Donate
Independent, objective, nonpartisan research

R 397MSR

Authors

R 397MSR

Tagged with:

Publication PDFs

Database

This is the content currently stored in the post and postmeta tables.

View live version

object(Timber\Post)#3741 (45) { ["ImageClass"]=> string(12) "Timber\Image" ["PostClass"]=> string(11) "Timber\Post" ["TermClass"]=> string(11) "Timber\Term" ["object_type"]=> string(4) "post" ["custom"]=> array(6) { ["_wp_attached_file"]=> string(12) "R_397MSR.pdf" ["wpmf_size"]=> string(7) "2468822" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(247320) "A Review of Local Government Revenue Data in California Michael A. Shires Melissa Glenn Habers February 1997 Copyright © 1997 Public Policy Institute of California, San Francisco, CA. All rights reserved. PPIC permits short sections of text, not to exceed three paragraphs, to be quoted without written permission, provided that full attribution is given to the source and the above copyright notice is included. Foreword As we were shaping the Institute’s initial research agenda, we talked with many leaders throughout the state. We asked them which public policy issues were most in need of attention by PPIC. The area most frequently suggested for study was governance—how California’s political institutions and processes affect policy outcomes. Public finance was a matter of particular concern. Proposition 13 and a succession of other revenue-and-expenditure-limiting initiatives were viewed as creating distortions in decisionmaking and accountability. Phrases such as “loss of control of the resource allocation process” and “a maddening proliferation of special districts” occurred often in our conversations with public officials. Many people have strong views about the governing process in California; yet it is difficult to judge how well the process is working or to measure its performance. Gaining insight into these issues is critical in a state as diverse and rapidly changing as California. This is why PPIC has, from the start, made governance one of its central areas of research. iii To successfully evaluate the governing process, we must have a reliable set of statistics on local government finances. The California State Controller’s Office publishes local public finance data on an annual basis, and the U.S. Census of Governments publishes comparable data every five years. If the Institute and others are to tackle—with credibility—the question of governance in the coming years, we must be certain that these published data are accurate and comprehensive. There are those who question the quality of these data, suggesting that the special fees and charges emerging in the aftermath of Proposition 13 may not have been captured in the data, thus leading to a substantial undercount of revenues raised at the local level. These concerns led Michael A. Shires and Melissa Glenn Haber to undertake the study described in the following report, A Review of Local Government Revenue Data in California. The authors reviewed more than 7,000 local government entities to verify that they were included in the reports published by the state Controller’s Office and reached the satisfying conclusion that the state data captured virtually all of the public entities generating revenues at the local level. In addition, they found a high degree of correspondence between the data compiled by the local government entities and the summary statistics published by the state. This degree of correspondence should assure both analysts and others interested in local public finance in California that, regardless of the findings reached, the databases used in the research accurately reflect the pattern of revenue-raising strategies throughout the state. With the passage of Proposition 218 in the fall of 1996, which requires a referendum on any attempt by local governments iv to raise revenue, the debate surrounding revenue-raising strategies is likely to become more heated. It is our intention to help inform the discussion, and our first contribution is this report. David W. Lyon President and CEO Public Policy Institute of California v Summary Background For more than two decades, Californians have engaged in a great debate over the size, shape and role of their state and local governments. This debate has involved a wide range of questions: How much of the state’s income should be spent by the public sector? Which goods and services should be provided by the state government, which by local governments? What is the appropriate level of state and local taxation? And perhaps most important, can citizens trust their elected officials to reflect their views on these policy issues or should such matters be governed by a combination of constitutionally imposed rules, limits and referenda? When this debate heated up in the 1970s, there was general agreement that Californians provided more goods and services through their state and local governments than all but one or two other states. This was the era when California led the nation in per-pupil spending for vii elementary and secondary education; when the California Master Plan for Higher Education promised state-funded postsecondary education to all of its citizens; and when the state engaged in massive infrastructure investment in its roads and waterways. These and other activities required considerable resources, and thus California had one of the highest state and local tax burdens in the nation. Beginning with the passage of Proposition 13 in 1978, California’s voters have used the initiative system to limit the size of their governments. Proposition 13 placed stringent constraints on state and local governments’ ability to raise revenues locally. Other initiatives have significantly affected the ability of elected officials to control the amount and mix of publicly provided goods and services. In effect, California has engaged in a great experiment over the past two decades: It has led the nation in the use of direct democracy to limit representative government. This is the first of a series of reports in which the Public Policy Institute of California (PPIC) will evaluate the effects of this experiment. How effective have the rules imposed by the initiative process been? Has the size of California’s state and local governments been truly limited, or have elected officials found ways around the limits? What effects have the initiatives had on the amount and quality of public goods and services? Given the transfer of many federal government activities to state and local governments, how will these constraints affect the implementation of changes such as welfare reform? These are all vital questions in the California policy debate. However, several key studies addressing these questions have raised concerns about the quality of the data available for evaluating issues relating to local public revenues. In addition, exploratory discussions with policymakers and analysts throughout the state have revealed that viii many do not use the data available on local public finance or, if they do, they do so hesitantly. These potential and actual users of the data cite three concerns: (1) the data do not capture all public entities in the state; (2) they do not accurately reflect the fiscal activity of those they do capture; and (3) they are not produced in a timely manner. This research thoroughly examines the first two concerns, analyzing the data produced by the California State Controller’s Office. The Controller’s Office, which provides the most timely and comprehensive data available on local finances, fields an annual survey of financial activity that collects revenue, expenditure, and debt information on every public entity in its database. All public entities are required by law to complete the survey and return it to the Controller’s Office within 90 days of the end of the fiscal year. This report evaluates the comprehensiveness, accuracy, and timeliness of the revenue data. Comprehensiveness of the Data: Are All Appropriate Entities Included? Because the initiative system has increasingly constrained local governments, some argue that the governments have found creative ways to go about their business. One of the criticisms of the state’s data systems is that they are unable to track this creativity. It is also argued that the data systems cannot always identify the institutions that spring up as a result of this creativity, and hence the studies that use the revenue data underestimate the actual size of public sector activity. Findings on the Comprehensiveness of the Data Overall, the data were found to be quite comprehensive, with the primary exception being Mello-Roos or community facility districts ix (CFDs), which are typically not included in the database. Analysis of a sample of CFDs showed that as few as 20 percent of these entities are included in reports to the State Controller’s Office. A follow-up analysis of all 233 CFDs in existence in 1991–92 found that the total revenue of these entities was only $283.5 million. In a local-government sector with revenues totaling nearly $95 billion, this represents a trivial source of error—0.3 percent—even if all CFDs were excluded. Since some of these districts are included in the reported information, the actual level of error introduced by the omission of CFDs is even lower. If one wishes to focus on these particular entities for a specific policy inquiry, the data are obviously inadequate. An exhaustive comparison of more than 7,000 entities appearing on the lists of local agency formation commissions and the California Debt Advisory Commission with the entities appearing in the Controller’s report identified only three non-CFD entities as missing from the Controller’s survey. The revenues for these entities totaled approximately $0.5 million—an inconsequential sum. Recommendations for Strengthening the Comprehensiveness of the Data While it can be concluded from these analyses that the Controller’s data are comprehensive—that they include virtually the full range of public entities generating revenues at the local level—there are some ways in which the data can be improved. Include Mello-Roos Districts. Specific instructions and questions should be included in the annual survey to ensure that community facility districts are included in the Controller’s reports. x Create a Mechanism for Identifying New Entities. The Controller’s Office should implement a watchdog-type mechanism for identifying new entities that might not be captured under the current reporting scheme. Accuracy of the Data: Do They Reflect What Actually Happens? Another issue raised about the quality of the data is whether the reported information accurately represents the activity of the reporting entity. Since the questionnaires are completed by each entity independently, the potential for variation in the interpretation of the instructions across more than 7,000 entities is significant. Because the surveys are usually submitted before annual audits are completed, the use of unaudited information in preparing the survey represents another area of concern. Findings on the Accuracy of the Data The Controller’s data are quite accurate. They were verified against third-party sources of information in two ways: (1) by comparing the data reported in the surveys to specific revenue series, such as sales and property taxes, and (2) by comparing the data reported by the entities with their audited information. When the total amounts reported by the entities for specific revenue series were compared with the amounts reported by the county auditor-controllers and the State Board of Equalization for property and sales taxes, respectively, the variance was very low—ranging from 0.0 percent to 5.1 percent. On a county-bycounty and city-by-city basis, the variation was also quite low. Furthermore, these averages were heavily influenced by a few outliers. xi Without these outliers, the difference between the two sets of information essentially disappeared. When the information reported to the Controller was compared with audited financial information, the overall variance was even lower. For counties it totaled 1.2 percent, for cities 1.4 percent, and for special districts 1.5 percent. On an entity-by-entity basis, the average percentage differences for counties was 1.4 percent, for cities 3.9 percent, and for special districts 0.9 percent, showing that the low variance identified in aggregate reflected low levels of variation at the individual entity level as well. Recommendations for Strengthening the Accuracy of the Data Even though there is a high level of accuracy in the data, it can be improved in some areas, as suggested below. Provide More Specific Instructions and Follow-Up Regarding Capital Project Funds, Debt Service Funds and Housing Authorities. Several types of activity were not consistently reported by all entities to the Controller’s Office. These three areas represented the greatest proportion of the overall variance in the data, and improvements in these areas would significantly improve the general quality of the data. Expand and Clarify the Reporting of Special Assessment Districts. The reporting on these increasingly popular revenue generating arrangements is spotty at best and does not appear to capture the full range of activity that occurs. Researchers and analysts trying to assess the impact of Proposition 218, which appeared on the November 1996 ballot, found the data quite inadequate.1 ____________ 1Proposition 218 requires a referendum on any attempt by local governments to raise revenues, service charges, or special assessments. xii Expand the Reporting of School District Information. Because school districts represent one of the largest categories of local revenues and expenditures, and because detailed district-level information is not published elsewhere, entity-level detail should be provided to the Controller for each district, county office, and joint powers agency in the state. It would also be helpful to include the specific numbers used by the state in calculating the Proposition 98–required expenditures each year.2 Such a service would provide a common source of information for analysts and decisionmakers. Provide Detailed Fiscal Information for Community College Districts. Community college districts are unique entities in the state. Even though they receive their funding largely under the Proposition 98 formula, they are considered part of the state’s postsecondary education sector. However, unlike the other public members of this sector, community colleges are organized and governed locally. Data on the community colleges would be useful for both informational and accountability purposes. Establish a Consistent Reporting Format for All Categories of Entities. There is significant variation in the way various types of entities report their information. As discussed regarding school districts above, but true on a much broader basis, the requirements and reporting structures associated with particular types of entities are almost always determined by the institutional and historical context of the reporting entity, rather than by concern for comparability and public accountability or the utility of the data for decisionmakers. ____________ 2Proposition 98 creates a floor on state spending and (indirectly) on local spending for K–14 education. xiii As a result, the statistics are difficult to compare and use—both across government entities and sometimes even within the same report. In addition to consistent reporting formats, it would be useful if the major tax revenue streams—such as sales and property taxes—were explicitly identified for each entity. In conjunction with a universal reporting format, it would be useful if the governance information for each dependent entity in the database were expanded to include explicit detail about what the parent entity is. Policy analysts and decisionmakers could then easily perform the type of aggregation that had to be done manually for this research. Timeliness of the Data This study did not examine in great detail the timeliness of the data. The problem is self-evident. Because data from each type of entity are published separately, the data become available at different times, and the delay can be as long as two years. This is problematic for policymakers and analysts who have much shorter and demanding time horizons. Recommendations for Strengthening the Timeliness of the Data The study suggests two ways in which the information in the Controller’s report might more quickly reach the policy and research communities. Institute Internet/Web–Based Submission. The Controller’s Office should use a more direct submission technology. Widespread access to the Web, coupled with its easy information-transfer capabilities, make it an ideal medium for transferring information between local governments and the State Controller’s Office. xiv Make the Data Immediately Available. In addition to getting the information to the Controller’s Office more quickly and in a more consistent format, the data could be made available before they have been reviewed.3 This would make the bulk of the information available 90 days after the end of the fiscal year—a much more useful timetable than the current one in which data are not released for more than a year. Broader Implications of the Findings Beyond the specific recommendations discussed above, the study arrived at some broader conclusions and implications. Reliability The revenue data accurately portray the range of activity occurring in the local government sector. Thus, decisionmakers and analysts can focus on the methodology and substantive questions involved in a policy argument and not worry about shortcomings in the data. Usability PPIC’s primary motivation for undertaking this study was to determine whether the State Controller’s data were of adequate quality and, if not, what changes might be initiated to make the data more usable. While we have recommended a number of changes, the data as they currently exist are usable for further research into more detailed aspects of state and local governance. ____________ 3Unreviewed data would be identified as such, and since the information is reviewed by the Controller’s Office, the identifier could be changed accordingly. Large and complex entities, which are more likely to significantly affect policy choices, should be reviewed first. xv PPIC is committed to studying the implications of various governance and finance choices at the local level for local, state and federal policy choices. It is critical to this line of research that one have confidence in the quality of local government finance data. This study has found that confidence in the data is well-placed. xvi Tables 1.1. Local Government in California, 1991–92 .......... 2.1. California Community Facility District Revenues, 1991–92 ............................... 2.2. Results of Search for Missing Entities Using the LAFCO Lists .................................. 2.3. Results of Sample Search for Missing Entities Using the CDAC List .............................. 3.1. Overall Summary of Comparison of Reported City Sales Tax Revenues, 1991–92 .................. 3.2. Overall Summary of Comparison of Reported County Sales Tax Revenues, 1991–92 .................. 3.3. Overall Summary of Comparison of Reported City Property Tax Revenues, 1991–92 ................ 3.4. Overall Summary of Comparison of Reported County Property Tax Revenues, 1991–92 ................ 3.5. Comparison of School District Tax Revenues, 1991–92 ............................... 4.1. Summary of Comparison of Reported Overall County Revenues for Study Sample, 1991–92 ............. 7 24 27 28 34 35 37 38 39 50 xvii 4.2. Overall Variance in Revenues by Revenue Category for Counties, 1991–92 ......................... 4.3. Summary of Comparison of Reported Overall City Revenues for Study Sample, 1991–92 ............. 4.4. Overall Variance in Revenues by Revenue Category for Cities, 1991–92 ........................... 4.5. Overall Variance in Revenues by Revenue Category for Cities, Without San Francisco, 1991–92 ........... 4.6. Revenues for School Districts in California, 1991–92 ... 4.7. Dependent Entities Included in County Reconciliations............................ 4.8. Summary of Comparison of Reported Overall Hospital District Revenues for Study Sample, 1991–92 ............................... 4.9. Summary of Comparison of Reported Overall Transit District Revenues for Study Sample, 1991–92 ........ 4.10. Summary of Comparison of Reported Overall Transportation Planning Agency Revenues for Study Sample, 1991–92 .......................... 4.11. Summary of Comparison of Reported Overall Water and Sanitation District Revenues for Study Sample, 1991–92 ............................... 4.12. Summary of Comparison of Reported Overall Airport, Electric Utility and Harbor and Port District Revenues for Study Sample, 1991–92 ................... 4.13. Summary of Comparison of Reported Overall Non-Enterprise District Revenues for Study Sample, 1991–92 ............................... 4.14. Summary of Comparison of Reported Overall Special District Revenues for Study Sample, 1991–92 ........ 4.15. Overall Variance in Other Financing Revenues, 1991–92 ............................... A.1. Comparison of City Sales Tax Revenues, All Cities Aggregated by County, 1991–92 ................ 52 56 59 59 63 66 71 72 74 77 78 79 80 81 98 xviii A.2. Summary Statistics for Differences in Reported City Sales Tax Revenues, All Cities, 1991–92 ........... 100 A.3. Comparison of County Sales Tax Revenues, All Counties, 1991–92 ......................... 101 A.4. Summary Statistics for Differences in Reported County Sales Tax Revenues, All Counties, 1991–92 ......... 102 B.1. Comparison of City Property Tax Revenues, All Cities Aggregated by County, 1991–92 ................ 104 B.2. Summary Statistics for Differences in Reported City Property Tax Revenues, All Cities, 1991–92 ......... 106 B.3. Comparison of County Property Tax Revenues, All Counties, 1991–92 ........................ 107 B.4. Summary Statistics for Differences in Reported County Property Tax Revenues, All Counties, 1991–92 ....... 108 E.1. Overall Variance in Revenues by County, PPIC Sample, 1991–92 ............................... 128 E.2. Summary Statistics for Differences in Reported Overall County Revenues, PPIC Sample, 1991–92..... 128 F.1. Overall Variance in Revenues by City, PPIC Sample, 1991–92 ............................... 130 F.2. Summary Statistics for Differences in Reported Overall City Revenues, PPIC Sample, 1991–92 ............ 131 G.1. Overall Variance in Revenues for Hospital Districts, 1991–92 ............................... 134 G.2. Overall Variance in Revenues for Transit Districts, 1991–92 ............................... 135 G.3. Overall Variance in Revenues for Transportation Planning Agencies, 1991–92 ................... 136 G.4. Overall Variance in Revenues for Water and Sanitation Districts, 1991–92 ......................... 137 xix G.5. Overall Variance in Revenues for Airport, Electric Utility and Harbor and Port Districts, 1991–92 ........... 138 G.6. Overall Variance in Revenues for Non-Enterprise Districts, 1991–92 ......................... 139 xx Acknowledgments The authors gratefully acknowledge the assistance and cooperation of a wide range of associates and colleagues who contributed to this report. Given the considerable volume of detail work and the complexity of the information systems and sources addressed, special thanks must be extended to the many individuals who made this analysis possible. Several people were invaluable in the conceptualization and development of this study. PPIC colleagues John Ellwood, Elisabeth Gerber and David Lyon provided insight and guidance during the development of the proposal for this study. Rudy Nothenberg and Randy Hamilton were instrumental during the conceptual phase in providing a local government perspective on the problem, as were Professors John Kirlin and Jeff Chapman of the University of Southern California, who also served as technical reviewers on this report. This project also benefited from the input and perspectives of several committee staff members of the California Legislature, and PPIC is indebted to Peter Detweiler for arranging their participation. xxi This report represents the synthesis of information from numerous sources within state and local government. As a result, it was highly dependent on the assistance and cooperation of staff in many state offices and local governments. The California State Controller’s staff were key to the preparation of a significant part of this report. They always made themselves available to answer questions about the data and their procedures. Special thanks go to Alice Fong and Wayne Beck, who fielded the brunt of the project’s inquiries. Jeff Reynolds of the State Board of Equalization was equally helpful, answering questions and providing valuable supporting documents for the county auditorcontroller’s data. Peter Schaafsma and the staff of the California Debt Advisory Commission provided assistance and data key to major parts of this study. The staff at the United States Bureau of the Census in Washington, D.C., have also been extremely helpful in the pursuit and analysis of the data in this study. Special mention goes to William Hulcher, who fielded numerous inquiries and hosted and arranged an informational visit to the Bureau’s offices. Another critical group of government participants were the numerous city, county and district managers and staff who provided copies of the audited financial statements, answered questions about them and sometimes went to great lengths to explain the subtleties of their particular entity’s finances to our project staff. Without the assistance and cooperation of these hundreds of individuals, this study would not have been possible. The largest volume of work in this project was handled by an extremely competent project team which deserves recognition here. Liz Britton and Charles Castel were invaluable in obtaining information and xxii providing follow-up, while Mehtab Bains, Jason Cohen, Mary Joy Espino, Pradeep Mathew and James McMillan performed the often exhausting financial reconciliations. This report has benefited from technical peer review by several colleagues and associates, including Kim Rueben, Fred Silva and John Ellwood of PPIC, John Kirlin and Jeff Chapman of USC, Marianne O’Malley of the Legislative Analyst’s Office, and Steven Frates and Eric Norby of Claremont McKenna College. Their comments, collectively and individually, have contributed to the quality of this report. Finally, and certainly not least, Mary Sprague deserves special credit and thanks for her efforts on this project. She stepped in half-way through this project and has done an excellent job of helping to keep it on track. She has contributed to this project in numerous ways, including coordinating staff, following up on the “one hundred and one” open items needing closure, tracking down that “one piece” of needed data and keeping the special districts portion of the data set straight. Without her assistance, this project may well never have ended. While all of these individuals and organizations have been instrumental in the planning, preparation and execution of this study, the research and views expressed in this report are entirely the authors’. They should in no way be construed as representing the views of anyone acknowledged herein or anyone else. xxiii 1. Introduction For two decades Californians have been engaged in a great debate over the proper size, shape and role of their state and local governments and whether representative government should or is capable of constraining the growth of the public sector. This debate has centered around such questions as: How much of the state’s income should be spent by the public sector? Which goods and services should be provided by the state government? By local governments? What is the appropriate level of state and local taxation? And perhaps, most important, can citizens trust their elected officials to reflect their views on these policy questions or should these issues be governed by a combination of constitutionally imposed rules, limits and referenda? When this debate heated up in the 1970s, there was general agreement that Californians chose to provide more goods and services through their state and local governments than all but one or two other states. This was the era when California led the nation in per-pupil spending for elementary and secondary education, when the California 1 Master Plan for Higher Education promised highly subsidized statefunded postsecondary education for all its citizens, and when the state engaged in massive infrastructure investment in areas ranging from its road system to the dams and aqueducts that shift water from the north to its population-rich south. These and other activities required considerable resources, and as a result from the close of World War II until the late 1970s the state had one of the highest state and local tax burdens in the nation. Beginning with the passage of Proposition 13 in 1978, California’s voters used the initiative system to place limits on the size of its governments. Proposition 13 placed stringent constraints on local government’s ability to determine and raise revenues locally. Other initiatives—such as Proposition 4 (the Gann Initiative), which placed limits on the overall levels of spending for all state and local governments; Proposition 98, which created a floor on state spending and (indirectly) local spending on K through 14 education; and Proposition 111, which modified the rules of Proposition 98 for those years in which state revenues were depressed by a recession—significantly affected the ability of state and especially local government elected officials to control the amount and mix of publicly provided goods and services. When elected officials turned to governmental institutions and mechanisms not constrained by these initiatives—such as fees, service charges and special assessments—to provide additional goods and services, the advocates of constraining representative government sought to pass new initiatives to control and limit this activity. Most recently, voters approved Proposition 218, which requires a referendum on any attempt of local governments to raise revenues, service charges or special assessments. 2 On a policy level, these initiatives reflect a desire to control and limit the size of California’s governments. But at a deeper level, they reflect a basic distrust of representative government: a belief that elected officials are, in practice, incapable of imposing the public’s will when it comes to taxation and spending. In effect, over the past two decades, California has been engaged in a great experiment. It has led the nation in the use of direct democracy to limit representative government. This is the first of a series of reports in which the Public Policy Institute of California (PPIC) will evaluate the effects of this experiment. How effective have the rules imposed by the initiative process been? Has the size of California’s state and local governments been truly limited or have elected officials found ways around the limits? What effects have the initiatives had on the amount and quality of public goods and services in areas ranging from elementary and secondary education to health services to welfare to corrections? Given the devolution of many federal government activities to the state and its local governments, how will these constraints affect the implementation of such changes as welfare reform? These are all vital questions for the California policy debate and policymaking process. As discussed in the next section, several studies of issues in this area have pointed to deficiencies in the data, describing important aspects of the local public revenue burden. Before the key policy questions can be addressed, therefore, it is critical to clear up any concerns about the quality of the available data. This report does just that, laying the groundwork for the future work of the Public Policy Institute of California by assessing the quality of the available fiscal data regarding California’s local governments. 3 The Quality of the Data Is Important In the wake of Proposition 13, several studies have been undertaken to assess its full effect on the level of local government expenditures. Several of these studies have found some level of reduced spending in the state and local levels of government. Studies by the California Taxpayers’ Association (Cal-Tax)1 and Steven Gold2 have found that local government revenues dropped from 3 to 20 percent between fiscal years 1977–78 and 1991–92. Others have found that local revenues have in fact risen since Proposition 13. For example, Gary Galles and James Long found that state and local real per-capita general revenues in fiscal 1989–90 were 6 percent higher than in 1977–78,3 while John Kirlin et al.4 found a small percentage increase in overall public revenues. Part of these differences arises from the use of different measures (revenues as a percentage of personal income versus real per- ____________ 1From California Taxpayers’ Association, “California Taxing and Spending,” CalTax Research, October 1994, pp. 3–9. Cal-Tax found that state and local own-source revenues declined from 16.7 percent of state personal income to 16.2 percent—a decrease of 3.0 percent. 2From Steven D. Gold, “California’s Budget from a National Perspective,” testimony presented at a hearing on Proposition 165, California Budget and Fiscal Review Committee, California Health and Human Services Committee, California Senate, October 1, 1992. He found that state and local tax revenues declined from 14.6 to 11.4 percent of state personal income—a decrease of 21.9 percent. This measure is a little different than that of the Cal-Tax and Kirlin studies because it looks only at tax revenues. The other two studies focus on a broader measure of local revenues that includes current service charges, fines and the sale of fixed assets. 3From Gary M. Galles and James E. Long, Proposition 13 and California Tax and Expenditure Trends, Sacramento, Calif.: The Howard Jarvis Taxpayers Foundation, May 1994, Executive Summary. 4From John J. Kirlin et al., “Fiscal Reform in California,” in California Fiscal Reform: A Plan for Action, Oakland, Calif.: California Business-Higher Education Forum, June 1994. Kirlin’s group found that state and local own-source revenues increased from 17.4 percent of state personal income to 18.1 percent—an increase of 4.0 percent. 4 capita revenues) and part arises from the use of different comparison years—1989–90 was near the peak of an economic cycle, while 1991–92 was the first year of a severe recession. Another point of difference between the studies was the kinds of public revenues included. Certain studies included quasi-public transactions like tuition and fee revenues from the California State University system, while others did not. The purpose of this report is not to address authoritatively the differences found in these works, but to address another important issue that has been raised in response to these works. In nearly every case, practitioners and analysts used either the U.S. Department of Commerce’s Census of Governments or the California State Controller’s series of Annual Reports of Financial Transactions. In the text of the reports and in subsequent one-on-one interviews with the authors, the issue has been raised regarding the quality of the data included in the reports. In the Western Political Quarterly, there was recently an exchange of articles debating the methodology and quality of the U.S. Census Bureau’s data.5 While no one proposes bounds for the quality of the data, significant concerns are raised. The quality of the data is clearly an important issue in this debate. For example, if the picture of local and state government finance included in the data is off by 10 percent in either of the endpoint years, the findings of only one of the studies described above would be robust ____________ 5See James Leigland, “The Census Bureau’s Role in Research on Special Districts: A Critique,” Western Political Quarterly, June 1990, Vol. 43, No. 2, pp. 367–380; Seymour Sacks, “The Census Bureau’s Role in Research on Special Districts: A Critique, A Necessary Rejoinder,” Western Political Quarterly, June 1990, Vol. 43, No. 2, pp. 381– 383; and James Leigland, “In Defense of a Preoccupation with Numbers,” Western Political Quarterly, June 1990, Vol. 43, No. 2, pp. 385–386. 5 enough to stand. With findings of changes in the revenue picture in the 3 to 6 percent range, it is important to be comfortable that the data underlying these comparisons have a error level that is lower. This is the goal of this report. Before turning to specific quality issues, it is important to identify the specific scope and range of the local government enterprise on which this report will focus. Exactly what is meant when referring to local government finance? The next section will provide a more in-depth discussion of the scope and scale of the local government enterprise. The Local Public Enterprise The local public enterprise refers to the collection of governments beneath the state level that raises revenues, expends monies in the public interest and sets policy for citizens of the local community. The boundaries of all of these entities are subordinate to the state, and this category includes all counties, cities, towns, special districts and school districts. It typically does not include non-profit corporations and public-private joint ventures that are not under the direct control of an elected or appointed body. For purposes of this report, the category “local government” does not include the state government or any of its agencies. Table 1.1 presents a summary of the local government sector in California in the year 1991–92. As this table shows, the size of this sector is considerable, representing overall total revenues nearly twice those in the state’s general 6 Table 1.1 Local Government in California, 1991–92 (revenues in billions of dollars) Entity Counties Cities School districts Transportation/planning agencies Community redevelopment agencies Special districts Grand total Number 58 466 1,005 81 381 4,995 6,986 Total Revenues 28.253 22.311 25.030 3.210 2.026 14.891 95.721 Own-Source Revenues 14.112 19.938 7.371 2.624 1.961 14.431 60.437 SOURCE: Annual Report of Financial Transactions Concerning Counties, Fiscal Year 1991–92; Annual Report of Financial Transactions Concerning Cities, Fiscal Year 1991–92; Annual Report of Financial Transactions Concerning Special Districts, Fiscal Year 1991–92; Annual Report of Financial Transactions Concerning School Districts, Fiscal Year 1991–92; Annual Report of Financial Transactions Concerning Redevelopment Agencies, Fiscal Year 1991–92; Annual Report of Financial Transactions Concerning Transportation Planning Agencies, Fiscal Year 1991–92. and special funds in 1991–92.6 Even on an own-source basis,7 the sector is large, accounting for $60 billion dollars. The local government sector is also highly fragmented, with nearly 7,000 entities—mostly special districts. Of the nearly 5,000 special districts, approximately 40 percent are dependent upon another entity for their governance and fiscal direction. These entities are typically governed by a board of supervisors, city council or school board and are ____________ 6State general and special fund revenues totaled $53.1 billion in 1991–92. It is important to note here, however, that these are not all own-source revenues. If all intergovernmental revenues reported by these entities are excluded, the sector has revenues of approximately $50 billion—almost exactly the same size as the state government itself. 7Own-source revenues are those revenues raised at the level of that specific government. Typically, they are the entity’s total revenues less transfers from other governmental bodies and agencies and thus include tax revenues, fines and forfeitures, interest and rents and current service charges. 7 entities such as redevelopment agencies and water and sewer enterprises.8 The remaining 3,000 or so have their own elected boards and include entities such as fire protection districts, utilities, airports and water districts. What Data Are Available? Sources of financial information available for this diversity of local institutions are limited. There are only two sources of comprehensive information on these local governments—the U.S. Department of Commerce’s Census of Governments and the State Controller’s Annual Report—and these two are closely related as will be seen below. A few other sources are available for selected categories of information. The California State Board of Equalization maintains detailed information on selected revenue series, and the California Department of Education maintains a detailed database on K–12 fiscal activity in the state. Census of Governments The U.S. Department of the Census produces a series on state and local finance every five years for every public entity it tracks in the nation. Detailed information is also published for larger entities every year. The five-year data set includes detailed revenue, expenditure and debt information on counties, cities, school districts and independent9 special districts. Dependent entities are included with the parent entity that retains governance control over it. ____________ 8About 90 percent of these dependent entities are governed by county boards of supervisors, and approximately 10 percent are governed by local city councils. Very few are governed by school boards. 9An independent special district is one that has independent governance (e.g., its own elected board) and/or independent revenue streams. 8 The raw inputs for the U.S. Census Bureau’s series, however, are not collected directly by the Census. Instead, the Census obtains the data tapes for the Controller’s Annual Reports on Financial Transactions (described below) and reorganizes the data to make the presentation more consistent with their national reporting format. According to the Census, only minor comparison and review activities are affected by the Census, so the bulk of the quality control associated with this information rests with the California State Controller’s Office. This is due in part to staff limitations and in part to their sense of the high quality of the data received from the state.10 California Controller’s Annual Reports on Financial Transactions The California State Controller collects annual information on local government finance within the state. The Controller maintains and distributes an annual survey of financial activity that includes revenue, expenditure and debt information for every public entity in its database.11 All public entities are required by law to collect this information and to provide completed surveys to the Controller’s office within 90 days of the end of the fiscal year. ____________ 10One original purpose of this study was to provide detailed comparisons between the data reported by the Bureau of the Census and those reported by the California Controller’s Office. The Census data for the most recent Census of Governments survey year (1991–92) have not been released as of the press time of this report and are not expected until early 1997. As a result, that detailed comparison was not possible. Even though the Census uses the Controller’s data as its primary input, conclusions about the quality and comprehensiveness of the Controller’s data are not and should not be automatically applied to the Census data. 11A copy of one of the Controller’s questionnaires—the questionnaire for counties—is provided in Appendix H. The other questionnaires are similar, although the categories and level of detail vary according to the language of the particular legislation establishing the category’s reporting requirements. 9 These questionnaires are then entered into a database system within the Controller’s Office and reviewed for quality and consistency with the prior year’s reported activity. Inconsistencies are identified and investigated. If problems are identified or suspected, the public entity’s audited financial information is also requested as it becomes available. The compiled information is then published in a series of reports. There are separate series for counties, cities, redevelopment agencies, special districts, school districts and transportation planning agencies. Care is taken to be certain that there is no duplicate reporting of revenues. California State Board of Equalization The State Board of Equalization, among other responsibilities, is the primary tax collector and distributor for the sales tax, the fuel tax, the alcoholic beverage tax and the cigarette tax. For example, it collects the sales taxes from all business in the state and then distributes the appropriate shares to each entity that imposes a portion of that tax. In this capacity, it retains detailed information on the distribution of each of these taxes. Detail on state, city and county receipts of each of these taxes12 is reported in the State Board of Equalization’s Annual Report. Additionally, the board collects and reports detailed information on the state property tax. This information is actually obtained from surveys entitled Annual Report of Property Taxes submitted by each county auditor-controller to the Division of Local Government Fiscal Affairs, Special Districts Unit of the State Controller’s Office. This information is collected independently of the annual questionnaires ____________ 12Cities and counties receive revenues for only some of the taxes administered by the Board of Equalization. The sales tax is the largest of these. 10 described above. Summaries are presented each year in the Board of Equalization’s Annual Report. California Department of Education The California Department of Education collects detailed information on school district finances in California. These reports include the activities of school districts, county superintendents of schools and joint powers agencies (JPAs)13 between county offices and school districts. These data are collected by the Department of Education using the J-200 series, J-400 series and J-600 series of reports. Only summary information is published by the Department of Education—the detailed information for each entity is not published separately, although the Department of Education advertises its willingness to provide detailed information to anyone for a nominal fee. This information is not collected by any other state or federal agency, including the California State Controller’s Office. While the Controller’s Office does publish information on school districts, the data in the State Controller’s Annual Report on Financial Transactions Concerning School Districts are compiled from summary information provided by the California State Superintendent of Public Instruction and the California State Department of Education. ____________ 13A joint powers agency is typically a collaborative entity that provides a specific service for a group of independent agencies who benefit from the economies of scale of having a larger pool of participants. The classic example of a JPA is a pooled risk selfinsurance entity. Each individual entity is unwilling to carry the risks associated with selfinsurance, but if 10 districts pool their risk, then it becomes an attractive fiscal alternative. 11 Local Governments Finally, local governments retain information on their own and, sometimes, other entities’ fiscal activity. A range of entities exist that can provide information on local government finance. Counties, for example, collect all the property taxes for a given county. County local agency formation commissions (LAFCOs) are responsible for arbitrating boundary disputes, approving annexations and allocating resources for government reorganizations within each county, and although they retain information on what kind of entities exist within a county, they do not retain fiscal information. Parent entities nearly always retain information on the fiscal activities of subordinate, dependent entities. Each entity is also required by law to maintain detailed records on its own fiscal activities. This information is retained in the form of budgets14 and in annual audits—the results of which are reported in comprehensive annual financial reports (CAFRs). Summary of Data Sources In summary, there are only three sources of comprehensive information about the fiscal activities of local governments: the Census of Governments, the Controller’s Annual Report series and the entities themselves. Since the data underlying the Census reports are largely those included in the Controller’s reports, it is appropriate that this study will focus on the data in the Controller’s reports. It should be noted, however, that the U.S. Census Bureau does perform significant reorganizations of the data reported in the Controller’s Annual Financial Transactions and that any conclusions about one data set do not ____________ 14The information for a given year for an entity’s actual revenues and expenditures is often included in the proposed budget two to three years later as a reference point. 12 necessarily carry over to the other. Because of the unavailability of the Census reports, this report necessarily focuses on the Controller’s numbers. Issues About the Quality of Local Government Data In discussions with many of the policymakers in the State of California, one fact became evident—nearly all of the likely consumers and users of the available data on state and local finance did not use the available series, and, if they did, they did so hesitantly. Private investment firms, for example, did not directly use the information. Other researchers, largely academics, often deferred important research questions because of a lack of confidence in the available data. Three sets of issues were raised by these consumers as explanations of their non-use of the data: 1. There is a range of public activity and entities that are not included in the reported information. 2. There were concerns about whether the data reported accurately reflect the actual fiscal activity of the entities. 3. The information is not produced quickly or timely enough. The purpose of this project is to assess the available data along each of these dimensions, with a particular focus on the first two issues—the comprehensiveness and the accuracy of the data. Comprehensiveness of the Data: Are All Appropriate Entities Included? One of the biggest criticisms of the state’s data systems is that they are not capable of tracking the creativity of local governments. Because 13 the various propositions have greatly increased the constraints and limits on local governments, some argue that these governments have become creative in finding ways of going about their business. These individuals also argue that the state’s data systems do not and cannot keep abreast of the institutions that come about as a result of this creativity. As a result, the data do not capture these new activities, and the research that uses these data therefore underestimates the actual size of public sector activity. Suppose, for example, that, in response to the voter approval requirements under Proposition 13, a city used a law that allowed the creation of a community facility district (CFD) to build a new civic center.15 In year one, the CFD issues bonds and builds the civic center—which is exclusively operated by the city. In year two, the CFD collects property tax assessments (parcel taxes) from those in the CFD and pays the debt service on the bonds. If the CFD was not required to report its revenues or for some reason was not included, then the Controller’s reports would subsequently underestimate the true size of the public sector’s activity in that city. This is precisely the kind of issue that is raised by practitioners and experts. To address this concern, this study will examine whether there is in fact public activity that is not included in the Controller’s reports, with a specific focus on whether the reports include the full range of entities they should and how complete the coverage is within those categories. ____________ 15These are also known as Mello-Roos districts. As discussed in Chapter 2, this law would allow the city to draw the borders of the district in such a way as to assure the voter support necessary for passage. 14 Accuracy of the Data: Do the Reported Data Match What Actually Happens? Another issue raised in the policy debate regarding the quality of the data is whether the reported information accurately represents the activity of the reporting entity. Since the data questionnaires are completed by each entity independently, the potential for variation in the interpretation of the instructions across more than 6,500 entities is significant.16 For example, there is some variation in the way that entities of the same type account for different revenues and expenditures. For example, one district may report a property tax that it receives from the county tax collector and that it passes on to another entity as a tax revenue and an intergovernmental transfer expenditure. A similar district may report the transaction as an intergovernmental transfer revenue and expenditure. Another district of the same type may not report it at all because it has nothing to do with the collection or expenditure of the tax. There is enough latitude in both the Controller’s instructions and the rules that govern public accounting17 to take any one of the above approaches, and cases of all three approaches were identified in the course of this study. Finally, the Controller’s questionnaires are due within 90 days of the fiscal year end, while the audits of most entities are not due until 12 ____________ 16If these errors occur independently, it can still be the case that any two errors could cancel each other out and that the overall totals would be correct. To the extent that someone would want to look at individual entities, he or she would care about these errors. Also, it is possible that there is a systematic underreporting of data brought about by the design of the reporting systems. 17These are the Generally Accepted Accounting Practices (GAAP) rules by which public agencies’ financial statements are prepared during audits. 15 months after the year end.18 This means that the Controller’s surveys may be completed with budgeted numbers, not audited numbers. These numbers represent the best guess as to what happened in the year recently ended, but audits are necessary to make certain that all fiscal transactions are properly and consistently reported. Anecdotes of the consequences of such variability in accuracy abound. One scholar points to several hundred million dollars of revenues missing from a southern California special district. Another story recounts the example of a certain special district in central California whose Controller’s questionnaire was completed by the spouse of one of its board members between clients in her hair salon. All of these examples raise concerns regarding the accuracy of the data. For public policy researchers, such as the Public Policy Institute of California, to use these data with any confidence, these accuracy issues must be addressed definitively. Since this is the only data set with comprehensive information regarding local government entities and their finances, this report will look in some detail at the accuracy of the Controller’s data. Timeliness of the Data: Are They Available Soon Enough To Be Useful? Even accurate data may have limited usefulness to practitioners if they are not available quickly enough. The Controller’s reports have been criticized for the length of time it takes them to become available. In part, the delay is understandable. While the completed surveys are due within 90 days of the fiscal year end (typically by September 30), ____________ 18The one major exception to this standard is the redevelopment agency whose audit due date coincides with the 90-day due date of the Controller’s questionnaire. 16 there is a tremendous amount of work to be done before the published reports become available. The survey data must be input into the data system; the data entry must be quality controlled; the information must be reviewed for consistency and accuracy with prior years and then the final report must be compiled, reviewed, proofed and published. For the 58 counties in the state, this can be done quite quickly, but for the report that includes information on the state’s nearly 5,000 special districts, this task takes much more time. These reports are often not available for more than 18 months (or two fiscal year budget cycles) from the end of the reported fiscal year. This timeline is problematic for legislators and committee staff who must make policy and budget decisions in advance of a fiscal year. Especially in rapidly changing times, such as the recession of the early 1990s, two-year-old information is not useful. Decisionmakers need to know what happened as soon as possible and the sooner the better. There are no alternative sources of information regarding state and local finance. Policy analysts and academics, to the extent that they are also doing current policy research and analysis, need current data as well.19 This aspect of the quality of the data is self-evident in the timing of the release of the various reports. The county and city reports are typically issued much earlier than the special districts reports, but in no case soon enough for the state’s decisionmakers. This study will not focus on documenting this timing but does include some ____________ 19The Census of Governments reports are released much later. For example, the 1991–92 detailed information was due to be released in the spring of 1996. Because of budget cuts, however, this release has been delayed until late 1996 or early 1997—if it is released at all. 17 recommendations in Chapter 5 for ways of improving the timeliness of the data. This Report Due to these concerns about the quality of the data, this report was commissioned. Its purpose was and is to take a careful and critical look at the Controller’s data. If the data were found to be reasonably comprehensive and accurate, then the debate over changes in local public finance can focus on substantive methodological and value-oriented issues. PPIC could then turn to addressing important questions of fact and policy using the data. If the data were found to be problematic, then it was hoped that prescriptions for corrective measures could be identified that would make them usable. As will be seen in the balance of this report, the data are largely comprehensive and accurate. Focus on Revenues This report focuses on the revenue side of the financial statement, as opposed to looking at expenditures or debt. This is mostly the result of this study’s origins in the debate over the public revenue or tax burden in California. Revenues are also one of the simplest series of data included in the Controller’s reports—making such an analysis more feasible.20 It is also true that local governments typically balance their revenues and expenditures, and therefore a reconciliation of one category will, at least ____________ 20The opportunity for different interpretations of expenditure categories is greater because of the major variation in the specific characteristics of local programs. Revenues are simpler because the types and character of the sources of resources for local public entities are quite consistent within a revenue category. 18 at the highest level of aggregation, provide a validation of the other category.21 This report does not address the debt issue—an area that represents the bulk of the private sector’s interests in the data. The complexity of this area coupled with the lack of good secondary sources of information were the primary reasons for this choice. As some preliminary work shows, accurate and complete reporting in CAFRs and other agencies’ reports of debt activity are quite poor. This report will provide some preliminary findings of just one crude debt category in Chapter 4, and this is a recommended area of further research in Chapter 5. Organization of This Report This report centers around the comprehensiveness and accuracy of the Controller’s data and is organized around those tasks. Chapter 2 focuses on the comprehensiveness question. It presents findings on the quantity and magnitude of entities that are missing from the Controller’s data set. Chapters 3 and 4 then turn to the accuracy of the data. The former compares some of the individual categories of information included in the Controller’s reports with third-party reports of that same information, while the latter compares the reported information for individual entities against those included in the respective entities’ audited financial reports. In general, the body of the report includes summary descriptions of this report’s findings, while detailed tables are included in the ____________ 21Local governments are not allowed to run fiscal deficits. 19 appendixes. In each case, references are provided in the text and notes for each table in the body of the report, including the location of the detailed information underlying the summaries reported. 20 2. Are the Data Comprehensive: What Entities Are Missing? One of the main complaints about the Controller’s data centers around the concern that they may not include significant numbers of entities that should be included in the public sector. Failure to include entities would have two important implications: (1) the activity of these entities would escape public reporting and its implicit public accountability and (2) analyses using the Controller’s data would underestimate the full extent of public activity. The purpose of this portion of the study was to ascertain the extent of the omitted entity problem and to devise corrections for the data if the problem was found to be significant. Two approaches were taken to identify the public entities that are missing from the Controller’s reports. First, numerous individuals identified Mello-Roos or community facility districts (CFDs)1 as one ____________ 1Mello-Roos refers to the name of the two California legislators—Senator Henry Mello and Assemblyman Michael Roos—who sponsored the legislation that allows for 21 category of entities that were routinely excluded from the Controller’s reports. As a result, this study conducted a small survey of CFDs with an eye toward understanding precisely how many had been omitted from the Controller’s report. A census of these districts was also performed to identify the precise amount of revenues likely to be unreported. The second portion of the research into missing entities focused on non-CFD entities that might be missing from the Controller’s data set. This entailed comparing lists of public agencies from different sources to the entities included in the Controller’s reports. This phase of the research focused predominantly on the lists provided by county local agency formation commissions (LAFCOs) and the California Debt Advisory Commission (CDAC). Community Facility Districts Are Largely Missing Community facility districts allow local governments to issue bonds to cover the cost of financing new public facilities. Like those issued by special assessment districts, the bonds are repaid primarily through a special parcel tax on property within the district. Unlike special assessment district financing, CFDs do require approval from a supermajority of affected voters, but the district lines can be drawn to improve the chance of approval. In undeveloped areas with fewer than 12 registered voters, only landowners may vote. Most of the early CFDs were established through landowner vote. Although Mello-Roos financing began in the early 1980s, only a few CFDs were established in the initial years. By 1991–92, however, there were 233 different CFDs: 115 located in cities, 40 in counties, 52 in ____________________________________________________ the creation of community facility districts. These will be discussed in the following section. 22 school districts, and 26 in special districts (primarily redevelopment agencies and water districts). Mello-Roos bonds are not backed by the full faith and credit of their “parent” entity, and their revenues are not subject to the parent entity’s Gann appropriations limits. In addition, the Controller does not require the parent entities to include the CFD revenues in their report of fiscal activity. As a result, the special tax and other money earned for the CFDs represent a large class of local government revenues that is missing from the Controller’s report. How Many CFDs Are Not Included? As the research into Mello-Roos districts progressed, it became clear that, even though the Controller had no provision for their inclusion in its report and questionnaire formats, some parent entities did include them in their reported revenues. In response to this problem, 45 of the state’s 233 CFDs were randomly selected and surveyed to ascertain if their revenues were included in the Controller’s report in some way. The first discovery of this survey was the immense diversity in the positions of those who handle the bookkeeping and reporting for these districts. Sometimes the appropriate person was found in the accounting/finance department of the parent entity, sometimes they were found in a developer’s office and other times in an accounting firm’s office. These individuals were not always able to tell whether they were included in their parent entity’s finances or not. There was also a timing issue— because this study focuses on 1991–92, there was not always someone who knew what was done in 1991–92. From this small sample, it is estimated that approximately 20 percent of the CFDs are included in their parent’s reported information in the 23 Controller’s reports. These estimated 50 or so CFDs are affiliated with both large and small entities, although small parents included them more often than did larger parents. It is clear, therefore, that there is some degree of missing CFD activity in the Controller’s reports. How Big Is the Missing CFD Problem? To size this missing entity problem, nearly2 every CFD in existence in California in 1991–92 was contacted by PPIC. Information on all revenues the entity received in 1991–92 was requested. Since these are largely financing entities, these revenues are primarily property tax assessments and interest, with some contributions and fees from developers. This information was often aggregated with other activities in audited reports,3 so extensive follow-up was required to obtain a clear understanding of each CFD’s revenue picture. The findings of that census are presented in Table 2.1. Table 2.1 California Community Facility District Revenues, 1991–92 Revenue Category Property taxes Interest Other Total revenues Total Revenues $219,265,383 59,729,504 4,518,994 $283,513,881 ____________ 2Only five CFDs were not contacted because a contact person could not be identified. Their revenues were imputed for purposes of this portion of the analysis by using amounts from comparably sized CFDs. 3CFDs were sometimes aggregated with other debt service activities and sometimes with benefit assessment and special assessment districts. 24 All of the community facilities districts in the state have combined revenues of only $283.5 million.4 Because of difficulties in attributing interest revenues directly to the CFD in some cases, the interest revenues are likely underestimated by a small amount.5 The revenues associated with these CFDs represent a negligible part of the overall parent in each case. Even with a 10 percent increase in interest revenues, the resulting total would only represent a drop in the bucket—a mere 0.3 percent— relative to the size of the local economy presented before in Table 1.1. Community facility districts, therefore, do not represent a major problem in terms of overall effect on the size of local public revenues. Clearly, if one was interested in specific issues associated with these entities, the lack of data would be problematic. The existence of one category of missing entity does give rise to the question, “What other public entities may be missing?” The second portion of this chapter will focus on this problem. What Other Entities Are Missing? To find out what other public entities may be missing from the Controller’s reports, every list of public entities that could be obtained was compared with the list of entities included in the Controller’s reports. Entities were reviewed to see if they were directly included as ____________ 4A review of the number of CFDs formed since 1991–92 shows a 31 percent increase in the number of CFDs to date. If the revenues associated with these entities were proportional to those in existence in 1991–92, the total revenues for CFDs, which would be largely missing from the Controller’s numbers, would be $371,069,638—still a relatively small amount. 5In some cases, CFD interest revenues were lumped together with interest revenues from other activities of the parent entity. In other cases, the interest revenues were mixed together with the CFD’s interest revenues. It was impossible to separate these amounts, so they were excluded from the reported totals. This happened with 26 CFDs. 25 separate entities or indirectly included through the financial information reported by a parent entity. The two main sources used to locate missing entities were the county local agency formation commissions and the California Debt Advisory Commission. In addition, individual city and county audited financial statements and lists from some county auditor-controllers were consulted, but neither of these sources uncovered any entities that were not also covered by the LAFCOs and CDAC lists. With the exception of the Mello-Roos districts, very few entities were discovered that were not already tracked by the Controller. Missing Entities Identified from County LAFCO Lists Every California county, with the exception of Alpine and San Francisco, has a local agency formation commission that tracks changes in special districts and their spheres of influence, including changes in boundaries and the creation of new districts. Lists of special districts were obtained from all 56 LAFCOs in the state. Unfortunately, some of the LAFCOs do not keep detailed records, and, as a result, it was impossible to know when the entities were created and whether they were in existence in fiscal year 1991–92. The lists from the 56 LAFCOs were checked against the list of entities included in the Controller’s report, and 129 entities were identified that did not appear on the Controller’s list. In most cases, there was no address, telephone number or contact name provided. As a result, a telephone search was instituted for each of these entities through directory assistance, the county auditor-controller, city budget offices and other city officials, and other related special districts. Information was obtained for the vast majority of these potentially missing entities, but 26 there were still 10 that could not be reached or located. The findings of that search are presented in Table 2.2. Nearly two-thirds of the entities on the list turned out to be subsidiary agencies of larger entities, with their financial activity included in the parent entity’s report. In addition, 31 were formed after 1991–92 or had no financial activity in that year, and five were private water companies. Two entities were identified that did not seem to report their fiscal activity to the Controller: the Sonoma County Law Library and Reclamation District #813 in Sacramento County. The total revenues for these two entities were $382,299 in 1991–92.6 Missing Entities Identified from the CDAC List Any government entity or private agency that issues tax-free bonds must report the sale to CDAC. The list of all entities that issued debt between 1982 and 1992 was compared with the Controller’s lists, and 251 entities were identified that were not listed separately in the Controller’s report. A random sample of 40 of these entities was drawn Table 2.2 Results of Search for Missing Entities Using the LAFCO Lists Description Reported to the Controller Not in existence in 1991–92 or a private company Did not report to the Controller No information available Grand total Number 81 36 2 10 129 Percentage 63% 28% 2% 7% 100% ____________ 6The total revenues for the Sonoma County Law Library were $361,870 and for Reclamation District #813 were $20,429. 27 for further exploration.7 A telephone search similar to that described above for the LAFCO search was undertaken. The results of that search are presented in Table 2.3. Of the 40 potentially missing entities contacted, 26 were included in some parent entity’s report to the Controller, seven did not exist in 1991–92, information was not available on two,8 and only one (the Riverbank Public Financing Authority in Stanislaus County) did not report its activity. The remaining four could not be reached. The only identified missing entity’s income was $152,000 in tax revenues. An Overview of Missing Entities In the context of the findings in this chapter, it can be concluded that the Controller’s reports do include essentially all of the public revenue activity in the state. Combining the CFD, LAFCO and CDAC results, the total of missing revenues identified is less than $300 million dollars.9 In the context of the local government sector’s nearly $96 Table 2.3 Results of Sample Search for Missing Entities Using the CDAC List Description Reported to the Controller Not in existence in 1991–92 or a private company Did not report to the Controller No information available Grand total Number 26 7 1 6 40 Percentage 65% 18% 2% 15% 100% ____________ 7A sampling approach was chosen because of the difficulty encountered in reaching these entities to obtain detailed information on their reporting status and revenues. 8The entities were reached, but they were not able to provide the necessary data. 9The amounts identified in each of the three sections above, when combined, total $284,048,180. If this sample was extrapolated across the population, the total error would be $284,849,930. 28 billion in revenues, these missing entities represent a negligible amount of activity. As a result, unless one is looking specifically at policy issues that are closely associated with CFDs, the data are comprehensive enough for all intents and purposes. The next chapter will address the other major issue associated with the quality of the data—their accuracy. 29 3. How Accurate Are the Data: A Statewide Comparison of Sales and Property Taxes One approach to reviewing the Controller’s data is to take revenue streams for which there is good third-party reporting and compare that reporting with the amounts reported in the annual reports. In such an approach, both statewide and, when available, more detailed reports of revenues can be compared with those given in the Controller’s reports. This chapter describes the findings of such an exercise for two important local revenue streams—property and sales taxes. Both property and sales taxes are important sources of revenues for counties, cities and other local districts.1 Both of these local revenue sources, however, are collected and distributed by other levels of government. In the case of sales taxes, local businesses pay the sales taxes ____________ 1Sales taxes account for 1.3 percent and 10.0 percent of county and city overall revenues, respectively. Property taxes account for 23.4 percent of county revenues and 9.6 percent of total city revenues. 31 they collect from consumers directly to the California State Board of Equalization. The Board of Equalization then distributes these sales taxes to the state, county, city and local transportation districts. For property taxes, the revenues are collected by the county and distributed from there to each county, city, school and special district. A report is filed by each county’s auditor-controller to a division of the State Controller’s Office, which is in turn incorporated into the California State Board of Equalization’s Annual Report. These reports, inasmuch as they are prepared by independent offices within each county’s government, represent an important third-party reference against which the self-reported numbers in the Controller’s reports can be compared. The county auditor-controller’s reports, filed annually, include detailed entity-by-entity listings of the property taxes billed within each county. It is important to note that some level of error is expected between the sources for property tax revenues, inasmuch as the county auditor-controller’s reports detail taxes and levies billed while the entities in the Controller’s reports are reporting the amounts they have received. The comparison is useful, however, to get a sense of major errors in selfreporting in the Controller’s reports. The inclusion of information from slightly different points in the tax collection process also allows for a broader view of the quality of the data. In this portion of the analysis, these sales and property tax revenues reported by the individual entities in the Controller’s survey are compared with third-party sources. For sales taxes, the reported amounts2 are compared with the amounts reported by the California ____________ 2The reported amounts are drawn from Controller’s Annual Report of Financial Transactions Concerning Counties, Fiscal Year 1991–92, and Controller’s Annual Report of Financial Transactions Concerning Cities, Fiscal Year 1991–92. 32 State Board of Equalization for sales taxes in the appendix tables of the Board’s Annual Report for the Year Ending June 30, 1992. Reported property taxes are compared with the amounts listed by each county’s auditor-controller in its Annual Report of Property Taxes.3 This chapter is organized simply. The first section will examine sales taxes on a statewide basis and the second portion will focus on property taxes. Each section will address the types of entities that receive that type of revenue. Sales Taxes Only four types of entities in California receive sales taxes—the state government, county governments, city governments and local transit agencies. Since the focus of this analysis is on the local government data, the portion of sales tax attributed to the state government is not examined in detail. The sales taxes received by the latter group, local transit agencies, are reported separately in both the Board of Equalization’s Annual Report and the Controller’s report and are subsequently not considered here. This leaves the two largest categories of local recipients for state sales tax revenues—cities and counties. The specific findings for those two groups are presented below. ____________ 3These surveys are actually collected by a different division within the State Controller’s Office. They represent a third-party source of information because they are completed by a different office within local government—in many cases an office headed by an elected official. There is no joint handling of the reported information within the Controller’s Office. As a result, the information included in the Annual Financial Transactions series and the Annual Report of Property Taxes is separate and independent. 33 Cities Cities levy 1.00 percent sales tax on all taxable sales within their borders. Some cities share part of their levy with other entities and report only the portion they retain, explaining the variation in reported sales tax rates across California cities.4 Table 3.1 presents the results of comparing the sales tax revenues reported by cities in the Controller’s report with those reported by the Board of Equalization. This table includes no county revenues. As Table 3.1 shows, the amount of variation in aggregate between the Controller’s and Board of Equalization’s reports is negligible. On a county-by-county comparison, the largest variation in total sales tax revenues to cities is only 4.6 percent.5 Moreover, 82 percent of all cities reported sales taxes revenues to the Controller that were within 3 percent Table 3.1 Overall Summary of Comparison of Reported City Sales Tax Revenues, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Board of Equalization Difference Percentage difference Sales Tax Revenues 2,401,082,947 2,401,772,801 689,854 0.0% NOTE: See Table A.1 for underlying detail and additional references. ____________ 4In certain cases, there are pass-through arrangements with some subordinate and dependent entities whereby a portion of that levy is allocated to another government. The remaining portion of variance arises from special taxes for transit purposes that can be imposed at the county level. 5Table B.1 presents these findings on a county-level basis. In this approach the sales taxes for all cities in a county are added together and then compared. 34 of those reported by the Board of Equalization, and 88 percent reported to within 5 percent. Even on a city-by-city basis, the individual variation is quite small— averaging only 1.9 percent. There are only six cities in the state for which the individual differences between the Controller’s and Board of Equalization’s reported numbers are more than a million dollars— Alhambra, Azusa, Berkeley, Hayward, Sacramento and San Diego. In combination, their net variance is only $7.7 million out of total revenues of $189.3 million, or 4.1 percent. Clearly the data correspond closely between the two data sets. Counties The total percentage variation in the county sales taxes is a bit higher, although from a much smaller dollar total. The comparison of the Controller’s and the Board of Equalization’s sales tax series for counties is presented in Table 3.2. As this table shows, the overall level of variation is a bit higher than that found in the cities—5.1 percent. At the same time, individual counties’ variation is generally much lower. Table 3.2 Overall Summary of Comparison of Reported County Sales Tax Revenues, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Board of Equalization Difference Percentage difference Sales Tax Revenues 322,601,996 306,124,396 16,477,600 5.1% NOTE: See Table A.3 for underlying detail and additional references. 35 The amounts reported in the Controller’s surveys and the amounts reported by the Board of Equalization correspond closely, with 74 percent of counties showing differences of less than 3 percent, and 82 percent with differences of less than 5 percent. The variation reflected above is concentrated in just a few counties. Eighty-one percent of the variance arises from the six counties whose individual variation is greater than one million dollars—Alameda, Fresno, Sacramento, San Bernardino, Sonoma and Stanislaus. In each of these cases, the amount reported to the Controller is greater than that reported by the Board of Equalization. The data were examined to see if cities within these overreporting counties may be underreporting, thus offsetting the overreporting—no such offsetting pattern was identified. In this context, it is important to remember that 1991–92 was the first year of the recession and that, in many cases, actual sales tax revenues were often much lower than anticipated in county budgets. Since the Controller’s report is more likely to use budgetary numbers, it is possible that in these individual cases, the numbers reported are in fact higher than what actually occurred. Property Taxes The reporting and review of property tax revenues are complicated by the diversity of the entities that receive the property taxes. There is little uniformity in how the information on property tax revenues is reported across the various groups that receive them in the Controller’s reports. This section will present detailed findings regarding property tax revenues for two of the four major categories of entities that receive them: cities and counties. Aggregate data are reviewed for a third category: school districts. A review of the fourth group, special districts, 36 is not included because the Controller’s report for special districts does not always separately report property tax revenues. Cities A summary of the comparison for cities is presented in Table 3.3. As this table shows, the amounts reported in the Controller’s report6 and those reported by the county auditor-controllers7 correspond closely. In fact, nearly all of the statewide variance can be attributed to cities in Los Angeles County, and more specifically to the Cities of Long Beach and Los Angeles whose combined property tax revenues totaled more than $677 million.8 Without the $70.1 million variance from cities in Table 3.3 Overall Summary of Comparison of Reported City Property Tax Revenues, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports County auditor-controllers Difference Percentage difference Property Tax Revenues 2,959,177,966 2,844,810,343 114,367,623 3.9% NOTE: See Table B.1 or underlying detail and additional references. ____________ 6The specific revenue categories from the Controller’s reports include countywide taxes, less-than-countywide taxes, prior-year taxes, special district augmentations and voter-approved indebtedness. 7The specific series from the Annual Report on Property Taxes include all levies and taxes associated with county and less-than-countywide activities. It does not include the homeowners property tax relief amounts, which are typically categorized as intergovernmental revenues. 8The reporting of property taxes in Los Angeles has been the subject of much concern and review. A state audit completed by the State Auditor-General’s Office and entitled An Analysis of the Deficiency in the 1984–85 State School Fund (April 16, 1985, Report P-530) identified specific problems in the way supplemental assessments were handled. This could account for some of the variances identified here. 37 Los Angeles County, the overall variance would be $44.3 million on total revenues of $1.97 billion or 2.2 percent. These variances are negligible. Counties The information for property taxes for counties is presented in Table 3.4. As is the case in the cities, the bulk of the variance in reported county property tax can be attributed to Los Angeles County. Without the variance raised by Los Angeles County, overall variance drops to $83.5 million on total revenues of $3.373 billion or 2.5 percent. In percentage terms, almost all county auditor-controllers reported property tax revenues for their counties that were very close to amounts reported separately by the county to the State Controller’s Office. Fifty-four percent reported amounts within 3 percent of the amount reported to the State Controller, while nearly three-fourths were within 5 percent. Most of those not falling within 5 percent of the amounts reported by the county auditor-controllers were small counties where relatively small dollar variances represent large percentage variances. Table 3.4 Overall Summary of Comparison of Reported County Property Tax Revenues, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports County auditor-controllers Difference Percentage difference Property Tax Revenues 5,593,901,820 5,353,466,793 240,435,027 4.3% NOTE: See Table B.3 for underlying detail and additional references. 38 School Districts Reporting of school district finances by the Controller in general and property taxes in particular are very sketchy at best. The Controller’s Annual Report 1991–92 on Financial Transactions Concerning School Districts of California is a restatement of the information collected by the State Superintendent of Public Instruction and the California Department of Education on its J-200 series of reports. The categories included in this report are defined in the California School Accounting Manual, 1992 Edition, and do not correspond to any of the categorization included in the Controller’s other reports. Property taxes, however, are a major component of the revenues for K–12 school districts in California, and they are reported at the grossest level in the Controller’s report. The comparison of these reported numbers to those described in the Annual Report on Property Taxes filed by each county’s auditor-controller are presented in Table 3.5. Table 3.5 Comparison of School District Tax Revenues, 1991–92 (dollars unless otherwise indicated) Source Controller County auditor-controllers Variance Percentage variance Reported Property Taxes 5,309,829,960 5,365,498,883 –55,668,923 1.0% SOURCE: Controller’s data are from the Controller’s Annual Report of Financial Transactions Concerning School Districts of California, Fiscal Year 1991–92, p. IX. The county auditor-controllers’ data are from the original surveys submitted to the Controller’s Office entitled Annual Report on Property Taxes. These surveys are used to prepare the summary tables presented in State Board of Equalization’s Annual Report for the Year Ending June 30, 1992, Tables 4–5, pp. A-4 to A-19. 39 Overall, this level of variance is quite low. Unfortunately, no more detailed analysis of the district-level activity was possible because of the unavailability of detailed information in the Controller’s report. Furthermore, no detailed information, on a district-by-district basis, is published by the Department of Education. Overview It is evident that, when considered separately, the sales and property tax revenue information reported independently to the Controller by each entity corresponds quite closely to the amounts reported separately by the California State Board of Equalization and each county’s respective auditor-controller’s office. 40 4. How Accurate Are the Data: A Review of Data Reported by Individual Entities Another serious issue raised by consumers, users and potential users of the data was that the numbers reported by entities were sometimes, at best, poor estimates of actual activity. Anecdotes ranged from missing property taxes in the millions of dollars to a hairdresser in a small town completing the annual survey for a special district on whose board her spouse served. Others pointed to the fact that private sector firms such as Moody’s and Dunn & Bradstreet, whose livelihoods depend on good data, do not use the data included in the Controller’s reports.1 The overall concern about consistency voiced by practitioners was not so much at the aggregate level, but at the local level where individual ____________ 1It turns out that these firms generate their own information based on specific, current data provided by the entities requesting ratings for prospective debt issues. They typically focus only on the entity in question and do not have an overall statewide database. Their primary concerns about the Controller’s data seem to center more on the timeliness of the data than their quality. 41 public entities provided the responses to the Controller’s questionnaire. With such a broad and diverse range of public activities and accounting methodologies at the local level, there are concerns that any set of instructions could and would be interpreted differently by each individual entity. Looking at the questionnaire process from the local entity perspective, it is easy to see how this problem can arise. A copy of a typical questionnaire is included in Appendix H. An individual city manager does not have the comparative benefit of looking at what other cities are doing when completing the questionnaire. As a result, he or she would be more likely to follow the guidelines set forth in the city’s own budget, which is the result of local decisionmaking, than to consider the questions from the perspective of statewide consistency. Once this variation has been introduced, there is no good way for state data managers to identify this variation. In other cases, there is some question as to which point local entities should recognize revenues. Some entities, such as counties, receive funds, such as the property tax, which they then pass on to other entities. Whether these revenues should be accounted for as own-source revenues or intergovernmental transfers is another area of ambiguity. The result of these and other ambiguities is a perception that the information contained in the Controller’s reports is not consistent or complete. This chapter presents the findings regarding the quality of the data contained in the Controller’s reports. The first portion of this chapter will describe the general methodology used to prepare the findings detailed in the balance of the chapter. The remaining sections provide detailed descriptions of a major undertaking in which the Controller’s numbers were compared with audited numbers. 42 General Methodology and Issues For this part of the analysis, the goal was to see how well the information contained in the Controller’s reports reflects reality. This required comparing the reported information with a third-party representation of activity during the year. The third-party standards against which the Controller’s reports were compared were the comprehensive annual financial reports (CAFRs) for each entity. CAFRs are the results of annual audits required of all governments in the state. These reports are prepared by private sector accounting firms and reflect the application of the audit process and Generally Accepted Accounting Practices (GAAP) to the entity’s activities during the year. Since these audits do not have to be completed for a year after the fiscal year end, the results of these audits are not generally available for the Controller’s 90day deadline. In general, a sampling list was prepared for each category of entity— counties, cities and special districts. These samples were stratified by overall size. The largest entities in each category were all selected. A random sample of the remaining entities in each category was selected. The final sample contained entities from all sizes and revenue levels. The detailed sampling methodologies for each category are discussed in Appendix C. For each entity in the sample, a call was made to that entity requesting the CAFR for the 1991–92 fiscal year. Follow-up calls were made until the CAFR was obtained. Upon receipt, the values in the CAFRs were compared with those reported in the Controller’s reports. Since CAFRs typically define government entities to include all dependent entities over which they exercise control and the Controller’s reports include many dependent entities in their special districts report, it 43 was necessary in some cases to combine entities reported separately in the Controller’s reports to obtain an entity comparable to those listed in the CAFR. For example, a county’s CAFR would include all of the combined activities over which the county Board of Supervisors exercised control— such as a transit agency, a redevelopment agency, a water agency and any number of county service areas. Each of these latter agencies would be included separately in either the Controller’s Annual Report of Financial Transactions Concerning Special Districts of California or the corresponding book for redevelopment agencies. To compare these entities, it is necessary to combine the amounts reported in the Controller’s report for each entity with that of the county. It is this combined entity that is compared with the audited amounts reported in the CAFR. The specific reconciliation methodologies are given in Appendix D. Because of time and fiscal constraints, the comparisons in this study focused on only revenues for public entities in the sample for the fiscal year 1991–92. This raises several issues about the generalizability of this research to other fiscal categories, especially expenditures, and to other fiscal years, as well as comparability to other entities not in the sample. Moreover, there is a certain level of variability implicit in the differences between public fund and GAAP accounting. Each of these concerns is addressed in more detail below. Revenues Versus Expenditures as a Comparison Point This analysis focuses on the categorization and levels of revenues received by local entities in California and does not address the expenditure or capital portions of the equation. The focus on revenues 44 arises in part from the original motivation of this study—to validate available data before answering questions about the state’s changing public revenue burden, a task that will require a high level of confidence that these public revenues are accurately reflected in the available data. Additionally, revenues are typically more identifiable and usually defined by external forces. For example, sales and property taxes are determined by eligible sales volume and property values, respectively. Expenditures, however, are endogenous to the public policy process and hence more subject to manipulation and creativity. Revenue categories tend also to be more unambiguous and explicitly categorized across entities and activities, while expenditure categories are much more ambiguous. In aggregate, however, local entities tend to expend whatever revenues they receive in a given year, and aggregate revenues are a reasonable proxy for aggregate spending. This chapter looks closely at the aggregate revenues. As will be discussed below, categorization issues are difficult for both revenues and expenditures. Using 1991–92 as a Base Point For reasons of financial practicality, a choice was made to study one year in depth rather than reviewing several years more shallowly. It is critical to be confident that this study answers the question: “How well do the available data reflect public activity?” To be truly useful, a study of data quality over time would have to review data quality at least back to before the adoption of Proposition 13. If one ties into the Census of Governments, this would require going back to 1977. The cost of performing an adequate review of the quality of the data over such a long period of time would have been prohibitive. There is 45 the further issue of the availability of the CAFRs from each of the local entities. It was difficult enough to obtain the 1991–92 CAFRs from many entities, and prior year reports were often unavailable—a factor that would greatly increase the difficulty associated with completing this study and likely render it prohibitively expensive. Recency is also of concern when trying to reconcile the amounts listed in the two reports. While it is possible to find someone in a county, city or special district who can comment on their reported numbers for a recent year, it is next to impossible to obtain clarification and explanations for much older reports. In many cases, this information has been shipped away to storage or even destroyed. In light of these difficulties, it was decided to study one year in detail and 1991–92 was chosen. This year was selected because it represented the most recent year for which the U.S. Department of Commerce’s Census of Governments data were likely to be available.2 It is also recent enough that most entities would have their CAFRs readily available for this year. Over time, there have been some changes in the way that the State Controller’s Office handles the collection and aggregation of the numbers reported. In 1982–83, for example, the entire state government underwent a change in the basic accounting methodologies used to report revenues. Most of these changes were associated with the timing ____________ 2The original intent was to compare and reconcile the two data sets. The Census of Governments produces the comprehensive census of local government financial activity every five years in years ending in two and seven. Subsequent budget cuts have called into question whether the detailed 1991–92 data will in fact be made available. The data set was not available at the time this report went to press. If and when this data set does become available, a technical update comparing these data with the Controller’s data may be prepared. For the reasons discussed in Chapter 1, however, it is expected that these two data sets will map closely. 46 of when revenues are recognized and affected primarily the two- to threeyear window during which they were implemented. They had little longterm impact. Periodically, the instructions and requirements are updated and improved. Such an effort is currently under way and was most recently done on a large scale in 1984, when much of the current system was automated. In a pilot exercise to test the variability of the data over time, one entity was evaluated in 1976–77 and 1981–82. The results of this exercise were inconclusive, but they generally paralleled the findings for the 1991–92 fiscal year.3 The entity selected was the City and County of San Francisco, and, as the findings below will show, this particular entity had an exceptionally high level of variation and it was an entity for which explanations of differences between the audits and reports were not forthcoming. Brief reviews were also done for two other major entities, and the level of variation was comparable. Because of the limited nature of these exercises, only the results from the 1991–92 analysis are presented here.4 Some Variance Is Anticipated One important consequence of the methodology chosen in this part of the study is that a certain level of variance is to be expected. Public ____________ 3The city in question showed approximately the same level of variability in the prior years as in 1991–92. Contact with the city, however, failed to obtain confirmation as to the explanations of that variability. 4Given that the Controller has, over time, revised the methodologies and, in some limited cases, the reporting of information in the Controller’s Annual Financial Transactions series, there are some reasons to be concerned about the full reliability of earlier years. For the purposes and goals of this study–to validate the data with a view toward looking at local government creativity–it was decided that a recent year would be much more useful. The user is cautioned when using data from other years, however, that there may be individual quirks in other years that this study does not bring to light. 47 fund accounting reports are being compared with GAAP-based accounting reports. There are differences in the conventions of these two accounting techniques that ensure that the numbers will usually not match exactly. For one difference, public fund accounting is essentially on a cash basis—local entities recognize cash receipts as revenues when they receive them and recognize expenditures when they make payment.5 The audited reports are reported on an accrual basis—funds are recognized in the year they are earned. This difference will have an effect in years when the revenue stream is either increasing or decreasing and the amounts carried over from the prior year differ from the amounts carried into the following year. To better understand this concept, consider the following hypothetical example.6 Dream City has monthly sales tax revenues of $10 million. The city receives the monies for each month on the first day of the following month. If the city is on a cash basis, in a given year it would receive 12 times $10 million or $120 million. The first payment, received on July 1, however, was not for revenues received in this year, but for revenues from the last month of the prior year.7 If monthly revenues remain flat over the period, however, the additional revenues received from last year would exactly equal the revenues not received in the current year. If, however, revenues from the last month of the prior year totaled $10 million and revenues from the last month of ____________ 5This is not universally true for all public entities in the study. In some cases, entities use a modified cash approach where some revenues and expenditures are accrued to the fiscal year in question. 6This example is a simplified stylization to demonstrate how the methodologies differ. Actual accounting practice is more elaborate and complicated. 7The last month of this year’s revenues will be paid in the following fiscal year. 48 the current year totaled $15 million, there would be an apparent discrepancy between the two. On a cash basis, sales tax revenues would still be $120 million. On an accrual basis, however, revenues would total $125 million (11 times $10 million plus $15 million). If one compared the cash-based report with the accrual-based report, there would be a discrepancy of $5 million. This is just one type of difference that can arise when comparing the entity public fund accounts with the audited reports. Because of these types of variations, a certain level of variation is expected in these reconciliations. The exact amount of variation attributable to these types of errors in each entity is nearly impossible to precisely quantify because of the multitude of accounting positions that can be taken by a particular enterprise. Discussions with experts in the exploratory phase of this study placed this expected error within the 1 to 5 percent range—but it could be higher or lower. Some entities clearly used audited information to complete the Controller’s questionnaire, and their variance is essentially zero. Others are higher. If the variance is unexplainable or not attributable, it is so identified. Overall Findings Overall, this research found that the amounts reported in the Controller’s reports closely correspond to the revenues reported in the audited financial statements. The level of variation, given the complexity and range of activities within each group of entities, was remarkably low. Both the local governments, which provide this information, and the State Controller’s Office, which compiles and reviews it, should be commended for the overall accuracy of these data. 49 Specific Findings: Counties Fifteen of the state’s 58 counties were examined and reconciled to identify the level of variation between the numbers reported to the Controller and the amounts reported in their audited financial statements. The sample was drawn by selecting the 7 counties with the highest revenues in the state and then a random sample of 8 of the remaining 50.8 See Appendix C for more details regarding the sampling methodology. Overall Variance by County Overall, the aggregate difference between the audited amounts and the reported amounts is relatively small. Table 4.1 presents a summary of the results for the 15 counties in the sample. These counties reported $22.22 billion of revenues to the State Controller, and the audited financial statements included $22.50 billion of revenues. The difference of $276,318,630 represents a mere 1.2 percent of total revenues—a Table 4.1 Summary of Comparison of Reported Overall County Revenues for Study Sample, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Audited financial reports Difference Percentage difference Overall Revenues 22,220,798,294 22,497,116,924 276,318,630 1.2% NOTE: See Table E.1 for underlying detail and additional references. ____________ 8The Controller’s convention of excluding the City and County of San Francisco was followed for these reconciliations. Hence the eligible pool of counties is 57, not 58. Note that San Francisco is included in the city sample below. 50 variance level that is well within that attributable to differences in accounting methodologies. The range of the overall variance within counties in the sample is from 0.0 to 3.1 percent. In most cases, the larger percentage variances are in the smallest counties where relatively small dollar differences can account for large percentage variances. The one exception to this is San Diego County, which shows a variance of 3.0 percent. Almost half ($25.8 million) of this variance is attributable to property contributions from property owners in conjunction with a development project. These are routinely included as part of certain Controller’s special district questionnaires, but there is no such line in the county questionnaire. These were reported as miscellaneous revenues in the audited numbers. Categorical/Classification Variance Another dimension of the values reported in the Controller’s report is whether the categories in which the revenues are reported correspond to those in the audited reports. For example, are counties really reporting all revenues from the use of money and property? These revenues are typically rents and interest from various property holdings and investments. Table 4.2 reports the net overall variance broken down by revenue category. As this table shows there are some differences between the classification of revenues in the Controller’s reports and in the audited financial reports. The specific character of these classification differences, however, should be downplayed. For example, GAAP does not specify a separate category for special benefit assessments for counties, and the 51 Table 4.2 Overall Variance in Revenues by Revenue Category for Counties, 1991–92 (dollars unless otherwise indicated) Category Controller’s Report CAFR Percentage Difference Differencea Taxes 5,231,000,836 Special benefit assessments 9,850,947 Licenses and permits 147,798,329 Fines and forfeitures 204,368,563 Revenues from the use of money and property 563,547,755 Intergovernmental funds 10,236,203,283 Current services 3,250,273,883 Other revenues 2,577,754,698 Total 22,220,798,294 5,217,461,331 0 164,859,169 210,305,264 771,650,447 10,332,996,948 3,379,490,060 2,420,353,705 22,497,116,924 13,539,505 9,850,947 –17,060,840 –5,936,701 –208,102,692 –96,793,665 –129,216,177 157,400,993 –276,318,630 0.3% 100.0% –11.5% –2.9% –36.9% –0.9% –4.0% 6.1% –1.2% aPercentage difference is the difference as a percentage of the Controller’s reported amounts. CAFRs subsequently do not separate these revenues. If the first two tax categories are combined, the overall variance for taxes rises to 0.4 percent. It is also important to look at both the absolute and percentage variances. Note that the “licenses and permits” category has an 11.5 percent variance but it totals only $17 million. Compare this with the 4.0 percent in the “current services” category, which accounts for $129 million. There is likely a significant amount of simple categorization issues where entities have chosen to include activity reported in a specific category in the CAFR as “other revenues.” Note that the Controller’s report reflects significantly more revenues in this category than the audited financials. Since the Controller’s questionnaire is more specific 52 than the guidelines presented under GAAP, it is likely that the amounts in the Controller’s report are actually more consistently reported. Overall, however, it is clear that the specific categories do not point to major classification problems for counties. The one category raising a systematic issue is the “revenue from the use of money and property” category. Here relatively large variances are manifest in both absolute and percentage terms. One contributing factor is the reporting of net interest income in the county enterprise activities. Another possible explanation is the interest revenues associated with the missing community facilities districts identified in Chapter 2.9 Under the nonoperating revenue portion of the county enterprise activities is a line for interest revenues. Several counties in our sample reported these interest revenues to the Controller net of interest expenses for the enterprise activities. As a result, both interest revenues and expenses are understated and the resulting variance appears in Table 4.2. Issues Identified In the course of these comparisons, the following four general issues were identified that represent areas where improvements can be made in the reported information for counties: • In some cases, counties excluded both capital projects and debt service funds from the amounts reported to the State Controller’s Office. This accounts in part for the underreporting of “revenues from the use of money and property” identified in Table 4.2. ____________ 9Although this would only represent a small proportion of the overall difference. 53 • As discussed above, some county enterprise activities report interest revenues net of interest expenses in the non-operating revenue section of the county enterprise questionnaire. This underreports the total interest revenues for the entity. • There is no consistency in how contributions from property owners are treated between entities. In some cases, private citizens (usually developers) will contribute property in the forms of streets, sewers, utility lines, etc. to the county in exchange for the provision of a service, such as the formation of a special taxing area. In the case of certain special districts, there are specific instructions on how this should be handled. In the case of at least one county, these revenues were not reported to the Controller because there was not a place or instruction to accommodate such in-kind contributions. This topic will be raised again in the section below discussing consistency among the various reports. • As discussed in Chapter 2, the reporting of Mello-Roos district revenues is very inconsistent. Some counties report portions of the receipts as tax revenues, some report only the interest earned and others do not report them at all. The full effect of these estimates is given in Chapter 2, but a more consistent methodology for addressing these entities should be introduced. In a broader sense, however, the revenues reported for counties in the sample track the audited numbers quite closely. An aggregate variation of 1.2 percent indicates that the overall quality of the county data is quite good. When the above four types of corrections are made to the data, that rate drops to 1.0 percent. 54 Specific Findings: Cities The values reported to the Controller by cities were also compared with their respective CAFRs. In this portion of the study, 52 cities were studied at a detailed level (out of a total of 466 for the state). The cities were selected first by population10 and then at random from the remaining cities.11 The largest 15 cities in the state by population were selected,12 as well as all 14 of the cities in Alameda County.13,14 The remaining 25 cities were selected at random from the rest of the cities in California.15 A subsequent review showed that the sample encompasses the entire state geographically and includes cities of all sizes. See Appendix C for more detail on the city sample. As with the counties, each city’s CAFR was obtained via telephone and compared with the values reported to the State Controller in the Annual Report of Financial Transactions Concerning Cities of California, Fiscal Year 1991–92. The detailed methodology for reconciling the two sets of reports is included in Appendix D. In cases where the research ____________ 10Population and total revenues correlated closely. Since the purpose of the study was to identify errors (if any) in the revenue data, population was chosen as the final criteria instead of revenues reported to the Controller’s Office. 11For purposes of consistency and comparability, the City and County of San Francisco was treated as a city. 12The City of San Bernardino was not included in the final results because the city’s Comprehensive Annual Financial Report was not available in time to be included in the study. 13Originally, this study was to obtain information on all of the entities in two to four counties. After completing a preliminary study of Alameda County, it was determined that there were good reasons to turn to the statewide approach instead. These options and choices are discussed in greater detail in Appendix A. 14The findings presented in this section are generally robust, even when the oversampling of Alameda County is reversed. 15Twenty-eight cities were selected at random, but only 25 CAFRs were received in time to be included in this analysis. 55 team could not explain the variance between the two reports, the cities were contacted for further explanation. Overall Variance The overall variance between the Controller’s and audited reports for the cities in the sample was quite low—only 1.4 percent, as shown in Table 4.3. The Controller reported revenues of $16.30 billion for the 52 cities in the sample and the audited financials included $16.08 billion for an overall variance of $226,575,549. There was one major outlier—the City and County of San Francisco—which contributes significantly to the variance and biases it in a direction opposite of the overall trend. In general, the CAFRs show more revenues than the amounts reported to the Controller. The City and County of San Francisco, however, reports $291,096,334 more to the Controller than is on its CAFR.16 Note that this is more than the Table 4.3 Summary of Comparison of Reported Overall City Revenues for Study Sample, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Audited financial reports Difference Percentage difference Overall Revenues 16,302,293,293 16,075,717,744 226,575,549 1.4% NOTE: See Table F.1 for underlying detail and additional references. ____________ 16There is some reason to expect a higher degree of variation for San Francisco. As a combined government (both a city and a county), its range of activity and institutions are much more complex than a typical city. Furthermore, the inconsistencies between the reporting formats for cities and counties are emphasized because San Francisco’s activities have to be classified into one of the two formats. 56 overall variance reported above. Numerous attempts to obtain explanations of this variance from the City and County of San Francisco were unsuccessful.17 Erring toward the conservative approach, this variance is left in the aggregate comparison. If this outlier is omitted, however, the overall aggregate variance would total only $64,520,785 or 0.5 percent. The overall variance totals for the cities in the sample are presented in Table 4.3. These levels of variance for the cities in the sample are quite low. More than half of the cities had variances of less than 2.5 percent. Seventy percent of the cities show overall variances of less than 5 percent, and 80 percent reflect variances of less than 10 percent. Specific explanations were identified for most of the cities with variances greater than 3 percent. The range of percentage variation is from 0.0 percent to 87.2 percent. Of the 10 cities with variances greater than 10 percent, nearly all of the variance can be attributed to one of three issues: (1) housing authorities are not consistently included in the Controller’s reports; (2) debt service and capital projects activity are not always reported in the Controller’s questionnaire; and (3) some cities generate overhead estimates of general management to various city departments that are then reported as “quasi-external transactions.” This is not done consistently, however. The issues identified in this analysis are discussed in greater detail below. ____________ 17Several offices within the city were contacted to obtain explanations. The staff that responded were unable to reconcile the information and attributed the differences to “differences in the accounting methodologies.” Given the closeness of most other cities in the sample, this explanation does not appear adequate. 57 Two of the three outliers—Irwindale (87.2 percent) and Westmorland (31.1 percent)—have unique circumstances in 1991–92. Irwindale reports $102,500,000 in other revenues associated with a bond issuance undertaken several years before to fund a sports stadium. The bond proceeds were never spent and the bonds were paid off in 1991–92. The proceeds were recognized as revenues in the Controller’s reports but did not really reflect new revenues in 1991–92. They should not have been reported as revenues to the Controller and are consequently recognized as a variance in this analysis. The City of Westmorland apparently does not include water and sewer enterprise revenues when reporting to the State Controller. The variance in the third outlier, Sacramento (27.5 percent), is attributable to the exclusion of its housing authority in the Controller’s report. Categorical/Classification Variance The second dimension of variance is associated with how revenues are categorized. The overall variance for the sample by revenue category is presented in Table 4.4. As discussed above, the report from the City and County of San Francisco introduces an abnormal amount of variance that biases the results. To obtain a better overall sense of the data, therefore, it is useful to look at the data without the effect of the City and County of San Francisco. Table 4.5 presents the same summary as Table 4.4, without San Francisco. This table shows that the application of the classifications in both the Controller’s and GAAP methodologies are fairly consistent. As with counties, many cities’ CAFRs do not have separate categories for “special 58 Table 4.4 Overall Variance in Revenues by Revenue Category for Cities, 1991–92 (dollars unless otherwise indicated) Category Controller’s Report CAFR Percentage Difference Differencea Taxes 5,399,824,479 Special benefit 130,522,897 assessments Licenses and permits 144,852,321 Fines and forfeitures 198,783,522 Revenues from the use of money and property 900,008,611 Intergovernmental funds 1,767,194,569 Current services 6,964,310,811 Other revenues 796,796,083 5,468,054,306 –68,229,827 79,423,852 51,099,045 211,474,171 –66,621,850 146,036,789 52,746,733 931,213,463 2,035,941,895 6,748,566,129 455,007,139 –31,204,852 –268,747,326 215,744,682 341,788,944 –1.3% 39.1% –46.0% 26.5% –3.5% –15.2% –3.1% 42.9% Overall 16,075,717,744 16,075,717,744 226,575,549 1.4% aPercentage difference is the difference as a percentage of the Controller’s reported amounts. Table 4.5 Overall Variance in Revenues by Revenue Category for Cities, Without San Francisco, 1991–92 (dollars unless otherwise indicated) Category Controller’s Report Taxes 4,446,001,944 Special benefit assessments 130,522,897 Licenses and permits 128,546,662 Fines and forfeitures 149,250,412 Revenues from the use of money and property 779,003,664 Intergovernmental funds 1,168,490,709 Current services 5,920,706,513 Other revenues 626,255,158 Overall 13,348,777,959 CAFR 4,529,404,306 Percentage Difference Differencea –83,402,362 –1.9% 79,423,852 51,099,045 155,097,171 –26,550,509 146,036,789 3,213,623 39.1% –20.7% 2.2% 825,904,463 –46,900,799 1,408,195,895 –239,705,186 5,854,659,129 66,047,384 414,577,139 211,678,019 13,413,298,744 –64,520,785 –6.0% –20.5% 1.1% 33.8% –0.5% aPercentage difference is the difference as a percentage of the Controller’s reported amounts. 59 benefit assessments.” If these are rolled up into the tax totals, the overall in the “taxes” category declines to $32,303,317 or –0.7 percent. If San Francisco is included, this drops further to $17,130,782 or –0.3 percent. There appear to be some offsetting classification differences between the Controller and GAAP definitions of “intergovernmental funds,” “current services” and “other revenues.” Overall, the classification differences are quite minor. Issues Identified Numerous issues have been identified in the city portion of this analysis. Some of the issues that affect cities directly are detailed below. Some broader issues that affect the reporting of all public entities will be discussed later in this chapter. • One of the most frequent areas of variance in this analysis is the treatment of city housing authorities and their associated federal and state revenues. These entities are not consistently included in the numbers reported to the Controller for cities. Most cities seem to include them, and both the questionnaire and follow-on discussions with the Controller’s Office indicate that they should be included in the city questionnaires. Yet several major cities do not include them. Additionally, specific instructions must be given to cities that clarify that they should be included. • Many cities failed to report debt service funds and, occasionally, capital projects funds in the amounts reported to the Controller. The Controller’s Office has subsequently instituted a clarification in its instructions to cities that has made it more explicit that these amounts should be included. Compliance with these instructions should be verified and the instructions reiterated if they still are not being followed. 60 • Some cities generated estimates of interdepartmental charges for overhead services provided by the city manager and general administrative departments and reported these amounts as “quasi-external transactions” to the Controller. A careful review of the instructions leaves open such an interpretation. One city indicated that these calculations were produced and used exclusively for the Controller’s survey. It does not appear that such activity is the goal of the specific set of instructions in question, but they should be further clarified. If, however, these cities are meeting the intent of this portion of the survey,18 then it should be made more explicit so that all cities perform the necessary calculations and report the appropriate numbers. • In the case of some city-operated enterprises, interest revenues are often reported net of interest expenses. This is the same issue raised in the county discussion above. It understates both interest revenues and interest expenses. • Occasionally city public financing authority revenues are not reported to the Controller. This happened once in a small city. Since these are not reported elsewhere, however, these public revenues are unreported. • As discussed above, one city used existing resources (proceeds from bonds issued in a prior year) to pay off the outstanding debt. In the process of doing this, it reported revenues to the Controller totaling the amount of the previously issued debt plus earnings. While the reporting of the earnings was appropriate, the reporting of the bond revenues was inappropriate. • Special benefit assessments are inconsistently and poorly reported. The definition provided is ambiguous and variously ____________ 18If this choice is open to decision, it is recommended that the Controller not institute such a calculation. These are merely accounting entries and do not represent substantive or even directly programmatic revenues of any kind. 61 interpreted. In one city, street and lighting assessments were not reported to the Controller. In another case, one city reported a significant amount of debt proceeds as special benefit assessment revenues. Concurrently, the actual special assessments to pay off these bonds were reported as revenue as well. As a result, these bonds were double counted. Recent investigations by numerous state and private entities, motivated by the introduction of Proposition 218, have found that the data reported in this portion of the survey are at best incomplete. A more explicit set of instructions needs to be provided to make this information as complete and relevant as possible. • Finally, several cities omitted enterprise activities that were consistently reported on other cities’ surveys. One city omitted its transit enterprise, while another omitted both its water and sewer enterprises. As a result, these revenues were excluded from the final reported numbers and were not picked up elsewhere. By and large, however, the data on cities are excellent. Only the housing authority, capital projects and debt service fund issues were pervasive across several entities (three or more). The remaining issues occurred in only one or two cases (out of 52) and, while worthy of mention, were not common. Specific Findings: School Districts The Controller provides summary information for the state’s 1,005 school districts in California in its report entitled Annual Report of Financial Transactions Concerning School Districts of California. The State Superintendent of Public Instruction is required by state law to provide summary information to the State Controller’s Office for 62 compilation of this report.19 The information in this report is provided in a very aggregated form—listing revenues and expenditures by category only for the whole state and by entity category.20 Total revenues and expenditures are provided, unaudited by the State Controller, for each district and K–12 entity in the state. These revenues are reported in Table 4.6. The categories included in this report conform to the account classifications prescribed in the California School Accounting Manual, 1992 Edition. As such, they are not consistent with any of the other reporting conventions included in the Controller’s Annual Financial Transactions report series. For example, the “state aid” presented in Table 4.6 is largely “aid from other governmental agencies”—the state government in this case. The “local taxes” category represents local Table 4.6 Revenues for School Districts in California, 1991–92 Revenue Source State aid Local taxes All other Total federal Other state Other local Other financing sources Total revenue Revenues $11,205,074,937 5,309,829,960 1,607,079 1,955,746,196 4,498,593,156 2,059,578,526 2,115,733,687 $27,146,163,541 SOURCE: Controller’s Annual Report of Financial Transactions Concerning School Districts, Fiscal Year 1991–92, p. IX. ____________ 19These requirements are specified in Section 53892.1 of the government code. 20The categories separated in the report are “School Districts,” “County Schools,” and “Joint Powers Agencies.” 63 property tax revenues. For consistency and clarity, these differences should be addressed. Another aspect of this report is that it provides only summary information for school district revenues and expenditures. There is no entity-by-entity presentation of information as is done with all other public entities in the state. Because school districts account for a major portion of local government spending in the state, this absence of detailed information should be addressed. Finally, there is no separate reporting of community college districts in the Controller’s reports. While the California Postsecondary Education Commission does publish some aggregate information for these entities, there is no district-by-district reporting of the type described above. Inasmuch as these community college districts, like elementary and secondary school districts, are local public entities with locally elected boards and local voter accountability, their fiscal information should be directly available to local voters, analysts and decisionmakers. Specific Findings: Dependent Special Districts Dependent special districts are those entities whose governance and/or finances are controlled by the governing body of another entity. Most often these entities are controlled by a city council or a county board of supervisors, although there are occasions when these entities are controlled by school boards or the governing boards of independent special districts. In 1991–92, the Controller obtained and reported financial information on 1,856 dependent special districts in its Annual Report on Financial Transactions Concerning Special Districts, 1991–92. 64 Obtaining audited financial information separately for these entities is not possible. Because CAFRs often include these entities, however, it is possible and even necessary to include them as part of the reconciliation of their parent entities. As a result, this study implicitly reviews numerous dependent special districts in the course of reconciling cities and counties. This study also performed a “spot check” on redevelopment agencies to test whether they, in fact, do rely on audited numbers, as required by law, in their submissions to the State Controller’s Office. Dependent Entities Reviewed as Part of City and County Reconciliations In the course of preparing the results for the counties and cities described above, it was necessary to aggregate dependent entities into the reconciliations to provide comparable entities between the two reporting formats. This process means that the above documented portion of the study effectively reviewed a significant number—more than onefourth—of the 1,856 dependent special districts listed in the Controller’s annual reports. In the course of reconciling the 52 cities presented in the immediately preceding section, 43 out of the state’s 491 redevelopment agencies were reconciled as part of the process. These redevelopment agencies are reported separately in the Annual Report of Financial Transactions Concerning Redevelopment Agencies of California, Fiscal Year 1991–92. This is in addition to the specific review of redevelopment agencies described later in this chapter. In addition, 11 other dependent entities were reviewed as part of the city reconciliation. This would include any entity explicitly included in a 65 city’s CAFR and showing activity in the Annual Report of Financial Transactions Concerning Special Districts of California, Fiscal Year 1991– 92. These 11 entities include the Estero Municipal Improvement District, the Livermore Recreation and Park District, the Modesto Municipal Sewer District, the Napa County Housing District, the Parking Authority of Oakdale, the Redding Area Bus Authority, the Automated Regional Justice Information System, the San Diego Data Processing Corporation, the Ventura County Water District, the Simi Valley Sanitation District and the Yucca Valley Recreation and Park District. Reconciling the 15 counties incorporated many more districts— nearly 500. The specific types of entities are reported in Table 4.7. Table 4.7 Dependent Entities Included in County Reconciliations County Alameda Humboldt Los Angeles Orange Placer Plumas Riverside Sacramento San Benito San Bernardino San Diego San Luis Obispo Santa Clara Siskiyou Tehama Total Redevelopment Agencies 0 0 1 1 1 0 1 1 0 1 1 1 0 0 0 8 Enterprise Districts 1 1 3 2 2 1 5 4 1 2 3 3 2 3 0 33 County Service Areas 16 1 1 9 5 6 67 2 18 54 77 13 0 1 0 270 Other Dependent Districts 4 12 76 4 7 6 4 15 1 15 14 9 5 3 5 180 Totals 21 14 81 16 15 13 77 22 20 72 95 26 7 7 5 491 66 Each of these entities was obtained from another source in the Controller’s annual reports and added to the basic amounts included in Table 6 of the Annual Report of Financial Transactions Concerning Counties of California, Fiscal Year 1991–92. The redevelopment agencies, as was the case with those encountered in the city reconciliations, were obtained from the Annual Report of Financial Transactions Concerning Redevelopment Agencies of California, Fiscal Year 1991–92. The county enterprise activities were obtained from Tables 10 through 13 of the Annual Report of Financial Transactions Concerning Counties of California, Fiscal Year 1991–92. The county service areas were all obtained from the appropriate tables in the Annual Report of Financial Transactions Concerning Special Districts of California, Fiscal Year 1991–92. In total, 51 or 10.4 percent of the 491 redevelopment agencies in the state were reviewed in this process, and 483 or 26.5 percent of the state’s 1,856 dependent special districts were reviewed. While the precision of these reconciliations is less than that of the independent special districts described below, they all received a significant level of review. Because of the aggregation issue, no summary statistics can be provided for the values reported for these entities. Additional Reviews of Redevelopment Agencies In addition to reviewing 51 redevelopment agencies as part of the general review process, 10 redevelopment agencies were specifically reconciled against their audited financial statements. In each case, the amounts reported to the Controller coincided precisely with those found in their comprehensive audited financial reports. This can be largely attributed to the 90-day deadline for completing annual audits of 67 redevelopment agencies, which coincides with the filing deadline for the Controller’s questionnaire. Since these two deadlines are the same, the information provided is typically the audited information.21 Specific Findings: Independent Special Districts Independent special districts are special local government entities formed under a variety of sections of the state code to perform a wide range of functions. Individual independent special districts typically have narrowly defined areas of responsibility, such as providing transit, airport, hospital or water services. Others serve financial needs, such as providing insurance services or capital resources, and yet another category of these districts allows the joint provision of services among multiple cities and counties, such as library districts. Within this framework, the process of reviewing the data was complex. As was the case with cities and counties, audited financial statements were obtained from the largest special districts within each functional category as well as a random selection from the remaining districts. The specific criteria and a list of the entities sampled are included in Appendix C. The revenues reported in these audited financial statements were compared with those reported in the Controller’s report. The specific methodologies and criteria used during the review and reconciliation process are included in Appendix D. The findings of these reconciliations are presented in the following portion of this chapter. It is organized by the functional types of the ____________ 21In the case of counties, cities and special districts, the Controller’s deadline remains the same, while the audit deadline is 12 months after the fiscal year end. It is subsequently less common for the information provided to the Controller to be based on the audited data. 68 special districts involved. Summaries will be provided for each of the following types of special districts included in the Controller’s Annual Report on Financial Transactions Concerning Special Districts: hospital districts, transit districts, transportation planning agencies, water and sanitation districts, other enterprise districts (airport, electrical utility and harbor port districts) and non-enterprise districts. An overview of the findings for special districts will then also be provided. As was the case above for counties and cities, a detailed table corresponding to each of the summary tables is provided in an appendix of this report (in this case Appendix G). This section will focus on overall variance and not include the comparison along categorical lines that was provided for the more complex general government entities above. Special districts are fiscally much simpler entities and typically have only two general revenue categories: operating revenues and non-operating revenues. Operating revenues are those revenues that the districts obtain as a direct result of the specific services they provide. For an airport this would include landing fees, while in a water district this may include connection fees, sales of water and even sales of sewer/sanitation services. This most closely corresponds to current service charges in the general government entities, and, in fact, when there are directly operated municipal equivalents to enterprise districts, these revenues are included in the current service charges section of their operating statements.22 Nonoperating revenues typically represent all other categories of revenues, ____________ 22In some non-enterprise special districts, there is a category explicitly called “charges for current services.” In these cases, these revenues are treated as operating revenues. 69 including tax and intergovernmental revenues.23 For example, in most cases, but not all, any property taxes received by the special district will be represented as a separate entry in this category. Differentiating between these two categories within this analysis does not provide any new insights to the findings of this research. As the following tables will show, the level of variation found within these special districts is very low, and if there is any variation in the operating revenue category it is almost always offset dollar-for-dollar by a complementary difference in the non-operating category. This points to a simple classification issue that is likely more due to the lack of detail included in the comprehensive annual financial report than any errors or inconsistencies in reporting to the Controller’s Office. Hospital Districts Hospital districts are local districts that usually govern and oversee a major hospital in a municipality, county or region. These districts are paralleled in the county context by the 26 dependent hospital enterprise activities that were included in Table 10 of the Annual Report on Financial Transactions Concerning Counties. Since these hospital entities are dependent on the county and were subsequently included in the dependent entity reconciliations discussed above, this section focuses exclusively on the 13 independent special districts included in the Annual Report on Financial Transactions Concerning Special Districts. The overview of the findings for hospital districts is provided in Table 4.8. As this table shows, the overall variation between the two reports for hospital reports was minimal. In fact, the largest individual ____________ 23This is “typical” because some entities also include narrow categories of tax revenues and intergovernmental transfers as operating revenues. 70 Table 4.8 Summary of Comparison of Reported Overall Hospital District Revenues for Study Sample, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Audited Financial reports Difference Percentage difference Overall Revenues 1,484,490,882 1,487,334,570 2,843,688 0.2% NOTE: See Table G.1 for underlying detail and additional references. variant on a percentage basis was Mt. Diablo Hospital District, with only a 1.8 percent difference. The total sample represented 63 percent of the total hospital district revenue in 1991–92. Although very few of the hospitals had reports to the Controller that exactly matched their audited financial statements, the difference was not large: On average, the audited revenues were 0.5 percent higher or lower than the revenues reported to the Controller. There did not seem to be any systematic differences between the Controller’s reports and the CAFRs except in one difference of categorization of net revenue: The Controller counts the “provision for bad debts” as negative income, whereas the CAFRs universally recognize it as an expense. Transit Districts Transit districts are local districts that govern and oversee bus and rail transit systems in a municipality, county or region. These districts are paralleled in the county context by the 41 dependent transit enterprise activities that were included in Table 12 of the Annual Report 71 on Financial Transactions Concerning Counties. Since these transit districts are dependent on the county and were subsequently included in the dependent entity reconciliations discussed above, this section focuses exclusively on the 13 independent special districts included in the Annual Report on Financial Transactions Concerning Special Districts. The sample for transit districts included 18 of the 54 transit districts in the Controller’s report. This sample was chosen first by selecting the 12 districts with the highest revenues. The remaining six districts were either part of the original Alameda sample,24 included in the CAFRs of related transportation entities or selected as part of the larger sample. The revenues of these 18 districts represent nearly 94 percent of all transit district revenues in the state. As Table 4.9 shows, the total revenues reported in the Controller’s report were extremely close to those given by the CAFRs. The aggregate results shown in Table 4.9 do hide some variation between the districts, some of whose CAFRs show higher and some lower Table 4.9 Summary of Comparison of Reported Overall Transit District Revenues for Study Sample, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Audited financial reports Difference Percentage difference Overall Revenues 1,696,997,635 1,699,606,416 2,608,781 0.2% NOTE: See Table G.2 for underlying detail and additional references. ____________ 24The initial phase of this study focused on all of the public entities in Alameda County. See Appendix C for a more detailed discussion. 72 numbers than the Controller’s report. Although 10 districts showed no difference between the Controller and audited financial statements, the districts’ audited numbers on average were 1.3 percent higher or lower than the Controller’s report. The largest difference was in the San Mateo Transit District, whose report to the Controller seems to have been 9 percent too low for two reasons: The district did not report a gain on the sale of assets, and the Controller’s questionnaire included net, rather than gross, interest figures. Transportation Planning Agencies The 81 entities included under the transportation planning agency banner are either transportation planning agencies, agencies that have a transit planning function or agencies that have the authority to expend locally raised sales taxes on transportation needs. As such, this category also includes regional organizations of governments, such as the Southern California Association of Governments, as well as county traffic authorities, county transportation commissions, county transportation authorities, service authorities for freeway emergencies and selected others. The 19 entities included in the sample account for 93 percent of all transportation planning agency revenues in the state. Unlike the other categories included in this section on special districts, this category of entities includes a portion of the range of activity reported in the Controller’s Annual Report on Financial Transactions Concerning Transportation Planning Agencies of California. The transportation planning agencies are included here, in large part, because of criticisms that the amounts reported in the transportation and transit-related reports should represent particular areas of concern. 73 The findings for the sample of entities in this category are presented in Table 4.10. As this table shows, the overall level of variation is low, totaling only 4.6 percent, but higher than other special districts. Much of this variation is attributable to a single entity within the sample, as discussed below. As noted above, transportation authorities or planning agencies provide planning oversight for the transit districts in a region. One main function of the transportation authorities is to distribute transportation planning dollars from the state and other local entities, such as the local 1/4 cent sales tax (the Local Transportation Fund, or LTF) and State Transit Assistance (STA) funds. In 1991–92 most planning agencies did not count these funds as their own revenue, because the funds were passed on entirely to the local transit authorities. Nevertheless, the Controller requires that the entities report the LTF and STA funds in Table 4.10 Summary of Comparison of Reported Overall Transportation Planning Agency Revenues for Study Sample, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Audited financial reports Difference Percentage difference Overall Revenues 2,992,542,356 2,853,440,046 139,102,310 4.6% NOTE: See Table G.3 for underlying detail and additional references. 74 their report of fiscal transactions.25 As a result, these pass-through funds were included in the Controller’s report, but there was no audit information to verify the accuracy of the amount of these funds. Similarly, the interest figures in the CAFRs were lower than in the Controller’s report because the CAFR figures did not include the interest on the LTF and STA money. If the entities’ estimates of the pass-through funds are accurate, the difference between the audited numbers and the Controller’s report are minimal—with one major exception. The San Bernardino Association of Governments (SANBAG), which had a 60 percent difference between the two amounts, reported 137 million dollars more to the Controller than showed up on its annual audit. The difference was largely attributable to the inclusion of revenue note and bond proceeds as revenues in the Controller’s report. Without SANBAG, the overall difference was only $2,226,417 or 0.1 percent. The Southern California Association of Governments also shows a large percentage variance (7.5 percent). This variance is due to the inclusion of internal service fund revenues in the numbers reported to the Controller, which should have been excluded.26 ____________ 25While it may seem like double counting to include the LTF and STA dollars here, it is necessary in order to show the complete picture of fiscal transactions in California. Since transportation agencies are the first agencies to actually receive these monies, and in light of the explicitly stated public policy goal of reporting these funds as revenues to these agencies, they should be reported as revenues at this point. Subsequent agencies that receive LTF and STA dollars as a result of their distribution by a transportation planning agency should report these revenues as intergovernmental transfers. 26Internal service funds are an accounting technique whereby administrative costs for various activities are billed out to various other departments. For example, a computer support department may bill the planning department for the installation of a computer system. The computer department would thus show revenue in its internal service fund and the planning department would have an expenditure. While useful for internal cost accounting purposes, the transaction reflects an internal accounting allocation mechanism rather than a real increase in the city’s overall revenues. As such, 75 Water and Sanitation Districts This category of special districts includes all water, sewer and sanitation districts. These entities provide water, water treatment and waste disposal services to customers. They have been combined here for convenience of presentation and are actually reported separately in Tables 22 and 23 of the Annual Report on Financial Transactions Concerning Special Districts for 1991–92. California’s 806 water, water utility, irrigation and sewer districts make up the plurality of special districts in California, and, in fact, the first independent special districts in the state were water districts. Most of these are independent.27 The sample included 30 water, sewer, irrigation and sanitation districts including the 10 largest such districts in the state. Revenues within this category of special districts are highly fragmented, but the sample includes more than 26 percent of all the reported revenues for these entities in the state. The summary of the findings for the sample in this category is presented in Table 4.11. More than half of the districts seem to have used audited numbers for their final report to the Controller. When the Controller’s report was not exactly the same as the audit, there was no clear reason for the discrepancy, although the difference was almost always concentrated in the “other” category. The one major exception to this was the Oro Loma Sanitary District, which reported nearly $8 million more in revenues to the Controller ____________________________________________________ these funds are and should systematically be excluded from the reported revenue amounts. 27Four water and sanitation districts are included as county enterprise activities in the Annual Report on Financial Transactions Concerning Counties. Numerous cities have dependent water and sanitation enterprises. 76 Table 4.11 Summary of Comparison of Reported Overall Water and Sanitation District Revenues for Study Sample, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Audited financial reports Difference Percentage difference Overall Revenues 1,443,508,721 1,430,320,834 13,187,887 0.9% NOTE: See Table G.4 for underlying detail and additional references. than showed up on its audited financial report. The difference is attributable to the inclusion of a balance in an equity account as revenue when only the change in that equity position should have been included.28 Without this district’s variance, overall variance would be $5,205,407 or 0.4 percent. Airport, Electric Utility and Harbor and Port Districts The entities included in this subsection are again aggregated to simplify the presentation of the findings. The district descriptions are quite self-explanatory in terms of the function of each district. Airport, electric utility and harbor and port districts are included in Tables 18, 19 and 20, respectively, of the Annual Report on Financial Transactions Concerning Special Districts for 1991–92. The overall summary findings for these three types of entities are given in Table 4.12. ____________ 28Oro Loma Sanitary District is a participant and equity holder in the East Bay Dischargers’ Authority. Its equity stake in the Authority is approximately $8.7 million, with an annual change (a loss) of $100,000. Instead of incorporating the year’s flow (the loss) in its report to the Controller, the district included the stock (the full value of its stake in the authority). 77 Table 4.12 Summary of Comparison of Reported Overall Airport, Electric Utility and Harbor and Port District Revenues for Study Sample, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Audited financial reports Difference Percentage difference Airport Electric Utility Harbor and Districts Districts Port Districts Combined 12,324,045 1,203,210,790 132,322,371 1,347,857,206 12,077,250 1,203,658,374 132,322,372 1,348,057,996 246,795 447,584 1 200,790 2.0% 0.0% 0.0% 0.0% NOTE: See Table G.5 for underlying detail and additional references. The differences for each district type are extremely small. All of the variation in the airport districts could be attributed to the Monterey Peninsula Airport District, which had an overall variation of 5.2 percent. No explanation was available from the district for this variance. The electric utility and harbor and port districts showed no variance. Non-Enterprise Districts Non-enterprise districts—the largest category of district—are those government entities that are not run on a for-profit basis. These entities typically receive the largest part of their funding from taxes and intergovernmental grants rather than through service charges. Examples of non-enterprise districts include self-insurance authorities, cemetery districts, flood control districts, parks and recreation and library districts. The sample of non-enterprise districts consisted of 64 districts and included entities of every functional type. The summary of the findings for non-enterprise districts is included in Table 4.13. Although more than half of the districts seem to have used audited numbers in filling out the final report to the Controller, the Controller’s 78 Table 4.13 Summary of Comparison of Reported Overall Non-Enterprise District Revenues for Study Sample, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Audited financial reports Difference Percentage difference Overall Revenues 595,653,711 597,529,625 1,875,914 0.3% NOTE: See Table G.6 for underlying detail and additional references. report was 1.2 percent lower overall than the total given by the audited financial statements. This total does hide some variation among the minority of districts whose report to the Controller did not match their audits. On average, the audits of the non-enterprise districts were 1.1 percent higher or lower than the Controller’s report, and the average difference was $107,563 on average revenues of more than $9 million. In most cases where there was variation, all or most of the difference was due to excluded funds. That is, while the districts are supposed to include all revenues no matter what purpose they are reserved for, some districts neglected to include revenues earmarked for debt service, capital projects, or other special projects in their report to the Controller even though the forms explicitly ask for all funds. These excluded funds accounted for 77 percent of the difference between the Controller’s reports and the audited financial statements. Special Districts: Overall Findings Overall, the accuracy of the special district data was quite remarkable. Table 4.14 summarizes the findings for special districts 79 Table 4.14 Summary of Comparison of Reported Overall Special District Revenues for Study Sample, 1991–92 (dollars unless otherwise indicated) Special District Group Controller’s Audited Financial Report Report Difference Hospital districts 1,484,490,882 Transit districts 1,696,997,635 Transportation planning agencies 2,992,542,356 Water and sanitation districts 1,443,508,721 Airport, electric utility and harbor and port districts 1,347,857,206 Non-enterprise districts 598,229,191 1,487,334,570 1,699,606,416 2,853,440,046 1,430,320,834 2,843,688 2,608,781 139,102,310 13,187,887 1,348,057,996 598,105,105 200,790 1,875,914 Overall Sample Totals 9,561,050,511 9,416,289,487 144,761,024 NOTE: See Tables 4.8 through 4.13 above. Percentage Difference 0.2% 0.2% 4.6% 0.9% 0.0% 0.3% 1.5% overall. As it shows, the overall variation identified in the 156 districts reviewed was only 1.5 percent. The revenues presented in the two series correspond closely. Most of the difference is accounted for by one entity in the transportation planning agencies group, the San Bernardino Association of Governments. If it is omitted from these findings, the total variance drops to $7,885,131 or 0.1 percent of the total. Based on these findings, the special district data in the Annual Report on Financial Transactions Concerning Special Districts and Annual Report on Financial Transactions Concerning Transportation Planning Agencies are very accurate—probably even more so than the amounts reported for cities and counties.29 ____________ 29As discussed in the individual sections, this is likely because of the use of audited information by many special districts to prepare their Controller’s questionnaires. 80 Preliminary Findings: Debt Data Are Much More Problematic As a quick and preliminary measure of the quality of the debt data included in the Controller’s reports, the study tracked one narrow category of debt activity—debt proceeds. The goal was to identify how well the debt information reported to the Controller tracked with the detailed information included in the CAFRs. These findings are presented in Table 4.15. As this table shows, the amount of variation between the two sources is considerable. A large proportion of the variance in the city sample is due to Los Angeles, which had a difference of $782 million—a problem that has already been identified by the state. The overall variance within the city sample drops to 21.2 percent if Los Angeles is excluded. Debt and debt proceeds were not the focus of this study, but it is important to note that this preliminary look at the debt numbers points to significant reporting differences, in complete contrast to the revenue reporting. As will be discussed in the final chapter, additional research is recommended to address this problem. Table 4.15 Overall Variance in Other Financing Revenues, 1991–92 (dollars unless otherwise indicated) Group Controller Audited Report Percentage Differencea Differenceb Cities sampled 1,854,534,605 1,263,528,344 591,006,261 31.9% Counties sampled 815,876,077 682,453,027 133,423,050 16.4% Sample total 2,670,410,682 1,945,981,371 724,429,311 27.1% aDifference is presented in absolute terms. Accordingly, the overall total listed does not equal the sum of the column, but the difference between the overall amounts reported. bPercentage difference is the difference as a percentage of the controller’s reported amounts. 81 5. Summary, Recommendations and Conclusions This chapter presents a summary of the findings of this study as well as the implications of those findings. These implications are approached from two different directions. First, a practical approach is taken to the data, and a series of recommendations is presented to address parts of each of the quality issues raised in Chapter 1—timeliness, comprehensiveness and accuracy. Second, a broader view of the implications of the findings is taken. This section will present specific implications of the findings regarding the quality of the data to policymakers and analysts interested in California state and local policy. The credit for the quality of the data must be shared between the Controller’s Office and the myriad of local governments who supply the Controller with accurate information. 83 Summary of Findings A general statement of this report’s findings is that the Controller’s data are very good. As one colleague commented, “I’m increasingly impressed with the quality of the Controller’s data.” This is not to say that the data are perfect—there are clearly areas where the consistency of the reporting can be significantly improved, as will be discussed in this report’s recommendations below. Comprehensiveness of the Data The data were found to be quite comprehensive. However, it is clear that community facility districts are typically not included. A survey of a sample of the CFDs in the state found that only between 20 and 35 percent of these entities are included in their parent entity’s report to the State Controller’s Office. The balance are not included. A census of the 233 CFDs in existence in 1991–92 found that the combined revenues of all of these entities totaled only $283.5 million. In a local government sector whose revenues totaled nearly $95 billion, this represents a trivial source of error—0.3 percent—even if all CFDs were excluded. Since some proportion of these districts is included in the reported information, the actual level of error introduced by the omission of CFDs is even lower. Note, however, that if one wishes to focus on these particular entities for a specific policy inquiry, the data are wholly inadequate. Beyond CFDs, an effort was made to identify other entities that may be missing from the Controller’s reports. After an exhaustive (more than 7,000 entities) comparison of local agency formation commission and California Debt Advisory Commission lists to the entities reported in the Controller’s reports, three non-CFD entities were found to be missing 84 from the Controller’s entity list.1 The revenues for these entities totaled approximately $0.5 million—an inconsequential level of difference. It can be concluded from this exercise that the Controller’s data are comprehensive—that they include the full range of public entities that generate revenues at the local level. At the same time, one major category of entity is generally missing from the revenue data, but its absence’s effect on the overall revenue picture is negligible. Accuracy of the Data Along the accuracy dimension, the data test well. When the amounts reported overall by individual entities were compared with the amounts reported by the State Board of Equalization for sales and property taxes, the variance was very low—ranging from 0.0 percent to 5.1 percent. On a county-by-county and city-by-city basis, the variation was also quite low. Furthermore, this average was heavily influenced by a few outliers. Without these outliers, the difference between the two sets of information essentially disappeared. In comparing the information reported to the Controller with audited financial information, the overall totals were even better. For counties the variance was 1.2 percent, for cities 1.4 percent and for special districts 1.5 percent. At the individual comparison level, the absolute average percentage differences for each were 1.4 percent, 3.9 percent,2 and 0.9 percent, respectively, showing that aggregation implicit in the overall totals reflects the overall trend in the data. In general, these ____________ 1The methodology used allows for the existence of additional non-reporting entities, but a review of the candidates indicates that their overall effect would also be inconsequential. 2This value excludes the three outliers, Irwindale, Westmorland and Sacramento, as specifically discussed in Chapter 4. 85 data are highly accurate in their portrayal of the revenues raised and received by these entities. In summary, this study has found the revenue data included in the Controller’s reports to be quite correct and complete. There were, however, areas where the process and specifics could be improved. The next section discusses some specific suggestions for improving the data to render them as useful, complete and correct as possible. Recommendations These recommendations arise as a result of this exhaustive review of the State Controller’s reports. These recommendations are generally organized around the areas where concerns were raised: timeliness, comprehensiveness and accuracy and provide specific suggestions about how the information provided could be strengthened along each of those specific dimensions. Timeliness As discussed in Chapter 1, this report did not provide an analysis of the timeliness of the data. The issue is key, however, to the utility of the data to state decisionmakers and policy analysts. For this reason, the following two suggestions are offered in hopes of making the availability of the information in the Controller’s report much more timely. • Institute Internet/Web-based submission. The Controller’s Office should move to a more direct submission technology for the information in the questionnaires. A direct World Wide Web–based submission form would go a long way toward this goal. While a few entities are now submitting their questionnaires on diskette, an on-line form could significantly enhance the availability and ease of electronic submission for 86 local governments. The general and increasingly widespread availability of access to the Web coupled with its easy information transfer capabilities make it an ideal medium for transferring information between local governments and the State Controller’s Office.3 Furthermore, such an approach could facilitate the implementation of a broader, more standardized format as proposed in the discussion under accuracy below. This approach would also remove the data entry and quality control of that data entry from the Controller’s Office, significantly accelerating the preparation of the data for review and freeing up valuable resources for the review stage. The data would also immediately and automatically be in a consistent format across entities in the same category. • Make data immediately available. Beyond getting the information to the Controller’s Office more quickly and in a more consistent format, the data could be made immediately available to policy analysts and decisionmakers—even before the data have been reviewed.4 This would make the bulk of the information available in an unreviewed format 90 days after the fiscal year end—a much more useful timetable than more than a year later. Granted there is the prospect that unreviewed and unaudited data will contain errors, but the timeliness of the information could more than offset the risks associated with ____________ 3Typically, one of the greatest concerns associated with using Web submission techniques is security. This is not an issue in this case because all of the reported data are already in the public domain. 4Unreviewed data provided in this format would be so marked, indicating to users and consumers that it had not been reviewed. As the information is reviewed by the Controller’s office, the notation could be changed to reflect that development. In such a strategy, it is recommended that large and complex entities, which may be more likely to significantly affect policy choices, should be reviewed first. 87 mistakes by specific entities.5 To provide a metric of the reliability of a specific entity’s unreviewed data, an additional field could be provided by the Controller’s Office for each entity that reports the magnitude and types of information that were corrected in the prior year’s report.6 Comprehensiveness The Controller’s data are quite comprehensive in their reporting of revenue activity in the state. There are two recommendations, however, that arose as a result of this analysis: • Provide specific instructions for Mello-Roos districts. Specific instructions and questions should be provided to assure the inclusion of community facility districts in the Controller’s reports. While some of these entities are already reported, the vast majority are not. It is critical, inasmuch as decisionmakers and analysts care about this financing entity, to have accurate and complete information regarding their activity. • Create a mechanism for identifying new entities. The Controller’s Office should also implement a watchdog-type mechanism for identifying new entities that would not normally be captured under the current reporting scheme. While this study did not find many at this time, the increased constraints ____________ 5Note that this study reviewed the published reports from the State Controller’s Office and subsequently included all of the corrections that that office identified in the course of their review of the data. Information on the quality of the data before this review was not available, and clearly the prevalence of a need for significant review and correction of the data should be a factor in the implementation of this recommendation. Note, however, that the immediate publication of information could serve as an incentive for local entities to provide the information more accurately as its unreviewed quality comes under greater scrutiny. 6Also note that it is recommended that the reported information be left on-line in subsequent years, preferably in a side-by-side format, so that changes over time can be reviewed easily. 88 and pressures on local governments will assure more creative behaviors in the future. The Controller should have an eye toward making sure that these entities report their activity as they come into existence. Accuracy The general content of the Controller’s reports was found to be highly accurate. However, there were several areas where specific recommendations for improvements can be made. These are detailed below. • Provide more specific instructions and follow-up regarding capital project funds, debt service funds and housing authorities. Several types of activity were not consistently reported by all entities to the Controller’s Office. Often the specific variation identified between the Controller’s report and the audited financial report could be tracked to the noninclusion of either capital projects funds, debt service funds or housing authorities, or some combination of the three. Even in the case of counties, where separate columns are provided for debt service and capital projects funds, these revenues were not always included. It is recommended that these instructions be further strengthened and a systematic effort be made to ensure that all are included. • Expand and clarify reporting of special assessment districts. One category of activity for which the inadequacy of reporting has come to the forefront in light of current events is special assessment districts. Researchers and analysts trying to assess the effects of Proposition 218 on the November 1996 ballot found the data not up to the task. The reporting on these increasingly (in recent years) popular revenue-generating 89 arrangements is spotty at best and does not seem to capture the full range of activity that occurs. Because of the lack of a third-party information source, this study did not directly quantify the revenues that should be classified and reported as special benefit assessments. Discussions with other analysts looking at special benefit assessments indicate that the reported revenues are very low. It would seem from this research, however, that these revenues are not missing; they are simply included as property taxes and not separated out as special benefit assessments. Specific instructions and procedures should be established to track these activities. • Expand the reporting of school district information. The Controller’s report on school districts in California is very cursory. It reports revenues and expenditures only at the most aggregated levels—providing revenue and expenditure detail by type and category only at the statewide level and district-specific revenues and expenditures only in total. Inasmuch as school districts represent one of the largest categories of local revenues and expenditures, the detailed information on these entities should be published and made available for public accountability, review and discussion. It is recommended, therefore, that the Controller provide in its annual report entitylevel detail for each district, county office and joint powers agency in the state. This would require an expanded provision of information by the California State Superintendent of Public Instruction. The Department of Education could provide this information in a format immediately available for paper or electronic publication. It would also be helpful to include the specific numbers used by the state for calculating the Proposition 98–required expenditures each year. Such a service would provide a common source of information for analysts and decisionmakers. 90 • Provide detailed fiscal information for community college districts. Community college districts are unique entities in the state. Even though they receive their funding largely under the Proposition 98 formula, they participate as part of the state’s postsecondary education sector. Unlike the other public members of the state’s postsecondary education sector, they are organized and governed locally. Inasmuch as this research envisions the Controller’s data as a centralized database for both informational and accountability purposes, the information on these largely local entities should be as available as the local elementary school district, city, county or mosquito abatement district. As a result, it is recommended that these districts’ financial activity be reported in the Controller’s data system. • Establish a consistent report format for all categories of entities. Currently, there is significant variety in the way that the diverse types of entities report their specific information. As discussed regarding school districts above, but true on a much broader basis, the requirements and reporting structures associated with individual entity type are almost always determined by the specific institutional and historical context of the reporting entity and less with a view toward comparability and public accountability and with even less of a view toward utility for decisionmakers. As a result, the various reports are very difficult to compare and use—both across government entities and sometimes even within the same report. In the Annual Report on Financial Transactions Concerning Special Districts, for example, transit districts use a format driven by their reporting requirements for the U.S. Department of Transportation, while hospital districts use the requirements specified in the (California) Hospital 91 Disclosure Act.7 Although there are some distinct advantages to making the reporting process as convenient as possible for reporting entities, some minor changes could be implemented to improve the usefulness and comparability of the data. To provide a specific example of how this lack of comparability affects reporting, there is variation in how different entity types handle the major category of property tax revenues. Some have a separate line indicating property tax revenues, while others include them under “other non-operating revenues.” There are also differences between the reporting format for enterprises that are considered part of the county general government—and are reported in the enterprise tables in the Controller’s annual report on counties—and those that are separate dependent and independent districts listed in the special districts report. Even the detailed categories between such general governments as cities and counties vary somewhat. For example, the sale of a fixed asset in a city is included under “other revenues.” In a county it would be included under “other financing sources.” The reporting format for school districts, because it comes from a different source, is completely different. It would be invaluable to have consistent reporting formats between entities for comparability purposes. It would also be useful to have the major tax revenue streams—such as sales and property taxes—explicitly identified for each entity. Such an approach would also provide a quick measure of the quality of the information provided. More generic information, such as addresses, telephone numbers, miscellaneous demographic information (or hyperlinks to associated databases in an Internet-type approach) and contact people, would also be useful additions to these databases. ____________ 7This act commences with Section 440 of the Health and Safety Code. 92 In conjunction with a universal report format, one may wish to expand the governance information for each dependent entity in the database. In most cases at the current time, the governance information is limited to the parent entity’s generic category (e.g., county, city, special district). It would be far more useful to include precisely who the parent entity is. One could include an identifying code or, in an Internet-type application, a hyperlink to the parent entity. This would make the information far more useful for policy analysts and decisionmakers who would like to perform the type of aggregation that had to be done manually for the purposes of this study. Overall Issues To Be Considered in Implementation Overall, these recommendations represent minor enhancements to the quality of the data currently available. There are issues to be considered in the implementation of any and all of these ideas, however. There are effects on the backward comparability of the data as well as changes in the associated workload for both the Controller’s Office and reporting entities. Any changes that affect the data reported in the Controller’s study will clearly affect their direct comparability to prior years. This happens every time improvements are made to data systems. The effects in this case, however, would likely be minimal, as the low variances identified in this report would suggest. Most of the recommendations center around properly classifying activities and providing consistent reporting formats between entities. The expanded inclusion of Mello-Roos districts could be included as a separate line item, minimizing the effect their inclusion would have on other revenue categories. The expanded and more consistent reporting 93 of capital project funds, debt service funds and housing authorities would mean that the activity reported would more closely correspond to what is actually happening. Since this type of variation is included in the comparisons presented in Chapter 4, it is clear that the overall effect would be minimal. These changes, especially the Web-related recommendations and the format standardization, would have implications for both the Controller’s Office and the reporting entities. On the Controller’s side, these changes would require some investment of time and resources in assessing both the appropriate reporting formats and developing the new information technologies to implement them. On the reporting entity’s side, each would have to find access to the Web and make the transition to completing an on-line form in a new format. Providing a cost estimate of either of these activities is beyond the scope of this report, but it is believed that the resulting improvements in the timelines and usefulness of the data would be worth the investment. The process of implementing the new format may well serve as the ideal mechanism for implementing the new and improved instructions as well. As entities review their reporting procedures to comply with the new format, they could also verify that they are including all of the types of activity desired by the State Controller’s Office. Conclusions: The Broader Implications of the Findings Beyond these specific recommendations, there are some broader conclusions and implications that arise as a result of this study’s findings. These findings have greater importance to the policy context in which this report was introduced. They are detailed below. 94 • Reliability. The revenue data are reliable for understanding the range of activity occurring in the local government sector. As a consequence, concerns about the data underlying the various studies on California’s public revenue burden can be dismissed. This then allows the decisionmaker and analyst to focus on the methodology and values involved in the policy argument instead of wondering whether the data are comprehensive and accurate enough to understand what is happening. The levels of variation identified in this study are well below the threshold change levels identified in the various published studies addressing the revenue burden issue, indicating that the changes they measure are real, subject to the constraints of their individual methodologies. • Usability. This study also indicates that the State Controller’s data are usable for further research into more detailed aspects of state and local governance. This was one of PPIC’s main concerns going into this study—if in fact the data were not of adequate quality, what, if any, corrections could be made to the data to make them usable. PPIC is committed to studying the implications of various governance and finance choices at the local level for local, state and federal policy and policy initiatives. It was and is critical to that line of research to have confidence in the quality of the local government finance data, and this study has found that such confidence is well placed in these data. • Areas for further work. This study has preliminarily identified one important area, beyond the revenue data, where the local finance data appear to have more significant problems. This is the area of debt reporting. Our preliminary review of reported debt proceeds indicated a relatively high level of variation between audited and reported numbers. To the extent that debt becomes an increasingly important part of the local finance and governance policy environment, this variation should be explored further. 95 Appendix A Detailed Sales Tax Comparisons This appendix contains the detailed results of the findings about sales taxes for counties and cities summarized in Tables 3.1 and 3.2. This detailed information and the statistical summaries of the differences in the entity-level data are provided for the reader’s reference. City Sales Tax Comparisons The detailed results presented in Table A.1 are summarized in Table 3.1. Note that the detail is presented in a semi-aggregated form. The specific comparisons for each city have been aggregated and reported by county in order to reduce the scale of the presentation. As an example, the entry in Table A.1 for Alameda County represents the sum of the sales tax revenues for all 14 cities in Alameda County. 97 Table A.1 Comparison of City Sales Tax Revenues, All Cities Aggregated by County, 1991–92 (dollars unless otherwise indicated) County Alameda Amador Butte Calaveras Colusa Contra Costa Del Norte El Dorado Fresno Glenn Humboldt Imperial Inyo Kern Kings Lake Lassen Los Angeles Madera Marin Mendocino Merced Modoc Mono Monterey Napa Nevada Orange Placer Plumas Riverside Sacramento San Benito San Bernardino Controller’s Reports 123,948,143 1,216,461 11,334,205 492,664 1,027,142 67,305,843 907,680 4,814,661 49,040,452 1,059,685 7,880,963 7,775,969 1,226,606 33,240,227 4,780,709 1,498,033 1,158,866 718,324,138 3,615,025 22,389,742 3,676,833 7,989,985 416,515 880,723 24,415,488 7,160,538 2,489,195 255,966,614 12,495,001 130,563 76,994,488 37,246,209 1,576,978 101,133,583 Board of Equalization 122,405,442 1,225,000 11,202,766 485,964 1,023,129 67,270,892 907,680 4,721,106 48,978,443 1,059,962 7,830,331 7,871,996 1,226,606 32,852,889 4,777,709 1,495,487 1,152,466 726,217,730 3,615,025 22,405,463 3,701,569 7,992,777 416,515 841,466 24,487,627 7,135,435 2,494,695 257,509,919 12,431,858 131,113 77,848,667 40,044,544 1,579,072 102,666,564 Difference 1,542,701 –8,539 131,439 6,700 4,013 34,951 0 93,555 62,009 –277 50,632 –96,027 0 387,338 3,000 2,546 6,400 –7,893,592 0 –15,721 –24,736 –2,792 0 39,257 –72,139 25,103 –5,500 –1,543,305 63,143 –550 –854,179 –2,798,335 –2,094 –1,532,981 Percentage Differencea 1.2% –0.7% 1.2% 1.4% 0.4% 0.1% 0.0% 1.9% 0.1% 0.0% 0.6% –1.2% 0.0% 1.2% 0.1% 0.2% 0.6% –1.1% 0.0% –0.1% –0.7% 0.0% 0.0% 4.5% –0.3% 0.4% –0.2% –0.6% 0.5% –0.4% –1.1% –7.5% –0.1% –1.5% 98 Table A.1—continued County San Diego San Francisco San Joaquin San Luis Obispo San Mateo Santa Barbara Santa Clara Santa Cruz Shasta Sierra Siskiyou Solano Sonoma Stanislaus Sutter Tehama Tulare Tuolumne Ventura Yolo Yuba Grand total Controller’s Reports 210,539,303 83,379,910 30,842,093 13,769,325 70,107,782 24,239,204 175,041,439 13,395,933 12,123,812 47,668 2,488,313 24,466,476 30,236,569 24,574,650 4,206,089 2,584,917 17,098,200 929,028 50,875,970 12,567,507 1,958,832 2,401,082,947 Board of Equalization 200,858,348 83,747,510 30,547,671 13,726,925 68,650,568 24,045,784 175,485,491 13,321,630 12,120,462 47,668 2,466,286 24,417,655 29,972,344 24,689,495 4,099,752 2,540,500 16,944,141 932,171 50,533,574 12,658,087 1,958,832 2,401,772,801 Difference 9,680,955 –367,600 294,422 42,400 1,457,214 193,420 –444,052 74,303 3,350 0 22,027 48,821 264,225 –114,845 106,337 44,417 154,059 –3,143 342,396 –90,580 0 –689,854 Percentage Differencea 4.6% –0.4% 1.0% 0.3% 2.1% 0.8% –0.3% 0.6% 0.0% 0.0% 0.9% 0.2% 0.9% –0.5% 2.5% 1.7% 0.9% –0.3% 0.7% –0.7% 0.0% 0.0% SOURCE: Controller’s data are from the Annual Report of Financial Transactions Concerning Cities, Fiscal Year 1991–92, Tables 2–7, pp. 4–481. The Board of Equalization data are from the State Board of Equalization’s Annual Report for the Year Ending June 30, 1992, Table 21A, pp. A-26 to A-29. NOTES: Alpine, Mariposa and Trinity Counties have no incorporated cities. aPercentage difference is the difference as a percentage of the Controller’s reported amounts. The summary statistics for these data, presented in Table A.2, also point to the closeness of the two datasets. The absolute average percentage difference between the two reported series is only 1.9 percent. 99 Table A.2 Summary Statistics for Differences in Reported City Sales Tax Revenues, All Cities, 1991–92 Description Average of absolute variance Largest underreporting Largest overreporting Proportion with < 3% difference Proportion with < 5% difference Dollars $89,543 $2,774,700 $10,401,694 Percentage 1.9% 46.1% 30.2% 82% 88% NOTE: See Table A.1 for underlying detail and additional references. County Sales Tax Comparisons The results presented in Table A.3 are the details for each county in the state. This information is summarized in Table 3.2 of this report. The summary statistics in Table A.4 show the average absolute percentage difference is only 3.4 percent. The average county received more than $11 million in sales taxes, and the average difference of $627,466 between the Controller’s and Board of Equalization reports accounts for 3.4 percent of this average. 100 Table A.3 Comparison of County Sales Tax Revenues, All Counties, 1991–92 (dollars unless otherwise indicated) County Alameda Alpine Amador Butte Calaveras Colusa Contra Costa Del Norte El Dorado Fresno Glenn Humboldt Imperial Inyo Kern Kings Lake Lassen Los Angeles Madera Marin Mariposa Mendocino Merced Modoc Mono Monterey Napa Nevada Orange Placer Plumas Riverside Sacramento San Benito San Bernardino San Diego San Joaquin Controller Reports 13,054,052 234,739 866,861 3,025,252 1,604,612 575,569 7,765,324 473,008 4,040,193 10,489,412 599,164 1,855,499 1,488,276 727,567 17,074,043 1,247,120 1,560,219 619,558 33,136,115 2,508,173 2,200,199 1,153,713 2,712,116 2,442,469 146,844 307,560 4,273,025 3,056,962 3,460,396 11,709,981 6,369,103 1,243,634 12,269,864 64,310,319 699,226 15,019,249 9,970,643 6,010,472 Board of Equalization 10,331,992 234,739 866,861 3,025,252 1,284,818 575,969 7,634,485 473,008 4,090,396 8,338,868 621,364 1,855,499 1,488,276 727,567 17,074,044 1,247,120 1,567,919 619,558 33,576,615 2,508,174 2,179,444 1,153,713 2,716,866 2,442,469 147,444 307,560 4,273,025 2,947,662 3,462,396 11,208,781 6,355,066 1,243,634 12,269,864 63,034,180 701,826 9,966,519 10,188,010 6,010,472 Variance 2,722,060 0 0 0 319,794 –400 130,839 0 –50,203 2,150,544 –22,200 0 0 0 –1 0 –7,700 0 –440,500 –1 20,755 0 –4,750 0 –600 0 0 109,300 –2,000 501,200 14,037 0 0 1,276,139 –2,600 5,052,730 –217,367 0 Percentage Variancea 20.9% 0.0% 0.0% 0.0% 19.9% –0.1% 1.7% 0.0% –1.2% 20.5% –3.7% 0.0% 0.0% 0.0% 0.0% 0.0% –0.5% 0.0% –1.3% 0.0% 0.9% 0.0% –0.2% 0.0% –0.4% 0.0% 0.0% 3.6% –0.1% 4.3% 0.2% 0.0% 0.0% 2.0% –0.4% 33.6% –2.2% 0.0% 101 Table A.3—continued County San Luis Obispo San Mateo Santa Barbara Santa Clara Santa Cruz Shasta Sierra Siskiyou Solano Sonoma Stanislaus Sutter Tehama Trinity Tulare Tuolumne Ventura Yolo Yuba Total Controller Reports 3,122,567 13,801,957 7,929,501 2,948,976 4,663,342 2,233,855 93,771 471,761 1,639,557 8,392,929 10,121,273 1,262,661 694,635 544,907 4,678,791 2,527,442 4,813,481 1,325,862 1,034,197 322,601,996 Board of Equalization 3,117,166 12,879,333 7,632,501 2,903,508 4,663,342 2,247,526 93,771 471,761 1,639,554 7,192,929 8,000,090 1,273,161 740,901 517,486 4,719,491 2,527,442 4,457,070 1,261,712 1,034,197 306,124,396 Variance 5,401 922,624 297,000 45,468 0 –13,671 0 0 3 1,200,000 2,121,183 –10,500 –46,266 27,421 –40,700 0 356,411 64,150 0 16,477,600 Percentage Variancea 0.2% 6.7% 3.7% 1.5% 0.0% –0.6% 0.0% 0.0% 0.0% 14.3% 21.0% –0.8% –6.7% 5.0% –0.9% 0.0% 7.4% 4.8% 0.0% 5.1% SOURCE: Controller’s data are from the Annual Report of Financial Transactions Concerning Counties, Fiscal Year 1991–92, Tables 6 and 9–13, pp. 14–32 and 100–137. The Board of Equalization data are from the State Board of Equalization’s Annual Report for the Year Ending June 30, 1992, Table 21A, pp. A-26 to A-29. aPercentage difference is the difference as a percentage of the Controller’s reported amounts. Table A.4 Summary Statistics for Differences in Reported County Sales Tax Revenues, All Counties, 1991–92 Description Average of absolute variance Largest underreporting Largest overreporting Proportion with < 3% difference Proportion with < 5% difference Dollars $627,466 $440,500 $5,052,730 Percentage 3.4% 6.7% 33.6% 74% 82% NOTE: See Table A.3 for underlying detail and additional references. 102 Appendix B Detailed Property Tax Comparisons This appendix contains the detailed results of the findings about property taxes for counties and cities summarized in Tables 3.3 and 3.4. This detailed information and the accompanying statistical summaries of the entity-level differences are provided for the reader’s reference. City Property Tax Comparisons The detailed results presented in Table B.1 are summarized in Table 3.3. Note that the detail is presented here in a semi-aggregated form. The specific comparisons for each city have been aggregated and reported by county in order to reduce the scale of the presentation. As an example, the entry in Table B.1 for Alameda County represents the sum of the property tax revenues for all 14 cities in Alameda County. 103 Table B.1 Comparison of City Property Tax Revenues, All Cities Aggregated by County, 1991–92 (dollars unless otherwise indicated) County Alameda Amador Butte Calaveras Colusa Contra Costa Del Norte El Dorado Fresno Glenn Humboldt Imperial Inyo Kern Kings Lake Lassen Los Angeles Madera Marin Mendocino Merced Modoc Mono Monterey Napa Nevada Orange Placer Plumas Riverside Sacramento San Benito San Bernardino San Diego San Francisco San Joaquin Controller 183,318,614 854,018 5,331,149 136,446 672,113 64,354,459 116,006 4,386,556 44,982,289 872,332 2,312,403 4,761,422 315,009 22,021,512 2,663,279 994,217 462,395 986,053,936 1,901,496 28,108,606 1,198,245 6,075,167 183,140 534,221 17,343,348 9,039,163 1,726,207 221,862,960 10,519,272 190,009 52,958,420 57,901,184 738,173 69,092,707 224,227,255 522,289,889 30,035,255 County AuditorController 182,609,682 878,759 5,279,639 132,330 638,654 65,869,851 118,199 4,318,961 44,488,494 899,176 2,193,000 4,661,794 296,976 21,612,090 2,760,219 978,642 437,211 915,967,524 1,770,172 26,371,367 1,165,921 6,081,295 203,047 538,127 17,285,664 8,728,000 1,643,236 218,185,979 10,024,605 137,139 51,674,136 56,434,993 779,469 65,128,268 217,996,342 511,334,716 29,524,769 Variance 708,932 –24,741 51,510 4,116 33,459 –1,515,392 –2,193 67,595 493,795 –26,844 119,403 99,628 18,033 409,422 –96,940 15,575 25,184 70,086,412 131,324 1,737,239 32,324 –6,128 –19,907 –3,906 57,684 311,163 82,971 3,676,981 494,667 52,870 1,284,284 1,466,191 –41,296 3,964,439 6,230,913 10,955,173 510,486 Percentage Variancea 0.4% –2.9% 1.0% 3.0% 5.0% –2.4% –1.9% 1.5% 1.1% –3.1% 5.2% 2.1% 5.7% 1.9% –3.6% 1.6% 5.4% 7.1% 6.9% 6.2% 2.7% –0.1% –10.9% –0.7% 0.3% 3.4% 4.8% 1.7% 4.7% 27.8% 2.4% 2.5% –5.6% 5.7% 2.8% 2.1% 1.7% 104 Table B.1—continued County San Luis Obispo San Mateo Santa Barbara Santa Clara Santa Cruz Shasta Sierra Siskiyou Solano Sonoma Stanislaus Sutter Tehama Tulare Tuolumne Ventura Yolo Yuba Grand total Controller 14,026,768 69,276,568 15,385,732 127,334,980 11,033,897 5,762,481 25,280 1,373,680 30,526,994 21,590,280 14,280,490 2,982,701 1,185,415 8,605,449 335,386 37,555,299 16,276,240 1,087,454 2,959,177,966 County Auditor- Controller Variance 13,796,855 229,913 66,949,327 2,327,241 13,897,870 1,487,862 123,859,452 3,475,528 9,480,608 1,553,289 5,404,143 358,338 24,392 888 1,449,634 –75,954 29,750,803 776,191 20,845,006 745,274 13,812,462 468,028 3,010,322 –27,621 1,265,706 –80,291 8,008,687 596,762 322,274 13,112 37,401,022 154,277 15,338,012 938,228 1,045,322 42,132 2,844,810,343 114,367,623 Percentage Variancea 1.6% 3.4% 9.7% 2.7% 14.1% 6.2% 3.5% –5.5% 2.5% 3.5% 3.3% –0.9% –6.8% 6.9% 3.9% 0.4% 5.8% 3.9% 3.9% SOURCE: Controller’s data are from the Annual Report of Financial Transactions Concerning Cities, Fiscal Year 1991–92, Tables 2–7, pp. 4–481. The county auditorcontroller data are from the original surveys submitted to the Controller’s Office Annual Report on Property Taxes. These surveys are used to prepare the summary tables presented in State Board of Equalization’s Annual Report for the Year Ending June 30, 1992, Tables 4–15, pp. A-4 to A-19. It is also worth noting that these amounts are very close to those reported by the Board of Equalization but that, because of the specific criteria reported in each category, these amounts are never explicitly reported in the Annual Report in a manner such that they correspond to those reported above for comparability. NOTES: Alpine, Mariposa and Trinity Counties have no incorporated cities. aPercentage difference is the difference as a percentage of the Controller’s reported amounts. Table B.2 contains the statistical summaries at the county-aggregated level for the differences in reported property taxes for cities. As a result of some difficulties with the source documents, it is not possible to provide these statistics for the fully disaggregated detailed data. 105 Table B.2 Summary Statistics for Differences in Reported City Property Tax Revenues, All Cities, 1991–92 Description Average of absolute variance Largest underreporting Largest overreporting Proportion with < 3% difference Proportion with < 5% difference Dollars $2,149,174 $1,515,392 $70,086,412 Percentage 4.2% 10.9% 27.8% 45% 69% NOTES: These comparisons are based on the county-level aggregations because it was not feasible to disaggregate the data included in the Board of Equalization reports. See Table B.1 for underlying detail and additional references. County Property Tax Comparisons The results presented in Table B.3 are the details for each county in the state. This information is summarized in Table 3.4 of this report. Table B.4 presents the summary statistics for these data at the detailed county level. These statistics further attest to the close agreement between the two data sets. 106 Table B.3 Comparison of County Property Tax Revenues, All Counties, 1991–92 (dollars unless otherwise indicated) County Alameda Alpine Amador Butte Calaveras Colusa Contra Costa Del Norte El Dorado Fresno Glenn Humboldt Imperial Inyo Kern Kings Lake Lassen Los Angeles Madera Marin Mariposa Mendocino Merced Modoc Mono Monterey Napa Nevada Orange Placer Plumas Riverside Sacramento San Benito San Bernardino San Diego San Joaquin Controller 244,631,962 1,302,100 8,426,311 16,976,687 6,260,557 4,853,537 153,612,873 2,136,772 28,233,034 87,750,635 4,011,468 17,196,633 18,077,125 8,261,674 136,644,112 14,576,330 10,876,670 3,662,935 2,221,118,329 13,032,443 54,822,482 2,716,103 16,199,034 26,816,199 2,244,483 6,633,710 50,734,910 24,181,607 14,677,843 326,392,260 41,678,575 4,423,707 191,197,422 182,197,235 4,528,019 190,760,765 346,181,240 90,229,793 County AuditorController Variance 239,690,240 1,233,386 7,997,242 16,813,050 6,203,279 4,762,462 149,602,417 2,153,954 26,912,738 80,532,082 3,984,571 16,403,295 17,015,635 8,022,820 138,367,835 14,183,427 10,846,596 3,562,188 2,064,141,394 12,419,819 54,354,253 2,790,041 15,068,978 26,050,773 2,004,100 5,928,187 48,770,233 23,218,833 14,299,544 329,176,489 40,451,075 4,753,792 173,240,222 179,094,318 4,536,123 183,641,238 341,951,387 87,429,755 4,941,722 68,714 429,069 163,637 57,278 91,075 4,010,456 –17,182 1,320,296 7,218,553 26,897 793,338 1,061,490 238,854 –1,723,723 392,903 30,074 100,747 156,976,935 612,624 468,229 –73,938 1,130,056 765,426 240,383 705,523 1,964,677 962,774 378,299 –2,784,229 1,227,500 –330,085 17,957,200 3,102,917 –8,104 7,119,527 4,229,853 2,800,038 Percentage Variancea 2.0% 5.3% 5.1% 1.0% 0.9% 1.9% 2.6% –0.8% 4.7% 8.2% 0.7% 4.6% 5.9% 2.9% –1.3% 2.7% 0.3% 2.8% 7.1% 4.7% 0.9% –2.7% 7.0% 2.9% 10.7% 10.6% 3.9% 4.0% 2.6% –0.9% 2.9% –7.5% 9.4% 1.7% –0.2% 3.7% 1.2% 3.1% 107 Table B.3—continued County San Luis Obispo San Mateo Santa Barbara Santa Clara Santa Cruz Shasta Sierra Siskiyou Solano Sonoma Stanislaus Sutter Tehama Trinity Tulare Tuolumne Ventura Yolo Yuba Grand total Controller 56,709,520 130,932,923 67,837,663 298,713,084 33,765,816 19,270,632 1,940,527 7,102,578 51,298,525 82,777,370 41,306,545 10,981,944 7,225,813 2,244,366 45,147,936 11,350,279 120,449,624 18,245,367 8,343,734 5,593,901,820 County Auditor- Controller Variance 56,347,036 362,484 126,798,154 4,134,769 67,675,213 162,450 292,380,335 6,332,749 33,072,656 693,160 18,562,664 707,968 1,926,673 13,854 6,446,262 656,316 50,854,789 443,736 81,462,994 1,314,376 39,268,442 2,038,103 10,777,681 204,263 6,998,041 227,772 2,185,865 58,501 41,247,236 3,900,700 9,860,309 1,489,970 120,971,069 –521,445 17,243,793 1,001,574 7,779,810 563,924 5,353,466,793 240,435,027 Percentage Variancea 0.6% 3.2% 0.2% 2.1% 2.1% 3.7% 0.7% 9.2% 0.9% 1.6% 4.9% 1.9% 3.2% 2.6% 8.6% 13.1% –0.4% 5.5% 6.8% 4.3% SOURCE: Controller’s data are from the Controller’s Annual Report of Financial Transactions Concerning Cities, Fiscal Year 1991–92, Tables 2–7, pp. 4–481. The county auditor-controller data are from the original surveys submitted to the Controller’s Office Annual Report of Property Taxes. These surveys are used to prepare the summary tables presented in State Board of Equalization’s Annual Report for the Year Ending June 30, 1992, Tables 4–15, pp. A-4 to A-19. aPercentage difference is the difference as a percentage of the Controller’s reported amounts. Table B.4 Summary Statistics for Differences in Reported County Property Tax Revenues, All Counties, 1991–92 Description Average of absolute variance Largest underreporting Largest overreporting Proportion with < 3% difference Proportion with < 5% difference Dollars $4,218,158 $2,784,229 $156,976,935 Percentage 3.2% 7.5% 13.1% 54% 74% NOTE: See Table B.3 for underlying detail and additional references. 108 Appendix C Selecting a Statewide Sample of Public Entities The sampling approach used in this study is key to the level of confidence one would have in its findings. In this appendix, the specifics of the sampling methodology of the study are presented. There were two approaches to sampling in this study. The initial concept was to select several counties in the state and to obtain detailed information from every entity within that county. After completing a study of a pilot county under this approach, the decision was made to change to a partially stratified, random sample of entities. Initially, the belief was that there were specific benefits to having the detail on a geographically and politically clustered group because there would be sufficient interplays and exchanges of resources between them, which would in turn facilitate understanding of the intergovernmental relationships and provide economies of scale. The first county selected 109 under these criteria was Alameda County. There were three reasons it was chosen: (1) its close proximity to PPIC, which would aid in information gathering; (2) its diversity in terms of the types and size of entities; and (3) its relatively moderate size and ordinal ranking when looking at the state across a wide range of demographic criteria. Several other counties were also identified for the next phase of research at this point. A pilot county, Alameda County, was chosen to test and verify the reasoning behind this methodology. The Initial Sample: Alameda County PPIC staff traveled to the county government and to each city within Alameda County to obtain CAFRs. A parallel telephone effort was instituted to obtain the CAFR for each special district that had activity within the county. Comprehensive audited financial statements were collected from the 20 independent districts, 11 joint powers agencies, and the three transportation planning agencies that had any activity in Alameda County, as well as all 14 cities and the county government itself. Information was not gathered separately on the redevelopment agencies or dependent special districts because the finances of such entities were included in the cities’ financial statements. The total number of special districts collected was slightly larger than either the Controller’s or Census of Governments’ list of special districts because this effort included entities that were administratively housed in other counties but had some activity in the county, while two other organizations were assigned nexus in Alameda County by virtue of the location of their district headquarters. These audited financial reports were then compared with the Controller’s reports using the methodology described in Appendix D. 110 After completing this study of Alameda County, the decision to use the county as the selection criterion was revisited. It was found that there were in fact no comparative benefits to using such a geographically clustered sampling approach. The level of shared information between entities and the aggregated level of detail in the reports did not allow the realization of some of the hoped for cross-comparisons of fiscal information. The Final Sample Consequently, the balance of the state sampling was stratified first by size, then through random selection. This was done by selecting the largest entities in a category (cities, counties, special districts). In general, this was done by listing the entities in a category in descending revenue order (as reported in the Controller’s reports). A cutoff point was then selected where the magnitude of revenues fell off. For example, the seven largest counties in the state were included in the sample. The largest, Los Angeles County, had more than $10 billion in revenues. The next largest had just under $2 billion in revenues.1 All seven had more than $1 billion in revenues. Alameda County was included because of its size and because it was the pilot county. The remaining seven counties were selected at random from the balance of the counties in the state. The strength of this methodology was that it allowed the study to make certain to address the bulk of the activity in the state while still allowing for the possibility that smaller entities may actually report the information either better (they have less activity and complexity) or ____________ 1For purposes of this analysis, the joint entity of the City and County of San Francisco is treated as a city. If treated as a county, it would actually be the next largest after Los Angeles County, with approximately $3 billion in revenues. 111 worse (they have fewer resources to commit to the provision of the information) than their large, urban cousins. Since dollars were the key dimension of this project, it was critical to make certain that the large entities in each category were included. A low percentage error in Los Angeles County with its more than $10 billion in revenues will have a much more significant effect on any research findings than Plumas County’s $27 million or San Benito County’s $31 million in revenues. However, to the extent that one wishes to make comparisons and understand the differential effects of policy on various local governments, it is just as important to know that the information included for Plumas and San Benito Counties is accurate. As a result, the sample of entities from across the state was composed of a complete census of the largest entities, as well as a random sample of the smaller counties, cities and special districts. The detailed samples are given in the subsections below. Counties Sampled As described above, the seven largest counties and Alameda County were included in the sample. The remaining counties were sampled at random from the rest of the counties in the state. The counties included in the sample were Alameda, Humboldt, Los Angeles, Orange, Placer, Plumas, Riverside, Sacramento, San Benito, San Bernardino, San Diego, San Luis Obispo, Santa Clara, Siskiyou and Tehama. Cities Sampled A similar approach was followed with cities. The sample initially included the 14 cities in Alameda County. With the expansion of the sample to a stratified statewide sample, the largest 15 cities (by population) were selected. An additional 29 cities were selected at 112 random from the rest for a total of 54 cities.2 These cities were Alameda, Albany, Anaheim, Bakersfield, Berkeley, Carlsbad, Chowchilla, Delano, Dublin, Emeryville, Eureka, Fontana, Foster City, Fremont, Fresno, Glendale, Hayward, Huntington Beach, Irwindale, La Habra, La Quinta, Lindsay, Livermore, Long Beach, Los Angeles, Manteca, Modesto, Morro Bay, Napa, Newark, Oakdale, Oakland, Parlier, Piedmont, Pleasanton, Redding, Redwood City, Riverside, Sacramento, San Bernardino, San Diego, San Francisco, San Jose, San Leandro, San Mateo, Santa Ana, Simi Valley, South Pasadena, Stockton, Union City, Villa Park, Westmorland, Woodside and Yucca Valley. The information from only two cities on this list was not received in time to be included in the analysis in this report, reducing the total sample to 52 cities.3 The sample, while oversampling Alameda County, was reviewed across several key dimensions to ensure a robust sample. These dimensions included region of the state, size, government type and age of the city. The sample contains a wide range of cities within each of these categories, and as such it contains a reasonable cross-section of the state’s cities. Special Districts Sampled The special district sample was drawn in a fashion similar to that for the cities and counties. First, data for all of the independent special districts in Alameda County were collected. These 34 special districts ____________ 2Note that there was some overlap between the Alameda County cities group and the largest cities group—three cities were in both samples. 3The two cities that were not received in time for inclusion were San Bernardino and Chowchilla. 113 included 20 independent districts, 11 joint power agencies, and three transportation planning agencies. In addition to these Alameda County entities, 141 more special districts were selected for review. These were selected first by district type and then, as with cities and counties above, by size and finally at random. The districts with the largest revenues in each category were chosen first. Because significant concerns had been voiced about transit districts and transportation planning authorities, these two categories were oversampled. After selecting the largest districts in each category, 45 more districts were sampled at random from the remaining districts. The response rate for this sample was very high. Of the 175 districts selected, information was obtained on 155. Of the 20 missing districts, 6 reported they were not in existence and 14 refused or were unable to provide audited financial information. These 14 entities were broadly distributed throughout the sample categories and also varied by size. As a result, these missing entities are not believed to affect research design. The final sample was composed of 155 special districts. Nineteen of the state’s transportation planning agencies and 18 transit districts were included, as well as 30 water/sewer/irrigation districts, 11 hospitals, 13 other enterprise districts (public utility districts, airports, and ports) and 64 non-enterprise districts (e.g., park and recreation districts, flood control, self-insurance). The specific districts are included by district category below. Hospital Districts. Antelope Valley Hospital District, Eden Township Hospital District, El Camino Hospital District, Kaweah Delta Hospital District, Mount Diablo Hospital District, Palomar Pomerado Hospital District, Salinas Valley Memorial Hospital District, Sequoia 114 Hospital District, Tri-City Hospital District, Valley Health System and Washington Township Hospital District. Transit Districts. Alameda-Contra Costa Transit District, Bay Area Rapid Transit District, Central Contra Costa Transit Authority, Golden Gate Bridge, Highway and Transportation District, Livermore/Amador Valley Transit Authority, Long Beach Public Transportation Company, Marin County Transit District, North County Transit District, Orange County Transit, Sacramento Regional Transit District, San Diego Transit Corporation, San Diego Trolley, Inc., San Mateo County Transit District, Santa Barbara Metropolitan Transit District, Santa Cruz Metropolitan Transit District, Southern California Rapid Transit District and Stockton Metropolitan Transit District. Transportation Planning Agencies. Alameda County Transportation Authority, Association of Bay Area Governments, Contra Costa Transportation Authority, Fresno County Transportation Authority, Imperial County Local Transportation Authority, Los Angeles County Transportation Commission, Madera County Transportation Authority, Metropolitan Transportation Commission, Orange County Transportation Authority, Riverside County Transportation Commission, Sacramento Area Council of Governments, Sacramento Transportation Authority, San Bernardino Association of Governments, San Diego Association of Governments, San Diego Metropolitan Transit Development Board, San Francisco County Transportation Authority, San Mateo County Transportation Authority, Santa Clara County Traffic Authority and Southern California Association of Governments. Water and Sanitation Districts. Alameda County Water District, Byron-Bethany Irrigation District, Central Contra Costa Sanitary 115 District, Central Marin Sanitation Agency, Costa Mesa Sanitary District, Dublin San Ramon Service District, East Bay Dischargers Authority, Eastern Municipal Water District, Fairfield-Suisun Sewer District, Garden Grove Sanitary District, Irvine Ranch Water District, Ivanhoe Public Utility District, Jurupa Community Services District, Kern County Water Agency, La Habra Heights County Water District, Leucadia County Water District, Livermore-Amador Valley Water Management Agency, Mesa Consolidated Water District, Metropolitan Water District of Southern California, Oro Loma Sanitary District, Sacramento Regional County Sanitation District, San Benito County Water District, San Diego County Water Authority, Shafter-Wasco Irrigation District, South Tahoe Public Utility District, Thermalito Irrigation District, Tri-Valley Wastewater Authority, Union Sanitary District, Ventura Regional Sanitation District and West Kern Water District. Airport, Electric Utility and Harbor and Port Districts. East Kern Airport District, Lassen Municipal Utility District, Monterey Peninsula Airport District, M-S-R Public Power Agency, Northern California Power Agency, Sacramento Municipal Utility District, Sacramento-Yolo Port District, San Diego Unified Port District, Santa Maria Public Airport District, Southern California Public Power Authority, Stockton Port District, Transmission Agency of Northern California and Truckee-Donner Public Utility District. Non-Enterprise Districts. Alameda County Mosquito Abatement District, Alameda County Resource Conservation District, American River Fire Protection District, Auburn Cemetery District, Auburn Recreation and Park District, Bay Area Air Quality Management District, Bay Area Housing Authority Risk Management Agency, Bay 116 Area Library and Information System, Brannan-Andrus Levee Maintenance District, Broadmoor Police Protection District, California Housing Authority Risk Management Agency, California Joint Powers Risk Management Authority, Chino Valley Independent Fire District, City of San Francisco Downtown Parking Corporation, City of San Francisco Uptown Parking Corporation, Clovis Memorial District, Conejo Recreation and Park District, Dry Creek Storm Water District, East Bay Regional Park District, East Bay Schools Insurance Group, Eastern Plumas Fire Protection District, East Side Mosquito Abatement District, Elk Grove Community Services District, Fairview Fire Protection District, Fresno Metropolitan Flood Control District, Gilsizer County Drainage District, Glenn County Mosquito Abatement District, Half Moon Bay Fire Protection District, Hayward Area Recreation and Park District, Inland Empire Schools Insurance Authority, Jamestown Cemetery District, Kings River Resource Conservation District, Livermore Area Recreation and Park District, Madera Cemetery District, Marin-Sonoma Mosquito Abatement District, Mariposa County Resource Conservation District, Midpeninsula Regional Open Space District, North Coast Schools Insurance District, North Coast Schools Medical Insurance Group, North County Cemetery District, Orange County Cemetery District, Orange County Vector Control District, Ortega Trail Recreation and Park District, Palos Verdes Library District, Panoche Drainage District, Reclamation District #999, Reclamation District #2025, Reclamation District #2091, Resource Conservation District of Santa Monica, Riverside-Corona Resources Conservation District, Riverside County Flood Control District, Rocklin Placer Library Authority, SacramentoYolo Mosquito and Vector Control District, San Ramon Fire Protection 117 District, Santa Barbara County Health Care District, Santa Maria Valley Water Conservation District, Schools Insurance Authority, Self-Insured School District of Kern, Self-Insured School District of Kern II, South Coast Air Quality Management District, Spreckels Memorial District, Surfside Colony Community Services District, Temecula Public Cemetery District and Tulare County Pest Control District. 118 Appendix D Financial Statement Reconciliation Methodology Reconciling the Controller’s reports to audited financial reports was a key portion of this analysis of the accuracy of the Controller’s reports, the results of which were presented in Chapter 4. This appendix describes the methodology used to reconcile the two sets of information. The series used in the reconciliations for the Controller’s reports were those in the published Annual Report on Financial Transactions Concerning . . . books. These are the final adjusted numbers reported by the Controller’s Office for the various entities. For the audited financial reports, amounts were captured from each entity’s comprehensive 119 audited financial report.1 The next section provides more detail on the CAFR. Some Background on the Comprehensive Audited Financial Report The CAFR reports the findings of an annual audit required of public governments in California. These reports are prepared by independent public accounting firms and follow a format prescribed by Generally Accepted Accounting Practices (GAAP).2 This format contains general instructions regarding the content and format of the audits to be performed and the specific content and format of the report. This section briefly highlights some of the issues that are relevant to the reconciliation process. First, the range of activity included in a given CAFR encompasses all the activity over which that entity’s governing board has control. For this reason, the activities of dependent entities, such as redevelopment agencies, whose entire governing board is typically the city council or the county board of supervisors, are included in the CAFR’s reported activity. Other types of dependent entities include libraries, community facility districts and housing authorities. Local-run enterprise activities, such as water, sewer and transit districts, are also included in the parent entity’s report, although they are listed separately in the report. Since the Controller reports many of these activities separately as dependent special districts in the Annual Report on Financial Transactions ____________ 1For the small number of entities that did not have CAFRs, any other audited numbers were collected or, when no audits were possible, the budget figures that showed the actual financial activity of the entity for the 1991–92 fiscal year were collected. 2These practices are established and enforced by a national professional board, the Financial Accounting Standards Board. 120 Concerning Special Districts, it is often necessary to add these amounts to the original entity’s amounts to assure an appropriate comparison. Second, in accordance with GAAP, the CAFR reports several types of funds within each entity. These fund categories reflect the different types of monies that a local government can have. One group of categories refers to the limitations placed on the use of funds. Funds in this category include general funds,3 special funds,4 capital projects funds5 and debt service funds.6 Others refer to the type of activity or enterprise that generates the revenues, such as enterprise7 and internal service funds.8 Another group identifies revenues from the perspective of ____________ 3General funds are the general “checking accounts” for local governments. These funds represent the unrestricted revenues by which the bulk of the activity of local government is funded. Most programs and activities over which the local government has full discretion are funded from these accounts. 4Special funds are funds legally or contractually obligated to fund specific activities within the entity, but that do not fall into the specific categories listed later in this paragraph. Some examples of these funds are cigarette taxes, which are earmarked for anti-smoking campaigns, and certain transit taxes, which are set aside for transit development and expansion. 5Capital projects funds are funds specifically earmarked to pay for the development and operation of capital projects within the entity. 6Debt service funds are those which include the revenues and expenditures for activities specifically earmarked to retire debt for which the entity is obligated. 7Enterprise funds contain the activity of wholly owned subsidiaries that generate revenues for the provision of specific public services within the local government entity. These are usually locally run utilities, hospitals, airports and transit activities. These funds are reported separately in part because the enterprise’s rates and charges are often set as a function of its cost of providing services. Providing an audited account of its revenues and expenditures facilitates the rate-setting process. 8Internal service funds include the activities of internal entity departments charged to other departments. For example, a city’s computer department may “charge” the cable enterprise for installing a new computer system. The city would reflect these charges as revenues to the internal service fund. These are tracked in part for internal accounting purposes and in part to make certain that these “revenues and expenditures,” which remain entirely within the city, are correctly identified and do not overrepresent the city’s overall revenues. They should not be included in the amounts reported to the Controller because they do not reflect revenues from outside the entity’s internal governmental structure. 121 the entity’s level of control over the funds—expendable9 and nonexpendable10 trust funds. Many independent special purpose districts also make a distinction between operating11 and non-operating funds.12 Each of these funds are reported across a range of revenue categories. Each category represents a way in which local governments can obtain revenues. These revenue categories include taxes, fines and forfeitures, permits, revenues from the use of money and property, current service charges and miscellaneous revenues. It is convenient that the Controller’s reports also use these same distinctions. The definitions of activity within each, however, are not exactly the same between the two reports. GAAP allows for much more variation in interpreting which revenue category best suits a specific activity.13 The Controller’s instructions are much more explicit. As such, there is a significant chance that there will be a different revenue category reported for a specific activity between the two reports. Finally, each CAFR typically includes notes that expand upon the information presented in the financial tables themselves. It is in the notes, for example, that the specific reporting entity is defined and that one ascertains which dependent districts are included in the accounting ____________ 9Expendable trust funds are monies left to the local government by a third party over which the entity has significant discretion in its use. 10Non-expendable trust funds are resources held and owned by a local government for a special purpose. These funds are constrained by contractual or legal obligations of the entity. The largest funds in this category are pension funds. Occasionally, CFD revenues are included here. 11Operating funds are those raised by a special purpose district that come about as a direct result of that entity’s primary activity or activities. 12Non-operating funds are those raised by a special purpose district that do not come about as a direct result of that entity’s primary activity or activities. 13This is due in part to the national character of these specifications. They must be broad and flexible enough to encompass a much larger range of activities than the California State Controller Office must track. 122 reconciliation. These notes also explain any activities that are unique to the entity and contain detailed descriptions about its outstanding debt and its pension funds and requirements. The Reconciliation Process The goal of the reconciliation process is to compare the information in the Controller’s report to the audited financial statements. The general philosophy used is to (a) create comparable entities, (b) compare the two sets of information by revenue category, and (c) seek explanations for any variation identified and to make corrections, if appropriate. There is also a desire to be consistent across all of the entities in the sample—one of the goals of this analysis is to identify how consistently the Controller’s instructions are interpreted and applied by the more than 6,500 local government entities in the state. The specific steps in the reconciliation are listed below: 1. A specific summary of the CAFR’s reported audited numbers was prepared. These summaries were listed by revenue category. The general, special, capital projects and debt service funds were always included. In addition, enterprise funds were also included. Internal service funds and non-expendable trust funds were always excluded. The inclusion of expendable trust funds was decided on a case-by-case basis.14 2. The Controller’s comparable entity was developed. First, the information in the published Controller’s report for the entity was ____________ 14The general criteria focused on whether the city had complete control of the revenues and from where the monies came. 123 aggregated into the same revenue categories as those reported in the CAFR.15 The entity list in the notes section of the CAFR was reviewed to see if there were any dependent entities that were included separately in the Controller’s reports. If any such entities were identified, their revenues were identified in the appropriate Controller’s report and added to those of the entity itself to create a comparable entity. 3. The Controller-based entity was then compared with the CAFR. The comparison was made on a revenue category-by-category basis. Variation between the two reports was identified both in each category and overall. 4. If the difference between the CAFR and the Controller’s numbers was greater than three percent, or in cases where all of the difference was concentrated in a single revenue category, an attempt was made to contact that entity for an explanation of the discrepancy. If the explanations provided required any corrections, these were addressed at this point. If satisfactory explanations could not be determined or if the resulting variation was still higher than the desired threshold listed above, the variance was left in the analysis. Therefore, any variance listed is a conservative estimate of the level of variation between the reports. It is possible that additional proper explanations are available for the variance reported—they were simply beyond the scope of the resources available to obtain them.16 ____________ 15The same general revenue categories were used for each entity type. 16Local governments were contacted at least four times before the unexplained variance was simply left in the analysis. 124 These reconciliations were performed by experienced accounting staff and reviewed for accuracy, completeness and consistency by the project leaders. The final results of these reconciliations were then compiled for this analysis. 125 Appendix E Detailed Comparison of Controller’s Reports to Audited Reports: Counties This appendix contains the detailed county-by-county comparisons of overall revenues summarized in Table 4.1. Table E.1 refers to each county in the sample and provides detailed revenues in each source as well as the absolute and percentage variance for each. Table E.2 presents a summary for county revenue differences. 127 Table E.1 Overall Variance in Revenues by County, PPIC Sample, 1991–92 (dollars unless otherwise indicated) County Controller’s Report CAFR Percentage Differencea Differenceb Alameda Humboldt Los Angeles Orange Placer Plumas Riverside Sacramento San Benito San Bernardino San Diego San Luis Obispo Santa Clara Siskiyou Tehama Sample total 1,336,376,878 107,295,843 10,637,724,062 1,851,222,881 160,121,836 27,222,179 1,277,362,193 1,266,267,170 31,233,474 1,491,348,401 1,925,142,944 215,138,708 1,794,561,975 49,274,479 50,505,271 22,220,798,294 1,335,938,000 109,735,034 10,804,509,000 1,894,906,668 160,416,275 27,403,031 1,295,303,000 1,263,491,000 31,888,493 1,493,286,000 1,982,789,000 221,376,757 1,774,369,907 50,823,204 50,881,555 22,497,116,924 438,878 2,439,191 166,784,938 43,683,787 294,439 180,852 17,940,807 2,776,170 655,019 1,937,599 57,646,056 6,238,049 20,192,068 1,548,725 376,284 276,318,630 0.0% 2.3% 1.6% 2.4% 0.2% 0.7% 1.4% 0.2% 2.1% 0.1% 3.0% 2.9% 1.1% 3.1% 0.7% 1.2% aDifference is presented in absolute terms. bPercentage difference is the difference as a percentage of the Controller’s reported amounts. Table E.2 Summary Statistics for Differences in Reported Overall County Revenues, PPIC Sample, 1991–92 Description Average of absolute variance Largest underreporting Largest overreporting Proportion with < 3% difference Proportion with < 5% difference Dollars $21,542,191 $166,784,938 $20,192,068 Percentage 1.5% 3.1% 1.1% 94% 100% NOTE: See Table E.1 for underlying detail and additional references. 128 Appendix F Detailed Comparison of Controller’s Reports to Audited Reports: Cities This appendix contains the detailed city-by-city comparisons of overall revenues summarized in Table 4.3. Table F.1 refers to each city in the sample and provides detailed revenues in each source as well as the absolute and percentage variance for each. Each city is presented individually, therefore the first line, which refers to Alameda, is for the City of Alameda, not the cities in the County of Alameda as is the case in Appendixes C and B. Table F.2 presents a summary of these results. 129 Table F.1 Overall Variance in Revenues by City, PPIC Sample, 1991–92 (dollars unless otherwise indicated) City Alameda Albany Anaheim Bakersfield Berkeley Carlsbad Delano Dublin Emeryville Eureka Fontana Foster City Fremont Fresno Glendale Hayward Huntington Beach Irwindale La Habra La Quinta Lindsay Livermore Long Beach Los Angeles Manteca Modesto Morro Bay Napa Newark Oakdale Oakland Parlier Piedmont Pleasanton Redding Redwood City Riverside Controller’s Report 96,025,964 10,227,244 511,914,234 117,843,042 142,505,892 74,442,368 15,070,563 16,087,477 19,208,296 24,755,649 85,408,186 42,603,768 90,531,341 274,671,462 258,066,930 87,878,928 149,271,364 117,362,428 29,478,535 20,476,844 5,465,245 60,307,576 827,583,052 6,170,652,329 26,748,613 103,283,396 9,802,143 45,929,676 21,288,827 7,926,170 556,730,768 2,618,047 6,281,386 48,865,069 113,207,966 69,505,589 402,358,554 CAFR 111,645,495 9,998,842 523,276,000 119,864,561 156,990,000 74,340,421 15,382,547 16,138,379 20,899,499 25,664,069 85,555,000 43,472,367 103,078,610 274,830,935 255,217,407 88,692,248 151,474,000 15,008,649 29,698,246 21,025,675 6,148,812 60,026,563 742,496,000 6,254,840,000 26,893,659 103,857,197 9,867,466 49,782,607 22,038,047 8,346,875 566,136,000 2,708,047 6,740,534 51,169,280 115,653,900 73,319,405 383,439,183 Differencea 15,619,531 228,402 11,361,766 2,021,519 14,484,108 101,947 311,984 50,902 1,691,203 908,420 146,814 868,599 12,547,269 159,473 2,849,523 813,320 Percentage Differenceb 16.3% 2.2% 2.2% 1.7% 10.2% 0.1% 2.1% 0.3% 8.8% 3.7% 0.2% 2.0% 13.9% 0.1% 1.1% 0.9% 2,202,636 102,353,779 219,711 548,831 683,567 281,013 85,087,052 84,187,671 145,046 573,801 65,323 3,852,931 749,220 420,705 9,405,232 90,000 459,148 2,304,211 2,445,934 3,813,816 18,919,371 1.5% 87.2% 0.7% 2.7% 12.5% 0.5% 10.3% 1.4% 0.5% 0.6% 0.7% 8.4% 3.5% 5.3% 1.7% 3.4% 7.3% 4.7% 2.2% 5.5% 4.7% 130 Table F.1—continued City Sacramento San Diego San Francisco San Jose San Leandro San Mateo Santa Ana Simi Valley South Pasadena Stockton Union City Villa Park Westmorland Woodside Yucca Valley Sample total Controller’s Report 346,860,999 1,106,748,531 2,953,515,334 659,063,058 61,997,779 66,574,216 197,294,635 61,128,118 13,304,341 142,378,224 23,249,959 2,038,084 767,903 2,447,101 2,540,090 16,302,293,293 CAFR 442,230,000 1,096,171,000 2,662,419,000 655,473,813 61,122,886 64,242,104 195,667,691 63,348,185 13,304,339 162,513,000 25,116,930 2,023,507 1,006,664 2,812,755 2,549,345 16,075,717,744 Differencea 95,369,001 10,577,531 291,096,334 3,589,245 874,893 2,332,112 1,626,944 2,220,067 2 20,134,776 1,866,971 14,577 238,761 365,654 9,255 226,575,549a Percentage Differenceb 27.5% 1.0% 9.9% 0.5% 1.4% 3.5% 0.8% 3.6% 0.0% 14.1% 8.0% 0.7% 31.1% 14.9% 0.4% 1.4% aDifference is presented in absolute terms. Accordingly, the overall total listed does not equal the sum of the column, but the difference between the overall amounts reported. bPercentage difference is the difference as a percentage of the Controller’s reported amounts. Table F.2 Summary Statistics for Differences in Reported Overall City Revenues, PPIC Sample, 1991–92 Description Average of absolute variance Largest underreporting Largest overreporting Proportion with < 3% difference Proportion with < 5% difference Dollars $15,640,190 $95,359,001 $291,096,334 Percentage 6.7% 31.1% 87.2% 58% 71% NOTE: See Table F.1 for underlying detail and additional references. 131 Appendix G Detailed Comparison of Controller’s Reports to Audited Reports: Special Districts This appendix provides tables detailing the findings reported in Tables 4.8 through 4.14, as well as some summary statistics of the differences at the entity level. It is organized to parallel the order of the tables included in the body of the report—hospital districts, transit districts, transportation planning agencies, water and sanitation districts, miscellaneous enterprise districts and, finally, non-enterprise districts. There is no detail provided for the overall special district summary provided in Table 4.14. This simply represents an accumulation of the values reported in each of the other special district tables. 133 Hospital Districts Table G.1 Overall Variance in Revenues for Hospital Districts, 1991–92 (dollars unless otherwise indicated) District Name Controller Antelope Valley 106,723,755 Eden Township 86,895,638 El Camino 177,653,644 Kaweah Delta 97,169,845 Mt. Diablo 132,163,864 Palomar Pomerado 211,514,468 Salinas Valley Memorial 116,332,768 Sequoia 151,752,099 Tri-City 148,898,251 Valley Health System 134,834,487 Washington Township 120,552,063 Sample total 1,484,490,882 Audited Report 106,799,025 85,982,000 180,572,140 97,235,574 134,564,000 211,514,468 116,332,768 151,019,000 148,351,861 134,569,734 120,394,000 1,487,334,570 Differencea 75,270 913,638 2,918,496 65,729 2,400,136 0 Percentage Differenceb 0.1% 1.1% 1.6% 0.1% 1.8% 0.0% 0 733,099 546,390 0.0% 0.5% 0.4% 264,753 0.2% 158,063 2,843,688 0.1% 0.2% aDifference is presented in absolute terms. Accordingly, the overall total listed does not equal the sum of the column, but the difference between the overall amounts reported. bPercentage difference is the difference as a percentage of the Controller’s reported amounts. 134 Transit Districts Table G.2 Overall Variance in Revenues for Transit Districts, 1991–92 (dollars unless otherwise indicated) District Name Controller Audited Report Percentage Differencea Differenceb Alameda-Contra Costa Transit District 143,891,257 143,990,000 98,743 0.1% BART 285,013,596 285,007,000 6,596 0.0% Central Contra Costa Transit Authority 14,756,837 14,757,000 163 0.0% Golden Gate Bridge, Highway and Transportation District 28,426,761 28,633,000 206,239 0.7% Livermore/Amador Valley Transit Authority 4,317,530 4,333,996 16,466 0.4% Long Beach Public Transportation Company 31,496,982 31,496,982 0 0.0% Marin Co. Transit District 1,621,438 1,596,551 24,887 1.5% North Co. Transit District 26,732,864 26,736,997 4,133 0.0% Orange Co. Transit 115,303,777 115,303,777 0 0.0% Sacramento Regional Transit District 49,730,820 49,730,821 1 0.0% San Diego Transit Corporation 54,696,544 54,696,545 1 0.0% San Diego Trolley, Inc. 19,451,926 19,451,926 0 0.0% San Mateo Co. Transit District 81,494,904 89,078,369 7,583,465 9.3% Santa Barbara Metropolitan Transit District 8,333,022 8,333,023 1 0.0% Santa Clara Co. Transit District 166,980,278 164,425,009 2,555,269 1.5% Santa Cruz Metropolitan Transit District 21,226,070 21,225,171 899 0.0% Southern California Rapid Transit District 632,754,288 630,989,000 1,765,288 0.3% Stockton Metropolitan Transit District 10,768,741 9,821,249 947,492 8.8% Sample total 1,696,997,635 1,699,606,416 2,608,781 0.2% aDifference is presented in absolute terms. Accordingly, the overall total listed does not equal the sum of the column, but the difference between the overall amounts reported. bPercentage difference is the difference as a percentage of the Controller’s reported amounts. 135 Transportation Planning Agencies Table G.3 Overall Variance in Revenues for Transportation Planning Agencies, 1991–92 (dollars unless otherwise indicated) District Name Alameda County Transportation Authority Association of Bay Area Governments Contra Costa Transportation Authority Fresno County Transportation Authority Imperial Co. Local Transportation Authority Los Angeles Co. Transportation Commission Madera Co. Transportation Authority Metropolitan Transportation Commission Orange Co. Transportation Authority Riverside Co. Transportation Commission Sacramento Area Council of Governments Sacramento Transportation Authority San Bernardino Association of Governments San Diego Association of Governments San Diego Metropolitan Transit Development Board San Francisco Co. Transportation Authority San Mateo Co. Transportation Authority Santa Clara Co. Traffic Authority Southern California Association of Governments Sample total Controller 76,845,000 5,897,594 53,063,000 38,251,629 5,178,213 1,314,435,023 3,062,472 202,866,248 228,268,952 80,236,093 39,595,972 50,377,539 227,375,321 226,446,352 215,022,840 47,012,962 40,950,805 125,023,000 12,633,341 2,992,542,356 Audited Report 76,845,000 5,936,311 53,063,000 37,929,779 5,227,609 1,315,707,384 3,062,473 199,036,029 235,623,782 80,235,824 39,277,620 50,377,539 90,499,428 224,010,537 211,837,676 47,012,000 40,950,805 125,121,000 11,686,250 2,853,440,046 Percentage Differencea Differenceb 0 38,717 0 0.0% 0.7% 0.0% 321,850 0.8% 49,396 1.0% 1,272,361 1 0.1% 0.0% 3,830,219 7,354,830 1.9% 3.2% 269 0.0% 318,352 0 0.8% 0.0% 136,875,893 60.2% 2,435,815 1.1% 3,185,164 1.5% 962 0.0% 0 0.0% 98,000 0.1% 947,091 139,102,310 7.5% 4.6% aDifference is presented in absolute terms. Accordingly, the overall total listed does not equal the sum of the column, but the difference between the overall amounts reported. bPercentage difference is the difference as a percentage of the Controller’s reported amounts. 136 Water and Sanitation Districts Table G.4 Overall Variance in Revenues for Water and Sanitation Districts, 1991–92 (dollars unless otherwise indicated) District Name Audited Controller Report Percentage Differencea Differenceb Alameda Co. Water District Byron-Bethany Irrigation District Central Contra Costa Sanitary District Central Marin Sanitation Agency Costa Mesa Sanitary District Dublin San Ramon Service District East Bay Dischargers Authority Eastern Municipal Water District Fairfield-Suisun Sewer District Garden Grove Sanitary District Irvine Ranch Water District Ivanhoe Public Utility District Jurupa Community Services District Kern Co. Water Agency La Habra Heights Co. Water District Leucadia Co. Water District Livermore-Amador Valley Water Mgt Agency Mesa Consolidated Water District Metropolitan Water District of So. Calif. Oro Loma Sanitary District Sacramento Regional Co. Sanitation District San Benito Co. Water District San Diego Co. Water Authority Shafter-Wasco Irrigation District South Tahoe Public Utility District Thermalito Irrigation District Tri-Valley Wastewater Authority Union Sanitary District Ventura Regional Sanitation District West Kern Water District 34,480,701 1,021,186 46,913,101 4,987,586 4,793,411 13,910,506 2,088,715 123,970,053 18,345,184 5,858,264 144,887,000 762,146 8,262,648 105,333,695 2,821,486 6,719,834 2,027,482 12,593,135 549,699,920 19,068,690 68,593,725 4,113,366 168,637,074 1,698,589 18,227,697 1,085,582 405,632 33,033,989 29,138,618 10,029,706 34,480,701 1,000,998 46,906,900 4,987,018 4,797,964 13,910,506 2,061,883 125,039,572 17,863,867 5,858,264 144,887,000 762,146 8,277,443 105,333,695 2,821,486 6,770,395 2,027,482 12,593,135 543,802,000 11,086,210 68,593,726 4,093,992 168,746,000 1,698,589 18,227,697 1,085,582 405,632 33,034,000 29,137,245 10,029,706 0 20,188 6,201 568 4,553 0 26,832 1,069,519 481,317 0 0 0 14,795 0 0 50,561 0 0 5,897,920 7,982,480 1 19,374 108,926 0 0 0 0 11 1,373 0 0.0% 2.0% 0.0% 0.0% 0.1% 0.0% 1.3% 0.9% 2.6% 0.0% 0.0% 0.0% 0.2% 0.0% 0.0% 0.8% 0.0% 0.0% 1.1% 41.9% 0.0% 0.5% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Sample total 1,443,508,721 1,430,320,834 13,187,887 0.9% aDifference is presented in absolute terms. Accordingly, the overall total listed does not equal the sum of the column, but the difference between the overall amounts reported. bPercentage difference is the difference as a percentage of the Controller’s reported amounts. 137 Airport, Electric Utility and Harbor and Port Districts Table G.5 Overall Variance in Revenues for Airport, Electric Utility and Harbor and Port Districts, 1991–92 (dollars unless otherwise indicated) District Name Controller East Kern Airport District 5,349,678 Lassen Municipal Utility District 14,336,889 Monterey Peninsula Airport District 4,722,181 M-S-R Public Power Agency 68,462,793 Northern California Power Agency 181,838,260 Sacramento Municipal Utility District 683,046,027 Sacramento-Yolo Port District 12,086,055 San Diego Unified Port District 109,183,738 Santa Maria Public Airport District 2,252,186 Southern California Public Power Authority 245,085,477 Stockton Port District 11,052,578 Transmission Agency of Northern California 46,723 Truckee-Donner Public Utility District 10,394,621 Sample total 1,347,857,206 Audited Report 5,349,678 14,336,892 4,475,385 70,625,000 180,258,000 682,982,000 12,086,055 109,183,738 2,252,187 245,085,000 11,052,579 46,723 10,324,759 1,348,057,996 Differencea 0 3 246,796 2,162,207 1,580,260 64,027 0 0 1 477 1 0 69,862 200,790 Percentage Differenceb 0.0% 0.0% 5.2% 3.2% 0.9% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.7% 0.0% aDifference is presented in absolute terms. Accordingly, the overall total listed does not equal the sum of the column, but the difference between the overall amounts reported. bPercentage difference is the difference as a percentage of the Controller’s reported amounts. 138 Non-Enterprise Districts Table G.6 Overall Variance in Revenues for Non-Enterprise Districts, 1991–92 (dollars unless otherwise indicated) District Name Controller Alameda Co. Mosquito Abatement District 1,046,595 Alameda Co. Resource Conservation District 66,912 American River Fire Protection District 17,329,094 Auburn Cemetery District 380,817 Auburn Recreation and Park District 2,261,380 Bay Area Air Quality Management District 26,707,906 Bay Area Housing Authority Risk Mgt Agency 2,228,398 Bay Area Library and Information System 770,476 Brannan-Andrus Levee Maintenance District 128,952 Broadmoor Police Protection District 750,275 Calif. Housing Authority Risk Mgt Agency 1,671,639 Calif. Joint Powers Risk Mgt. Authority 10,426,588 Chino Valley Independent Fire District 10,113,062 City of San Francisco Downtown Parking Co. 3,331,569 City of San Francisco Uptown Parking Co. 5,509,476 Clovis Memorial District 728,882 Conejo Recreation and Park District 10,818,803 Dry Creek Storm Water District 88,722 East Bay Regional Park District 52,037,070 East Bay Schools Insurance Group 1,398,876 Audited Report 1,069,772 64,638 17,334,006 380,817 2,185,931 28,864,381 2,228,398 770,476 125,553 750,275 1,648,684 10,426,588 10,113,062 3,331,569 5,509,477 728,881 10,818,803 89,822 52,645,762 1,398,876 Differencea Percentage Differenceb 23,177 2.2% 2,274 3.4% 4,912 0 0.0% 0.0% 75,449 3.5% 2,156,475 8.1% 0 0.0% 0 0.0% 3,399 2.6% 0 0.0% 22,955 1.4% 0 0.0% 0 0.0% 0 0.0% 1 0.0% 1 0.0% 0 1,100 608,692 0 0.0% 1.2% 1.2% 0.0% 139 Table G.6—continued District Name Eastern Plumas Fire Protection District Controller 46,762 Audited Report 46,762 Differencea Percentage Differenceb 0 0.0% 140 Appendix H Copy of the Controller’s Annual Questionnaire for Counties" ["_edit_lock"]=> string(12) "1540320463:1" } ["___content":protected]=> string(102) "

R 397MSR

" ["_permalink":protected]=> string(98) "https://www.ppic.org/publication/a-review-of-local-government-revenue-data-in-california/r_397msr/" ["_next":protected]=> array(0) { } ["_prev":protected]=> array(0) { } ["_css_class":protected]=> NULL ["id"]=> int(8094) ["ID"]=> int(8094) ["post_author"]=> string(1) "1" ["post_content"]=> string(0) "" ["post_date"]=> string(19) "2017-05-20 02:34:51" ["post_excerpt"]=> string(0) "" ["post_parent"]=> int(3189) ["post_status"]=> string(7) "inherit" ["post_title"]=> string(8) "R 397MSR" ["post_type"]=> string(10) "attachment" ["slug"]=> string(8) "r_397msr" ["__type":protected]=> NULL ["_wp_attached_file"]=> string(12) "R_397MSR.pdf" ["wpmf_size"]=> string(7) "2468822" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(247320) "A Review of Local Government Revenue Data in California Michael A. Shires Melissa Glenn Habers February 1997 Copyright © 1997 Public Policy Institute of California, San Francisco, CA. All rights reserved. PPIC permits short sections of text, not to exceed three paragraphs, to be quoted without written permission, provided that full attribution is given to the source and the above copyright notice is included. Foreword As we were shaping the Institute’s initial research agenda, we talked with many leaders throughout the state. We asked them which public policy issues were most in need of attention by PPIC. The area most frequently suggested for study was governance—how California’s political institutions and processes affect policy outcomes. Public finance was a matter of particular concern. Proposition 13 and a succession of other revenue-and-expenditure-limiting initiatives were viewed as creating distortions in decisionmaking and accountability. Phrases such as “loss of control of the resource allocation process” and “a maddening proliferation of special districts” occurred often in our conversations with public officials. Many people have strong views about the governing process in California; yet it is difficult to judge how well the process is working or to measure its performance. Gaining insight into these issues is critical in a state as diverse and rapidly changing as California. This is why PPIC has, from the start, made governance one of its central areas of research. iii To successfully evaluate the governing process, we must have a reliable set of statistics on local government finances. The California State Controller’s Office publishes local public finance data on an annual basis, and the U.S. Census of Governments publishes comparable data every five years. If the Institute and others are to tackle—with credibility—the question of governance in the coming years, we must be certain that these published data are accurate and comprehensive. There are those who question the quality of these data, suggesting that the special fees and charges emerging in the aftermath of Proposition 13 may not have been captured in the data, thus leading to a substantial undercount of revenues raised at the local level. These concerns led Michael A. Shires and Melissa Glenn Haber to undertake the study described in the following report, A Review of Local Government Revenue Data in California. The authors reviewed more than 7,000 local government entities to verify that they were included in the reports published by the state Controller’s Office and reached the satisfying conclusion that the state data captured virtually all of the public entities generating revenues at the local level. In addition, they found a high degree of correspondence between the data compiled by the local government entities and the summary statistics published by the state. This degree of correspondence should assure both analysts and others interested in local public finance in California that, regardless of the findings reached, the databases used in the research accurately reflect the pattern of revenue-raising strategies throughout the state. With the passage of Proposition 218 in the fall of 1996, which requires a referendum on any attempt by local governments iv to raise revenue, the debate surrounding revenue-raising strategies is likely to become more heated. It is our intention to help inform the discussion, and our first contribution is this report. David W. Lyon President and CEO Public Policy Institute of California v Summary Background For more than two decades, Californians have engaged in a great debate over the size, shape and role of their state and local governments. This debate has involved a wide range of questions: How much of the state’s income should be spent by the public sector? Which goods and services should be provided by the state government, which by local governments? What is the appropriate level of state and local taxation? And perhaps most important, can citizens trust their elected officials to reflect their views on these policy issues or should such matters be governed by a combination of constitutionally imposed rules, limits and referenda? When this debate heated up in the 1970s, there was general agreement that Californians provided more goods and services through their state and local governments than all but one or two other states. This was the era when California led the nation in per-pupil spending for vii elementary and secondary education; when the California Master Plan for Higher Education promised state-funded postsecondary education to all of its citizens; and when the state engaged in massive infrastructure investment in its roads and waterways. These and other activities required considerable resources, and thus California had one of the highest state and local tax burdens in the nation. Beginning with the passage of Proposition 13 in 1978, California’s voters have used the initiative system to limit the size of their governments. Proposition 13 placed stringent constraints on state and local governments’ ability to raise revenues locally. Other initiatives have significantly affected the ability of elected officials to control the amount and mix of publicly provided goods and services. In effect, California has engaged in a great experiment over the past two decades: It has led the nation in the use of direct democracy to limit representative government. This is the first of a series of reports in which the Public Policy Institute of California (PPIC) will evaluate the effects of this experiment. How effective have the rules imposed by the initiative process been? Has the size of California’s state and local governments been truly limited, or have elected officials found ways around the limits? What effects have the initiatives had on the amount and quality of public goods and services? Given the transfer of many federal government activities to state and local governments, how will these constraints affect the implementation of changes such as welfare reform? These are all vital questions in the California policy debate. However, several key studies addressing these questions have raised concerns about the quality of the data available for evaluating issues relating to local public revenues. In addition, exploratory discussions with policymakers and analysts throughout the state have revealed that viii many do not use the data available on local public finance or, if they do, they do so hesitantly. These potential and actual users of the data cite three concerns: (1) the data do not capture all public entities in the state; (2) they do not accurately reflect the fiscal activity of those they do capture; and (3) they are not produced in a timely manner. This research thoroughly examines the first two concerns, analyzing the data produced by the California State Controller’s Office. The Controller’s Office, which provides the most timely and comprehensive data available on local finances, fields an annual survey of financial activity that collects revenue, expenditure, and debt information on every public entity in its database. All public entities are required by law to complete the survey and return it to the Controller’s Office within 90 days of the end of the fiscal year. This report evaluates the comprehensiveness, accuracy, and timeliness of the revenue data. Comprehensiveness of the Data: Are All Appropriate Entities Included? Because the initiative system has increasingly constrained local governments, some argue that the governments have found creative ways to go about their business. One of the criticisms of the state’s data systems is that they are unable to track this creativity. It is also argued that the data systems cannot always identify the institutions that spring up as a result of this creativity, and hence the studies that use the revenue data underestimate the actual size of public sector activity. Findings on the Comprehensiveness of the Data Overall, the data were found to be quite comprehensive, with the primary exception being Mello-Roos or community facility districts ix (CFDs), which are typically not included in the database. Analysis of a sample of CFDs showed that as few as 20 percent of these entities are included in reports to the State Controller’s Office. A follow-up analysis of all 233 CFDs in existence in 1991–92 found that the total revenue of these entities was only $283.5 million. In a local-government sector with revenues totaling nearly $95 billion, this represents a trivial source of error—0.3 percent—even if all CFDs were excluded. Since some of these districts are included in the reported information, the actual level of error introduced by the omission of CFDs is even lower. If one wishes to focus on these particular entities for a specific policy inquiry, the data are obviously inadequate. An exhaustive comparison of more than 7,000 entities appearing on the lists of local agency formation commissions and the California Debt Advisory Commission with the entities appearing in the Controller’s report identified only three non-CFD entities as missing from the Controller’s survey. The revenues for these entities totaled approximately $0.5 million—an inconsequential sum. Recommendations for Strengthening the Comprehensiveness of the Data While it can be concluded from these analyses that the Controller’s data are comprehensive—that they include virtually the full range of public entities generating revenues at the local level—there are some ways in which the data can be improved. Include Mello-Roos Districts. Specific instructions and questions should be included in the annual survey to ensure that community facility districts are included in the Controller’s reports. x Create a Mechanism for Identifying New Entities. The Controller’s Office should implement a watchdog-type mechanism for identifying new entities that might not be captured under the current reporting scheme. Accuracy of the Data: Do They Reflect What Actually Happens? Another issue raised about the quality of the data is whether the reported information accurately represents the activity of the reporting entity. Since the questionnaires are completed by each entity independently, the potential for variation in the interpretation of the instructions across more than 7,000 entities is significant. Because the surveys are usually submitted before annual audits are completed, the use of unaudited information in preparing the survey represents another area of concern. Findings on the Accuracy of the Data The Controller’s data are quite accurate. They were verified against third-party sources of information in two ways: (1) by comparing the data reported in the surveys to specific revenue series, such as sales and property taxes, and (2) by comparing the data reported by the entities with their audited information. When the total amounts reported by the entities for specific revenue series were compared with the amounts reported by the county auditor-controllers and the State Board of Equalization for property and sales taxes, respectively, the variance was very low—ranging from 0.0 percent to 5.1 percent. On a county-bycounty and city-by-city basis, the variation was also quite low. Furthermore, these averages were heavily influenced by a few outliers. xi Without these outliers, the difference between the two sets of information essentially disappeared. When the information reported to the Controller was compared with audited financial information, the overall variance was even lower. For counties it totaled 1.2 percent, for cities 1.4 percent, and for special districts 1.5 percent. On an entity-by-entity basis, the average percentage differences for counties was 1.4 percent, for cities 3.9 percent, and for special districts 0.9 percent, showing that the low variance identified in aggregate reflected low levels of variation at the individual entity level as well. Recommendations for Strengthening the Accuracy of the Data Even though there is a high level of accuracy in the data, it can be improved in some areas, as suggested below. Provide More Specific Instructions and Follow-Up Regarding Capital Project Funds, Debt Service Funds and Housing Authorities. Several types of activity were not consistently reported by all entities to the Controller’s Office. These three areas represented the greatest proportion of the overall variance in the data, and improvements in these areas would significantly improve the general quality of the data. Expand and Clarify the Reporting of Special Assessment Districts. The reporting on these increasingly popular revenue generating arrangements is spotty at best and does not appear to capture the full range of activity that occurs. Researchers and analysts trying to assess the impact of Proposition 218, which appeared on the November 1996 ballot, found the data quite inadequate.1 ____________ 1Proposition 218 requires a referendum on any attempt by local governments to raise revenues, service charges, or special assessments. xii Expand the Reporting of School District Information. Because school districts represent one of the largest categories of local revenues and expenditures, and because detailed district-level information is not published elsewhere, entity-level detail should be provided to the Controller for each district, county office, and joint powers agency in the state. It would also be helpful to include the specific numbers used by the state in calculating the Proposition 98–required expenditures each year.2 Such a service would provide a common source of information for analysts and decisionmakers. Provide Detailed Fiscal Information for Community College Districts. Community college districts are unique entities in the state. Even though they receive their funding largely under the Proposition 98 formula, they are considered part of the state’s postsecondary education sector. However, unlike the other public members of this sector, community colleges are organized and governed locally. Data on the community colleges would be useful for both informational and accountability purposes. Establish a Consistent Reporting Format for All Categories of Entities. There is significant variation in the way various types of entities report their information. As discussed regarding school districts above, but true on a much broader basis, the requirements and reporting structures associated with particular types of entities are almost always determined by the institutional and historical context of the reporting entity, rather than by concern for comparability and public accountability or the utility of the data for decisionmakers. ____________ 2Proposition 98 creates a floor on state spending and (indirectly) on local spending for K–14 education. xiii As a result, the statistics are difficult to compare and use—both across government entities and sometimes even within the same report. In addition to consistent reporting formats, it would be useful if the major tax revenue streams—such as sales and property taxes—were explicitly identified for each entity. In conjunction with a universal reporting format, it would be useful if the governance information for each dependent entity in the database were expanded to include explicit detail about what the parent entity is. Policy analysts and decisionmakers could then easily perform the type of aggregation that had to be done manually for this research. Timeliness of the Data This study did not examine in great detail the timeliness of the data. The problem is self-evident. Because data from each type of entity are published separately, the data become available at different times, and the delay can be as long as two years. This is problematic for policymakers and analysts who have much shorter and demanding time horizons. Recommendations for Strengthening the Timeliness of the Data The study suggests two ways in which the information in the Controller’s report might more quickly reach the policy and research communities. Institute Internet/Web–Based Submission. The Controller’s Office should use a more direct submission technology. Widespread access to the Web, coupled with its easy information-transfer capabilities, make it an ideal medium for transferring information between local governments and the State Controller’s Office. xiv Make the Data Immediately Available. In addition to getting the information to the Controller’s Office more quickly and in a more consistent format, the data could be made available before they have been reviewed.3 This would make the bulk of the information available 90 days after the end of the fiscal year—a much more useful timetable than the current one in which data are not released for more than a year. Broader Implications of the Findings Beyond the specific recommendations discussed above, the study arrived at some broader conclusions and implications. Reliability The revenue data accurately portray the range of activity occurring in the local government sector. Thus, decisionmakers and analysts can focus on the methodology and substantive questions involved in a policy argument and not worry about shortcomings in the data. Usability PPIC’s primary motivation for undertaking this study was to determine whether the State Controller’s data were of adequate quality and, if not, what changes might be initiated to make the data more usable. While we have recommended a number of changes, the data as they currently exist are usable for further research into more detailed aspects of state and local governance. ____________ 3Unreviewed data would be identified as such, and since the information is reviewed by the Controller’s Office, the identifier could be changed accordingly. Large and complex entities, which are more likely to significantly affect policy choices, should be reviewed first. xv PPIC is committed to studying the implications of various governance and finance choices at the local level for local, state and federal policy choices. It is critical to this line of research that one have confidence in the quality of local government finance data. This study has found that confidence in the data is well-placed. xvi Tables 1.1. Local Government in California, 1991–92 .......... 2.1. California Community Facility District Revenues, 1991–92 ............................... 2.2. Results of Search for Missing Entities Using the LAFCO Lists .................................. 2.3. Results of Sample Search for Missing Entities Using the CDAC List .............................. 3.1. Overall Summary of Comparison of Reported City Sales Tax Revenues, 1991–92 .................. 3.2. Overall Summary of Comparison of Reported County Sales Tax Revenues, 1991–92 .................. 3.3. Overall Summary of Comparison of Reported City Property Tax Revenues, 1991–92 ................ 3.4. Overall Summary of Comparison of Reported County Property Tax Revenues, 1991–92 ................ 3.5. Comparison of School District Tax Revenues, 1991–92 ............................... 4.1. Summary of Comparison of Reported Overall County Revenues for Study Sample, 1991–92 ............. 7 24 27 28 34 35 37 38 39 50 xvii 4.2. Overall Variance in Revenues by Revenue Category for Counties, 1991–92 ......................... 4.3. Summary of Comparison of Reported Overall City Revenues for Study Sample, 1991–92 ............. 4.4. Overall Variance in Revenues by Revenue Category for Cities, 1991–92 ........................... 4.5. Overall Variance in Revenues by Revenue Category for Cities, Without San Francisco, 1991–92 ........... 4.6. Revenues for School Districts in California, 1991–92 ... 4.7. Dependent Entities Included in County Reconciliations............................ 4.8. Summary of Comparison of Reported Overall Hospital District Revenues for Study Sample, 1991–92 ............................... 4.9. Summary of Comparison of Reported Overall Transit District Revenues for Study Sample, 1991–92 ........ 4.10. Summary of Comparison of Reported Overall Transportation Planning Agency Revenues for Study Sample, 1991–92 .......................... 4.11. Summary of Comparison of Reported Overall Water and Sanitation District Revenues for Study Sample, 1991–92 ............................... 4.12. Summary of Comparison of Reported Overall Airport, Electric Utility and Harbor and Port District Revenues for Study Sample, 1991–92 ................... 4.13. Summary of Comparison of Reported Overall Non-Enterprise District Revenues for Study Sample, 1991–92 ............................... 4.14. Summary of Comparison of Reported Overall Special District Revenues for Study Sample, 1991–92 ........ 4.15. Overall Variance in Other Financing Revenues, 1991–92 ............................... A.1. Comparison of City Sales Tax Revenues, All Cities Aggregated by County, 1991–92 ................ 52 56 59 59 63 66 71 72 74 77 78 79 80 81 98 xviii A.2. Summary Statistics for Differences in Reported City Sales Tax Revenues, All Cities, 1991–92 ........... 100 A.3. Comparison of County Sales Tax Revenues, All Counties, 1991–92 ......................... 101 A.4. Summary Statistics for Differences in Reported County Sales Tax Revenues, All Counties, 1991–92 ......... 102 B.1. Comparison of City Property Tax Revenues, All Cities Aggregated by County, 1991–92 ................ 104 B.2. Summary Statistics for Differences in Reported City Property Tax Revenues, All Cities, 1991–92 ......... 106 B.3. Comparison of County Property Tax Revenues, All Counties, 1991–92 ........................ 107 B.4. Summary Statistics for Differences in Reported County Property Tax Revenues, All Counties, 1991–92 ....... 108 E.1. Overall Variance in Revenues by County, PPIC Sample, 1991–92 ............................... 128 E.2. Summary Statistics for Differences in Reported Overall County Revenues, PPIC Sample, 1991–92..... 128 F.1. Overall Variance in Revenues by City, PPIC Sample, 1991–92 ............................... 130 F.2. Summary Statistics for Differences in Reported Overall City Revenues, PPIC Sample, 1991–92 ............ 131 G.1. Overall Variance in Revenues for Hospital Districts, 1991–92 ............................... 134 G.2. Overall Variance in Revenues for Transit Districts, 1991–92 ............................... 135 G.3. Overall Variance in Revenues for Transportation Planning Agencies, 1991–92 ................... 136 G.4. Overall Variance in Revenues for Water and Sanitation Districts, 1991–92 ......................... 137 xix G.5. Overall Variance in Revenues for Airport, Electric Utility and Harbor and Port Districts, 1991–92 ........... 138 G.6. Overall Variance in Revenues for Non-Enterprise Districts, 1991–92 ......................... 139 xx Acknowledgments The authors gratefully acknowledge the assistance and cooperation of a wide range of associates and colleagues who contributed to this report. Given the considerable volume of detail work and the complexity of the information systems and sources addressed, special thanks must be extended to the many individuals who made this analysis possible. Several people were invaluable in the conceptualization and development of this study. PPIC colleagues John Ellwood, Elisabeth Gerber and David Lyon provided insight and guidance during the development of the proposal for this study. Rudy Nothenberg and Randy Hamilton were instrumental during the conceptual phase in providing a local government perspective on the problem, as were Professors John Kirlin and Jeff Chapman of the University of Southern California, who also served as technical reviewers on this report. This project also benefited from the input and perspectives of several committee staff members of the California Legislature, and PPIC is indebted to Peter Detweiler for arranging their participation. xxi This report represents the synthesis of information from numerous sources within state and local government. As a result, it was highly dependent on the assistance and cooperation of staff in many state offices and local governments. The California State Controller’s staff were key to the preparation of a significant part of this report. They always made themselves available to answer questions about the data and their procedures. Special thanks go to Alice Fong and Wayne Beck, who fielded the brunt of the project’s inquiries. Jeff Reynolds of the State Board of Equalization was equally helpful, answering questions and providing valuable supporting documents for the county auditorcontroller’s data. Peter Schaafsma and the staff of the California Debt Advisory Commission provided assistance and data key to major parts of this study. The staff at the United States Bureau of the Census in Washington, D.C., have also been extremely helpful in the pursuit and analysis of the data in this study. Special mention goes to William Hulcher, who fielded numerous inquiries and hosted and arranged an informational visit to the Bureau’s offices. Another critical group of government participants were the numerous city, county and district managers and staff who provided copies of the audited financial statements, answered questions about them and sometimes went to great lengths to explain the subtleties of their particular entity’s finances to our project staff. Without the assistance and cooperation of these hundreds of individuals, this study would not have been possible. The largest volume of work in this project was handled by an extremely competent project team which deserves recognition here. Liz Britton and Charles Castel were invaluable in obtaining information and xxii providing follow-up, while Mehtab Bains, Jason Cohen, Mary Joy Espino, Pradeep Mathew and James McMillan performed the often exhausting financial reconciliations. This report has benefited from technical peer review by several colleagues and associates, including Kim Rueben, Fred Silva and John Ellwood of PPIC, John Kirlin and Jeff Chapman of USC, Marianne O’Malley of the Legislative Analyst’s Office, and Steven Frates and Eric Norby of Claremont McKenna College. Their comments, collectively and individually, have contributed to the quality of this report. Finally, and certainly not least, Mary Sprague deserves special credit and thanks for her efforts on this project. She stepped in half-way through this project and has done an excellent job of helping to keep it on track. She has contributed to this project in numerous ways, including coordinating staff, following up on the “one hundred and one” open items needing closure, tracking down that “one piece” of needed data and keeping the special districts portion of the data set straight. Without her assistance, this project may well never have ended. While all of these individuals and organizations have been instrumental in the planning, preparation and execution of this study, the research and views expressed in this report are entirely the authors’. They should in no way be construed as representing the views of anyone acknowledged herein or anyone else. xxiii 1. Introduction For two decades Californians have been engaged in a great debate over the proper size, shape and role of their state and local governments and whether representative government should or is capable of constraining the growth of the public sector. This debate has centered around such questions as: How much of the state’s income should be spent by the public sector? Which goods and services should be provided by the state government? By local governments? What is the appropriate level of state and local taxation? And perhaps, most important, can citizens trust their elected officials to reflect their views on these policy questions or should these issues be governed by a combination of constitutionally imposed rules, limits and referenda? When this debate heated up in the 1970s, there was general agreement that Californians chose to provide more goods and services through their state and local governments than all but one or two other states. This was the era when California led the nation in per-pupil spending for elementary and secondary education, when the California 1 Master Plan for Higher Education promised highly subsidized statefunded postsecondary education for all its citizens, and when the state engaged in massive infrastructure investment in areas ranging from its road system to the dams and aqueducts that shift water from the north to its population-rich south. These and other activities required considerable resources, and as a result from the close of World War II until the late 1970s the state had one of the highest state and local tax burdens in the nation. Beginning with the passage of Proposition 13 in 1978, California’s voters used the initiative system to place limits on the size of its governments. Proposition 13 placed stringent constraints on local government’s ability to determine and raise revenues locally. Other initiatives—such as Proposition 4 (the Gann Initiative), which placed limits on the overall levels of spending for all state and local governments; Proposition 98, which created a floor on state spending and (indirectly) local spending on K through 14 education; and Proposition 111, which modified the rules of Proposition 98 for those years in which state revenues were depressed by a recession—significantly affected the ability of state and especially local government elected officials to control the amount and mix of publicly provided goods and services. When elected officials turned to governmental institutions and mechanisms not constrained by these initiatives—such as fees, service charges and special assessments—to provide additional goods and services, the advocates of constraining representative government sought to pass new initiatives to control and limit this activity. Most recently, voters approved Proposition 218, which requires a referendum on any attempt of local governments to raise revenues, service charges or special assessments. 2 On a policy level, these initiatives reflect a desire to control and limit the size of California’s governments. But at a deeper level, they reflect a basic distrust of representative government: a belief that elected officials are, in practice, incapable of imposing the public’s will when it comes to taxation and spending. In effect, over the past two decades, California has been engaged in a great experiment. It has led the nation in the use of direct democracy to limit representative government. This is the first of a series of reports in which the Public Policy Institute of California (PPIC) will evaluate the effects of this experiment. How effective have the rules imposed by the initiative process been? Has the size of California’s state and local governments been truly limited or have elected officials found ways around the limits? What effects have the initiatives had on the amount and quality of public goods and services in areas ranging from elementary and secondary education to health services to welfare to corrections? Given the devolution of many federal government activities to the state and its local governments, how will these constraints affect the implementation of such changes as welfare reform? These are all vital questions for the California policy debate and policymaking process. As discussed in the next section, several studies of issues in this area have pointed to deficiencies in the data, describing important aspects of the local public revenue burden. Before the key policy questions can be addressed, therefore, it is critical to clear up any concerns about the quality of the available data. This report does just that, laying the groundwork for the future work of the Public Policy Institute of California by assessing the quality of the available fiscal data regarding California’s local governments. 3 The Quality of the Data Is Important In the wake of Proposition 13, several studies have been undertaken to assess its full effect on the level of local government expenditures. Several of these studies have found some level of reduced spending in the state and local levels of government. Studies by the California Taxpayers’ Association (Cal-Tax)1 and Steven Gold2 have found that local government revenues dropped from 3 to 20 percent between fiscal years 1977–78 and 1991–92. Others have found that local revenues have in fact risen since Proposition 13. For example, Gary Galles and James Long found that state and local real per-capita general revenues in fiscal 1989–90 were 6 percent higher than in 1977–78,3 while John Kirlin et al.4 found a small percentage increase in overall public revenues. Part of these differences arises from the use of different measures (revenues as a percentage of personal income versus real per- ____________ 1From California Taxpayers’ Association, “California Taxing and Spending,” CalTax Research, October 1994, pp. 3–9. Cal-Tax found that state and local own-source revenues declined from 16.7 percent of state personal income to 16.2 percent—a decrease of 3.0 percent. 2From Steven D. Gold, “California’s Budget from a National Perspective,” testimony presented at a hearing on Proposition 165, California Budget and Fiscal Review Committee, California Health and Human Services Committee, California Senate, October 1, 1992. He found that state and local tax revenues declined from 14.6 to 11.4 percent of state personal income—a decrease of 21.9 percent. This measure is a little different than that of the Cal-Tax and Kirlin studies because it looks only at tax revenues. The other two studies focus on a broader measure of local revenues that includes current service charges, fines and the sale of fixed assets. 3From Gary M. Galles and James E. Long, Proposition 13 and California Tax and Expenditure Trends, Sacramento, Calif.: The Howard Jarvis Taxpayers Foundation, May 1994, Executive Summary. 4From John J. Kirlin et al., “Fiscal Reform in California,” in California Fiscal Reform: A Plan for Action, Oakland, Calif.: California Business-Higher Education Forum, June 1994. Kirlin’s group found that state and local own-source revenues increased from 17.4 percent of state personal income to 18.1 percent—an increase of 4.0 percent. 4 capita revenues) and part arises from the use of different comparison years—1989–90 was near the peak of an economic cycle, while 1991–92 was the first year of a severe recession. Another point of difference between the studies was the kinds of public revenues included. Certain studies included quasi-public transactions like tuition and fee revenues from the California State University system, while others did not. The purpose of this report is not to address authoritatively the differences found in these works, but to address another important issue that has been raised in response to these works. In nearly every case, practitioners and analysts used either the U.S. Department of Commerce’s Census of Governments or the California State Controller’s series of Annual Reports of Financial Transactions. In the text of the reports and in subsequent one-on-one interviews with the authors, the issue has been raised regarding the quality of the data included in the reports. In the Western Political Quarterly, there was recently an exchange of articles debating the methodology and quality of the U.S. Census Bureau’s data.5 While no one proposes bounds for the quality of the data, significant concerns are raised. The quality of the data is clearly an important issue in this debate. For example, if the picture of local and state government finance included in the data is off by 10 percent in either of the endpoint years, the findings of only one of the studies described above would be robust ____________ 5See James Leigland, “The Census Bureau’s Role in Research on Special Districts: A Critique,” Western Political Quarterly, June 1990, Vol. 43, No. 2, pp. 367–380; Seymour Sacks, “The Census Bureau’s Role in Research on Special Districts: A Critique, A Necessary Rejoinder,” Western Political Quarterly, June 1990, Vol. 43, No. 2, pp. 381– 383; and James Leigland, “In Defense of a Preoccupation with Numbers,” Western Political Quarterly, June 1990, Vol. 43, No. 2, pp. 385–386. 5 enough to stand. With findings of changes in the revenue picture in the 3 to 6 percent range, it is important to be comfortable that the data underlying these comparisons have a error level that is lower. This is the goal of this report. Before turning to specific quality issues, it is important to identify the specific scope and range of the local government enterprise on which this report will focus. Exactly what is meant when referring to local government finance? The next section will provide a more in-depth discussion of the scope and scale of the local government enterprise. The Local Public Enterprise The local public enterprise refers to the collection of governments beneath the state level that raises revenues, expends monies in the public interest and sets policy for citizens of the local community. The boundaries of all of these entities are subordinate to the state, and this category includes all counties, cities, towns, special districts and school districts. It typically does not include non-profit corporations and public-private joint ventures that are not under the direct control of an elected or appointed body. For purposes of this report, the category “local government” does not include the state government or any of its agencies. Table 1.1 presents a summary of the local government sector in California in the year 1991–92. As this table shows, the size of this sector is considerable, representing overall total revenues nearly twice those in the state’s general 6 Table 1.1 Local Government in California, 1991–92 (revenues in billions of dollars) Entity Counties Cities School districts Transportation/planning agencies Community redevelopment agencies Special districts Grand total Number 58 466 1,005 81 381 4,995 6,986 Total Revenues 28.253 22.311 25.030 3.210 2.026 14.891 95.721 Own-Source Revenues 14.112 19.938 7.371 2.624 1.961 14.431 60.437 SOURCE: Annual Report of Financial Transactions Concerning Counties, Fiscal Year 1991–92; Annual Report of Financial Transactions Concerning Cities, Fiscal Year 1991–92; Annual Report of Financial Transactions Concerning Special Districts, Fiscal Year 1991–92; Annual Report of Financial Transactions Concerning School Districts, Fiscal Year 1991–92; Annual Report of Financial Transactions Concerning Redevelopment Agencies, Fiscal Year 1991–92; Annual Report of Financial Transactions Concerning Transportation Planning Agencies, Fiscal Year 1991–92. and special funds in 1991–92.6 Even on an own-source basis,7 the sector is large, accounting for $60 billion dollars. The local government sector is also highly fragmented, with nearly 7,000 entities—mostly special districts. Of the nearly 5,000 special districts, approximately 40 percent are dependent upon another entity for their governance and fiscal direction. These entities are typically governed by a board of supervisors, city council or school board and are ____________ 6State general and special fund revenues totaled $53.1 billion in 1991–92. It is important to note here, however, that these are not all own-source revenues. If all intergovernmental revenues reported by these entities are excluded, the sector has revenues of approximately $50 billion—almost exactly the same size as the state government itself. 7Own-source revenues are those revenues raised at the level of that specific government. Typically, they are the entity’s total revenues less transfers from other governmental bodies and agencies and thus include tax revenues, fines and forfeitures, interest and rents and current service charges. 7 entities such as redevelopment agencies and water and sewer enterprises.8 The remaining 3,000 or so have their own elected boards and include entities such as fire protection districts, utilities, airports and water districts. What Data Are Available? Sources of financial information available for this diversity of local institutions are limited. There are only two sources of comprehensive information on these local governments—the U.S. Department of Commerce’s Census of Governments and the State Controller’s Annual Report—and these two are closely related as will be seen below. A few other sources are available for selected categories of information. The California State Board of Equalization maintains detailed information on selected revenue series, and the California Department of Education maintains a detailed database on K–12 fiscal activity in the state. Census of Governments The U.S. Department of the Census produces a series on state and local finance every five years for every public entity it tracks in the nation. Detailed information is also published for larger entities every year. The five-year data set includes detailed revenue, expenditure and debt information on counties, cities, school districts and independent9 special districts. Dependent entities are included with the parent entity that retains governance control over it. ____________ 8About 90 percent of these dependent entities are governed by county boards of supervisors, and approximately 10 percent are governed by local city councils. Very few are governed by school boards. 9An independent special district is one that has independent governance (e.g., its own elected board) and/or independent revenue streams. 8 The raw inputs for the U.S. Census Bureau’s series, however, are not collected directly by the Census. Instead, the Census obtains the data tapes for the Controller’s Annual Reports on Financial Transactions (described below) and reorganizes the data to make the presentation more consistent with their national reporting format. According to the Census, only minor comparison and review activities are affected by the Census, so the bulk of the quality control associated with this information rests with the California State Controller’s Office. This is due in part to staff limitations and in part to their sense of the high quality of the data received from the state.10 California Controller’s Annual Reports on Financial Transactions The California State Controller collects annual information on local government finance within the state. The Controller maintains and distributes an annual survey of financial activity that includes revenue, expenditure and debt information for every public entity in its database.11 All public entities are required by law to collect this information and to provide completed surveys to the Controller’s office within 90 days of the end of the fiscal year. ____________ 10One original purpose of this study was to provide detailed comparisons between the data reported by the Bureau of the Census and those reported by the California Controller’s Office. The Census data for the most recent Census of Governments survey year (1991–92) have not been released as of the press time of this report and are not expected until early 1997. As a result, that detailed comparison was not possible. Even though the Census uses the Controller’s data as its primary input, conclusions about the quality and comprehensiveness of the Controller’s data are not and should not be automatically applied to the Census data. 11A copy of one of the Controller’s questionnaires—the questionnaire for counties—is provided in Appendix H. The other questionnaires are similar, although the categories and level of detail vary according to the language of the particular legislation establishing the category’s reporting requirements. 9 These questionnaires are then entered into a database system within the Controller’s Office and reviewed for quality and consistency with the prior year’s reported activity. Inconsistencies are identified and investigated. If problems are identified or suspected, the public entity’s audited financial information is also requested as it becomes available. The compiled information is then published in a series of reports. There are separate series for counties, cities, redevelopment agencies, special districts, school districts and transportation planning agencies. Care is taken to be certain that there is no duplicate reporting of revenues. California State Board of Equalization The State Board of Equalization, among other responsibilities, is the primary tax collector and distributor for the sales tax, the fuel tax, the alcoholic beverage tax and the cigarette tax. For example, it collects the sales taxes from all business in the state and then distributes the appropriate shares to each entity that imposes a portion of that tax. In this capacity, it retains detailed information on the distribution of each of these taxes. Detail on state, city and county receipts of each of these taxes12 is reported in the State Board of Equalization’s Annual Report. Additionally, the board collects and reports detailed information on the state property tax. This information is actually obtained from surveys entitled Annual Report of Property Taxes submitted by each county auditor-controller to the Division of Local Government Fiscal Affairs, Special Districts Unit of the State Controller’s Office. This information is collected independently of the annual questionnaires ____________ 12Cities and counties receive revenues for only some of the taxes administered by the Board of Equalization. The sales tax is the largest of these. 10 described above. Summaries are presented each year in the Board of Equalization’s Annual Report. California Department of Education The California Department of Education collects detailed information on school district finances in California. These reports include the activities of school districts, county superintendents of schools and joint powers agencies (JPAs)13 between county offices and school districts. These data are collected by the Department of Education using the J-200 series, J-400 series and J-600 series of reports. Only summary information is published by the Department of Education—the detailed information for each entity is not published separately, although the Department of Education advertises its willingness to provide detailed information to anyone for a nominal fee. This information is not collected by any other state or federal agency, including the California State Controller’s Office. While the Controller’s Office does publish information on school districts, the data in the State Controller’s Annual Report on Financial Transactions Concerning School Districts are compiled from summary information provided by the California State Superintendent of Public Instruction and the California State Department of Education. ____________ 13A joint powers agency is typically a collaborative entity that provides a specific service for a group of independent agencies who benefit from the economies of scale of having a larger pool of participants. The classic example of a JPA is a pooled risk selfinsurance entity. Each individual entity is unwilling to carry the risks associated with selfinsurance, but if 10 districts pool their risk, then it becomes an attractive fiscal alternative. 11 Local Governments Finally, local governments retain information on their own and, sometimes, other entities’ fiscal activity. A range of entities exist that can provide information on local government finance. Counties, for example, collect all the property taxes for a given county. County local agency formation commissions (LAFCOs) are responsible for arbitrating boundary disputes, approving annexations and allocating resources for government reorganizations within each county, and although they retain information on what kind of entities exist within a county, they do not retain fiscal information. Parent entities nearly always retain information on the fiscal activities of subordinate, dependent entities. Each entity is also required by law to maintain detailed records on its own fiscal activities. This information is retained in the form of budgets14 and in annual audits—the results of which are reported in comprehensive annual financial reports (CAFRs). Summary of Data Sources In summary, there are only three sources of comprehensive information about the fiscal activities of local governments: the Census of Governments, the Controller’s Annual Report series and the entities themselves. Since the data underlying the Census reports are largely those included in the Controller’s reports, it is appropriate that this study will focus on the data in the Controller’s reports. It should be noted, however, that the U.S. Census Bureau does perform significant reorganizations of the data reported in the Controller’s Annual Financial Transactions and that any conclusions about one data set do not ____________ 14The information for a given year for an entity’s actual revenues and expenditures is often included in the proposed budget two to three years later as a reference point. 12 necessarily carry over to the other. Because of the unavailability of the Census reports, this report necessarily focuses on the Controller’s numbers. Issues About the Quality of Local Government Data In discussions with many of the policymakers in the State of California, one fact became evident—nearly all of the likely consumers and users of the available data on state and local finance did not use the available series, and, if they did, they did so hesitantly. Private investment firms, for example, did not directly use the information. Other researchers, largely academics, often deferred important research questions because of a lack of confidence in the available data. Three sets of issues were raised by these consumers as explanations of their non-use of the data: 1. There is a range of public activity and entities that are not included in the reported information. 2. There were concerns about whether the data reported accurately reflect the actual fiscal activity of the entities. 3. The information is not produced quickly or timely enough. The purpose of this project is to assess the available data along each of these dimensions, with a particular focus on the first two issues—the comprehensiveness and the accuracy of the data. Comprehensiveness of the Data: Are All Appropriate Entities Included? One of the biggest criticisms of the state’s data systems is that they are not capable of tracking the creativity of local governments. Because 13 the various propositions have greatly increased the constraints and limits on local governments, some argue that these governments have become creative in finding ways of going about their business. These individuals also argue that the state’s data systems do not and cannot keep abreast of the institutions that come about as a result of this creativity. As a result, the data do not capture these new activities, and the research that uses these data therefore underestimates the actual size of public sector activity. Suppose, for example, that, in response to the voter approval requirements under Proposition 13, a city used a law that allowed the creation of a community facility district (CFD) to build a new civic center.15 In year one, the CFD issues bonds and builds the civic center—which is exclusively operated by the city. In year two, the CFD collects property tax assessments (parcel taxes) from those in the CFD and pays the debt service on the bonds. If the CFD was not required to report its revenues or for some reason was not included, then the Controller’s reports would subsequently underestimate the true size of the public sector’s activity in that city. This is precisely the kind of issue that is raised by practitioners and experts. To address this concern, this study will examine whether there is in fact public activity that is not included in the Controller’s reports, with a specific focus on whether the reports include the full range of entities they should and how complete the coverage is within those categories. ____________ 15These are also known as Mello-Roos districts. As discussed in Chapter 2, this law would allow the city to draw the borders of the district in such a way as to assure the voter support necessary for passage. 14 Accuracy of the Data: Do the Reported Data Match What Actually Happens? Another issue raised in the policy debate regarding the quality of the data is whether the reported information accurately represents the activity of the reporting entity. Since the data questionnaires are completed by each entity independently, the potential for variation in the interpretation of the instructions across more than 6,500 entities is significant.16 For example, there is some variation in the way that entities of the same type account for different revenues and expenditures. For example, one district may report a property tax that it receives from the county tax collector and that it passes on to another entity as a tax revenue and an intergovernmental transfer expenditure. A similar district may report the transaction as an intergovernmental transfer revenue and expenditure. Another district of the same type may not report it at all because it has nothing to do with the collection or expenditure of the tax. There is enough latitude in both the Controller’s instructions and the rules that govern public accounting17 to take any one of the above approaches, and cases of all three approaches were identified in the course of this study. Finally, the Controller’s questionnaires are due within 90 days of the fiscal year end, while the audits of most entities are not due until 12 ____________ 16If these errors occur independently, it can still be the case that any two errors could cancel each other out and that the overall totals would be correct. To the extent that someone would want to look at individual entities, he or she would care about these errors. Also, it is possible that there is a systematic underreporting of data brought about by the design of the reporting systems. 17These are the Generally Accepted Accounting Practices (GAAP) rules by which public agencies’ financial statements are prepared during audits. 15 months after the year end.18 This means that the Controller’s surveys may be completed with budgeted numbers, not audited numbers. These numbers represent the best guess as to what happened in the year recently ended, but audits are necessary to make certain that all fiscal transactions are properly and consistently reported. Anecdotes of the consequences of such variability in accuracy abound. One scholar points to several hundred million dollars of revenues missing from a southern California special district. Another story recounts the example of a certain special district in central California whose Controller’s questionnaire was completed by the spouse of one of its board members between clients in her hair salon. All of these examples raise concerns regarding the accuracy of the data. For public policy researchers, such as the Public Policy Institute of California, to use these data with any confidence, these accuracy issues must be addressed definitively. Since this is the only data set with comprehensive information regarding local government entities and their finances, this report will look in some detail at the accuracy of the Controller’s data. Timeliness of the Data: Are They Available Soon Enough To Be Useful? Even accurate data may have limited usefulness to practitioners if they are not available quickly enough. The Controller’s reports have been criticized for the length of time it takes them to become available. In part, the delay is understandable. While the completed surveys are due within 90 days of the fiscal year end (typically by September 30), ____________ 18The one major exception to this standard is the redevelopment agency whose audit due date coincides with the 90-day due date of the Controller’s questionnaire. 16 there is a tremendous amount of work to be done before the published reports become available. The survey data must be input into the data system; the data entry must be quality controlled; the information must be reviewed for consistency and accuracy with prior years and then the final report must be compiled, reviewed, proofed and published. For the 58 counties in the state, this can be done quite quickly, but for the report that includes information on the state’s nearly 5,000 special districts, this task takes much more time. These reports are often not available for more than 18 months (or two fiscal year budget cycles) from the end of the reported fiscal year. This timeline is problematic for legislators and committee staff who must make policy and budget decisions in advance of a fiscal year. Especially in rapidly changing times, such as the recession of the early 1990s, two-year-old information is not useful. Decisionmakers need to know what happened as soon as possible and the sooner the better. There are no alternative sources of information regarding state and local finance. Policy analysts and academics, to the extent that they are also doing current policy research and analysis, need current data as well.19 This aspect of the quality of the data is self-evident in the timing of the release of the various reports. The county and city reports are typically issued much earlier than the special districts reports, but in no case soon enough for the state’s decisionmakers. This study will not focus on documenting this timing but does include some ____________ 19The Census of Governments reports are released much later. For example, the 1991–92 detailed information was due to be released in the spring of 1996. Because of budget cuts, however, this release has been delayed until late 1996 or early 1997—if it is released at all. 17 recommendations in Chapter 5 for ways of improving the timeliness of the data. This Report Due to these concerns about the quality of the data, this report was commissioned. Its purpose was and is to take a careful and critical look at the Controller’s data. If the data were found to be reasonably comprehensive and accurate, then the debate over changes in local public finance can focus on substantive methodological and value-oriented issues. PPIC could then turn to addressing important questions of fact and policy using the data. If the data were found to be problematic, then it was hoped that prescriptions for corrective measures could be identified that would make them usable. As will be seen in the balance of this report, the data are largely comprehensive and accurate. Focus on Revenues This report focuses on the revenue side of the financial statement, as opposed to looking at expenditures or debt. This is mostly the result of this study’s origins in the debate over the public revenue or tax burden in California. Revenues are also one of the simplest series of data included in the Controller’s reports—making such an analysis more feasible.20 It is also true that local governments typically balance their revenues and expenditures, and therefore a reconciliation of one category will, at least ____________ 20The opportunity for different interpretations of expenditure categories is greater because of the major variation in the specific characteristics of local programs. Revenues are simpler because the types and character of the sources of resources for local public entities are quite consistent within a revenue category. 18 at the highest level of aggregation, provide a validation of the other category.21 This report does not address the debt issue—an area that represents the bulk of the private sector’s interests in the data. The complexity of this area coupled with the lack of good secondary sources of information were the primary reasons for this choice. As some preliminary work shows, accurate and complete reporting in CAFRs and other agencies’ reports of debt activity are quite poor. This report will provide some preliminary findings of just one crude debt category in Chapter 4, and this is a recommended area of further research in Chapter 5. Organization of This Report This report centers around the comprehensiveness and accuracy of the Controller’s data and is organized around those tasks. Chapter 2 focuses on the comprehensiveness question. It presents findings on the quantity and magnitude of entities that are missing from the Controller’s data set. Chapters 3 and 4 then turn to the accuracy of the data. The former compares some of the individual categories of information included in the Controller’s reports with third-party reports of that same information, while the latter compares the reported information for individual entities against those included in the respective entities’ audited financial reports. In general, the body of the report includes summary descriptions of this report’s findings, while detailed tables are included in the ____________ 21Local governments are not allowed to run fiscal deficits. 19 appendixes. In each case, references are provided in the text and notes for each table in the body of the report, including the location of the detailed information underlying the summaries reported. 20 2. Are the Data Comprehensive: What Entities Are Missing? One of the main complaints about the Controller’s data centers around the concern that they may not include significant numbers of entities that should be included in the public sector. Failure to include entities would have two important implications: (1) the activity of these entities would escape public reporting and its implicit public accountability and (2) analyses using the Controller’s data would underestimate the full extent of public activity. The purpose of this portion of the study was to ascertain the extent of the omitted entity problem and to devise corrections for the data if the problem was found to be significant. Two approaches were taken to identify the public entities that are missing from the Controller’s reports. First, numerous individuals identified Mello-Roos or community facility districts (CFDs)1 as one ____________ 1Mello-Roos refers to the name of the two California legislators—Senator Henry Mello and Assemblyman Michael Roos—who sponsored the legislation that allows for 21 category of entities that were routinely excluded from the Controller’s reports. As a result, this study conducted a small survey of CFDs with an eye toward understanding precisely how many had been omitted from the Controller’s report. A census of these districts was also performed to identify the precise amount of revenues likely to be unreported. The second portion of the research into missing entities focused on non-CFD entities that might be missing from the Controller’s data set. This entailed comparing lists of public agencies from different sources to the entities included in the Controller’s reports. This phase of the research focused predominantly on the lists provided by county local agency formation commissions (LAFCOs) and the California Debt Advisory Commission (CDAC). Community Facility Districts Are Largely Missing Community facility districts allow local governments to issue bonds to cover the cost of financing new public facilities. Like those issued by special assessment districts, the bonds are repaid primarily through a special parcel tax on property within the district. Unlike special assessment district financing, CFDs do require approval from a supermajority of affected voters, but the district lines can be drawn to improve the chance of approval. In undeveloped areas with fewer than 12 registered voters, only landowners may vote. Most of the early CFDs were established through landowner vote. Although Mello-Roos financing began in the early 1980s, only a few CFDs were established in the initial years. By 1991–92, however, there were 233 different CFDs: 115 located in cities, 40 in counties, 52 in ____________________________________________________ the creation of community facility districts. These will be discussed in the following section. 22 school districts, and 26 in special districts (primarily redevelopment agencies and water districts). Mello-Roos bonds are not backed by the full faith and credit of their “parent” entity, and their revenues are not subject to the parent entity’s Gann appropriations limits. In addition, the Controller does not require the parent entities to include the CFD revenues in their report of fiscal activity. As a result, the special tax and other money earned for the CFDs represent a large class of local government revenues that is missing from the Controller’s report. How Many CFDs Are Not Included? As the research into Mello-Roos districts progressed, it became clear that, even though the Controller had no provision for their inclusion in its report and questionnaire formats, some parent entities did include them in their reported revenues. In response to this problem, 45 of the state’s 233 CFDs were randomly selected and surveyed to ascertain if their revenues were included in the Controller’s report in some way. The first discovery of this survey was the immense diversity in the positions of those who handle the bookkeeping and reporting for these districts. Sometimes the appropriate person was found in the accounting/finance department of the parent entity, sometimes they were found in a developer’s office and other times in an accounting firm’s office. These individuals were not always able to tell whether they were included in their parent entity’s finances or not. There was also a timing issue— because this study focuses on 1991–92, there was not always someone who knew what was done in 1991–92. From this small sample, it is estimated that approximately 20 percent of the CFDs are included in their parent’s reported information in the 23 Controller’s reports. These estimated 50 or so CFDs are affiliated with both large and small entities, although small parents included them more often than did larger parents. It is clear, therefore, that there is some degree of missing CFD activity in the Controller’s reports. How Big Is the Missing CFD Problem? To size this missing entity problem, nearly2 every CFD in existence in California in 1991–92 was contacted by PPIC. Information on all revenues the entity received in 1991–92 was requested. Since these are largely financing entities, these revenues are primarily property tax assessments and interest, with some contributions and fees from developers. This information was often aggregated with other activities in audited reports,3 so extensive follow-up was required to obtain a clear understanding of each CFD’s revenue picture. The findings of that census are presented in Table 2.1. Table 2.1 California Community Facility District Revenues, 1991–92 Revenue Category Property taxes Interest Other Total revenues Total Revenues $219,265,383 59,729,504 4,518,994 $283,513,881 ____________ 2Only five CFDs were not contacted because a contact person could not be identified. Their revenues were imputed for purposes of this portion of the analysis by using amounts from comparably sized CFDs. 3CFDs were sometimes aggregated with other debt service activities and sometimes with benefit assessment and special assessment districts. 24 All of the community facilities districts in the state have combined revenues of only $283.5 million.4 Because of difficulties in attributing interest revenues directly to the CFD in some cases, the interest revenues are likely underestimated by a small amount.5 The revenues associated with these CFDs represent a negligible part of the overall parent in each case. Even with a 10 percent increase in interest revenues, the resulting total would only represent a drop in the bucket—a mere 0.3 percent— relative to the size of the local economy presented before in Table 1.1. Community facility districts, therefore, do not represent a major problem in terms of overall effect on the size of local public revenues. Clearly, if one was interested in specific issues associated with these entities, the lack of data would be problematic. The existence of one category of missing entity does give rise to the question, “What other public entities may be missing?” The second portion of this chapter will focus on this problem. What Other Entities Are Missing? To find out what other public entities may be missing from the Controller’s reports, every list of public entities that could be obtained was compared with the list of entities included in the Controller’s reports. Entities were reviewed to see if they were directly included as ____________ 4A review of the number of CFDs formed since 1991–92 shows a 31 percent increase in the number of CFDs to date. If the revenues associated with these entities were proportional to those in existence in 1991–92, the total revenues for CFDs, which would be largely missing from the Controller’s numbers, would be $371,069,638—still a relatively small amount. 5In some cases, CFD interest revenues were lumped together with interest revenues from other activities of the parent entity. In other cases, the interest revenues were mixed together with the CFD’s interest revenues. It was impossible to separate these amounts, so they were excluded from the reported totals. This happened with 26 CFDs. 25 separate entities or indirectly included through the financial information reported by a parent entity. The two main sources used to locate missing entities were the county local agency formation commissions and the California Debt Advisory Commission. In addition, individual city and county audited financial statements and lists from some county auditor-controllers were consulted, but neither of these sources uncovered any entities that were not also covered by the LAFCOs and CDAC lists. With the exception of the Mello-Roos districts, very few entities were discovered that were not already tracked by the Controller. Missing Entities Identified from County LAFCO Lists Every California county, with the exception of Alpine and San Francisco, has a local agency formation commission that tracks changes in special districts and their spheres of influence, including changes in boundaries and the creation of new districts. Lists of special districts were obtained from all 56 LAFCOs in the state. Unfortunately, some of the LAFCOs do not keep detailed records, and, as a result, it was impossible to know when the entities were created and whether they were in existence in fiscal year 1991–92. The lists from the 56 LAFCOs were checked against the list of entities included in the Controller’s report, and 129 entities were identified that did not appear on the Controller’s list. In most cases, there was no address, telephone number or contact name provided. As a result, a telephone search was instituted for each of these entities through directory assistance, the county auditor-controller, city budget offices and other city officials, and other related special districts. Information was obtained for the vast majority of these potentially missing entities, but 26 there were still 10 that could not be reached or located. The findings of that search are presented in Table 2.2. Nearly two-thirds of the entities on the list turned out to be subsidiary agencies of larger entities, with their financial activity included in the parent entity’s report. In addition, 31 were formed after 1991–92 or had no financial activity in that year, and five were private water companies. Two entities were identified that did not seem to report their fiscal activity to the Controller: the Sonoma County Law Library and Reclamation District #813 in Sacramento County. The total revenues for these two entities were $382,299 in 1991–92.6 Missing Entities Identified from the CDAC List Any government entity or private agency that issues tax-free bonds must report the sale to CDAC. The list of all entities that issued debt between 1982 and 1992 was compared with the Controller’s lists, and 251 entities were identified that were not listed separately in the Controller’s report. A random sample of 40 of these entities was drawn Table 2.2 Results of Search for Missing Entities Using the LAFCO Lists Description Reported to the Controller Not in existence in 1991–92 or a private company Did not report to the Controller No information available Grand total Number 81 36 2 10 129 Percentage 63% 28% 2% 7% 100% ____________ 6The total revenues for the Sonoma County Law Library were $361,870 and for Reclamation District #813 were $20,429. 27 for further exploration.7 A telephone search similar to that described above for the LAFCO search was undertaken. The results of that search are presented in Table 2.3. Of the 40 potentially missing entities contacted, 26 were included in some parent entity’s report to the Controller, seven did not exist in 1991–92, information was not available on two,8 and only one (the Riverbank Public Financing Authority in Stanislaus County) did not report its activity. The remaining four could not be reached. The only identified missing entity’s income was $152,000 in tax revenues. An Overview of Missing Entities In the context of the findings in this chapter, it can be concluded that the Controller’s reports do include essentially all of the public revenue activity in the state. Combining the CFD, LAFCO and CDAC results, the total of missing revenues identified is less than $300 million dollars.9 In the context of the local government sector’s nearly $96 Table 2.3 Results of Sample Search for Missing Entities Using the CDAC List Description Reported to the Controller Not in existence in 1991–92 or a private company Did not report to the Controller No information available Grand total Number 26 7 1 6 40 Percentage 65% 18% 2% 15% 100% ____________ 7A sampling approach was chosen because of the difficulty encountered in reaching these entities to obtain detailed information on their reporting status and revenues. 8The entities were reached, but they were not able to provide the necessary data. 9The amounts identified in each of the three sections above, when combined, total $284,048,180. If this sample was extrapolated across the population, the total error would be $284,849,930. 28 billion in revenues, these missing entities represent a negligible amount of activity. As a result, unless one is looking specifically at policy issues that are closely associated with CFDs, the data are comprehensive enough for all intents and purposes. The next chapter will address the other major issue associated with the quality of the data—their accuracy. 29 3. How Accurate Are the Data: A Statewide Comparison of Sales and Property Taxes One approach to reviewing the Controller’s data is to take revenue streams for which there is good third-party reporting and compare that reporting with the amounts reported in the annual reports. In such an approach, both statewide and, when available, more detailed reports of revenues can be compared with those given in the Controller’s reports. This chapter describes the findings of such an exercise for two important local revenue streams—property and sales taxes. Both property and sales taxes are important sources of revenues for counties, cities and other local districts.1 Both of these local revenue sources, however, are collected and distributed by other levels of government. In the case of sales taxes, local businesses pay the sales taxes ____________ 1Sales taxes account for 1.3 percent and 10.0 percent of county and city overall revenues, respectively. Property taxes account for 23.4 percent of county revenues and 9.6 percent of total city revenues. 31 they collect from consumers directly to the California State Board of Equalization. The Board of Equalization then distributes these sales taxes to the state, county, city and local transportation districts. For property taxes, the revenues are collected by the county and distributed from there to each county, city, school and special district. A report is filed by each county’s auditor-controller to a division of the State Controller’s Office, which is in turn incorporated into the California State Board of Equalization’s Annual Report. These reports, inasmuch as they are prepared by independent offices within each county’s government, represent an important third-party reference against which the self-reported numbers in the Controller’s reports can be compared. The county auditor-controller’s reports, filed annually, include detailed entity-by-entity listings of the property taxes billed within each county. It is important to note that some level of error is expected between the sources for property tax revenues, inasmuch as the county auditor-controller’s reports detail taxes and levies billed while the entities in the Controller’s reports are reporting the amounts they have received. The comparison is useful, however, to get a sense of major errors in selfreporting in the Controller’s reports. The inclusion of information from slightly different points in the tax collection process also allows for a broader view of the quality of the data. In this portion of the analysis, these sales and property tax revenues reported by the individual entities in the Controller’s survey are compared with third-party sources. For sales taxes, the reported amounts2 are compared with the amounts reported by the California ____________ 2The reported amounts are drawn from Controller’s Annual Report of Financial Transactions Concerning Counties, Fiscal Year 1991–92, and Controller’s Annual Report of Financial Transactions Concerning Cities, Fiscal Year 1991–92. 32 State Board of Equalization for sales taxes in the appendix tables of the Board’s Annual Report for the Year Ending June 30, 1992. Reported property taxes are compared with the amounts listed by each county’s auditor-controller in its Annual Report of Property Taxes.3 This chapter is organized simply. The first section will examine sales taxes on a statewide basis and the second portion will focus on property taxes. Each section will address the types of entities that receive that type of revenue. Sales Taxes Only four types of entities in California receive sales taxes—the state government, county governments, city governments and local transit agencies. Since the focus of this analysis is on the local government data, the portion of sales tax attributed to the state government is not examined in detail. The sales taxes received by the latter group, local transit agencies, are reported separately in both the Board of Equalization’s Annual Report and the Controller’s report and are subsequently not considered here. This leaves the two largest categories of local recipients for state sales tax revenues—cities and counties. The specific findings for those two groups are presented below. ____________ 3These surveys are actually collected by a different division within the State Controller’s Office. They represent a third-party source of information because they are completed by a different office within local government—in many cases an office headed by an elected official. There is no joint handling of the reported information within the Controller’s Office. As a result, the information included in the Annual Financial Transactions series and the Annual Report of Property Taxes is separate and independent. 33 Cities Cities levy 1.00 percent sales tax on all taxable sales within their borders. Some cities share part of their levy with other entities and report only the portion they retain, explaining the variation in reported sales tax rates across California cities.4 Table 3.1 presents the results of comparing the sales tax revenues reported by cities in the Controller’s report with those reported by the Board of Equalization. This table includes no county revenues. As Table 3.1 shows, the amount of variation in aggregate between the Controller’s and Board of Equalization’s reports is negligible. On a county-by-county comparison, the largest variation in total sales tax revenues to cities is only 4.6 percent.5 Moreover, 82 percent of all cities reported sales taxes revenues to the Controller that were within 3 percent Table 3.1 Overall Summary of Comparison of Reported City Sales Tax Revenues, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Board of Equalization Difference Percentage difference Sales Tax Revenues 2,401,082,947 2,401,772,801 689,854 0.0% NOTE: See Table A.1 for underlying detail and additional references. ____________ 4In certain cases, there are pass-through arrangements with some subordinate and dependent entities whereby a portion of that levy is allocated to another government. The remaining portion of variance arises from special taxes for transit purposes that can be imposed at the county level. 5Table B.1 presents these findings on a county-level basis. In this approach the sales taxes for all cities in a county are added together and then compared. 34 of those reported by the Board of Equalization, and 88 percent reported to within 5 percent. Even on a city-by-city basis, the individual variation is quite small— averaging only 1.9 percent. There are only six cities in the state for which the individual differences between the Controller’s and Board of Equalization’s reported numbers are more than a million dollars— Alhambra, Azusa, Berkeley, Hayward, Sacramento and San Diego. In combination, their net variance is only $7.7 million out of total revenues of $189.3 million, or 4.1 percent. Clearly the data correspond closely between the two data sets. Counties The total percentage variation in the county sales taxes is a bit higher, although from a much smaller dollar total. The comparison of the Controller’s and the Board of Equalization’s sales tax series for counties is presented in Table 3.2. As this table shows, the overall level of variation is a bit higher than that found in the cities—5.1 percent. At the same time, individual counties’ variation is generally much lower. Table 3.2 Overall Summary of Comparison of Reported County Sales Tax Revenues, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Board of Equalization Difference Percentage difference Sales Tax Revenues 322,601,996 306,124,396 16,477,600 5.1% NOTE: See Table A.3 for underlying detail and additional references. 35 The amounts reported in the Controller’s surveys and the amounts reported by the Board of Equalization correspond closely, with 74 percent of counties showing differences of less than 3 percent, and 82 percent with differences of less than 5 percent. The variation reflected above is concentrated in just a few counties. Eighty-one percent of the variance arises from the six counties whose individual variation is greater than one million dollars—Alameda, Fresno, Sacramento, San Bernardino, Sonoma and Stanislaus. In each of these cases, the amount reported to the Controller is greater than that reported by the Board of Equalization. The data were examined to see if cities within these overreporting counties may be underreporting, thus offsetting the overreporting—no such offsetting pattern was identified. In this context, it is important to remember that 1991–92 was the first year of the recession and that, in many cases, actual sales tax revenues were often much lower than anticipated in county budgets. Since the Controller’s report is more likely to use budgetary numbers, it is possible that in these individual cases, the numbers reported are in fact higher than what actually occurred. Property Taxes The reporting and review of property tax revenues are complicated by the diversity of the entities that receive the property taxes. There is little uniformity in how the information on property tax revenues is reported across the various groups that receive them in the Controller’s reports. This section will present detailed findings regarding property tax revenues for two of the four major categories of entities that receive them: cities and counties. Aggregate data are reviewed for a third category: school districts. A review of the fourth group, special districts, 36 is not included because the Controller’s report for special districts does not always separately report property tax revenues. Cities A summary of the comparison for cities is presented in Table 3.3. As this table shows, the amounts reported in the Controller’s report6 and those reported by the county auditor-controllers7 correspond closely. In fact, nearly all of the statewide variance can be attributed to cities in Los Angeles County, and more specifically to the Cities of Long Beach and Los Angeles whose combined property tax revenues totaled more than $677 million.8 Without the $70.1 million variance from cities in Table 3.3 Overall Summary of Comparison of Reported City Property Tax Revenues, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports County auditor-controllers Difference Percentage difference Property Tax Revenues 2,959,177,966 2,844,810,343 114,367,623 3.9% NOTE: See Table B.1 or underlying detail and additional references. ____________ 6The specific revenue categories from the Controller’s reports include countywide taxes, less-than-countywide taxes, prior-year taxes, special district augmentations and voter-approved indebtedness. 7The specific series from the Annual Report on Property Taxes include all levies and taxes associated with county and less-than-countywide activities. It does not include the homeowners property tax relief amounts, which are typically categorized as intergovernmental revenues. 8The reporting of property taxes in Los Angeles has been the subject of much concern and review. A state audit completed by the State Auditor-General’s Office and entitled An Analysis of the Deficiency in the 1984–85 State School Fund (April 16, 1985, Report P-530) identified specific problems in the way supplemental assessments were handled. This could account for some of the variances identified here. 37 Los Angeles County, the overall variance would be $44.3 million on total revenues of $1.97 billion or 2.2 percent. These variances are negligible. Counties The information for property taxes for counties is presented in Table 3.4. As is the case in the cities, the bulk of the variance in reported county property tax can be attributed to Los Angeles County. Without the variance raised by Los Angeles County, overall variance drops to $83.5 million on total revenues of $3.373 billion or 2.5 percent. In percentage terms, almost all county auditor-controllers reported property tax revenues for their counties that were very close to amounts reported separately by the county to the State Controller’s Office. Fifty-four percent reported amounts within 3 percent of the amount reported to the State Controller, while nearly three-fourths were within 5 percent. Most of those not falling within 5 percent of the amounts reported by the county auditor-controllers were small counties where relatively small dollar variances represent large percentage variances. Table 3.4 Overall Summary of Comparison of Reported County Property Tax Revenues, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports County auditor-controllers Difference Percentage difference Property Tax Revenues 5,593,901,820 5,353,466,793 240,435,027 4.3% NOTE: See Table B.3 for underlying detail and additional references. 38 School Districts Reporting of school district finances by the Controller in general and property taxes in particular are very sketchy at best. The Controller’s Annual Report 1991–92 on Financial Transactions Concerning School Districts of California is a restatement of the information collected by the State Superintendent of Public Instruction and the California Department of Education on its J-200 series of reports. The categories included in this report are defined in the California School Accounting Manual, 1992 Edition, and do not correspond to any of the categorization included in the Controller’s other reports. Property taxes, however, are a major component of the revenues for K–12 school districts in California, and they are reported at the grossest level in the Controller’s report. The comparison of these reported numbers to those described in the Annual Report on Property Taxes filed by each county’s auditor-controller are presented in Table 3.5. Table 3.5 Comparison of School District Tax Revenues, 1991–92 (dollars unless otherwise indicated) Source Controller County auditor-controllers Variance Percentage variance Reported Property Taxes 5,309,829,960 5,365,498,883 –55,668,923 1.0% SOURCE: Controller’s data are from the Controller’s Annual Report of Financial Transactions Concerning School Districts of California, Fiscal Year 1991–92, p. IX. The county auditor-controllers’ data are from the original surveys submitted to the Controller’s Office entitled Annual Report on Property Taxes. These surveys are used to prepare the summary tables presented in State Board of Equalization’s Annual Report for the Year Ending June 30, 1992, Tables 4–5, pp. A-4 to A-19. 39 Overall, this level of variance is quite low. Unfortunately, no more detailed analysis of the district-level activity was possible because of the unavailability of detailed information in the Controller’s report. Furthermore, no detailed information, on a district-by-district basis, is published by the Department of Education. Overview It is evident that, when considered separately, the sales and property tax revenue information reported independently to the Controller by each entity corresponds quite closely to the amounts reported separately by the California State Board of Equalization and each county’s respective auditor-controller’s office. 40 4. How Accurate Are the Data: A Review of Data Reported by Individual Entities Another serious issue raised by consumers, users and potential users of the data was that the numbers reported by entities were sometimes, at best, poor estimates of actual activity. Anecdotes ranged from missing property taxes in the millions of dollars to a hairdresser in a small town completing the annual survey for a special district on whose board her spouse served. Others pointed to the fact that private sector firms such as Moody’s and Dunn & Bradstreet, whose livelihoods depend on good data, do not use the data included in the Controller’s reports.1 The overall concern about consistency voiced by practitioners was not so much at the aggregate level, but at the local level where individual ____________ 1It turns out that these firms generate their own information based on specific, current data provided by the entities requesting ratings for prospective debt issues. They typically focus only on the entity in question and do not have an overall statewide database. Their primary concerns about the Controller’s data seem to center more on the timeliness of the data than their quality. 41 public entities provided the responses to the Controller’s questionnaire. With such a broad and diverse range of public activities and accounting methodologies at the local level, there are concerns that any set of instructions could and would be interpreted differently by each individual entity. Looking at the questionnaire process from the local entity perspective, it is easy to see how this problem can arise. A copy of a typical questionnaire is included in Appendix H. An individual city manager does not have the comparative benefit of looking at what other cities are doing when completing the questionnaire. As a result, he or she would be more likely to follow the guidelines set forth in the city’s own budget, which is the result of local decisionmaking, than to consider the questions from the perspective of statewide consistency. Once this variation has been introduced, there is no good way for state data managers to identify this variation. In other cases, there is some question as to which point local entities should recognize revenues. Some entities, such as counties, receive funds, such as the property tax, which they then pass on to other entities. Whether these revenues should be accounted for as own-source revenues or intergovernmental transfers is another area of ambiguity. The result of these and other ambiguities is a perception that the information contained in the Controller’s reports is not consistent or complete. This chapter presents the findings regarding the quality of the data contained in the Controller’s reports. The first portion of this chapter will describe the general methodology used to prepare the findings detailed in the balance of the chapter. The remaining sections provide detailed descriptions of a major undertaking in which the Controller’s numbers were compared with audited numbers. 42 General Methodology and Issues For this part of the analysis, the goal was to see how well the information contained in the Controller’s reports reflects reality. This required comparing the reported information with a third-party representation of activity during the year. The third-party standards against which the Controller’s reports were compared were the comprehensive annual financial reports (CAFRs) for each entity. CAFRs are the results of annual audits required of all governments in the state. These reports are prepared by private sector accounting firms and reflect the application of the audit process and Generally Accepted Accounting Practices (GAAP) to the entity’s activities during the year. Since these audits do not have to be completed for a year after the fiscal year end, the results of these audits are not generally available for the Controller’s 90day deadline. In general, a sampling list was prepared for each category of entity— counties, cities and special districts. These samples were stratified by overall size. The largest entities in each category were all selected. A random sample of the remaining entities in each category was selected. The final sample contained entities from all sizes and revenue levels. The detailed sampling methodologies for each category are discussed in Appendix C. For each entity in the sample, a call was made to that entity requesting the CAFR for the 1991–92 fiscal year. Follow-up calls were made until the CAFR was obtained. Upon receipt, the values in the CAFRs were compared with those reported in the Controller’s reports. Since CAFRs typically define government entities to include all dependent entities over which they exercise control and the Controller’s reports include many dependent entities in their special districts report, it 43 was necessary in some cases to combine entities reported separately in the Controller’s reports to obtain an entity comparable to those listed in the CAFR. For example, a county’s CAFR would include all of the combined activities over which the county Board of Supervisors exercised control— such as a transit agency, a redevelopment agency, a water agency and any number of county service areas. Each of these latter agencies would be included separately in either the Controller’s Annual Report of Financial Transactions Concerning Special Districts of California or the corresponding book for redevelopment agencies. To compare these entities, it is necessary to combine the amounts reported in the Controller’s report for each entity with that of the county. It is this combined entity that is compared with the audited amounts reported in the CAFR. The specific reconciliation methodologies are given in Appendix D. Because of time and fiscal constraints, the comparisons in this study focused on only revenues for public entities in the sample for the fiscal year 1991–92. This raises several issues about the generalizability of this research to other fiscal categories, especially expenditures, and to other fiscal years, as well as comparability to other entities not in the sample. Moreover, there is a certain level of variability implicit in the differences between public fund and GAAP accounting. Each of these concerns is addressed in more detail below. Revenues Versus Expenditures as a Comparison Point This analysis focuses on the categorization and levels of revenues received by local entities in California and does not address the expenditure or capital portions of the equation. The focus on revenues 44 arises in part from the original motivation of this study—to validate available data before answering questions about the state’s changing public revenue burden, a task that will require a high level of confidence that these public revenues are accurately reflected in the available data. Additionally, revenues are typically more identifiable and usually defined by external forces. For example, sales and property taxes are determined by eligible sales volume and property values, respectively. Expenditures, however, are endogenous to the public policy process and hence more subject to manipulation and creativity. Revenue categories tend also to be more unambiguous and explicitly categorized across entities and activities, while expenditure categories are much more ambiguous. In aggregate, however, local entities tend to expend whatever revenues they receive in a given year, and aggregate revenues are a reasonable proxy for aggregate spending. This chapter looks closely at the aggregate revenues. As will be discussed below, categorization issues are difficult for both revenues and expenditures. Using 1991–92 as a Base Point For reasons of financial practicality, a choice was made to study one year in depth rather than reviewing several years more shallowly. It is critical to be confident that this study answers the question: “How well do the available data reflect public activity?” To be truly useful, a study of data quality over time would have to review data quality at least back to before the adoption of Proposition 13. If one ties into the Census of Governments, this would require going back to 1977. The cost of performing an adequate review of the quality of the data over such a long period of time would have been prohibitive. There is 45 the further issue of the availability of the CAFRs from each of the local entities. It was difficult enough to obtain the 1991–92 CAFRs from many entities, and prior year reports were often unavailable—a factor that would greatly increase the difficulty associated with completing this study and likely render it prohibitively expensive. Recency is also of concern when trying to reconcile the amounts listed in the two reports. While it is possible to find someone in a county, city or special district who can comment on their reported numbers for a recent year, it is next to impossible to obtain clarification and explanations for much older reports. In many cases, this information has been shipped away to storage or even destroyed. In light of these difficulties, it was decided to study one year in detail and 1991–92 was chosen. This year was selected because it represented the most recent year for which the U.S. Department of Commerce’s Census of Governments data were likely to be available.2 It is also recent enough that most entities would have their CAFRs readily available for this year. Over time, there have been some changes in the way that the State Controller’s Office handles the collection and aggregation of the numbers reported. In 1982–83, for example, the entire state government underwent a change in the basic accounting methodologies used to report revenues. Most of these changes were associated with the timing ____________ 2The original intent was to compare and reconcile the two data sets. The Census of Governments produces the comprehensive census of local government financial activity every five years in years ending in two and seven. Subsequent budget cuts have called into question whether the detailed 1991–92 data will in fact be made available. The data set was not available at the time this report went to press. If and when this data set does become available, a technical update comparing these data with the Controller’s data may be prepared. For the reasons discussed in Chapter 1, however, it is expected that these two data sets will map closely. 46 of when revenues are recognized and affected primarily the two- to threeyear window during which they were implemented. They had little longterm impact. Periodically, the instructions and requirements are updated and improved. Such an effort is currently under way and was most recently done on a large scale in 1984, when much of the current system was automated. In a pilot exercise to test the variability of the data over time, one entity was evaluated in 1976–77 and 1981–82. The results of this exercise were inconclusive, but they generally paralleled the findings for the 1991–92 fiscal year.3 The entity selected was the City and County of San Francisco, and, as the findings below will show, this particular entity had an exceptionally high level of variation and it was an entity for which explanations of differences between the audits and reports were not forthcoming. Brief reviews were also done for two other major entities, and the level of variation was comparable. Because of the limited nature of these exercises, only the results from the 1991–92 analysis are presented here.4 Some Variance Is Anticipated One important consequence of the methodology chosen in this part of the study is that a certain level of variance is to be expected. Public ____________ 3The city in question showed approximately the same level of variability in the prior years as in 1991–92. Contact with the city, however, failed to obtain confirmation as to the explanations of that variability. 4Given that the Controller has, over time, revised the methodologies and, in some limited cases, the reporting of information in the Controller’s Annual Financial Transactions series, there are some reasons to be concerned about the full reliability of earlier years. For the purposes and goals of this study–to validate the data with a view toward looking at local government creativity–it was decided that a recent year would be much more useful. The user is cautioned when using data from other years, however, that there may be individual quirks in other years that this study does not bring to light. 47 fund accounting reports are being compared with GAAP-based accounting reports. There are differences in the conventions of these two accounting techniques that ensure that the numbers will usually not match exactly. For one difference, public fund accounting is essentially on a cash basis—local entities recognize cash receipts as revenues when they receive them and recognize expenditures when they make payment.5 The audited reports are reported on an accrual basis—funds are recognized in the year they are earned. This difference will have an effect in years when the revenue stream is either increasing or decreasing and the amounts carried over from the prior year differ from the amounts carried into the following year. To better understand this concept, consider the following hypothetical example.6 Dream City has monthly sales tax revenues of $10 million. The city receives the monies for each month on the first day of the following month. If the city is on a cash basis, in a given year it would receive 12 times $10 million or $120 million. The first payment, received on July 1, however, was not for revenues received in this year, but for revenues from the last month of the prior year.7 If monthly revenues remain flat over the period, however, the additional revenues received from last year would exactly equal the revenues not received in the current year. If, however, revenues from the last month of the prior year totaled $10 million and revenues from the last month of ____________ 5This is not universally true for all public entities in the study. In some cases, entities use a modified cash approach where some revenues and expenditures are accrued to the fiscal year in question. 6This example is a simplified stylization to demonstrate how the methodologies differ. Actual accounting practice is more elaborate and complicated. 7The last month of this year’s revenues will be paid in the following fiscal year. 48 the current year totaled $15 million, there would be an apparent discrepancy between the two. On a cash basis, sales tax revenues would still be $120 million. On an accrual basis, however, revenues would total $125 million (11 times $10 million plus $15 million). If one compared the cash-based report with the accrual-based report, there would be a discrepancy of $5 million. This is just one type of difference that can arise when comparing the entity public fund accounts with the audited reports. Because of these types of variations, a certain level of variation is expected in these reconciliations. The exact amount of variation attributable to these types of errors in each entity is nearly impossible to precisely quantify because of the multitude of accounting positions that can be taken by a particular enterprise. Discussions with experts in the exploratory phase of this study placed this expected error within the 1 to 5 percent range—but it could be higher or lower. Some entities clearly used audited information to complete the Controller’s questionnaire, and their variance is essentially zero. Others are higher. If the variance is unexplainable or not attributable, it is so identified. Overall Findings Overall, this research found that the amounts reported in the Controller’s reports closely correspond to the revenues reported in the audited financial statements. The level of variation, given the complexity and range of activities within each group of entities, was remarkably low. Both the local governments, which provide this information, and the State Controller’s Office, which compiles and reviews it, should be commended for the overall accuracy of these data. 49 Specific Findings: Counties Fifteen of the state’s 58 counties were examined and reconciled to identify the level of variation between the numbers reported to the Controller and the amounts reported in their audited financial statements. The sample was drawn by selecting the 7 counties with the highest revenues in the state and then a random sample of 8 of the remaining 50.8 See Appendix C for more details regarding the sampling methodology. Overall Variance by County Overall, the aggregate difference between the audited amounts and the reported amounts is relatively small. Table 4.1 presents a summary of the results for the 15 counties in the sample. These counties reported $22.22 billion of revenues to the State Controller, and the audited financial statements included $22.50 billion of revenues. The difference of $276,318,630 represents a mere 1.2 percent of total revenues—a Table 4.1 Summary of Comparison of Reported Overall County Revenues for Study Sample, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Audited financial reports Difference Percentage difference Overall Revenues 22,220,798,294 22,497,116,924 276,318,630 1.2% NOTE: See Table E.1 for underlying detail and additional references. ____________ 8The Controller’s convention of excluding the City and County of San Francisco was followed for these reconciliations. Hence the eligible pool of counties is 57, not 58. Note that San Francisco is included in the city sample below. 50 variance level that is well within that attributable to differences in accounting methodologies. The range of the overall variance within counties in the sample is from 0.0 to 3.1 percent. In most cases, the larger percentage variances are in the smallest counties where relatively small dollar differences can account for large percentage variances. The one exception to this is San Diego County, which shows a variance of 3.0 percent. Almost half ($25.8 million) of this variance is attributable to property contributions from property owners in conjunction with a development project. These are routinely included as part of certain Controller’s special district questionnaires, but there is no such line in the county questionnaire. These were reported as miscellaneous revenues in the audited numbers. Categorical/Classification Variance Another dimension of the values reported in the Controller’s report is whether the categories in which the revenues are reported correspond to those in the audited reports. For example, are counties really reporting all revenues from the use of money and property? These revenues are typically rents and interest from various property holdings and investments. Table 4.2 reports the net overall variance broken down by revenue category. As this table shows there are some differences between the classification of revenues in the Controller’s reports and in the audited financial reports. The specific character of these classification differences, however, should be downplayed. For example, GAAP does not specify a separate category for special benefit assessments for counties, and the 51 Table 4.2 Overall Variance in Revenues by Revenue Category for Counties, 1991–92 (dollars unless otherwise indicated) Category Controller’s Report CAFR Percentage Difference Differencea Taxes 5,231,000,836 Special benefit assessments 9,850,947 Licenses and permits 147,798,329 Fines and forfeitures 204,368,563 Revenues from the use of money and property 563,547,755 Intergovernmental funds 10,236,203,283 Current services 3,250,273,883 Other revenues 2,577,754,698 Total 22,220,798,294 5,217,461,331 0 164,859,169 210,305,264 771,650,447 10,332,996,948 3,379,490,060 2,420,353,705 22,497,116,924 13,539,505 9,850,947 –17,060,840 –5,936,701 –208,102,692 –96,793,665 –129,216,177 157,400,993 –276,318,630 0.3% 100.0% –11.5% –2.9% –36.9% –0.9% –4.0% 6.1% –1.2% aPercentage difference is the difference as a percentage of the Controller’s reported amounts. CAFRs subsequently do not separate these revenues. If the first two tax categories are combined, the overall variance for taxes rises to 0.4 percent. It is also important to look at both the absolute and percentage variances. Note that the “licenses and permits” category has an 11.5 percent variance but it totals only $17 million. Compare this with the 4.0 percent in the “current services” category, which accounts for $129 million. There is likely a significant amount of simple categorization issues where entities have chosen to include activity reported in a specific category in the CAFR as “other revenues.” Note that the Controller’s report reflects significantly more revenues in this category than the audited financials. Since the Controller’s questionnaire is more specific 52 than the guidelines presented under GAAP, it is likely that the amounts in the Controller’s report are actually more consistently reported. Overall, however, it is clear that the specific categories do not point to major classification problems for counties. The one category raising a systematic issue is the “revenue from the use of money and property” category. Here relatively large variances are manifest in both absolute and percentage terms. One contributing factor is the reporting of net interest income in the county enterprise activities. Another possible explanation is the interest revenues associated with the missing community facilities districts identified in Chapter 2.9 Under the nonoperating revenue portion of the county enterprise activities is a line for interest revenues. Several counties in our sample reported these interest revenues to the Controller net of interest expenses for the enterprise activities. As a result, both interest revenues and expenses are understated and the resulting variance appears in Table 4.2. Issues Identified In the course of these comparisons, the following four general issues were identified that represent areas where improvements can be made in the reported information for counties: • In some cases, counties excluded both capital projects and debt service funds from the amounts reported to the State Controller’s Office. This accounts in part for the underreporting of “revenues from the use of money and property” identified in Table 4.2. ____________ 9Although this would only represent a small proportion of the overall difference. 53 • As discussed above, some county enterprise activities report interest revenues net of interest expenses in the non-operating revenue section of the county enterprise questionnaire. This underreports the total interest revenues for the entity. • There is no consistency in how contributions from property owners are treated between entities. In some cases, private citizens (usually developers) will contribute property in the forms of streets, sewers, utility lines, etc. to the county in exchange for the provision of a service, such as the formation of a special taxing area. In the case of certain special districts, there are specific instructions on how this should be handled. In the case of at least one county, these revenues were not reported to the Controller because there was not a place or instruction to accommodate such in-kind contributions. This topic will be raised again in the section below discussing consistency among the various reports. • As discussed in Chapter 2, the reporting of Mello-Roos district revenues is very inconsistent. Some counties report portions of the receipts as tax revenues, some report only the interest earned and others do not report them at all. The full effect of these estimates is given in Chapter 2, but a more consistent methodology for addressing these entities should be introduced. In a broader sense, however, the revenues reported for counties in the sample track the audited numbers quite closely. An aggregate variation of 1.2 percent indicates that the overall quality of the county data is quite good. When the above four types of corrections are made to the data, that rate drops to 1.0 percent. 54 Specific Findings: Cities The values reported to the Controller by cities were also compared with their respective CAFRs. In this portion of the study, 52 cities were studied at a detailed level (out of a total of 466 for the state). The cities were selected first by population10 and then at random from the remaining cities.11 The largest 15 cities in the state by population were selected,12 as well as all 14 of the cities in Alameda County.13,14 The remaining 25 cities were selected at random from the rest of the cities in California.15 A subsequent review showed that the sample encompasses the entire state geographically and includes cities of all sizes. See Appendix C for more detail on the city sample. As with the counties, each city’s CAFR was obtained via telephone and compared with the values reported to the State Controller in the Annual Report of Financial Transactions Concerning Cities of California, Fiscal Year 1991–92. The detailed methodology for reconciling the two sets of reports is included in Appendix D. In cases where the research ____________ 10Population and total revenues correlated closely. Since the purpose of the study was to identify errors (if any) in the revenue data, population was chosen as the final criteria instead of revenues reported to the Controller’s Office. 11For purposes of consistency and comparability, the City and County of San Francisco was treated as a city. 12The City of San Bernardino was not included in the final results because the city’s Comprehensive Annual Financial Report was not available in time to be included in the study. 13Originally, this study was to obtain information on all of the entities in two to four counties. After completing a preliminary study of Alameda County, it was determined that there were good reasons to turn to the statewide approach instead. These options and choices are discussed in greater detail in Appendix A. 14The findings presented in this section are generally robust, even when the oversampling of Alameda County is reversed. 15Twenty-eight cities were selected at random, but only 25 CAFRs were received in time to be included in this analysis. 55 team could not explain the variance between the two reports, the cities were contacted for further explanation. Overall Variance The overall variance between the Controller’s and audited reports for the cities in the sample was quite low—only 1.4 percent, as shown in Table 4.3. The Controller reported revenues of $16.30 billion for the 52 cities in the sample and the audited financials included $16.08 billion for an overall variance of $226,575,549. There was one major outlier—the City and County of San Francisco—which contributes significantly to the variance and biases it in a direction opposite of the overall trend. In general, the CAFRs show more revenues than the amounts reported to the Controller. The City and County of San Francisco, however, reports $291,096,334 more to the Controller than is on its CAFR.16 Note that this is more than the Table 4.3 Summary of Comparison of Reported Overall City Revenues for Study Sample, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Audited financial reports Difference Percentage difference Overall Revenues 16,302,293,293 16,075,717,744 226,575,549 1.4% NOTE: See Table F.1 for underlying detail and additional references. ____________ 16There is some reason to expect a higher degree of variation for San Francisco. As a combined government (both a city and a county), its range of activity and institutions are much more complex than a typical city. Furthermore, the inconsistencies between the reporting formats for cities and counties are emphasized because San Francisco’s activities have to be classified into one of the two formats. 56 overall variance reported above. Numerous attempts to obtain explanations of this variance from the City and County of San Francisco were unsuccessful.17 Erring toward the conservative approach, this variance is left in the aggregate comparison. If this outlier is omitted, however, the overall aggregate variance would total only $64,520,785 or 0.5 percent. The overall variance totals for the cities in the sample are presented in Table 4.3. These levels of variance for the cities in the sample are quite low. More than half of the cities had variances of less than 2.5 percent. Seventy percent of the cities show overall variances of less than 5 percent, and 80 percent reflect variances of less than 10 percent. Specific explanations were identified for most of the cities with variances greater than 3 percent. The range of percentage variation is from 0.0 percent to 87.2 percent. Of the 10 cities with variances greater than 10 percent, nearly all of the variance can be attributed to one of three issues: (1) housing authorities are not consistently included in the Controller’s reports; (2) debt service and capital projects activity are not always reported in the Controller’s questionnaire; and (3) some cities generate overhead estimates of general management to various city departments that are then reported as “quasi-external transactions.” This is not done consistently, however. The issues identified in this analysis are discussed in greater detail below. ____________ 17Several offices within the city were contacted to obtain explanations. The staff that responded were unable to reconcile the information and attributed the differences to “differences in the accounting methodologies.” Given the closeness of most other cities in the sample, this explanation does not appear adequate. 57 Two of the three outliers—Irwindale (87.2 percent) and Westmorland (31.1 percent)—have unique circumstances in 1991–92. Irwindale reports $102,500,000 in other revenues associated with a bond issuance undertaken several years before to fund a sports stadium. The bond proceeds were never spent and the bonds were paid off in 1991–92. The proceeds were recognized as revenues in the Controller’s reports but did not really reflect new revenues in 1991–92. They should not have been reported as revenues to the Controller and are consequently recognized as a variance in this analysis. The City of Westmorland apparently does not include water and sewer enterprise revenues when reporting to the State Controller. The variance in the third outlier, Sacramento (27.5 percent), is attributable to the exclusion of its housing authority in the Controller’s report. Categorical/Classification Variance The second dimension of variance is associated with how revenues are categorized. The overall variance for the sample by revenue category is presented in Table 4.4. As discussed above, the report from the City and County of San Francisco introduces an abnormal amount of variance that biases the results. To obtain a better overall sense of the data, therefore, it is useful to look at the data without the effect of the City and County of San Francisco. Table 4.5 presents the same summary as Table 4.4, without San Francisco. This table shows that the application of the classifications in both the Controller’s and GAAP methodologies are fairly consistent. As with counties, many cities’ CAFRs do not have separate categories for “special 58 Table 4.4 Overall Variance in Revenues by Revenue Category for Cities, 1991–92 (dollars unless otherwise indicated) Category Controller’s Report CAFR Percentage Difference Differencea Taxes 5,399,824,479 Special benefit 130,522,897 assessments Licenses and permits 144,852,321 Fines and forfeitures 198,783,522 Revenues from the use of money and property 900,008,611 Intergovernmental funds 1,767,194,569 Current services 6,964,310,811 Other revenues 796,796,083 5,468,054,306 –68,229,827 79,423,852 51,099,045 211,474,171 –66,621,850 146,036,789 52,746,733 931,213,463 2,035,941,895 6,748,566,129 455,007,139 –31,204,852 –268,747,326 215,744,682 341,788,944 –1.3% 39.1% –46.0% 26.5% –3.5% –15.2% –3.1% 42.9% Overall 16,075,717,744 16,075,717,744 226,575,549 1.4% aPercentage difference is the difference as a percentage of the Controller’s reported amounts. Table 4.5 Overall Variance in Revenues by Revenue Category for Cities, Without San Francisco, 1991–92 (dollars unless otherwise indicated) Category Controller’s Report Taxes 4,446,001,944 Special benefit assessments 130,522,897 Licenses and permits 128,546,662 Fines and forfeitures 149,250,412 Revenues from the use of money and property 779,003,664 Intergovernmental funds 1,168,490,709 Current services 5,920,706,513 Other revenues 626,255,158 Overall 13,348,777,959 CAFR 4,529,404,306 Percentage Difference Differencea –83,402,362 –1.9% 79,423,852 51,099,045 155,097,171 –26,550,509 146,036,789 3,213,623 39.1% –20.7% 2.2% 825,904,463 –46,900,799 1,408,195,895 –239,705,186 5,854,659,129 66,047,384 414,577,139 211,678,019 13,413,298,744 –64,520,785 –6.0% –20.5% 1.1% 33.8% –0.5% aPercentage difference is the difference as a percentage of the Controller’s reported amounts. 59 benefit assessments.” If these are rolled up into the tax totals, the overall in the “taxes” category declines to $32,303,317 or –0.7 percent. If San Francisco is included, this drops further to $17,130,782 or –0.3 percent. There appear to be some offsetting classification differences between the Controller and GAAP definitions of “intergovernmental funds,” “current services” and “other revenues.” Overall, the classification differences are quite minor. Issues Identified Numerous issues have been identified in the city portion of this analysis. Some of the issues that affect cities directly are detailed below. Some broader issues that affect the reporting of all public entities will be discussed later in this chapter. • One of the most frequent areas of variance in this analysis is the treatment of city housing authorities and their associated federal and state revenues. These entities are not consistently included in the numbers reported to the Controller for cities. Most cities seem to include them, and both the questionnaire and follow-on discussions with the Controller’s Office indicate that they should be included in the city questionnaires. Yet several major cities do not include them. Additionally, specific instructions must be given to cities that clarify that they should be included. • Many cities failed to report debt service funds and, occasionally, capital projects funds in the amounts reported to the Controller. The Controller’s Office has subsequently instituted a clarification in its instructions to cities that has made it more explicit that these amounts should be included. Compliance with these instructions should be verified and the instructions reiterated if they still are not being followed. 60 • Some cities generated estimates of interdepartmental charges for overhead services provided by the city manager and general administrative departments and reported these amounts as “quasi-external transactions” to the Controller. A careful review of the instructions leaves open such an interpretation. One city indicated that these calculations were produced and used exclusively for the Controller’s survey. It does not appear that such activity is the goal of the specific set of instructions in question, but they should be further clarified. If, however, these cities are meeting the intent of this portion of the survey,18 then it should be made more explicit so that all cities perform the necessary calculations and report the appropriate numbers. • In the case of some city-operated enterprises, interest revenues are often reported net of interest expenses. This is the same issue raised in the county discussion above. It understates both interest revenues and interest expenses. • Occasionally city public financing authority revenues are not reported to the Controller. This happened once in a small city. Since these are not reported elsewhere, however, these public revenues are unreported. • As discussed above, one city used existing resources (proceeds from bonds issued in a prior year) to pay off the outstanding debt. In the process of doing this, it reported revenues to the Controller totaling the amount of the previously issued debt plus earnings. While the reporting of the earnings was appropriate, the reporting of the bond revenues was inappropriate. • Special benefit assessments are inconsistently and poorly reported. The definition provided is ambiguous and variously ____________ 18If this choice is open to decision, it is recommended that the Controller not institute such a calculation. These are merely accounting entries and do not represent substantive or even directly programmatic revenues of any kind. 61 interpreted. In one city, street and lighting assessments were not reported to the Controller. In another case, one city reported a significant amount of debt proceeds as special benefit assessment revenues. Concurrently, the actual special assessments to pay off these bonds were reported as revenue as well. As a result, these bonds were double counted. Recent investigations by numerous state and private entities, motivated by the introduction of Proposition 218, have found that the data reported in this portion of the survey are at best incomplete. A more explicit set of instructions needs to be provided to make this information as complete and relevant as possible. • Finally, several cities omitted enterprise activities that were consistently reported on other cities’ surveys. One city omitted its transit enterprise, while another omitted both its water and sewer enterprises. As a result, these revenues were excluded from the final reported numbers and were not picked up elsewhere. By and large, however, the data on cities are excellent. Only the housing authority, capital projects and debt service fund issues were pervasive across several entities (three or more). The remaining issues occurred in only one or two cases (out of 52) and, while worthy of mention, were not common. Specific Findings: School Districts The Controller provides summary information for the state’s 1,005 school districts in California in its report entitled Annual Report of Financial Transactions Concerning School Districts of California. The State Superintendent of Public Instruction is required by state law to provide summary information to the State Controller’s Office for 62 compilation of this report.19 The information in this report is provided in a very aggregated form—listing revenues and expenditures by category only for the whole state and by entity category.20 Total revenues and expenditures are provided, unaudited by the State Controller, for each district and K–12 entity in the state. These revenues are reported in Table 4.6. The categories included in this report conform to the account classifications prescribed in the California School Accounting Manual, 1992 Edition. As such, they are not consistent with any of the other reporting conventions included in the Controller’s Annual Financial Transactions report series. For example, the “state aid” presented in Table 4.6 is largely “aid from other governmental agencies”—the state government in this case. The “local taxes” category represents local Table 4.6 Revenues for School Districts in California, 1991–92 Revenue Source State aid Local taxes All other Total federal Other state Other local Other financing sources Total revenue Revenues $11,205,074,937 5,309,829,960 1,607,079 1,955,746,196 4,498,593,156 2,059,578,526 2,115,733,687 $27,146,163,541 SOURCE: Controller’s Annual Report of Financial Transactions Concerning School Districts, Fiscal Year 1991–92, p. IX. ____________ 19These requirements are specified in Section 53892.1 of the government code. 20The categories separated in the report are “School Districts,” “County Schools,” and “Joint Powers Agencies.” 63 property tax revenues. For consistency and clarity, these differences should be addressed. Another aspect of this report is that it provides only summary information for school district revenues and expenditures. There is no entity-by-entity presentation of information as is done with all other public entities in the state. Because school districts account for a major portion of local government spending in the state, this absence of detailed information should be addressed. Finally, there is no separate reporting of community college districts in the Controller’s reports. While the California Postsecondary Education Commission does publish some aggregate information for these entities, there is no district-by-district reporting of the type described above. Inasmuch as these community college districts, like elementary and secondary school districts, are local public entities with locally elected boards and local voter accountability, their fiscal information should be directly available to local voters, analysts and decisionmakers. Specific Findings: Dependent Special Districts Dependent special districts are those entities whose governance and/or finances are controlled by the governing body of another entity. Most often these entities are controlled by a city council or a county board of supervisors, although there are occasions when these entities are controlled by school boards or the governing boards of independent special districts. In 1991–92, the Controller obtained and reported financial information on 1,856 dependent special districts in its Annual Report on Financial Transactions Concerning Special Districts, 1991–92. 64 Obtaining audited financial information separately for these entities is not possible. Because CAFRs often include these entities, however, it is possible and even necessary to include them as part of the reconciliation of their parent entities. As a result, this study implicitly reviews numerous dependent special districts in the course of reconciling cities and counties. This study also performed a “spot check” on redevelopment agencies to test whether they, in fact, do rely on audited numbers, as required by law, in their submissions to the State Controller’s Office. Dependent Entities Reviewed as Part of City and County Reconciliations In the course of preparing the results for the counties and cities described above, it was necessary to aggregate dependent entities into the reconciliations to provide comparable entities between the two reporting formats. This process means that the above documented portion of the study effectively reviewed a significant number—more than onefourth—of the 1,856 dependent special districts listed in the Controller’s annual reports. In the course of reconciling the 52 cities presented in the immediately preceding section, 43 out of the state’s 491 redevelopment agencies were reconciled as part of the process. These redevelopment agencies are reported separately in the Annual Report of Financial Transactions Concerning Redevelopment Agencies of California, Fiscal Year 1991–92. This is in addition to the specific review of redevelopment agencies described later in this chapter. In addition, 11 other dependent entities were reviewed as part of the city reconciliation. This would include any entity explicitly included in a 65 city’s CAFR and showing activity in the Annual Report of Financial Transactions Concerning Special Districts of California, Fiscal Year 1991– 92. These 11 entities include the Estero Municipal Improvement District, the Livermore Recreation and Park District, the Modesto Municipal Sewer District, the Napa County Housing District, the Parking Authority of Oakdale, the Redding Area Bus Authority, the Automated Regional Justice Information System, the San Diego Data Processing Corporation, the Ventura County Water District, the Simi Valley Sanitation District and the Yucca Valley Recreation and Park District. Reconciling the 15 counties incorporated many more districts— nearly 500. The specific types of entities are reported in Table 4.7. Table 4.7 Dependent Entities Included in County Reconciliations County Alameda Humboldt Los Angeles Orange Placer Plumas Riverside Sacramento San Benito San Bernardino San Diego San Luis Obispo Santa Clara Siskiyou Tehama Total Redevelopment Agencies 0 0 1 1 1 0 1 1 0 1 1 1 0 0 0 8 Enterprise Districts 1 1 3 2 2 1 5 4 1 2 3 3 2 3 0 33 County Service Areas 16 1 1 9 5 6 67 2 18 54 77 13 0 1 0 270 Other Dependent Districts 4 12 76 4 7 6 4 15 1 15 14 9 5 3 5 180 Totals 21 14 81 16 15 13 77 22 20 72 95 26 7 7 5 491 66 Each of these entities was obtained from another source in the Controller’s annual reports and added to the basic amounts included in Table 6 of the Annual Report of Financial Transactions Concerning Counties of California, Fiscal Year 1991–92. The redevelopment agencies, as was the case with those encountered in the city reconciliations, were obtained from the Annual Report of Financial Transactions Concerning Redevelopment Agencies of California, Fiscal Year 1991–92. The county enterprise activities were obtained from Tables 10 through 13 of the Annual Report of Financial Transactions Concerning Counties of California, Fiscal Year 1991–92. The county service areas were all obtained from the appropriate tables in the Annual Report of Financial Transactions Concerning Special Districts of California, Fiscal Year 1991–92. In total, 51 or 10.4 percent of the 491 redevelopment agencies in the state were reviewed in this process, and 483 or 26.5 percent of the state’s 1,856 dependent special districts were reviewed. While the precision of these reconciliations is less than that of the independent special districts described below, they all received a significant level of review. Because of the aggregation issue, no summary statistics can be provided for the values reported for these entities. Additional Reviews of Redevelopment Agencies In addition to reviewing 51 redevelopment agencies as part of the general review process, 10 redevelopment agencies were specifically reconciled against their audited financial statements. In each case, the amounts reported to the Controller coincided precisely with those found in their comprehensive audited financial reports. This can be largely attributed to the 90-day deadline for completing annual audits of 67 redevelopment agencies, which coincides with the filing deadline for the Controller’s questionnaire. Since these two deadlines are the same, the information provided is typically the audited information.21 Specific Findings: Independent Special Districts Independent special districts are special local government entities formed under a variety of sections of the state code to perform a wide range of functions. Individual independent special districts typically have narrowly defined areas of responsibility, such as providing transit, airport, hospital or water services. Others serve financial needs, such as providing insurance services or capital resources, and yet another category of these districts allows the joint provision of services among multiple cities and counties, such as library districts. Within this framework, the process of reviewing the data was complex. As was the case with cities and counties, audited financial statements were obtained from the largest special districts within each functional category as well as a random selection from the remaining districts. The specific criteria and a list of the entities sampled are included in Appendix C. The revenues reported in these audited financial statements were compared with those reported in the Controller’s report. The specific methodologies and criteria used during the review and reconciliation process are included in Appendix D. The findings of these reconciliations are presented in the following portion of this chapter. It is organized by the functional types of the ____________ 21In the case of counties, cities and special districts, the Controller’s deadline remains the same, while the audit deadline is 12 months after the fiscal year end. It is subsequently less common for the information provided to the Controller to be based on the audited data. 68 special districts involved. Summaries will be provided for each of the following types of special districts included in the Controller’s Annual Report on Financial Transactions Concerning Special Districts: hospital districts, transit districts, transportation planning agencies, water and sanitation districts, other enterprise districts (airport, electrical utility and harbor port districts) and non-enterprise districts. An overview of the findings for special districts will then also be provided. As was the case above for counties and cities, a detailed table corresponding to each of the summary tables is provided in an appendix of this report (in this case Appendix G). This section will focus on overall variance and not include the comparison along categorical lines that was provided for the more complex general government entities above. Special districts are fiscally much simpler entities and typically have only two general revenue categories: operating revenues and non-operating revenues. Operating revenues are those revenues that the districts obtain as a direct result of the specific services they provide. For an airport this would include landing fees, while in a water district this may include connection fees, sales of water and even sales of sewer/sanitation services. This most closely corresponds to current service charges in the general government entities, and, in fact, when there are directly operated municipal equivalents to enterprise districts, these revenues are included in the current service charges section of their operating statements.22 Nonoperating revenues typically represent all other categories of revenues, ____________ 22In some non-enterprise special districts, there is a category explicitly called “charges for current services.” In these cases, these revenues are treated as operating revenues. 69 including tax and intergovernmental revenues.23 For example, in most cases, but not all, any property taxes received by the special district will be represented as a separate entry in this category. Differentiating between these two categories within this analysis does not provide any new insights to the findings of this research. As the following tables will show, the level of variation found within these special districts is very low, and if there is any variation in the operating revenue category it is almost always offset dollar-for-dollar by a complementary difference in the non-operating category. This points to a simple classification issue that is likely more due to the lack of detail included in the comprehensive annual financial report than any errors or inconsistencies in reporting to the Controller’s Office. Hospital Districts Hospital districts are local districts that usually govern and oversee a major hospital in a municipality, county or region. These districts are paralleled in the county context by the 26 dependent hospital enterprise activities that were included in Table 10 of the Annual Report on Financial Transactions Concerning Counties. Since these hospital entities are dependent on the county and were subsequently included in the dependent entity reconciliations discussed above, this section focuses exclusively on the 13 independent special districts included in the Annual Report on Financial Transactions Concerning Special Districts. The overview of the findings for hospital districts is provided in Table 4.8. As this table shows, the overall variation between the two reports for hospital reports was minimal. In fact, the largest individual ____________ 23This is “typical” because some entities also include narrow categories of tax revenues and intergovernmental transfers as operating revenues. 70 Table 4.8 Summary of Comparison of Reported Overall Hospital District Revenues for Study Sample, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Audited Financial reports Difference Percentage difference Overall Revenues 1,484,490,882 1,487,334,570 2,843,688 0.2% NOTE: See Table G.1 for underlying detail and additional references. variant on a percentage basis was Mt. Diablo Hospital District, with only a 1.8 percent difference. The total sample represented 63 percent of the total hospital district revenue in 1991–92. Although very few of the hospitals had reports to the Controller that exactly matched their audited financial statements, the difference was not large: On average, the audited revenues were 0.5 percent higher or lower than the revenues reported to the Controller. There did not seem to be any systematic differences between the Controller’s reports and the CAFRs except in one difference of categorization of net revenue: The Controller counts the “provision for bad debts” as negative income, whereas the CAFRs universally recognize it as an expense. Transit Districts Transit districts are local districts that govern and oversee bus and rail transit systems in a municipality, county or region. These districts are paralleled in the county context by the 41 dependent transit enterprise activities that were included in Table 12 of the Annual Report 71 on Financial Transactions Concerning Counties. Since these transit districts are dependent on the county and were subsequently included in the dependent entity reconciliations discussed above, this section focuses exclusively on the 13 independent special districts included in the Annual Report on Financial Transactions Concerning Special Districts. The sample for transit districts included 18 of the 54 transit districts in the Controller’s report. This sample was chosen first by selecting the 12 districts with the highest revenues. The remaining six districts were either part of the original Alameda sample,24 included in the CAFRs of related transportation entities or selected as part of the larger sample. The revenues of these 18 districts represent nearly 94 percent of all transit district revenues in the state. As Table 4.9 shows, the total revenues reported in the Controller’s report were extremely close to those given by the CAFRs. The aggregate results shown in Table 4.9 do hide some variation between the districts, some of whose CAFRs show higher and some lower Table 4.9 Summary of Comparison of Reported Overall Transit District Revenues for Study Sample, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Audited financial reports Difference Percentage difference Overall Revenues 1,696,997,635 1,699,606,416 2,608,781 0.2% NOTE: See Table G.2 for underlying detail and additional references. ____________ 24The initial phase of this study focused on all of the public entities in Alameda County. See Appendix C for a more detailed discussion. 72 numbers than the Controller’s report. Although 10 districts showed no difference between the Controller and audited financial statements, the districts’ audited numbers on average were 1.3 percent higher or lower than the Controller’s report. The largest difference was in the San Mateo Transit District, whose report to the Controller seems to have been 9 percent too low for two reasons: The district did not report a gain on the sale of assets, and the Controller’s questionnaire included net, rather than gross, interest figures. Transportation Planning Agencies The 81 entities included under the transportation planning agency banner are either transportation planning agencies, agencies that have a transit planning function or agencies that have the authority to expend locally raised sales taxes on transportation needs. As such, this category also includes regional organizations of governments, such as the Southern California Association of Governments, as well as county traffic authorities, county transportation commissions, county transportation authorities, service authorities for freeway emergencies and selected others. The 19 entities included in the sample account for 93 percent of all transportation planning agency revenues in the state. Unlike the other categories included in this section on special districts, this category of entities includes a portion of the range of activity reported in the Controller’s Annual Report on Financial Transactions Concerning Transportation Planning Agencies of California. The transportation planning agencies are included here, in large part, because of criticisms that the amounts reported in the transportation and transit-related reports should represent particular areas of concern. 73 The findings for the sample of entities in this category are presented in Table 4.10. As this table shows, the overall level of variation is low, totaling only 4.6 percent, but higher than other special districts. Much of this variation is attributable to a single entity within the sample, as discussed below. As noted above, transportation authorities or planning agencies provide planning oversight for the transit districts in a region. One main function of the transportation authorities is to distribute transportation planning dollars from the state and other local entities, such as the local 1/4 cent sales tax (the Local Transportation Fund, or LTF) and State Transit Assistance (STA) funds. In 1991–92 most planning agencies did not count these funds as their own revenue, because the funds were passed on entirely to the local transit authorities. Nevertheless, the Controller requires that the entities report the LTF and STA funds in Table 4.10 Summary of Comparison of Reported Overall Transportation Planning Agency Revenues for Study Sample, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Audited financial reports Difference Percentage difference Overall Revenues 2,992,542,356 2,853,440,046 139,102,310 4.6% NOTE: See Table G.3 for underlying detail and additional references. 74 their report of fiscal transactions.25 As a result, these pass-through funds were included in the Controller’s report, but there was no audit information to verify the accuracy of the amount of these funds. Similarly, the interest figures in the CAFRs were lower than in the Controller’s report because the CAFR figures did not include the interest on the LTF and STA money. If the entities’ estimates of the pass-through funds are accurate, the difference between the audited numbers and the Controller’s report are minimal—with one major exception. The San Bernardino Association of Governments (SANBAG), which had a 60 percent difference between the two amounts, reported 137 million dollars more to the Controller than showed up on its annual audit. The difference was largely attributable to the inclusion of revenue note and bond proceeds as revenues in the Controller’s report. Without SANBAG, the overall difference was only $2,226,417 or 0.1 percent. The Southern California Association of Governments also shows a large percentage variance (7.5 percent). This variance is due to the inclusion of internal service fund revenues in the numbers reported to the Controller, which should have been excluded.26 ____________ 25While it may seem like double counting to include the LTF and STA dollars here, it is necessary in order to show the complete picture of fiscal transactions in California. Since transportation agencies are the first agencies to actually receive these monies, and in light of the explicitly stated public policy goal of reporting these funds as revenues to these agencies, they should be reported as revenues at this point. Subsequent agencies that receive LTF and STA dollars as a result of their distribution by a transportation planning agency should report these revenues as intergovernmental transfers. 26Internal service funds are an accounting technique whereby administrative costs for various activities are billed out to various other departments. For example, a computer support department may bill the planning department for the installation of a computer system. The computer department would thus show revenue in its internal service fund and the planning department would have an expenditure. While useful for internal cost accounting purposes, the transaction reflects an internal accounting allocation mechanism rather than a real increase in the city’s overall revenues. As such, 75 Water and Sanitation Districts This category of special districts includes all water, sewer and sanitation districts. These entities provide water, water treatment and waste disposal services to customers. They have been combined here for convenience of presentation and are actually reported separately in Tables 22 and 23 of the Annual Report on Financial Transactions Concerning Special Districts for 1991–92. California’s 806 water, water utility, irrigation and sewer districts make up the plurality of special districts in California, and, in fact, the first independent special districts in the state were water districts. Most of these are independent.27 The sample included 30 water, sewer, irrigation and sanitation districts including the 10 largest such districts in the state. Revenues within this category of special districts are highly fragmented, but the sample includes more than 26 percent of all the reported revenues for these entities in the state. The summary of the findings for the sample in this category is presented in Table 4.11. More than half of the districts seem to have used audited numbers for their final report to the Controller. When the Controller’s report was not exactly the same as the audit, there was no clear reason for the discrepancy, although the difference was almost always concentrated in the “other” category. The one major exception to this was the Oro Loma Sanitary District, which reported nearly $8 million more in revenues to the Controller ____________________________________________________ these funds are and should systematically be excluded from the reported revenue amounts. 27Four water and sanitation districts are included as county enterprise activities in the Annual Report on Financial Transactions Concerning Counties. Numerous cities have dependent water and sanitation enterprises. 76 Table 4.11 Summary of Comparison of Reported Overall Water and Sanitation District Revenues for Study Sample, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Audited financial reports Difference Percentage difference Overall Revenues 1,443,508,721 1,430,320,834 13,187,887 0.9% NOTE: See Table G.4 for underlying detail and additional references. than showed up on its audited financial report. The difference is attributable to the inclusion of a balance in an equity account as revenue when only the change in that equity position should have been included.28 Without this district’s variance, overall variance would be $5,205,407 or 0.4 percent. Airport, Electric Utility and Harbor and Port Districts The entities included in this subsection are again aggregated to simplify the presentation of the findings. The district descriptions are quite self-explanatory in terms of the function of each district. Airport, electric utility and harbor and port districts are included in Tables 18, 19 and 20, respectively, of the Annual Report on Financial Transactions Concerning Special Districts for 1991–92. The overall summary findings for these three types of entities are given in Table 4.12. ____________ 28Oro Loma Sanitary District is a participant and equity holder in the East Bay Dischargers’ Authority. Its equity stake in the Authority is approximately $8.7 million, with an annual change (a loss) of $100,000. Instead of incorporating the year’s flow (the loss) in its report to the Controller, the district included the stock (the full value of its stake in the authority). 77 Table 4.12 Summary of Comparison of Reported Overall Airport, Electric Utility and Harbor and Port District Revenues for Study Sample, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Audited financial reports Difference Percentage difference Airport Electric Utility Harbor and Districts Districts Port Districts Combined 12,324,045 1,203,210,790 132,322,371 1,347,857,206 12,077,250 1,203,658,374 132,322,372 1,348,057,996 246,795 447,584 1 200,790 2.0% 0.0% 0.0% 0.0% NOTE: See Table G.5 for underlying detail and additional references. The differences for each district type are extremely small. All of the variation in the airport districts could be attributed to the Monterey Peninsula Airport District, which had an overall variation of 5.2 percent. No explanation was available from the district for this variance. The electric utility and harbor and port districts showed no variance. Non-Enterprise Districts Non-enterprise districts—the largest category of district—are those government entities that are not run on a for-profit basis. These entities typically receive the largest part of their funding from taxes and intergovernmental grants rather than through service charges. Examples of non-enterprise districts include self-insurance authorities, cemetery districts, flood control districts, parks and recreation and library districts. The sample of non-enterprise districts consisted of 64 districts and included entities of every functional type. The summary of the findings for non-enterprise districts is included in Table 4.13. Although more than half of the districts seem to have used audited numbers in filling out the final report to the Controller, the Controller’s 78 Table 4.13 Summary of Comparison of Reported Overall Non-Enterprise District Revenues for Study Sample, 1991–92 (dollars unless otherwise indicated) Description Controller’s reports Audited financial reports Difference Percentage difference Overall Revenues 595,653,711 597,529,625 1,875,914 0.3% NOTE: See Table G.6 for underlying detail and additional references. report was 1.2 percent lower overall than the total given by the audited financial statements. This total does hide some variation among the minority of districts whose report to the Controller did not match their audits. On average, the audits of the non-enterprise districts were 1.1 percent higher or lower than the Controller’s report, and the average difference was $107,563 on average revenues of more than $9 million. In most cases where there was variation, all or most of the difference was due to excluded funds. That is, while the districts are supposed to include all revenues no matter what purpose they are reserved for, some districts neglected to include revenues earmarked for debt service, capital projects, or other special projects in their report to the Controller even though the forms explicitly ask for all funds. These excluded funds accounted for 77 percent of the difference between the Controller’s reports and the audited financial statements. Special Districts: Overall Findings Overall, the accuracy of the special district data was quite remarkable. Table 4.14 summarizes the findings for special districts 79 Table 4.14 Summary of Comparison of Reported Overall Special District Revenues for Study Sample, 1991–92 (dollars unless otherwise indicated) Special District Group Controller’s Audited Financial Report Report Difference Hospital districts 1,484,490,882 Transit districts 1,696,997,635 Transportation planning agencies 2,992,542,356 Water and sanitation districts 1,443,508,721 Airport, electric utility and harbor and port districts 1,347,857,206 Non-enterprise districts 598,229,191 1,487,334,570 1,699,606,416 2,853,440,046 1,430,320,834 2,843,688 2,608,781 139,102,310 13,187,887 1,348,057,996 598,105,105 200,790 1,875,914 Overall Sample Totals 9,561,050,511 9,416,289,487 144,761,024 NOTE: See Tables 4.8 through 4.13 above. Percentage Difference 0.2% 0.2% 4.6% 0.9% 0.0% 0.3% 1.5% overall. As it shows, the overall variation identified in the 156 districts reviewed was only 1.5 percent. The revenues presented in the two series correspond closely. Most of the difference is accounted for by one entity in the transportation planning agencies group, the San Bernardino Association of Governments. If it is omitted from these findings, the total variance drops to $7,885,131 or 0.1 percent of the total. Based on these findings, the special district data in the Annual Report on Financial Transactions Concerning Special Districts and Annual Report on Financial Transactions Concerning Transportation Planning Agencies are very accurate—probably even more so than the amounts reported for cities and counties.29 ____________ 29As discussed in the individual sections, this is likely because of the use of audited information by many special districts to prepare their Controller’s questionnaires. 80 Preliminary Findings: Debt Data Are Much More Problematic As a quick and preliminary measure of the quality of the debt data included in the Controller’s reports, the study tracked one narrow category of debt activity—debt proceeds. The goal was to identify how well the debt information reported to the Controller tracked with the detailed information included in the CAFRs. These findings are presented in Table 4.15. As this table shows, the amount of variation between the two sources is considerable. A large proportion of the variance in the city sample is due to Los Angeles, which had a difference of $782 million—a problem that has already been identified by the state. The overall variance within the city sample drops to 21.2 percent if Los Angeles is excluded. Debt and debt proceeds were not the focus of this study, but it is important to note that this preliminary look at the debt numbers points to significant reporting differences, in complete contrast to the revenue reporting. As will be discussed in the final chapter, additional research is recommended to address this problem. Table 4.15 Overall Variance in Other Financing Revenues, 1991–92 (dollars unless otherwise indicated) Group Controller Audited Report Percentage Differencea Differenceb Cities sampled 1,854,534,605 1,263,528,344 591,006,261 31.9% Counties sampled 815,876,077 682,453,027 133,423,050 16.4% Sample total 2,670,410,682 1,945,981,371 724,429,311 27.1% aDifference is presented in absolute terms. Accordingly, the overall total listed does not equal the sum of the column, but the difference between the overall amounts reported. bPercentage difference is the difference as a percentage of the controller’s reported amounts. 81 5. Summary, Recommendations and Conclusions This chapter presents a summary of the findings of this study as well as the implications of those findings. These implications are approached from two different directions. First, a practical approach is taken to the data, and a series of recommendations is presented to address parts of each of the quality issues raised in Chapter 1—timeliness, comprehensiveness and accuracy. Second, a broader view of the implications of the findings is taken. This section will present specific implications of the findings regarding the quality of the data to policymakers and analysts interested in California state and local policy. The credit for the quality of the data must be shared between the Controller’s Office and the myriad of local governments who supply the Controller with accurate information. 83 Summary of Findings A general statement of this report’s findings is that the Controller’s data are very good. As one colleague commented, “I’m increasingly impressed with the quality of the Controller’s data.” This is not to say that the data are perfect—there are clearly areas where the consistency of the reporting can be significantly improved, as will be discussed in this report’s recommendations below. Comprehensiveness of the Data The data were found to be quite comprehensive. However, it is clear that community facility districts are typically not included. A survey of a sample of the CFDs in the state found that only between 20 and 35 percent of these entities are included in their parent entity’s report to the State Controller’s Office. The balance are not included. A census of the 233 CFDs in existence in 1991–92 found that the combined revenues of all of these entities totaled only $283.5 million. In a local government sector whose revenues totaled nearly $95 billion, this represents a trivial source of error—0.3 percent—even if all CFDs were excluded. Since some proportion of these districts is included in the reported information, the actual level of error introduced by the omission of CFDs is even lower. Note, however, that if one wishes to focus on these particular entities for a specific policy inquiry, the data are wholly inadequate. Beyond CFDs, an effort was made to identify other entities that may be missing from the Controller’s reports. After an exhaustive (more than 7,000 entities) comparison of local agency formation commission and California Debt Advisory Commission lists to the entities reported in the Controller’s reports, three non-CFD entities were found to be missing 84 from the Controller’s entity list.1 The revenues for these entities totaled approximately $0.5 million—an inconsequential level of difference. It can be concluded from this exercise that the Controller’s data are comprehensive—that they include the full range of public entities that generate revenues at the local level. At the same time, one major category of entity is generally missing from the revenue data, but its absence’s effect on the overall revenue picture is negligible. Accuracy of the Data Along the accuracy dimension, the data test well. When the amounts reported overall by individual entities were compared with the amounts reported by the State Board of Equalization for sales and property taxes, the variance was very low—ranging from 0.0 percent to 5.1 percent. On a county-by-county and city-by-city basis, the variation was also quite low. Furthermore, this average was heavily influenced by a few outliers. Without these outliers, the difference between the two sets of information essentially disappeared. In comparing the information reported to the Controller with audited financial information, the overall totals were even better. For counties the variance was 1.2 percent, for cities 1.4 percent and for special districts 1.5 percent. At the individual comparison level, the absolute average percentage differences for each were 1.4 percent, 3.9 percent,2 and 0.9 percent, respectively, showing that aggregation implicit in the overall totals reflects the overall trend in the data. In general, these ____________ 1The methodology used allows for the existence of additional non-reporting entities, but a review of the candidates indicates that their overall effect would also be inconsequential. 2This value excludes the three outliers, Irwindale, Westmorland and Sacramento, as specifically discussed in Chapter 4. 85 data are highly accurate in their portrayal of the revenues raised and received by these entities. In summary, this study has found the revenue data included in the Controller’s reports to be quite correct and complete. There were, however, areas where the process and specifics could be improved. The next section discusses some specific suggestions for improving the data to render them as useful, complete and correct as possible. Recommendations These recommendations arise as a result of this exhaustive review of the State Controller’s reports. These recommendations are generally organized around the areas where concerns were raised: timeliness, comprehensiveness and accuracy and provide specific suggestions about how the information provided could be strengthened along each of those specific dimensions. Timeliness As discussed in Chapter 1, this report did not provide an analysis of the timeliness of the data. The issue is key, however, to the utility of the data to state decisionmakers and policy analysts. For this reason, the following two suggestions are offered in hopes of making the availability of the information in the Controller’s report much more timely. • Institute Internet/Web-based submission. The Controller’s Office should move to a more direct submission technology for the information in the questionnaires. A direct World Wide Web–based submission form would go a long way toward this goal. While a few entities are now submitting their questionnaires on diskette, an on-line form could significantly enhance the availability and ease of electronic submission for 86 local governments. The general and increasingly widespread availability of access to the Web coupled with its easy information transfer capabilities make it an ideal medium for transferring information between local governments and the State Controller’s Office.3 Furthermore, such an approach could facilitate the implementation of a broader, more standardized format as proposed in the discussion under accuracy below. This approach would also remove the data entry and quality control of that data entry from the Controller’s Office, significantly accelerating the preparation of the data for review and freeing up valuable resources for the review stage. The data would also immediately and automatically be in a consistent format across entities in the same category. • Make data immediately available. Beyond getting the information to the Controller’s Office more quickly and in a more consistent format, the data could be made immediately available to policy analysts and decisionmakers—even before the data have been reviewed.4 This would make the bulk of the information available in an unreviewed format 90 days after the fiscal year end—a much more useful timetable than more than a year later. Granted there is the prospect that unreviewed and unaudited data will contain errors, but the timeliness of the information could more than offset the risks associated with ____________ 3Typically, one of the greatest concerns associated with using Web submission techniques is security. This is not an issue in this case because all of the reported data are already in the public domain. 4Unreviewed data provided in this format would be so marked, indicating to users and consumers that it had not been reviewed. As the information is reviewed by the Controller’s office, the notation could be changed to reflect that development. In such a strategy, it is recommended that large and complex entities, which may be more likely to significantly affect policy choices, should be reviewed first. 87 mistakes by specific entities.5 To provide a metric of the reliability of a specific entity’s unreviewed data, an additional field could be provided by the Controller’s Office for each entity that reports the magnitude and types of information that were corrected in the prior year’s report.6 Comprehensiveness The Controller’s data are quite comprehensive in their reporting of revenue activity in the state. There are two recommendations, however, that arose as a result of this analysis: • Provide specific instructions for Mello-Roos districts. Specific instructions and questions should be provided to assure the inclusion of community facility districts in the Controller’s reports. While some of these entities are already reported, the vast majority are not. It is critical, inasmuch as decisionmakers and analysts care about this financing entity, to have accurate and complete information regarding their activity. • Create a mechanism for identifying new entities. The Controller’s Office should also implement a watchdog-type mechanism for identifying new entities that would not normally be captured under the current reporting scheme. While this study did not find many at this time, the increased constraints ____________ 5Note that this study reviewed the published reports from the State Controller’s Office and subsequently included all of the corrections that that office identified in the course of their review of the data. Information on the quality of the data before this review was not available, and clearly the prevalence of a need for significant review and correction of the data should be a factor in the implementation of this recommendation. Note, however, that the immediate publication of information could serve as an incentive for local entities to provide the information more accurately as its unreviewed quality comes under greater scrutiny. 6Also note that it is recommended that the reported information be left on-line in subsequent years, preferably in a side-by-side format, so that changes over time can be reviewed easily. 88 and pressures on local governments will assure more creative behaviors in the future. The Controller should have an eye toward making sure that these entities report their activity as they come into existence. Accuracy The general content of the Controller’s reports was found to be highly accurate. However, there were several areas where specific recommendations for improvements can be made. These are detailed below. • Provide more specific instructions and follow-up regarding capital project funds, debt service funds and housing authorities. Several types of activity were not consistently reported by all entities to the Controller’s Office. Often the specific variation identified between the Controller’s report and the audited financial report could be tracked to the noninclusion of either capital projects funds, debt service funds or housing authorities, or some combination of the three. Even in the case of counties, where separate columns are provided for debt service and capital projects funds, these revenues were not always included. It is recommended that these instructions be further strengthened and a systematic effort be made to ensure that all are included. • Expand and clarify reporting of special assessment districts. One category of activity for which the inadequacy of reporting has come to the forefront in light of current events is special assessment districts. Researchers and analysts trying to assess the effects of Proposition 218 on the November 1996 ballot found the data not up to the task. The reporting on these increasingly (in recent years) popular revenue-generating 89 arrangements is spotty at best and does not seem to capture the full range of activity that occurs. Because of the lack of a third-party information source, this study did not directly quantify the revenues that should be classified and reported as special benefit assessments. Discussions with other analysts looking at special benefit assessments indicate that the reported revenues are very low. It would seem from this research, however, that these revenues are not missing; they are simply included as property taxes and not separated out as special benefit assessments. Specific instructions and procedures should be established to track these activities. • Expand the reporting of school district information. The Controller’s report on school districts in California is very cursory. It reports revenues and expenditures only at the most aggregated levels—providing revenue and expenditure detail by type and category only at the statewide level and district-specific revenues and expenditures only in total. Inasmuch as school districts represent one of the largest categories of local revenues and expenditures, the detailed information on these entities should be published and made available for public accountability, review and discussion. It is recommended, therefore, that the Controller provide in its annual report entitylevel detail for each district, county office and joint powers agency in the state. This would require an expanded provision of information by the California State Superintendent of Public Instruction. The Department of Education could provide this information in a format immediately available for paper or electronic publication. It would also be helpful to include the specific numbers used by the state for calculating the Proposition 98–required expenditures each year. Such a service would provide a common source of information for analysts and decisionmakers. 90 • Provide detailed fiscal information for community college districts. Community college districts are unique entities in the state. Even though they receive their funding largely under the Proposition 98 formula, they participate as part of the state’s postsecondary education sector. Unlike the other public members of the state’s postsecondary education sector, they are organized and governed locally. Inasmuch as this research envisions the Controller’s data as a centralized database for both informational and accountability purposes, the information on these largely local entities should be as available as the local elementary school district, city, county or mosquito abatement district. As a result, it is recommended that these districts’ financial activity be reported in the Controller’s data system. • Establish a consistent report format for all categories of entities. Currently, there is significant variety in the way that the diverse types of entities report their specific information. As discussed regarding school districts above, but true on a much broader basis, the requirements and reporting structures associated with individual entity type are almost always determined by the specific institutional and historical context of the reporting entity and less with a view toward comparability and public accountability and with even less of a view toward utility for decisionmakers. As a result, the various reports are very difficult to compare and use—both across government entities and sometimes even within the same report. In the Annual Report on Financial Transactions Concerning Special Districts, for example, transit districts use a format driven by their reporting requirements for the U.S. Department of Transportation, while hospital districts use the requirements specified in the (California) Hospital 91 Disclosure Act.7 Although there are some distinct advantages to making the reporting process as convenient as possible for reporting entities, some minor changes could be implemented to improve the usefulness and comparability of the data. To provide a specific example of how this lack of comparability affects reporting, there is variation in how different entity types handle the major category of property tax revenues. Some have a separate line indicating property tax revenues, while others include them under “other non-operating revenues.” There are also differences between the reporting format for enterprises that are considered part of the county general government—and are reported in the enterprise tables in the Controller’s annual report on counties—and those that are separate dependent and independent districts listed in the special districts report. Even the detailed categories between such general governments as cities and counties vary somewhat. For example, the sale of a fixed asset in a city is included under “other revenues.” In a county it would be included under “other financing sources.” The reporting format for school districts, because it comes from a different source, is completely different. It would be invaluable to have consistent reporting formats between entities for comparability purposes. It would also be useful to have the major tax revenue streams—such as sales and property taxes—explicitly identified for each entity. Such an approach would also provide a quick measure of the quality of the information provided. More generic information, such as addresses, telephone numbers, miscellaneous demographic information (or hyperlinks to associated databases in an Internet-type approach) and contact people, would also be useful additions to these databases. ____________ 7This act commences with Section 440 of the Health and Safety Code. 92 In conjunction with a universal report format, one may wish to expand the governance information for each dependent entity in the database. In most cases at the current time, the governance information is limited to the parent entity’s generic category (e.g., county, city, special district). It would be far more useful to include precisely who the parent entity is. One could include an identifying code or, in an Internet-type application, a hyperlink to the parent entity. This would make the information far more useful for policy analysts and decisionmakers who would like to perform the type of aggregation that had to be done manually for the purposes of this study. Overall Issues To Be Considered in Implementation Overall, these recommendations represent minor enhancements to the quality of the data currently available. There are issues to be considered in the implementation of any and all of these ideas, however. There are effects on the backward comparability of the data as well as changes in the associated workload for both the Controller’s Office and reporting entities. Any changes that affect the data reported in the Controller’s study will clearly affect their direct comparability to prior years. This happens every time improvements are made to data systems. The effects in this case, however, would likely be minimal, as the low variances identified in this report would suggest. Most of the recommendations center around properly classifying activities and providing consistent reporting formats between entities. The expanded inclusion of Mello-Roos districts could be included as a separate line item, minimizing the effect their inclusion would have on other revenue categories. The expanded and more consistent reporting 93 of capital project funds, debt service funds and housing authorities would mean that the activity reported would more closely correspond to what is actually happening. Since this type of variation is included in the comparisons presented in Chapter 4, it is clear that the overall effect would be minimal. These changes, especially the Web-related recommendations and the format standardization, would have implications for both the Controller’s Office and the reporting entities. On the Controller’s side, these changes would require some investment of time and resources in assessing both the appropriate reporting formats and developing the new information technologies to implement them. On the reporting entity’s side, each would have to find access to the Web and make the transition to completing an on-line form in a new format. Providing a cost estimate of either of these activities is beyond the scope of this report, but it is believed that the resulting improvements in the timelines and usefulness of the data would be worth the investment. The process of implementing the new format may well serve as the ideal mechanism for implementing the new and improved instructions as well. As entities review their reporting procedures to comply with the new format, they could also verify that they are including all of the types of activity desired by the State Controller’s Office. Conclusions: The Broader Implications of the Findings Beyond these specific recommendations, there are some broader conclusions and implications that arise as a result of this study’s findings. These findings have greater importance to the policy context in which this report was introduced. They are detailed below. 94 • Reliability. The revenue data are reliable for understanding the range of activity occurring in the local government sector. As a consequence, concerns about the data underlying the various studies on California’s public revenue burden can be dismissed. This then allows the decisionmaker and analyst to focus on the methodology and values involved in the policy argument instead of wondering whether the data are comprehensive and accurate enough to understand what is happening. The levels of variation identified in this study are well below the threshold change levels identified in the various published studies addressing the revenue burden issue, indicating that the changes they measure are real, subject to the constraints of their individual methodologies. • Usability. This study also indicates that the State Controller’s data are usable for further research into more detailed aspects of state and local governance. This was one of PPIC’s main concerns going into this study—if in fact the data were not of adequate quality, what, if any, corrections could be made to the data to make them usable. PPIC is committed to studying the implications of various governance and finance choices at the local level for local, state and federal policy and policy initiatives. It was and is critical to that line of research to have confidence in the quality of the local government finance data, and this study has found that such confidence is well placed in these data. • Areas for further work. This study has preliminarily identified one important area, beyond the revenue data, where the local finance data appear to have more significant problems. This is the area of debt reporting. Our preliminary review of reported debt proceeds indicated a relatively high level of variation between audited and reported numbers. To the extent that debt becomes an increasingly important part of the local finance and governance policy environment, this variation should be explored further. 95 Appendix A Detailed Sales Tax Comparisons This appendix contains the detailed results of the findings about sales taxes for counties and cities summarized in Tables 3.1 and 3.2. This detailed information and the statistical summaries of the differences in the entity-level data are provided for the reader’s reference. City Sales Tax Comparisons The detailed results presented in Table A.1 are summarized in Table 3.1. Note that the detail is presented in a semi-aggregated form. The specific comparisons for each city have been aggregated and reported by county in order to reduce the scale of the presentation. As an example, the entry in Table A.1 for Alameda County represents the sum of the sales tax revenues for all 14 cities in Alameda County. 97 Table A.1 Comparison of City Sales Tax Revenues, All Cities Aggregated by County, 1991–92 (dollars unless otherwise indicated) County Alameda Amador Butte Calaveras Colusa Contra Costa Del Norte El Dorado Fresno Glenn Humboldt Imperial Inyo Kern Kings Lake Lassen Los Angeles Madera Marin Mendocino Merced Modoc Mono Monterey Napa Nevada Orange Placer Plumas Riverside Sacramento San Benito San Bernardino Controller’s Reports 123,948,143 1,216,461 11,334,205 492,664 1,027,142 67,305,843 907,680 4,814,661 49,040,452 1,059,685 7,880,963 7,775,969 1,226,606 33,240,227 4,780,709 1,498,033 1,158,866 718,324,138 3,615,025 22,389,742 3,676,833 7,989,985 416,515 880,723 24,415,488 7,160,538 2,489,195 255,966,614 12,495,001 130,563 76,994,488 37,246,209 1,576,978 101,133,583 Board of Equalization 122,405,442 1,225,000 11,202,766 485,964 1,023,129 67,270,892 907,680 4,721,106 48,978,443 1,059,962 7,830,331 7,871,996 1,226,606 32,852,889 4,777,709 1,495,487 1,152,466 726,217,730 3,615,025 22,405,463 3,701,569 7,992,777 416,515 841,466 24,487,627 7,135,435 2,494,695 257,509,919 12,431,858 131,113 77,848,667 40,044,544 1,579,072 102,666,564 Difference 1,542,701 –8,539 131,439 6,700 4,013 34,951 0 93,555 62,009 –277 50,632 –96,027 0 387,338 3,000 2,546 6,400 –7,893,592 0 –15,721 –24,736 –2,792 0 39,257 –72,139 25,103 –5,500 –1,543,305 63,143 –550 –854,179 –2,798,335 –2,094 –1,532,981 Percentage Differencea 1.2% –0.7% 1.2% 1.4% 0.4% 0.1% 0.0% 1.9% 0.1% 0.0% 0.6% –1.2% 0.0% 1.2% 0.1% 0.2% 0.6% –1.1% 0.0% –0.1% –0.7% 0.0% 0.0% 4.5% –0.3% 0.4% –0.2% –0.6% 0.5% –0.4% –1.1% –7.5% –0.1% –1.5% 98 Table A.1—continued County San Diego San Francisco San Joaquin San Luis Obispo San Mateo Santa Barbara Santa Clara Santa Cruz Shasta Sierra Siskiyou Solano Sonoma Stanislaus Sutter Tehama Tulare Tuolumne Ventura Yolo Yuba Grand total Controller’s Reports 210,539,303 83,379,910 30,842,093 13,769,325 70,107,782 24,239,204 175,041,439 13,395,933 12,123,812 47,668 2,488,313 24,466,476 30,236,569 24,574,650 4,206,089 2,584,917 17,098,200 929,028 50,875,970 12,567,507 1,958,832 2,401,082,947 Board of Equalization 200,858,348 83,747,510 30,547,671 13,726,925 68,650,568 24,045,784 175,485,491 13,321,630 12,120,462 47,668 2,466,286 24,417,655 29,972,344 24,689,495 4,099,752 2,540,500 16,944,141 932,171 50,533,574 12,658,087 1,958,832 2,401,772,801 Difference 9,680,955 –367,600 294,422 42,400 1,457,214 193,420 –444,052 74,303 3,350 0 22,027 48,821 264,225 –114,845 106,337 44,417 154,059 –3,143 342,396 –90,580 0 –689,854 Percentage Differencea 4.6% –0.4% 1.0% 0.3% 2.1% 0.8% –0.3% 0.6% 0.0% 0.0% 0.9% 0.2% 0.9% –0.5% 2.5% 1.7% 0.9% –0.3% 0.7% –0.7% 0.0% 0.0% SOURCE: Controller’s data are from the Annual Report of Financial Transactions Concerning Cities, Fiscal Year 1991–92, Tables 2–7, pp. 4–481. The Board of Equalization data are from the State Board of Equalization’s Annual Report for the Year Ending June 30, 1992, Table 21A, pp. A-26 to A-29. NOTES: Alpine, Mariposa and Trinity Counties have no incorporated cities. aPercentage difference is the difference as a percentage of the Controller’s reported amounts. The summary statistics for these data, presented in Table A.2, also point to the closeness of the two datasets. The absolute average percentage difference between the two reported series is only 1.9 percent. 99 Table A.2 Summary Statistics for Differences in Reported City Sales Tax Revenues, All Cities, 1991–92 Description Average of absolute variance Largest underreporting Largest overreporting Proportion with < 3% difference Proportion with < 5% difference Dollars $89,543 $2,774,700 $10,401,694 Percentage 1.9% 46.1% 30.2% 82% 88% NOTE: See Table A.1 for underlying detail and additional references. County Sales Tax Comparisons The results presented in Table A.3 are the details for each county in the state. This information is summarized in Table 3.2 of this report. The summary statistics in Table A.4 show the average absolute percentage difference is only 3.4 percent. The average county received more than $11 million in sales taxes, and the average difference of $627,466 between the Controller’s and Board of Equalization reports accounts for 3.4 percent of this average. 100 Table A.3 Comparison of County Sales Tax Revenues, All Counties, 1991–92 (dollars unless otherwise indicated) County Alameda Alpine Amador Butte Calaveras Colusa Contra Costa Del Norte El Dorado Fresno Glenn Humboldt Imperial Inyo Kern Kings Lake Lassen Los Angeles Madera Marin Mariposa Mendocino Merced Modoc Mono Monterey Napa Nevada Orange Placer Plumas Riverside Sacramento San Benito San Bernardino San Diego San Joaquin Controller Reports 13,054,052 234,739 866,861 3,025,252 1,604,612 575,569 7,765,324 473,008 4,040,193 10,489,412 599,164 1,855,499 1,488,276 727,567 17,074,043 1,247,120 1,560,219 619,558 33,136,115 2,508,173 2,200,199 1,153,713 2,712,116 2,442,469 146,844 307,560 4,273,025 3,056,962 3,460,396 11,709,981 6,369,103 1,243,634 12,269,864 64,310,319 699,226 15,019,249 9,970,643 6,010,472 Board of Equalization 10,331,992 234,739 866,861 3,025,252 1,284,818 575,969 7,634,485 473,008 4,090,396 8,338,868 621,364 1,855,499 1,488,276 727,567 17,074,044 1,247,120 1,567,919 619,558 33,576,615 2,508,174 2,179,444 1,153,713 2,716,866 2,442,469 147,444 307,560 4,273,025 2,947,662 3,462,396 11,208,781 6,355,066 1,243,634 12,269,864 63,034,180 701,826 9,966,519 10,188,010 6,010,472 Variance 2,722,060 0 0 0 319,794 –400 130,839 0 –50,203 2,150,544 –22,200 0 0 0 –1 0 –7,700 0 –440,500 –1 20,755 0 –4,750 0 –600 0 0 109,300 –2,000 501,200 14,037 0 0 1,276,139 –2,600 5,052,730 –217,367 0 Percentage Variancea 20.9% 0.0% 0.0% 0.0% 19.9% –0.1% 1.7% 0.0% –1.2% 20.5% –3.7% 0.0% 0.0% 0.0% 0.0% 0.0% –0.5% 0.0% –1.3% 0.0% 0.9% 0.0% –0.2% 0.0% –0.4% 0.0% 0.0% 3.6% –0.1% 4.3% 0.2% 0.0% 0.0% 2.0% –0.4% 33.6% –2.2% 0.0% 101 Table A.3—continued County San Luis Obispo San Mateo Santa Barbara Santa Clara Santa Cruz Shasta Sierra Siskiyou Solano Sonoma Stanislaus Sutter Tehama Trinity Tulare Tuolumne Ventura Yolo Yuba Total Controller Reports 3,122,567 13,801,957 7,929,501 2,948,976 4,663,342 2,233,855 93,771 471,761 1,639,557 8,392,929 10,121,273 1,262,661 694,635 544,907 4,678,791 2,527,442 4,813,481 1,325,862 1,034,197 322,601,996 Board of Equalization 3,117,166 12,879,333 7,632,501 2,903,508 4,663,342 2,247,526 93,771 471,761 1,639,554 7,192,929 8,000,090 1,273,161 740,901 517,486 4,719,491 2,527,442 4,457,070 1,261,712 1,034,197 306,124,396 Variance 5,401 922,624 297,000 45,468 0 –13,671 0 0 3 1,200,000 2,121,183 –10,500 –46,266 27,421 –40,700 0 356,411 64,150 0 16,477,600 Percentage Variancea 0.2% 6.7% 3.7% 1.5% 0.0% –0.6% 0.0% 0.0% 0.0% 14.3% 21.0% –0.8% –6.7% 5.0% –0.9% 0.0% 7.4% 4.8% 0.0% 5.1% SOURCE: Controller’s data are from the Annual Report of Financial Transactions Concerning Counties, Fiscal Year 1991–92, Tables 6 and 9–13, pp. 14–32 and 100–137. The Board of Equalization data are from the State Board of Equalization’s Annual Report for the Year Ending June 30, 1992, Table 21A, pp. A-26 to A-29. aPercentage difference is the difference as a percentage of the Controller’s reported amounts. Table A.4 Summary Statistics for Differences in Reported County Sales Tax Revenues, All Counties, 1991–92 Description Average of absolute variance Largest underreporting Largest overreporting Proportion with < 3% difference Proportion with < 5% difference Dollars $627,466 $440,500 $5,052,730 Percentage 3.4% 6.7% 33.6% 74% 82% NOTE: See Table A.3 for underlying detail and additional references. 102 Appendix B Detailed Property Tax Comparisons This appendix contains the detailed results of the findings about property taxes for counties and cities summarized in Tables 3.3 and 3.4. This detailed information and the accompanying statistical summaries of the entity-level differences are provided for the reader’s reference. City Property Tax Comparisons The detailed results presented in Table B.1 are summarized in Table 3.3. Note that the detail is presented here in a semi-aggregated form. The specific comparisons for each city have been aggregated and reported by county in order to reduce the scale of the presentation. As an example, the entry in Table B.1 for Alameda County represents the sum of the property tax revenues for all 14 cities in Alameda County. 103 Table B.1 Comparison of City Property Tax Revenues, All Cities Aggregated by County, 1991–92 (dollars unless otherwise indicated) County Alameda Amador Butte Calaveras Colusa Contra Costa Del Norte El Dorado Fresno Glenn Humboldt Imperial Inyo Kern Kings Lake Lassen Los Angeles Madera Marin Mendocino Merced Modoc Mono Monterey Napa Nevada Orange Placer Plumas Riverside Sacramento San Benito San Bernardino San Diego San Francisco San Joaquin Controller 183,318,614 854,018 5,331,149 136,446 672,113 64,354,459 116,006 4,386,556 44,982,289 872,332 2,312,403 4,761,422 315,009 22,021,512 2,663,279 994,217 462,395 986,053,936 1,901,496 28,108,606 1,198,245 6,075,167 183,140 534,221 17,343,348 9,039,163 1,726,207 221,862,960 10,519,272 190,009 52,958,420 57,901,184 738,173 69,092,707 224,227,255 522,289,889 30,035,255 County AuditorController 182,609,682 878,759 5,279,639 132,330 638,654 65,869,851 118,199 4,318,961 44,488,494 899,176 2,193,000 4,661,794 296,976 21,612,090 2,760,219 978,642 437,211 915,967,524 1,770,172 26,371,367 1,165,921 6,081,295 203,047 538,127 17,285,664 8,728,000 1,643,236 218,185,979 10,024,605 137,139 51,674,136 56,434,993 779,469 65,128,268 217,996,342 511,334,716 29,524,769 Variance 708,932 –24,741 51,510 4,116 33,459 –1,515,392 –2,193 67,595 493,795 –26,844 119,403 99,628 18,033 409,422 –96,940 15,575 25,184 70,086,412 131,324 1,737,239 32,324 –6,128 –19,907 –3,906 57,684 311,163 82,971 3,676,981 494,667 52,870 1,284,284 1,466,191 –41,296 3,964,439 6,230,913 10,955,173 510,486 Percentage Variancea 0.4% –2.9% 1.0% 3.0% 5.0% –2.4% –1.9% 1.5% 1.1% –3.1% 5.2% 2.1% 5.7% 1.9% –3.6% 1.6% 5.4% 7.1% 6.9% 6.2% 2.7% –0.1% –10.9% –0.7% 0.3% 3.4% 4.8% 1.7% 4.7% 27.8% 2.4% 2.5% –5.6% 5.7% 2.8% 2.1% 1.7% 104 Table B.1—continued County San Luis Obispo San Mateo Santa Barbara Santa Clara Santa Cruz Shasta Sierra Siskiyou Solano Sonoma Stanislaus Sutter Tehama Tulare Tuolumne Ventura Yolo Yuba Grand total Controller 14,026,768 69,276,568 15,385,732 127,334,980 11,033,897 5,762,481 25,280 1,373,680 30,526,994 21,590,280 14,280,490 2,982,701 1,185,415 8,605,449 335,386 37,555,299 16,276,240 1,087,454 2,959,177,966 County Auditor- Controller Variance 13,796,855 229,913 66,949,327 2,327,241 13,897,870 1,487,862 123,859,452 3,475,528 9,480,608 1,553,289 5,404,143 358,338 24,392 888 1,449,634 –75,954 29,750,803 776,191 20,845,006 745,274 13,812,462 468,028 3,010,322 –27,621 1,265,706 –80,291 8,008,687 596,762 322,274 13,112 37,401,022 154,277 15,338,012 938,228 1,045,322 42,132 2,844,810,343 114,367,623 Percentage Variancea 1.6% 3.4% 9.7% 2.7% 14.1% 6.2% 3.5% –5.5% 2.5% 3.5% 3.3% –0.9% –6.8% 6.9% 3.9% 0.4% 5.8% 3.9% 3.9% SOURCE: Controller’s data are from the Annual Report of Financial Transactions Concerning Cities, Fiscal Year 1991–92, Tables 2–7, pp. 4–481. The county auditorcontroller data are from the original surveys submitted to the Controller’s Office Annual Report on Property Taxes. These surveys are used to prepare the summary tables presented in State Board of Equalization’s Annual Report for the Year Ending June 30, 1992, Tables 4–15, pp. A-4 to A-19. It is also worth noting that these amounts are very close to those reported by the Board of Equalization but that, because of the specific criteria reported in each category, these amounts are never explicitly reported in the Annual Report in a manner such that they correspond to those reported above for comparability. NOTES: Alpine, Mariposa and Trinity Counties have no incorporated cities. aPercentage difference is the difference as a percentage of the Controller’s reported amounts. Table B.2 contains the statistical summaries at the county-aggregated level for the differences in reported property taxes for cities. As a result of some difficulties with the source documents, it is not possible to provide these statistics for the fully disaggregated detailed data. 105 Table B.2 Summary Statistics for Differences in Reported City Property Tax Revenues, All Cities, 1991–92 Description Average of absolute variance Largest underreporting Largest overreporting Proportion with < 3% difference Proportion with < 5% difference Dollars $2,149,174 $1,515,392 $70,086,412 Percentage 4.2% 10.9% 27.8% 45% 69% NOTES: These comparisons are based on the county-level aggregations because it was not feasible to disaggregate the data included in the Board of Equalization reports. See Table B.1 for underlying detail and additional references. County Property Tax Comparisons The results presented in Table B.3 are the details for each county in the state. This information is summarized in Table 3.4 of this report. Table B.4 presents the summary statistics for these data at the detailed county level. These statistics further attest to the close agreement between the two data sets. 106 Table B.3 Comparison of County Property Tax Revenues, All Counties, 1991–92 (dollars unless otherwise indicated) County Alameda Alpine Amador Butte Calaveras Colusa Contra Costa Del Norte El Dorado Fresno Glenn Humboldt Imperial Inyo Kern Kings Lake Lassen Los Angeles Madera Marin Mariposa Mendocino Merced Modoc Mono Monterey Napa Nevada Orange Placer Plumas Riverside Sacramento San Benito San Bernardino San Diego San Joaquin Controller 244,631,962 1,302,100 8,426,311 16,976,687 6,260,557 4,853,537 153,612,873 2,136,772 28,233,034 87,750,635 4,011,468 17,196,633 18,077,125 8,261,674 136,644,112 14,576,330 10,876,670 3,662,935 2,221,118,329 13,032,443 54,822,482 2,716,103 16,199,034 26,816,199 2,244,483 6,633,710 50,734,910 24,181,607 14,677,843 326,392,260 41,678,575 4,423,707 191,197,422 182,197,235 4,528,019 190,760,765 346,181,240 90,229,793 County AuditorController Variance 239,690,240 1,233,386 7,997,242 16,813,050 6,203,279 4,762,462 149,602,417 2,153,954 26,912,738 80,532,082 3,984,571 16,403,295 17,015,635 8,022,820 138,367,835 14,183,427 10,846,596 3,562,188 2,064,141,394 12,419,819 54,354,253 2,790,041 15,068,978 26,050,773 2,004,100 5,928,187 48,770,233 23,218,833 14,299,544 329,176,489 40,451,075 4,753,792 173,240,222 179,094,318 4,536,123 183,641,238 341,951,387 87,429,755 4,941,722 68,714 429,069 163,637 57,278 91,075 4,010,456 –17,182 1,320,296 7,218,553 26,897 793,338 1,061,490 238,854 –1,723,723 392,903 30,074 100,747 156,976,935 612,624 468,229 –73,938 1,130,056 765,426 240,383 705,523 1,964,677 962,774 378,299 –2,784,229 1,227,500 –330,085 17,957,200 3,102,917 –8,104 7,119,527 4,229,853 2,800,038 Percentage Variancea 2.0% 5.3% 5.1% 1.0% 0.9% 1.9% 2.6% –0.8% 4.7% 8.2% 0.7% 4.6% 5.9% 2.9% –1.3% 2.7% 0.3% 2.8% 7.1% 4.7% 0.9% –2.7% 7.0% 2.9% 10.7% 10.6% 3.9% 4.0% 2.6% –0.9% 2.9% –7.5% 9.4% 1.7% –0.2% 3.7% 1.2% 3.1% 107 Table B.3—continued County San Luis Obispo San Mateo Santa Barbara Santa Clara Santa Cruz Shasta Sierra Siskiyou Solano Sonoma Stanislaus Sutter Tehama Trinity Tulare Tuolumne Ventura Yolo Yuba Grand total Controller 56,709,520 130,932,923 67,837,663 298,713,084 33,765,816 19,270,632 1,940,527 7,102,578 51,298,525 82,777,370 41,306,545 10,981,944 7,225,813 2,244,366 45,147,936 11,350,279 120,449,624 18,245,367 8,343,734 5,593,901,820 County Auditor- Controller Variance 56,347,036 362,484 126,798,154 4,134,769 67,675,213 162,450 292,380,335 6,332,749 33,072,656 693,160 18,562,664 707,968 1,926,673 13,854 6,446,262 656,316 50,854,789 443,736 81,462,994 1,314,376 39,268,442 2,038,103 10,777,681 204,263 6,998,041 227,772 2,185,865 58,501 41,247,236 3,900,700 9,860,309 1,489,970 120,971,069 –521,445 17,243,793 1,001,574 7,779,810 563,924 5,353,466,793 240,435,027 Percentage Variancea 0.6% 3.2% 0.2% 2.1% 2.1% 3.7% 0.7% 9.2% 0.9% 1.6% 4.9% 1.9% 3.2% 2.6% 8.6% 13.1% –0.4% 5.5% 6.8% 4.3% SOURCE: Controller’s data are from the Controller’s Annual Report of Financial Transactions Concerning Cities, Fiscal Year 1991–92, Tables 2–7, pp. 4–481. The county auditor-controller data are from the original surveys submitted to the Controller’s Office Annual Report of Property Taxes. These surveys are used to prepare the summary tables presented in State Board of Equalization’s Annual Report for the Year Ending June 30, 1992, Tables 4–15, pp. A-4 to A-19. aPercentage difference is the difference as a percentage of the Controller’s reported amounts. Table B.4 Summary Statistics for Differences in Reported County Property Tax Revenues, All Counties, 1991–92 Description Average of absolute variance Largest underreporting Largest overreporting Proportion with < 3% difference Proportion with < 5% difference Dollars $4,218,158 $2,784,229 $156,976,935 Percentage 3.2% 7.5% 13.1% 54% 74% NOTE: See Table B.3 for underlying detail and additional references. 108 Appendix C Selecting a Statewide Sample of Public Entities The sampling approach used in this study is key to the level of confidence one would have in its findings. In this appendix, the specifics of the sampling methodology of the study are presented. There were two approaches to sampling in this study. The initial concept was to select several counties in the state and to obtain detailed information from every entity within that county. After completing a study of a pilot county under this approach, the decision was made to change to a partially stratified, random sample of entities. Initially, the belief was that there were specific benefits to having the detail on a geographically and politically clustered group because there would be sufficient interplays and exchanges of resources between them, which would in turn facilitate understanding of the intergovernmental relationships and provide economies of scale. The first county selected 109 under these criteria was Alameda County. There were three reasons it was chosen: (1) its close proximity to PPIC, which would aid in information gathering; (2) its diversity in terms of the types and size of entities; and (3) its relatively moderate size and ordinal ranking when looking at the state across a wide range of demographic criteria. Several other counties were also identified for the next phase of research at this point. A pilot county, Alameda County, was chosen to test and verify the reasoning behind this methodology. The Initial Sample: Alameda County PPIC staff traveled to the county government and to each city within Alameda County to obtain CAFRs. A parallel telephone effort was instituted to obtain the CAFR for each special district that had activity within the county. Comprehensive audited financial statements were collected from the 20 independent districts, 11 joint powers agencies, and the three transportation planning agencies that had any activity in Alameda County, as well as all 14 cities and the county government itself. Information was not gathered separately on the redevelopment agencies or dependent special districts because the finances of such entities were included in the cities’ financial statements. The total number of special districts collected was slightly larger than either the Controller’s or Census of Governments’ list of special districts because this effort included entities that were administratively housed in other counties but had some activity in the county, while two other organizations were assigned nexus in Alameda County by virtue of the location of their district headquarters. These audited financial reports were then compared with the Controller’s reports using the methodology described in Appendix D. 110 After completing this study of Alameda County, the decision to use the county as the selection criterion was revisited. It was found that there were in fact no comparative benefits to using such a geographically clustered sampling approach. The level of shared information between entities and the aggregated level of detail in the reports did not allow the realization of some of the hoped for cross-comparisons of fiscal information. The Final Sample Consequently, the balance of the state sampling was stratified first by size, then through random selection. This was done by selecting the largest entities in a category (cities, counties, special districts). In general, this was done by listing the entities in a category in descending revenue order (as reported in the Controller’s reports). A cutoff point was then selected where the magnitude of revenues fell off. For example, the seven largest counties in the state were included in the sample. The largest, Los Angeles County, had more than $10 billion in revenues. The next largest had just under $2 billion in revenues.1 All seven had more than $1 billion in revenues. Alameda County was included because of its size and because it was the pilot county. The remaining seven counties were selected at random from the balance of the counties in the state. The strength of this methodology was that it allowed the study to make certain to address the bulk of the activity in the state while still allowing for the possibility that smaller entities may actually report the information either better (they have less activity and complexity) or ____________ 1For purposes of this analysis, the joint entity of the City and County of San Francisco is treated as a city. If treated as a county, it would actually be the next largest after Los Angeles County, with approximately $3 billion in revenues. 111 worse (they have fewer resources to commit to the provision of the information) than their large, urban cousins. Since dollars were the key dimension of this project, it was critical to make certain that the large entities in each category were included. A low percentage error in Los Angeles County with its more than $10 billion in revenues will have a much more significant effect on any research findings than Plumas County’s $27 million or San Benito County’s $31 million in revenues. However, to the extent that one wishes to make comparisons and understand the differential effects of policy on various local governments, it is just as important to know that the information included for Plumas and San Benito Counties is accurate. As a result, the sample of entities from across the state was composed of a complete census of the largest entities, as well as a random sample of the smaller counties, cities and special districts. The detailed samples are given in the subsections below. Counties Sampled As described above, the seven largest counties and Alameda County were included in the sample. The remaining counties were sampled at random from the rest of the counties in the state. The counties included in the sample were Alameda, Humboldt, Los Angeles, Orange, Placer, Plumas, Riverside, Sacramento, San Benito, San Bernardino, San Diego, San Luis Obispo, Santa Clara, Siskiyou and Tehama. Cities Sampled A similar approach was followed with cities. The sample initially included the 14 cities in Alameda County. With the expansion of the sample to a stratified statewide sample, the largest 15 cities (by population) were selected. An additional 29 cities were selected at 112 random from the rest for a total of 54 cities.2 These cities were Alameda, Albany, Anaheim, Bakersfield, Berkeley, Carlsbad, Chowchilla, Delano, Dublin, Emeryville, Eureka, Fontana, Foster City, Fremont, Fresno, Glendale, Hayward, Huntington Beach, Irwindale, La Habra, La Quinta, Lindsay, Livermore, Long Beach, Los Angeles, Manteca, Modesto, Morro Bay, Napa, Newark, Oakdale, Oakland, Parlier, Piedmont, Pleasanton, Redding, Redwood City, Riverside, Sacramento, San Bernardino, San Diego, San Francisco, San Jose, San Leandro, San Mateo, Santa Ana, Simi Valley, South Pasadena, Stockton, Union City, Villa Park, Westmorland, Woodside and Yucca Valley. The information from only two cities on this list was not received in time to be included in the analysis in this report, reducing the total sample to 52 cities.3 The sample, while oversampling Alameda County, was reviewed across several key dimensions to ensure a robust sample. These dimensions included region of the state, size, government type and age of the city. The sample contains a wide range of cities within each of these categories, and as such it contains a reasonable cross-section of the state’s cities. Special Districts Sampled The special district sample was drawn in a fashion similar to that for the cities and counties. First, data for all of the independent special districts in Alameda County were collected. These 34 special districts ____________ 2Note that there was some overlap between the Alameda County cities group and the largest cities group—three cities were in both samples. 3The two cities that were not received in time for inclusion were San Bernardino and Chowchilla. 113 included 20 independent districts, 11 joint power agencies, and three transportation planning agencies. In addition to these Alameda County entities, 141 more special districts were selected for review. These were selected first by district type and then, as with cities and counties above, by size and finally at random. The districts with the largest revenues in each category were chosen first. Because significant concerns had been voiced about transit districts and transportation planning authorities, these two categories were oversampled. After selecting the largest districts in each category, 45 more districts were sampled at random from the remaining districts. The response rate for this sample was very high. Of the 175 districts selected, information was obtained on 155. Of the 20 missing districts, 6 reported they were not in existence and 14 refused or were unable to provide audited financial information. These 14 entities were broadly distributed throughout the sample categories and also varied by size. As a result, these missing entities are not believed to affect research design. The final sample was composed of 155 special districts. Nineteen of the state’s transportation planning agencies and 18 transit districts were included, as well as 30 water/sewer/irrigation districts, 11 hospitals, 13 other enterprise districts (public utility districts, airports, and ports) and 64 non-enterprise districts (e.g., park and recreation districts, flood control, self-insurance). The specific districts are included by district category below. Hospital Districts. Antelope Valley Hospital District, Eden Township Hospital District, El Camino Hospital District, Kaweah Delta Hospital District, Mount Diablo Hospital District, Palomar Pomerado Hospital District, Salinas Valley Memorial Hospital District, Sequoia 114 Hospital District, Tri-City Hospital District, Valley Health System and Washington Township Hospital District. Transit Districts. Alameda-Contra Costa Transit District, Bay Area Rapid Transit District, Central Contra Costa Transit Authority, Golden Gate Bridge, Highway and Transportation District, Livermore/Amador Valley Transit Authority, Long Beach Public Transportation Company, Marin County Transit District, North County Transit District, Orange County Transit, Sacramento Regional Transit District, San Diego Transit Corporation, San Diego Trolley, Inc., San Mateo County Transit District, Santa Barbara Metropolitan Transit District, Santa Cruz Metropolitan Transit District, Southern California Rapid Transit District and Stockton Metropolitan Transit District. Transportation Planning Agencies. Alameda County Transportation Authority, Association of Bay Area Governments, Contra Costa Transportation Authority, Fresno County Transportation Authority, Imperial County Local Transportation Authority, Los Angeles County Transportation Commission, Madera County Transportation Authority, Metropolitan Transportation Commission, Orange County Transportation Authority, Riverside County Transportation Commission, Sacramento Area Council of Governments, Sacramento Transportation Authority, San Bernardino Association of Governments, San Diego Association of Governments, San Diego Metropolitan Transit Development Board, San Francisco County Transportation Authority, San Mateo County Transportation Authority, Santa Clara County Traffic Authority and Southern California Association of Governments. Water and Sanitation Districts. Alameda County Water District, Byron-Bethany Irrigation District, Central Contra Costa Sanitary 115 District, Central Marin Sanitation Agency, Costa Mesa Sanitary District, Dublin San Ramon Service District, East Bay Dischargers Authority, Eastern Municipal Water District, Fairfield-Suisun Sewer District, Garden Grove Sanitary District, Irvine Ranch Water District, Ivanhoe Public Utility District, Jurupa Community Services District, Kern County Water Agency, La Habra Heights County Water District, Leucadia County Water District, Livermore-Amador Valley Water Management Agency, Mesa Consolidated Water District, Metropolitan Water District of Southern California, Oro Loma Sanitary District, Sacramento Regional County Sanitation District, San Benito County Water District, San Diego County Water Authority, Shafter-Wasco Irrigation District, South Tahoe Public Utility District, Thermalito Irrigation District, Tri-Valley Wastewater Authority, Union Sanitary District, Ventura Regional Sanitation District and West Kern Water District. Airport, Electric Utility and Harbor and Port Districts. East Kern Airport District, Lassen Municipal Utility District, Monterey Peninsula Airport District, M-S-R Public Power Agency, Northern California Power Agency, Sacramento Municipal Utility District, Sacramento-Yolo Port District, San Diego Unified Port District, Santa Maria Public Airport District, Southern California Public Power Authority, Stockton Port District, Transmission Agency of Northern California and Truckee-Donner Public Utility District. Non-Enterprise Districts. Alameda County Mosquito Abatement District, Alameda County Resource Conservation District, American River Fire Protection District, Auburn Cemetery District, Auburn Recreation and Park District, Bay Area Air Quality Management District, Bay Area Housing Authority Risk Management Agency, Bay 116 Area Library and Information System, Brannan-Andrus Levee Maintenance District, Broadmoor Police Protection District, California Housing Authority Risk Management Agency, California Joint Powers Risk Management Authority, Chino Valley Independent Fire District, City of San Francisco Downtown Parking Corporation, City of San Francisco Uptown Parking Corporation, Clovis Memorial District, Conejo Recreation and Park District, Dry Creek Storm Water District, East Bay Regional Park District, East Bay Schools Insurance Group, Eastern Plumas Fire Protection District, East Side Mosquito Abatement District, Elk Grove Community Services District, Fairview Fire Protection District, Fresno Metropolitan Flood Control District, Gilsizer County Drainage District, Glenn County Mosquito Abatement District, Half Moon Bay Fire Protection District, Hayward Area Recreation and Park District, Inland Empire Schools Insurance Authority, Jamestown Cemetery District, Kings River Resource Conservation District, Livermore Area Recreation and Park District, Madera Cemetery District, Marin-Sonoma Mosquito Abatement District, Mariposa County Resource Conservation District, Midpeninsula Regional Open Space District, North Coast Schools Insurance District, North Coast Schools Medical Insurance Group, North County Cemetery District, Orange County Cemetery District, Orange County Vector Control District, Ortega Trail Recreation and Park District, Palos Verdes Library District, Panoche Drainage District, Reclamation District #999, Reclamation District #2025, Reclamation District #2091, Resource Conservation District of Santa Monica, Riverside-Corona Resources Conservation District, Riverside County Flood Control District, Rocklin Placer Library Authority, SacramentoYolo Mosquito and Vector Control District, San Ramon Fire Protection 117 District, Santa Barbara County Health Care District, Santa Maria Valley Water Conservation District, Schools Insurance Authority, Self-Insured School District of Kern, Self-Insured School District of Kern II, South Coast Air Quality Management District, Spreckels Memorial District, Surfside Colony Community Services District, Temecula Public Cemetery District and Tulare County Pest Control District. 118 Appendix D Financial Statement Reconciliation Methodology Reconciling the Controller’s reports to audited financial reports was a key portion of this analysis of the accuracy of the Controller’s reports, the results of which were presented in Chapter 4. This appendix describes the methodology used to reconcile the two sets of information. The series used in the reconciliations for the Controller’s reports were those in the published Annual Report on Financial Transactions Concerning . . . books. These are the final adjusted numbers reported by the Controller’s Office for the various entities. For the audited financial reports, amounts were captured from each entity’s comprehensive 119 audited financial report.1 The next section provides more detail on the CAFR. Some Background on the Comprehensive Audited Financial Report The CAFR reports the findings of an annual audit required of public governments in California. These reports are prepared by independent public accounting firms and follow a format prescribed by Generally Accepted Accounting Practices (GAAP).2 This format contains general instructions regarding the content and format of the audits to be performed and the specific content and format of the report. This section briefly highlights some of the issues that are relevant to the reconciliation process. First, the range of activity included in a given CAFR encompasses all the activity over which that entity’s governing board has control. For this reason, the activities of dependent entities, such as redevelopment agencies, whose entire governing board is typically the city council or the county board of supervisors, are included in the CAFR’s reported activity. Other types of dependent entities include libraries, community facility districts and housing authorities. Local-run enterprise activities, such as water, sewer and transit districts, are also included in the parent entity’s report, although they are listed separately in the report. Since the Controller reports many of these activities separately as dependent special districts in the Annual Report on Financial Transactions ____________ 1For the small number of entities that did not have CAFRs, any other audited numbers were collected or, when no audits were possible, the budget figures that showed the actual financial activity of the entity for the 1991–92 fiscal year were collected. 2These practices are established and enforced by a national professional board, the Financial Accounting Standards Board. 120 Concerning Special Districts, it is often necessary to add these amounts to the original entity’s amounts to assure an appropriate comparison. Second, in accordance with GAAP, the CAFR reports several types of funds within each entity. These fund categories reflect the different types of monies that a local government can have. One group of categories refers to the limitations placed on the use of funds. Funds in this category include general funds,3 special funds,4 capital projects funds5 and debt service funds.6 Others refer to the type of activity or enterprise that generates the revenues, such as enterprise7 and internal service funds.8 Another group identifies revenues from the perspective of ____________ 3General funds are the general “checking accounts” for local governments. These funds represent the unrestricted revenues by which the bulk of the activity of local government is funded. Most programs and activities over which the local government has full discretion are funded from these accounts. 4Special funds are funds legally or contractually obligated to fund specific activities within the entity, but that do not fall into the specific categories listed later in this paragraph. Some examples of these funds are cigarette taxes, which are earmarked for anti-smoking campaigns, and certain transit taxes, which are set aside for transit development and expansion. 5Capital projects funds are funds specifically earmarked to pay for the development and operation of capital projects within the entity. 6Debt service funds are those which include the revenues and expenditures for activities specifically earmarked to retire debt for which the entity is obligated. 7Enterprise funds contain the activity of wholly owned subsidiaries that generate revenues for the provision of specific public services within the local government entity. These are usually locally run utilities, hospitals, airports and transit activities. These funds are reported separately in part because the enterprise’s rates and charges are often set as a function of its cost of providing services. Providing an audited account of its revenues and expenditures facilitates the rate-setting process. 8Internal service funds include the activities of internal entity departments charged to other departments. For example, a city’s computer department may “charge” the cable enterprise for installing a new computer system. The city would reflect these charges as revenues to the internal service fund. These are tracked in part for internal accounting purposes and in part to make certain that these “revenues and expenditures,” which remain entirely within the city, are correctly identified and do not overrepresent the city’s overall revenues. They should not be included in the amounts reported to the Controller because they do not reflect revenues from outside the entity’s internal governmental structure. 121 the entity’s level of control over the funds—expendable9 and nonexpendable10 trust funds. Many independent special purpose districts also make a distinction between operating11 and non-operating funds.12 Each of these funds are reported across a range of revenue categories. Each category represents a way in which local governments can obtain revenues. These revenue categories include taxes, fines and forfeitures, permits, revenues from the use of money and property, current service charges and miscellaneous revenues. It is convenient that the Controller’s reports also use these same distinctions. The definitions of activity within each, however, are not exactly the same between the two reports. GAAP allows for much more variation in interpreting which revenue category best suits a specific activity.13 The Controller’s instructions are much more explicit. As such, there is a significant chance that there will be a different revenue category reported for a specific activity between the two reports. Finally, each CAFR typically includes notes that expand upon the information presented in the financial tables themselves. It is in the notes, for example, that the specific reporting entity is defined and that one ascertains which dependent districts are included in the accounting ____________ 9Expendable trust funds are monies left to the local government by a third party over which the entity has significant discretion in its use. 10Non-expendable trust funds are resources held and owned by a local government for a special purpose. These funds are constrained by contractual or legal obligations of the entity. The largest funds in this category are pension funds. Occasionally, CFD revenues are included here. 11Operating funds are those raised by a special purpose district that come about as a direct result of that entity’s primary activity or activities. 12Non-operating funds are those raised by a special purpose district that do not come about as a direct result of that entity’s primary activity or activities. 13This is due in part to the national character of these specifications. They must be broad and flexible enough to encompass a much larger range of activities than the California State Controller Office must track. 122 reconciliation. These notes also explain any activities that are unique to the entity and contain detailed descriptions about its outstanding debt and its pension funds and requirements. The Reconciliation Process The goal of the reconciliation process is to compare the information in the Controller’s report to the audited financial statements. The general philosophy used is to (a) create comparable entities, (b) compare the two sets of information by revenue category, and (c) seek explanations for any variation identified and to make corrections, if appropriate. There is also a desire to be consistent across all of the entities in the sample—one of the goals of this analysis is to identify how consistently the Controller’s instructions are interpreted and applied by the more than 6,500 local government entities in the state. The specific steps in the reconciliation are listed below: 1. A specific summary of the CAFR’s reported audited numbers was prepared. These summaries were listed by revenue category. The general, special, capital projects and debt service funds were always included. In addition, enterprise funds were also included. Internal service funds and non-expendable trust funds were always excluded. The inclusion of expendable trust funds was decided on a case-by-case basis.14 2. The Controller’s comparable entity was developed. First, the information in the published Controller’s report for the entity was ____________ 14The general criteria focused on whether the city had complete control of the revenues and from where the monies came. 123 aggregated into the same revenue categories as those reported in the CAFR.15 The entity list in the notes section of the CAFR was reviewed to see if there were any dependent entities that were included separately in the Controller’s reports. If any such entities were identified, their revenues were identified in the appropriate Controller’s report and added to those of the entity itself to create a comparable entity. 3. The Controller-based entity was then compared with the CAFR. The comparison was made on a revenue category-by-category basis. Variation between the two reports was identified both in each category and overall. 4. If the difference between the CAFR and the Controller’s numbers was greater than three percent, or in cases where all of the difference was concentrated in a single revenue category, an attempt was made to contact that entity for an explanation of the discrepancy. If the explanations provided required any corrections, these were addressed at this point. If satisfactory explanations could not be determined or if the resulting variation was still higher than the desired threshold listed above, the variance was left in the analysis. Therefore, any variance listed is a conservative estimate of the level of variation between the reports. It is possible that additional proper explanations are available for the variance reported—they were simply beyond the scope of the resources available to obtain them.16 ____________ 15The same general revenue categories were used for each entity type. 16Local governments were contacted at least four times before the unexplained variance was simply left in the analysis. 124 These reconciliations were performed by experienced accounting staff and reviewed for accuracy, completeness and consistency by the project leaders. The final results of these reconciliations were then compiled for this analysis. 125 Appendix E Detailed Comparison of Controller’s Reports to Audited Reports: Counties This appendix contains the detailed county-by-county comparisons of overall revenues summarized in Table 4.1. Table E.1 refers to each county in the sample and provides detailed revenues in each source as well as the absolute and percentage variance for each. Table E.2 presents a summary for county revenue differences. 127 Table E.1 Overall Variance in Revenues by County, PPIC Sample, 1991–92 (dollars unless otherwise indicated) County Controller’s Report CAFR Percentage Differencea Differenceb Alameda Humboldt Los Angeles Orange Placer Plumas Riverside Sacramento San Benito San Bernardino San Diego San Luis Obispo Santa Clara Siskiyou Tehama Sample total 1,336,376,878 107,295,843 10,637,724,062 1,851,222,881 160,121,836 27,222,179 1,277,362,193 1,266,267,170 31,233,474 1,491,348,401 1,925,142,944 215,138,708 1,794,561,975 49,274,479 50,505,271 22,220,798,294 1,335,938,000 109,735,034 10,804,509,000 1,894,906,668 160,416,275 27,403,031 1,295,303,000 1,263,491,000 31,888,493 1,493,286,000 1,982,789,000 221,376,757 1,774,369,907 50,823,204 50,881,555 22,497,116,924 438,878 2,439,191 166,784,938 43,683,787 294,439 180,852 17,940,807 2,776,170 655,019 1,937,599 57,646,056 6,238,049 20,192,068 1,548,725 376,284 276,318,630 0.0% 2.3% 1.6% 2.4% 0.2% 0.7% 1.4% 0.2% 2.1% 0.1% 3.0% 2.9% 1.1% 3.1% 0.7% 1.2% aDifference is presented in absolute terms. bPercentage difference is the difference as a percentage of the Controller’s reported amounts. Table E.2 Summary Statistics for Differences in Reported Overall County Revenues, PPIC Sample, 1991–92 Description Average of absolute variance Largest underreporting Largest overreporting Proportion with < 3% difference Proportion with < 5% difference Dollars $21,542,191 $166,784,938 $20,192,068 Percentage 1.5% 3.1% 1.1% 94% 100% NOTE: See Table E.1 for underlying detail and additional references. 128 Appendix F Detailed Comparison of Controller’s Reports to Audited Reports: Cities This appendix contains the detailed city-by-city comparisons of overall revenues summarized in Table 4.3. Table F.1 refers to each city in the sample and provides detailed revenues in each source as well as the absolute and percentage variance for each. Each city is presented individually, therefore the first line, which refers to Alameda, is for the City of Alameda, not the cities in the County of Alameda as is the case in Appendixes C and B. Table F.2 presents a summary of these results. 129 Table F.1 Overall Variance in Revenues by City, PPIC Sample, 1991–92 (dollars unless otherwise indicated) City Alameda Albany Anaheim Bakersfield Berkeley Carlsbad Delano Dublin Emeryville Eureka Fontana Foster City Fremont Fresno Glendale Hayward Huntington Beach Irwindale La Habra La Quinta Lindsay Livermore Long Beach Los Angeles Manteca Modesto Morro Bay Napa Newark Oakdale Oakland Parlier Piedmont Pleasanton Redding Redwood City Riverside Controller’s Report 96,025,964 10,227,244 511,914,234 117,843,042 142,505,892 74,442,368 15,070,563 16,087,477 19,208,296 24,755,649 85,408,186 42,603,768 90,531,341 274,671,462 258,066,930 87,878,928 149,271,364 117,362,428 29,478,535 20,476,844 5,465,245 60,307,576 827,583,052 6,170,652,329 26,748,613 103,283,396 9,802,143 45,929,676 21,288,827 7,926,170 556,730,768 2,618,047 6,281,386 48,865,069 113,207,966 69,505,589 402,358,554 CAFR 111,645,495 9,998,842 523,276,000 119,864,561 156,990,000 74,340,421 15,382,547 16,138,379 20,899,499 25,664,069 85,555,000 43,472,367 103,078,610 274,830,935 255,217,407 88,692,248 151,474,000 15,008,649 29,698,246 21,025,675 6,148,812 60,026,563 742,496,000 6,254,840,000 26,893,659 103,857,197 9,867,466 49,782,607 22,038,047 8,346,875 566,136,000 2,708,047 6,740,534 51,169,280 115,653,900 73,319,405 383,439,183 Differencea 15,619,531 228,402 11,361,766 2,021,519 14,484,108 101,947 311,984 50,902 1,691,203 908,420 146,814 868,599 12,547,269 159,473 2,849,523 813,320 Percentage Differenceb 16.3% 2.2% 2.2% 1.7% 10.2% 0.1% 2.1% 0.3% 8.8% 3.7% 0.2% 2.0% 13.9% 0.1% 1.1% 0.9% 2,202,636 102,353,779 219,711 548,831 683,567 281,013 85,087,052 84,187,671 145,046 573,801 65,323 3,852,931 749,220 420,705 9,405,232 90,000 459,148 2,304,211 2,445,934 3,813,816 18,919,371 1.5% 87.2% 0.7% 2.7% 12.5% 0.5% 10.3% 1.4% 0.5% 0.6% 0.7% 8.4% 3.5% 5.3% 1.7% 3.4% 7.3% 4.7% 2.2% 5.5% 4.7% 130 Table F.1—continued City Sacramento San Diego San Francisco San Jose San Leandro San Mateo Santa Ana Simi Valley South Pasadena Stockton Union City Villa Park Westmorland Woodside Yucca Valley Sample total Controller’s Report 346,860,999 1,106,748,531 2,953,515,334 659,063,058 61,997,779 66,574,216 197,294,635 61,128,118 13,304,341 142,378,224 23,249,959 2,038,084 767,903 2,447,101 2,540,090 16,302,293,293 CAFR 442,230,000 1,096,171,000 2,662,419,000 655,473,813 61,122,886 64,242,104 195,667,691 63,348,185 13,304,339 162,513,000 25,116,930 2,023,507 1,006,664 2,812,755 2,549,345 16,075,717,744 Differencea 95,369,001 10,577,531 291,096,334 3,589,245 874,893 2,332,112 1,626,944 2,220,067 2 20,134,776 1,866,971 14,577 238,761 365,654 9,255 226,575,549a Percentage Differenceb 27.5% 1.0% 9.9% 0.5% 1.4% 3.5% 0.8% 3.6% 0.0% 14.1% 8.0% 0.7% 31.1% 14.9% 0.4% 1.4% aDifference is presented in absolute terms. Accordingly, the overall total listed does not equal the sum of the column, but the difference between the overall amounts reported. bPercentage difference is the difference as a percentage of the Controller’s reported amounts. Table F.2 Summary Statistics for Differences in Reported Overall City Revenues, PPIC Sample, 1991–92 Description Average of absolute variance Largest underreporting Largest overreporting Proportion with < 3% difference Proportion with < 5% difference Dollars $15,640,190 $95,359,001 $291,096,334 Percentage 6.7% 31.1% 87.2% 58% 71% NOTE: See Table F.1 for underlying detail and additional references. 131 Appendix G Detailed Comparison of Controller’s Reports to Audited Reports: Special Districts This appendix provides tables detailing the findings reported in Tables 4.8 through 4.14, as well as some summary statistics of the differences at the entity level. It is organized to parallel the order of the tables included in the body of the report—hospital districts, transit districts, transportation planning agencies, water and sanitation districts, miscellaneous enterprise districts and, finally, non-enterprise districts. There is no detail provided for the overall special district summary provided in Table 4.14. This simply represents an accumulation of the values reported in each of the other special district tables. 133 Hospital Districts Table G.1 Overall Variance in Revenues for Hospital Districts, 1991–92 (dollars unless otherwise indicated) District Name Controller Antelope Valley 106,723,755 Eden Township 86,895,638 El Camino 177,653,644 Kaweah Delta 97,169,845 Mt. Diablo 132,163,864 Palomar Pomerado 211,514,468 Salinas Valley Memorial 116,332,768 Sequoia 151,752,099 Tri-City 148,898,251 Valley Health System 134,834,487 Washington Township 120,552,063 Sample total 1,484,490,882 Audited Report 106,799,025 85,982,000 180,572,140 97,235,574 134,564,000 211,514,468 116,332,768 151,019,000 148,351,861 134,569,734 120,394,000 1,487,334,570 Differencea 75,270 913,638 2,918,496 65,729 2,400,136 0 Percentage Differenceb 0.1% 1.1% 1.6% 0.1% 1.8% 0.0% 0 733,099 546,390 0.0% 0.5% 0.4% 264,753 0.2% 158,063 2,843,688 0.1% 0.2% aDifference is presented in absolute terms. Accordingly, the overall total listed does not equal the sum of the column, but the difference between the overall amounts reported. bPercentage difference is the difference as a percentage of the Controller’s reported amounts. 134 Transit Districts Table G.2 Overall Variance in Revenues for Transit Districts, 1991–92 (dollars unless otherwise indicated) District Name Controller Audited Report Percentage Differencea Differenceb Alameda-Contra Costa Transit District 143,891,257 143,990,000 98,743 0.1% BART 285,013,596 285,007,000 6,596 0.0% Central Contra Costa Transit Authority 14,756,837 14,757,000 163 0.0% Golden Gate Bridge, Highway and Transportation District 28,426,761 28,633,000 206,239 0.7% Livermore/Amador Valley Transit Authority 4,317,530 4,333,996 16,466 0.4% Long Beach Public Transportation Company 31,496,982 31,496,982 0 0.0% Marin Co. Transit District 1,621,438 1,596,551 24,887 1.5% North Co. Transit District 26,732,864 26,736,997 4,133 0.0% Orange Co. Transit 115,303,777 115,303,777 0 0.0% Sacramento Regional Transit District 49,730,820 49,730,821 1 0.0% San Diego Transit Corporation 54,696,544 54,696,545 1 0.0% San Diego Trolley, Inc. 19,451,926 19,451,926 0 0.0% San Mateo Co. Transit District 81,494,904 89,078,369 7,583,465 9.3% Santa Barbara Metropolitan Transit District 8,333,022 8,333,023 1 0.0% Santa Clara Co. Transit District 166,980,278 164,425,009 2,555,269 1.5% Santa Cruz Metropolitan Transit District 21,226,070 21,225,171 899 0.0% Southern California Rapid Transit District 632,754,288 630,989,000 1,765,288 0.3% Stockton Metropolitan Transit District 10,768,741 9,821,249 947,492 8.8% Sample total 1,696,997,635 1,699,606,416 2,608,781 0.2% aDifference is presented in absolute terms. Accordingly, the overall total listed does not equal the sum of the column, but the difference between the overall amounts reported. bPercentage difference is the difference as a percentage of the Controller’s reported amounts. 135 Transportation Planning Agencies Table G.3 Overall Variance in Revenues for Transportation Planning Agencies, 1991–92 (dollars unless otherwise indicated) District Name Alameda County Transportation Authority Association of Bay Area Governments Contra Costa Transportation Authority Fresno County Transportation Authority Imperial Co. Local Transportation Authority Los Angeles Co. Transportation Commission Madera Co. Transportation Authority Metropolitan Transportation Commission Orange Co. Transportation Authority Riverside Co. Transportation Commission Sacramento Area Council of Governments Sacramento Transportation Authority San Bernardino Association of Governments San Diego Association of Governments San Diego Metropolitan Transit Development Board San Francisco Co. Transportation Authority San Mateo Co. Transportation Authority Santa Clara Co. Traffic Authority Southern California Association of Governments Sample total Controller 76,845,000 5,897,594 53,063,000 38,251,629 5,178,213 1,314,435,023 3,062,472 202,866,248 228,268,952 80,236,093 39,595,972 50,377,539 227,375,321 226,446,352 215,022,840 47,012,962 40,950,805 125,023,000 12,633,341 2,992,542,356 Audited Report 76,845,000 5,936,311 53,063,000 37,929,779 5,227,609 1,315,707,384 3,062,473 199,036,029 235,623,782 80,235,824 39,277,620 50,377,539 90,499,428 224,010,537 211,837,676 47,012,000 40,950,805 125,121,000 11,686,250 2,853,440,046 Percentage Differencea Differenceb 0 38,717 0 0.0% 0.7% 0.0% 321,850 0.8% 49,396 1.0% 1,272,361 1 0.1% 0.0% 3,830,219 7,354,830 1.9% 3.2% 269 0.0% 318,352 0 0.8% 0.0% 136,875,893 60.2% 2,435,815 1.1% 3,185,164 1.5% 962 0.0% 0 0.0% 98,000 0.1% 947,091 139,102,310 7.5% 4.6% aDifference is presented in absolute terms. Accordingly, the overall total listed does not equal the sum of the column, but the difference between the overall amounts reported. bPercentage difference is the difference as a percentage of the Controller’s reported amounts. 136 Water and Sanitation Districts Table G.4 Overall Variance in Revenues for Water and Sanitation Districts, 1991–92 (dollars unless otherwise indicated) District Name Audited Controller Report Percentage Differencea Differenceb Alameda Co. Water District Byron-Bethany Irrigation District Central Contra Costa Sanitary District Central Marin Sanitation Agency Costa Mesa Sanitary District Dublin San Ramon Service District East Bay Dischargers Authority Eastern Municipal Water District Fairfield-Suisun Sewer District Garden Grove Sanitary District Irvine Ranch Water District Ivanhoe Public Utility District Jurupa Community Services District Kern Co. Water Agency La Habra Heights Co. Water District Leucadia Co. Water District Livermore-Amador Valley Water Mgt Agency Mesa Consolidated Water District Metropolitan Water District of So. Calif. Oro Loma Sanitary District Sacramento Regional Co. Sanitation District San Benito Co. Water District San Diego Co. Water Authority Shafter-Wasco Irrigation District South Tahoe Public Utility District Thermalito Irrigation District Tri-Valley Wastewater Authority Union Sanitary District Ventura Regional Sanitation District West Kern Water District 34,480,701 1,021,186 46,913,101 4,987,586 4,793,411 13,910,506 2,088,715 123,970,053 18,345,184 5,858,264 144,887,000 762,146 8,262,648 105,333,695 2,821,486 6,719,834 2,027,482 12,593,135 549,699,920 19,068,690 68,593,725 4,113,366 168,637,074 1,698,589 18,227,697 1,085,582 405,632 33,033,989 29,138,618 10,029,706 34,480,701 1,000,998 46,906,900 4,987,018 4,797,964 13,910,506 2,061,883 125,039,572 17,863,867 5,858,264 144,887,000 762,146 8,277,443 105,333,695 2,821,486 6,770,395 2,027,482 12,593,135 543,802,000 11,086,210 68,593,726 4,093,992 168,746,000 1,698,589 18,227,697 1,085,582 405,632 33,034,000 29,137,245 10,029,706 0 20,188 6,201 568 4,553 0 26,832 1,069,519 481,317 0 0 0 14,795 0 0 50,561 0 0 5,897,920 7,982,480 1 19,374 108,926 0 0 0 0 11 1,373 0 0.0% 2.0% 0.0% 0.0% 0.1% 0.0% 1.3% 0.9% 2.6% 0.0% 0.0% 0.0% 0.2% 0.0% 0.0% 0.8% 0.0% 0.0% 1.1% 41.9% 0.0% 0.5% 0.1% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Sample total 1,443,508,721 1,430,320,834 13,187,887 0.9% aDifference is presented in absolute terms. Accordingly, the overall total listed does not equal the sum of the column, but the difference between the overall amounts reported. bPercentage difference is the difference as a percentage of the Controller’s reported amounts. 137 Airport, Electric Utility and Harbor and Port Districts Table G.5 Overall Variance in Revenues for Airport, Electric Utility and Harbor and Port Districts, 1991–92 (dollars unless otherwise indicated) District Name Controller East Kern Airport District 5,349,678 Lassen Municipal Utility District 14,336,889 Monterey Peninsula Airport District 4,722,181 M-S-R Public Power Agency 68,462,793 Northern California Power Agency 181,838,260 Sacramento Municipal Utility District 683,046,027 Sacramento-Yolo Port District 12,086,055 San Diego Unified Port District 109,183,738 Santa Maria Public Airport District 2,252,186 Southern California Public Power Authority 245,085,477 Stockton Port District 11,052,578 Transmission Agency of Northern California 46,723 Truckee-Donner Public Utility District 10,394,621 Sample total 1,347,857,206 Audited Report 5,349,678 14,336,892 4,475,385 70,625,000 180,258,000 682,982,000 12,086,055 109,183,738 2,252,187 245,085,000 11,052,579 46,723 10,324,759 1,348,057,996 Differencea 0 3 246,796 2,162,207 1,580,260 64,027 0 0 1 477 1 0 69,862 200,790 Percentage Differenceb 0.0% 0.0% 5.2% 3.2% 0.9% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.7% 0.0% aDifference is presented in absolute terms. Accordingly, the overall total listed does not equal the sum of the column, but the difference between the overall amounts reported. bPercentage difference is the difference as a percentage of the Controller’s reported amounts. 138 Non-Enterprise Districts Table G.6 Overall Variance in Revenues for Non-Enterprise Districts, 1991–92 (dollars unless otherwise indicated) District Name Controller Alameda Co. Mosquito Abatement District 1,046,595 Alameda Co. Resource Conservation District 66,912 American River Fire Protection District 17,329,094 Auburn Cemetery District 380,817 Auburn Recreation and Park District 2,261,380 Bay Area Air Quality Management District 26,707,906 Bay Area Housing Authority Risk Mgt Agency 2,228,398 Bay Area Library and Information System 770,476 Brannan-Andrus Levee Maintenance District 128,952 Broadmoor Police Protection District 750,275 Calif. Housing Authority Risk Mgt Agency 1,671,639 Calif. Joint Powers Risk Mgt. Authority 10,426,588 Chino Valley Independent Fire District 10,113,062 City of San Francisco Downtown Parking Co. 3,331,569 City of San Francisco Uptown Parking Co. 5,509,476 Clovis Memorial District 728,882 Conejo Recreation and Park District 10,818,803 Dry Creek Storm Water District 88,722 East Bay Regional Park District 52,037,070 East Bay Schools Insurance Group 1,398,876 Audited Report 1,069,772 64,638 17,334,006 380,817 2,185,931 28,864,381 2,228,398 770,476 125,553 750,275 1,648,684 10,426,588 10,113,062 3,331,569 5,509,477 728,881 10,818,803 89,822 52,645,762 1,398,876 Differencea Percentage Differenceb 23,177 2.2% 2,274 3.4% 4,912 0 0.0% 0.0% 75,449 3.5% 2,156,475 8.1% 0 0.0% 0 0.0% 3,399 2.6% 0 0.0% 22,955 1.4% 0 0.0% 0 0.0% 0 0.0% 1 0.0% 1 0.0% 0 1,100 608,692 0 0.0% 1.2% 1.2% 0.0% 139 Table G.6—continued District Name Eastern Plumas Fire Protection District Controller 46,762 Audited Report 46,762 Differencea Percentage Differenceb 0 0.0% 140 Appendix H Copy of the Controller’s Annual Questionnaire for Counties" ["_edit_lock"]=> string(12) "1540320463:1" ["post_date_gmt"]=> string(19) "2017-05-20 09:34:51" ["comment_status"]=> string(4) "open" ["ping_status"]=> string(6) "closed" ["post_password"]=> string(0) "" ["post_name"]=> string(8) "r_397msr" ["to_ping"]=> string(0) "" ["pinged"]=> string(0) "" ["post_modified"]=> string(19) "2017-05-20 02:34:51" ["post_modified_gmt"]=> string(19) "2017-05-20 09:34:51" ["post_content_filtered"]=> string(0) "" ["guid"]=> string(50) "http://148.62.4.17/wp-content/uploads/R_397MSR.pdf" ["menu_order"]=> int(0) ["post_mime_type"]=> string(15) "application/pdf" ["comment_count"]=> string(1) "0" ["filter"]=> string(3) "raw" ["status"]=> string(7) "inherit" ["attachment_authors"]=> bool(false) }