchise Tax Board R-5 SUMMARY RESULTS IN BRIEF T he Franchise Tax Board (board) is one of the primary tax-collecting agencies in the State. For fiscal years 1990-91 through 1997-98, it collected an average of Audit Highlights . . . $20 billion in Personal Income Tax (PIT) revenues and $5 billion in Bank and Corporations (B&C) tax revenues, annually. To Our review of the Franchise increase the board’s audit revenues, the Legislature authorized Tax Board’s (board) budget 362 new audit positions for the board’s audit branch between augmentation for audit staff revealed the following: fiscal years 1992-93 and 1995-96. The board projected a $993 million return on the State’s $73 million investment in (cid:254) Although audit revenues these additional positions. increased overall by $558 million, this increase did not result from the Since one interpretation of a revenue increase is an increase additional staffing. in revenues from prior years, we computed the growth in (cid:254) audit revenues before and after the staffing increases. Our Revenues actually de- creased by $128.6 million analysis isolates the impact of the additional audit positions from prior years in those by eliminating revenues from audits, such as follow-ups on areas where we would Internal Revenue Service (IRS) leads and audits with potential expect the board to assign for large-dollar assessments, that the board would complete even new staff. if it did not have the additional staff. (cid:254) Despite adding 100 to 130 new audit staff, hours We determined that the board’s revenue increases of $558 million spent actually performing audits dropped from would have occurred regardless of the added positions since the fiscal year 1992-93 to increases came from audit types that traditionally receive high 1997-98. staffing priority because of their potential for very high returns. In fact, when we isolated the impact of the new audit positions from Finally, the board anticipates that changes in Internal the continuing efforts of the entire audit branch, we found that Revenue Service (IRS) revenues actually decreased by $128.6 million from prior years in operations will result in a the audits where we would expect the board to assign new staff. significant decrease in revenues from IRS leads. Several factors have contributed to a decrease in revenues in these types of audits. However, one significant reason is that the board is not spending additional time on these revenue-generating audits. Instead, although the total hours for the entire audit branch increased to reflect 100 to 130 additional audit staff, the number of hours spent performing audits dropped. The board disagrees with our assessment of its performance and asserts that it has not only met its projection of $993 million in increased revenue but has exceeded it by an additional $490 million. To determine these amounts, the board used a differing interpretation of a revenue increase from audits, a 1 method which the board asserts was understood by readers of its budget documents. The board’s analysis compares budgeted revenues to actual assessments; however, it does not isolate the benefits of the additional staff. We believe the board’s budgeting concept that forecasts future audit revenues by estimating the effects of changes in tax laws, changes in the economy, and other relevant factors is defensible. However, the board’s analysis does not fully describe its actual revenue resulting from the State’s investment in the new positions because it did not exclude the effects of IRS leads and audits with potential for large-dollar assessments that it would have completed even without the additional staff. To fully describe the actual revenue it received as a result of the State’s investment in new positions, the board’s analysis should indicate, by category of revenue, the hours to be charged to specific types of audits and the audit revenues projected to result from each type if the staffing increase is approved or denied. The board then needs to compare these projected hours and revenues to subsequent actual hours and revenues by type of audit. The board anticipates that changes in IRS operations will result in a decrease in the leads from this source and reduce audit revenues by at least 30 percent. Based on fiscal year 1997-98 data, each 10 percent drop in IRS leads could result in a $41 million decrease in audit revenues annually. Not only do many of the board’s audit assessments stem from IRS leads, but the costs associated with audits from these leads are lower, thus providing a greater return for each dollar it spends. For example, during fiscal years 1992-93 through 1997-98, audit assessments from IRS leads averaged $374 million annually, but the board spent only $12 million each year to generate these revenues. The Legislature asked the board to report on the benefits and costs of its audit program; however, it did not request informa- tion specific enough to fully assess the revenues resulting from the board’s 362 additional audit positions. Furthermore, the board’s report did not include essential assessment information such as all costs of its audit program. 2 C A L I F O R N I A S T A T E A U D I T O R RECOMMENDATIONS The Franchise Tax Board (board) should do the following: • The board’s budget documents should clearly indicate whether the board will use additional personnel hours for mandatory activities, such as filing enforcement and tax return processing, or for audit activities that are discretion- ary. If the additional hours will be used for audits, the budget documents should explicitly show, by category of revenue, the hours that will be charged to discretionary audits as well as the audit revenues that are projected to result from each type of audit with or without the staffing increase. • In subsequent years’ budget documents, the board should compare these projections to actual hours and revenues by type of audit. • If the board intends to request funding for auditors to gener- ate additional revenue, it should use these resources to supplement, rather than supplant, the auditors it has in the field. However, if the board later determines the resources can produce a greater benefit in support functions, it should report this to the Legislature. • The board should continue to monitor changes in audit revenues resulting from fewer IRS leads and either shift existing staff or request additional staff accordingly to maintain tax revenues. AGENCY COMMENTS The Franchise Tax Board (board) disagrees with the methodol- ogy we used to analyze the revenues generated by its additional audit staff. The board asserts that actual audit assessments should be compared to budgeted assessments to determine the benefit of increased staffing. n 3 Blank page inserted for reproduction purposes only. 4 C A L I F O R N I A S T A T E A U D I T O R INTRODUCTION BACKGROUND T he Franchise Tax Board (board) is one of the primary tax-collecting agencies in the State. For fiscal years 1990-91 through 1997-98, it collected an average of $20 billion in Personal Income Tax (PIT) revenues and $5 billion in Bank and Corporations (B&C) tax revenues annually. These taxes provide approximately 62 percent of the State’s General Fund revenues, which support educational, health, welfare, and other basic services to the public. As Figure 1 below indicates, the board collected $34 billion in taxes for fiscal year 1997-98, including taxes self-assessed by individuals and companies, and additional taxes assessed from audits. FIGURE 1 Tax Revenues the Board Collected Since Fiscal Year 1990-91 $35 Total Revenues PIT Revenues 30 B&C Revenues 25 20 15 10 5 0 Source: Governor’s Budget Summaries. 5 39-2991 49-3991 59-4991 69-5991 79-6991 89-7991 2 9 -1 9 9 1 19-0991 Fiscal Year )snoillib ni( seuneveR A three-member governing body consisting of the state control- ler, the director of the Department of Finance, and the chair of the Board of Equalization oversees the board’s activities, which are under the direct administration of an executive officer. With an estimated budget of $388 million for fiscal year 1998-99, the board employs 5,600 staff in its central office, 16 field offices, and satellite offices in Illinois, New York, and Texas. This report focuses on the board’s audit branch. THE AUDIT BRANCH REVIEWS TAX RETURNS TO ENSURE TAXPAYERS HAVE COMPLIED WITH THE LAW The board’s audit branch is responsible for examining tax returns to determine the accuracy of tax liabilities taxpayers calculated they owe. Audit staff issue notices of proposed changes to these self-assessed liabilities where appropriate and resolve some of the ensuing taxpayer disputes. The audit branch examines over 1.1 million tax returns annually. Figure 2 below displays the additional assessments the audit branch generated in the last eight fiscal years. FIGURE 2 Assessments the Board’s Audit Branch Generated Fiscal Years 1990-91 Through 1997-98 $1,200 Total Assessments B&C Assessments PIT Assessments 1,000 800 600 400 200 Source: Franchise Tax Board data for use in its annual Operations Report for each fiscal year. 6 C A L I F O R N I A S T A T E A U D I T O R 39-2991 49-3991 59-4991 69-5991 79-6991 89-7991 2 9 -1 9 9 1 )snoillim ni( stnemssessA 19-0991 Fiscal Year To maximize audit staff’s productivity, the board has developed a system for identifying those tax returns most likely to have the largest additional assessments. Based on past experience, the audit branch evaluates certain key tax return characteristics as well as information from outside sources, such as the IRS, and ranks returns based on the resources they would require for an audit and the revenue they might generate. Tax returns that could generate more revenue have the greatest priority for assignment. Historically, audits resulting from IRS leads have had very high returns, generating an average of $30 in revenue for every $1 of cost incurred. These audits have generated 40 percent of the board’s total additional assessments, yet they constitute only 15 percent of total auditing costs. The board has also placed a high priority on those B&C audits that have his- torically yielded very large assessments. In addition to identifying tax returns with higher estimated assessments, the board’s work plan is designed to provide an audit presence in the taxpaying community as a means of encouraging compliance with tax laws and increasing overall self-assessed revenues. In addition, the board identifies new or potentially new audit issues. Although it may be difficult to quantify the precise effect of maintaining an audit presence, there can be significant monetary value in making taxpayers aware of the potential for an audit. Further, this audit presence can help instill confidence in the fairness of state government by reassuring taxpayers who have always complied with the law that those who do not comply are at risk. SCOPE AND METHODOLOGY The Joint Legislative Audit Committee requested that the Bureau of State Audits review the board’s audit program. The specific purpose of our review was to determine the accuracy, reliability, and responsiveness of a report the board provided to the Legisla- ture regarding the benefits and costs of its audit program. We were also to independently assess these benefits and costs. Finally, we were asked to determine how much of the board’s additional assessments and costs result from IRS leads, and how much more value the board adds through its own audit efforts. To understand the board’s responsibilities related to the PIT and B&C tax audit programs, we reviewed the relevant laws, regula- tions, and policies and interviewed key staff. 7 In assessing the reliability, accuracy, and responsiveness of the board’s report to the Legislature on the benefits and costs of its audits, we compared the information the board reported to the kinds of information the Legislature requested, vouched selected data to accounting records and supporting schedules, and assessed the reasonableness of certain estimates used. We also conducted a general review of costs included in the data reported to determine if any significant costs were omitted. To independently assess the return on the board’s investment in its audit operations, we analyzed changes in related assessments before and after it received funding for additional auditors. We also compared the change in revenues to the anticipated changes described in budget documents. We interviewed key staff and assessed the board’s use of its auditors to determine the reasons for shortfalls in anticipated revenues. We did not attempt to determine the impact on tax revenues or additional assessments of changes in the economy or changes in tax law. Finally, to determine the impact of IRS audit leads, we separately identified the additional assessments from these leads and their related costs. n 8 C A L I F O R N I A S T A T E A U D I T O R CHAPTER 1 An Increase in Audit Revenues Did Not Result From Additional Time Spent Performing Audits CHAPTER SUMMARY R evenues from audits of Personal Income Tax (PIT) and Bank and Corporation (B&C) tax returns by the Franchise Tax Board (board) have provided the State an important source of funding at relatively little cost, returning approxi- mately $11 in assessments for every $1 of costs. Since one interpretation of a revenue increase is an increase in revenues from prior years, we computed the growth in audit revenues before and after staffing increases. Further, our analysis isolates the impact of the additional audit positions by eliminating revenues from audits, such as follow-ups on Internal Revenue Service (IRS) leads and audits with potential for large-dollar assessments, that the board would complete even if it did not have the additional staff. We determined that the board’s revenue increases of $558 million would have occurred without the new staff, since the increases came from audits with a high staffing priority because of their potential for large assessments. In fact, when we isolated the impact of the new audit positions from the continuing efforts of the entire audit branch, we found that revenues actually decreased by $128.6 million from prior years in those areas where we would expect the board to assign new staff. Several factors have contrib- uted to a decrease in revenues in these audit types. However, one significant reason is that the board is not spending additional time on these revenue-generating audits. In addition, according to the board, changes in tax law reduced B&C tax revenues substantially. The board disagrees with our assessment of its performance and asserts that it has not only met its projection of $993 million in increased revenue but has exceeded these expectations by an additional $490 million. To determine these amounts, the board used a differing interpretation of a revenue increase from audits, a method which the board asserts was understood by readers of its budget documents. The board’s analysis uses 9 budgeted revenues as the basis of comparison to actual assess- ments; however, the board did not isolate the benefits derived from the additional staff. We believe the board’s budgeting concept that forecasts future audit revenues by estimating the effects of changes in tax laws, changes in the economy, and other relevant factors is defensible. However, the board’s analysis does not fully describe the actual revenue it received as a result of the State’s investment in the new positions, since it did not exclude the effects of IRS leads and audits with potential for large-dollar assessments that the board would have completed even if it did not have the additional staff. To fully describe the actual revenue it received as a result of the State’s investment in new positions, the board’s analysis should indicate, by category of revenue, the hours to be charged to specific types of audits and the audit revenues projected to result from each type if the staffing increase is approved or denied. The board then needs to compare these projected hours and revenues to subsequent actual hours and revenues by type of audit. THE FRANCHISE TAX BOARD HAS HAD A POSITIVE RETURN ON ITS AUDIT PROGRAM The board’s audits have been an important source of revenue to the State at relatively little cost, returning approximately $11 in assessments for every $1 of cost. To ensure that the return on its investment in audits remains high, the board tracks the return it receives for each of the primary types of audits it performs and prioritizes its workload based on potential returns. As we discuss in subsequent sections, increases in revenues were limited to audit types with historically high returns. The board conducts three primary types of audits for both PIT Historically, the returns and B&C tax programs: field audits, desk audits, and audits that on the different audit follow up on IRS leads. Historically, the returns on each audit types varied dramatically. type have differed dramatically. For example, from fiscal years 1992-93 through 1997-98, the average return on PIT audits stemming from IRS leads was $28 for every $1 of cost while the average return on PIT field audits was only $4 for every $1 of cost. Table 1 summarizes the average return for each type of audit during this period. Refer to the Appendix for the returns during individual fiscal years. It is important to note that in compiling this information, the board did not adjust for the amount of these assessments it could actually collect. 10 C A L I F O R N I A S T A T E A U D I T O R TABLE 1 Average Return for Fiscal Years 1992-93 Through 1997-98 by Audit Type Average Return Per Description Dollar of Cost Personal Income Tax Desk audits $ 3.83 IRS follow-up audits 27.73 Field audits 3.76 PIT Audit Average $ 9.87 Bank and Corporation Tax IRS follow-up audits $49.77 Desk audits 5.11 Field audits 12.10 B & C Average $12.30 Average of All Audits $11.08 Source: Exhibit D of the Franchise Tax Board’s annual Operations Reports for fiscal years 1992-93 through 1997-98. Note: We included the costs of auditing refund requests in these calculations, but did not adjust the rate of return to account for the reductions in refunds resulting from these audits. Refund reductions are included in our analysis in Table 2. Within each of these audit types, the board creates numerous individual audit models with varying rates of return. As stated in the Introduction, when selecting tax returns for audit, the board identifies audit models likely to have the highest return and ensures that these audits receive highest priority. Because the returns associated with audits stemming from IRS leads have historically been very high, the board has given these audits a high staffing priority. The board has also placed a high priority on those B&C audits that have historically yielded very large assessments. The board’s policy is to assign staff to audits based on priority. Thus, we can expect it would have used its existing positions to ensure these high return audits were completed rather than requesting the additional positions to fill vacancies in these areas. 11 THERE WAS NO INCREASE IN REVENUES FROM THE TYPE OF AUDITS NEW STAFF PERFORM From fiscal years 1992-93 through 1995-96, the board received authorization for 362 additional auditor positions. In justifying more staff, the board indicated in budget documents that it could generate $993 million more in assessments from the additional audits the staff would perform. Audit revenues did increase by $558 million, but we believe revenues would have increased even without the new staff since the increases oc- curred in audit types that have a high priority. To determine the incremental benefit of the additional staff, we isolated the revenues and costs that existing staff would have created. In fact, we found revenues actually decreased on the lower-dollar audits to which the board would have assigned the additional staff. We discuss below the reasons for the lack of revenue. The 362 new positions the board received were intended to increase state revenues through additional audits of PIT and B&C returns. However, not all of the additional positions generate audit revenues because only some are directly involved in performing audits. The additional positions include tax auditors, tax technicians, legal counsel, temporary help, and support staff needed to keep up with the growth in the number of audits. Tax auditors, for example, generate audit revenues by reviewing the income tax returns of individuals or businesses and identifying any tax underpayments. On the other hand, support positions, such as telecommunications systems analysts, provide necessary services to enable tax auditors to work effi- ciently and thus complete more audits. In justifying the Revenues from those audit $73 million cost of the 362 new positions, the board’s budget types we would expect documents estimated that the additional positions would new staff to perform increase revenues by over $993 million. actually decreased from prior years by According to the board, it first staffs high-return audits, such $128.6 million since the as those resulting from IRS leads or B&C field audits resulting in authorization of new large assessments, and requests new staff for lower-priority audit positions. audits. Therefore, we have eliminated revenues of audit types with the highest returns from our analysis, which appears in Table 2, to isolate the return applicable to the new auditors. Further, our analysis reduces the board’s annual assessments by 20 percent to reflect uncollectible assessments. According to the board’s calculations, it collected 65 percent of fiscal year 1993-94 PIT assessments within the following five years. Although the board does not have collection data for B&C assessments, it 12 C A L I F O R N I A S T A T E A U D I T O R estimates that it collects 86 percent of these assessments within five years. We believe that 80 percent is a generous estimate of all collectible assessments. These estimates are reflected in the total adjusted audit revenues for PIT and B&C taxes in Table 2. Table 2 compares the adjusted annual audit revenues to a base- year average, which is the average of adjusted audit revenues for the two years before the board received the 362 new positions. This comparison reveals that audit revenues from the types of audits additional staff would likely perform have actually decreased by $128.6 million since the new audit positions were authorized. TABLE 2 Additional Audit Positions Did Not Generate Additional Returns Total Adjusted Increase (Decrease) Estimated Increase Audit Revenues in Adjusted Audit Total Number of in Budgeted Fiscal Year for PIT and B&C Revenues Over Base Positions Added Expenditures 1990-91 $ 453,076,090 1991-92 445,793,957 Base Year Average 449,435,023 1992-93 432,554,289 $ (16,880,734) 153 $ 5,833,208 1993-94 477,796,192 28,361,169 52 8,000,514 1994-95 411,642,848 (37,792,175) 88 14,917,514 1995-96 387,536,554 (61,898,469) 69 15,009,514 1996-97 427,254,302 (22,180,721) 14,509,514 1997-98 431,215,090 (18,219,933) 14,509,514 Total for Fiscal Years 1992-93 to 1997-98 $2,567,999,275* $(128,610,863) 362 $72,779,778 Source: Franchise Tax Board’s annual Operations Reports for fiscal years 1992-93 through 1997-98, budget documents, and additional data from the board. * Total adjusted audit revenues include reductions to claims for refund. The Number of Audit Hours, Changes in Tax Laws, and a Slow Economy Have Contributed to the Lack of Revenues Several factors, not all of which are under the board’s control, have contributed to the lack of audit revenues from sources other than IRS leads and large-dollar assessments. A significant 13 factor that is at least partially under the board’s control is the number of hours actually spent working on audits. As shown in Figure 3, the number of PIT and B&C audit hours did not in- crease as a result of adding these new positions. FIGURE 3 Despite Added Positions, Hours Spent on Audits Did Not Increase Total Audit Branch Hours 2,000 Total Audit Hours B&C Audit Hours PIT Audit Hours 1,500 1,000 500 0 Source: Data provided by the Franchise Tax Board. The PIT and B&C hours do not reflect supervisory hours, which the board was not able to separate from the audit branch’s total hours. The board explained that it would not expect audit assessments to increase during the first year because of the time-consuming learning curve for new audit staff. Although we agree that the full benefit from the new auditors may not have been realized in their first years, an increase in audit hours certainly should have occurred by fiscal year 1997-98 to reflect the 153 positions received in fiscal year 1992-93. Instead, although the total hours for the entire audit branch increased to reflect 100 to 130 additional audit staff, the number of hours spent perform- ing audits dropped. 14 C A L I F O R N I A S T A T E A U D I T O R 59-4991 69-5991 79-6991 89-7991 4 9 -3 9 9 1 )sdnasuoht ni( sruoH 39-2991 Fiscal Year We asked the board to explain how it used the additional audit Although the hours branch positions it had received since fiscal year 1992-93, but it spent by the entire unit was unable to tell us where it had assigned all the new staff. In increased to reflect 100 a report to the Legislature, however, the board presented limited to 130 additional staff, data on its assignment of audit staff. For example, the report audit hours dropped. stated that the board redirected 30 audit positions to informa- tion technology projects. Although these positions may serve an important support function for audits, they do not directly generate revenue. According to the board, it has had a policy of maintaining necessary funding for its mandatory workloads, such as filing enforcement and tax return processing, by redirecting resources from discretionary workloads, such as audits. The board also stated that turnover of experienced audit staff has resulted in fewer audit hours and a decrease in additional assessments. The board has further suggested that overall slow growth in PIT and B&C revenues during the early to mid-1990s, which was not under its control, contributed to lower audit revenues than projected. Although we agree that growth in these revenues was slow, it may not have had a significant impact on more recent audit assessments. Specifically, revenue data indicates that, although PIT and B&C revenues remained stagnant in the early part of the decade, PIT revenues, which constitute a much larger portion of the total, have been increasing since fiscal year 1993-94. The board also stated that legislative changes that allow tax credits for certain corporate activities, a reduced tax rate for some corporations, the elimination of a tax fee, and a deduction for net operating losses, all contributed to lower B&C tax rev- enues. The board estimated that collectively these legislative changes reduced B&C taxes by $1 billion annually. However, we have not independently reviewed the basis for the board’s estimates. Further, according to the board, at least one of these changes in tax law began having an effect before the new staff were added. 15 THE BOARD BELIEVES TOTAL AUDIT ASSESSMENTS HAVE EXCEEDED PROJECTIONS SINCE THE INCREASE IN AUDIT STAFF The board believes that we have not accurately assessed the effect of its new staff on its audit revenues. Although we believe our analysis of the board’s audit revenues gives a valuable per- spective based on a comparison to prior-year revenues, we also believe the board’s own assessment of its performance gives another important perspective. The board asserts that actual assessments should be compared to budgeted assessments rather than actual prior-year revenues. Calculating revenues in this manner would indicate the board has exceeded expectations for increases in revenues. In the board’s opinion, its perspective is the one shared by those who read its budget documents such as the Legislature and its staff. Each year, the board analyzes its workload to estimate the The board’s position on amount of audit revenues existing staff should generate. It projected revenues differs takes into account factors such as changes in tax law or in from ours. the State’s economy that could have a significant impact on tax revenues. If the resulting workload analysis indicates that there is not sufficient staff to perform all audits with a return on investment ratio of $5-to-$1 or greater, it may request addi- tional staff. The board asserts that in the budget documents justifying new staff, it based projected increases in assessments on the estimated return of the additional audits new auditors can perform, not on assessment increases over the prior year. The board believes that readers of the budget documents understand this basis for projected revenues in its budget docu- ments. Table 3 summarizes the board’s position, which differs from ours. Thus, according to the board, its audits actually exceeded the $993 million promised in the budget by an additional $490 million, the amount by which actual assessments exceeded total budgeted assessments. 16 C A L I F O R N I A S T A T E A U D I T O R TABLE 3 The Board’s Comparison of Anticipated Audit Assessments to Actual Assessments (Dollars in Millions) Increase (Decrease) in Actual Over Total Budgeted Audit Total Actual Audit Budgeted Audit Fiscal Year Assessments Assessments Assessments 1992-93 $ 855 $ 862 $ 7 1993-94 864 1,039 175 1994-95 806 851 45 1995-96 978 1,125 147 1996-97 1,063 1,308 245 1997-98 1,055 926 (129) Totals $5,621* $6,111 $490 Source: Franchise Tax Board’s Analysis of Audit Augmentation Costs and Net Audit Assessments. * Includes the $993 million increase anticipated by the board. THE BOARD’S ANALYSIS OF THE PERFORMANCE OF ITS AUDIT BRANCH DOES NOT SEPARATELY IDENTIFY THE BENEFITS FROM NEW STAFF We do not dispute that overall audit assessments have increased since the board received the new staff or that audit assessments have exceeded budgeted amounts. However, we do not believe that the board’s analysis separately measures the benefits the State derived as a result of the board’s budget increases. Increases in staff over a base-line budget and estimates of revenue generated from the increase call for an incremental, or marginal, analysis. In this type of analysis—similar to the one we employed in our calculation of the revenues additional staff generated—the value derived from the added resources is isolated from what would have occurred without the increase. The board’s estimated audit revenues are for the entire audit Our analysis fully staff. To more accurately assess the contribution of the new describes the actual auditors, it is more appropriate to analyze the additional costs revenues the board and actual revenues associated with the new positions only, as generated from its new we have done. We believe the board’s analysis does not fully positions. describe the actual revenue it gained as a result of the State’s investment in the new positions. 17 Further, the board’s assessment of its performance does not clearly follow from the information in its budget documents. These documents do not indicate specifically how the board planned to use the staff or to assess their proposed benefit. Although the board’s analysis indicated it exceeded budgeted revenues, it did so primarily through the use of existing, not additional, staff. RECOMMENDATIONS The Franchise Tax Board’s (board) budget documents should clearly indicate whether it will use additional personnel hours for mandatory activities, such as filing enforcement and tax return processing, or for audit activities that are discretionary. If the additional hours will be used for audits, the budget docu- ments should explicitly show, by category of revenue, the hours that will be charged to discretionary audits as well as the audit revenues that are projected to result from each type of audit with or without the staffing increase. In subsequent years’ budget documents, the board should compare these projections to actual hours and revenues by type of audit to determine the benefit of additional staff. If the board receives funding to increase discretionary auditing activities to generate more revenue, it should dedicate these resources to auditing. If the board later determines that the resources would have greater benefit if used elsewhere, it should notify the Legislature of its plans to shift resources. n 18 C A L I F O R N I A S T A T E A U D I T O R CHAPTER 2 Internal Revenue Service Leads Contribute Significantly to Tax Assessments, but the Board Is Concerned That These Leads May Be Reduced in the Future CHAPTER SUMMARY M any audit assessments stem from leads the Internal Revenue Service (IRS) provides to the Franchise Tax Board (board). Additionally, the board’s costs for audits resulting from these leads are lower than costs associated with its independent audit efforts. For example, over the last six years, the board’s average return on Personal Income Tax (PIT) audits stemming from IRS leads was $28 for every $1 of cost it incurred, while its independent PIT audit efforts only returned $4 for every $1 of cost. From fiscal years 1992-93 through 1997- 98, audit assessments from IRS leads averaged $374 million annually, representing 40 percent of all audit assessments the board generated. However, the board anticipates that potential changes in IRS operations will result in a substantial reduction in the volume of these leads, decreasing audit revenue by at least 30 percent in fiscal year 1998-99 and in each of the next few fiscal years. According to the board, it has already experienced a 40 percent drop in the revenue agent reports it receives from the IRS. Based on fiscal year 1997-98 data, each 10 percent drop in IRS leads could result in a $41 million decrease in annual audit revenues. IRS staff, on the other hand, were unable to assess what, if any, changes would occur in the quantity of leads provided. BACKGROUND The board receives audit leads from the IRS in two ways. The IRS sends the board revenue agent reports, which disclose additional federal tax liabilities identified during agents’ field audits. The IRS also sends electronic files with information from its Underreporter Program. The board reviews these leads and 19 pursues only those with a potential impact on state taxes exceeding certain minimum thresholds for a return on the costs it will incur for the follow up. THE BOARD GENERATES SIGNIFICANT AUDIT ASSESSMENTS FROM IRS LEADS AT MINIMAL COST A large portion of the board’s audit assessments stem from IRS Seventy-one percent of leads, and the costs related to these audits are lower than costs all PIT assessments are associated with the board’s independent audit efforts. Audit derived from IRS leads, assessments stemming from IRS leads for fiscal years 1992-93 yet consume only through 1997-98 averaged $374 million, but cost the board only 25 percent of PIT audit $12 million annually. costs. As Figure 4 shows, from fiscal years 1992-93 through 1997-98, an average of 40 percent of total audit assessments stemmed from these leads, while only 15 percent of total audit costs were related to them. Moreover, 71 percent of audit assessments generated from PIT were derived from IRS leads, yet only 25 percent of total PIT audit costs stemmed from these leads. According to the board’s annual operating reports, during this period it averaged $28 for every $1 of cost on PIT audits stem- ming from IRS leads, while the return on its own independent PIT audit efforts averaged only $4 for every $1 of cost. The board’s annual operating reports indicate that during the same period, its return on Bank and Corporation (B&C) tax audits stemming from IRS leads averaged $50 for every $1 of cost, while its independent B&C field audit efforts returned $12 for every $1 of cost and its B&C desk audits returned only $5 for every $1 of cost. However, the majority of B&C audit assessments result from the board’s own field audit efforts and not IRS leads. As Figure 4 indicates, during fiscal years 1992-93 through 1997-98, an average of 79 percent of B&C audit assess- ments came from the board’s own field audits, while only 14 percent resulted from IRS leads. 20 C A L I F O R N I A S T A T E A U D I T O R FIGURE 4 Assessments and Costs Related to Each Type of Audit Fiscal Years 1992-93 Through 1997-98 Field audits Costs Desk audits IRS follow-up audits Assessments $ in billions $5.6 $.51 $2.5 $.26 $3.1 $.25 100% 80 60 40 20 0 Total PIT B&C Source: Franchise Tax Board’s annual Operations Reports for fiscal years 1992-93 through 1997-98. A POTENTIAL REDUCTION IN IRS LEADS MAY SIGNIFICANTLY DECREASE AUDIT REVENUES The board has raised concerns about potential declines in the volume of IRS leads, and, therefore audit revenues, because of changes in IRS operations. The IRS has not projected the effect of any changes in its operations on leads to the board. If the board’s concerns are justified, the impact on audit revenues could be significant. 21 The board anticipates that in fiscal year 1998-99 and beyond, operational changes within the IRS will reduce the volume of revenue agent reports the board receives by 30 percent to 50 percent. According to the board, since March 1998, it has experienced a 40 percent drop in revenue agent reports. Although fluctuations in the revenues from these reports are not unusual, the board expects this reduction in volume to decrease audit revenue by at least 30 percent in fiscal year 1998-99 and in each of the next few fiscal years. Based on fiscal year 1997-98 data, each 10 percent drop in IRS leads could result in a $41 million decrease in annual audit revenues. When we contacted representatives of the IRS to inquire about Each 10 percent drop in these changes, we were told that although the agency is now IRS leads could result in a focusing more on customer service, it has not projected the $41 million decrease in effect of any changes on leads to the board. annual audit revenues. Although the decrease in IRS leads may have a significant impact on future audit revenues, the impact on the board’s total tax revenues would be less significant because audit rev- enues represent a relatively small portion of total tax revenues. For example, for fiscal year 1997-98, audit revenues totaled $851 million, representing only 3 percent of total tax revenues of $34 billion. The board’s projected decrease of 30 percent in audit revenues would represent less than a 1 percent decrease in total tax revenues the board collects. Nevertheless, IRS leads result in an important source of revenue to the State. RECOMMENDATION The Franchise Tax Board (board) should continue to monitor changes in audit revenues resulting from fewer IRS leads. It may also have to propose a shift of existing staff from activities related to IRS leads to other productive audits or request addi- tional staff to ensure that it is able to sustain its contribution to the State’s revenue needs. n 22 C A L I F O R N I A S T A T E A U D I T O R CHAPTER 3 The Franchise Tax Board’s Report on Audits Does Not Contain Sufficient Information to Assess the Return on Its Investments in Its Audits CHAPTER SUMMARY T he Legislature required the Franchise Tax Board (board) to report on the benefits and costs of its audit program. However, the request did not solicit key information that would allow for a comparison of revenues before and after the board added the 362 audit positions. Additionally, the board’s report did not provide essential information, including all costs of external vendors used in developing information technology projects or departmental overhead costs, such as rent and utilities, totaling $75 million for fiscal years 1992-93 through 1996-97. Furthermore, the board’s inconsistent presentation of the benefits and costs of its audits contributes to the difficulties in assessing the return of its audit program. BACKGROUND In the past, the board has justified budget augmentations to increase its staff on the basis of revenue impact. The Legislature approved these additional auditors with the apparent understanding that the board would realize at least a $5-to-$1 return for tax auditing activity. However, the Legislative Analyst’s Office, which independently reviews the State’s budget for the Legislature, became concerned that the board was not meeting the $5-to-$1 return on its audit program and recom- mended that the board provide the Legislature with additional information. Based on this recommendation, the Legislature required the board to submit a report detailing its audit costs and revenues by November 1, 1997. The purpose of obtaining this information was to enable the Legislature to assess the board’s return on its audit program and, in particular, on the new audit staff. 23 A REPORT THE BOARD PREPARED FOR THE LEGISLATURE DOES NOT CONTAIN ALL AUDIT COSTS The Legislature requested the board to disclose the total direct and indirect costs of its audit program since fiscal year 1992-93. Although the board’s report contains direct costs, and the collection costs related to its audit program, it does not contain all of the indirect costs. As a result, the reader of the report cannot accurately determine the board’s return on its invest- ment in audits. For example, the audit costs do not include the complete cost Audit costs excluded from of information technology projects. The costs disclosed in the the board’s report totaled report include the audit branch’s allocated portion of costs $75 million for fiscal associated with the in-house development of certain informa- years 1992-93 through tion technology projects. However, the costs of external vendors 1996-97. used in developing these projects are not included in the report. Additionally, the audit costs reported do not include other departmental overhead costs, such as rent and utilities. The audit branch’s allocated portion of these costs totaled $75 million for fiscal years 1992-93 through 1996-97, the years included in the board’s report. Additionally, the board’s presentation of the benefits and costs of its audit program makes it even more difficult to calculate an accurate return on its investment in audits. For example, although the report expresses the cumulative collections of fiscal year 1992-93 net assessments over a five-year period, it does not provide the cumulative costs of collecting these assess- ments. Only the costs incurred in the original fiscal year are presented. This inconsistent treatment does not equitably match the benefits with the true costs of the audit program for any given year. ALTHOUGH USEFUL, THE LEGISLATURE’S REQUEST TO THE BOARD COULD HAVE BEEN MORE FOCUSED Although it did request useful information from the board, the Legislature did not require certain data, and, therefore, the report did not provide all information needed to address legislative concerns. A primary intent of the report was to provide data that would permit an assessment of the board’s methodology when making revenue impact calculations justify- ing the addition of audit staff. 24 C A L I F O R N I A S T A T E A U D I T O R Although we recognize the need to balance the board’s ease of gathering information against the usefulness of the information, the type of information the Legislature requested was too gen- eral to clearly demonstrate the impact on revenue from changes in staff. Additional data could have made the report more valuable to the Legislature and other readers. For example, providing the amount of revenue generated before the board received any of the 362 new positions would allow the reader to determine the incremental change in audit revenues that occurred after the new staff were authorized. In addition, the Legislature could have required the board to explicitly demonstrate the benefit derived from all its audit activities, as well as that derived just from the new staff, thus providing a better understanding of the true rate of return on audit activities. Finally, the Legislature could have required the board to report exactly how it used the new staff. With this information, users of the report would have learned that the board assigned very few of the new staff to revenue-generating audit positions. RECOMMENDATION In any future request for program information from the Fran- chise Tax Board, the Legislature should tailor the request to more specifically address its concerns. By doing so, the Legislature can gather relevant information that is more valuable to it and other interested parties. We conducted this review under the authority vested in the California State Auditor by Section 8543 et seq. of the California Government Code and according to generally accepted governmental auditing standards. We limited our review to those areas specified in the audit scope section of this report. Respectfully submitted, KURT R. SJOBERG State Auditor Date: March 18, 1999 Staff: Lois Benson, Audit Principal, CPA Michael Tilden, CPA Debra Maus, CPA 25 Blank page inserted for reproduction purposes only. 26 C A L I F O R N I A S T A T E A U D I T O R 27 APPENDIX Fiscal Year 1992-93 Fiscal Year 1993-94 Fiscal Year 1994-95 Net Net Net Assessments Assessments Assessments Net Per Dollar Net Per Dollar Net Per Dollar Description Assessments Costs of Cost Assessments Costs of Cost Assessments Costs of Cost Personal Income Tax Audit Activities: Desk audits $72,350,279 $18,016,591 $ 4.02 $ 72,481,019 $20,179,560 $ 3.59 $72,521,500 $29,197,471 $ 2.48 IRS follow-up audits 253,182,720 10,690,515 23.68 215,103,169 13,465,380 15.97 234,242,595 12,591,051 18.60 Field audits 30,826,462 7,543,866 4.09 37,358,335 10,066,610 3.71 25,635,732 14,088,297 1.82 PIT Audit Subtotals $356,359,461 $36,250,972 $ 9.83 $ 324,942,523 $43,711,550 $ 7.43 $332,399,827 $55,876,819 $ 5.95 Bank & Corporation Tax Audit Activities: Desk audits $126,420,239 $ 4,065,500 $ 31.10 $ 28,866,476 $ 5,193,341 $ 5.56 $15,802,957 $ 5,393,235 $ 2.93 IRS follow-up audits 78,456,142 697,881 112.42 92,889,584 595,056 156.10 67,869,209 317,738 213.60 Field audits 262,345,881 27,260,493 9.62 559,433,361 28,136,434 19.88 355,641,856 29,234,452 12.17 B&C Audit Subtotals $467,222,262 $32,023,874 $ 14.59 $ 681,189,421 $33,924,831 $ 20.08 $439,314,022 $34,945,425 $ 12.57 Total desk audits $198,770,518 $22,082,091 $ 9.00 $ 101,347,495 $25,372,901 $ 3.99 $88,324,457 $34,590,706 $ 2.55 Total IRS follow-up audits 331,638,862 11,388,396 29.12 307,992,753 14,060,436 21.90 302,111,804 12,908,789 23.40 Total field audits 293,172,343 34,804,359 8.42 596,791,696 38,203,044 15.62 381,277,588 43,322,749 8.80 Total, All Audits $823,581,723 $68,274,846 $ 12.06 $1,006,131,944 $77,636,381 $ 12.96 $771,713,849 $90,822,244 $ 8.50 28 C A L I F O R N I A S T A T E A U D I T O R Fiscal Year 1995-96 Fiscal Year 1996-97 Fiscal Year 1997-98 Net Net Net Assessments Assessments Assessments Net Per Dollar Net Per Dollar Net Per Dollar Description Assessments Costs of Cost Assessments Costs of Cost Assessments Costs of Cost Personal Income Tax Audit Activities: Desk audits $ 75,197,249 $ 23,149,061 $3.25 $ 125,272,190 $18,584,040 $6.74 $ 65,748,724 $16,970,439 $ 3.87 IRS follow-up audits 428,869,457 10,481,893 40.92 328,254,628 10,008,388 32.80 340,397,536 7,671,541 44.37 Field audits 51,574,009 14,178,928 3.64 70,783,680 10,857,404 6.52 28,709,761 8,400,690 3.42 PIT Audit Subtotals $ 555,640,715 $ 47,809,882 $11.62 $ 524,310,498 $39,449,832 $13.29 $434,856,021 $33,042,670 $13.16 Bank & Corporation Tax Audit Activities: Desk audits $ 10,658,444 $ 7,486,577 $1.42 $ 10,341,357 $ 8,523,808 $1.21 $ 14,986,595 $ 9,842,264 $ 1.52 IRS follow-up audits 45,766,446 1,935,262 23.65 91,960,547 1,812,347 50.74 65,152,738 3,523,743 18.49 Field audits 470,787,761 29,405,065 16.01 474,409,137 42,656,038 11.12 335,823,783 46,537,381 7.22 B&C Audit Subtotals $ 527,212,651 $ 38,826,904 $13.58 $ 576,711,041 $52,992,193 $10.88 $415,963,116 $59,903,388 $ 6.94 Total desk audits $ 85,855,693 $ 30,635,638 $2.80 $ 135,613,547 $27,107,848 $5.00 $ 80,735,319 $26,812,703 $ 3.01 Total IRS follow-up audits 474,635,903 12,417,155 38.22 420,215,175 11,820,735 35.55 405,550,274 11,195,284 36.23 Total field audits 522,361,770 43,583,993 11.99 545,192,817 53,513,442 10.19 364,533,544 54,938,071 6.64 Total, All Audits $1,082,853,366 $ 86,636,786 $12.50 $1,101,021,539 $92,442,025 $11.91 $850,819,137 $92,946,058 $ 9.15 29 89-7991 hguorhT 39-2991 sraeY lacsiF teN egarevA stnemssessA egarevA ralloD reP egarevA teN tsoC fo tsoC stnemssessA noitpircseD xaT emocnI lanosreP :seitivitcA tiduA 38.3 $ 76.391,610,12$ 71.061,595,08 $ stidua kseD 37.72 00.821,818,01 38.053,800,003 stidua pu-wollof SRI 67.3 38.569,558,01 71.366,418,04 stidua dleiF 78.9 $ 05.782,096,24$ 71.471,814,124$ slatotbuS tiduA TIP xaT noitaroproC & knaB :seitivitcA tiduA 11.5 $ 05.787,057,6 $ 00.876,215,43 $ stidua kseD 77.94 38.733,084,1 33.444,286,37 stidua pu-wollof SRI 01.21 38.346,178,33 05.692,047,904 stiduA dleiF 03.21$ 71.967,201,24$ 38.814,539,715$ slatotbuS tiduA C&B 51.4 $ 71.189,667,72$ 71.838,701,511$ stidua ksed latoT 93.03 38.564,892,21 71.597,096,373 stidua pu-wollof SRI latoT 70.01 76.906,727,44 76.959,455,054 stidua dleif latoT 80.11$ 76.650,397,48$ 00.395,353,939$ stiduA llA ,latoT Agency’s response provided as text only: M E M O R A N D U M To: Kurt R. Sjoberg, State Auditor Date: March 11, 1999 Bureau of State Audits 555 Capitol Mall, Suite 300 Sacramento, CA 95814 From: Gerald H. Goldberg Subject: Draft Bureau of State Audit Report Thank you for the opportunity to review the draft audit report prepared by your staff for the Joint Legislative Audit Committee. As you are aware, we do not agree with the methodology utilized by your staff to analyze the revenue produced by Franchise Tax Board’s (FTB) program. We appreciate our analysis being included in your report and your finding that our methodology for determining revenues is defendable. Following are specific comments to the report and the recommendations: CBR METHODOLOGY—We understood the primary purpose of the report to 1* be an audit of the methodology used by the FTB in the computation of the Cost Benefit Ratio (CBR) which is used to determine the most beneficial audits to perform. Apparently, there was a concern that all costs were not included in the computation of CBRs. Chapter 3 of your report discusses CBRs, but there is no mention of this issue in the Executive Summary. CBRs are determined by FTB on three different levels: Annual Operations Report – This is the information summarized in the Appendix to your report. The CBRs in the Operations Report are based on total net assessments finalized in the fiscal year compared to the total program cost. As shown in your report, FTB’s audit program has averaged over $11 in net assessments for every $1 of cost over the last six years. Audit Workplan – The computation of CBRs in the workplan was initiated only for the purpose of prioritizing audit workloads so staff resources could be applied to those audits that would provide the highest return per dollar cost. These CBRs compare projected net assessments to program costs on a workload model basis. Prioritizing the discretionary workload models on the basis of these CBRs allows us to identify any workloads that are unbudgeted at the $5 to $1 threshold or above. *California State Auditor’s comments on this response begin on page R-5 R-1 Budget Change Proposal (BCP) – BCPs include only the additional, variable costs needed to complete the additional audits identified in the workplan. These costs are substantially less than the full program cost (variable plus fixed costs) that are used for the CBRs calculated above. Comparing the increase in net assessments to the BCP cost gives the true picture of what it is costing the State to produce the additional revenue. For example, the Audit Program BCP in 1995/96 added 107,494 hours of audit production at a total BCP cost of $2.9 million. In the following two years, those audit resources produced 23 $17.3 million in 1996/97 and $20.6 million in 1997/98. This results in a return on investment, or CBR, of $5.97 to $1 in 1996/97 and $7.10 to $1 in 1997/98. STAFFING ANALYSIS – The assumptions made in the report have ignored the entire workplan process and as a result have erroneously concluded that 3 FTB revenues could have been produced without the additional 362 positions. The audit report should have focused on the entire audit program instead of only the discretionary audit portion. Since 1981, FTB has prioritized its audit workloads in a workplan that has been provided to the Department of Finance and Legislative Analyst each year. Whenever there has been an augmentation to the audit staff, it is because the workplan has shown, in detail, how existing staff is going to be utilized for mandatory/indirect and discretionary workloads and how much discretionary workload exceeds our budgeted resources. Over the period of 1992/93 – 1997/98, audit hours have been redirected to successfully implement new audit technologies, provide technology support and to increase audit training. The automation of the Revenue Agents Report (RAR) Inventory system and the implementation of the Audit Pass-Through-Entities Automated Selection and Support System (PASS) project have dramatically changed the way we do business and allowed us to maintain or surpass revenue projections with few additional discretionary hours. Obviously, the addition of this technology requires 3 redirection to provide training and support. All of these redirections were identified before implementation and were accepted when the BCP augmentations were approved. Without the additional positions, discretionary audit levels could not have been maintained and a significant portion of each year’s revenue production would have been lost. The workplans are also used by Department of Finance and the Legislative Analyst to make their revenue forecasts and to establish the budgeted level of revenues expected each year from the board. As shown in Chapter 1, page 12 of your report, our actual performance over the past six years has exceeded budgeted revenues by almost $500 million using the same or less in discretionary hours. This efficiency can be attributed to the technology improvements stated above. Net assessments per hour increased from $976 for 92/93 to $1426 and $1143 per hour for 96/97 and 97/98, respectively. R-2 REVENUE ANALYSIS — As stated in the report, we do not agree with the methodology used by your staff to analyze revenue. Our annual budget is computed based on projected revenue not historical revenue. There are a number of factors that govern the planned budgeted audit revenue. These include, but are not limited to, tax law changes, tax regulations, case law and precedent, economic conditions and self-compliance. All of these factors are considered in projecting audit program revenue. Your analysis did not include any of these factors. A statistical study would need to control for each of these factors. If these factors are not identified and their impacts measured, it is not possible to use your methodology to make a useful evaluation. A far 1 better evaluation is to compare audit revenue planned, which has taken into consideration these factors, to actual audit revenue results. The analysis of planned to actuals shows that FTB exceeded projected revenue for the period reviewed. RECOMMENDATIONS – • The board’s budget documents should clearly differentiate whether the additional personnel hours will be used for mandatory activities such as taxpayer assistance and legal support or audit activities that are discretionary. If the additional hours will be used for audits, the budget documents should explicitly show, by category of revenue, the hours that will be charged to discretionary audits and the audit revenues that are projected to result from each type of audit if the staffing increase is approved and if it is not approved. As explained previously, FTB already provides an audit program workplan that shows how all budgeted audit resources are planned to be used and how many additional personnel hours are needed to process all discretionary 4 audits at or above the $5 to $1 CBR. Audit Program BCPs separately identify hours to be used for support from hours to be used for discretionary audits and the workplan shows the discretionary audit models and the revenues expected from each. An extra effort will be made in the future to make sure legislative members and their consultants have this detail in addition to the Department of Finance and the Legislative Analyst. • In subsequent years’ budget documents, the board should then compare these projections to subsequent actual hours and revenues by type of audit revenue. We agree with this recommendation. • If the board intends to request funding for auditors to generate additional revenue, it should use these resources to supplement, rather than supplant, the auditors it has in the field. However, if the board determines later that the resources can produce a greater benefit in support functions, the board should report this to the Legislature. R-3 As described above, any redirection of budgeted audit staff is fully detailed in the workplan before any decision is made to add additional staff. To the extent that the board proposes to redirect newly budgeted staff in subsequent years, the workplans will report the redirection and provide the necessary justification. • The board should continue to monitor changes in the level of audit revenues resulting from fewer Internal Revenue Service leads and shift existing staff or request additional staff accordingly to maintain state revenues. We agree. Again, we appreciate the opportunity to provide you this response. If you need any further data or would like to discuss any of the issues above, feel free to contact me at 845-4543. (Signed by: Gerald H. Goldberg) Executive Officer cc: Hon. Kathleen Connell Hon. Johan Klehs Hon. B. Timothy Gage Marcy Jo Mandel James R. Tucker Annette Porini Connie Squires R-4 COMMENTS California State Auditor’s Comments on the Response From the Franchise Tax Board T o provide clarity and perspective, we are commenting on the Franchise Tax Board’s (board) response to our audit report. The number corresponds to the number we have placed in the response. 1 We believe the letter requesting the audit clearly asks for an analysis of the cause and effect relationship between the marginal cost and marginal benefit of additional auditors by asking “. . . do extra auditors really generate an additional $5 in revenue for every $1 expended?” In response to this question, we analyzed both the marginal revenue and marginal costs the board used in calculating its return on investing in additional auditors. We found the board’s budgeting concepts used to prioritize audit workloads and fore- cast future audit revenues, which includes the estimated effects of various factors such as tax law and economic changes, is defen- sible. However, we found its assertion that the $490 million increase in revenues it attributes to 362 additional audit staff is incorrect. The board’s use of its budgeting concept to prioritize workloads is appropriate. As discussed on page 16 of our report, this process assists in identifying those audits with a certain level of expected return. However, to identify the actual benefit derived from new staff, the board must adjust its budgeting concept for workloads performed by existing staff. It does not make such adjustments. In its response, as well as throughout the audit, the board was unable to explain its logic as to why the incremental revenues of $490 million should be attributed to the efforts of the new staff. 2 The board’s statistics are misleading. Subsequent to receiving its response, we contacted board staff for the data supporting its calculation of its returns on investment. The data indicates that for fiscal years 1995-96, 1996-97, and 1997-98, revenues resulting from the seven lowest priority audit models were $26.9 million, $17.3 million, and $20.6 million, respectively, from audit hours invested of 56,436, 31,291, and 25,331, respectively. Thus, the R-5 hours invested in these seven audit models and the revenues gener- ated from these models decreased from fiscal year 1995-96. Therefore, the board’s response attributing the revenues generated from these seven models to the efforts of new staff actually highlights the flaw in its analysis. Given the board’s logic, it can continually justify addi- tional funding for auditors, use this funding for any purpose, and then refer to the results of these seven audit models where it invests minimal time, but receives a positive return. Even if the funding is used for purposes the board deems higher priority than audits, this calculation of the audit return on investment should not be used to justify the funding. Rather, funding in non-audit areas should be justified on its own merits. 3 On page 16, we describe the board’s work plan and how it affects the staffing of its operations. The board ignores the fact that we did assess the staffing as it relates to the entire audit program. We present the results on pages 13-15 of our report. The graph and discussion on those pages relate not only to the discretionary audits, but also to the entire audit program. Our discussion and graph are based on information the board provided to us on the number of new audit branch positions requested and on the changes in total audit branch hours and the actual hours spent on performing discre- tionary audits between fiscal years 1992-93 and 1997-98. Although the board received authorization for 362 new audit branch positions, its own data shows that it used no more than 100 to 130 of those positions in the audit branch. More importantly, hours spent on discretionary audits actually dropped, despite the increase in staff. The board’s argument about the benefits derived from audit branch staff other than those performing audits would be more compelling if it had been able to tell us how it used the 362 new authorized posi- tions, or even the 100 to 130 staff who worked in the audit branch, and the extent of the benefits derived specifically from these staff. In fact, the board could not. 4 The board believes its present work plans meet the intent of our recommendation. To further clarify our recommendation on page 18, we modified the format we suggest be provided to policy makers when considering augmenting the board’s audit staff. We have added columns for prior-year actual data and current-year estimated data that were not previously included in our recommendation, but that we believe are also needed. Since the board believes it already has the necessary information and that its budgeting method establishes the benefits resulting from the investment in additional audit resources, it should not object to arraying the data in this format. R-6 Prior-Year Current-Year Proposed Without Proposed With Actual Estimate Staffing Increase Staffing Increase Actual Actual Estimated Estimated Proposed Projected Proposed Projected Activity Hours Revenues Hours Revenues Hours Revenues Hours Revenues Mandatory Activities: Filing enforcement Tax return processing Etc. Discretionary Activities: Personal Income Tax Desk audits Field audits IRS follow-up audits Bank and Corporation Tax Desk audits: Large assessment audits Smaller assessment audits Field Audits: Large assessment audits Smaller assessment audits IRS follow-up audits R-7 Blank page inserted for reproduction purposes only. R-8