CSA

Summary

California State Auditor · 2002-124 · 2002-01-01

Read the report at California State Auditor ↗

Franchise Tax Board: Its Performance Measures Are Insufficient to Justify Requests for New Audit or Collection Program Staff May 2003 2002-124 rotiduA etatS ainrofilaC S T I D U A E T A T S F O U A E R U B The first five copies of each California State Auditor report are free. Additional copies are $3 each, payable by check or money order. 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 TTY (916) 445-0033 OR This report is also available on the World Wide Web http://www.bsa.ca.gov/bsa/ The California State Auditor is pleased to announce the availability of an online subscription service. For information on how to subscribe, please contact the Information Technology Unit at (916) 445-0255, ext. 456, or visit our Web site at www.bsa.ca.gov/bsa Alternate format reports available upon request. Permission is granted to reproduce reports. � � � �� May 13, 2003 2002-124 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 and collection program positions. This report concludes the board’s performance measures are not sufficient to justify requests for new audit or collection program staff. The board uses a variety of performance measures, with various cost components, for its audit and collection programs, potentially causing confusion about prospective and actual program results. Board documents requesting new staff for its programs typically refer to a projected return of at least $5 of benefit for each $1 of cost for these staff. In fact, between fiscal years 1998–99 and 2001–02, the board’s 340 newest audit positions returned only $2.71 in assessments— potential revenue—for each $1 of cost, without adjusting for the likelihood or cost of collecting the assessed amounts. Due to various limitations in collection program data, including a change in accounting treatment, we were unable to determine the incremental revenue resulting from the newest 175 collection program positions. However, the board has allowed some approved collection program positions to go unfilled in order to achieve savings to pay for other expenses, such as merit salary adjustments. Respectfully submitted, ELAINE M. HOWLE State Auditor �� Franchise Tax Board: Its Performance Measures Are Insufficient to Justify Requests for New Audit or Collection Program Staff CONTENTS Summary 1 Introduction 7 Chapter 1 The Board’s Various Performance Measures May Create Confusion About the Return on Audit and Collection Activities 15 Recommendations 24 Chapter 2 The Board’s Performance Measures Are Insufficient to Justify Requests for New Audit Program Staff 27 Recommendations 39 Chapter 3 The Board’s Performance Measures Are Insufficient to Justify Requests for New Collection Program Staff 41 Recommendations 54 Appendix A Historical Returns of the Audit Program By Audit Type 55 Appendix B Collection Program Actual Revenue and Cost Data 65 Appendix C Proposed Template for a Comparison of Actual to Projected Returns for Specific Audit Workload Types 67 Response to the Audit Franchise Tax Board 71 California State Auditor’s Comments on the Response From the Franchise Tax Board 77 California State Auditor Report 2002-124 11 SUMMARY RESULTS IN BRIEF A primary revenue-generating agency for the State, the Franchise Tax Board (board) processes individual and corporation tax returns, audits certain tax returns for errors, and collects delinquent taxes. Between fiscal years Audit Highlights . . . 1990–91 and 2001–02, the board provided an average of Our review of the Franchise $31 billion in annual tax revenues to the State, over 60 percent Tax Board’s (board) audit and of the State’s General Fund that pays for education, health, collection activities revealed welfare, and other public services. Although many taxes are the following: self-assessed by individuals and companies, the board’s audit þ The board does not program reviews the accuracy of tax returns, assessing addi- always describe the tional taxes when appropriate. In turn, the collection program differing cost components pursues delinquent taxpayers identified by the board’s various of its various performance measures, potentially assessment activities. leading to confusion about program results. The variety of performance measurements the board uses for þ Between fiscal years the audit and collection programs can confuse decision makers 1998–99 and 2001–02, such as the Department of Finance (Finance) and the Legislature recently acquired audit about the programs’ projected and actual results. A complete staff returned $2.71 in performance measure compares all the benefits of a program assessments for each $1 of cost. with all the costs of producing them. However, in budget docu- ments describing the projected benefits that will result from þ Because of limitations in new staff, the board excludes some departmental overhead costs board data, we could without disclosing this exclusion. However, the board’s subse- not isolate the return on 175 new collection quently published historical reports of program results include program positions. all costs. Further, the board’s budget documents do not disclose a substantial overlap in benefits that are stated once by the audit þ The board’s process for assessing the program as assessments and again by the collection program as incremental benefit of those same tax assessments are collected as revenue. The board recently acquired audit believes Finance and the Legislative Analyst’s Office (LAO), and collection program which advises the Legislature on budgetary matters, are aware positions is flawed. of the overlap. Finance confirmed that it is aware of the overlap, þ The board allows some but indicates it would support a refinement of these measure- collection program ments to better capture the benefits. The LAO is also aware of positions to remain this overlap and considers it when interpreting the board’s data. unfilled in order to pay for other expenses. To increase revenues, the board received authorization for an additional 340 net audit positions between fiscal years 1992–93 and 2001–02. The board justified many of the new positions with cost-benefit ratios (CBR) that projected returns of at least $5 in audit assessments for every $1 of cost. In contrast, our California State Auditor Report 2002-124 11 review found that for every $1 of cost, the 340 audit positions returned only 79 cents in assessments over the period although the return on the additional positions improved to $2.71 for fiscal years 1998–99 through 2001–02. Changes in the economy probably affected the return on these audit positions, but a major cause of the low return is that despite having additional staff, the board did not increase the number of hours staff spent performing audits. These hours differed little in fiscal year 2001–02 from those in fiscal year 1992–93. The collection program added 175 collection program positions between fiscal years 1998–99 and 2001–02, projecting increased revenue of $179 million over that period. However, because of limitations in board data, we could not determine the return on the 175 collection program positions. Although sufficiently demonstrating the overall cost-effectiveness of its audit and collection programs, the board’s process for assessing the incremental benefit of recently acquired audit and collection program positions is flawed. The board lacks suffi- cient data and uses an inadequate methodology to determine whether increases in audit assessments or collection program revenues resulted from additional positions. Rather than using an incremental approach to isolate assessment or revenue pools likely to have been affected by additional audit or collec- tion program positions, the board compares its total projected audit assessments against its total actual audit assessments and its total projected collection program revenue against its total actual collection program revenues. At the highest level of analysis, the board can demonstrate that the audit division returns about $10 in assessments, or potential revenue, for every $1 of cost, and the collection program returns around $19 in revenue for every $1 of cost. However, the board lacks a persuasive analysis to show that additional tax assessments and cash receipts to the State came from the workloads that incremental staff would likely have been assigned to work. Recently, the board has justified collection program staffing requests based on a process that prioritizes workload according to a cost-benefit ratio. However, the board actually assigns staff based on risk and yield factors calculated by its new Accounts Receivable Collection System (ARCS). This leaves the board unable to adequately demonstrate the reliability of its cost- benefit approach. Now that the collection program has nearly two years of experience using the new system, the board is developing an alternative methodology for justifying collection 22 California State Auditor Report 2002-124 California State Auditor Report 2002-124 33 program staffing needs that uses data from ARCS to better reflect the manner in which the board actually assigns collection program staff. Finally, the board is not using all of its funding for collection program salaries to actually fill authorized positions, but is instead using some of the funding for other costs. In fiscal year 1999–2000, separate merit salary adjustment (MSA) funding for the board was ended, leaving the board to find another way to pay for its MSAs. In fiscal years 2000–01 and 2001–02, the board’s savings on salaries increased by roughly 5 percent. To achieve these savings the board has left unfilled some collec- tion program positions even though the board’s budget control language requires it to fill them expeditiously. RECOMMENDATIONS To more completely and clearly reveal its programs’ costs and benefits, the board should consider using the complete measurement of the audit program’s performance that we have described in Table 3 on page 17. This measurement compares all the benefits—the total revenues that result over time from the auditors’ assessments of additional taxes—with the total costs to produce them, including the costs of collection. Thus, the board would treat the collection program as another service center for audits. If it determines that its current information system cannot produce the data necessary for such a measurement, the board should consider the needs of a complete measurement when it upgrades or changes its current information system. If the board decides not to use the complete measurement and continues to use separate performance measurements for the audit and collection programs, it should do the following: • In budget change documents and other reports given to external decision makers, the board should explicitly disclose the elements not included in the cost components of various performance measures used to assess the audit and collection programs. The board should also disclose the effect of those excluded elements. Further, the board should disclose the overlap in benefits claimed by its audit and collection programs. 22 California State Auditor Report 2002-124 California State Auditor Report 2002-124 33 • To provide useful information to decision makers when requesting additional audit positions, the board should use a format, recommended in our prior report and shown on page 22, that details the types of activities new auditors will perform as well as the projected assessments and historical assessments resulting from these activities. Additionally, the board should revise its supporting audit workplan to include the actual returns of each of the specific workload types for the most recently completed fiscal year. • To track the accuracy over time of its calculations of the prospective CBRs for individual audit workload types, the board should compare these prospective CBRs against actual returns annually. The board should make the results available to Finance and the LAO, and should also include them in the board’s annual report to the Legislature on the results of its audit and collection activities. If the board believes this information is confidential, it can cloak the identity of the individual audit workloads in its annual report to the Legis- lature as we have done in Appendix C. Moreover, the board should use the results of the comparison in future calculations of prospective CBRs. To demonstrate the effectiveness of new collection program positions, the board should develop a methodology for measuring the benefit of these positions by isolating the return resulting from the additional positions and comparing it against a base year. To more accurately represent the process that assigns work to collection program staff, the board should continue to develop a methodology based on data from ARCS to justify new collection program staffing requests. For the board to be consistent with the intent of budget control language and Finance, it should not as a long-term strategy leave collection program positions unfilled beyond the normal time it takes to fill a position. AGENCY COMMENTS The board indicates it will implement most of the recommen- dations made in the report. It agrees that improvements can be made to increase the usability of information it provides to Finance and the Legislature, asserting that it has already begun a project to better capture revenue and cost data. The 44 California State Auditor Report 2002-124 California State Auditor Report 2002-124 55 board disagrees with the methodology we used to analyze the additional assessments generated by the new audit staff. The board contends that because our analysis fails to take into consideration tax law changes, tax regulations, case law and precedent, economic conditions, and self-compliance it does not recognize the full value of the new audit positions. We provide comments to clarify and add perspective to the board’s response to the audit. n 44 California State Auditor Report 2002-124 California State Auditor Report 2002-124 55 66 California State Auditor Report 2002-124 California State Auditor Report 2002-124 77 INTRODUCTION BACKGROUND The Franchise Tax Board (board) is one of the primary tax-collecting agencies in the State. For fiscal years 1990–91 through 2001–02, the board annually collected an average of $26 billion in personal income tax revenues and $5 billion in corporation tax revenues. Corporation taxes include amounts from corporations, limited liability partnerships, banks, and other business entities. The combined personal income and corporation taxes provide over 60 percent of the State’s General Fund revenues, which support educational, health, welfare, and other basic services to the public. As Figure 1 on the following page indicates, the board collected $38 billion in tax revenue for fiscal year 2001–02, including taxes self-assessed by individuals and companies when they filed their tax returns, and additional taxes assessed from the board’s audit and filing enforcement activities. A three-member governing body composed of the state control- ler, the director of the Department of Finance (Finance), and the chair of the State Board of Equalization oversees the board’s activities, which are under the direct administration of an execu- tive officer. With a proposed budget of $438 million for fiscal year 2002–03, the board employs about 5,500 staff in its central office; three regional offices; 11 district offices in California; and satellite offices in Illinois, New York, and Texas. This report focuses on the board’s audit and tax collection programs. THE AUDIT PROGRAM REVIEWS TAX RETURNS TO ENSURE TAXPAYERS HAVE COMPLIED WITH THE LAW The board’s audit program examines tax returns to determine if taxpayers have accurately calculated the amounts they owe the State. When they find errors, the audit staff issue notices of corrections. The audit program examines about 400,000 tax returns each year. Between fiscal years 1998–99 and 2001–02, the audit program averaged $950 million each year in additional assessments. 66 California State Auditor Report 2002-124 California State Auditor Report 2002-124 77 FIGURE 1 The Board’s Tax Revenues Fiscal Years 1990–91 Through 2001–02 �� 88 California State Auditor Report 2002-124 California State Auditor Report 2002-124 99 �� Source: Governor’s Budget Summaries for indicated fiscal years. To maximize audit staff’s productivity, the board has developed a workplan process for identifying tax returns most likely to produce the largest additional assessments. Based on past experience, the audit branch evaluates selected returns, using certain key character- istics of tax returns as well as information from outside sources such as the Internal Revenue Service (IRS), then ranks returns based on the estimated cost of auditing them and the amount of assessments they might generate. Tax returns that could generate the most benefit relative to cost have the greatest priority for assignment. For example, audits resulting from IRS leads have historically led to very high returns, generating an average of $32 in tax assessments for every $1 of cost incurred for fiscal years 1992–93 through 2001–02. The board has also placed a high priority on certain corporation tax audits that have the potential to yield very large assessments. The board generally tries to staff all audits with projected returns of at least $5 in assessments for every $1 of cost. In addition to identifying tax returns with higher estimated assessments, the board’s workplan is designed to aid in identifying new audit issues and provide an audit presence in the taxpaying community as a means of encouraging compliance with tax laws and increasing overall self-assessed revenues. 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 for an audit. Although the value of this audit presence cannot be quantified, we believe it can be significant. The amount of time the board spends on these audits with lower returns varies from year to year, and according to the Legislative Analyst’s Office (LAO) the board also has a certain degree of administrative flexibility to redirect staff during the year, but should notify the LAO and the Legislature when it does so. We discuss the board’s audit workplan process further in Chapter 2. For fiscal years 1992–93 through 2002–03, the board’s audit program requested new positions, generally using the $5 to $1 return ratio as justification to the Legislature. As Table 1 indicates, the board received approval for 385 additional audit program positions. TABLE 1 The Audit Program’s Additional Authorized Positions Fiscal Years 1992–93 to 2002–03 Fiscal Year Additional Positions 1992–93 153 1993–94 52 1994–95 88 1995–96 69 1996–97 1997–98 1998–99 1999–2000 -22 2000–01 2001–02 2002–03 45 Total Additional Positions 385 Sources: Governor’s Budgets, budget change documents, and Final Changebooks for fiscal years 1998–99 to 2002–03; Bureau of State Audits Report 98118.2 for prior years. 88 California State Auditor Report 2002-124 California State Auditor Report 2002-124 99 THE TAX COLLECTION PROGRAM PURSUES OUTSTANDING ASSESSMENTS GENERATED THROUGH OTHER BOARD ACTIVITIES The Accounts Receivable Management Division administers the board’s tax collection program, which collects delinquent taxes established through the board’s self-assessment, audit, settle- ment, and filing enforcement activities. Filing enforcement staff identify and issue tax assessments to individuals and busi- ness entities that have not filed a required return; settlement activities are a streamlined, alternative method of resolving tax disputes. Because collections result from accounts receivable generated by these other activities, the success of the collection program depends partly on the quality of the information these other activities provide. According to preliminary information the board compiled, in fiscal year 2001–02 collection program activities produced an additional $1.7 billion in receipts, about 4 percent of the board’s total tax revenues for the year. Collections on delinquent taxes become the collection program’s responsibility roughly 120 days after they are assessed. After assessing a delinquent amount, the audit program sends the taxpayer a notice requesting payment. If the taxpayer does not remit the payment within the 120 days, the collection program then attempts to collect the outstanding amount through its automated billing and collection process. If the automated system also fails to collect, the account is generally referred to collectors who then pursue the outstanding amount, thereby attempting to ensure that nonvoluntary taxpayers also contribute the appropriate amount to the State. Collecting delinquent tax amounts can take several years, if they are collected at all. Like the audit program, the collection program has a process for prioritizing collection workloads. Based on historical experience, the collection program estimates the probable return on account types and attempts to target its resources toward those promis- ing the highest return and greatest likelihood of collection. It sometimes helps to justify its requests for additional staff by estimating a return of at least $5 in collectible revenue for every $1 of cost. As Table 2 indicates, the board has received 337 new collection program positions since fiscal year 1998–99. 1100 California State Auditor Report 2002-124 California State Auditor Report 2002-124 1111 TABLE 2 The Board’s Newly Authorized Collection Program Positions Fiscal Years 1998–99 to 2002–03 Fiscal Year Additional Positions 1998–99 42 1999–2000 7 2000–01 3 2001–02 123 2002–03 162* Total Additional Positions 337 Sources: Governor’s Budgets, budget change documents, and Final Changebooks for indicated fiscal years. * Only 77 of these positions are permanent additions. WE PREVIOUSLY REVIEWED THE BOARD’S AUDIT PROGRAM In 1998, the Joint Legislative Audit Committee (committee) asked us to review the benefits and costs of the board’s audit program. In response, we issued a report in March 1999 titled Franchise Tax Board: Its Revenue From Audits Has Increased, but the Increase Did Not Result From Additional Time Spent Performing Audits. In general, we concluded that overall assessments from audits had increased, but the increase did not result from the 362 additional audit positions the board received from fiscal years 1992–93 through 1995–96. In fact, we found that assessments had actually decreased from prior years in those areas where we would have expected the board to assign new staff. We concluded that one significant reason for this decrease was that the board had not assigned all the new staff to the audit program. SCOPE AND METHODOLOGY The committee requested that the Bureau of State Audits (bureau) review the board’s audit and collection programs, identifying additional audit and collection program positions given to the board between fiscal years 1998–99 and 2001–02, assessing the board’s calculation of the costs and benefits of these positions, and determining whether the board uses these positions as the Legislature intended. The committee also asked us to determine whether new audit positions given to the 1100 California State Auditor Report 2002-124 California State Auditor Report 2002-124 1111 board between fiscal years 1992–93 and 1997–98 are used as audit personnel and to assess the board’s calculation of the costs and benefits of those positions as well. Also, the committee asked us to determine whether the board has documentary evidence of the costs and benefits of its audit and collection programs and to review its methodology for calculating these costs and benefits. Finally, the committee asked us to determine whether a point of diminishing returns exists where additional audit and collection program positions do not generate a $1 to $5 cost-benefit ratio (CBR) and, if so, to determine the board’s actions to shift those positions to other activities1. To understand the board’s responsibilities related to its personal income tax and corporation tax audit and collection programs, we reviewed the relevant laws and policies and interviewed key staff. To identify additional audit and collection program positions the board received between fiscal years 1998–99 and 2001–02, we interviewed key staff and examined various budget documents finding that the board did not receive any new audit positions during this period. In assessing the board’s calcula- tion of the costs and benefits of the new collection positions, we analyzed staffing requests and other supporting documentation. We also tried to assess the incremental return on new collection program positions but were unable to do so because certain data were unavailable. To determine whether the board uses these collection program positions as the Legislature intended, we pri- marily compared actual hours to budgeted workload hours that included hours for the new staff and interviewed key staff about the results of our comparison. To determine the assignment of new audit staff given to the board between fiscal years 1992–93 and 1997–98, we inter- viewed key staff and analyzed audit branch hours. To assess the board’s calculation of the costs and benefits of audit positions for fiscal years 1992–93 through 2001–02, we analyzed changes in related assessments before and after it received funding for these additional auditors. We did not attempt to quantify how changes in the economy affected tax revenues or assessments. The board discloses the costs and benefits of its audit and collection programs in annual operations reports, which are the source for much of the data in our various analyses. We did 1 Our review deals with personal income and corporation taxes. The board also collects certain insurance taxes, but the amounts are very small in comparison to the board’s overall revenues. 1122 California State Auditor Report 2002-124 California State Auditor Report 2002-124 1133 not audit the board’s operations reports and did not validate the assessment amounts by examining individual tax returns or actual assessments. However, to gain some assurance that information in the operations reports is accurate and complete, we reviewed for reasonableness the board’s methodology in preparing them, and we traced selected costs and revenues to accounting records and traced assessments to supporting schedules. We also compared assessment and cost data provided as estimates in our previous audit to amounts subsequently published in the operations reports and asked the board to explain significant differences. Only fiscal years 1995–96 through 1997–98 were revised from estimates included in our 1999 report. Therefore, for these three years certain amounts contained in our prior report have been updated in this report. When information we currently report is not final, we have identified it as draft information. To determine whether a point of diminishing returns exists where additional staff do not generate a $1 to $5 CBR and, if so, to examine the board’s actions to shift those staff to other activities, we interviewed key staff, reviewed work plans, and examined the nature of the CBR. This included a review of the costs and benefits incorporated in the audit and collection programs’ workplans. Finally, we reviewed budget documents supporting the board’s request for additional audit program positions for fiscal year 2002–03, and we interviewed board and Finance staff to deter- mine whether the board implemented a recommendation from our previous audit that it provide more specific information on the results of prior staff increases and on its planned use of new audit positions when requesting additional positions. n 1122 California State Auditor Report 2002-124 California State Auditor Report 2002-124 1133 1144 California State Auditor Report 2002-124 California State Auditor Report 2002-124 1155 CHAPTER 1 The Board’s Various Performance Measures May Create Confusion About the Return on Audit and Collection Activities CHAPTER SUMMARY The Franchise Tax Board (board) uses a variety of measure- ments to gauge audit and collection program performance and to assign workloads to staff. Most of these mea- surements take into account some of the costs and related benefits for program activities, but the various measurements may include differing calculations of costs, which the board does not always fully describe in public documents. As a result, misunderstandings of the board’s performance may arise. Ide- ally, a performance measure should compare all the benefits of a program with all the costs of producing them. However, when the board’s budget documents project a return of at least $5 in benefits, whether assessments or revenues, for each $1 of cost for new positions, the projected return does not reflect allocated costs for departmental overhead, such as rent and utilities, and the understated costs are not disclosed. Further, the audit and collection programs have some overlap in the benefits they claim. After 120 days, tax assessments the audit program claims as benefits become the collection program’s accounts receivable, which, if collected, are also counted as benefits of the collection program. If the board continues to use incomplete performance measurements, decision makers should be aware of the specific benefit and cost elements that are included in the board’s various audit and collection program performance measures and consider the overlap in claimed benefits. Although the board’s current resource request format for new audit positions provides decision makers with more detail regarding audit workloads than the board typically provided prior to our 1999 report, its current format is still insufficient to demonstrate both the workload types to which the board intends to assign new staff and the historical return on those workloads. In addition, historical actual returns on the specific workloads are not measured against the projections used to justify the staff increases. Without 1144 California State Auditor Report 2002-124 California State Auditor Report 2002-124 1155 this information, decision makers are left without an accurate tool against which to measure whether the board’s staffing increases return their projected assessments. THE BEST MEASUREMENT OF PERFORMANCE COMPARES ALL BENEFITS WITH THE COSTS REQUIRED TO PRODUCE THEM Because of how the board compiles information, its perfor- mance measurements of its audit and collection programs do not provide a full picture of either program. As explained later, the board separately measures the performance of the audit and Not all of the board’s collection programs, with some overlap of claimed benefits. various performance Further, not all the board’s various performance measurements measurements of its audit include the full costs of those programs. and collection programs include the full costs of In contrast, complete measurement of a program’s performance those programs. would compare all the quantifiable benefits the program produces to all the costs required to produce those benefits. As discussed in the Introduction, the audit program produces some benefits that are difficult to quantify. For example, some audits are designed to identify new audit issues and others to provide an audit presence in the taxpaying community as a means of encouraging compliance with the laws and increasing overall self-assessed revenues. Nevertheless, a complete measurement of the performance of the board’s audit program, for example, would compare the benefits—the total revenues that result over time from the auditors’ assessments of additional taxes— with the total costs to produce them. Total costs would then include not only auditors’ salaries and benefits, but also salaries and benefits of audit program and other staff who support the auditors’ activities, as well as a proportionate share of departmental overhead costs. The costs of actually collecting assessments would then become a part of the performance measure of the audit program. The Legislative Analyst’s Office (LAO), which provides the Legislature with analyses of budget issues, made a similar point in a fiscal year 1997–98 report when it stated that the full cost of collecting audit assessments should be considered as part of the overall and true cost of generating returns through audits. Similarly, other board programs, such as the self-assessment and filing enforcement activities, that require the services of the collection program to realize their full program benefits would also incorporate the costs of collection in the measures of their performance. Because the collection program serves to realize the full benefits of these other 1166 California State Auditor Report 2002-124 California State Auditor Report 2002-124 1177 programs, it would not have a separate performance measure. Table 3 below provides a simple illustration of the complete performance measurement of the board’s various programs. TABLE 3 Composition of Complete Performance Measurements for the Board’s Various Assessment Activities Total Assessment Total Collection Costs Costs Total Costs (Including (Including to Assess and Cost-Benefit Departmental Departmental Total Revenues Collect Ratio Assessment Origin Overhead) Overhead) (Cash Receipts) (Columns 1 + 2) (Columns 3 ÷ 4) Column 1 Column 2 Column 3 Column 4 Column 5 Audit Return Processing Filing Enforcement THE BOARD USES A VARIETY OF PERFORMANCE MEASURES AND DOES NOT ALWAYS DESCRIBE THEIR DIFFERENCES IN PUBLIC DOCUMENTS The board uses various measurements to gauge audit and col- lection program performance, assign workloads to staff, and report to outside interested parties, including the Legislature and Department of Finance (Finance), for budgeting and other purposes. Although they usually account for program costs and related benefits, the various measures differ in the composi- tion of their costs and benefits. The board does not always fully describe what is behind its various calculations, increasing the potential for misunderstanding, particularly for external users who make decisions based on budget documents. Table 4 on the following page describes measures the board commonly uses, indicates the composition of each measure, and discloses how the board uses each. 1166 California State Auditor Report 2002-124 California State Auditor Report 2002-124 1177 TABLE 4 Various Performance Measures the Board Uses Reference to Audit Program Collection Program Source Use in This (Measure) Use Costs Benefits Costs Benefits Report Operations Used for Total actual Net assessments Total actual Actual revenue. Figure 2, p.29 Reports (annual internal and expenditures, (potential expenditures, Table 8, p. 38 program return) external including revenue). including historical departmental departmental Figure 5, p.43 reporting of overhead overhead program results. allocation. allocation. App. A.1, p.57 App. B.1, p. 66 Budget Change Prospective Estimated Projected net Estimated Projected Table 9. p.45 Documents estimates of expenditures, assessments and expenditures, revenue. (expenditures the effect of excluding some cash receipts. excluding some and revenues additional departmental departmental associated requested overhead. overhead. with additional funding for resources) positions or other resources. Provided to external decision makers in order to justify increased funding. Workplan (cost- Prospective Estimated Projected net Estimated Projected Table 8, p.38 benefit ratio) detailed expenditures for assessments for expenditures for revenue for App. C.1, p.68 estimate of each workload each workload each workload each workload results of type, excluding type. type, excluding type. specific types departmental departmental of workloads. overhead. overhead. Audit program uses internally to determine which projects to assign. Frequently used as support for budget change documents. ARCS (risk-yield Prospective N/A N/A Risk calculation Estimated yield. Figure 7, p. 50 calculation) measure used (ARCS is not internally to designed to determine quantify all nature of work costs). for collection program staff. 1188 California State Auditor Report 2002-124 California State Auditor Report 2002-124 1199 As the table indicates, the board includes different costs and benefits in its measures of historical and prospective performance for both its audit and collection programs. The historical measures reported in the board’s annual operations reports are calculated using full costs—direct program costs such as audit and collection personnel costs, allocated indirect costs from the board’s central service units, and allocated departmental overhead costs, such as rent and utilities. Prospective measures used for budget change purposes, on the other hand, include some, but not all, overhead costs. Other prospective measures, used primarily for internal prioritizing of audit program workload but also for supporting the audit and collection programs’ budget change documents, do not reflect departmental overhead costs in their calculation. Although internal users may be aware of these differences, external users, such as the Legislature and Finance, may not, and the board does not always explicitly include a description in public documents of the cost elements of each measure. Given the board’s use of separate performance measures for the Audit program benefits audit and collection programs, it is also important to distinguish are assessments, or between the types of benefits each program provides. In general, potential revenue, whereas audit program benefits are assessments for additional taxes, rep- collection program resenting only potential revenue. In contrast, collection program benefits consist of actual benefits are actual revenues collected, or cash receipts, which revenue collected. are based on the audit assessments as well as receivables other programs establish. Thus, with the board’s separate reporting of audit and collection program performance, it is inaccurate to think of cash receipts as the primary benefit of both audit and collection activity. Budget documents referring to prospective cost-benefit ratios (CBR) can be especially confusing. As Table 4 shows, the board employs a CBR analysis of its audit and certain collection pro- gram workloads. The CBR’s purpose is to quantitatively compare the anticipated costs of performing audit or collection activi- ties to their corresponding estimated benefits. Independently, each program creates a workplan ranking its projects in order of estimated return. The audit program then assigns its staff to the various workload types, generally beginning with those having the highest anticipated return. As discussed more fully in Chap- ter 3, the collection program does not use its workplan to assign work to staff. However, both the audit and collection programs have used the workplans to identify the need for additional staff to pursue all workload types at or above a 1:5 CBR, signifying $1 of cost for every $5 of benefit returned. 1188 California State Auditor Report 2002-124 California State Auditor Report 2002-124 1199 However, an estimated CBR of 1:5 does not mean a given work- load will generate $5 of additional tax revenue to the State, or even $5 in additional tax assessments, for every $1 invested because neither program includes departmental overhead costs in calculating its CBRs.2 The board uses these CBRs in its budget documents to help justify additional staff, and we do not believe this use of the CBR is appropriate without adequate dis- closure of the omission of departmental overhead costs, which can be significant. For fiscal year 2000–01, such costs totaled $81 million, approximately 20 percent of the board’s total costs, so including each program’s proportionate share of full overhead costs would significantly reduce the projected CBR of additional requested staff positions. By not explicitly disclosing the differ- ences in these calculations, the board risks confusing decision makers, who may then make decisions based on conflicting or incomplete information. THE AUDIT AND COLLECTION PROGRAMS HAVE SOME OVERLAP IN CLAIMED BENEFITS The board’s performance measures for its audit and collection programs also suffer from a partial overlap in claimed benefits, another potential source of confusion about returns on costs. The board includes all tax deficiencies assessed through its audit program as benefits in its historical and prospective measures of audit program performance. However, taxpayers do not remit many of these assessments within 120 days, after which the assessments become the collection program’s responsibility. Consequently, all audit assessments that the collection program ultimately expects to collect are also counted as benefits in the Some assessments that collection program’s prospective performance measures—and are counted as benefits actual receipts are similarly double-counted in its historical of the audit program performance measures. For example, the board asserts that its are also considered collection program collected almost $115 million in personal collection program income tax audit assessments in fiscal year 2001–02. However, benefits when the revenue these audit assessments were already claimed as benefits of the is ultimately collected. audit program. While acknowledging this overlap exists, the board does not discuss it in budget documents. However, the board believes that the LAO and Finance are aware of the overlap. Finance has confirmed that it is aware of the overlap, but indicates it would support a refinement of these measure- ments to better capture the benefits. The LAO is also aware of 2 When the board uses the CBR as a tool to prioritize work for staff, omitting departmental overhead is not a problem as we would expect such costs to be spread proportionally among all workloads. As a result, the order of priority would not change. 2200 California State Auditor Report 2002-124 California State Auditor Report 2002-124 2211 this overlap and considers it when interpreting the board’s data. Nevertheless, we believe it is important that the board indicate in the documents available to external decision makers that the overlap exists, thus lessening the likelihood of an expectation that the audit and collection programs will each generate actual cash receipts comparable to their performance measures. THE BOARD’S BUDGET CHANGE DOCUMENTS DO NOT SHOW HOW NEW AUDIT POSITIONS HAVE MET PROJECTED RESULTS When requesting audit program positions, the board uses a format that does not show projected and actual hours spent by audit staff on the various audit types, nor does it detail each type’s related assessments. We believe that without the data to make these detailed comparisons, the board cannot reason- ably justify its requests for additional audit staff. In our 1999 report we recommended that the board use a specific template we developed to provide decision makers with information that is relevant and specific to the additional positions requested. As illustrated in Table 5 on the following page, the template we recommended separates the audit program work by audit types, contains historical assessment returns for each of these audit types, and compares the projected assessments with and without the benefit of additional positions. Thus, the template requires information specific to those audit types to which new staff would be assigned. However, after we released our report, the board and Finance met and agreed on the use of a different template, or “matrix.” To the best of Finance’s recollection, the changes made to the matrix were based on the board’s concern about workload cat- egory size and possible abridgement of confidentiality of data. Finance accepted the changes proposed by the board based on the understanding that Finance could request additional infor- mation, as needed. Although the matrix includes many of the features we recommended—such as prior year actual data, cur- rent year estimated data, and projected hours and assessments with and without staffing increases—it does not detail historical and projected hours and assessments by audit type as we had recommended. Rather, the board’s matrix summarizes all desk, field, and Internal Revenue Service (IRS) follow-up audit activ- ity in a single category, “Audit Direct,” which obscures the very different returns on each of the personal income tax and corporation tax audit types. 2200 California State Auditor Report 2002-124 California State Auditor Report 2002-124 2211 TABLE 5 Template We Recommended for the Board’s Use When Requesting Additional Staff Without Staffing Prior Year Current Year Increase With Staffing Increase Actual Actual Estimated Estimated Proposed Projected Proposed Projected Activity Hours Assessments Hours Assessments Hours Assessments Hours Assessments Direct Audit Activities: Personal Income Tax Desk audits Field audits IRS follow-up audits Corporation Tax Desk audits Field audits IRS follow-up audits Therefore, the board’s matrix does not focus on the audit types new auditors would be assigned to, the projected return of those audit types, or even their historical returns. As discussed earlier, the board generally staffs the various audit types beginning with those having the highest anticipated return. Therefore, exist- ing staff would probably fill the high-return audits, while new positions would most likely pursue lower-return audit types such as desk and personal income tax field audits. However, how the board plans to use new auditors is not apparent in the board’s matrix because it only includes very high-level information on its proposed use of new staff. The board used its matrix to support a request for 45 additional audit positions in its 2002–03 fiscal year budget change proposal, which was approved by the Legislature. However, we continue to believe more detailed information is crucial to evaluating the need for additional staff. Also, without more detail, the board and decision makers have no relevant measure of the subsequent return on the audit types where additional staff would be assigned. 2222 California State Auditor Report 2002-124 California State Auditor Report 2002-124 2233 In addition to its matrix, the board provides Finance its audit workplan as supporting information for its budget change pro- In its requests for new posals. In fact, the board provides Finance its preliminary and audit staff, the board revised workplans every year. The workplan includes projected does not provide the returns for more specific workload types than even our template historical returns on the does. For example, the fiscal year 2002–03 workplan lists more workloads to which the than 10 separate categories and their prospective CBRs related to new staff would probably personal income tax desk audits and 14 for corporation tax field be assigned. audits. The board uses this information to assign work to staff. However, the workplan does not contain the historical returns of the specific workload types. We believe that the workplan would be more useful in evaluating the board’s need for addi- tional staff if it contained the actual returns and full costs of each specific workload type for the most recently completed fiscal year. According to Finance and the LAO, they are not dis- satisfied with what the board has been providing since they can ask for information beyond the matrix and workplan. In fact, they have historically received any information they requested. However, Finance and the LAO have no objection to having the board routinely provide, as opposed to providing in response to a special request, more detailed information as long as any confi- dentiality concerns the board may have are addressed. Additionally, we believe the board should compare actual to projected CBRs for specific workload types to measure the reasonableness of its prospective CBRs. The board does not currently make this comparison, but it already collects much of the necessary information about actual results—actual assessments and staff time spent on each of the detailed audit types. Also, the board records actual costs for the broader category of audit type—personal income tax field audits, for example. Thus, the only information the board needs for a comparison of projected and actual CBRs at the workplan level of detail is the appropriate distribution of costs. Since the board also tracks salary costs associated with each individual audit workload, it can use this information to allocate audit costs proportionately. Therefore, information that the board would need to assess the reasonableness of its detailed audit CBRs is readily available. Appendix C presents a template the board could use to compare actual returns to projected CBRs for individual audit types. We believe this information should be made available to Finance and the LAO, and it should be included in the board’s annual report to the Legislature that summarizes the results of its audit and collection programs. The LAO agrees with our suggestion, as does Finance, subject to any confidentiality concerns the board may have with providing this data. 2222 California State Auditor Report 2002-124 California State Auditor Report 2002-124 2233 If the board performed this comparison annually, it would have the necessary information to identify those prospective CBRs that are not reasonably close to actual results and revise them accordingly. As a result, it would have both a more effective tool for assigning workload to staff and a more defendable and accurate projection of the benefits that would accrue from adding new audit positions. As we discuss in Chapter 3, the collection program also has difficulty in demonstrating the effectiveness of new positions. RECOMMENDATIONS To more completely and clearly reveal its programs’ costs and benefits, the board should consider using the complete measure- ment of the audit program’s performance that we have described in Table 3 on page 17. This measurement compares all the benefits—the total revenues that result over time from the auditors’ assessments of additional taxes—with the total costs to produce them, including the costs of collection. Thus, the board would treat the collection program as another service center for audits. If it determines that its current information system cannot produce the data necessary for such a measurement, the board should consider the needs of a complete measurement when it upgrades or changes its current information system. If the board decides not to use the complete measurement and continues to use separate performance measurements for the audit and collection programs, it should do the following: • In budget change documents and other reports given to exter- nal decision makers, the board should explicitly disclose the elements not included in the cost components of various performance measures used to assess the audit and collection programs. The board should also disclose the effect of those excluded elements. Further, the board should disclose the over- lap in benefits claimed by its audit and collection programs. • To provide useful information to decision makers when requesting additional audit positions, the board should use a format, recommended in our prior report and shown on page 22, that details the types of activities new auditors will perform as well as the projected assessments and historical assessments resulting from these activities. Additionally, the board should revise its supporting audit workplan to include the actual returns of each of the specific workload types for the most recently completed fiscal year. 2244 California State Auditor Report 2002-124 California State Auditor Report 2002-124 2255 • To track the accuracy over time of its calculations of the prospective CBRs for individual audit workload types, the board should compare these prospective CBRs against actual returns annually. The board should make the results avail- able to Finance and the LAO and should also include them in the board’s annual report to the Legislature on the results of its audit and collection activities. If the board believes this information is confidential, it can cloak the identity of the individual audit workloads in its annual report to the Legis- lature as we have done in Appendix C. Moreover, the board should use the results of the comparison in future calculations of prospective CBRs. n 2244 California State Auditor Report 2002-124 California State Auditor Report 2002-124 2255 2266 California State Auditor Report 2002-124 California State Auditor Report 2002-124 2277 CHAPTER 2 The Board’s Performance Measures Are Insufficient to Justify Requests for New Audit Program Staff CHAPTER SUMMARY The Franchise Tax Board’s (board) total assessments from audits of personal income and corporation tax returns have provided the State an important source of potential revenue at relatively little cost, returning around $10 in assess- ments for every $1 of costs. Also, total audit program assessments of $10.4 billion for fiscal years 1992–93 through 2001–02 have exceeded budgeted amounts by almost $1.1 billion. However, this increase is not attributable to the 340 net audit positions the board received in those years. We computed the growth in audit assessments before and after these staffing increases. Our analysis isolates the impact of these new audit positions by eliminating audit assessments the board would complete even without the additional staff, such as follow-ups of Internal Revenue Service (IRS) leads and audits with possible large-dollar assessments. When we isolated the impact of these positions from the continuing efforts of the entire audit division for fiscal years 1992–93 through 2001–02, we found that assessments increased by $101 million at a cost of $127 million, returning 79 cents in assessments for every $1 spent. More recently, return on the new positions improved to $2.71 in assessments for each $1 of cost during the last four years of the above period. However, the $101 million of assessments has not been reduced for uncollectible accounts, and the $127 million of costs has not been increased to include the cost of collecting the accounts. Thus, the amount of additional cash receipts related to the additional cash investment is unknown. We believe economic conditions have affected the return of these positions. Another significant cause for the low return is that the board did not devote additional hours to assessment- generating audits. As we reported in a prior audit, in fiscal year 1995–96, after the board added 362 staff positions over four years, total hours for the audit division increased dramatically, but the number of hours spent on audits remained relatively unchanged. The board cannot identify where it assigned all the 2266 California State Auditor Report 2002-124 California State Auditor Report 2002-124 2277 additional staff, but in a 1997–98 report to the Legislature the board revealed that it had redirected 30 audit positions to information technology projects, which do not directly create assessments. Since fiscal year 1999–2000, budget control language now requires the board to secure prior approval from the Department of Finance (Finance) to reassign direct audit or collection program positions to other tasks. THE BOARD HAS HAD A POSITIVE RETURN ON ITS AUDIT PROGRAM, BUT SOME TYPES OF AUDITS HAVE HIGHER RETURNS THAN OTHERS The board’s audit program measures its performance in terms of audit assessments, which represent potential revenue. It tracks assessments for each of the primary types of audits it performs and prioritizes its workload based on potential returns. Although the amount of audit assessments fluctuated between fiscal years 1992–93 and 2001–02, on average, the board returned about $10 in assessments for every $1 of cost over the period. However, there is a wide range of returns among the various types of audits. The board conducts three primary types of audits on both personal income and corporation tax returns: field audits, desk For each $1 of cost, the audits, and audits that follow up on IRS leads. Desk audits are audits of corporation tax performed at the board’s Sacramento central office, while field returns produced about audits are conducted elsewhere—at a taxpayer’s place of busi- $59 in assessments for ness, for example. Historically, the returns between audit types IRS follow-up audits, but have differed dramatically, according to the board’s calculations. only $3 for desk audits. For example, from fiscal years 1992–93 through 2001–02, the average return on personal income tax audits stemming from IRS leads was $27 in assessments for every $1 of cost, whereas the average return on personal income tax desk audits was only $4. The various kinds of corporation tax audits show similar differences in average returns: for each $1 of cost, IRS follow-up audits returned on average $59 in assessments, while desk audits returned only $3. Figure 2 on the following page summarizes the board’s data on the average return for each type of audit during this period, and Appendix A details the returns for individual fiscal years. As described in Chapter 1, the historical returns presented in this figure reflect all costs including departmental overhead for the audit program. In compiling these data, the board indicates it 2288 California State Auditor Report 2002-124 California State Auditor Report 2002-124 2299 included some of the costs of auditing refund requests, but it did not include the resulting reductions in refunds. As a result, the benefits are somewhat understated. FIGURE 2 Average Assessments Per Dollar of Cost by Primary Audit Type Fiscal Years 1992–93 Through 2001–02 �� 2288 California State Auditor Report 2002-124 California State Auditor Report 2002-124 2299 �� Source: Franchise Tax Board’s annual operations reports for fiscal years 1992–93 through 2001–02. Data for fiscal years 2000–01 and 2001–02 are draft figures. Within each of these primary audit types, the board groups tax returns into numerous, discrete audit models with varying cal- culated rates of return. When selecting tax returns for audit, the board identifies audit models likely to have the highest returns and generally assigns these audits the highest priority. For exam- ple, 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 staffing those corporation tax audits that are likely to yield very large assessments. Therefore, we expect that the board generally uses its existing positions, not new posi- tions, to ensure these high-return audits are completed. THE BOARD BELIEVES THE AUDIT PROGRAM IS SUCCESSFUL WHEN TOTAL AUDIT ASSESSMENTS EXCEED PROJECTIONS The board measures the performance of the audit program as a whole, blending the contributions of both existing and new staff and comparing actual to projected assessments. Using this measure of performance, the audit program as a whole has been successful. As Table 6 on the following page indicates, actual audit assessments have exceeded estimates by a total of $1.1 billion between fiscal years 1992–93 and 2001–02. According to the board, the Legislature and Finance believe that measuring the performance of the audit program as a whole is an appropriate measure of the audit program’s performance, and we concur that the measure is important. However, this measure of performance fails to identify the incremental return from additional audit positions. While Finance believes that comparing total actual assessments to projections is an appropriate measure of the audit program’s overall performance, it also thinks that knowledge about mar- ginal performance would be useful since the overall return can vary for many different reasons and can mask the marginal performance. Although it agrees that comparing total actual Determining the audit assessments to projections provides good information, incremental contribution the Legislative Analyst’s Office also believes this may not be an of new auditors is adequate measure of the program’s performance without addi- important, but the tional information. board’s analysis falls short of this goal. Although its budget change documents justify new staff based on projected increases in assessments new auditors can generate less the cost of the new auditors, the board does not separately identify the workloads, such as desk audit returns or personal income tax field audit returns, that additional staff would be assigned to, and compare the incremental return from these workloads to the incremental cost of the additional auditors. 3300 California State Auditor Report 2002-124 California State Auditor Report 2002-124 3311 TABLE 6 The Board’s Comparison of Projected and Actual Audit Assessments Fiscal Years 1992–93 to 2001–02 (in Millions) Increase (Decrease) in Projected Audit Actual Audit Actual Over Budgeted Fiscal Year Assessments Assessments* Audit Assessments 1992–93 $ 855 $ 862 $ 7 1993–94 864 1,039 175 1994–95 806 851 45 1995–96 978 957 (21) 1996–97 1,063 1,252 189 1997–98 1,055 1,310 255 1998–99 1,034 1,207 173 1999–2000 908 1,043 135 2000–01 835 785† (50) 2001–02 842 1,057† 215 Totals $9,240 $10,363 $1,123 Source: Franchise Tax Board’s summary based on its annual operations reports, production reports, and workplans for fiscal years 1992–93 through 2001–02. * Actual assessments include reductions to claims for refund. Claims data are from the board’s production reports. † Data for fiscal years 2000–01 and 2001–02 are draft figures. THE RETURN FROM NEW POSITIONS WAS ORIGINALLY NEGATIVE, BUT HAS INCREASED RECENTLY Between fiscal years 1992–93 and 2001–02, the board received 340 net audit positions. Our analysis of the incremental benefit from these new positions for the entire period indicates they did not generate audit assessments that exceeded their cost, but the return has improved overall since fiscal year 1998–99. To determine the incremental benefit of the 340 net audit positions, we isolated their budgeted costs and the actual assess- ments associated with the audits to which the board would have likely assigned the new staff. From fiscal years 1992–93 through 1995–96, the board received authorization for 362 additional audit positions. We reported on the incremental benefit of these positions in our prior report, discussed in the Introduction. The board did not receive any more positions during fiscal years 1996–97 through 2001–02; it lost 22 unfilled audit positions 3300 California State Auditor Report 2002-124 California State Auditor Report 2002-124 3311 during fiscal year 1999–2000. The board received 45 audit posi- tions in fiscal year 2002–03, but it is too early to assess the effect of those positions, so we did not include them in our analysis. The 340 net positions the board received were intended to increase assessments through additional audits of personal income tax and corporation tax returns. However, not all of the additional positions directly generate audit assessments because only some are involved in performing audits. The additional positions also include tax technicians, legal counsel, temporary help, and support staff needed to keep up with the growth in the number of auditors. These support positions provide necessary services that enable tax auditors to work efficiently and com- plete more audits. The board’s policy is to first assign staff to audits with high projected returns, such as those resulting from IRS leads or corporation tax audits expected to result in large assessments, and to request new staff for lower-priority audits. Since the board’s policy is to staff these high-priority audits first, without the benefit of additional positions, our analysis adjusted the total assessment figures by eliminating assessments from such high-priority audits to isolate the return applicable to the new auditors. We also included as a benefit of the audit program the amount of the program’s reductions to requested refunds. Table 7 compares our adjusted annual audit assessments to a base year average, which we calculated averaging adjusted audit Although the new audit assessments for the two years before the board received the positions initially had a new positions. This comparison reveals that assessments from negative return for each the types of audits additional staff would likely perform have $1 of cost, their return increased by $101 million since the positions were authorized improved to $2.71 recently. beginning in fiscal year 1992–93, but that this gain was offset by costs of $127 million for the new positions, making the potential return on these positions negative, only 79 cents for every $1 of cost. It is important to note that the return on the additional positions shows improvement over the more recent fiscal years in our analysis. Between fiscal years 1998–99 and 2001–02, the new positions produced average assessments of $2.71 for every $1 of cost. However, the $101 million of assessments has not been reduced for uncollectible accounts, and the $127 million has not been increased to include the cost of collecting the accounts. Therefore, the amount of additional cash receipts related to the cash investment in the new positions is unknown. Historically, board records show that it has collected only 23 percent of personal income tax assessments 3322 California State Auditor Report 2002-124 California State Auditor Report 2002-124 3333 3322 California State Auditor Report 2002-124 California State Auditor Report 2002-124 3333 7 ELBAT mrofreP ylekiL dluoW ffatS lanoitiddA tahT stiduA no nruteR ot stnemtsujdA )esaerceD( esaercnI detsujdA latoT ytiroirP hgiH evomeR lanosreP latoT esaercnI detamitsE tiduA detsujdA ni xaT emocnI lanosreP stnemssessA tiduA dna xaT emocnI detegduB ni revO stnemssessA xaT noitaroproC dna snoitcudeR ddA dna xaT noitaroproC rebmuN latoT no nruteR serutidnepxE egarevA raeY esaB stnemssessA tiduA sdnufeR rof smialC ot stnemssessA tiduA snoitisoP fo tnemtsevnI )sdnasuohT ni( )sdnasuohT ni( )sdnasuohT ni( )sdnasuohT ni( )sdnasuohT ni( deddA raeY lacsiF 690,155 $ )718,302( $ 319,457 $ 19–0991 347,255 )419,734( 756,099 29–1991 919,155 )668,023( 587,278 egarevA raeY esaB )06.3( $ 338,5 $ )679,02( $ 349,035 )936,292( 285,328 351 39–2991 36.4 100,8 670,73 599,885 )731,714( 231,600,1 25 49–3991 )58.3( 819,41 )563,75( 455,494 )061,772( 417,177 88 59–4991 )87.71( 010,51 )019,662( 900,582 )520,036( 430,519 96 *69–5991 )93.4( 015,41 )676,36( 342,884 )704,555( 056,340,1 *79–6991 14.22 015,41 261,523 180,778 )020,853( 101,532,1 *89–7991 29.41 015,41 454,612 373,867 )734,533( 018,301,1 99–8991 24.01 003,31 075,831 984,096 )219,672( 104,769 )22( 0002–9991 )31.22( 003,31 )672,492( 346,752 )132,834( †478,596 10–0002 35.6 003,31 248,68 167,836 )143,743( †201,689 20–1002 sraey lacsfi rof slatoT hguorht 39–2991 )46.0( 287,27 )986,64( 528,462,3 )883,035,2( 312,597,5 263 89–7991 sraey lacsfi rof slatoT hguorht 99–8991 17.2 014,45 095,741 662,553,2 )129,793,1( 781,357,3 )22( 20–1002 sraey lacsfi rof slatoT hguorht 39–2991 97. $ 291,721$ 109,001$ 190,026,5$ )903,829,3($ 004,845,9$ 043 20–1002 .atad lanoitidda dna ,stnemucod tegdub ,2.81189 tropeR stiduA etatS fo uaeruB ,20–1002 hguorht 39–2991 sraey lacsfi rof stroper snoitarepo launna s’draoB xaT esihcnarF :secruoS .stroper noitcudorp s’draob eht morf era atad smialC ot eud etaruccani erew draob eht yb dedivorp ylsuoiverp stnuoma esuaceb ,2.81189 ,troper tidua roirp ruo ni detneserp stnuoma morf reffid sraey eseht rof stnemssessa detsujdA * .srorre metsys gnitnuocca .serugfi tfard era 20–1002 dna 10–0002 sraey lacsfi rof ataD † in the first year, 50 percent by the end of three years, and 58 percent by the end of five years. The board indicates it does not have a system in place to track the collection rate of corporation tax audit assessments. However, based on a five-month study the board completed in 1998, it determined that the collection rate on these assessments was 86 percent. According to the board, it has no reason to believe that the collection rate has changed since the study was completed. Our analysis does not account for unquantifiable benefits provided by the audit program. The overall intent of the audit program is to establish an audit presence in the taxpaying community as a means of encouraging compliance with tax laws and increasing overall self-assessed revenue. When requesting additional positions, the board provides the dual justification of projected increases to General Fund revenues, as well as increased audit presence. The Number of Audit Hours and Changes in the Economy Have Affected the Amount of Assessments Several factors contributed to the low return on the new audit positions. We believe some factors not in the board’s control— changes in the economy, for example—have affected the magni- tude of assessments. However, as our prior audit indicated, the number of hours actually spent performing audits is a significant factor that is at least partially under the board’s control. Even after allowing for delays caused by time required to hire and train the new staff, we would expect to see an increase in audit hours. From 1992–93 through 1995–96, the board received the following additional positions annually: 153, 52, 88, and 69. Yet as Figure 3 on page 35 shows, the number of personal income tax and corporation tax audit hours did not increase accordingly. Although total hours for the entire audit division increased suf- ficiently to reflect a limited number of additional audit staff, the number of hours actually spent performing audits in fiscal year 2001–02 differed little from hours spent in fiscal year 1992–93. As reported in our previous audit, the board could not explain where it assigned all of the new staff, but it did give some infor- mation on the assignments in a fiscal year 1997–98 report to the Legislature. The report disclosed that the board had redirected 30 audit positions to information technology projects. While serving an important support function for audits, these positions do not directly generate assessments. 3344 California State Auditor Report 2002-124 California State Auditor Report 2002-124 3355 FIGURE 3 Hours Spent on Personal Income Tax and Corporation Tax Audits Fiscal Years 1992–93 Through 2001–02 �� 3344 California State Auditor Report 2002-124 California State Auditor Report 2002-124 3355 �� Sources: The board’s summary of Audit Division hours for fiscal years 1998–99 through 2001–02; Bureau of State Audits Report 98118.2 for prior fiscal years. Enacted in fiscal year 1999–2000 and each subsequent year, budget control language referred to as “Provision 1” requires the board to secure prior approval from Finance before transferring direct auditing or collection program positions to other areas. Since the provision’s establishment, the board asserts that it has not redirected any positions out of direct audit activities. However, in the course of Finance’s review of the board’s budget change proposals in the fall of 2000, it found that the board had redirected five positions from auditing and collection activities in violation of Provision 1. As a result, Finance required the board to return these positions to audit and collection activity. At that time, Finance also notified the Leg- islature of the issue and the remedies it took. As shown in Figure 4, direct audit hours for each auditor have gradually increased since fiscal year 1997–98. FIGURE 4 Annual Productivity Per Auditor �� 3366 California State Auditor Report 2002-124 California State Auditor Report 2002-124 3377 �� Source: The board’s summary of Audit Division hours. Note: According to the board, the difference between available hours and direct audit hours represents time spent on tasks such as training, supervision, audit selection, and other projects. *After allowances for various kinds of leave. PROSPECTIVE AUDIT PROGRAM COST-BENEFIT RATIOS DO NOT REFLECT HISTORICAL PERFORMANCE Another potential cause for the lower than expected return from additional auditors is the different ways costs are figured in the pro- spective CBRs used to help justify new positions and in the histori- cal data that report actual performance. As discussed in Chapter 1, the board’s historical performance measure of returns includes the full effect of indirect costs, including departmental overhead, but the prospective CBRs do not. Thus, when the full departmental Including departmental overhead costs are taken into account, certain prospective CBRs overhead cost reduces drop below the anticipated return of $5 in assessments generated for the calculated benefit every $1 of cost. to less than $5 for each $1 of cost for certain Table 8 on the following page illustrates the difference between audit types. the historical returns of general audit types reported in operations reports and their projected returns given in the board’s workplans, and shows the effect of reducing the projected CBRs by the amount of departmental overhead costs. For fiscal years 1998–99 through 2001–02, the board returned $9.50 in audit assessments for every $1 of audit costs. However, over the same period, the board pro- jected a return of almost $12 for every $1 invested, which is more than 20 percent higher. When we deflated the board’s projections by actual departmental overhead costs for the four-year span, we found that had the board included full departmental overhead costs, the total actual return in assessments would closely resemble the projections. For the period, the board projected a return of $11.84 in assessments per $1 of cost; departmental overhead costs reduce this figure to $9.48, which is close to the actual return of $9.50 in assessments for each $1 of cost. When individual audit types are examined, however, the variance is much greater, and the workplan projections fail to mirror historical return. The types of audits to which we would expect new staff to be assigned—desk audits and personal income tax field audits—do not meet the $5 or greater projected returns. For example, the average assessment per $1 invested in personal income tax desk audits over the period was $3.87, whereas the board estimated that they would return $6.36. Even after deflating the workplan projections by departmental overhead costs, actual assessments per dollar of cost were still $1.75 less than originally projected. The board believes that these differences generally arise from adjustments the audit program makes to historical data ultimately reported in operations reports. According to the board, the adjust- ments are made to correct misallocated charges and miscoded revenue and to better match costs to benefits. The audit program then uses the modified data as the basis for the projections that are reflected in its workplans and in Table 8 on the following page. We find it difficult to understand the board’s explanation: Why would the board publish information in its operations reports that it believes is inaccurate? Further, if the audit program corrects errors in the financial reporting system when it recalculates the basis for projections, we would expect that the board would use the corrected 3366 California State Auditor Report 2002-124 California State Auditor Report 2002-124 3377 data in the operations reports, which it publishes after it prepares the workplans. Finally, we do not understand why the board toler- ates errors in such a vital function as its financial reporting system, which is the basis for the published cost data. TABLE 8 Comparison of Projected and Actual Returns on Audit Types Fiscal Years 1998–99 Through 2001–02 Four-Year Average Remaining Less Adjusted Difference Not Workplan Application of Workplan Explained by Operations Projected Departmental Projected Departmental Report Actual Description Return Overhead* Return Overhead Return* Personal Income Tax Audit activities: Desk audits $ 6.36 $0.74 $ 5.62 $ 1.75 $ 3.87 IRS follow-up audits 33.96 3.45 30.51 3.85 26.66 Field audits 5.10 0.76 4.34 0.11 4.23 Personal income tax total 10.14 1.28 8.86 0.68 8.18 Corporation Tax Audit activities: Desk audits 6.51 2.09 4.42 3.14 1.28 IRS follow-up audits 31.22 2.55 28.67 (40.18) 68.85 Field audits 13.01 3.32 9.69 (1.07) 10.76 Corporation tax total 13.34 3.39 9.95 (0.42) 10.37 Total desk audits 6.39 1.06 5.33 2.50 2.83 Total IRS follow-up audits 33.19 3.19 30.00 (4.56) 34.56 Total field audits 10.75 2.44 8.31 (0.99) 9.30 Total all audits $11.84 $2.36 $ 9.48 $ (0.02) $ 9.50 Sources: The board’s annual audit workplans and operations reports for fiscal years 1998–99 through 2001–02. * Data for fiscal years 2000–01 and 2001–02 are draft figures. In contrast to the projections for desk audits and personal income tax field audits, the board’s average four-year projected return of about $29 for IRS follow-up audits of corporation taxes was less than half the historical return of almost $69 for the same period. As Appendix A indicates, the four-year historical returns, from fiscal years 1998–99 through 2001–02, for this type of audit were approximately $69, $63, $64, and $78, and since fiscal year 1992–93 have ranged between about 3388 California State Auditor Report 2002-124 California State Auditor Report 2002-124 3399 $25 and $214. When we asked why it consistently established targeted returns well below the recent historical rate, the board explained that it is dependent on audits forwarded from the IRS, leaving the audit selection criteria beyond the board’s control and resulting in wide fluctuations in assessments from this workload. The board notes that recently the IRS has chosen to do fewer audits with higher CBRs, but in the year prior to the period depicted in Table 8 the corporation tax IRS follow-up audits returned about $25. Given the unpredictability of IRS policies and performance, the board has chosen to be conserva- tive in its revenue projections for this workload. Regardless of the rationale, the effect of the low projection is Low projections of the to make it more likely that the board will meet or approach return on IRS follow-up its overall projected assessments. Because the board measures audits make it more likely the success of the audit program as a whole—by determining that the board will meet whether actual assessments exceed estimated assessments for the total assessment goals. program as a whole—the low projection for corporation tax IRS follow-ups enables the board to exceed its projections in total, even though those audit types to which we would expect the board to assign new staff return below $5 in assessments. In Chapter 1, we reiterated a recommendation from our prior audit about the format the board should use when requesting additional audit staff. Our suggestion was that the format should detail the types of activities new auditors will per- form, as well as the projected assessments and historical assessments resulting from these activities. The analysis above of four-year projected and actual returns further underscores the value of this recommendation. RECOMMENDATIONS If the board believes that information it publishes in its opera- tions reports is not accurate, even though it is based on the board’s financial accounting system, the board should do the following: • Ensure that its financial accounting system reports accurate information, and • Correct data it believes to be inaccurate before it publishes the information in its operations reports. n 3388 California State Auditor Report 2002-124 California State Auditor Report 2002-124 3399 4400 California State Auditor Report 2002-124 California State Auditor Report 2002-124 4411 CHAPTER 3 The Board’s Performance Measures Are Insufficient to Justify Requests for New Collection Program Staff CHAPTER SUMMARY Despite the positive overall return of its tax collection program (collection program), the Franchise Tax Board (board) is unable to demonstrate that additional collec- tion program positions have returned their projected revenue. Between fiscal years 1998–99 and 2001–02, the board received an additional 175 collection program positions, promising a total of $179 million in increased revenue over that period. However, the board’s analysis of the return on its new positions does not demonstrate that increases in collection program revenue resulted from the additional staff because it does not isolate the workloads of the positions and compare the pro- jected returns over a base year against actual results. Similarly, we were unable to measure the incremental benefit of the new positions because of the board’s limited data on workload cat- egories and a change in the board’s accounting for revenues. Also, although the board justified its fiscal year 2001–02 staff increase using a workplan process that prioritizes accounts based on a cost-benefit ratio (CBR), the board actually assigns work to staff according to risk and yield factors calculated by its new Accounts Receivable Collection System (ARCS). The director of the board’s special programs bureau, who has broad knowledge of the collection program, indicates that the board has continued to use its workplan process for justifying staff increases because, until recently, ARCS had not amassed sufficient historical data to allow the board to use it to justify new staff. However, until the board can use the same data to both prioritize workload to assign staff and to justify new collection program positions, it cannot have reliable data to determine the effectiveness of new staff. Finally, the board is not fully using its collection program salary funding to fill authorized positions. In fiscal year 1999–2000, separate merit salary adjustment (MSA) funding for the board was ended, leaving the board to find another way to pay for its MSAs. In fiscal years 2000–01 and 2001–02 the actual 4400 California State Auditor Report 2002-124 California State Auditor Report 2002-124 4411 collection program hours worked decreased by 5 percent and the board applied the additional savings toward, among other expenses, MSAs. To achieve this additional savings, the board has allowed some collection program positions to remain unfilled, even some that the board’s budget control language requires it to fill expeditiously. THE FRANCHISE TAX BOARD HAS HAD A POSITIVE RETURN ON ITS COLLECTION PROGRAM Providing about 4 percent of the board’s total tax revenue, the collection program returned around $19 in actual revenue for every $1 of cost between fiscal years 1998–99 and 2001–02. The board’s overall historical assessment of its collection program appears to include all appropriate costs and benefits, with the exception of certain revenues that the board claims were incorrectly counted as collection program revenue for years prior to fiscal year 2000–01. As indicated in Chapter 1, the board measures the historical overall collection program return by comparing all collection program revenues to all related costs. Costs include all direct col- lection program costs, including staff salaries and benefits, and indirect support costs incurred in other areas—for example, the technology, legal, or various administrative branches. Finally, the board includes an overhead allocation for departmental costs such as rent and utilities. In its annual operations reports, the board compares the resulting cost figure to collection pro- gram revenue to determine the return on each dollar of cost. At a very high level, as depicted in Appendix B, the board tracks the revenue it receives from different collection program Manual collection workload types. The board records collection program revenue activities involve as either personal income tax or corporation tax and further more intensive staff subdivides these revenue categories by automated and manual intervention, such as collection activities. It classifies collection program revenue as locating the taxpayer, automated when it results from computer-generated activities identifying valuable such as sending bills and notices, levying bank accounts, and assets, and seeking to garnishing wages, in addition to any staff interaction such as arrange payment. responding to phone calls or other correspondence that results from automated notices. Manual collection activities involve more intensive staff intervention, such as locating the taxpayer, identifying valuable assets, and seeking to arrange payment. According to the director of the board’s special programs bureau, about 15 percent of the roughly 1 million collection program accounts processed annually require manual involvement. 4422 California State Auditor Report 2002-124 California State Auditor Report 2002-124 4433 Historically, returns between types of collection program activ- ity have differed significantly. For example, between fiscal years 1998–99 and 2001–02, the average return on automated collec- tion program activities for corporation taxes was about $52 of revenue for every $1 of cost, whereas the average return on the manual collection program activities for personal income taxes was approximately $10 for every $1 of cost. Figure 5 summarizes average returns for high-level groupings of collection program workloads during this period. FIGURE 5 Average Return Per Dollar of Cost by Collection Program Revenue Type Fiscal Years 1998–99 Through 2001–02 �� 4422 California State Auditor Report 2002-124 California State Auditor Report 2002-124 4433 �� Source: Franchise Tax Board’s annual operations reports for fiscal years 1998–99 through 2001–02. Data for fiscal years 2000–01 and 2001–02 are draft figures. We were unable to calculate returns per dollar of cost for collec- tion program workloads by their origins within the board, such as those originating from audit assessments, because the board does not track all the necessary information at this level of detail. Also, we did not adjust the data in Figure 5 or Figure 6 on page 48, for what the board states is a change in accounting treatment in fiscal year 2000–01 that is partly responsible for a drop of approximately $247 million in personal income tax revenues credited to the collection program. As discussed more fully later, the board states that collection program revenues in prior years were overstated. MEASURING THE INCREMENTAL BENEFIT OF ADDITIONAL COLLECTION PROGRAM STAFF PROVES ELUSIVE Between fiscal years 1998–99 and 2001–02, the board received authorization for an additional 175 collection program posi- Because of various tions, promising to increase collection program revenue by limitations to the $179 million during the period. However, in part because the board’s data, we cannot board cannot quantify the effect of a change in accounting determine the incremental treatment for personal income tax collection program revenues revenue generated by new in fiscal year 2000–01, we cannot verify whether this promise collection program staff. was fulfilled. Moreover, in its own evaluation of the return on these additional positions, the board does not isolate revenue pools potentially affected by the new positions to compare their projected and actual return. Rather, to demonstrate the impact of additional collection program positions, the board com- pares total actual collection program revenue to total projected revenue for each fiscal year. Similar to our concerns with the board’s measurement of audit program performance, we find this type of an analysis inconclusive because it does not measure the incremental benefit of new positions and does not establish a base year for comparison. Finally, we were unable to perform our own analysis of the incremental benefit of the new collection program positions both because of the change in accounting treatment mentioned above and because the board lacks sufficient detailed revenue and cost data for workload categories. The Board Received Authorization for 175 Collection Program Positions for Varying Purposes in Recent Years Job functions vary for the 175 new collection program positions authorized for the board. In fiscal year 1998–99, the board received 42 positions to continue implementing the automated ARCS system begun in fiscal year 1997–98. In fiscal years 1999–2000 and 2000–01, the board netted an additional 10 collection program positions to develop an automated system for the collection of limited liability corporation taxes. For fiscal year 2001–02, the board received 4444 California State Auditor Report 2002-124 California State Auditor Report 2002-124 4455 123 positions, justified by the board’s analysis that account types projected to have returns at or above $5 for every $1 spent were understaffed. Table 9 summarizes the board’s total projected revenues and costs associated with these increases in staffing and other resources. For example, the board’s estimated costs for the 42 positions the board received in fiscal year 1998–99 amount to $10.1 million over four years. Other related costs for fiscal years 1998–99 and 1999–2000 primarily consist of expected consult- ing fees through fiscal year 2001–02 for the implementation of information technology systems that are intended to assist with the collections.3 TABLE 9 Estimated Expenditures and Projected Revenues for 175 New Collection Program Positions and Other Resources 1998–99 Through 2001–02 Estimated Expenditures Position Total Projected Fiscal Year Increase Positions Other Costs Revenues 1998–99* 42 $10,147,000 $18,477,389 $109,500,000 1999–2000† 7 1,063,427 171,818 2000–01 3 260,400 16,000,000 2001–02 123 7,658,000 53,800,000 Totals 175 $19,128,827 $18,649,207 $179,300,000 Source: Franchise Tax Board’s budget change documents for fiscal years 1997–98 through 2001–02. * Estimated expenditures include significant ARCS implementation costs in fiscal years 1998–99 and 1999–2000; however, the expenditures do not include costs associated with the fiscal year 1997–98 budget change proposal for ARCS. Projected revenues are not solely the result of the 42 positions, but are the aggregate amount attributable to ARCS, for which the board had already received 39 positions in fiscal year 1997–98. † In fiscal year 1999–2000, the board received 22 positions and $1,792,000 to develop an automated return validation, billing, and accounting system for the limited liability corporation tax program. Seven of these were collection program positions. The table proportionately reflects the estimated expenditures for the seven positions, but shows all the projected revenue, which the board considers attributable solely to collection program efforts. 3 Table 9 does not reflect additional collection program positions received in fiscal year 2002–03 because actual data on the potential effect of these positions are not yet available. In fiscal year 2002–03, the board gained a net 77 additional permanent positions and 51 three-year, limited-term positions to pursue collection program workloads that the board determined have a 1:5 or higher CBR. The board also received 34 one-year, limited-term positions to accelerate the payment of delinquent accounts. 4444 California State Auditor Report 2002-124 California State Auditor Report 2002-124 4455 Limitations in the Board’s Data Prevent an Analysis of the Return on Its Investment in Additional Staff The board’s analysis of its return on the additional collection program positions received between fiscal years 1998–99 and 2001–02, depicted in Table 10, does not demonstrate that any increases in collection program revenue were the result of the additional resources. For example, it does not show that the new positions in fiscal year 2001–02 actually generated the $53.8 million in projected revenues. The board evaluates the return on its new collection program positions simply by comparing total projected revenue to total actual revenue. For example, for the fiscal year 2001–02 positions, because the variance between total projected revenue and total actual revenue ($83.5 million) exceeds the board’s projected increase in revenue resulting from new positions ($53.8 million), the board considers its new positions as having returned their promised benefit. We believe TABLE 10 The Board’s Analysis of the Return on Its New Collection Program Positions Fiscal Years 1998–99 Through 2001–02 Number of Projected Positions Revenue Actual Revenue Variance Added* 1998–99 Base (without new positions) $1,687,134,677 Augmentation (from new positions) 16,000,000 Total (with new positions) 1,703,134,677 $1,893,057,216 $189,922,539 41.5 1999–2000 Base (without new positions) 1,793,745,209 Augmentation (from new positions) 8,000,000† Total (with new positions) 1,801,745,209 1,805,963,513 4,218,304 7.0 2000–01‡ Base (without new positions) 1,602,016,316 Augmentation (from new positions) 8,000,000 Total (with new positions) 1,610,016,316 1,608,578,192 (1,438,124) 5.0 2001–02‡ Base (without new positions) 1,574,100,000 Augmentation (from new positions) 53,800,000 Total (with new positions) $1,627,900,000 $1,711,399,022 $ 83,499,022 119.0 Source: Franchise Tax Board’s analysis based on workplans and operations reports. Actual revenue for fiscal years 2000–01 and 2001–02 are draft figures. * Number of positions differs slightly from those in our analysis because the board does not round or net positions, as we do in Table 9, and does not include four positions for fiscal year 2001–02 that were part of the same budgetary action providing the 119 positions. † Based on our analysis of budget documents, we disagree with this $8 million increase to the board’s revenue base. ‡ Revenue for these years is affected by the change in accounting for certain revenues discussed on pages 47 and 48. 4466 California State Auditor Report 2002-124 California State Auditor Report 2002-124 4477 a more appropriate analysis focuses on revenue pools potentially affected by the new staff. For example, if the board believes that new positions will result in additional revenue from certain manual workloads, a more effective way to demonstrate the return on the positions would be to isolate the workloads within this classification where it expects additional revenue and compare its projected increases over a base year against actual results. Because the board does not track collection program revenues or related costs beyond high-level categories, we could not isolate to a sufficient level of detail potentially affected revenue pools. Appendix B displays the level of revenue and cost detail the board can provide. Our effort to isolate the effect attributable to the new collection program staff was further complicated by the board’s change in the accounting treatment of certain personal income tax revenues, described below, and its inability to document the impact of the change. As a result, we were unable to conduct an independent analysis of the incremental return from each staffing and other resource increase for the collection program for fiscal years 1998–99 through 2001–02. Despite the increase in collection program staff and projected revenue, in fiscal year 2001–02 the board collected only $2 mil- lion more in collection program revenue than it did in fiscal year 1997–98. Again, we believe it is probable that economic factors influenced collections. Also, according to the board, this compari- son does not reflect the true return of these positions because of a A change in the board’s change in the board’s method of accounting for certain collection treatment of certain program revenues that occurred in fiscal year 2000–01. The board revenues complicates explains that, in fiscal year 2000–01, it changed its accounting an analysis of the for certain personal income tax revenues. Under the board’s prior returns on new collection revenue reporting system, if a taxpayer had a collection program program positions. account with an outstanding balance, any money received from that taxpayer would be credited as collection program revenue, even if the taxpayer simply remitted payment of a current-year liability. For example, prior to fiscal year 2000–01, if a taxpayer owed back taxes from a prior-year audit and remitted a payment for current-year taxes, this payment would be counted as col- lection program revenue rather than current-year self-assessed revenue. Using the new system, the board would appropriately identify the payment as current-year self-assessed revenue, which would not be credited to the collection program. Although the board informed us that the subsequent reduction in reported collection program revenue from fiscal year 1999–2000 was $247 million, it was unable to tell us how much of the 4466 California State Auditor Report 2002-124 California State Auditor Report 2002-124 4477 FIGURE 6 Collection Program Revenues for Personal Income and Corporation Taxes Fiscal Years 1995–96 Through 2001–02 �� 4488 California State Auditor Report 2002-124 California State Auditor Report 2002-124 4499 �� reduction was caused by the change in accounting. Figure 6 illustrates that a decline in personal income tax collection program revenue did occur in fiscal year 2000–01. According to the board, after the passage of time and the physical moves of personnel and files, the board cannot locate the documentation prepared during the 1998–2000 time period describing the exact components of the change. �� Source: Franchise Tax Board’s operations reports for indicated fiscal years. Data for fiscal years 2000–01 and 2001–02 are draft figures. THE BOARD’S JUSTIFICATION FOR NEW COLLECTION PROGRAM POSITIONS DOES NOT REFLECT ITS CURRENT PROCESS FOR ASSIGNING WORK Unlike the audit program, which both justifies new positions and assigns work based on a workplan process that prioritizes work according to a CBR, the collection program currently uses a similar workplan process only to justify its increases in collection program positions. In actually assigning work, the board relies on the recently implemented ARCS to rank accounts according to various risk and yield factors that predict the likelihood of collection as well as the ultimate amount the system expects to collect. According to the director of the board’s special programs bureau, now that the collection program has nearly two years of collecting experience using ARCS, analysis is under way to The board uses a use data from that system to justify future staffing needs. This manually prepared goal is appropriate, but the board should also ensure that its workplan to identify revised process considers all the costs of collection in addition to and justify collection the risk and yield components when justifying future increases program staffing in positions. needs, but relies on factors calculated by Each year, for budgeting purposes, the board prepares a collec- its automated system to tion program resource workplan for the upcoming fiscal year. actually assign work The board budgets its staff resources in terms of hours. Within to staff. the workplan the board assigns available staff hours to various direct, support, and overhead collection program activities, budgeting staff hours to required workloads first and apply- ing the remaining staff hours to discretionary workloads. The board’s final allocation is to certain accounts requiring manual intervention, which the board prioritizes according to CBRs. The board used the above process to justify additional collection program positions in fiscal years 2001–02 and 2002–03 based on work projected to generate at least $5 of revenue for each $1 of cost the board identified following its workplan process. While waiting for the new ARCS to amass sufficient historical data to provide more detailed and reliable revenue projections, the board continued to use the workplan process to justify addi- tional positions, relying on actual revenue data from fiscal year 1998–99 that it adjusted each year for changes in tax law and the economy. The board completed implementation of ARCS for personal income and corporation taxes in fiscal year 2000–01. 4488 California State Auditor Report 2002-124 California State Auditor Report 2002-124 4499 Although it justifies additional staffing needs based on the workplan process, the board actually allocates collection program work based on automated risk-yield calculations performed by ARCS. In contrast to the workplan process that uses CBRs based on estimated costs and projected revenue, ARCS prioritizes collecting activities by likelihood of collection (risk) and the amount the system expects to collect (yield). Figure 7 provides a simple illustration of how ARCS determines collecting activities. When an account enters ARCS for collection, the system scores its risk using a wide range of account characteristics, such as account balance, taxpayer filing history, time since the last payment, the number of payments made in recent years, the accuracy of the taxpayer’s address, and other factors. Based on the risk score, ARCS classifies the account as high, moderate, or low risk. The system also computes each account’s anticipated yield, which it also classifies as high, moderate, or low. FIGURE 7 Accounts Receivable Collection System (ARCS) Determines Collection Actions Based on Risk and Yield Yield Low Medium High Low Risk Medium High Generally collected through automated activity. Automated activity, with some staff activity generally reacting to taxpayer questions or needs. Generally require more proactive staff activity. Using this risk and yield information, ARCS assigns accounts to the functional area best suited to the characteristics of each case. A functional area is a predefined set of actions designed to collect on accounts displaying similar characteristics. Low- risk accounts include those that will likely result in payment with minimal intervention from collection program staff. For example, a taxpayer will receive a computer-generated 5500 California State Auditor Report 2002-124 California State Auditor Report 2002-124 5511 notice with an explanation of the debt and is expected to pay promptly. High-risk accounts require more vigorous collection activity such as staff intervention to locate the taxpayers or their assets. According to the manager of the board’s program administration and analysis section, as additional information about an existing account enters ARCS, the system continuously rescores the risk and yield factors. The director of the board’s special programs bureau has indicated The board plans to refine that, until recently, the newly implemented ARCS had not its process for justifying accumulated sufficient historical performance data to allow the new collection program board to refine its process for justifying additional collection staff to reflect current program staffing requests and that, in the interim, the board has workflow strategies. used its workplan process for budgeting purposes. According to the same director, now that the collection program has nearly two years of collecting experience using ARCS, analysis is under way to use data from that system to justify future staffing requests that reflect current workflow strategies. In fact, the board plans to use a revised process to forecast and validate collection program performance data beginning with fiscal year 2003–04. As the board considers alternatives for justifying additional staff- ing requests based on ARCS data, it should ensure that its revised process includes costs as well as the risk and yield components. The board could use its risk and yield calculations to project the revenue related to staffing requests, while estimating the related costs according to a model that includes all costs, as well as an allocation for departmental overhead. Thus, the goal is to have staffing requests for the collection program reflect how the collection program staff completes its work and also to give a complete picture of the projected revenues and estimated costs against which decision makers could measure performance. THE BOARD LEAVES SOME APPROVED COLLECTION PROGRAM POSITIONS UNFILLED The board is not using all of its funding for collection program salaries to actually fill authorized positions, but is instead using some funding for other costs. Periodically, the board rewards employees for meritorious performance through pay increases, or MSAs, above the initial salary funding for their positions. Before fiscal year 1999–2000, the board received budget augmentations to fund its MSAs, but beginning in fiscal year 1999–2000, the board’s MSA funding ended. The difference between the total hours collection program staff worked and 5500 California State Auditor Report 2002-124 California State Auditor Report 2002-124 5511 the total budgeted hours for the collection program increased by 5 percent shortly after the board lost its separate funding for MSAs. In the same fiscal year, however, the budget control lan- guage required the board to “expeditiously fill” direct collection program positions. When the board requests both additional positions and the associated funding, it makes a standard state adjustment for “salary savings” that automatically reduces its funding request by 5 percent to account for funding it is unlikely to spend on the additional positions during the normal course of business because of turnover or the time it takes to fill new positions. For example, if the board requests 100 positions that cost $50,000 each, it will request funding to pay only for the equivalent of 95 positions, or $4,750,000, even though it receives authorization for the full 100 positions. Thus, budgeted hours as shown in Figure 8 already reflect this standard 5 percent reduction, and the difference between budgeted and actual hours reflects additional savings. The board indicates that expenses it finances through these savings fluctuate from year to year because of retirements, new hires, the time it takes to hire staff, MSAs, and other unfunded salary adjustments arising from upgrading and reclassifying staff. Since the loss of its separate MSA funding, the board requires each branch to achieve savings to pay for the branch employees’ MSAs, allowing them to realize the savings from unfilled positions. The The board believes state board believes state departments must leave positions vacant or departments must leave they will overspend their salaries and wage budgets. However, positions vacant to avoid Government Code Section 12439 requires that positions that overspending salaries and are continuously vacant for six months be eliminated and the wage budgets. Department of Finance (Finance) recently began eliminating those positions in state departments. Finance believes that it is inappropriate as a long-term strategy to leave positions unfilled without addressing the underlying budget problem. However, in short-term situations, in order to not overexpend their budgets, departments may be required to hold positions open longer than normal recruitment times. In general, Finance’s position is that given a stable base of positions, normal turnover means that employees who receive MSAs and are at higher salary costs in a range are eventually replaced with employees who are at the bottom step. Where this approach does not work is when a department is steadily adding employees who at the bottom step. Then, as that group of new positions “ages” in class, it has higher salary costs driven by MSAs. Finance occasionally funds MSAs when departments can demonstrate real need; however, it generally guards against funding voluntary upgrades that did not come through the budget change proposal process. 5522 California State Auditor Report 2002-124 California State Auditor Report 2002-124 5533 FIGURE 8 Budgeted and Actual Collection Program Workload Hours Fiscal Years 1998–99 Through 2001–02 �� 5522 California State Auditor Report 2002-124 California State Auditor Report 2002-124 5533 �� Source: Franchise Tax Board’s summary based on workplans. *Budgeted hours already reflect the standard state funding reduction of 5 percent. For the board to be consistent with the intent of the budget control language and with Finance, it should not as a long-term strategy leave collection program positions unfilled beyond the normal time it takes to fill a vacant position. This should prevent an increased gap between budgeted and actual hours. Figure 8, which is based on information the board provided to us, illustrates that prior to losing its funding for MSAs, the collection program’s actual hours worked were about 5 percent lower than budgeted hours, and the gap subsequently increased by an additional 5 percent. Therefore, in addition to the stan- dard funding reduction of 5 percent, by fiscal year 2000–01 the board was applying salary and wage funding related to the 10 percent gap in hours to finance MSAs and other expenditures. RECOMMENDATIONS To better measure the effectiveness of its additional positions, the collection program should develop a methodology for determining the incremental return of new collection program positions received in any given year. This type of analysis should isolate changes over a base year in revenue pools that are affected by the new positions and compare the resulting revenue against all costs resulting from the new positions. To more accurately represent how it actually allocates resources, the collection program should continue to develop a method- ology based on ARCS for justifying future collection program positions. The revised process should include all relevant costs, including an allocation for departmental overhead, in addition to ARCS’ risk and yield factors. The estimated expenditures and projected revenues related to each new staffing request should be easy to compare against actual results. For the board to be consistent with the intent of budget control language and Finance, it should not as a long-term strategy leave collection program positions unfilled beyond the normal time it takes to fill a position. 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 government auditing standards. We limited our review to those areas specified in the audit scope section of this report. Respectfully submitted, ELAINE M. HOWLE State Auditor Date: May 13, 2003 Staff: Lois Benson, CPA, Audit Principal Michael Tilden, CPA Michael K. Adjemian Almis Udrys 5544 California State Auditor Report 2002-124 California State Auditor Report 2002-124 5555 APPENDIX A Historical Returns of the Audit Program by Audit Type This appendix provides a breakdown of the Franchise Tax Board’s (board) return on its audit program for fiscal years 1992–93 through 2001–02 by audit type. The board defines its return on audit activities based on net assessments per dollar of cost. As explained in Chapter 2, the board con- ducts three primary types of audits for both its personal income tax and corporation tax programs: field audits, desk audits, and audits that follow up on Internal Revenue Service (IRS) leads. As displayed in this appendix, over the last 10 years IRS follow- up audits have generated the highest returns. For example, personal income tax IRS follow-up audits have averaged $27.06 in net assessments per $1 of cost, with returns for individual fiscal years ranging from $15.97 to $44.37 in net assessments per $1 of cost. Similarly, corporation tax IRS follow-up audits averaged $59.30 in net assessments per $1 of cost, with returns ranging from $25.28 to $213.42 in net assessments per $1 of cost. The board’s corporation tax field audits have also performed well, averaging $11.76 in net assessments per $1 of cost over the 10-year period. Because its best-performing types of audits receive the highest staffing priority, the board would pursue these workloads regardless of whether it received any new auditors. Conversely, over the last 10 years, the board’s lowest-performing types have been personal income tax desk and field audits and corporation tax desk audits with average returns of net assessments per $1 of cost of $3.85, $3.66, and $2.99, respectively. In fact, at times corporation tax desk audits did not generate net assessments in excess of cost. For example, in fiscal years 1995–96 and 2000–01, corporation tax desk audits only returned $0.59 and $0.81 in net assessments for every $1 of cost, respectively. Again, because the board’s higher-return audit types receive staffing priority, new auditors would likely be used to pursue these lower-return audit types. Furthermore, the returns described above and presented in this appendix are averages for the various audit types, and each audit type contains numerous specific workloads. Therefore, the returns 5544 California State Auditor Report 2002-124 California State Auditor Report 2002-124 5555 associated with the lowest performing workloads that new staff would likely be assigned to would generally be lower than the averages presented in this appendix. Finally, it is important to note that net assessments presented in this appendix have not been reduced for uncollectible accounts, and costs have not been increased to include collection costs. Both of these adjustments would reduce the returns. 5566 California State Auditor Report 2002-124 California State Auditor Report 2002-124 5577 5566 California State Auditor Report 2002-124 California State Auditor Report 2002-124 5577 1.A XIDNEPPA epyT tiduA yb margorP tiduA eht fo snruteR lacirotsiH 49–3991 raeY lacsiF 39–2991 raeY lacsiF reP stnemssessA teN stsoC stnemssessA teN reP stnemssessA teN stsoC stnemssessA teN tsoC fo ralloD )sdnasuohT ni( )sdnasuohT ni( tsoC fo ralloD )sdnasuohT ni( )sdnasuohT ni( noitpircseD xaT emocnI lanosreP :seitivitca tiduA 95.3 $ 081,02$ 184,27 $ 20.4 $ 610,81$ 153,27 $ stidua kseD 79.51 564,31 301,512 86.32 196,01 381,352 stidua pu-wollof SRI 17.3 760,01 953,73 90.4 445,7 628,03 stidua dleiF tidua xat emocni lanosreP 34.7 217,34 349,423 38.9 152,63 063,653 slatotbus xaT noitaroproC :seitivitca tiduA 65.5 391,5 668,82 90.13 660,4 024,621 stidua kseD 21.651 595 098,29 04.211 896 654,87 stidua pu-wollof SRI 88.91 631,82 334,955 26.9 062,72 643,262 stidua dleiF tidua xat noitaroproC 80.02 429,33 981,186 95.41 420,23 222,764 slatotbus 99.3 373,52 743,101 00.9 280,22 177,891 stidua ksed latoT 19.12 060,41 399,703 21.92 983,11 936,133 stidua pu-wollof SRI latoT 26.51 302,83 297,695 24.8 408,43 271,392 stidua dlefi latoT 69.21 $ 636,77$ 231,600,1$ 60.21$ 572,86$ 285,328$ stidua lla latoT egap txen eht no deunitnoc 5588 California State Auditor Report 2002-124 California State Auditor Report 2002-124 5599 69–5991 raeY lacsiF 59–4991 raeY lacsiF reP stnemssessA teN stsoC stnemssessA teN reP stnemssessA teN stsoC stnemssessA teN tsoC fo ralloD )sdnasuohT ni( )sdnasuohT ni( tsoC fo ralloD )sdnasuohT ni( )sdnasuohT ni( noitpircseD xaT emocnI lanosreP :seitivitca tiduA 52.3 $ 941,32$ 791,57 $ 84.2 $ 791,92$ 125,27 $ stidua kseD 19.04 284,01 968,824 06.81 195,21 342,432 stidua pu-wollof SRI 61.3 203,61 475,15 28.1 980,41 636,52 stidua dleiF tidua xat emocni lanosreP 31.11 339,94 046,555 59.5 778,55 004,233 slatotbus xaT noitaroproC :seitivitca tiduA 95.0 784,7 024,4 39.2 393,5 308,51 stidua kseD 55.43 539,1 958,66 24.312 813 968,76 stidua pu-wollof SRI 97.9 534,92 511,882 71.21 432,92 246,553 stidua dleiF tidua xat noitaroproC 52.9 758,83 493,953 75.21 549,43 413,934 slatotbus 06.2 636,03 716,97 55.2 095,43 423,88 stidua ksed latoT 29.93 714,21 827,594 04.32 909,21 211,203 stidua pu-wollof SRI latoT 34.7 737,54 986,933 08.8 323,34 872,183 stidua dlefi latoT 13.01$ 097,88$ 430,519$ 05.8$ 228,09$ 417,177$ stidua lla latoT 5588 California State Auditor Report 2002-124 California State Auditor Report 2002-124 5599 89–7991 raeY lacsiF 79–6991 raeY lacsiF reP stnemssessA teN stsoC stnemssessA teN reP stnemssessA teN stsoC stnemssessA teN tsoC fo ralloD )sdnasuohT ni( )sdnasuohT ni( tsoC fo ralloD )sdnasuohT ni( )sdnasuohT ni( noitpircseD xaT emocnI lanosreP :seitivitca tiduA 78.3 $ 079,61$ 947,56 $ 47.6 $ 485,81$ 272,521 $ stidua kseD 73.44 276,7 793,043 03.92 302,11 452,823 stidua pu-wollof SRI 21.2 445,31 017,82 91.5 446,31 487,07 stidua dleiF tidua xat emocni lanosreP 93.11 681,83 658,434 70.21 134,34 013,425 slatotbus xaT noitaroproC :seitivitca tiduA 42.1 248,9 481,21 92.2 425,8 784,91 stidua kseD 82.52 425,3 190,98 98.73 218,1 466,86 stidua pu-wollof SRI 18.41 981,74 079,896 08.9 800,44 981,134 stidua dleiF tidua xat noitaroproC 22.31 555,06 542,008 65.9 443,45 043,915 slatotbus 19.2 318,62 339,77 43.5 801,72 957,441 stidua ksed latoT 63.83 591,11 884,924 05.03 510,31 819,693 stidua pu-wollof SRI latoT 89.11 337,06 086,727 17.8 256,75 379,105 stidua dlefi latoT 15.21$ 147,89$ 101,532,1$ 76.01$ 577,79$ 056,340,1$ stidua lla latoT egap txen eht no deunitnoc 6600 California State Auditor Report 2002-124 California State Auditor Report 2002-124 6611 99–8991 raeY lacsiF 89–7991 hguorhT 39–2991 raeY lacsiF teN egarevA teN egarevA reP stnemssessA teN stsoC stnemssessA teN reP stnemssessA tsoC egarevA stnemssessA tsoC fo ralloD )sdnasuohT ni( )sdnasuohT ni( tsoC fo ralloD )sdnasuohT ni( )sdnasuohT ni( noitpircseD xaT emocnI lanosreP :seitivitca tiduA 01.5 $ 304,41 $ 374,37 $ 38.3 $ 610,12$ 595,08 $ stidua kseD 69.43 807,6 294,432 32.72 710,11 800,003 stidua pu-wollof SRI 63.3 032,41 868,74 62.3 235,21 518,04 stidua dleiF tidua xat emocni lanosreP 70.01 143,53 338,553 64.9 565,44 814,124 slatotbus xaT noitaroproC :seitivitca tiduA 83.2 751,7 150,71 11.5 157,6 035,43 stidua kseD 07.86 518,1 096,421 32.25 084,1 503,77 stidua pu-wollof SRI 22.01 633,95 632,606 56.21 112,43 616,234 stidua dleiF tidua xat noitaroproC 59.01 803,86 779,747 38.21 244,24 154,445 slatotbus 02.4 065,12 425,09 51.4 767,72 521,511 stidua ksed latoT 41.24 325,8 281,953 91.03 794,21 313,773 stidua pu-wollof SRI latoT 98.8 665,37 401,456 31.01 347,64 134,374 stidua dlefi latoT 56.01$ 946,301$ 018,301,1$ 01.11$ 700,78$ 968,569$ stidua lla latoT 6600 California State Auditor Report 2002-124 California State Auditor Report 2002-124 6611 10–0002 raeY lacsiF 00–9991 raeY lacsiF reP stnemssessA teN stsoC stnemssessA teN reP stnemssessA teN stsoC stnemssessA teN tsoC fo ralloD )sdnasuohT ni( )sdnasuohT ni( tsoC fo ralloD )sdnasuohT ni( )sdnasuohT ni( noitpircseD xaT emocnI lanosreP :seitivitca tiduA 38.2 $ 051,12$ 048,95 $ 65.3 $ 580,81$ 293,46 $ stidua kseD 50.03 243,7 416,022 66.91 678,8 705,471 stidua pu-wollof SRI 63.4 714,11 157,94 41.3 547,41 643,64 stidua dleiF tidua xat emocni lanosreP 72.8 909,93 502,033 48.6 607,14 542,582 slatotbus xaT noitaroproC :seitivitca tiduA 18.0 686,51 117,21 81.1 388,21 491,51 stidua kseD 39.36 335,1 000,89 85.26 464,1 216,19 stidua pu-wollof SRI 33.6 682,04 859,452 82.41 582,04 053,575 stidua dleiF tidua xat noitaroproC 63.6 505,75 966,563 94.21 236,45 651,286 slatotbus 79.1 638,63 155,27 75.2 869,03 685,97 stidua ksed latoT 09.53 578,8 416,813 47.52 043,01 911,662 stidua pu-wollof SRI latoT 98.5 307,15 907,403 03.11 030,55 696,126 stidua dlefi latoT 41.7 $ 414,79$ 478,596$ 40.01$ 833,69$ 104,769$ stidua lla latoT egap txen eht no deunitnoc 6622 California State Auditor Report 2002-124 California State Auditor Report 2002-124 6633 20–1002 hguorht 99–8991 raeY lacsiF 20–1002 raeY lacsiF teN egarevA teN egarevA reP stnemssessA tsoC egarevA stnemssessA reP stnemssessA teN stsoC stnemssessA teN tsoC fo ralloD )sdnasuohT nI( )sdnasuohT nI( tsoC fo ralloD )sdnasuohT nI( )sdnasuohT nI( noitpircseD xaT emocnI lanosreP :seitivitca tiduA 78.3 $ 396,81$ 563,27 $ 43.4 $ 431,12$ 557,19 $ stidua kseD 66.62 861,7 201,191 64.32 647,5 597,431 stidua pu-wollof SRI 32.4 561,31 237,55 44.6 662,21 269,87 stidua dleiF tidua xat emocni lanosreP 81.8 620,93 991,913 08.7 641,93 215,503 slatotbus xaT noitaroproC :seitivitca tiduA 82.1 385,21 821,61 43.1 806,41 955,91 stidua kseD 58.86 156,1 376,311 93.87 197,1 883,041 stidua pu-wollof SRI 67.01 664,54 792,984 14.21 859,14 346,025 stidua dleiF tidua xat noitaroproC 73.01 007,95 890,916 66.11 753,85 095,086 slatotbus 38.2 672,13 394,88 11.3 247,53 413,111 stidua ksed latoT 65.43 918,8 577,403 15.63 735,7 381,572 stidua pu-wollof SRI latoT 03.9 136,85 920,545 60.11 422,45 506,995 stidua dlefi latoT 05.9 $ 627,89$ 792,839$ 11.01$ 305,79$ 201,689$ stidua lla latoT 6622 California State Auditor Report 2002-124 California State Auditor Report 2002-124 6633 20–1002 hguorhT 39–2991 raeY lacsiF teN egarevA teN egarevA reP stnemssessA tsoC egarevA stnemssessA tsoC fo ralloD )sdnasuohT ni( )sdnasuohT ni( noitpircseD xaT emocnI lanosreP :seitivitca tiduA 58.3 $ 780,02$ 303,77 $ stidua kseD 60.72 774,9 644,652 stidua pu-wollof SRI 66.3 587,21 187,64 stidua dleiF tidua xat emocni lanosreP 99.8 943,24 035,083 slatotbus xaT noitaroproC :seitivitca tiduA 99.2 480,9 071,72 stidua kseD 03.95 945,1 258,19 stidua pu-wollof SRI 67.11 217,83 882,554 stidua dleiF tidua xat noitaroproC 46.11 543,94 013,475 slatotbus 85.3 171,92 374,401 stidua ksed latoT 95.13 620,11 892,843 stidua pu-wollof SRI latoT 57.9 794,15 960,205 stidua dlefi latoT 14.01$ 496,19$ 048,459$ stidua lla latoT stnemssessA .20–1002 hguorht 39–2991 sraey lacsfi rof stroper snoitarepo launna s’draob ehT :ecruoS .serugfi tfard era 20–1002 dna 10–0002 sraey lacsfi rof ataD .tidua fo epyt hcae ot detaler era stsoc dna ten eht ,oslA .eunever etats lautca tneserper ton od stnemssessa eseht ,1 retpahC ni denialpxe sA :etoN ro ytilibitcelloc rof detsujda ton era tub ,daehrevo latnemtraped tcefler tsoc fo rallod rep stnemssessa .stsoc noitcelloc 6644 California State Auditor Report 2002-124 California State Auditor Report 2002-124 6655 APPENDIX B Collection Program Actual Revenue and Cost Data As noted in Chapter 3, the Franchise Tax Board (board) generally does not track actual collection program revenues or related costs beyond high-level categories. Appendix B.1 on the following page displays the level of revenue and cost detail that the board is able to provide for fiscal year 2001–02. Although the board is able to provide revenue detail by assessment origin for total personal income taxes, it is unable to further divide these revenues into automated and manual activities, nor does it track corporation tax revenue in this detail. In addition, the board does not track any costs by origin, which prevents the calculation of a cost-benefit ratio by assessment type. 6644 California State Auditor Report 2002-124 California State Auditor Report 2002-124 6655 6666 California State Auditor Report 2002-124 California State Auditor Report 2002-124 6677 1.B XIDNEPPA ataD tsoC dna euneveR margorP noitcelloC oitaR tfieneB-tsoC tsoC euneveR latoT launaM detamotuA latoT launaM detamotuA latoT launaM detamotuA xaT 02.51$ 99.8$ 10.42$ 000,206,66$ 000,840,93$ 000,455,72$ 000,676,210,1$ 000,570,153$ 000,106,166 $ emocnI lanosreP — — — A/N A/N A/N 000,371,627 A/N A/N nruteR — — — A/N A/N A/N 000,758,411 A/N A/N tiduA — — — A/N A/N A/N 000,472,761 A/N A/N tnemecrofne gniliF — — — A/N A/N A/N 000,273,4 A/N A/N rehtO 25.82 19.21 66.05 000,694,42 000,263,41 000,431,01 000,327,896 000,583,581 000,833,315 noitaroproC — — — A/N A/N A/N A/N A/N A/N nruteR — — — A/N A/N A/N A/N A/N A/N tiduA — — — A/N A/N A/N A/N A/N A/N tnemecrofne gniliF — — — A/N A/N A/N A/N A/N A/N rehtO 97.81$ 40.01$ 81.13$ 000,890,19$ 000,014,35$ 000,886,73$ 000,993,117,1$ 000,064,635$ 000,939,471,1$ slatoT .20–1002 raey lacsfi rof troper snoitarepo tfard dna setamitse s’draoB xaT esihcnarF :secruoS .elbaliava ton ataD = A/N APPENDIX C Proposed Template for a Comparison of Actual to Projected Returns for Specific Audit Workload Types The following pages provide the template that we recom- mend in Chapter 1 for the Franchise Tax Board’s (board) use in comparing its audit workplan’s projected cost-ben- efit ratios (CBR) to the corresponding actual CBR, or return on investment, by individual audit workloads, an evaluation that the board does not currently make. As discussed in Chapter 1, we believe that this template would aid the board in assessing the reasonableness of its prospective audit CBRs, identifying those projections that do not reasonably match actual results, and adjusting future prospective CBRs accordingly. Using this information, the board would have a more accurate and effec- tive tool for assigning workload to staff and estimating the additional benefits that proposed new auditors would provide. Each of the board’s audit types contains numerous specific workloads such as audits of partnerships, estates and trusts, and corporations. For this public report, we have removed the names of the specific workloads to protect the confidentiality of the board’s workplans. The template includes the planned assess- ments and cost for the individual audit workloads from the board’s 2001–02 workplan. The board already collects much of the information about actual results necessary to complete this comparison, compiling actual hours charged and assessments by individual audit workload, while accounting for actual costs at a higher level: audit type. Therefore, the board needs only a method of allocating costs to each workload. A relatively simple allocation would use information the board already has—salary costs and hours charged for each specific workload—to distribute total costs proportionately. Finally, the board could compute an actual return on investment per individual workload and com- pare it to the corresponding prospective CBR in order to measure the reasonableness of the projection. 6666 California State Auditor Report 2002-124 California State Auditor Report 2002-124 6677 6688 California State Auditor Report 2002-124 California State Auditor Report 2002-124 6699 1.C XIDNEPPA sepyT daolkroW tiduA cfiicepS rof snruteR detcejorP ot lautcA fo nosirapmoC rof etalpmeT desoporP -tsoC nalpkroW no nruteR dennalP ecnairaV oitaR tfieneB tnemtsevnI stsoC lautcA latoT stnemssessA lautcA tsoC dennalP stnemssessA xaT emocnI lanosreP dleiF — 63.4 — — 711,841,03 $ 774,023,2 $ 008,801,01 $ A daolkroW — 73.21 — — 143,480,9 902,055,1 000,081,91 B daolkroW — 17.6 — — 519,844,6 108,636,2 005,296,71 C daolkroW — 40.0 — — 426,1 499,971 000,7 D daolkroW — 80.0 — — 214,564 465,918 000,96 E daolkroW — 19.4 — — 808,018,9 149,853,2 066,375,11 F daolkroW — 29.4 — — 553,218,9 288,390,3 005,702,51 G daolkroW — 19.4 — — — 110,301 004,505 H daolkroW — 87.2 — — — 867,921 633,063 I daolkroW — 90.1 — — — 961,88 497,59 J daolkroW — 36.5 — 801,937,9 $ 275,177,56 618,082,31 099,997,47 slatoT kseD — 35.4 — — 832,625,02 742,289,1 059,389,8 A daolkroW — 31.6 — — 436,436,2 463,754 000,508,2 B daolkroW — 17.6 — — 128,367,2 750,031,1 005,285,7 C daolkroW — 10.5 — — 732,974,7 667,302 000,020,1 D daolkroW — 99.0 — — 014,967 253,206 000,495 E daolkroW — 71.1 — — 605,05 888,243 000,004 F daolkroW — 40.0 — — 696 041,77 000,3 G daolkroW — 53.11 — — 989,286,92 600,389,1 000,005,22 H daolkroW — 79.9 — — 153,056,43 017,318,3 038,620,83 I daolkroW — 91.6 — — 544,405,41 099,713,2 158,053,41 J daolkroW — 10.12 — — 419,3 887,83 000,518 K daolkroW — 19.01 — — 046,852 994,14 598,254 L daolkroW — 41.0 — — 967,157,1 997,668 000,021 M daolkroW — 19.4 — — 236,402,4 579,010,1 041,069,4 N daolkroW — 29.4 — — 592,502,4 059,523,1 005,715,6 O daolkroW — 19.4 — — — 741,44 006,612 P daolkroW — 87.2 — — — 516,55 034,451 Q daolkroW — 90.1 — — — 787,73 450,14 R daolkroW — PN — — 797,215,6 S daolkroW — 17.6 — 872,447,81$ 473,999,921$ 080,233,61$ 057,345,901$ slatoT .nalpkrow 20–1002 raey lacsfi s’draob eht ni dennalp toN = PN 6688 California State Auditor Report 2002-124 California State Auditor Report 2002-124 6699 tsoC nalpkroW no nruteR dennalP ecnairaV oitaR tfieneB tnemtsevnI stsoC lautcA latoT stnemssessA lautcA tsoC dennalP stnemssessA SRI — 31.01 — — 657,633,1 $ 906,381 $ 000,068,1 $ A daolkroW — 05.15 — — 924,598,05 821,728 000,595,24 B daolkroW — 79.8 — — 383,946,2 875,58 000,867 C daolkroW — 70.63 — — 954,875,6 065,38 987,310,3 D daolkroW — 70.51 — — 233,265,7 152,883 000,058,5 E daolkroW — 48.73 — — 879,054,71 887,736 571,431,42 F daolkroW — 45.81 — — 781,752,21 537,751 000,529,2 G daolkroW — 28.87 — — 684,389,2 004,431 270,395,01 H daolkroW — 01.37 — — 220,407,71 213,692 025,956,12 I daolkroW — 98.52 — — 410,364,7 795,453 000,081,9 J daolkroW — 39.83 — 311,408,5 $ 640,188,621 859,841,3 655,875,221 slatoT smialC — 58.7 — — 825,639,2 212,103 314,463,2 A daolkroW — 58.7 — — 830,603,1 669,331 785,150,1 B daolkroW — PN — — 313,692 C daolkroW — 92.7 — — 730,231 633,442 009,187,1 D daolkroW — PN — — 455,9 E daolkroW — 65.2 — — 214,677 545,741 223,773 F daolkroW — PN — — — G daolkroW — 47.6 — 645,783,2 288,654,5 950,728 222,575,5 slatoT xaT noitaroproC dleiF — 18.31 — — 959,560,921 614,083,8 005,267,511 A daolkroW — 86.5 — — 837,006,3 887,109,1 000,008,01 B daolkroW — 95.5 — — 586,760,93 593,414,4 005,296,42 C daolkroW — 22.12 — — 886,656,24 290,369,2 000,488,26 D daolkroW — 45.11 — — 800,802$ 000,004,2 E daolkroW — 01.51 — — 717,580,631 202,818,4 000,567,27 F daolkroW — 60.6 — — 262,410,21 896,712,1 000,083,7 G daolkroW — 79.5 — — 243,672,73 787,180,5 005,633,03 H daolkroW — 02.32 — — 429,836,32 518,345,1 000,028,53 I daolkroW — 13.21 — — — 739,921 000,006,1 J daolkroW — 15.7 — — 283,150,81 814,899,1 000,000,51 K daolkroW 27.1 306,1 814,537 000,562,1 L daolkroW — — — PN )682,523( M daolkroW — 04.11 — 288,985,53$ 410,431,144$ 479,293,33$ 005,507,083$ slatoT egap txen eht no deunitnoc .nalpkrow 20–1002 raey lacsfi s’draob eht ni dennalp toN = PN 7700 California State Auditor Report 2002-124 California State Auditor Report 2002-124 7711 tsoC nalpkroW no nruteR dennalP ecnairaV oitaR tfieneB tnemtsevnI stsoC lautcA latoT stnemssessA lautcA tsoC dennalP stnemssessA kseD — 48.7 — — 569,929,7 $ 770,911,1 $ 000,877,8 $ A daolkroW — 22.51 — — 144,973,1 063,002 005,940,3 B daolkroW — 19.1 — — — 441,162 000,005 C daolkroW — 26.7 — — 713,921,2 092,964 000,575,3 D daolkroW — PN — — )076,18( E daolkroW — PN — — — F daolkroW — PN — — 444,52 G daolkroW — PN — — — H daolkroW — PN — — — I daolkroW — 10.0 — — 021,012 454,601 000,1 J daolkroW — 83.7 — 679,915,21$ 716,295,11 523,651,2 005,309,51 slatoT SRI — 68.627 — — 209,033,51 382,5$ 000,048,3 A daolkroW — 67.577 — — 788,264,15 069,3 000,270,3 B daolkroW — 31.13 — — 943,365,46 774,866,1 000,539,15 C daolkroW — 49.71 — — 708,805,81 025,046 000,094,11 D daolkroW — 22.23 — — — 090,721 000,590,4 E daolkroW PN 249,968 F daolkroW — — — — 44.03 — 486,585,1 788,537,051 033,544,2 000,234,47 slatoT smialC — 12.9 — — 701,356,62 436,809,2 000,297,62 A daolkroW — 52.01 — — 020,161,41 373,331,2 000,768,12 B daolkroW — 18.5 — — 744,230,91 143,893,1 000,721,8 C daolkroW 16.3 108,423,1 940,973 000,863,1 D daolkroW — — — — 35.8 — 492,194,6 573,171,16 793,918,6 000,451,85 slatoT dennalP-noN — PN — — )902,813( A daolkroW — PN — — 713,7 B daolkroW — PN — — 496,731,3 C daolkroW — PN — — — D daolkroW — PN — — 805,274 E daolkroW — PN — — 510,061 F daolkroW — PN — — — G daolkroW — PN — — 227,854,2 H daolkroW — 47.01 — 188,168,29$ 418,066,899$ 939,204,87$ 815,296,148$ slatoT dnarG .serugfi tfard era atad troper snoitarepo ehT .20–1002 raey lacsfi rof sdrocer gnitnuocca SRATSLAC dna ,stroper noitcudorp ,troper snoitarepo ,nalpkrow s’draob ehT :secruoS .nalpkrow 20–1002 raey lacsfi s’draob eht ni dennalp toN = PN M E M O R A N D U M To: Elaine M. Howle, State Auditor* Bureau of State Audits 555 Capitol Mall, Suite 300 Sacramento, CA 95814 Date: May 5, 2003 From: Gerald H. Goldberg Subject: Draft Bureau of State Audits Report Thank you for the opportunity to review the draft audit report prepared by your staff for the Joint Legislative Audit Committee. We appreciate your recommendations for improving the usability of information we provide to the Department of Finance and the Legislature. We concur that improvements can be made. On our own initiative we have begun a project that will use activity based accounting to better capture revenue and cost stream data. Franchise Tax Board’s total assessments from audits of personal income and corporation tax returns have provided the State an important source of potential revenue at relatively little cost, returning around $10 in assessments for every $1 of cost. In addition, the collection program returns around $19 in revenue for every $1 of cost. Following are specific comments to the report and the recommendations: Bureau of State Audits (BSA) Recommendation: To more completely and clearly reveal its programs’ costs and benefits, the board should consider using the complete measure- ment of the audit program’s performance that we have described in Table 3 on page 19. This measurement compares all the benefits – the total revenues that result over time from the auditors’ assessments of additional taxes – with the total costs to produce them, including the costs of collection. Thus the board would treat the collection program as another service center for audits. If it determines that its current information system cannot produce the data necessary for such a measurement, the board should consider the needs of a complete mea- surement when it upgrades or changes its current information system. Franchise Tax Board (FTB) Response: While BSA’s report notes that there are significant benefits of the audit program, which cannot be measured, it chooses to focus its attention primarily on recognizing additional costs. Centering attention only on costs within the CBRs 1 tends to obscure the central purpose of the audit program, namely to ensure that taxpayers assess the correct amount of tax and to provide a level of assurance to self compliant taxpay- ers that non compliance is being recognized and dealt with. * California State Auditor’s comments begin on page 77. 7700 California State Auditor Report 2002-124 California State Auditor Report 2002-124 7711 Draft Bureau of State Audits Report May 5, 2003 Page 2 The CBR schedule’s primary purpose is to provide a format for prioritizing audit workloads and to provide management with performance measures for program decisions. A secondary role it plays is to give a sense of the relative value of audits that have been identified but that cannot be undertaken because of resource constraints. It is not meant to be a financial accounting document. 2 As noted previously to the BSA, the CBRs have historically excluded certain costs such as departmental overhead and costs of collection. The costs of collection have been excluded at the direction of the LAO in its 1981 study, The Allocation of Audit Resources: An Analysis of California’s Tax Audit Programs, since to include them could well influence the choice of which audits to undertake. Audit models that have a high cost to collect, and represent the most egre- gious non-compliant taxpayers would have lower CBRs and could possibly fall below the CBR 1 threshold. This would then have the impact of discouraging compliance efforts on the taxpayers who are in the most need of it, and serve to degrade the primary purpose of the audit program. The selection of the traditional level of funding CBRs at the 5:1 threshold was proposed and accepted with these conventions in place. The threshold seeks to maintain a balance between audit intrusiveness and effective audit presence. Inclusion of additional cost factors such as departmental overhead and collection costs should cause a reevaluation of the reasonable- ness of that 5:1 CBR threshold, since it would impact the number of audits undertaken, and result in a change in the current program balance. BSA Recommendation: If the board decides not to currently use the complete measurement and continues to use separate performance measurements for the audit and collection pro- grams, it should do the following: • In budget change documents and other reports given to external decision makers, the board should explicitly disclose the elements not included in the cost components of various performance measures used to assess the audit and collection programs. The board should also disclose the effect of those excluded elements. Further, the board should disclose the overlap in benefits claimed by its audit and collection programs. FTB Response: We agree. • To provide useful information to decision makers when requesting additional audit posi- tions, the board should use a format, recommended in our prior report, that details the types of activities new auditors will perform as well as the projected assessments and historical returns resulting from these activities. Additionally, the board should revise its supporting audit workplans to include the actual returns of each of the specific workload types for the most recently completed fiscal year. FTB Response: FTB considered the format recommended by BSA in the prior report and determined that the actual user of the report should be consulted on their needs. Therefore, at a meeting with Department of Finance on April 7, 1999, and under the scrutiny of a representa- tive of the BSA we discussed options for a workplan matrix. While considering the BSA matrix, 7722 California State Auditor Report 2002-124 California State Auditor Report 2002-124 7733 Draft Bureau of State Audits Report May 5, 2003 Page 3 the Department of Finance desired a number of modifications, which included additional detail 3 for indirect workloads, and a higher summary of direct audit workloads. The resulting matrix included both projected assessments and historical returns at the agreed upon levels. The representative from BSA who attended that meeting agreed with our efforts to work together on this issue. Indeed, our 60 day response letter to the 1999 BSA audit addressed to Mr. Kurt R. Sjoberg reported that the comments from the BSA representative indicated that he was satis- fied with our representation of how we use our audit hours, and that our matrix provided the format to aid in the communication of resource needs between FTB, DOF, and the Legislature. As result of this iterative process, we assumed in good faith that our workplan was appropriate on two levels. It met our customer’s (DOF) stated needs, and it had the BSA’s approval. It now appears that the BSA wishes for us to utilize its original format. If this is what the BSA requires, we will certainly comply. However, we would like to clarify the workload categories that BSA suggests. Under Corpo- ration Tax, both Field and Desk audits are broken down into large assessments and small assessment audits. Normally we do not classify our audits in this manner, and we generally do not know at the outset whether an individual audit will yield a large or small assessment. We 4 would suggest using the alternate sub categories of non-apportioning and apportioning audits. • To track the accuracy over time of its calculations of the prospective CBRs for individual audit workload types, the board should compare these prospective CBRs against actual returns annually. The board should make the results available to Finance and the Legis- lative Analyst’s Office, which advises the Legislature on budgetary matters, and should also include them in the board’s annual report to the Legislature on the results of its audit and collection activities. Moreover, the board should use the results of the comparison in future calculations of prospective CBRs. FTB Response: We concur. We will compare projected CBRs with actual returns annually and will make them available to Finance and the Legislative Analyst’s Office. Due to the confidential nature of the workplan, specific workloads will not be identified by name on the report. BSA Recommendation: If the board believes that information it publishes in its operations reports is not accurate, even though it is based on the board’s financial accounting system, the board should do the following: • Ensure that its financial accounting system reports accurate information, and • Correct data it believes to be inaccurate before it publishes the information in its opera- tions reports. FTB Response: The financial accounting system used by FTB is the California Statewide Accounting and Reporting System (CALSTARS), which was developed and is administered by the Department of Finance. FTB captures costs within CALSTARS in the same structure as the department’s appropriation budget – by program and element. The system captures costs 7722 California State Auditor Report 2002-124 California State Auditor Report 2002-124 7733 Draft Bureau of State Audits Report May 5, 2003 Page 4 as assigned by the particular program areas. These costs are captured through direct charges to the program; program timesheet reporting; and overhead allocations. Together these dif- ferent cost elements are combined to reflect the financial status and reports for the various programs administered by FTB (Tax, Child Support, etc.) 5 As such, the CALSTARS financial accounting system is sound in design, functionality and integrity. It is in full compliance with statewide directives and regulations. The reports gener- ated from the system reflect the financial condition of FTB based on programmatic information provided through the aforementioned methods. The financial information from CALSTARS is reported monthly and can be used by the vari- ous organizational units and program areas in various ways, such as for monitoring costs, developing workplans, and making projections. As the CALSTARS information is reviewed, it is possible that any discrepancies or adjustments could be identified. The adjustments are reviewed by the financial staff and, if appropriate, subsequent corrections are made to update the system. The FTB Operations Report is based, only in part, on the financial reports from the CAL- STARS system. In addition, other data or adjustments are included for purposes of develop- ing the Operations Report. In response to the BSA recommendations, we are reviewing our procedures to ensure the Operations Report is accurate, timely, and verifiable with information sources and reports, such as workplans and financial statements. The department will focus on improving its processes to better capture and validate data that goes into the automated financial reporting system. For example, the department is evaluating how indirect costs are distributed. In addition, an important tool the department will use is the recently completed Activity Based Costing (ABC) model. This model was designed to improve the reporting of cost information and its use will enable the department to more accurately identify costs to activities, processes, and programs. BSA Recommendation: To demonstrate the effectiveness of new collection program posi- tions, the board should develop a methodology for measuring the benefit of these positions by isolating the return from the additional positions and comparing it against a base year. FTB Response: The audit is correct in stating that we do not typically report year-end results by isolating revenue from the new positions we have acquired. However, we now have the capability to present results according to this methodology, and will follow it in future annual summaries of results. As noted in the Audit Report, we do use a cost-to-benefit ratio (CBR) process. Since the 1980s we have used CBRs to prioritize workloads and assess our future staffing needs. This process projects the total number of staff we will need annually to collect the revenue gener- ated from accounts that meet Department of Finance’s (DOF) CBR criteria. For example, for fiscal year 2001/02, we projected that we could produce $1.63 billion in revenue from cases 7744 California State Auditor Report 2002-124 California State Auditor Report 2002-124 7755 Draft Bureau of State Audits Report May 5, 2003 Page 5 meeting DOF CBR criteria, if we had 1,024 PYs. As we had budget authority for only 905 PYs, we submitted a justification for additional staff to produce the projected revenue. The proposal was approved. We leveraged our new collection system (ARCS) to implement our workplan and to generate the promised revenue. ARCS is designed to implement and support our work plan process through continuously monitoring and managing collection account activities. The system’s automated account selection process maximizes collections by using yield as the criterion for selecting the “next best” account to work. Drawing from this pool of “best” accounts, the system assigns individual cases to collectors, and determines the next best manual or auto- mated action to resolve the account. ARCS’ automated assignment and decision processes keep costs of collections as low as possible. Our actual results for fiscal year 2001/02 show that we produced $1.71 billion in revenue, exceeding our revenue projections for the year. 6 We believe these results demonstrate the effectiveness of our new collection system as well as the methodology we used to estimate the benefit of new positions. BSA Recommendation: To more accurately represent the process that assigns work to col- lection program staff, the board should continue developing a methodology based on data from its Accounts Receivable Collection System to justify new collection program staffing requests. FTB Response: We strongly agree with this recommendation, and have been waiting for sufficient data to become available from ARCS to allow us to implement this methodology. We recently completed an extensive analysis, based on two full years experience in using the new system. We are pleased to report that this analysis now allows us to leverage ARCS data more fully, and we are using that data for our fiscal year 2003/04 work plan. BSA Recommendation: For the board to be consistent with the intent of budget control language and Finance, it should not leave collection program positions unfilled beyond the normal time it takes to fill a position as a long-term strategy. FTB Response: We agree, and strive to fill all collection positions as quickly as possible. We do, however, periodically encounter challenges that may prevent us from filling positions, such as delayed budget enactment, hiring freezes, and limited training resources, especially during years of large augmentations of new positions. We make every reasonable effort to keep all positions filled and as of March 2003, for fiscal year 2002/03, our vacancy rate for tax collection positions is consistent with our salary savings goal. Additional FTB Comments: We would like to reiterate our concerns expressed at the time of the 1999 BSA Audit about Table 7, Return on Audits that Additional Staff Would Be Likely to Perform. This table is highly selective of the factors that it presents as impacting the 7 return on audits that new staff would likely perform. It fails to take into consideration tax law changes, tax regulations, case law and precedent, economic conditions and self-compliance. By failing to modify the base years in the table to include the impact of these factors, the analysis does not recognize the full value of the new audit positions. 7744 California State Auditor Report 2002-124 California State Auditor Report 2002-124 7755 Draft Bureau of State Audits Report May 5, 2003 Page 6 Again, we appreciate the opportunity to provide you with this response. If you need any fur- ther information or would like to discuss any of the issues above, please feel free to contact Philip Yu at 845-3388. (Signed by: John W. Davis for) Executive Officer 7766 California State Auditor Report 2002-124 California State Auditor Report 2002-124 7777 COMMENTS California State Auditor’s Comments on the Response From the Franchise Tax Board To provide clarity and perspective, we are commenting on the Franchise Tax Board’s (board) response to our audit report. The numbers correspond with the numbers we have placed in the board’s response. 1 Our suggestion to show full costs associated with audit assess- ments does not preclude the board from auditing returns with less than $5 in anticipated benefits for each $1 of cost. Our recommendation addresses how the board describes its work, not what work it chooses to do. However, if the board has pro- grammatic reasons for requesting new audit positions, it should disclose those reasons as the justification for new positions. For example, if the board wants new auditors to provide additional assurance to self-compliant taxpayers that noncompliance is being recognized and dealt with, the board should explicitly describe in its budget documents the goal of the new positions. Further, the board should clarify that the monetary value of this enhanced audit presence cannot be measured. 2 We recognize the various uses of the cost-benefit ratios (CBRs) and believe that, for purposes of allocating workload to staff, it is not essential to include departmental overhead in the board’s workplan CBRs. However, a complete measurement of the audit program’s performance reported to external decision makers would compare the benefits—the total revenues that result over time from the auditors’ assessments of additional taxes—with the total costs to produce them, including the costs of collecting the assessments. As stated on page 16, the Legislative Analyst’s Office made a similar point in a fiscal year 1997–98 report when it stated that the full cost of collecting audit assessments should be considered as part of the overall and true cost of generating returns through audits. 3 In its 60-day response to the recommendations made in our 1999 report, the board indicated that it modified the matrix format to provide additional detail of total hours used and additional workload breakdowns. The board also acknowledged the agreement reached in the April 7, 1999, meeting with the 7766 California State Auditor Report 2002-124 California State Auditor Report 2002-124 7777 Department of Finance and the Bureau of State Audits that the workload categories would be modified over time to reflect program changes. However, the board did not indicate in the 60-day response that it planned to modify the matrix to show only a higher summary of direct audit workloads. 4 We agree with the board that it would be difficult to anticipate at the outset the size of a future audit assessment. Accordingly, we have modified our proposed template in Table 5 to delete the subcategories for large and smaller assessment audits. 5 We agree with the board’s comments that the California State- wide Accounting and Reporting System (CALSTARS) which many state departments use, is sound in design, functionality, and integrity. We did not intend to suggest the board should abandon its use. CALSTARS is, however, dependent on the qual- ity of the information input into the system, and this is what we understand the board to have been commenting on when it cited miscoding and misallocation. Thus, we believe the board’s proposal to focus on improving its processes to better capture and validate data that goes into the system is appropriate. 6 We disagree with the board’s argument that, when its collections exceeded its projected revenue for fiscal year 2001–02, the board demonstrated the benefit of the new positions it received in that year. As Figure 8 indicates, the actual hours spent on collections decreased in fiscal year 2001–02, so that the board exceeded its projected revenue with less, rather than more, staff effort for the year. As a result, we question the board’s premise that the fund- ing for additional staff resulted in the increase in revenue. 7 We agree that Table 7 does not take into account all the possible factors that could affect audit assessments, and we discuss two additional factors in the report: the effect of the economy (page 34) and the need for maintaining an audit presence in the taxpaying community, resulting in the board’s conducting audits for which it does not necessarily anticipate a $5 benefit for each $1 of cost (pages 9, 16 and 34). As we noted on page 34, our analysis does not account for such unquantifiable benefits provided by the audit program. Similarly, the effect of changes in tax law and regulations is not readily quantifiable. 7788 California State Auditor Report 2002-124 California State Auditor Report 2002-124 7799 cc: Members of the Legislature Office of the Lieutenant Governor Milton Marks Commission on California State Government Organization and Economy Department of Finance Attorney General State Controller State Treasurer Legislative Analyst Senate Office of Research California Research Bureau Capitol Press 7788 California State Auditor Report 2002-124 California State Auditor Report 2002-124 7799