Donate
Independent, objective, nonpartisan research

R 998MSR

Authors

R 998MSR

Tagged with:

Publication PDFs

Database

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

View live version

object(Timber\Post)#3742 (44) { ["ImageClass"]=> string(12) "Timber\Image" ["PostClass"]=> string(11) "Timber\Post" ["TermClass"]=> string(11) "Timber\Term" ["object_type"]=> string(4) "post" ["custom"]=> array(5) { ["_wp_attached_file"]=> string(12) "R_998MSR.pdf" ["wpmf_size"]=> string(6) "207550" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(143008) "Has Proposition 13 Delivered? The Changing Tax Burden in California Michael A. Shires John Ellwood Mary Sprague 1998 Copyright © 1998 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 The passage of Proposition 13 on June 6, 1978, signaled a dramatic change in California politics as voters placed a cap on the amount of taxes they were willing to pay. Perhaps no other initiative in the past 20 years has been subject to greater scrutiny or provoked more controversy. The policy issues involved, including the question of equity, the fiscal relationship between state and local governments, and the level and quality of government services, are large and important concerns that need to be analyzed with an objective and independent eye. They are exactly the kinds of issues that the Public Policy Institute of California was founded to study. In this report, Michael Shires, John Ellwood, and Mary Sprague answer one of the most prominent questions in the policy debate: How much do we pay to our state and local governments in California and how has that amount changed since Proposition 13? The authors conclude that the answer depends on whether you account for inflation. In nominal dollars per capita, public revenues in California have more iii than doubled in the 20 years since the passage of Proposition 13. But, controlling for inflation, per capita revenues have declined by 16 percent. The authors point out, however, that although the revenue burden is lower today than it was in 1978, it has risen significantly since 1981. This volume is one of a series that PPIC is publishing on the status of public finance in California as part of an overall program of research in governance and public finance. Related reports will soon be available, addressing such issues as the fragmentation of local government, longterm trends in county finances, patterns in state and local government revenues, and the consequences of the 1990’s recession on property valuations under Proposition 13. In addition, a recent PPIC book by Mark Baldassare examines in great detail the bankruptcy of Orange County in 1994. His analysis traces the bankruptcy directly to the exigencies of Proposition 13 and the unending pressure to raise revenues from local sources. The book, When Government Fails: The Orange County Bankruptcy, is available through the University of California Press. We trust that this growing body of research and findings will reduce the level of disagreement about Proposition 13 and its aftermath and set the stage for a more informed public dialogue. David W. Lyon President and CEO Public Policy Institute of California iv Summary Over the past two decades, Californians have engaged in an almost continuous debate over the proper size of their governments. Beginning in 1978 with the adoption of Proposition 13, they have used the direct legislative procedures of the initiative process to limit the growth of their governments’ taxing and spending powers and to mandate that significant portions of their governments’ revenues be devoted to specific purposes. The annual budget processes of most of the state’s governments have also focused on these issues. This report explores the recent history of this debate by analyzing California’s revenue burden since 1978. In it, we will answer the question, How much do we pay to our state and local governments in California and how have those amounts changed since Proposition 13? Unfortunately, there is a lot of debate over how to answer this question. For one thing, citizens do not agree on what should be counted as tax or nontax revenue or on how to measure the burden of taxes and other government revenues. Some revenues are clearly taxes— v the income tax, the property tax, and the sales tax, for example. But should all fees charged by governments be included in the revenue burden? For example, if a student pays fees to a state college, should those fees be included in the state revenue burden even if that student could have attended a private college or gone to an out-of-state public university? How should the income of publicly owned hospitals be viewed, or fees for other services where governments compete with private firms to provide a service? How should we think about fees for a service that the government provides through a monopoly or taxes that the government converts into user fees? Citizens also disagree about the correct way to measure the size of state and local government revenue burdens. For example, if the dollar amount of taxes doubles over a decade, has the revenue burden doubled? Some would say yes. Others would say that the effects of inflation should be taken into account when answering this question. Still others would point out that since average income has also doubled over this period, the actual “burden” has remained unchanged. The Public Revenue Burden in California Since Proposition 13 We find that the public revenue burden in California has declined since Proposition 13, but that it has been increasing since the early 1980s. Figure S.1 illustrates this primary finding of our research. As the figure shows, the public revenue burden in California fell to about 76 percent of its 1978 levels in 1981 and then rose to about 90 percent of its 1978 levels by 1992. Although the decline from 1978 to 1981 is not wholly attributable to Proposition 13—there was a state-level tax cut and vi Percentage 100 90 80 70 60 10 0 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 Figure S.1—Composite Measure of Overall Public Revenues in California a statewide recession between 1978 and 1981—it was certainly a major contributing factor. We also know that the public revenue burden has risen from the 76 percent level in 1981 to 85 percent in 1995. This growth has been largely due to increases in local taxes and charges. These findings are relatively robust, no matter what measure of the public revenue burden is used—whether the revenue burden is considered as a percentage of overall income or on a per-person, inflation-adjusted basis. At the same time, we note that there seems to be some concern on the part of the electorate about the growing public revenue burden, as demonstrated by the passage of Proposition 218 in November 1996. This initiative placed supermajority voting requirements on many local assessments and charges. Figure S.2 may shed some light on this issue. As shown in this figure, the average public revenues per person in the vii Dollars 3,500 3,000 2,500 Current 2,000 1,500 1,000 Real 500 0 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 Figure S.2—Per Capita Public Revenues, in Real (1978) and Current Dollars state have declined when adjusted for inflation. However, in absolute terms, the revenue burden has increased dramatically over the past two decades, rising from $1,500 per person in 1978 to nearly $3,000 in 1995. Local governments are also charging residents for services that were free in the past, such as access to local parks. The result is a public finance system that appears to cost more to sustain while simultaneously charging more for the services it does provide. Implications for California Public Policy These findings have several implications for public policy in California. First, the pervasiveness of the study’s findings across measures and time shows that how to measure the changing revenue burden should not be the focal point of the debate—our analysis showed similar results with both measures. Furthermore, even if alternative measures of the revenue burden are used, the results are consistent. We viii hope that this finding will clarify the confusion introduced by several previous studies on the topic and will allow the debate to focus instead on the critical issue of the appropriate size of state and local governments. Beyond this direct contribution, we believe that this study speaks to the future prospects for the public sector in California. Perhaps in response to the growth in the size of the state and local sectors since the implementation of Proposition 13, the voters of California have placed significant additional constraints on the public sector through the passage of Proposition 218, which significantly constrains the ability of local government to creatively expand its revenues through several of the mechanisms that were the mainstays of revenue growth during the late 1980s and 1990s. Interestingly enough, this initiative seems to be more of a response to “general government growth” than to specific local needs, because voters seem to be passing many of the measures that Proposition 218 has brought to the ballot. It is likely, however, that the requirements imposed by Proposition 218 and their attendant logistical, political, and fiscal costs will slow the growth of new local revenue streams in the future—especially in periods of recession. In light of the rates of growth in real revenues identified in this study, it follows then that the resources to fund expansions in the level of services provided at the state and local levels will grow, at best, slowly. There does seem to be some hope for specific programs and initiatives, as voters have in recent years been much more receptive to local bonds and finance measures. Because a large share of state and local revenues is derived from taxes, it is unlikely that state and local governments will experience funding shortfalls during periods of high economic growth, when tax coffers swell. In recessionary periods, however, the ability to raise additional ix revenues through increased license fees, service charges, and user fees— state and local governments’ response during the last recession—will be constrained by both Proposition 218 and the extensive use of such fees during the last recession. This combination could leave California’s state and local budgets sensitive to economic shocks and could result in reductions in public support for discretionary programs, such as higher education, with the onset of a future recession. This study has shown that Proposition 13 has contributed to a significant rollback of the public revenue burden. Although the effects of this rollback continue today, its full effect has been partially offset by growth in the public revenue burden in the intervening years. It is possible and even probable, however, that Proposition 218 will limit state and local governments’ ability to continue this growth and that the future size of the public revenue burden will remain at its current levels into the future. x Contents Foreword ..................................... Summary..................................... Figures ...................................... Tables ....................................... Acknowledgments ............................... iii v xiii xv xvii 1. INTRODUCTION ........................... The Debate over State and Local Finance ............. An Anatomy of Public Finance in California ........... How Can Revenue Burdens Differ? ................. What Others Have Found ....................... Steven Gold ............................... California Taxpayers’ Association ................. Kirlin et al. ............................... California Legislative Analyst’s Office .............. Why We Need Another Study ................... Our Study ................................. Organization of This Report ...................... 2. DEFINING THE STATE AND LOCAL REVENUE BURDEN IN CALIFORNIA ..................... The Public Revenue Burden in California: A Series of Choices ................................ A Taxonomy of Revenue Types .................... 1 3 6 11 12 14 14 15 16 16 17 18 19 20 23 xi Figures S.1. Composite Measure of Overall Public Revenues in California ............................... vii S.2. Per Capita Public Revenues, in Real (1978) and Current Dollars................................. viii 1.1. Reported Public Revenues of California State and Local Governments, by Revenue Base, FY 1994–95 ........ 9 2.1. Overall Share of State and Local Revenues, by Revenue Type .................................. 33 4.1. Composite Measure of Overall Public Revenues in California ............................... 58 4.2. Two Measures of Public Revenues in California....... 59 4.3. Per Capita Public Revenues, in Real (1978) and Current Dollars................................. 61 B.1. Annual Growth in California Real per Capita Personal Income................................. 74 xiii Tables 1.1. FY 1994–95 Total Public Revenues in California, by Level of Government Receiving the Revenue ......... 1.2. Percentage Share of FY 1994–95 Revenues from Various Sources, by Level of Government Receiving the Revenue ................................ 1.3. Selected Findings of Prior Studies on Public Revenues in California, as a Percentage Share of Personal Income.... 2.1. Public Revenues in California for All Levels of Government, and Their Percentage Share, by Level Receiving the Revenue ....................... 2.2. Public Revenues in California for All Levels of Government, and Their Percentage Share, by Revenue Type .................................. 2.3. Revenues Excluded from Our Estimate of Public Revenues in California for All Levels of Government, by Revenue Type ............................ 2.4. Remaining Public Revenues in California for All Levels of Government, and Their Percentage Share, by Revenue Type .................................. 3.1. Public Revenues in California as a Share of Personal Income for All Levels of State and Local Government ... 8 10 13 21 32 39 41 49 xv 3.2. Tax Revenues in California as a Share of Personal Income for All Levels of State and Local Government ... 3.3. TAR Revenues in California as a Share of Personal Income for All Levels of State and Local Government ... 3.4. Per Capita Public Revenues, Tax Revenues, and TAR Revenues in California for All Levels of State and Local Government ............................. 3.5. Real per Capita Public Revenues, Tax Revenues, and TAR Revenues in California for All Levels of State and Local Government ......................... A.1. Comparison of Estimated Revenues, Using Steven Gold’s Methodology, as a Percentage of Personal Income................................. A.2. Comparison of Estimated Revenues Using the Cal-Tax Methodology, 1986–87 ...................... A.3. Comparison of Estimated Revenues Using Kirlin et al.’s Methodology, 1986–87 ...................... C.1. Measures of California’s Public Revenues if General Services Are Included ....................... C.2. Measures of California’s Public Revenues if Enterprise Revenues Are Included....................... C.3. Measures of California’s Public Revenues if Both General Service Revenues and Enterprise Revenues Are Included ................................ 51 51 53 56 69 70 72 78 79 80 xvi Acknowledgments As is the case with any significant research project, the authors of this study are indebted to many individuals and organizations for their contributions and assistance. First and most important, we are indebted to the 13 individuals who early in the project provided us with the benefit of their wisdom and experience: Michael Coleman; John Decker, Resource Subcommittee, Assembly Budget Committee; Joel Fox, Howard Jarvis Taxpayers’ Association; Lenny Goldberg, California Tax Reform Association; Steve Kamp, Assembly Revenue and Taxation Committee; Stephen Kroes, California Taxpayers’ Association; Bob Leland, City of Fairfield; Anne Maitland, Senate Revenue and Taxation Committee; April Manatt, Senate Local Government Committee; Marianne O’Malley, California Legislative Analyst’s Office; Jean Ross, the California Budget Project; and Connie Squires, Department of Finance. We are also grateful for the time and efforts of those who served as formal reviewers of this study, including Mark Baldassare, PPIC; Jeff xvii 1. Introduction Over the past two decades, Californians have engaged in an almost continuous debate over the proper size of their governments. Beginning in 1978 with the adoption of Proposition 13, they have used the direct legislative procedures of the initiative process to limit the growth of their governments’ taxing and spending powers as well as to mandate that significant portions of their governments’ revenues be devoted to specific purposes. The annual budget processes of most of the state’s governments have also focused on these issues. This report explores the recent history of this debate by analyzing California’s revenue burden since 1978. In it, we answer the question, How much do we pay to our state and local governments in California and how have those amounts changed since Proposition 13? Unfortunately, there is a lot of debate over how to answer this question. For one thing, citizens do not agree on what should be counted as tax or nontax revenue or on how to measure the burden of taxes and other revenues. Some revenues are clearly taxes—the income 1 tax, the property tax, and the sales tax, for example. But should all fees charged by governments be included in the revenue burden? For example, if a student pays fees to a state college, should those fees be included in the state revenue burden even if that student could have attended a private or out-of-state college? How should the income of publicly owned hospitals be viewed, or fees for other services where governments compete with private firms to provide a service? How should we think about fees for a service that the government provides through a monopoly or taxes that the government converts into user fees? Citizens also disagree about the correct way to measure the size of state and local government revenue burdens.1 For example, if the dollar amount of taxes doubles over a decade, has the revenue burden doubled? Some would say yes. Others would say that the effects of inflation should be taken into account when answering this question. Still others would point out that since average income has also doubled over this period, the actual “burden” has remained unchanged. They would argue that the burden should be measured as a function of people’s ability to pay, much the same way that banks consider a person’s income when deciding whether to grant a loan—the higher the person’s income, the more he or she can borrow. Yet others would argue that the size of the burden and the services it funds should be considered exclusively on its own merits. ____________ 1Note that “public revenue burden” will be used interchangeably with the phrase “state and local revenue burden” throughout the report. Although the concept of the public revenue burden in its most general sense would include federal revenues, the state/federal debate is not the focus of this policy debate or of this study. Consequently, we will use the public revenue burden concept only to refer to the state and local portion of that burden. 2 In this report, we will discuss and define what should be included in the public revenue burden in California and explore the main measures of that burden. A companion background paper (Shires, 1998) discusses how different levels of California governments—the state, counties, cities, special districts, school districts, and public postsecondary education institutions—have relied on different sources of revenue over the past 20 years. This report and the companion background paper provide the reader with the information necessary to calculate the average revenue burden under a variety of definitions for a variety of classifications of revenues and taxes for each level of California state and local government for five fiscal years between the passage of Proposition 13 in 1978 and fiscal year 1994–95 (the most recent year for which full revenue data are available from the California State Controller’s Office). The Debate over State and Local Finance Whether they intend to or not, all governments—through their annual taxing and spending decisions—provide answers to the following questions: How large should the public sector be? The size of the revenues and budgets controlled by governments reflects the long-run preferences of the citizens within a jurisdiction. Governments can only spend monies that the voters authorize either directly through the ballot or indirectly through the election of representatives sympathetic to the imposition of taxes and fees to raise those monies. In California, which makes extensive use of the initiative process, voters often make their preferences directly known. Also, the size of the public sector differs across states and across jurisdictions within a given state, because some local 3 governments raise more revenue than others. The citizens of Hawaii, New York, and Minnesota, for example, are noted for their willingness to raise more revenues than their fellow citizens in New Hampshire and Texas. Which goods and services should be provided and how much should be spent on a given good or service? Governments differ as to whether they will provide a given good or service and how generous they will be in its provision. They must decide how much should be spent on elementary or secondary education, on postsecondary education, on police, on fire protection, on the correctional system, on public parks, and on welfare benefits and services The services they provide depend on the resources they receive. Which government should provide which good or service? Even when two states provide approximately the same goods and services they frequently differ as to which level of government should provide these goods and services. Should medical services for the poor and elderly be provided by the state government, by county governments, or by city governments? In California, even though these services are provided at the local level, the MediCal program that controls and administers them is housed at the state level. In New York, these services are both funded and administered by local governments. California also provides many goods and services through special governmental units, called special districts, that in some other states are provided by regular state and local governmental units. As a result of this variety of institutional frameworks, it is misleading to look only at the pattern of revenues and expenditures at any one level of government because activities are sometimes shifted from one governmental unit to another. Instead, one 4 must take a broader view and look at all levels of state and local government in concert. How do governments raise the revenue needed to fund these goods and services? Traditionally, governments have raised their revenues through taxation. At the state and local levels, the traditional taxes have been the income, property, and sales taxes. Over the years, state and local governments have imposed different taxes as well as different levels of those taxes. California, for example, relies on income, property, and sales taxes, whereas New Hampshire relies almost exclusively on the property tax. Some observers have claimed that in California, as traditional governmental units have been restrained in their ability to raise taxes, the provision of goods and services has been shifted to other units, especially special districts; and, as the taxing power of traditional governmental units has been constrained, they have found new ways, particularly through fees and exactions, to fund the provision of public sector goods and services. Who bears the burden of providing the revenues? Economists traditionally ask two sets of questions when analyzing the economic effects of taxation: (1) Does a given tax distort the efficiency of economic activity, and (2) who bears the burden of that tax? The question of who bears the burden of taxation is usually answered in terms of whether the tax burden is greater on one income class than another. For example, are taxes a greater proportion of the incomes of those at the upper income levels than of those at the lower income levels? The fact that different state and local governments raise different amounts of revenue through different mechanisms means that the revenue burden can differ not only across geographic jurisdictions but also across income 5 categories. Not only will a citizen of Tennessee pay lower taxes than an equivalent citizen of Hawaii, but even within a state, a citizen of a city or county that provides minimal services will likely pay less than the equivalent citizen of a city or county that is generous in its provision of goods and services. Who will benefit from the goods and services? Not only does who pays for state and local government services differ, but also who receives the services. In most state and local governments, a certain amount of redistribution occurs when governments raise revenues for certain programs from groups of residents who do not directly receive the benefits of the services funded by their taxes. The magnitude and role of this redistribution, whether intended or not, will differ not only across states but also across local jurisdictions within states. This report addresses only some of these questions—those related to the revenue side of the budgetary equation. It does not address the magnitude and distribution of state and local spending, the magnitude and purposes of public sector borrowing, or the broad issue of efficiency in government. Its focus on the revenue burden, however, is important and relevant for two reasons. First, at the state and local levels where governments are mandated to balance their operating budgets, the amount of revenue raised has a significant effect on the size and mix of spending. Second, the debate over the appropriate size of the revenue burden has been central to the ongoing California debate over the appropriate size and role of government. An Anatomy of Public Finance in California The United States is governed under a federal system. Under the American Constitution, sovereignty is jointly held by the national and 6 state governments. The national government is granted a set of specific powers under the Constitution, and all other powers are reserved for state governments and the people. Local governments2 have no sovereign power other than that granted to them by their states. Sometimes this power is provided for in the state’s constitution and sometimes by statute. Because local governments are creations of their respective states, their powers to raise revenues are set out in the state’s constitution and in state statutes. In a similar fashion, because national laws take precedence over state laws, it is sometimes the case that a state’s power to raise revenues is constrained by actions of the federal government. In practice, this federal system has created a complex set of fiscal relationships between and among the states, their local governments, and the national government. The powers of the national government are not fixed but have changed over time as the Supreme Court has interpreted the Constitution. This has meant that, since Franklin Roosevelt’s New Deal of the 1930s, the federal government has increasingly funded and regulated what in prior years were thought to be solely state and local activities. In addition to the increasing complexity of federal involvement in state and local finance, the institutional framework created by California has rendered the state’s public finance framework almost incomprehensible. In 1995, California had nearly 5,000 independent governments. In addition to the state government, it had 58 counties, 470 cities, 3,217 independent special districts,3 and 1,001 school ____________ 2The term “local governments” will refer throughout this report to governmental entities within and below a state government, including counties, cities, special districts, and school districts. 3Independent special districts are those governed by either independently elected boards or boards appointed by multiple jurisdictions. 7 districts. In addition, California funds three public postsecondary education systems: the University of California system (with nine campuses), the California State University system (with 22 campuses), and the California Community Colleges system (with 71 districts and 106 campuses). The situation is further complicated because the various governments of California receive grants and other transfers from both the federal government and the state. As indicated in Table 1.1, in fiscal year 1995 these governmental units reported more than $204 billion in total revenues. Although, as we will find in Chapter 2, this total includes several categories of revenues that we may wish to exclude from our estimation of the size of state and local revenue burdens in California,4 the size and scale is still considerable. California governments obtain their revenues through taxes, fees and fines, various types of activities for which a charge is levied (such as public utilities), and intergovernmental transfers. Figure 1.1 shows us Table 1.1 FY 1994–95 Total Public Revenues in California, by Level of Government Receiving the Revenue Government Entity State Counties Cities Independent special districts School districts Public postsecondary institutions Revenues ($ Billions) 85.6 31.9 30.8 12.7 27.7 15.9 % of State and Local Total 41.8 15.6 15.1 6.2 13.5 7.8 Total 204.6 100.0 SOURCES: Compiled from numerous state agency publications. ____________ 4In the case of intergovernmental revenues, adding these reported revenues can result in double-counting those revenues, as we will see in Chapter 2. 8 Service 5% Interest 2% Enterprise 11% Other 3% Property tax 10% Sales tax 10% Income tax 12% Intergovernmental transfers 39% Figure 1.1 Activity-based fees 8% Figure 1.1—Reported Public Revenues of California State and Local Governments, by Revenue Base, FY 1994–95 that these governments receive the largest share of their revenues from other levels of government, which account for 39 percent of state and local government revenues. Income, sales, and property taxes combined account for nearly one-third of these reported revenues, whereas regulatory fees account for 8 percent and enterprise activities (such as water and sewer services) account for another 11 percent. The level of reliance on any one of these revenue types differs with the level of government, as Table 1.2 shows. This table sets out the percentage of fiscal year 19955 revenues that resulted from each of these major categories for each level of California state and local government.6 ____________ 5In California, the fiscal year extends from July 1 to June 30 of each year. Hence, fiscal year 1995 is the period July 1, 1994, through June 30, 1995. 6Note that some of the revenue streams provided in this table, especially those relating to cities, counties, and special districts, do not directly correspond to the detailed totals reported for each entity by the California State Controller. This is due to the reclassification of dependent special districts’ revenues into their respective parent entities’ revenues. For example, the revenues from county service areas, which are created and 9 Table 1.2 Percentage Share of FY 1994–95 Revenues from Various Sources, by Level of Government Receiving the Revenue Revenue Type Property taxes Sales taxes Income taxes Regulatory fees Intergovernmental Interest Charges for non- enterprise services Charges for enterprise services Other Public Post- Special School secondary State Counties Cities Districts Districts Institutions 0.0 12.5 14.5 10.2 32.0 8.4 19.8 1.5 10.2 2.1 0.0 0.0 29.0 0.0 0.0 0.0 0.0 0.0 12.7 2.7 12.7 0.8 3.2 0.0 37.1 55.9 14.0 20.8 59.6 52.6 0.6 1.8 4.4 6.3 1.7 0.5 0.4 7.1 5.4 0.5 2.5 28.5 0.0 13.9 31.9 52.1 0.0 7.4 0.4 4.6 6.9 7.2 1.0 2.6 Total 100.0 100.0 100.0 100.0 100.0 100.0 NOTE: The City and County of San Francisco is included as a city. Dependent special district revenues have been included in parent entities’ revenues. One clear picture that emerges from the percentages shown in Table 1.2 is that although the state government raises most of its revenues from taxation, each of the other levels of government obtains less than a third of its revenues through traditional taxation. Counties and school districts depend highly on intergovernmental transfers of funds (primarily from the state), whereas independent special districts obtain more than 60 percent of their revenues from the enterprises they run (such as providing water and fire protection). California’s cities have the ____________________________________________________ managed by county boards of supervisors, are included as county revenues, not as special district revenues, as is done in the State Controller’s reports. Furthermore, the revenues for redevelopment agencies are also reported in the revenues of their respective parent entities, typically cities. As a result, although the proportions shown in this table do not directly correspond to other published sources, we believe that they more accurately reflect the distribution and control of the monies they represent. 10 most diverse sources of revenues; but even in their case, only one-quarter of their revenues are generated from locally levied taxes; over half come from activities, enterprises, and services provided for a fee. How Can Revenue Burdens Differ? The vast array of California governments that fund their activities through a variety of mechanisms means that two neighbors can face very different revenue burdens. The most obvious and familiar case occurs when residents are separated by a city or a county line. Frequently, adjoining cities or counties provide very different levels of services and, to fund these, levy very different levels and mixes of taxes and fees. But even within the same city or county, neighbors can face different revenue burdens. This happens because the boundaries of California’s special districts and school districts are not necessarily contiguous with the bounds of its cities and counties. Thus, two adjoining neighbors might find themselves located in different school districts, fire districts, flood control districts, mosquito abatement districts, and so on. These districts often have different fee structures and, sometimes, different tax levels. Neighbors can also face different revenue burdens depending on whether they take advantage of publicly provided services that are elective. If one neighbor regularly uses public transportation and another walks to and from work, the two would pay different amounts to the government. If one neighbor uses state or county parks each weekend and his neighbor stays at home, the former will bear a greater share of the revenue burden since he will regularly be paying park fees. Even when it comes to taxes, neighbors may face very different revenue burdens. Because Proposition 13 limits increases in the property 11 tax to 2 percent per year but allows full reassessment every time a property is sold, two neighbors in identical houses might pay very different property taxes if one bought his house in 1974 and the other bought his house in 1988. In another instance, one neighbor might buy a how-to book on sailing, and the other might take sailing lessons. The first would pay sales tax on the book but the second would pay no sales tax on the sailing lessons. For all of these reasons, it is difficult to calculate the revenue burden for each California citizen, but that is not a goal of this study. Rather, we wish to determine the size of the financial burden the government imposes on society and how it has changed since the passage of Proposition 13. What Others Have Found Clearly, thinking about Proposition 13’s effects on what people pay in taxes is not a new idea. Several studies have addressed this question. For example, Dunstan (1993) examined long-term trends in state, city, and county finances from 1975–76 to 1990–91, focusing on the major revenue and expenditure categories. This study, however, did not include the revenues and expenditures of three important categories of public entities—school districts, independent special districts, and public postsecondary education institutions. Sheffrin and Dresch (1995) used an economic approach to examine the tax burden, assuming that all taxes are eventually passed on to individuals in the form of higher prices for goods and higher rents. Although arguably an appropriate methodology, this study looked only at the three main categories of revenues—personal income tax, sales and use tax, and residential property tax—identifying who ended up paying 12 these taxes and how much they paid. The study did not include the many other forms of revenues that the public sector generates. Although these two studies appear to be directly relevant to our question, they looked only at a portion of the revenue burden. Four other studies, however, have more specifically addressed the question we are studying. Unfortunately, the underlying assumptions and results of these studies have differed widely and their findings have raised more questions than they have answered. Table 1.3 presents some selected findings from these studies. A detailed examination of this work helps explain why we have chosen to undertake this project. The need for our research arises, not from the failing of any of these studies but rather from the complexity and range of assumptions that are implicit in an examination of the state’s complicated state and local finance landscapes and the need to place the diverse findings of these previous studies in context. In addition, the recent passage of Proposition 218 seems to indicate that the public perceives that there is still a need to restrict local government’s ability to manage and increase its revenues. Since none of Table 1.3 Selected Findings of Prior Studies on Public Revenues in California, as a Percentage Share of Personal Income Study Gold California Taxpayers’ Association Kirlin et al. Legislative Analyst’s Office 1977–78 14.6 16.7 18.7 — 1990–91 11.3 16.0 — — 1991–92 — 16.2 19.0 16.6 SOURCES: Gold (1993); California Taxpayers’ Association (1994; 1991–92 estimate obtained separately from the California Taxpayers’ Association); Kirlin et al. (1994); Legislative Analyst’s Office (1995). 13 the previous work provides us with the most recent information available, our study serves a valuable function by including data from fiscal year 1994–95. Each study in Table 1.3 measured the public revenue burden with slightly different assumptions and concerns. We briefly discuss each of these studies below and then discuss in greater detail why we felt that this follow-on study was needed. Appendix A contains a more detailed discussion of our efforts to reconcile our findings to these previous studies. Steven Gold Steven Gold’s study was included as part of his testimony at a joint hearing of the California Senate’s Budget Committee and Human Services Committee on October 1, 1992. The study also served as the basis for his presentation at a symposium on “State Taxation of Business Activities” at UC Davis in March 1993. Gold’s approach to estimating the public revenue burden focused exclusively on taxes. As Table 1.3 shows, his research found that state and local taxes in California declined from 14.6 percent of personal income in 1977–78 to 11.2 percent in 1990–91. The analysis focused only on the tax portion of public revenues and did not include other miscellaneous revenue categories such as fees, interest, and fines. California Taxpayers’ Association California Taxpayers’ Association (1994) presented estimates of the public revenue burden in California, at least through 1990–91. Its analysis was based on U.S. Bureau of the Census reported amounts and showed a near return to pre-Proposition 13 revenue levels by 1990–91. 14 Although these findings are quite compelling, there are several issues to consider when using Census data. The Bureau of the Census recompiles State Controller’s data to include subsidiary entities’ revenues with their parents’ revenues while simultaneously converting California’s revenue category definitions into the Census’s more generic structure. Preliminary comparisons of the work done for this study to the Census’s reported revenue amounts indicate some significant variation in the totals and subtotals reported for the various revenue categories—including the total revenues reported by each level of government.7 Because of our exhaustive efforts to aggregate and classify the data at the lowest level of detail possible and the confidence we have in the data we are using,8 we believe that our estimates, which will be presented in Chapters 2 and 3, more appropriately reflect state and local government revenues. Kirlin et al. The Kirlin et al. study was prepared for the Task Force on California Fiscal Reform of the California Business—Higher Education Forum. It was published as Chapter Five in the Forum’s Task Force Member Report California Fiscal Reform: A Plan for Action. Tables 5.1 through 5.3b of the report (pp. 45–48) contain detailed estimates of the public revenue burden in California for the fiscal years 1977–78, 1988–89, and 1991–92. ____________ 7Although the possibility of inconsistencies and classification problems in the Census data have been raised by others (for example, see Leigland, 1990), this study is the first to do a detailed recalculation for California. For this study, it was necessary to keypunch and verify data from the 300 to 600 pages of published reports for each year studied. We have, therefore, excellent revenue data, which should correspond directly to the revenues reported by the Bureau of the Census. The reported Census revenues will be debated in another paper. 8See Shires and Glenn Haber (1997) for a detailed discussion of the quality of our data. 15 The authors’ data and the assumptions implicit in their study correspond most closely to those we use here, but there are some issues associated with their data that are problematic. First, because the 1991– 92 special district data were not available when they prepared their estimate of the state and local revenue burden, they had to use 1990–91 data for this important information. Furthermore, reporting inconsistencies in the data for redevelopment agencies, transit districts, school districts, and some public postsecondary institutions create a scenario where those estimates are not fully comparable from year to year. We have tried to correct for these inconsistencies in our study. California Legislative Analyst’s Office The Legislative Analyst’s Office estimate presented in Table 1.3 comes from the “State and Local Finance—Fiscal Overview” section in the Legislative Analyst’s Office’s 1995 Cal Guide: A Profile of State Programs and Finances. The estimate of the state’s public revenue burden was taken from a U.S. Department of Commerce estimate of the state’s public revenue burden. Since this analysis used Bureau of the Census data, it calls up the same issues that were raised about the California Taxpayers’ Association estimates above. Why We Need Another Study Even with the results of these four studies in hand, the subject needs to be reexamined for several reasons. First, we hope to resolve some of the uncertainties raised by the differing answers found in these studies. Second, we hope to provide a time series of the public revenue burden in California that is directly comparable from year to year. As the discussion of the four studies above demonstrates, significant data issues 16 have not previously been fully addressed. Our report will do so. We also hope to bring the findings of these studies up to date—since most of the studies were completed for years during or before the relatively severe recession of the early 1990s. Our updated study will also allow us to assess the effects of the many policy changes instituted as a result of the recession. Finally, we hope to fully address the question of how to best measure the public revenue burden in California. Our Study This study is the second of a three-part series on state and local finance and the effects of Proposition 13 on local citizens. Our first study (Shires and Glenn Haber, 1997), which examined the quality of the data available on the revenues of state and local governments in California, found the aggregate data to be both comprehensive and accurate. This report presents the findings of our overall assessment of the public revenue burden in California. It looks at the issue of the public revenue burden at the statewide level and in aggregate numbers, focusing on what should be included in that revenue burden and the best way to measure it. The third and final study in this series will focus on how Proposition 13 affected individuals in California. It will examine the effects on individual residents of the state of the changed fiscal structures and institutions documented in theses two reports. Because of the significant costs associated with preparing the data for each year, we will review only five years in our study. However, we believe that the five years we have selected will provide a good picture of what has happened to the public revenue burden in the years since 17 Proposition 13. This report looks in great detail at the public revenue burden in California for the fiscal years 1977–78, 1980–81, 1987–88, 1991–92, and 1994–95.9 Although there were many reasons for choosing these years, our primary reasons centered on the timing of Proposition 13 and subsequent California business cycles. We selected fiscal year 1978 as a baseline for the research because it was the year that Proposition 13 was passed by the voters. 1981 was selected because it falls soon after the full effects of the implementation of Proposition 13 were able to work their way through the public finance system. 1988 and 1992 were chosen for their comparability as points in the business cycle10 and 1995 because it is the most recent year for which data are available. For a more detailed explanation of our choice of these five years, see Appendix B. Organization of This Report This study has three remaining chapters. Chapter 2 explores the issue of what public revenues should be included in an estimate of the public revenue burden and Chapter 3 examines how we should measure the size of the public revenue burden. Chapter 4 then discusses the findings presented in these two chapters in the context of the policy debate presented in the introduction and the implications of these findings for citizens and policymakers. ____________ 9Throughout this report, the year described will refer to the fiscal year ending in that year unless otherwise noted. For example, a reference to “1978” would be referring to the fiscal year ending June 30, 1978, and is also written as 1977–78. 10Comparability across the business cycle refers to the fact that each of the two years, 1988 and 1992, represents a point in the business cycle (e.g., a period of economic expansion or recession) that was similar to our first two years, 1978 and 1981, respectively. 18 Chapman, USC School of Public Administration; Stephen Kroes, California Taxpayers’ Association; Paul Lewis, PPIC; and Marianne O’Malley, California Legislative Analyst’s Office. Their comments and suggestions for revisions and improvements to this report have been invaluable. We would also like to thank Marcela Cordon, who spent long hours keypunching and validating the raw data necessary to this study. The project also benefited from the patience and efforts of Michelle Chaffee, formerly of PPIC, who spent many days poring over the data tables for this study to make sure that we were using the right numbers. Finally, we also wish to acknowledge the contribution of PPIC’s Advisory Council and Board of Directors, who provided helpful direction, comments, and suggestions during our numerous briefings and discussions. Although this report reflects the contributions of many people, the authors are solely responsible for its content. xviii 2. Defining the State and Local Revenue Burden in California The question, What has happened to the state and local revenue burden in California since Proposition 13? has two parts: (1) What are public revenues in California and (2) how have they changed over time? In this chapter, we will explore the first part of the answer to that question: What should be included as public sector revenues when we think about the public revenue burden? First we will discuss the types of revenues that are commonly reported by governments and agencies at the state and local levels. To help us do that, we will develop a taxonomy or menu of public revenues into which we will subsequently group all public revenues. We will then look at this taxonomy and select only those revenues that appropriately belong in our study. As we will see, we may not wish to include certain types of revenues in our estimates. We will then present the totals of these revenues for the years we studied. 19 The Public Revenue Burden in California: A Series of Choices As the differences among the studies presented in Table 1.3 show, estimates of the overall size of the public revenue burden depend heavily on the types of revenues included. For example, one might include university fees in the public revenue burden because these institutions are indeed public. However, these institutions reflect one area where the public sector is in direct competition with the private sector and their fees represent a specific choice made by state residents to purchase education from the state instead of the private sector. Residents are not required to attend a public university and incur this fee. To determine the kinds of revenues that are claimed by state and local government entities, we looked at the standard reports that state and local governments provide which include details of their revenues. For the state, this source is Schedule 8 of the California Governor’s Budget Summary, which is available for each year.1 For cities, counties, special districts, and redevelopment agencies the appropriate source is the Annual Financial Transactions series published by the California State Controller’s office. For K–12 school districts, our source included the State Controller’s reports and a special data analysis prepared for us by the California Department of Education. Finally, we used information from the California Postsecondary Education Commission to examine the revenues of the state’s public postsecondary education institutions. The overall revenues reported by each level of state and local government for each of the years in our study are presented in Table 2.1. ____________ 1In 1978 and 1981, the corresponding schedule was Schedule 2. Federal intergovernmental revenues for the state were taken from Schedule 6 for 1978, from Schedule 3 for 1981, and from Schedule 9 for 1988, 1992, and 1995. 20 Table 2.1 Public Revenues in California for All Levels of Government, and Their Percentage Share, by Level Receiving the Revenue Level Receiving Revenue State 1978 1981 1988 23,202,407,119 32,386,866,000 52,915,676,000 39.9 41.8 39.8 County 9,182,173,418 11,002,462,589 19,629,130,409 15.8 14.2 14.7 City 8,472,134,749 11,329,778,338 20,248,790,547 14.6 14.6 15.2 Independent special districts 3,665,344,143 5,282,467,621 10,274,183,865 6.3 6.8 7.7 School districts 8,978,391,928 10,963,606,830 18,805,199,859 15.4 14.2 14.1 Public postsecondary education 4,630,040,287 6,509,053,000 11,376,175,000 8.0 8.4 8.5 1992 79,313,651,000 42.3 29,873,677,459 15.9 27,325,541,671 14.5 11,724,148,380 6.2 24,915,087,181 13.3 14,696,063,000 7.8 1995 85,622,226,000 41.8 31,858,760,981 15.6 30,796,774,219 15.1 12,726,996,423 6.2 27,674,571,941 13.5 15,905,264,000 7.8 Total 58,130,491,644 77,474,234,378 133,249,155,680 187,848,168,691 204,584,593,564 100.0 100.0 100.0 100.0 100.0 21 As this table shows, the public sector overall is large—totaling about $205 billion dollars in fiscal year 1995—and has more than tripled over the approximately 20 years shown in the table. Note that this table includes an entry for independent special districts but not for dependent special districts, because dependent special districts are included as part of their parent governments.2 Clearly, all levels of government grew significantly over this time period—as one might expect in a state where the population grew by more than 40 percent. Overall, these revenues grew about 252 percent— a significant portion of which can be accounted for by inflation. If one considers the shares of revenues generated by each level of government, it is clear that they are quite stable, with the exception of school districts. We see that the share of revenues reported by school districts declined from 15.4 percent to 13.5 percent of total state and local revenues. Part of the explanation for this is that school district revenues grew at a slower rate than overall revenues, increasing by only 208 percent from 1978 to 1995, whereas the revenues of some levels of government, such as cities, grew much more quickly—rising 264 percent. Although some may find this table interesting, it is too summary in nature to allow us to fully understand the public revenue burden in California. It contains all revenues reported by state and local ____________ 2This distinction includes redevelopment agencies as well as all special districts that are listed in the State Controller’s reports as being governed by a city council or board of supervisors. The specific revenues within each dependent district were classified individually (e.g., tax or intergovernmental), but they were designated as revenues to the parent government (usually a city or county) instead of being classified as special district revenues. As a result of this adjustment, the totals for cities and counties will be slightly higher than the raw number included in the State Controller’s reports. For a detailed discussion of the amount our revenues reclassified in this manner, please see Shires (1998). 22 governments, including many that we may not wish to include. For this reason, we must examine these revenues in greater detail. To do this, we have created a taxonomy to classify public revenues in California. The classification of revenues into this taxonomy is not an easy process because of the numerous inconsistencies in the way revenues are reported by different types of governments. To develop our taxonomy, we start with the standard revenue categories commonly used for cities and counties and then expand them to address some of the issues we raised above, especially with respect to regulation and service areas where the government is not the exclusive provider of a service. The resulting taxonomy is presented in the next section along with its application to the revenues shown in Table 2.1. A Taxonomy of Revenue Types The State Controller’s reports for cities and counties divide their revenues into eight categories: taxes, special benefit assessments, licenses and permits, fines and forfeitures, revenues from the use of money and property, intergovernmental funds, current service charges, and other revenues. There are some limitations to these categories, however. Some of them disguise aspects of the revenues that we care about. For example, current service charges include such diverse revenues as university fees, water revenues from a municipal utility, filing fees for zoning permits, copying charges for public records, and fire department charges for false alarms. To help us draw distinctions among these revenues and because of the differences in the way information is reported for the different levels of government, we have modified the taxonomy to include ten 23 categories. We have separated current service charges into three categories—enterprise revenues, other service revenues for which the government is the exclusive service provider, and revenues generated by government operations that compete with the private sector. We have also broadened revenues from the use of money and property, except interest, into this latter category.3 We have also broadened the licenses and permits category to include regulatory fees and charges. This change serves two purposes that the original classification did not easily allow: (1) It allows for easier classification of revenues from other levels of government (especially the state), and (2) it recognizes a category of revenues that the courts have, in the past, required governments to reinvest into the regulated activity (regulatory fees and charges). As a result of these changes, our taxonomy has the following categories: taxes; assessments; regulatory fees and charges; fines, penalties, and forfeitures; intergovernmental revenues; interest; enterprise revenues; service revenues from activities where the government is the exclusive provider; service revenues from general services; and other revenues. Below is a brief description of each of these revenue categories. Taxes Taxes are fees paid to the government by residents who choose to participate in an activity, such as owning property, buying general goods, earning income, buying gasoline or alcoholic beverages, participating in business activity, operating a franchise, staying in a hotel room, or purchasing electricity. ____________ 3We argue that, in general, the government is in fact competing with private savings and real estate markets when it earns interest and rents from its assets. An exception to this rule would arise, for example, in the case of rents in a port or airport where the government is the sole provider of space. 24 Assessments Assessments are generated as the result of specific voter action to pay for specific services. They differ from general property taxes because their level is based on an estimate of the benefit they will provide to a specific property and on the actual cost of the improvement, rather than on the overall value of the property. They are most often generated under the auspices of voter-approved ballot measures, as required by Proposition 13, and, more recently, Proposition 218. Unfortunately, as prior work by PPIC and others has shown,4 the amounts reported by local governments under this heading do not reflect the full range of assessment revenues received by local governments. The distinction is retained in our taxonomy, however, because these revenues are significantly different from the tax revenues described above. In the past, these revenues have been imposed both at the behest of the voters and through the intervention of locally elected officials, although the passage of Proposition 218 has resulted in assessments being handled much more like general property taxes than had previously been the case. Regulatory Fees and Charges This category is quite similar to the general tax category above in that it represents a revenue stream where the government charges a fee ____________ 4Shires and Glenn Haber (1997) found that revenues from Mello-Roos assessments were commonly unreported or were reported by local governments as part of overall property tax revenues. The size of the problem was not found to be large relative to overall state and local revenues, but it is significant if one wishes to focus exclusively on assessments. In general, the current state and local government revenue-reporting structure is not adequate to identify revenues from special assessments. This fact was recently highlighted by the presence of Proposition 218 on the November 1997 ballot. As they tried to estimate the effects of this initiative—which severely constrained assessments by local governments for specific purposes—on local government finance, policy analysts from all agencies and groups discovered that there was very little information available. 25 for permission to undertake some activity. Regulatory fees and charges include two types of revenues: (1) revenues generated to fund specific regulatory activities, such as the Public Utilities Commission and various licensing boards, and (2) revenues generated as the result of permits issued as part of a specific regulatory process, such as construction permits and fish and game licenses. Planning fees and animal licenses also fall in this category. In some ways these could be looked upon as taxes. A construction permit, for example, could be viewed as a tax on the activity of construction much as a business license is a tax on the activity of being in business. They are much less generic, however.5 Reporting these regulatory fees and revenues in a separate category allows us to consider revenues that are generated by the regulatory power of government. It also has the advantage of allowing us to either make the distinction or not. By reporting these revenues separately, if one is not comfortable with the distinction used here, he can simply combine these revenues with taxes and ignore the differentiation. Fines, Penalties, and Forfeitures These revenues represent payments to government by individuals who have violated provisions of the state and local codes. Included in this category are fines for traffic violations, penalties on late property taxes, and parking fines. These are effectively taxes on socially ____________ 5For example, owning a business is much more generic than building specific structures. We do recognize, however, that this category and the tax category represent a continuum and that some fees could be classified as either a tax or a regulatory fee under these categories. Because of general practice within the policy community and the specificity criteria we raised above, however, business licenses and franchise taxes are not included here, but are included rather as taxes. 26 unacceptable activity and serve a broader social purpose of providing deterrents to these activities. Intergovernmental Revenues These revenues represent transfers from other levels of government, either restricted or unrestricted. For example, intergovernmental revenues for school districts include general state appropriations and funding under the Proposition 98 guarantee6 as well as revenues earmarked for the construction of schools. It also includes federal revenue sharing, state support for a range of local programs including children’s centers and mental health, and county support for city programs. Since the goal of this report is to determine the total revenues that the public sector obtains from private individuals and industry, intergovernmental revenues should also include current service charge revenues that are billed to public clients—such as electrical revenues received by public utilities from other governments. In most cases, this is not easy or even possible to do. There is one major exception, however—self-insurance districts. Over the past 20 years, numerous governments and groups of governments have set up independent special districts to provide self-insurance programs or to pool resources to provide insurance. The revenues reported by these independent districts are reported as intergovernmental revenues in this analysis, since they are generated exclusively from other public entities. This distinction is ____________ 6Proposition 98, passed in the June 1988 election, sets minimum funding levels for the support of K–14 education in California. The “guarantee” is the amount required under the provisions of the proposition. 27 important because these revenues are significant, exceeding $1 billion per year in recent years. Interest In the case of “revenues from the use of money and property category,” we chose to separate out interest revenues and to report the other two common revenues in this category—rents and royalties—as general services. This distinction goes directly to the issue of the reason why governments generate these resources and whether there is private sector provision of the resources. In general, interest revenues arise from the holding of monies that are usually held for other governmental purposes. State and local governments do not generally pursue the lending of money and the earning of interest as a primary business activity, as a bank would.7 The revenues generated from interest, therefore, can be considered as a byproduct of other activities, and private sector competition probably does not have too much of an effect on whether those revenues are earned or not. Rents and royalties, however, almost certainly have strong private sector competition and, in some cases, government activity may actually be “crowding out” private sector activity. This revenue stream, therefore, seems to represent a general service to the community and, except where the government has a monopoly on the type of resource provided—such as an airport—should be categorized as a general service revenue. Where the government has an actual or de facto monopoly, this revenue should ____________ 7There have been some notable exceptions to this. See Baldassare (1998) for a detailed description and discussion of one such case. 28 be classified as a service revenue associated with the government’s exclusive provider status. Enterprise Revenues Enterprise revenues are those generated by such services as sewer, water, electric, gas, and transportation. They arise from local publicly owned monopolies and are typically provided by either an independent special district or a quasi-independent department within the government. This enterprise revenue distinction is important from a policy perspective when considering the fact that not all of the provision of goods and services in these categories comes from publicly owned monopolies. In the case of electric power, for example, the City of Los Angeles has a publicly owned enterprise—the Department of Water and Power—whereas much of Northern and Central California are served by a privately owned company—Pacific Gas and Electric. Especially with deregulation in the offing, it is important to distinguish these revenues for policy purposes. In general, the revenues from these entities are largely committed to the costs and production of the specific goods and services. Service Revenues—Exclusive Provider This category of revenues represents activities for which a government receives revenues and for which it is the sole provider of that service and for which the service revenue is not the result of an enterprise activity. It category includes state lottery revenues, charges for the holding of elections, tax collection fees, and the costs of specialized police and fire services. 29 Service Revenues—General These revenues are generated by activities that are also commonly provided by nongovernmental entities. The largest revenues in this category include rents, ambulance services, golf course fees, university fees, and hospital revenues. Other Revenue This category includes all other revenues that do not fall into the above-defined definitions or for which the detail to classify the revenues was unavailable. It also includes donations from private sources and revenues from discontinued special districts. What Is Not Included It is important to note that this analysis does not include revenues from bond proceeds. The issuance of debt is important to public policy in California, but it does not conceptually fit into our main research question. The funds generated by bonds do not reflect a true revenue to the local government—just as an individual is not taxed by the government for funds borrowed to purchase a house.8 What are included, however, are the revenues generated to pay off the debt. In many cases, such as special bonds for schools, new revenues are generated to pay for these debts, usually in the form of property tax revenues and special assessments. As noted above under taxes and assessments, these revenues are captured in our analysis. ____________ 8Note that if we were concerned with the expenditure side of government, then these sources of funds would be quite important because they would be used to fund the provision of some assets or services. For revenue purposes, however, the debt assumption and issuance is a nonevent. 30 Applying This Taxonomy to California Let us then apply this taxonomy to state revenues overall and ask the question, What precisely is the nature of state and local government revenues? Table 2.2 provides us with an understanding of the composition and character of the overall state and local revenue burden. This table shows us that the largest revenue source for state and local government is generally taxes, but that their importance declined overall immediately after Proposition 13 and has remained relatively flat since. The share of intergovernmental revenues, which peaked in the early 1980s, is slightly higher than pre-Proposition 13 levels, whereas service revenues—including enterprise and services that are both generally and exclusively provided—have risen to fill the void left by the decline in tax revenues, rising from about 11 percent of revenues in 1978 to about 16 percent today. Assessments, although they account for a relatively small share of overall revenues, have risen the most dramatically, increasing some 1,832 percent since 1978. Taxes and intergovernmental revenues are the two slowest growing categories, growing at 186 percent and 276 percent, respectively. Because they account for 77 percent of overall revenues, the overall growth of all revenues is 252 percent. Because of the uneven intervals between the years in our study, it is difficult to visualize the trends represented in Table 2.2. For this reason, it is helpful to look at these trends in graphic form, as in Figure 2.1. This figure gives a better sense of the overall trends in the sources of revenues over time—showing the long period of relatively little change in the shares of revenue categories. Even with the categorical distinctions, however, this figure does not tell the full story of overall state and local 31 Table 2.2 Public Revenues in California for All Levels of Government, and Their Percentage Share, by Revenue Type 32 Revenue Type Taxes Assessments Regulatory fees and charges Fines and forfeitures Interest Intergovernmental Enterprise revenues Exclusive provider General services Other Total 1978 26,409,666,294 45.5 32,850,839 0.1 1,390,237,527 2.4 252,254,534 0.4 856,891,141 1.5 21,694,128,405 37.3 4,476,246,475 7.7 201,040,942 0.3 1,683,879,855 2.9 1,133,295,632 1.9 58,130,491,644 100.0 1981 28,392,188,943 36.6 162,711,104 0.2 1,338,110,598 1.7 362,918,091 0.5 1,838,620,532 2.4 32,456,331,967 41.9 7,733,753,254 10.0 287,272,242 0.4 3,133,030,720 4.0 1,769,296,927 2.3 77,474,234,378 100.0 1988 52,504,319,843 39.5 1992 72,070,935,103 38.4 422,836,209 0.3 560,079,307 0.3 3,228,228,698 2.4 4,101,351,362 2.2 953,171,700 0.7 988,966,277 0.5 3,366,566,608 2.5 3,871,848,963 2.1 49,489,137,169 73,373,444,285 37.1 39.1 13,714,929,929 18,857,594,195 10.3 10.0 1,095,613,493 0.8 1,182,836,525 0.6 5,567,580,382 4.2 7,179,561,943 3.8 2,906,771,649 2.2 5,661,550,731 3.0 133,249,155,680 187,848,168,691 100.0 100.0 1995 75,540,377,001 37.0 634,712,424 0.3 5,497,714,659 2.7 1,080,762,038 0.5 3,751,894,129 1.8 81,478,340,742 39.8 22,061,142,572 10.8 1,468,815,789 0.7 8,152,364,018 4.0 4,918,470,192 2.4 204,584,593,564 100.0 Percentage 50 45 40 35 30 Taxes Intergovernmental 25 Enterprise All others 20 Regulatory Exclusive 15 10 5 0 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 Figure 2.1—Overall Share of State and Local Revenues, by Revenue Type revenues. It is important to recall that there has been tremendous growth in overall state and local revenues over this period. Before discussing how to measure this growth, we must deal with an important bookkeeping issue: Should all of these revenues be considered part of the public revenue burden or are there legitimate reasons for excluding some of them from our estimations? In the following section we reexamine our taxonomy of revenue types and consider why we may wish to exclude some of the revenue types from our estimate of the public revenue burden. The Major Choices in Calculating the Public Revenue Burden When estimating the size of state and local government, it is important to consider which revenue categories should be included. 33 Clearly, a case must be made for including or not including any specific revenue type when estimating the overall public revenue burden in California. We believe that an easy argument can be made for including nearly all reported categories of revenues, except three: intergovernmental revenues, enterprise revenues, and general service charges. Should We Include Intergovernmental Revenues? Although intergovernmental revenues are important in determining how much money a particular type of government has to spend, it confuses the issue if one is trying to determine how much money the public sector is raising from its citizens. In fact, as we try to sum total revenues, counting these revenues would effectively double-count them. Take the case of an intergovernmental transfer from the state to a city. The state recognizes the revenue when it receives it in the form of a sales tax, income tax, or some other revenue. It then chooses to give some amount to a city government, say for law enforcement purposes. The city then reports this money as intergovernmental revenue from the state. In reality, however, no money has changed hands from the private citizens to the city. Only an allocative, administrative event has occurred between the state and the city. It is important, therefore, to exclude such funds. We have excluded all revenues that were reported as transfers from other levels of state and local government. The above argument holds true for state and local intergovernmental transfers, but there is the further issue of intergovernmental transfers from the federal government. Clearly these revenues come from outside the state and local revenue stream and hence the argument against their inclusion due to double-counting is invalid. We must then turn back to 34 the purpose of our analysis for guidance. We wish to estimate the public revenue burden at the state and local levels in California. Would including these federal transfers further this purpose? From one perspective, the answer is yes. These revenues do represent revenues to the local government that may have come, ultimately, from the local taxpaying population. To include them requires that we address the more global issue of whether Californians are in fact net contributors or benefactors in the federal tax and budget system. Although some literature indicates that California is a net contributor, the validation of this research is beyond the scope of this report. In light of these issues, we have chosen to exclude all identifiable federal intergovernmental revenues from our estimates of the public revenue burden in California. Our argument turns on the fact that the decisions about the amount of these revenues are made outside California and hence do not arise as part of the power of the people of California to raise revenues for their own purposes. Should We Include Enterprise Revenues? Enterprise revenues are generated by state and local governments to provide significant public services, such as water, electricity, natural gas, airports, and ports. We may wish to exclude these revenues in a public revenue burden estimate for two reasons: (1) They generally fund specific services and produce little surplus, and (2) they are not consistently reported as public activities. With regard to the first point, these revenues usually reflect the true cost of doing business in providing a desired service. Therefore, these revenues are not generally available for other purposes. Although there is 35 usually some cross-subsidization of other public activities by enterprise activities, they typically have separate accounting and funding systems. The cross-subsidies that do occur take two forms: (1) in-lieu franchise fees and (2) direct budgetary transfers. The first type of subsidy is reported as a tax revenue by the parent entities. Because of limitations in the state’s reporting system for enterprise revenues, the second type of subsidy remains largely unreported.9 For this reason, many advocate the inclusion of public enterprise revenues in the totals as a way to ensure their inclusion. However, these revenues are relatively small compared to the overall revenues generated by enterprise activities, and the inclusion of all enterprise revenues as part of the public revenue burden would dramatically overstate the size of this cross-subsidy. It is better to exclude these enterprise revenues and slightly underestimate public revenues than to include all enterprise revenues and significantly overestimate the revenue burden. This leads to the second issue—their universality. People in nearly all localities have electricity, water, and sewer services. However, not all of these services are provided by governments. In many localities, they are provided by private companies. Still other communities, such as Sacramento, are served by independent utility districts. As a result, some of these revenues are reported as public revenues and others are not. The lack of comparability associated with these revenues, coupled with their cost-based nature, leads to the conclusion that they, too, should be ____________ 9Note that this was not always the case. In the 1960s and early 1970s, these subsidies of local governments were reported separately by the California State Controller’s Office. 36 excluded from our estimate of the state and local revenue burden in California.10 Should We Include General Service Revenues? General service revenues are derived from government services and activities that are provided in direct competition with the private sector. The largest of these revenues are university fees, university research revenues, and public hospital revenues. These revenues also include a variety of rents as well as revenues from the sale of assets, such as real estate. Why would we want to consider excluding these? These revenues represent areas where the choice to purchase the good from the state or local government is entirely discretionary. One can attend a private university, obtain research services from a private firm, seek out a private hospital, or even rent space from a private landlord. Unlike many other revenues that arise as a consequence of the police or monopoly power of the government, these are revenues for which there are private sector substitutes and that arise from public sector competition with the private sector. For this reason, it can be argued that these revenues should also be excluded, although the arguments are somewhat weaker than those for excluding intergovernmental and enterprise revenues. For the purposes of our estimation, we choose to omit these revenues from the state and local revenue total.11 ____________ 10We recognize that some would disagree with the exclusion of these revenues from our estimate of the public revenue burden. Consequently, we have included tables summarizing our main findings, but including these revenues, in Appendix C. 11As before with enterprise revenues, some would disagree with the exclusion of these revenues from our estimate of the public revenue burden. We have therefore presented our main findings with these revenues included in Appendix C. 37 Should We Exclude Any Other Revenues? Having excluded these three types of revenues, we ask if there are any other candidates for exclusion among the remaining revenue types? We must again turn to our main research question, What has happened to the public revenue burden since Proposition 13? What other types of revenues are largely discretionary on the payee side or relate to specific business-like activities? Certainly taxes, assessments, regulatory fees, and fines and forfeitures do not fit the discretionary category. In fact, all of these categories are rooted directly in the police power of the state. Nonenterprise service revenues resulting from a service provided exclusively by the state or local government, such as election services and tax collection services, also do not fit this profile. The government exercises the right granted to it by its citizens to exercise control over these revenues. The only remaining major revenue stream for which we have not yet accounted is interest. We could have easily made a case for excluding it, but we have chosen to include it. Our reasoning is that the government is not generally in the business of raising monies just to receive interest revenues and that these revenues are usually a by-product of other activities that it undertakes, such as financing major construction activities. How Large Are the Excluded Revenue Categories? Table 2.3 shows the revenues we exclude from our estimate of the public revenue burden in California. These totals constitute nearly half of all state and local revenues.12 It should also be noted that approximately two-thirds of the revenues in the general service revenues ____________ 12The amounts are provided here for readers wishing to include them. 38 Table 2.3 Revenues Excluded from Our Estimate of Public Revenues in California for All Levels of Government, by Revenue Type Revenue Type Intergovernmental revenues Enterprise revenues General service revenues 1978 21,694,128,405 4,476,246,475 1,683,879,855 1981 32,456,331,967 7,733,753,254 3,133,030,720 1988 49,489,137,169 13,714,929,929 5,567,580,382 1992 73,373,444,285 18,857,594,195 7,179,561,943 1995 81,478,340,742 22,061,142,572 8,152,364,018 39 Total 27,854,254,735 43,323,115,941 68,771,647,480 99,410,600,423 111,691,847,332 Percentage of total revenues 47.9 55.9 51.6 52.9 54.6 category are from the state’s public postsecondary education institutions and reflect a combination of student fees and external research revenues, further strengthening our argument for excluding them from our calculation. After making these adjustments, we are left with our estimated totals of state and local revenues in California. These amounts are provided in Table 2.4.13 Perhaps the most dramatic feature of this table is the large proportion of the overall state and local revenue pie that comes from taxes—more than 80 percent—and its persistence over time, even in the face of Proposition 13. Furthermore, even though assessments account for only a small portion of overall state and local revenues (only 0.7 percent in 1995), their growth is striking—rising by more than 1,800 percent over the 17 years in our study. It is also interesting to note the increase in regulatory fees in recent years and the increase in service revenues associated with activities where the government is the exclusive provider. One crucial question remains to be answered: How accurate are the data underlying the findings in Table 2.4? Before undertaking the present study, we addressed this issue (Shires, 1997). We found that the overall revenue totals reported in the California State Controller’s series Annual Report on Financial Transactions were quite accurate. We did ____________ 13Note that we do not call these revenues “own-source revenues” for California state and local governments. Own-source revenues, a measure often used in the state and local revenue burden debate, are revenues that a government generates from its own activities. Such estimates typically exclude intergovernmental revenues and enterprise revenues but include what we call general service revenues. The own-source distinction would be most comparable to revenues presented in Table C.1 in Appendix C. We believe that including general service revenues in such a definition of the public revenue burden is problematic, however, as detailed earlier in this chapter. 40 Table 2.4 Remaining Public Revenues in California for All Levels of Government, and Their Percentage Share, by Revenue Type Revenue Type Taxes Assessments Regulatory fees and charges Fines and forfeitures Interest Exclusive provider Other 1978 26,409,666,294 87.3 32,850,839 0.1 1,390,237,527 4.6 252,254,534 0.8 856,891,141 2.8 201,040,942 0.7 1,133,295,632 3.7 1981 28,392,188,943 83.1 162,711,104 0.5 1,338,110,598 3.9 362,918,091 1.1 1,838,620,532 5.4 287,272,242 0.8 1,769,296,927 5.2 1988 52,504,319,843 81.4 422,836,209 0.7 3,228,228,698 5.0 953,171,700 1.5 3,366,566,608 5.2 1,095,613,493 1.7 2,906,771,649 4.5 1992 72,070,935,103 81.6 560,079,307 0.6 4,101,351,362 4.6 988,966,277 1.1 3,871,848,963 4.4 1,182,836,525 1.3 5,661,550,731 6.4 1995 75,540,377,001 81.3 634,712,424 0.7 5,497,714,659 5.9 1,080,762,038 1.2 3,751,894,129 4.0 1,468,815,789 1.6 4,918,470,192 5.3 Total 30,276,236,909 34,151,118,437 64,477,508,200 88,437,568,268 92,892,746,232 100.0 100.0 100.0 100.0 100.0 find some specific reporting variations within the specific revenue categories that could affect the findings reported here, however. Specifically, we found that city and county intergovernmental revenues appeared to be underreported in the State Controller’s report.14 Because the overall totals were consistently quite accurate, we surmised that these underreported amounts were simply miscategorized in the final reports. Such a miscategorization would bias our reported revenue burden upward—that is, it would overreport public revenues—because we did not deduct enough revenues when removing intergovernmental revenues. A similar finding was made in the opposite direction for current services—a category that includes public enterprise revenues and the general service revenues listed above.15 These underreportings and overreportings appear to roughly offset each other and, in any case, are minor relative to the overall magnitude of the revenues involved in this analysis. As a result, we believe that the data we use here are of high quality. To conclude, Table 2.4 presents what we believe should be included in public policy discussion about the size of the revenue burden imposed by state and local governments in California. Although the total size of government under this estimate is significantly smaller than we previously thought, it is important to remember that the $205 billion total presented in Table 2.2 includes many revenues that were counted twice, as well as significant revenues that arose from enterprise activities. ____________ 14The study, which used 1992 data, found that city intergovernmental revenues were underreported by about 15.2 percent and county intergovernmental revenues by about 0.1 percent. Much of this underreporting was linked with federal housing and transportation subsidies. 15These revenues were overreported, most likely due to classification errors, by 4 percent in counties and 3 percent in cities. 42 Nonetheless, the $93 billion generated by state and local governments represents a lot of revenue. The next chapter, which discusses how best to measure changes in the public revenue burden in California, will place these revenues in context. 43 3. Measuring Changes in the State and Local Revenue Burden in California As we discussed in Chapter 1, there are two parts to our research question. Chapter 2 answered the question, What constitutes the public revenue burden? Now we turn to the second part of the question, How do we measure the changes that have occurred? In this chapter, we will discuss the different ways that the revenue burden is measured and the advantages and shortcomings of each. To do this, we will generate estimates of the revenue burden over time using two dominant methodologies and discuss briefly the meaning and limitations of each— a discussion that will be enhanced and expanded in the final chapter of this study. Implicit in our main policy question is the further question, How has the revenue burden changed “relative to what?” The obvious answer is relative to what residents paid before, or relative to time. However, measuring the revenue burden over time introduces a significant level of 45 complexity to the problem because California has changed dramatically over the past 20 years. Most notably, the state’s population has grown 41 percent, by more than nine million people. Inflation has driven overall consumer prices up nearly 150 percent, and the overall income of residents—measured by annual personal income—has more than tripled, growing by more than one-half trillion dollars. If we simply ask the question, Are we paying more to the state and local government today than yesterday, the answer is an unequivocal Yes! As Table 2.4 showed, the public revenue burden rose from $30 billion in 1978 to $93 billion in 1995—more than tripling! But this finding does not take into consideration the profound changes the state has undergone over the past 20 years. We need to consider the public revenue burden relative to the many changes that have reshaped the state and its fiscal landscape, especially its economy and its population. Because of the importance of income and population to the state’s fiscal options, the two most common approaches to measuring the size of the public revenue burden focus on gauging it relative to these two elements. Measures that consider revenues relative to income ask the question, How much of our income are we paying?—much the same as asking, How much of my salary do I pay on rent? Measures that focus on population ask the question, How much do people pay on average?— similar to asking, How much was the average rent in my building this month compared to last month? We discuss each of these measures below. 46 Income-Based Measures of the Public Revenue Burden The rationale for using income-based measures is that they measure an individual’s or society’s ability to pay for services. For example, paying $1,000 a month for rent is a much more difficult proposition for someone whose annual income is $15,000 than for someone whose income is $100,000. Similarly, citizens with low incomes would have a much smaller ability to pay their state a given amount of taxes than citizens with higher incomes.1 And, just as an individual’s income—and hence his ability to pay a given level of rent—changes over time, so the income of a state’s citizens changes over time, which affects its population’s ability to pay a given level of revenues. California’s per capita personal income, for example, rose from $8,951 in 1978 to $23,279 in 1995.2 When we consider the statewide income of California, the cleanest measure available is personal income. This measure, which is produced by the Regional Economics Information System, Bureau of Economic Analysis, United States Department of Commerce, measures the overall income of individuals in the state. Other possible candidates for measuring the wealth of Californians include the Gross State Product (GSP), which corresponds to the most commonly used measure of national income—the Gross Domestic Product (GDP), and total ____________ 1Also implicit in this discussion is the question of who is paying the taxes. For example, a wealthy state’s average taxes might be lower than another state’s, but that burden might be disproportionately concentrated in either higher- or lower-income taxpayers. The issues of taxpayer equity that this concentration introduces, however, are beyond the scope of this report. 2These amounts are calculated from California Department of Finance reports of California personal income and population. 47 household income. Because GSP and state personal income are closely related, the GSP measure should produce results that closely parallel those we present below using personal income. Since the measurement of GSP, however, is not as precise as the measurement of state personal income,3 we will use personal income in our analysis. We do not use household income because it fails to include the business-sector portion of the state’s income capacity. To provide an income-based measure of the public revenue burden, we simply divide the public revenue burden by the measure of wealth— in this case, personal income. We have done this in Table 3.1. As this table shows, the public revenue burden has varied significantly over the 17 years in our study. In 1978, when the voters approved Proposition 13, the public paid 15 percent of its income to state and local governments. By 1981, state and local governments’ share of personal income fell dramatically to 11.4 percent. Although a significant portion of this decline is probably attributable to Proposition 13, a statewide recession and significant state-level tax cuts also directly contributed. During the 1980s and early 1990s, however, the public revenue burden grew, rising to more than 13 percent of total personal income in 1992 when, in response to the recession, state and local policymakers increased revenues through fees and tax increases.4 The public revenue burden has since declined to approximately the same level it was in 1988, dropping to 12.5 percent of personal income. ____________ 3Estimates of GSP are typically generated by roughly allocating the national GDP estimates across 63 industries to states. State personal income, on the other hand, is measured by compiling the relevant income information for each state. 4Because we have included only five years in our analysis, it is not possible to ascertain when the public revenue burden peaked as a share of personal income. 48 Table 3.1 Public Revenues in California as a Share of Personal Income for All Levels of State and Local Government 49 Revenue Type Public revenues California personal income Percentage of personal income 1978 30,284,118,492 202,241,530,500 15.0 1981 34,151,118,437 298,482,837,000 11.4 1988 64,477,508,200 527,589,456,000 12.2 1992 88,437,568,268 671,172,174,000 13.2 1995 92,892,746,232 743,218,326,000 12.5 If one believes that Proposition 13 was exclusively about tax reform, and that increases in other revenues such as interest, fines, and user fees are not critical, we could take a similar look at only tax revenues in California. In Table 3.2, we see how the tax burden has changed during this same period, looking at taxes as a share of state personal income. Since taxes are the main component of state and local revenues, we find that the tax trends correspond closely to those of the public revenue burden overall. There is a significant decrease in tax revenues as a share of personal income in the period immediately after Proposition 13, again attributable not only to Proposition 13 but also to statewide tax reforms and a statewide recession. This decrease is then followed by a gradual increase through the 1980s and early 1990s, with a final decrease back to approximately 1988 levels by 1995. This measure corresponds most closely to that used by Gold in his analysis, which we discussed in Chapter 1, and our results are consistent with his. One might, however, want to take a broader view of the public revenue burden in California. Some participants in the debate over the size of state and local government argue that governments have found creative ways around the constraints imposed by Proposition 13, primarily through increases in regulatory fees and assessments. To illuminate this issue, we present a different measure of the revenue burden in Table 3.3, which includes three revenue streams in question: taxes, assessments, and regulatory fees. This measure examines the whole family of revenues that are taxes by nature, if not in their specific 50 Table 3.2 Tax Revenues in California as a Share of Personal Income for All Levels of State and Local Government Revenue Type 1978 1981 1988 1992 1995 Tax revenues 26,409,666,294 28,392,188,943 52,504,319,843 71,586,395,964 75,540,377,001 Percentage of personal income 13.1 9.5 10.0 10.7 10.2 51 Table 3.3 TAR Revenues in California as a Share of Personal Income for All Levels of State and Local Government Revenue Type 1978 1981 1988 TAR revenues 27,832,754,660 29,893,010,645 56,155,384,750 Percentage of personal income 13.8 10.0 10.6 1992 76,732,365,772 11.4 1995 81,672,804,084 11.0 implementation.5 For our discussion, we will call this group of revenues TAR revenues (taxes, assessments, and regulatory fees). The findings are similar to those presented in Tables 3.1 and 3.2 because TAR revenues account for a significant portion of overall public revenues (82 percent of public revenues in 1995). One interesting difference, however, is that TAR revenues do appear to fall proportionately less than tax revenues between 1992 and 1995, indicating the persistence of assessments and regulatory fees as a mechanism for financing government in this period. Population-Based Measures of the Public Revenue Burden Population-based measures of changes in the public revenue burden tell us how the revenue burden has changed for the average citizen over time. It tells us how much the average person gives to state and local government, not as a share of income, but in strict dollar terms. Proponents of this measure would argue that it provides a real estimate of the “price of government.” They would also argue that demands for public services are driven more by the number of people using the services than by changes in the population’s income. Hence, populationbased measures would provide a fairer representation of the resources available to government. These population-based measures, also described as per capita measures, are customarily generated by dividing public revenues by the ____________ 5In Chapter 2, we made distinctions between general taxes (taxes on broad and general activities), assessments (taxes levied to fund specific activities and purposes and applied to a limited group of taxpayers), and regulatory fees and assessments (charges for participating in or undertaking specific voluntary activities). In this measure, we are generally including all monies levied by the government (the broadest definition of a tax). 52 state population in a given year.6 We have done this in Table 3.4 for the three revenue streams introduced in Tables 3.1, 3.2, and 3.3: public revenues, tax revenues, and TAR revenues. This table describes the total tax revenues per person in the state. Since this estimate also includes children, typical taxpayers actually see a multiple of this number, as they pay the taxes for their entire household. As we see here, the taxpayer sees an increasing amount each year going to state and local governments as all three revenue series increase over the 17 years in our study—even between 1978 and 1981 when the implementation of Proposition 13 coincided with state-level tax reforms and a statewide recession. This table explains, at least in part, the ongoing series of initiatives designed to “reel in” the expansion of government. Taxpayers and voters have a sense that government is evergrowing. Although this table may accurately characterize what the average Californian sees from year to year, it does not necessarily provide a good population-based measure of the revenue burden in California. This is Table 3.4 Per Capita Public Revenues, Tax Revenues, and TAR Revenues in California for All Levels of State and Local Government (Current Dollars) Revenue Type Per capita public revenues Per capita tax revenues Per capita TAR revenues 1978 1,340 1,169 1,232 1981 1,421 1,182 1,244 1988 2,298 1,871 2,002 1992 2,864 2,334 2,485 1995 2,910 2,366 2,558 ____________ 6For this analysis, we have generated an estimate of the population in a given fiscal year by averaging the California Department of Finance’s July 1 estimates of the state’s population for the beginning of the fiscal year and for the following year. Hence, our fiscal year 1978 population estimate is an average of the Department of Finance’s estimates for July 1, 1977, and July 1, 1978. 53 because it fails to take into account inflation, the year-to-year increase in prices that both individuals and governments encounter when purchasing the resources necessary to produce their services. Food, rents, gas, and other commodities increase in price over time. Governments and businesses must pay higher salaries over time to allow their employees to keep up with these rising prices. This gradual increase in costs means that a dollar today does not buy nearly as much as a dollar did 17 years ago. If we examine the public revenue burden over time without taking into account the reduced buying power of more current dollars, then we will overestimate the amount of services that governments can provide in a given year. The most common way to account for inflation is to use price indices, which are tracked by the federal government, to adjust prior-year dollars for changes in the cost of living. A price index allows one to convert the cost of something in a former year into what it would cost this year. For example, suppose a loaf of bread cost 40 cents in 1978 and $1.00 today. We could describe the price of a loaf of bread either in 1978 dollars (40 cents) or in today’s dollars ($1.00). If we assumed that the cost of other things grew at approximately the same rate, then a dollar’s worth of goods in 1978 would be worth approximately $2.50 today and, alternatively, $2.50 in today’s dollars would buy goods today that would have cost $1.00 in 1978. When one makes these corrections for the effects of inflation, the resulting revenues are called real revenues and the amounts are reported in constant 19xx dollars, where 19xx is the year into whose dollars the other year’s dollar amounts are converted. The use of these price indices allows us to make year-to-year comparisons of the value of money. One widely used index is the 54 Consumer Price Index (CPI).7 The federal government conducts annual surveys of prices and prepares estimates of the annual change in prices. Conveniently for us, the federal government even computes the changes in California prices separately. We use this version of the CPI to adjust the per capita burden amounts in Table 3.3 for inflation converting all the other years into equivalent 1995 dollars.8 The results are shown in Table 3.5.9 As we can see from this table, the earlier years’ revenues increased, reflecting the increased buying power of a dollar in the earlier years. The trends we see in this table are quite different from those we saw in Table 3.4 and actually correspond quite closely to those we saw in Tables 3.2 and 3.3 when we were measuring the public revenue burden as a share of personal income. The real per capita revenue burden declines significantly between 1978 and 1981 and then rises over the 1980s and early 1990s before falling back to 1988 levels in 1995. ____________ 7Other price indices are possible candidates for use here. One strong candidate would be the national CPI deflator for government goods and services. Using this deflator introduces a conceptual problem, however. Inasmuch as the availability of public funds defines the resources available to purchase services (more resources often lead to larger increases in expenditures for the same units), it becomes problematic to use the amounts actually paid for those services to discount their costs. Because of this concern, we have chosen to use the broader inflation measure, the CPI. 8This conversion was done by dividing each year’s index by the fiscal year 1995 index. Note that the resulting dollars are not precisely the same as 1995 dollars because a true recalibration of the index would require reweighting the bundle of goods included in the index by their 1995 proportions instead of those already in use in the index (1982–84 in this case). Because the federal government does not provide annual recalibrations of these amounts, this approach is a close approximation and it is commonly used in public policy analyses. 9It is important to note that the use of another index may produce different findings than we show here. However, it would take a 13.5 percent increase in the total price change over the period to reverse the direction of the change in real per capita public revenues from a decrease to an increase over this time. This level of error in the price index is possible, but unlikely. 55 Table 3.5 Real per Capita Public Revenues, Tax Revenues, and TAR Revenues in California for All Levels of State and Local Government (Constant 1995 Dollars) Revenue Type Real per capita public revenues Real per capita tax revenues Real per capita TAR revenues 1978 3,305 2,882 3,037 1981 2,498 2,077 2,187 1988 2,944 2,397 2,564 1992 3,057 2,492 2,653 1995 2,910 2,366 2,558 We have uncovered an interesting story in this chapter. Although the amounts that taxpayers pay to government appear to have risen dramatically over the 17 post-Proposition 13 years in this study, a different picture develops when income and inflation are factored into the equation. In fact, both common measures of the magnitude of the public revenue burden, both income- and population-based, tell a remarkably similar story. Revenues seem to decline dramatically after the implementation of Proposition 13 and then rise gradually through the 1980s. For the years we studied, they peak in 1992 and, by 1995, fall back to levels comparable to those in 1988. In our final chapter, we will discuss in more detail the implications of these patterns for public policy and the policy debate surrounding the size of the public revenue burden in California. 56 4. Our Findings The goal of this study was to present a comprehensive and informative overview of the public revenue burden in California and to provide a definitive answer on how it has changed over time. We believe that the definition of the public revenue burden we presented in Chapter 2 and the results we presented in Chapter 3 from the application of that definition have accomplished both of these goals. What We Have Found Figure 4.1 shows how the public revenue burden changed over time. It fell dramatically in the years after Proposition 13 (to 76 percent of preProposition 13 levels) and then rose over the 1980s. By 1992, the public revenue burden had returned to about 90 percent of pre-Proposition 13 levels, and it then declined to about 85 percent of pre-Proposition 13 levels in 1995. 57 Percentage 100 90 80 70 60 10 0 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 Figure 4.1—Composite Measure of Overall Public Revenues in California The trendlines are similar for both income-based and population- based measures. The composite measure used in Figure 4.1 is simply an average of the two series relative to their values in 1978. The actual changes in the two measures—revenue burden as a percentage of personal income and real per capita revenues—track quite closely, as can be seen in Figure 4.2. Although the real per capita revenue burden rises a bit faster than the revenue burden as a share of personal income in the 1980s, the changes relative to their levels in 1978 are almost identical. Both measures seem to tell the same story. Proposition 13 contributed significantly, in combination with other policy and economic factors, to a 24 percent decrease in the size of state and local 58 Percent of personal income Real per capita income (dollars) 16 3,500 14 Real per capita revenue burden 3,000 12 Percentage of 2,500 personal income 10 2,000 8 1,500 6 4 1,000 2 500 00 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 Figure 4.2—Two Measures of Public Revenues in California government1—to the delight of some and the chagrin of others. During the 1980s, however, the losses/gains were eroded as the size of government grew by 12 percent between 1981 and 1988. The growth in the size of government continued until, and was likely accelerated by, the recession of the early 1990s. Although we do not know for certain when the size of post-Proposition 13 government peaked,2 it reached its highest point in our series in 1992 at 90 percent of its pre-Proposition 13 level—an increase of 19 percent since 1981. By 1995, the size of state ____________ 1Note that not all of this decrease can be attributed to Proposition 13. The combination of state tax reductions and a statewide recession also contributed to the decrease in state and local revenues in 1981. 2Since we analyzed only five years in this study, we have no way of knowing whether 1992 was the peak. This was the first year of the recession and was a year marked by significant expansions of several types of state revenues to make up for the losses in revenues generated by the recession. Other scholars’ data indicate that the peak may have occurred earlier, in 1990. 59 and local government had returned to approximately the same size it was in 1988. We can also look at the changing revenue burden in terms of the two measures described above. Measured as a share of personal income, the public revenue burden declined 17 percent from 1978 to 1995 but rose 9 percent from 1981 to 1995. Similarly, real per capita public revenues declined 12 percent from 1978 to 1995 but rose 16 percent from 1981 to 1995. The average annual change in the public revenue burden was –8.6 percent from 1978 to 1981 and then averaged +1.3 percent from 1981 to 1992 before declining 1.8 percent a year from 1992 to 1995. Some would argue that the changes we describe above are directly attributable to changes in the economic and business cycles in California—that is to say that they are simply the result of the changing fortunes in our economy—and as state personal income grew during the 1980s the state’s progressive tax system stepped in to produce the growth in the public revenue burden noted in our study.3 However, as we noted in Chapter 1, we chose only five years in our study to address just this issue. We refer the reader to Appendix B for the full discussion, but the general patterns that we have identified over the past two decades hold, even when we compare across comparable points in the business cycle. In November 1996, voters chose to significantly expand the constraints on their government, passing Proposition 218. If revenue burden levels had declined to prerecessionary levels, how then do our ____________ 3Note that the data underlying our analysis indicate that this is not the case. If we were to report only the change in the portion of the revenue burden raised by state government, we find that it remains flat at 7.1 percent of personal income between 1981 and 1988 and then rises to 7.7 percent of personal income in 1992 before returning to 7.2 percent in 1995. Most of the growth over this entire period occurs in cities, counties, and school districts. 60 numbers explain this occurrence? There are many likely explanations. However, a contributing factor was probably the 12 percent growth in state and local government revenues that we identified during the 1980s. As we discuss in Chapter 2, much of the growth in this period was in the form of assessments, increased local taxes, increased regulatory fees, and increased service charges. These high-profile revenues would be likely to create a direct response in the voting public. An additional issue that our study has brought to light is shown in Figure 4.3. This figure shows the per capita public revenue burden in both current and real terms. As we can see from this diagram, the average resident in the state sees an increasing number of his dollars going to state and local government. Dollars 3,500 3,000 2,500 Current 2,000 1,500 1,000 Real 500 0 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 Figure 4.3—Per Capita Public Revenues, in Real (1978) and Current Dollars 61 Even though real revenues4 declined slightly over this period, public revenues in noninflation adjusted dollars have risen continuously and dramatically, even in the period immediately following Proposition 13. As we discussed in Chapter 3, taxpayers and voters think in terms of current dollars. For example, we think about how our property tax bill went up 2 percent this year. We do not usually consider the fact that, although our property tax bill went up 2 percent, our wages went up 3 percent and the price of gasoline or bread may have risen by 3 percent as well. Thus, although real per capita public revenues are certainly a legitimate measure to use in the policy debate, their message is often lost on voters. The bottom line of our study is that, after removing the appropriate revenue types,5 both measures of the changes in the public revenue burden seem to tell the same story, and that story lends support to both of the major camps in the public policy debate. The fact is that Proposition 13 did immediately contribute to a significant reduction in the revenues available to state and local governments in California. It is also true that these public revenues have never returned to pre-Proposition 13 levels. It is also true, however, that state and local governments have been able to significantly mitigate the full effects of Proposition 13 and that ____________ 4To make it easier to follow, we have used 1978 dollars in this figure. 5Some will find it controversial to remove the three types of revenues that we have eliminated (intergovernmental, competitively provided service revenues, and public enterprise revenues), in large part because this is not the way it has been historically done. We believe, however, that the arguments for their removal are strong and robust and allow for a more accurate understanding of what has happened to the public revenue burden in California over the past two decades. 62 the public revenue burden has risen substantially and progressively since Proposition 13’s implementation in the late 1970s and early 1980s.6 What Our Findings Suggest to Current and Future Policy in California Our findings have several implications for public policy in California. First, the pervasiveness of the study’s findings across measures and time shows that how to measure the changing revenue burden should not be the focal point of the debate—our analysis showed similar results with both measures. Furthermore, even if alternative measures of the revenue burden are used, as in Appendix C, the findings are consistent. We hope that this finding will clarify the confusion introduced by several previous studies on the topic (see Table 1.3) and allow the debate to focus instead on the critical issue of the appropriate size of state and local governments. Beyond this direct contribution, we believe that this study speaks to the future prospects for the public sector in California. Perhaps in response to the growth in the size of the state and local sectors in the years (in absolute terms) since the implementation of Proposition 13 (as shown in Figure 4.3), the voters of California have placed significant additional constraints on the public sector through the passage of Proposition 218. The constraints imposed by this proposition will significantly increase the hurdles that local governments will have to clear to expand their revenues through several of the mechanisms that were the mainstays of revenue growth during the late 1980s and early 1990s. ____________ 6We should also note that, as other PPIC research has shown, state and local governments can transfer the costs of government to private citizens and taxpayers in other ways (see, for example, Dresch and Sheffrin, 1997). Unfortunately, because of data limitations, our study could not fully incorporate those costs into this analysis. 63 Interestingly enough, this initiative seems to be more of a response to “general government growth” than to specific local needs, because voters seem to be passing many of the measures that have been brought to the ballot as a result of Proposition 218. In light of the requirements imposed by Proposition 218 and their attendant logistical, political, and fiscal costs, it is likely that the growth of new local revenue streams will slow in the future. In light of the flat rate of growth in real revenues identified in this study, it follows that the resources to fund expansions in the level of services provided at the state and local levels will grow, at best, slowly. There does seem to be some hope for specific programs and initiatives, as voters have in recent years been much more receptive to local bonds and finance measures. Because a large share of state and local revenues is derived from taxes, it is unlikely that state and local governments will experience funding shortfalls during periods of high economic growth, when tax coffers swell. In recessionary periods, however, the ability to raise additional revenues through increased license fees, service charges, and user fees— state and local governments’ response during the last recession—will be constrained by both Proposition 218 and their extensive use during the last recession. This combination could leave California’s state and local budgets sensitive to economic shocks and could result in reductions in public support for discretionary programs, such as higher education, with the onset of a future recession. This study has shown that Proposition 13 has contributed to a significant rollback of the public revenue burden. The effects of this rollback continue today, but its full effect has been partially offset by growth in the public revenue burden in the intervening years. It is 64 possible and even probable, however, that Proposition 218 will limit state and local governments’ ability to continue this growth and that the future size of the public revenue burden will remain at its current levels into the future. 65 Appendix A Detailed Comparisons with Other Studies In the introduction to this report, we briefly discuss four studies that examined the magnitude of the state and local revenue burden in California. Before beginning our study, we felt we needed to understand precisely what each of these scholars and analysts did in theirs and how their assumptions and choices affected their findings. We examined each study in detail and sought to replicate their findings with our data. Only in this way could we be certain that any findings of ours that differed from theirs were truly the result of differences in our assumptions and not data-driven. We were able to replicate each study’s revenue burden estimates fairly closely. The most variation occurred between our numbers and Gold’s 1977–78 estimates. Our estimate was about 5 percent lower than his. We discuss below our replication of each study’s estimates. 67 Steven Gold’s Estimates Steven Gold’s revenue burden estimates were the lowest among the studies, because he counted only “tax revenue.” He gave no source for his revenue numbers, so we used State Controller data to recreate his figures. For 1990–91, we limited the categories we included in the revenue burden to the following “tax” categories from the State Controller data: taxes, special benefit assessments, and licenses and permits for cities and counties; property tax revenue for schools; taxes and assessments for nonenterprise special districts and redevelopment agencies; and local transportation funds and other locally funded sales tax for transportation planning agencies. For enterprise special districts, the categories were assessments, county allocation, special district augmentation fund, property tax, sales tax, and debt service taxes. State revenues for our estimates included major taxes and licenses, regulatory taxes and licenses, revenue from local agencies, and services to the public. We tried to capture the same sources of revenue in the 1977–78 revenue burden estimate as we did in the 1990–91 figure. Although there were fewer categories delineated in the State Controller’s books in 1977–78, it is reasonable to assume that revenue sources from most of the newer categories were included previously in one of the other broader categories. In 1977–78, enterprise special districts had only two relevant categories—taxes, and taxes and assessments—instead of six; city and county classifications were the same as in 1990–91 with the exception of the lack of the special benefit assessment category. The categories for nonenterprise special districts and schools were unchanged. There were no transportation planning agency revenues recorded in 1977–78. In addition to major taxes and licenses, state revenues for 1977–78 included those from the “other revenues” category, with the exception of interest 68 on investments, oil and gas revenues, penalties and interest on unemployment contributions, proceeds under unclaimed property tax, and sales of state property. As shown in Table A.1, both our estimate of the 1990–91 state revenue burden and Gold’s are the same. Local estimates for that year differ by less than 2 percent, or approximately $300 million ($25.2 billion compared to $24.9 billion). In 1977–78, our estimate is below Gold’s by about 5 percent, or $1.4 billion ($27.8 billion compared to $29.2 billion). Table A.1 Comparison of Estimated Revenues Using Steven Gold’s Methodology, as a Percentage of Personal Income Study Gold Our study Difference 1977–78 Total 14.62 13.92 0.70 1990–91 Local 3.97 4.02 –0.05 1990–91 State 7.28 7.28 0.00 California Taxpayers’ Association’s Estimates The California Taxpayers’ Association (Cal-Tax) cites the Bureau of the Census as the source for its revenue figures. Thus, to replicate its numbers, it was necessary to choose a year in which the Census published government financial transaction data. Since detailed Census data were not published in both 1977–78 and 1990–91,1 the most recent year that ____________ 1The Bureau of the Census publishes estimates of state and local revenues, by type of government for various revenue categories each year, but the actual detailed information necessary for us to reconcile our estimates to those of Cal-Tax is available only in the specific full Census years that occur in years ending in 7 and 2, hence 1986–87 and 1991–92. Note that the 1991–92 data were not available in time for us to include that year in our reconciliation. 69 Cal-Tax published the revenue burden, we used the most current Census year in which the Cal-Tax estimate coincided with the detailed Census publications—1986–87—to validate our methodology. For our calculations, own-source revenues were derived by subtracting intergovernmental revenue from total general revenue. The resulting estimates exclude utility revenues (water supply, electric power, gas supply, and transit) and employee retirement revenue. Cal-Tax’s own-source revenue estimates for cities, counties, and special districts in 1986–87 equaled our figures from the Census data. Differences between the sets of numbers within the two remaining classifications were minimal. Our own-source figure for the state was $43.25 billion whereas Cal-Tax’s was $43.19 billion—a $60 million difference. The school revenue calculations differed by $40 million, with our total as $5.22 billion and Cal-Tax’s as $5.26 billion. Since the discrepancies are tiny, as shown in Table A.2, the overall revenue burden estimate is the same for us and Cal-Tax at 15.71 percent. As discussed in Chapter 1, there are some significant methodological issues associated with using the Bureau of the Census’ reported revenue amounts. We will present a detailed discussion of these differences in a future study. Table A.2 Comparison of Estimated Revenues Using the Cal-Tax Methodology, 1986–87 (Public Revenues in Billions of Dollars) Study California Taxpayers’ Association Our study Local Revenues 31.45 31.49 State Revenues 43.25 43.19 Total Revenues 74.70 74.68 Difference –0.04 0.06 0.02 70 Legislative Analyst’s Office’s Estimates The Legislative Analyst’s Office document containing the revenue burden estimate indicates that theirs was obtained directly from the U.S. Department of Commerce. The replication approach was quite similar to the one we pursued with the California Taxpayers’ Association estimate above. Using 1991–92 Census numbers for own-source local and state revenues and personal income, we calculated the revenue burden for 1991–92 to be 16.66 percent. The variation between the two revenue estimates is negligible. It is possible that the difference is due to different personal income numbers in the denominator, because such numbers may have been revised since these original estimates were made. John Kirlin et al. Estimates John Kirlin et al. provide the most detailed breakdown of the revenues that constitute their revenue burden estimates. Since the State Controller data are listed as the source for their revenue burden estimates, we were able to evaluate their numbers in each category. A description of our evaluation for each category is given in Table A.3. As shown in the table, we were able to determine how Kirlin and his colleagues calculated most of their revenue numbers. The primary differences between their estimates and ours center on transit revenues that were reported by the State Board of Equalization but were difficult to trace in the State Controller’s reports. We were unable to ascertain with certainty that these revenues were actually unreported and not miscategorized in the State Controller’s reports and we therefore did not add them to our estimates of the public revenue burden in California. Otherwise the two estimates correspond quite well. 71 Table A.3 Comparison of Estimated Revenues Using Kirlin et al.’s Methodology, 1986–87 (Public Revenues in Millions of Dollars) Study 1991–92: Kirlin et al. Our study Special School and Cities and Districts Transit Counties and RDAsa Districts State 54,172 16,134 54,164 16,133 30,736 30,358 61,887 61,986 Total 162,929 162,644 Difference 8 1 378 –99 288 1977–78: Kirlin et al. Our study 16,800 5,955 16,787 5,955 8,446 8,187 20,223 20,201 51,424 51,130 Difference 13 0 259 22 294 NOTES: Our replication of the 1991–92 revenue estimate for counties does not include other financing sources for San Francisco ($155 million) because we know that Kirlin et al. exclude these revenues. We also do not include in the replication revenues from current service charges for self-insurance districts ($973 million) in our 1991–92 nonenterprise special district total, because Kirlin et al. exclude them. The sum of the 1977–78 revenues for each of Kirlin et al.’s entities is $51,424, but they list it as $51,167, in part because they omitted transit sales revenues in their total sales tax figure. aRedevelopment agencies. 72 Appendix B The Years Included in This Report Because of the complications associated with the way special district revenues are reported in California, it was necessary to keypunch a considerable amount of data from the State Controller’s Annual Report on Financial Transactions Concerning Special Districts for each of the years we wished to study. As a result of this significant cost, we selected only five years for our analysis: 1977–78, 1980–81, 1987–88, 1991–92, and 1994–95. The first fiscal year was chosen because it was the year that Proposition 13 passed. 1991–92 was chosen because several other studies reported revenue burdens for this fiscal year. 1994–95 was chosen because it was the most recent fiscal year for which the data were available. The two remaining years, 1980–81 and 1987–88, were selected because they represented points in the business cycle that corresponded to 1991–92 and 1977–78, respectively, as shown in Figure B.1. 73 Percentage 6 5 4 3 2• 1 0 • • –1 •–2 –3 –4 1975 1977 1979 1981 1983 1985 1987 1989 • 1991 1993 1995 Figure B.1—Annual Growth in California Real per Capita Personal Income As the figure shows, the state was in the depths of recession in both 1981 and 1992, and 1978 and 1988 were comparable points in periods of economic growth. This aspect of comparison may be particularly important when using the income-oriented measures of the public revenue burden—such as percentage of personal income—which are closely associated with changes in the business cycle. 1981 has the additional benefit of coming soon after Proposition 13 was fully implemented1 and before a series of accounting changes were instituted by state and local governments.2 ____________ 1The three-year delay after 1978, when Proposition 13 was passed by the voters, is actually ideal timing. Proposition 13 was initially implemented in fiscal year 1979, but significant revisions were subsequently instituted in fiscal year 1980. As a result, 1981 actually represents the first year after the full transition to the post-Proposition 13 world. 2Fiscal year 1981 falls before a series of accounting changes that were implemented in 1982 and 1983. Although these changes will not affect the longer term because the year-to-year variation largely nets out, there were noticeable effects on reported revenues during the transition years. 74 The most recent year for which data were available was 1995, but the changes in the public revenue burden from 1992 to 1995 may provide some insights into the state of the world when Proposition 218 was passed by the voters in November 1996. To compare the findings across the business cycle, Figure B.1 plots the progression of the California business cycle over time by presenting the annual growth in real per capita personal income in California. To help us understand the changes over time, it is useful to compare our composite indices for two pairs of years, the first pair corresponding to plateaus after economic peaks (1978 and 1988), and the second pair to a bottoming out of recessions (1981 and 1992). In the first case—the two economic midpoint years 1978 and 1988—the findings are consistent with our overall conclusions. The overall public revenue burden, as measured by our composite index, declines from 100 to 85 percent of 1978 levels. Although we cannot reach any conclusions about changes over the intervening years by using these two points, the findings regarding the two recession years add weight to our overall conclusions. The findings in the second comparison are more compelling. In the comparison of the two recession years, we see an increase from 76 percent of the 1978 level to 90 percent—again pointing to our finding that the revenue burden increased over this time, but to a level less than where state and local government revenues were before Proposition 13. Perhaps the most interesting aspect of this finding is the seeming inconsistency of outcomes. If we assume that a recession has the same repressive effect on state tax revenues, then the increases in the 1992 recession are so much the more remarkable. The state’s fiscal status was quite different during the two recessions, however. During the recession 75 in 1981, not only was Proposition 13 implemented—creating a significant demand for state resources among local governments—but the state instituted several tax cuts. During the 1992 recession, the state had to institute significant tax increases to keep the state solvent. This difference in state action goes a long way toward explaining the seeming inconsistency of the two findings, although the broader story of a reduced, but growing public revenue burden remains intact. 76 Appendix C Revenue Burden Measures Under Alternative Definitions In this appendix, we will reproduce the findings presented in Chapter 3 for the public revenue burden under three alternative scenarios: (1) restoring general service revenues to our definition of public revenues, (2) restoring enterprise revenues, and (3) restoring both. We will present summary statistics for each of three measures for each scenario: (1) the public revenue burden as a share of personal income, (2) the revenue burden adjusted for changes in the population, measured by per capita revenues, and (3) the revenue burden adjusted changes in both the population and inflation, measured by real per capita revenues. Restoring General Services Restoring those public revenues for which the public sector has direct private sector competition produces the results shown in Table C.1. 77 Table C.1 Measures of California’s Public Revenues if General Services Are Included Measure Public revenues ($ billions) Percentage share of personal income Per capita public revenues ($) Real per capita public revenues (1995 $) 1978 31.97 15.8 1,415 1981 37.28 12.5 1,552 3,489 2,727 1988 70.05 13.3 2,497 3,198 1992 1995 95.62 101.04 14.2 13.6 3,097 3,165 3,306 3,165 This formulation most closely corresponds to the own-source definition often used for convenience in policy debates. Here, the story remains much the same as we found with our primary definition of the public revenue burden. Public revenues decline immediately after Proposition 13 and then regain much of the decline through the early 1990s before declining once again in the mid-1990s. In real per capita terms, the growth is slightly higher in this formulation than it is in our primary definition. Real per capita revenues rise to nearly 95 percent of pre-Proposition 13 levels by 1992 before dropping off to 90 percent in 1995. This is because state and local governments significantly increased their prices for products they produced in competition with the private sector during the recession.1 Restoring Enterprise Revenues Some would argue that, although we should not include general service revenues, we should include enterprise revenues that result from ____________ 1For example, student fees at the University of California rose dramatically from $2,486 per year in 1992 to $4,111 in 1995 (see CPEC, 1996, for further details). 78 policies and pricing strategies instituted by public organizations.2 Restoring public revenues that arise from enterprise activities produces the results shown in Table C.2. The general pattern of the story is still consistent with our main findings in this study, although the thresholds are much higher. Because of their dependence on service charges, enterprise revenues declined only to 83 percent of pre-Proposition 13 levels by 1981 and then rose back to 93 percent of those levels in 1992 before dropping back to 90 percent in 1990. This step-up in the levels of change over time tells a story about the increasing cost of services provided by these entities and further reinforces the validity of our decision to exclude them from our estimate of the public revenue burden in California. Table C.2 Measures of California’s Public Revenues if Enterprise Revenues Are Included Measure Public revenues (billions) Percentage share of personal income Per capita public revenues ($) Real per capita public revenues (1995 $) 1978 1981 1988 1992 1995 34.75 41.88 78.19 107.30 114.96 17.2 14.0 14.8 16.0 15.5 1,538 1,743 2,787 3,475 3,601 3,793 3,064 3,570 3,709 3,601 Restoring Both General Services and Enterprise Revenues Finally, Table C.3 presents the changes in public revenues if one were to include both those public revenues for which the public sector ____________ 2Note that under this definition, one may wish to include, at the very least, the higher education enterprise, which is included in the general services category. We do not do so here for the sake of consistency. 79 Table C.3 Measures of California’s Public Revenues if Both General Service Revenues and Enterprise Revenues Are Included Measure Public revenues ($ billions) Percentage share of personal income Per capita public revenues ($) Real per capita public revenues (1995 $) 1978 1981 1988 1992 36.44 45.02 83.76 114.47 18.0 15.1 15.9 17.1 1,613 1,873 2,986 3,708 1995 123.11 16.6 3,856 3,976 3,293 3,824 3,958 3,856 has direct private sector competition and those that arise from enterprise activities. As one would expect, this has the effect of significantly expanding the estimate of the revenue burden to the point that we are now 96 percent of the way back to pre-Proposition 13 levels, as measured by real per capita income. We believe, however, that this measure is too inclusive and disguises key distinctions in revenues that should not be included in estimates of the public revenue burden in California. 80 References Baldassare, Mark, When Government Fails: The Orange County Bankruptcy, Public Policy Institute of California, San Francisco, California, 1998. California Postsecondary Education Commission (CPEC), Fiscal Profiles 1996, Sacramento, California, 1996. California State Controller, Annual Report on Financial Transactions Concerning Cities, Sacramento, California, 1978, 1981, 1988, 1992, and 1995. California State Controller, Annual Report on Financial Transactions Concerning Counties, Sacramento, California, 1978, 1981, 1988, 1992, and 1995. California State Controller, Annual Report on Financial Transactions Concerning Special Districts, Sacramento, California, 1978, 1981, 1988, 1992, and 1995. California Taxpayers’ Association, “California State and Local Revenue Growth: Effects of Proposition 13 Are Almost Gone,” Cal-Tax News, April 1, 1994. Dresch, Marla, and Steven M. Sheffrin, Who Pays for Development Fees and Exactions? Public Policy Institute of California, San Francisco, California, 1997. 81 Dunstan, Roger, Local Government Revenue and Expenditures Since Proposition 13: A Historical Primer, California Research Bureau, report CRB-IS-93-006, Sacramento, California, August 1993. Gold, Steven D., “California’s Tax Burden from a National Perspective,” Symposium on State Taxation of Business Activities, University of California at Davis, March 26, 1993. Kirlin, John J., Jeffrey I. Chapman, Peter Asmus, and Roy Thompson, “Fiscal Reform in California,” in California Fiscal Reform: A Plan for Action, California Business—Higher Education Forum, Oakland, California, June 1994. Legislative Analyst’s Office, “State and Local Finance—Fiscal Overview,” Cal Guide: A Profile of State Programs and Finances, Sacramento, California, January 1995. Leigland, James, “The Census Bureau’s Role in Research on Special Districts: A Critique,” Western Political Quarterly, Vol. 43, No. 2, June 1990, pp. 367–380. Sheffrin, Steven, and Marla Dresch, Estimating the Tax Burden in California, Technical Assistance Program Report, California Policy Seminar, University of California at Berkeley, 1995. Shires, Michael A., Patterns in California Government Revenues Since Proposition 13, Public Policy Institute of California, San Francisco, California, forthcoming 1998. Shires, Michael A., and Melissa Glenn Haber, A Review of Local Government Revenue Data in California, Public Policy Institute of California, San Francisco, California, 1997. 82 About the Authors MICHAEL A. SHIRES Michael A. Shires is a research fellow at the Public Policy Institute of California. Before joining PPIC, he was a doctoral fellow at RAND's Graduate School of Policy Studies, concentrating on domestic education policy, California fiscal policy, and international trade policy. He is the author of several publications on U.S. trade relations, higher education, and California state finance and education. He has also been active as a consultant to private companies on strategic planning, marketing research, and tax planning and preparation. He holds a B.A. in economics from the University of California, Los Angeles; an M.B.A. from UCLA’s Anderson Graduate School of Management; and a Ph.D. in public policy analysis from the RAND Graduate School. JOHN ELLWOOD John Ellwood is an adjunct fellow at the Public Policy Institute of California and a professor in the Richard and Rhoda Goldman School of Public Policy at the University of California, Berkeley. He served as PPIC's research director from 1996–1997. He has held academic positions at the University of Minnesota, Dartmouth College, and Princeton University. He also has extensive experience in government, including service as the special assistant to the director of the Congressional Budget Office from 1976 to 1980. He received an A.B. in political science from the Franklin and Marshall College and an M.A. and Ph.D. in political science from the Johns Hopkins University. MARY SPRAGUE Mary Sprague is a Ph.D. student in political science at the University of California, Berkeley. During 1996–1997 she worked as a research assistant at the Public Policy Institute of California. Previously, she spent a year as a Javits Fellow on the U.S. Senate Finance Committee. She holds an A.B. in public policy from Stanford University and an M.P.P. from the University of California, Berkeley. 85 Other PPIC Publications Dardia, Michael (1998), Subsidizing Redevelopment in California. Dresch, Marla, and Steven M. Sheffrin (1997), Who Pays for Development Fees and Exactions? Johnson, Hans P. (1996), Undocumented Immigration to California: 1980–1993. Lewis, Paul G. (1998), Deep Roots: Local Government Structure in California. Lewis, Paul G., and Mary Sprague (1997), Federal Transportation Policy and the Role of Metropolitan Planning Organizations in California. MaCurdy, Thomas, and Margaret O’Brien-Strain (1997), Who Will Be Affected by Welfare Reform in California? MaCurdy, Thomas, and Margaret O’Brien-Strain (1998), Reform Reversed? The Restoration of Welfare Benefits to Immigrants in California. Reed, Deborah, Melissa Glenn Haber, and Laura Mameesh (1996), The Distribution of Income in California. Reyes, Belinda I. (1997), Dynamics of Immigration: Return Migration to Western Mexico. Shires, Michael A., and Melissa Glenn Haber (1997), A Review of Local Government Revenue Data in California. Smolensky, Eugene, Eirik Evenhouse, and Siobhán Reilly (1997), Welfare Reform: A Primer in 12 Questions. Spetz, Joanne (1996), Nursing Staff Trends in California Hospitals: 1977 through 1995. Copies of PPIC publications may be ordered by e-mail (order@ppic.org) or phone (415) 291-4400. Full text versions are also available at: www.ppic.org. 86" } ["___content":protected]=> string(102) "

R 998MSR

" ["_permalink":protected]=> string(109) "https://www.ppic.org/publication/has-proposition-13-delivered-the-changing-tax-burden-in-california/r_998msr/" ["_next":protected]=> array(0) { } ["_prev":protected]=> array(0) { } ["_css_class":protected]=> NULL ["id"]=> int(8135) ["ID"]=> int(8135) ["post_author"]=> string(1) "1" ["post_content"]=> string(0) "" ["post_date"]=> string(19) "2017-05-20 02:35:09" ["post_excerpt"]=> string(0) "" ["post_parent"]=> int(3251) ["post_status"]=> string(7) "inherit" ["post_title"]=> string(8) "R 998MSR" ["post_type"]=> string(10) "attachment" ["slug"]=> string(8) "r_998msr" ["__type":protected]=> NULL ["_wp_attached_file"]=> string(12) "R_998MSR.pdf" ["wpmf_size"]=> string(6) "207550" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(143008) "Has Proposition 13 Delivered? The Changing Tax Burden in California Michael A. Shires John Ellwood Mary Sprague 1998 Copyright © 1998 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 The passage of Proposition 13 on June 6, 1978, signaled a dramatic change in California politics as voters placed a cap on the amount of taxes they were willing to pay. Perhaps no other initiative in the past 20 years has been subject to greater scrutiny or provoked more controversy. The policy issues involved, including the question of equity, the fiscal relationship between state and local governments, and the level and quality of government services, are large and important concerns that need to be analyzed with an objective and independent eye. They are exactly the kinds of issues that the Public Policy Institute of California was founded to study. In this report, Michael Shires, John Ellwood, and Mary Sprague answer one of the most prominent questions in the policy debate: How much do we pay to our state and local governments in California and how has that amount changed since Proposition 13? The authors conclude that the answer depends on whether you account for inflation. In nominal dollars per capita, public revenues in California have more iii than doubled in the 20 years since the passage of Proposition 13. But, controlling for inflation, per capita revenues have declined by 16 percent. The authors point out, however, that although the revenue burden is lower today than it was in 1978, it has risen significantly since 1981. This volume is one of a series that PPIC is publishing on the status of public finance in California as part of an overall program of research in governance and public finance. Related reports will soon be available, addressing such issues as the fragmentation of local government, longterm trends in county finances, patterns in state and local government revenues, and the consequences of the 1990’s recession on property valuations under Proposition 13. In addition, a recent PPIC book by Mark Baldassare examines in great detail the bankruptcy of Orange County in 1994. His analysis traces the bankruptcy directly to the exigencies of Proposition 13 and the unending pressure to raise revenues from local sources. The book, When Government Fails: The Orange County Bankruptcy, is available through the University of California Press. We trust that this growing body of research and findings will reduce the level of disagreement about Proposition 13 and its aftermath and set the stage for a more informed public dialogue. David W. Lyon President and CEO Public Policy Institute of California iv Summary Over the past two decades, Californians have engaged in an almost continuous debate over the proper size of their governments. Beginning in 1978 with the adoption of Proposition 13, they have used the direct legislative procedures of the initiative process to limit the growth of their governments’ taxing and spending powers and to mandate that significant portions of their governments’ revenues be devoted to specific purposes. The annual budget processes of most of the state’s governments have also focused on these issues. This report explores the recent history of this debate by analyzing California’s revenue burden since 1978. In it, we will answer the question, How much do we pay to our state and local governments in California and how have those amounts changed since Proposition 13? Unfortunately, there is a lot of debate over how to answer this question. For one thing, citizens do not agree on what should be counted as tax or nontax revenue or on how to measure the burden of taxes and other government revenues. Some revenues are clearly taxes— v the income tax, the property tax, and the sales tax, for example. But should all fees charged by governments be included in the revenue burden? For example, if a student pays fees to a state college, should those fees be included in the state revenue burden even if that student could have attended a private college or gone to an out-of-state public university? How should the income of publicly owned hospitals be viewed, or fees for other services where governments compete with private firms to provide a service? How should we think about fees for a service that the government provides through a monopoly or taxes that the government converts into user fees? Citizens also disagree about the correct way to measure the size of state and local government revenue burdens. For example, if the dollar amount of taxes doubles over a decade, has the revenue burden doubled? Some would say yes. Others would say that the effects of inflation should be taken into account when answering this question. Still others would point out that since average income has also doubled over this period, the actual “burden” has remained unchanged. The Public Revenue Burden in California Since Proposition 13 We find that the public revenue burden in California has declined since Proposition 13, but that it has been increasing since the early 1980s. Figure S.1 illustrates this primary finding of our research. As the figure shows, the public revenue burden in California fell to about 76 percent of its 1978 levels in 1981 and then rose to about 90 percent of its 1978 levels by 1992. Although the decline from 1978 to 1981 is not wholly attributable to Proposition 13—there was a state-level tax cut and vi Percentage 100 90 80 70 60 10 0 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 Figure S.1—Composite Measure of Overall Public Revenues in California a statewide recession between 1978 and 1981—it was certainly a major contributing factor. We also know that the public revenue burden has risen from the 76 percent level in 1981 to 85 percent in 1995. This growth has been largely due to increases in local taxes and charges. These findings are relatively robust, no matter what measure of the public revenue burden is used—whether the revenue burden is considered as a percentage of overall income or on a per-person, inflation-adjusted basis. At the same time, we note that there seems to be some concern on the part of the electorate about the growing public revenue burden, as demonstrated by the passage of Proposition 218 in November 1996. This initiative placed supermajority voting requirements on many local assessments and charges. Figure S.2 may shed some light on this issue. As shown in this figure, the average public revenues per person in the vii Dollars 3,500 3,000 2,500 Current 2,000 1,500 1,000 Real 500 0 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 Figure S.2—Per Capita Public Revenues, in Real (1978) and Current Dollars state have declined when adjusted for inflation. However, in absolute terms, the revenue burden has increased dramatically over the past two decades, rising from $1,500 per person in 1978 to nearly $3,000 in 1995. Local governments are also charging residents for services that were free in the past, such as access to local parks. The result is a public finance system that appears to cost more to sustain while simultaneously charging more for the services it does provide. Implications for California Public Policy These findings have several implications for public policy in California. First, the pervasiveness of the study’s findings across measures and time shows that how to measure the changing revenue burden should not be the focal point of the debate—our analysis showed similar results with both measures. Furthermore, even if alternative measures of the revenue burden are used, the results are consistent. We viii hope that this finding will clarify the confusion introduced by several previous studies on the topic and will allow the debate to focus instead on the critical issue of the appropriate size of state and local governments. Beyond this direct contribution, we believe that this study speaks to the future prospects for the public sector in California. Perhaps in response to the growth in the size of the state and local sectors since the implementation of Proposition 13, the voters of California have placed significant additional constraints on the public sector through the passage of Proposition 218, which significantly constrains the ability of local government to creatively expand its revenues through several of the mechanisms that were the mainstays of revenue growth during the late 1980s and 1990s. Interestingly enough, this initiative seems to be more of a response to “general government growth” than to specific local needs, because voters seem to be passing many of the measures that Proposition 218 has brought to the ballot. It is likely, however, that the requirements imposed by Proposition 218 and their attendant logistical, political, and fiscal costs will slow the growth of new local revenue streams in the future—especially in periods of recession. In light of the rates of growth in real revenues identified in this study, it follows then that the resources to fund expansions in the level of services provided at the state and local levels will grow, at best, slowly. There does seem to be some hope for specific programs and initiatives, as voters have in recent years been much more receptive to local bonds and finance measures. Because a large share of state and local revenues is derived from taxes, it is unlikely that state and local governments will experience funding shortfalls during periods of high economic growth, when tax coffers swell. In recessionary periods, however, the ability to raise additional ix revenues through increased license fees, service charges, and user fees— state and local governments’ response during the last recession—will be constrained by both Proposition 218 and the extensive use of such fees during the last recession. This combination could leave California’s state and local budgets sensitive to economic shocks and could result in reductions in public support for discretionary programs, such as higher education, with the onset of a future recession. This study has shown that Proposition 13 has contributed to a significant rollback of the public revenue burden. Although the effects of this rollback continue today, its full effect has been partially offset by growth in the public revenue burden in the intervening years. It is possible and even probable, however, that Proposition 218 will limit state and local governments’ ability to continue this growth and that the future size of the public revenue burden will remain at its current levels into the future. x Contents Foreword ..................................... Summary..................................... Figures ...................................... Tables ....................................... Acknowledgments ............................... iii v xiii xv xvii 1. INTRODUCTION ........................... The Debate over State and Local Finance ............. An Anatomy of Public Finance in California ........... How Can Revenue Burdens Differ? ................. What Others Have Found ....................... Steven Gold ............................... California Taxpayers’ Association ................. Kirlin et al. ............................... California Legislative Analyst’s Office .............. Why We Need Another Study ................... Our Study ................................. Organization of This Report ...................... 2. DEFINING THE STATE AND LOCAL REVENUE BURDEN IN CALIFORNIA ..................... The Public Revenue Burden in California: A Series of Choices ................................ A Taxonomy of Revenue Types .................... 1 3 6 11 12 14 14 15 16 16 17 18 19 20 23 xi Figures S.1. Composite Measure of Overall Public Revenues in California ............................... vii S.2. Per Capita Public Revenues, in Real (1978) and Current Dollars................................. viii 1.1. Reported Public Revenues of California State and Local Governments, by Revenue Base, FY 1994–95 ........ 9 2.1. Overall Share of State and Local Revenues, by Revenue Type .................................. 33 4.1. Composite Measure of Overall Public Revenues in California ............................... 58 4.2. Two Measures of Public Revenues in California....... 59 4.3. Per Capita Public Revenues, in Real (1978) and Current Dollars................................. 61 B.1. Annual Growth in California Real per Capita Personal Income................................. 74 xiii Tables 1.1. FY 1994–95 Total Public Revenues in California, by Level of Government Receiving the Revenue ......... 1.2. Percentage Share of FY 1994–95 Revenues from Various Sources, by Level of Government Receiving the Revenue ................................ 1.3. Selected Findings of Prior Studies on Public Revenues in California, as a Percentage Share of Personal Income.... 2.1. Public Revenues in California for All Levels of Government, and Their Percentage Share, by Level Receiving the Revenue ....................... 2.2. Public Revenues in California for All Levels of Government, and Their Percentage Share, by Revenue Type .................................. 2.3. Revenues Excluded from Our Estimate of Public Revenues in California for All Levels of Government, by Revenue Type ............................ 2.4. Remaining Public Revenues in California for All Levels of Government, and Their Percentage Share, by Revenue Type .................................. 3.1. Public Revenues in California as a Share of Personal Income for All Levels of State and Local Government ... 8 10 13 21 32 39 41 49 xv 3.2. Tax Revenues in California as a Share of Personal Income for All Levels of State and Local Government ... 3.3. TAR Revenues in California as a Share of Personal Income for All Levels of State and Local Government ... 3.4. Per Capita Public Revenues, Tax Revenues, and TAR Revenues in California for All Levels of State and Local Government ............................. 3.5. Real per Capita Public Revenues, Tax Revenues, and TAR Revenues in California for All Levels of State and Local Government ......................... A.1. Comparison of Estimated Revenues, Using Steven Gold’s Methodology, as a Percentage of Personal Income................................. A.2. Comparison of Estimated Revenues Using the Cal-Tax Methodology, 1986–87 ...................... A.3. Comparison of Estimated Revenues Using Kirlin et al.’s Methodology, 1986–87 ...................... C.1. Measures of California’s Public Revenues if General Services Are Included ....................... C.2. Measures of California’s Public Revenues if Enterprise Revenues Are Included....................... C.3. Measures of California’s Public Revenues if Both General Service Revenues and Enterprise Revenues Are Included ................................ 51 51 53 56 69 70 72 78 79 80 xvi Acknowledgments As is the case with any significant research project, the authors of this study are indebted to many individuals and organizations for their contributions and assistance. First and most important, we are indebted to the 13 individuals who early in the project provided us with the benefit of their wisdom and experience: Michael Coleman; John Decker, Resource Subcommittee, Assembly Budget Committee; Joel Fox, Howard Jarvis Taxpayers’ Association; Lenny Goldberg, California Tax Reform Association; Steve Kamp, Assembly Revenue and Taxation Committee; Stephen Kroes, California Taxpayers’ Association; Bob Leland, City of Fairfield; Anne Maitland, Senate Revenue and Taxation Committee; April Manatt, Senate Local Government Committee; Marianne O’Malley, California Legislative Analyst’s Office; Jean Ross, the California Budget Project; and Connie Squires, Department of Finance. We are also grateful for the time and efforts of those who served as formal reviewers of this study, including Mark Baldassare, PPIC; Jeff xvii 1. Introduction Over the past two decades, Californians have engaged in an almost continuous debate over the proper size of their governments. Beginning in 1978 with the adoption of Proposition 13, they have used the direct legislative procedures of the initiative process to limit the growth of their governments’ taxing and spending powers as well as to mandate that significant portions of their governments’ revenues be devoted to specific purposes. The annual budget processes of most of the state’s governments have also focused on these issues. This report explores the recent history of this debate by analyzing California’s revenue burden since 1978. In it, we answer the question, How much do we pay to our state and local governments in California and how have those amounts changed since Proposition 13? Unfortunately, there is a lot of debate over how to answer this question. For one thing, citizens do not agree on what should be counted as tax or nontax revenue or on how to measure the burden of taxes and other revenues. Some revenues are clearly taxes—the income 1 tax, the property tax, and the sales tax, for example. But should all fees charged by governments be included in the revenue burden? For example, if a student pays fees to a state college, should those fees be included in the state revenue burden even if that student could have attended a private or out-of-state college? How should the income of publicly owned hospitals be viewed, or fees for other services where governments compete with private firms to provide a service? How should we think about fees for a service that the government provides through a monopoly or taxes that the government converts into user fees? Citizens also disagree about the correct way to measure the size of state and local government revenue burdens.1 For example, if the dollar amount of taxes doubles over a decade, has the revenue burden doubled? Some would say yes. Others would say that the effects of inflation should be taken into account when answering this question. Still others would point out that since average income has also doubled over this period, the actual “burden” has remained unchanged. They would argue that the burden should be measured as a function of people’s ability to pay, much the same way that banks consider a person’s income when deciding whether to grant a loan—the higher the person’s income, the more he or she can borrow. Yet others would argue that the size of the burden and the services it funds should be considered exclusively on its own merits. ____________ 1Note that “public revenue burden” will be used interchangeably with the phrase “state and local revenue burden” throughout the report. Although the concept of the public revenue burden in its most general sense would include federal revenues, the state/federal debate is not the focus of this policy debate or of this study. Consequently, we will use the public revenue burden concept only to refer to the state and local portion of that burden. 2 In this report, we will discuss and define what should be included in the public revenue burden in California and explore the main measures of that burden. A companion background paper (Shires, 1998) discusses how different levels of California governments—the state, counties, cities, special districts, school districts, and public postsecondary education institutions—have relied on different sources of revenue over the past 20 years. This report and the companion background paper provide the reader with the information necessary to calculate the average revenue burden under a variety of definitions for a variety of classifications of revenues and taxes for each level of California state and local government for five fiscal years between the passage of Proposition 13 in 1978 and fiscal year 1994–95 (the most recent year for which full revenue data are available from the California State Controller’s Office). The Debate over State and Local Finance Whether they intend to or not, all governments—through their annual taxing and spending decisions—provide answers to the following questions: How large should the public sector be? The size of the revenues and budgets controlled by governments reflects the long-run preferences of the citizens within a jurisdiction. Governments can only spend monies that the voters authorize either directly through the ballot or indirectly through the election of representatives sympathetic to the imposition of taxes and fees to raise those monies. In California, which makes extensive use of the initiative process, voters often make their preferences directly known. Also, the size of the public sector differs across states and across jurisdictions within a given state, because some local 3 governments raise more revenue than others. The citizens of Hawaii, New York, and Minnesota, for example, are noted for their willingness to raise more revenues than their fellow citizens in New Hampshire and Texas. Which goods and services should be provided and how much should be spent on a given good or service? Governments differ as to whether they will provide a given good or service and how generous they will be in its provision. They must decide how much should be spent on elementary or secondary education, on postsecondary education, on police, on fire protection, on the correctional system, on public parks, and on welfare benefits and services The services they provide depend on the resources they receive. Which government should provide which good or service? Even when two states provide approximately the same goods and services they frequently differ as to which level of government should provide these goods and services. Should medical services for the poor and elderly be provided by the state government, by county governments, or by city governments? In California, even though these services are provided at the local level, the MediCal program that controls and administers them is housed at the state level. In New York, these services are both funded and administered by local governments. California also provides many goods and services through special governmental units, called special districts, that in some other states are provided by regular state and local governmental units. As a result of this variety of institutional frameworks, it is misleading to look only at the pattern of revenues and expenditures at any one level of government because activities are sometimes shifted from one governmental unit to another. Instead, one 4 must take a broader view and look at all levels of state and local government in concert. How do governments raise the revenue needed to fund these goods and services? Traditionally, governments have raised their revenues through taxation. At the state and local levels, the traditional taxes have been the income, property, and sales taxes. Over the years, state and local governments have imposed different taxes as well as different levels of those taxes. California, for example, relies on income, property, and sales taxes, whereas New Hampshire relies almost exclusively on the property tax. Some observers have claimed that in California, as traditional governmental units have been restrained in their ability to raise taxes, the provision of goods and services has been shifted to other units, especially special districts; and, as the taxing power of traditional governmental units has been constrained, they have found new ways, particularly through fees and exactions, to fund the provision of public sector goods and services. Who bears the burden of providing the revenues? Economists traditionally ask two sets of questions when analyzing the economic effects of taxation: (1) Does a given tax distort the efficiency of economic activity, and (2) who bears the burden of that tax? The question of who bears the burden of taxation is usually answered in terms of whether the tax burden is greater on one income class than another. For example, are taxes a greater proportion of the incomes of those at the upper income levels than of those at the lower income levels? The fact that different state and local governments raise different amounts of revenue through different mechanisms means that the revenue burden can differ not only across geographic jurisdictions but also across income 5 categories. Not only will a citizen of Tennessee pay lower taxes than an equivalent citizen of Hawaii, but even within a state, a citizen of a city or county that provides minimal services will likely pay less than the equivalent citizen of a city or county that is generous in its provision of goods and services. Who will benefit from the goods and services? Not only does who pays for state and local government services differ, but also who receives the services. In most state and local governments, a certain amount of redistribution occurs when governments raise revenues for certain programs from groups of residents who do not directly receive the benefits of the services funded by their taxes. The magnitude and role of this redistribution, whether intended or not, will differ not only across states but also across local jurisdictions within states. This report addresses only some of these questions—those related to the revenue side of the budgetary equation. It does not address the magnitude and distribution of state and local spending, the magnitude and purposes of public sector borrowing, or the broad issue of efficiency in government. Its focus on the revenue burden, however, is important and relevant for two reasons. First, at the state and local levels where governments are mandated to balance their operating budgets, the amount of revenue raised has a significant effect on the size and mix of spending. Second, the debate over the appropriate size of the revenue burden has been central to the ongoing California debate over the appropriate size and role of government. An Anatomy of Public Finance in California The United States is governed under a federal system. Under the American Constitution, sovereignty is jointly held by the national and 6 state governments. The national government is granted a set of specific powers under the Constitution, and all other powers are reserved for state governments and the people. Local governments2 have no sovereign power other than that granted to them by their states. Sometimes this power is provided for in the state’s constitution and sometimes by statute. Because local governments are creations of their respective states, their powers to raise revenues are set out in the state’s constitution and in state statutes. In a similar fashion, because national laws take precedence over state laws, it is sometimes the case that a state’s power to raise revenues is constrained by actions of the federal government. In practice, this federal system has created a complex set of fiscal relationships between and among the states, their local governments, and the national government. The powers of the national government are not fixed but have changed over time as the Supreme Court has interpreted the Constitution. This has meant that, since Franklin Roosevelt’s New Deal of the 1930s, the federal government has increasingly funded and regulated what in prior years were thought to be solely state and local activities. In addition to the increasing complexity of federal involvement in state and local finance, the institutional framework created by California has rendered the state’s public finance framework almost incomprehensible. In 1995, California had nearly 5,000 independent governments. In addition to the state government, it had 58 counties, 470 cities, 3,217 independent special districts,3 and 1,001 school ____________ 2The term “local governments” will refer throughout this report to governmental entities within and below a state government, including counties, cities, special districts, and school districts. 3Independent special districts are those governed by either independently elected boards or boards appointed by multiple jurisdictions. 7 districts. In addition, California funds three public postsecondary education systems: the University of California system (with nine campuses), the California State University system (with 22 campuses), and the California Community Colleges system (with 71 districts and 106 campuses). The situation is further complicated because the various governments of California receive grants and other transfers from both the federal government and the state. As indicated in Table 1.1, in fiscal year 1995 these governmental units reported more than $204 billion in total revenues. Although, as we will find in Chapter 2, this total includes several categories of revenues that we may wish to exclude from our estimation of the size of state and local revenue burdens in California,4 the size and scale is still considerable. California governments obtain their revenues through taxes, fees and fines, various types of activities for which a charge is levied (such as public utilities), and intergovernmental transfers. Figure 1.1 shows us Table 1.1 FY 1994–95 Total Public Revenues in California, by Level of Government Receiving the Revenue Government Entity State Counties Cities Independent special districts School districts Public postsecondary institutions Revenues ($ Billions) 85.6 31.9 30.8 12.7 27.7 15.9 % of State and Local Total 41.8 15.6 15.1 6.2 13.5 7.8 Total 204.6 100.0 SOURCES: Compiled from numerous state agency publications. ____________ 4In the case of intergovernmental revenues, adding these reported revenues can result in double-counting those revenues, as we will see in Chapter 2. 8 Service 5% Interest 2% Enterprise 11% Other 3% Property tax 10% Sales tax 10% Income tax 12% Intergovernmental transfers 39% Figure 1.1 Activity-based fees 8% Figure 1.1—Reported Public Revenues of California State and Local Governments, by Revenue Base, FY 1994–95 that these governments receive the largest share of their revenues from other levels of government, which account for 39 percent of state and local government revenues. Income, sales, and property taxes combined account for nearly one-third of these reported revenues, whereas regulatory fees account for 8 percent and enterprise activities (such as water and sewer services) account for another 11 percent. The level of reliance on any one of these revenue types differs with the level of government, as Table 1.2 shows. This table sets out the percentage of fiscal year 19955 revenues that resulted from each of these major categories for each level of California state and local government.6 ____________ 5In California, the fiscal year extends from July 1 to June 30 of each year. Hence, fiscal year 1995 is the period July 1, 1994, through June 30, 1995. 6Note that some of the revenue streams provided in this table, especially those relating to cities, counties, and special districts, do not directly correspond to the detailed totals reported for each entity by the California State Controller. This is due to the reclassification of dependent special districts’ revenues into their respective parent entities’ revenues. For example, the revenues from county service areas, which are created and 9 Table 1.2 Percentage Share of FY 1994–95 Revenues from Various Sources, by Level of Government Receiving the Revenue Revenue Type Property taxes Sales taxes Income taxes Regulatory fees Intergovernmental Interest Charges for non- enterprise services Charges for enterprise services Other Public Post- Special School secondary State Counties Cities Districts Districts Institutions 0.0 12.5 14.5 10.2 32.0 8.4 19.8 1.5 10.2 2.1 0.0 0.0 29.0 0.0 0.0 0.0 0.0 0.0 12.7 2.7 12.7 0.8 3.2 0.0 37.1 55.9 14.0 20.8 59.6 52.6 0.6 1.8 4.4 6.3 1.7 0.5 0.4 7.1 5.4 0.5 2.5 28.5 0.0 13.9 31.9 52.1 0.0 7.4 0.4 4.6 6.9 7.2 1.0 2.6 Total 100.0 100.0 100.0 100.0 100.0 100.0 NOTE: The City and County of San Francisco is included as a city. Dependent special district revenues have been included in parent entities’ revenues. One clear picture that emerges from the percentages shown in Table 1.2 is that although the state government raises most of its revenues from taxation, each of the other levels of government obtains less than a third of its revenues through traditional taxation. Counties and school districts depend highly on intergovernmental transfers of funds (primarily from the state), whereas independent special districts obtain more than 60 percent of their revenues from the enterprises they run (such as providing water and fire protection). California’s cities have the ____________________________________________________ managed by county boards of supervisors, are included as county revenues, not as special district revenues, as is done in the State Controller’s reports. Furthermore, the revenues for redevelopment agencies are also reported in the revenues of their respective parent entities, typically cities. As a result, although the proportions shown in this table do not directly correspond to other published sources, we believe that they more accurately reflect the distribution and control of the monies they represent. 10 most diverse sources of revenues; but even in their case, only one-quarter of their revenues are generated from locally levied taxes; over half come from activities, enterprises, and services provided for a fee. How Can Revenue Burdens Differ? The vast array of California governments that fund their activities through a variety of mechanisms means that two neighbors can face very different revenue burdens. The most obvious and familiar case occurs when residents are separated by a city or a county line. Frequently, adjoining cities or counties provide very different levels of services and, to fund these, levy very different levels and mixes of taxes and fees. But even within the same city or county, neighbors can face different revenue burdens. This happens because the boundaries of California’s special districts and school districts are not necessarily contiguous with the bounds of its cities and counties. Thus, two adjoining neighbors might find themselves located in different school districts, fire districts, flood control districts, mosquito abatement districts, and so on. These districts often have different fee structures and, sometimes, different tax levels. Neighbors can also face different revenue burdens depending on whether they take advantage of publicly provided services that are elective. If one neighbor regularly uses public transportation and another walks to and from work, the two would pay different amounts to the government. If one neighbor uses state or county parks each weekend and his neighbor stays at home, the former will bear a greater share of the revenue burden since he will regularly be paying park fees. Even when it comes to taxes, neighbors may face very different revenue burdens. Because Proposition 13 limits increases in the property 11 tax to 2 percent per year but allows full reassessment every time a property is sold, two neighbors in identical houses might pay very different property taxes if one bought his house in 1974 and the other bought his house in 1988. In another instance, one neighbor might buy a how-to book on sailing, and the other might take sailing lessons. The first would pay sales tax on the book but the second would pay no sales tax on the sailing lessons. For all of these reasons, it is difficult to calculate the revenue burden for each California citizen, but that is not a goal of this study. Rather, we wish to determine the size of the financial burden the government imposes on society and how it has changed since the passage of Proposition 13. What Others Have Found Clearly, thinking about Proposition 13’s effects on what people pay in taxes is not a new idea. Several studies have addressed this question. For example, Dunstan (1993) examined long-term trends in state, city, and county finances from 1975–76 to 1990–91, focusing on the major revenue and expenditure categories. This study, however, did not include the revenues and expenditures of three important categories of public entities—school districts, independent special districts, and public postsecondary education institutions. Sheffrin and Dresch (1995) used an economic approach to examine the tax burden, assuming that all taxes are eventually passed on to individuals in the form of higher prices for goods and higher rents. Although arguably an appropriate methodology, this study looked only at the three main categories of revenues—personal income tax, sales and use tax, and residential property tax—identifying who ended up paying 12 these taxes and how much they paid. The study did not include the many other forms of revenues that the public sector generates. Although these two studies appear to be directly relevant to our question, they looked only at a portion of the revenue burden. Four other studies, however, have more specifically addressed the question we are studying. Unfortunately, the underlying assumptions and results of these studies have differed widely and their findings have raised more questions than they have answered. Table 1.3 presents some selected findings from these studies. A detailed examination of this work helps explain why we have chosen to undertake this project. The need for our research arises, not from the failing of any of these studies but rather from the complexity and range of assumptions that are implicit in an examination of the state’s complicated state and local finance landscapes and the need to place the diverse findings of these previous studies in context. In addition, the recent passage of Proposition 218 seems to indicate that the public perceives that there is still a need to restrict local government’s ability to manage and increase its revenues. Since none of Table 1.3 Selected Findings of Prior Studies on Public Revenues in California, as a Percentage Share of Personal Income Study Gold California Taxpayers’ Association Kirlin et al. Legislative Analyst’s Office 1977–78 14.6 16.7 18.7 — 1990–91 11.3 16.0 — — 1991–92 — 16.2 19.0 16.6 SOURCES: Gold (1993); California Taxpayers’ Association (1994; 1991–92 estimate obtained separately from the California Taxpayers’ Association); Kirlin et al. (1994); Legislative Analyst’s Office (1995). 13 the previous work provides us with the most recent information available, our study serves a valuable function by including data from fiscal year 1994–95. Each study in Table 1.3 measured the public revenue burden with slightly different assumptions and concerns. We briefly discuss each of these studies below and then discuss in greater detail why we felt that this follow-on study was needed. Appendix A contains a more detailed discussion of our efforts to reconcile our findings to these previous studies. Steven Gold Steven Gold’s study was included as part of his testimony at a joint hearing of the California Senate’s Budget Committee and Human Services Committee on October 1, 1992. The study also served as the basis for his presentation at a symposium on “State Taxation of Business Activities” at UC Davis in March 1993. Gold’s approach to estimating the public revenue burden focused exclusively on taxes. As Table 1.3 shows, his research found that state and local taxes in California declined from 14.6 percent of personal income in 1977–78 to 11.2 percent in 1990–91. The analysis focused only on the tax portion of public revenues and did not include other miscellaneous revenue categories such as fees, interest, and fines. California Taxpayers’ Association California Taxpayers’ Association (1994) presented estimates of the public revenue burden in California, at least through 1990–91. Its analysis was based on U.S. Bureau of the Census reported amounts and showed a near return to pre-Proposition 13 revenue levels by 1990–91. 14 Although these findings are quite compelling, there are several issues to consider when using Census data. The Bureau of the Census recompiles State Controller’s data to include subsidiary entities’ revenues with their parents’ revenues while simultaneously converting California’s revenue category definitions into the Census’s more generic structure. Preliminary comparisons of the work done for this study to the Census’s reported revenue amounts indicate some significant variation in the totals and subtotals reported for the various revenue categories—including the total revenues reported by each level of government.7 Because of our exhaustive efforts to aggregate and classify the data at the lowest level of detail possible and the confidence we have in the data we are using,8 we believe that our estimates, which will be presented in Chapters 2 and 3, more appropriately reflect state and local government revenues. Kirlin et al. The Kirlin et al. study was prepared for the Task Force on California Fiscal Reform of the California Business—Higher Education Forum. It was published as Chapter Five in the Forum’s Task Force Member Report California Fiscal Reform: A Plan for Action. Tables 5.1 through 5.3b of the report (pp. 45–48) contain detailed estimates of the public revenue burden in California for the fiscal years 1977–78, 1988–89, and 1991–92. ____________ 7Although the possibility of inconsistencies and classification problems in the Census data have been raised by others (for example, see Leigland, 1990), this study is the first to do a detailed recalculation for California. For this study, it was necessary to keypunch and verify data from the 300 to 600 pages of published reports for each year studied. We have, therefore, excellent revenue data, which should correspond directly to the revenues reported by the Bureau of the Census. The reported Census revenues will be debated in another paper. 8See Shires and Glenn Haber (1997) for a detailed discussion of the quality of our data. 15 The authors’ data and the assumptions implicit in their study correspond most closely to those we use here, but there are some issues associated with their data that are problematic. First, because the 1991– 92 special district data were not available when they prepared their estimate of the state and local revenue burden, they had to use 1990–91 data for this important information. Furthermore, reporting inconsistencies in the data for redevelopment agencies, transit districts, school districts, and some public postsecondary institutions create a scenario where those estimates are not fully comparable from year to year. We have tried to correct for these inconsistencies in our study. California Legislative Analyst’s Office The Legislative Analyst’s Office estimate presented in Table 1.3 comes from the “State and Local Finance—Fiscal Overview” section in the Legislative Analyst’s Office’s 1995 Cal Guide: A Profile of State Programs and Finances. The estimate of the state’s public revenue burden was taken from a U.S. Department of Commerce estimate of the state’s public revenue burden. Since this analysis used Bureau of the Census data, it calls up the same issues that were raised about the California Taxpayers’ Association estimates above. Why We Need Another Study Even with the results of these four studies in hand, the subject needs to be reexamined for several reasons. First, we hope to resolve some of the uncertainties raised by the differing answers found in these studies. Second, we hope to provide a time series of the public revenue burden in California that is directly comparable from year to year. As the discussion of the four studies above demonstrates, significant data issues 16 have not previously been fully addressed. Our report will do so. We also hope to bring the findings of these studies up to date—since most of the studies were completed for years during or before the relatively severe recession of the early 1990s. Our updated study will also allow us to assess the effects of the many policy changes instituted as a result of the recession. Finally, we hope to fully address the question of how to best measure the public revenue burden in California. Our Study This study is the second of a three-part series on state and local finance and the effects of Proposition 13 on local citizens. Our first study (Shires and Glenn Haber, 1997), which examined the quality of the data available on the revenues of state and local governments in California, found the aggregate data to be both comprehensive and accurate. This report presents the findings of our overall assessment of the public revenue burden in California. It looks at the issue of the public revenue burden at the statewide level and in aggregate numbers, focusing on what should be included in that revenue burden and the best way to measure it. The third and final study in this series will focus on how Proposition 13 affected individuals in California. It will examine the effects on individual residents of the state of the changed fiscal structures and institutions documented in theses two reports. Because of the significant costs associated with preparing the data for each year, we will review only five years in our study. However, we believe that the five years we have selected will provide a good picture of what has happened to the public revenue burden in the years since 17 Proposition 13. This report looks in great detail at the public revenue burden in California for the fiscal years 1977–78, 1980–81, 1987–88, 1991–92, and 1994–95.9 Although there were many reasons for choosing these years, our primary reasons centered on the timing of Proposition 13 and subsequent California business cycles. We selected fiscal year 1978 as a baseline for the research because it was the year that Proposition 13 was passed by the voters. 1981 was selected because it falls soon after the full effects of the implementation of Proposition 13 were able to work their way through the public finance system. 1988 and 1992 were chosen for their comparability as points in the business cycle10 and 1995 because it is the most recent year for which data are available. For a more detailed explanation of our choice of these five years, see Appendix B. Organization of This Report This study has three remaining chapters. Chapter 2 explores the issue of what public revenues should be included in an estimate of the public revenue burden and Chapter 3 examines how we should measure the size of the public revenue burden. Chapter 4 then discusses the findings presented in these two chapters in the context of the policy debate presented in the introduction and the implications of these findings for citizens and policymakers. ____________ 9Throughout this report, the year described will refer to the fiscal year ending in that year unless otherwise noted. For example, a reference to “1978” would be referring to the fiscal year ending June 30, 1978, and is also written as 1977–78. 10Comparability across the business cycle refers to the fact that each of the two years, 1988 and 1992, represents a point in the business cycle (e.g., a period of economic expansion or recession) that was similar to our first two years, 1978 and 1981, respectively. 18 Chapman, USC School of Public Administration; Stephen Kroes, California Taxpayers’ Association; Paul Lewis, PPIC; and Marianne O’Malley, California Legislative Analyst’s Office. Their comments and suggestions for revisions and improvements to this report have been invaluable. We would also like to thank Marcela Cordon, who spent long hours keypunching and validating the raw data necessary to this study. The project also benefited from the patience and efforts of Michelle Chaffee, formerly of PPIC, who spent many days poring over the data tables for this study to make sure that we were using the right numbers. Finally, we also wish to acknowledge the contribution of PPIC’s Advisory Council and Board of Directors, who provided helpful direction, comments, and suggestions during our numerous briefings and discussions. Although this report reflects the contributions of many people, the authors are solely responsible for its content. xviii 2. Defining the State and Local Revenue Burden in California The question, What has happened to the state and local revenue burden in California since Proposition 13? has two parts: (1) What are public revenues in California and (2) how have they changed over time? In this chapter, we will explore the first part of the answer to that question: What should be included as public sector revenues when we think about the public revenue burden? First we will discuss the types of revenues that are commonly reported by governments and agencies at the state and local levels. To help us do that, we will develop a taxonomy or menu of public revenues into which we will subsequently group all public revenues. We will then look at this taxonomy and select only those revenues that appropriately belong in our study. As we will see, we may not wish to include certain types of revenues in our estimates. We will then present the totals of these revenues for the years we studied. 19 The Public Revenue Burden in California: A Series of Choices As the differences among the studies presented in Table 1.3 show, estimates of the overall size of the public revenue burden depend heavily on the types of revenues included. For example, one might include university fees in the public revenue burden because these institutions are indeed public. However, these institutions reflect one area where the public sector is in direct competition with the private sector and their fees represent a specific choice made by state residents to purchase education from the state instead of the private sector. Residents are not required to attend a public university and incur this fee. To determine the kinds of revenues that are claimed by state and local government entities, we looked at the standard reports that state and local governments provide which include details of their revenues. For the state, this source is Schedule 8 of the California Governor’s Budget Summary, which is available for each year.1 For cities, counties, special districts, and redevelopment agencies the appropriate source is the Annual Financial Transactions series published by the California State Controller’s office. For K–12 school districts, our source included the State Controller’s reports and a special data analysis prepared for us by the California Department of Education. Finally, we used information from the California Postsecondary Education Commission to examine the revenues of the state’s public postsecondary education institutions. The overall revenues reported by each level of state and local government for each of the years in our study are presented in Table 2.1. ____________ 1In 1978 and 1981, the corresponding schedule was Schedule 2. Federal intergovernmental revenues for the state were taken from Schedule 6 for 1978, from Schedule 3 for 1981, and from Schedule 9 for 1988, 1992, and 1995. 20 Table 2.1 Public Revenues in California for All Levels of Government, and Their Percentage Share, by Level Receiving the Revenue Level Receiving Revenue State 1978 1981 1988 23,202,407,119 32,386,866,000 52,915,676,000 39.9 41.8 39.8 County 9,182,173,418 11,002,462,589 19,629,130,409 15.8 14.2 14.7 City 8,472,134,749 11,329,778,338 20,248,790,547 14.6 14.6 15.2 Independent special districts 3,665,344,143 5,282,467,621 10,274,183,865 6.3 6.8 7.7 School districts 8,978,391,928 10,963,606,830 18,805,199,859 15.4 14.2 14.1 Public postsecondary education 4,630,040,287 6,509,053,000 11,376,175,000 8.0 8.4 8.5 1992 79,313,651,000 42.3 29,873,677,459 15.9 27,325,541,671 14.5 11,724,148,380 6.2 24,915,087,181 13.3 14,696,063,000 7.8 1995 85,622,226,000 41.8 31,858,760,981 15.6 30,796,774,219 15.1 12,726,996,423 6.2 27,674,571,941 13.5 15,905,264,000 7.8 Total 58,130,491,644 77,474,234,378 133,249,155,680 187,848,168,691 204,584,593,564 100.0 100.0 100.0 100.0 100.0 21 As this table shows, the public sector overall is large—totaling about $205 billion dollars in fiscal year 1995—and has more than tripled over the approximately 20 years shown in the table. Note that this table includes an entry for independent special districts but not for dependent special districts, because dependent special districts are included as part of their parent governments.2 Clearly, all levels of government grew significantly over this time period—as one might expect in a state where the population grew by more than 40 percent. Overall, these revenues grew about 252 percent— a significant portion of which can be accounted for by inflation. If one considers the shares of revenues generated by each level of government, it is clear that they are quite stable, with the exception of school districts. We see that the share of revenues reported by school districts declined from 15.4 percent to 13.5 percent of total state and local revenues. Part of the explanation for this is that school district revenues grew at a slower rate than overall revenues, increasing by only 208 percent from 1978 to 1995, whereas the revenues of some levels of government, such as cities, grew much more quickly—rising 264 percent. Although some may find this table interesting, it is too summary in nature to allow us to fully understand the public revenue burden in California. It contains all revenues reported by state and local ____________ 2This distinction includes redevelopment agencies as well as all special districts that are listed in the State Controller’s reports as being governed by a city council or board of supervisors. The specific revenues within each dependent district were classified individually (e.g., tax or intergovernmental), but they were designated as revenues to the parent government (usually a city or county) instead of being classified as special district revenues. As a result of this adjustment, the totals for cities and counties will be slightly higher than the raw number included in the State Controller’s reports. For a detailed discussion of the amount our revenues reclassified in this manner, please see Shires (1998). 22 governments, including many that we may not wish to include. For this reason, we must examine these revenues in greater detail. To do this, we have created a taxonomy to classify public revenues in California. The classification of revenues into this taxonomy is not an easy process because of the numerous inconsistencies in the way revenues are reported by different types of governments. To develop our taxonomy, we start with the standard revenue categories commonly used for cities and counties and then expand them to address some of the issues we raised above, especially with respect to regulation and service areas where the government is not the exclusive provider of a service. The resulting taxonomy is presented in the next section along with its application to the revenues shown in Table 2.1. A Taxonomy of Revenue Types The State Controller’s reports for cities and counties divide their revenues into eight categories: taxes, special benefit assessments, licenses and permits, fines and forfeitures, revenues from the use of money and property, intergovernmental funds, current service charges, and other revenues. There are some limitations to these categories, however. Some of them disguise aspects of the revenues that we care about. For example, current service charges include such diverse revenues as university fees, water revenues from a municipal utility, filing fees for zoning permits, copying charges for public records, and fire department charges for false alarms. To help us draw distinctions among these revenues and because of the differences in the way information is reported for the different levels of government, we have modified the taxonomy to include ten 23 categories. We have separated current service charges into three categories—enterprise revenues, other service revenues for which the government is the exclusive service provider, and revenues generated by government operations that compete with the private sector. We have also broadened revenues from the use of money and property, except interest, into this latter category.3 We have also broadened the licenses and permits category to include regulatory fees and charges. This change serves two purposes that the original classification did not easily allow: (1) It allows for easier classification of revenues from other levels of government (especially the state), and (2) it recognizes a category of revenues that the courts have, in the past, required governments to reinvest into the regulated activity (regulatory fees and charges). As a result of these changes, our taxonomy has the following categories: taxes; assessments; regulatory fees and charges; fines, penalties, and forfeitures; intergovernmental revenues; interest; enterprise revenues; service revenues from activities where the government is the exclusive provider; service revenues from general services; and other revenues. Below is a brief description of each of these revenue categories. Taxes Taxes are fees paid to the government by residents who choose to participate in an activity, such as owning property, buying general goods, earning income, buying gasoline or alcoholic beverages, participating in business activity, operating a franchise, staying in a hotel room, or purchasing electricity. ____________ 3We argue that, in general, the government is in fact competing with private savings and real estate markets when it earns interest and rents from its assets. An exception to this rule would arise, for example, in the case of rents in a port or airport where the government is the sole provider of space. 24 Assessments Assessments are generated as the result of specific voter action to pay for specific services. They differ from general property taxes because their level is based on an estimate of the benefit they will provide to a specific property and on the actual cost of the improvement, rather than on the overall value of the property. They are most often generated under the auspices of voter-approved ballot measures, as required by Proposition 13, and, more recently, Proposition 218. Unfortunately, as prior work by PPIC and others has shown,4 the amounts reported by local governments under this heading do not reflect the full range of assessment revenues received by local governments. The distinction is retained in our taxonomy, however, because these revenues are significantly different from the tax revenues described above. In the past, these revenues have been imposed both at the behest of the voters and through the intervention of locally elected officials, although the passage of Proposition 218 has resulted in assessments being handled much more like general property taxes than had previously been the case. Regulatory Fees and Charges This category is quite similar to the general tax category above in that it represents a revenue stream where the government charges a fee ____________ 4Shires and Glenn Haber (1997) found that revenues from Mello-Roos assessments were commonly unreported or were reported by local governments as part of overall property tax revenues. The size of the problem was not found to be large relative to overall state and local revenues, but it is significant if one wishes to focus exclusively on assessments. In general, the current state and local government revenue-reporting structure is not adequate to identify revenues from special assessments. This fact was recently highlighted by the presence of Proposition 218 on the November 1997 ballot. As they tried to estimate the effects of this initiative—which severely constrained assessments by local governments for specific purposes—on local government finance, policy analysts from all agencies and groups discovered that there was very little information available. 25 for permission to undertake some activity. Regulatory fees and charges include two types of revenues: (1) revenues generated to fund specific regulatory activities, such as the Public Utilities Commission and various licensing boards, and (2) revenues generated as the result of permits issued as part of a specific regulatory process, such as construction permits and fish and game licenses. Planning fees and animal licenses also fall in this category. In some ways these could be looked upon as taxes. A construction permit, for example, could be viewed as a tax on the activity of construction much as a business license is a tax on the activity of being in business. They are much less generic, however.5 Reporting these regulatory fees and revenues in a separate category allows us to consider revenues that are generated by the regulatory power of government. It also has the advantage of allowing us to either make the distinction or not. By reporting these revenues separately, if one is not comfortable with the distinction used here, he can simply combine these revenues with taxes and ignore the differentiation. Fines, Penalties, and Forfeitures These revenues represent payments to government by individuals who have violated provisions of the state and local codes. Included in this category are fines for traffic violations, penalties on late property taxes, and parking fines. These are effectively taxes on socially ____________ 5For example, owning a business is much more generic than building specific structures. We do recognize, however, that this category and the tax category represent a continuum and that some fees could be classified as either a tax or a regulatory fee under these categories. Because of general practice within the policy community and the specificity criteria we raised above, however, business licenses and franchise taxes are not included here, but are included rather as taxes. 26 unacceptable activity and serve a broader social purpose of providing deterrents to these activities. Intergovernmental Revenues These revenues represent transfers from other levels of government, either restricted or unrestricted. For example, intergovernmental revenues for school districts include general state appropriations and funding under the Proposition 98 guarantee6 as well as revenues earmarked for the construction of schools. It also includes federal revenue sharing, state support for a range of local programs including children’s centers and mental health, and county support for city programs. Since the goal of this report is to determine the total revenues that the public sector obtains from private individuals and industry, intergovernmental revenues should also include current service charge revenues that are billed to public clients—such as electrical revenues received by public utilities from other governments. In most cases, this is not easy or even possible to do. There is one major exception, however—self-insurance districts. Over the past 20 years, numerous governments and groups of governments have set up independent special districts to provide self-insurance programs or to pool resources to provide insurance. The revenues reported by these independent districts are reported as intergovernmental revenues in this analysis, since they are generated exclusively from other public entities. This distinction is ____________ 6Proposition 98, passed in the June 1988 election, sets minimum funding levels for the support of K–14 education in California. The “guarantee” is the amount required under the provisions of the proposition. 27 important because these revenues are significant, exceeding $1 billion per year in recent years. Interest In the case of “revenues from the use of money and property category,” we chose to separate out interest revenues and to report the other two common revenues in this category—rents and royalties—as general services. This distinction goes directly to the issue of the reason why governments generate these resources and whether there is private sector provision of the resources. In general, interest revenues arise from the holding of monies that are usually held for other governmental purposes. State and local governments do not generally pursue the lending of money and the earning of interest as a primary business activity, as a bank would.7 The revenues generated from interest, therefore, can be considered as a byproduct of other activities, and private sector competition probably does not have too much of an effect on whether those revenues are earned or not. Rents and royalties, however, almost certainly have strong private sector competition and, in some cases, government activity may actually be “crowding out” private sector activity. This revenue stream, therefore, seems to represent a general service to the community and, except where the government has a monopoly on the type of resource provided—such as an airport—should be categorized as a general service revenue. Where the government has an actual or de facto monopoly, this revenue should ____________ 7There have been some notable exceptions to this. See Baldassare (1998) for a detailed description and discussion of one such case. 28 be classified as a service revenue associated with the government’s exclusive provider status. Enterprise Revenues Enterprise revenues are those generated by such services as sewer, water, electric, gas, and transportation. They arise from local publicly owned monopolies and are typically provided by either an independent special district or a quasi-independent department within the government. This enterprise revenue distinction is important from a policy perspective when considering the fact that not all of the provision of goods and services in these categories comes from publicly owned monopolies. In the case of electric power, for example, the City of Los Angeles has a publicly owned enterprise—the Department of Water and Power—whereas much of Northern and Central California are served by a privately owned company—Pacific Gas and Electric. Especially with deregulation in the offing, it is important to distinguish these revenues for policy purposes. In general, the revenues from these entities are largely committed to the costs and production of the specific goods and services. Service Revenues—Exclusive Provider This category of revenues represents activities for which a government receives revenues and for which it is the sole provider of that service and for which the service revenue is not the result of an enterprise activity. It category includes state lottery revenues, charges for the holding of elections, tax collection fees, and the costs of specialized police and fire services. 29 Service Revenues—General These revenues are generated by activities that are also commonly provided by nongovernmental entities. The largest revenues in this category include rents, ambulance services, golf course fees, university fees, and hospital revenues. Other Revenue This category includes all other revenues that do not fall into the above-defined definitions or for which the detail to classify the revenues was unavailable. It also includes donations from private sources and revenues from discontinued special districts. What Is Not Included It is important to note that this analysis does not include revenues from bond proceeds. The issuance of debt is important to public policy in California, but it does not conceptually fit into our main research question. The funds generated by bonds do not reflect a true revenue to the local government—just as an individual is not taxed by the government for funds borrowed to purchase a house.8 What are included, however, are the revenues generated to pay off the debt. In many cases, such as special bonds for schools, new revenues are generated to pay for these debts, usually in the form of property tax revenues and special assessments. As noted above under taxes and assessments, these revenues are captured in our analysis. ____________ 8Note that if we were concerned with the expenditure side of government, then these sources of funds would be quite important because they would be used to fund the provision of some assets or services. For revenue purposes, however, the debt assumption and issuance is a nonevent. 30 Applying This Taxonomy to California Let us then apply this taxonomy to state revenues overall and ask the question, What precisely is the nature of state and local government revenues? Table 2.2 provides us with an understanding of the composition and character of the overall state and local revenue burden. This table shows us that the largest revenue source for state and local government is generally taxes, but that their importance declined overall immediately after Proposition 13 and has remained relatively flat since. The share of intergovernmental revenues, which peaked in the early 1980s, is slightly higher than pre-Proposition 13 levels, whereas service revenues—including enterprise and services that are both generally and exclusively provided—have risen to fill the void left by the decline in tax revenues, rising from about 11 percent of revenues in 1978 to about 16 percent today. Assessments, although they account for a relatively small share of overall revenues, have risen the most dramatically, increasing some 1,832 percent since 1978. Taxes and intergovernmental revenues are the two slowest growing categories, growing at 186 percent and 276 percent, respectively. Because they account for 77 percent of overall revenues, the overall growth of all revenues is 252 percent. Because of the uneven intervals between the years in our study, it is difficult to visualize the trends represented in Table 2.2. For this reason, it is helpful to look at these trends in graphic form, as in Figure 2.1. This figure gives a better sense of the overall trends in the sources of revenues over time—showing the long period of relatively little change in the shares of revenue categories. Even with the categorical distinctions, however, this figure does not tell the full story of overall state and local 31 Table 2.2 Public Revenues in California for All Levels of Government, and Their Percentage Share, by Revenue Type 32 Revenue Type Taxes Assessments Regulatory fees and charges Fines and forfeitures Interest Intergovernmental Enterprise revenues Exclusive provider General services Other Total 1978 26,409,666,294 45.5 32,850,839 0.1 1,390,237,527 2.4 252,254,534 0.4 856,891,141 1.5 21,694,128,405 37.3 4,476,246,475 7.7 201,040,942 0.3 1,683,879,855 2.9 1,133,295,632 1.9 58,130,491,644 100.0 1981 28,392,188,943 36.6 162,711,104 0.2 1,338,110,598 1.7 362,918,091 0.5 1,838,620,532 2.4 32,456,331,967 41.9 7,733,753,254 10.0 287,272,242 0.4 3,133,030,720 4.0 1,769,296,927 2.3 77,474,234,378 100.0 1988 52,504,319,843 39.5 1992 72,070,935,103 38.4 422,836,209 0.3 560,079,307 0.3 3,228,228,698 2.4 4,101,351,362 2.2 953,171,700 0.7 988,966,277 0.5 3,366,566,608 2.5 3,871,848,963 2.1 49,489,137,169 73,373,444,285 37.1 39.1 13,714,929,929 18,857,594,195 10.3 10.0 1,095,613,493 0.8 1,182,836,525 0.6 5,567,580,382 4.2 7,179,561,943 3.8 2,906,771,649 2.2 5,661,550,731 3.0 133,249,155,680 187,848,168,691 100.0 100.0 1995 75,540,377,001 37.0 634,712,424 0.3 5,497,714,659 2.7 1,080,762,038 0.5 3,751,894,129 1.8 81,478,340,742 39.8 22,061,142,572 10.8 1,468,815,789 0.7 8,152,364,018 4.0 4,918,470,192 2.4 204,584,593,564 100.0 Percentage 50 45 40 35 30 Taxes Intergovernmental 25 Enterprise All others 20 Regulatory Exclusive 15 10 5 0 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 Figure 2.1—Overall Share of State and Local Revenues, by Revenue Type revenues. It is important to recall that there has been tremendous growth in overall state and local revenues over this period. Before discussing how to measure this growth, we must deal with an important bookkeeping issue: Should all of these revenues be considered part of the public revenue burden or are there legitimate reasons for excluding some of them from our estimations? In the following section we reexamine our taxonomy of revenue types and consider why we may wish to exclude some of the revenue types from our estimate of the public revenue burden. The Major Choices in Calculating the Public Revenue Burden When estimating the size of state and local government, it is important to consider which revenue categories should be included. 33 Clearly, a case must be made for including or not including any specific revenue type when estimating the overall public revenue burden in California. We believe that an easy argument can be made for including nearly all reported categories of revenues, except three: intergovernmental revenues, enterprise revenues, and general service charges. Should We Include Intergovernmental Revenues? Although intergovernmental revenues are important in determining how much money a particular type of government has to spend, it confuses the issue if one is trying to determine how much money the public sector is raising from its citizens. In fact, as we try to sum total revenues, counting these revenues would effectively double-count them. Take the case of an intergovernmental transfer from the state to a city. The state recognizes the revenue when it receives it in the form of a sales tax, income tax, or some other revenue. It then chooses to give some amount to a city government, say for law enforcement purposes. The city then reports this money as intergovernmental revenue from the state. In reality, however, no money has changed hands from the private citizens to the city. Only an allocative, administrative event has occurred between the state and the city. It is important, therefore, to exclude such funds. We have excluded all revenues that were reported as transfers from other levels of state and local government. The above argument holds true for state and local intergovernmental transfers, but there is the further issue of intergovernmental transfers from the federal government. Clearly these revenues come from outside the state and local revenue stream and hence the argument against their inclusion due to double-counting is invalid. We must then turn back to 34 the purpose of our analysis for guidance. We wish to estimate the public revenue burden at the state and local levels in California. Would including these federal transfers further this purpose? From one perspective, the answer is yes. These revenues do represent revenues to the local government that may have come, ultimately, from the local taxpaying population. To include them requires that we address the more global issue of whether Californians are in fact net contributors or benefactors in the federal tax and budget system. Although some literature indicates that California is a net contributor, the validation of this research is beyond the scope of this report. In light of these issues, we have chosen to exclude all identifiable federal intergovernmental revenues from our estimates of the public revenue burden in California. Our argument turns on the fact that the decisions about the amount of these revenues are made outside California and hence do not arise as part of the power of the people of California to raise revenues for their own purposes. Should We Include Enterprise Revenues? Enterprise revenues are generated by state and local governments to provide significant public services, such as water, electricity, natural gas, airports, and ports. We may wish to exclude these revenues in a public revenue burden estimate for two reasons: (1) They generally fund specific services and produce little surplus, and (2) they are not consistently reported as public activities. With regard to the first point, these revenues usually reflect the true cost of doing business in providing a desired service. Therefore, these revenues are not generally available for other purposes. Although there is 35 usually some cross-subsidization of other public activities by enterprise activities, they typically have separate accounting and funding systems. The cross-subsidies that do occur take two forms: (1) in-lieu franchise fees and (2) direct budgetary transfers. The first type of subsidy is reported as a tax revenue by the parent entities. Because of limitations in the state’s reporting system for enterprise revenues, the second type of subsidy remains largely unreported.9 For this reason, many advocate the inclusion of public enterprise revenues in the totals as a way to ensure their inclusion. However, these revenues are relatively small compared to the overall revenues generated by enterprise activities, and the inclusion of all enterprise revenues as part of the public revenue burden would dramatically overstate the size of this cross-subsidy. It is better to exclude these enterprise revenues and slightly underestimate public revenues than to include all enterprise revenues and significantly overestimate the revenue burden. This leads to the second issue—their universality. People in nearly all localities have electricity, water, and sewer services. However, not all of these services are provided by governments. In many localities, they are provided by private companies. Still other communities, such as Sacramento, are served by independent utility districts. As a result, some of these revenues are reported as public revenues and others are not. The lack of comparability associated with these revenues, coupled with their cost-based nature, leads to the conclusion that they, too, should be ____________ 9Note that this was not always the case. In the 1960s and early 1970s, these subsidies of local governments were reported separately by the California State Controller’s Office. 36 excluded from our estimate of the state and local revenue burden in California.10 Should We Include General Service Revenues? General service revenues are derived from government services and activities that are provided in direct competition with the private sector. The largest of these revenues are university fees, university research revenues, and public hospital revenues. These revenues also include a variety of rents as well as revenues from the sale of assets, such as real estate. Why would we want to consider excluding these? These revenues represent areas where the choice to purchase the good from the state or local government is entirely discretionary. One can attend a private university, obtain research services from a private firm, seek out a private hospital, or even rent space from a private landlord. Unlike many other revenues that arise as a consequence of the police or monopoly power of the government, these are revenues for which there are private sector substitutes and that arise from public sector competition with the private sector. For this reason, it can be argued that these revenues should also be excluded, although the arguments are somewhat weaker than those for excluding intergovernmental and enterprise revenues. For the purposes of our estimation, we choose to omit these revenues from the state and local revenue total.11 ____________ 10We recognize that some would disagree with the exclusion of these revenues from our estimate of the public revenue burden. Consequently, we have included tables summarizing our main findings, but including these revenues, in Appendix C. 11As before with enterprise revenues, some would disagree with the exclusion of these revenues from our estimate of the public revenue burden. We have therefore presented our main findings with these revenues included in Appendix C. 37 Should We Exclude Any Other Revenues? Having excluded these three types of revenues, we ask if there are any other candidates for exclusion among the remaining revenue types? We must again turn to our main research question, What has happened to the public revenue burden since Proposition 13? What other types of revenues are largely discretionary on the payee side or relate to specific business-like activities? Certainly taxes, assessments, regulatory fees, and fines and forfeitures do not fit the discretionary category. In fact, all of these categories are rooted directly in the police power of the state. Nonenterprise service revenues resulting from a service provided exclusively by the state or local government, such as election services and tax collection services, also do not fit this profile. The government exercises the right granted to it by its citizens to exercise control over these revenues. The only remaining major revenue stream for which we have not yet accounted is interest. We could have easily made a case for excluding it, but we have chosen to include it. Our reasoning is that the government is not generally in the business of raising monies just to receive interest revenues and that these revenues are usually a by-product of other activities that it undertakes, such as financing major construction activities. How Large Are the Excluded Revenue Categories? Table 2.3 shows the revenues we exclude from our estimate of the public revenue burden in California. These totals constitute nearly half of all state and local revenues.12 It should also be noted that approximately two-thirds of the revenues in the general service revenues ____________ 12The amounts are provided here for readers wishing to include them. 38 Table 2.3 Revenues Excluded from Our Estimate of Public Revenues in California for All Levels of Government, by Revenue Type Revenue Type Intergovernmental revenues Enterprise revenues General service revenues 1978 21,694,128,405 4,476,246,475 1,683,879,855 1981 32,456,331,967 7,733,753,254 3,133,030,720 1988 49,489,137,169 13,714,929,929 5,567,580,382 1992 73,373,444,285 18,857,594,195 7,179,561,943 1995 81,478,340,742 22,061,142,572 8,152,364,018 39 Total 27,854,254,735 43,323,115,941 68,771,647,480 99,410,600,423 111,691,847,332 Percentage of total revenues 47.9 55.9 51.6 52.9 54.6 category are from the state’s public postsecondary education institutions and reflect a combination of student fees and external research revenues, further strengthening our argument for excluding them from our calculation. After making these adjustments, we are left with our estimated totals of state and local revenues in California. These amounts are provided in Table 2.4.13 Perhaps the most dramatic feature of this table is the large proportion of the overall state and local revenue pie that comes from taxes—more than 80 percent—and its persistence over time, even in the face of Proposition 13. Furthermore, even though assessments account for only a small portion of overall state and local revenues (only 0.7 percent in 1995), their growth is striking—rising by more than 1,800 percent over the 17 years in our study. It is also interesting to note the increase in regulatory fees in recent years and the increase in service revenues associated with activities where the government is the exclusive provider. One crucial question remains to be answered: How accurate are the data underlying the findings in Table 2.4? Before undertaking the present study, we addressed this issue (Shires, 1997). We found that the overall revenue totals reported in the California State Controller’s series Annual Report on Financial Transactions were quite accurate. We did ____________ 13Note that we do not call these revenues “own-source revenues” for California state and local governments. Own-source revenues, a measure often used in the state and local revenue burden debate, are revenues that a government generates from its own activities. Such estimates typically exclude intergovernmental revenues and enterprise revenues but include what we call general service revenues. The own-source distinction would be most comparable to revenues presented in Table C.1 in Appendix C. We believe that including general service revenues in such a definition of the public revenue burden is problematic, however, as detailed earlier in this chapter. 40 Table 2.4 Remaining Public Revenues in California for All Levels of Government, and Their Percentage Share, by Revenue Type Revenue Type Taxes Assessments Regulatory fees and charges Fines and forfeitures Interest Exclusive provider Other 1978 26,409,666,294 87.3 32,850,839 0.1 1,390,237,527 4.6 252,254,534 0.8 856,891,141 2.8 201,040,942 0.7 1,133,295,632 3.7 1981 28,392,188,943 83.1 162,711,104 0.5 1,338,110,598 3.9 362,918,091 1.1 1,838,620,532 5.4 287,272,242 0.8 1,769,296,927 5.2 1988 52,504,319,843 81.4 422,836,209 0.7 3,228,228,698 5.0 953,171,700 1.5 3,366,566,608 5.2 1,095,613,493 1.7 2,906,771,649 4.5 1992 72,070,935,103 81.6 560,079,307 0.6 4,101,351,362 4.6 988,966,277 1.1 3,871,848,963 4.4 1,182,836,525 1.3 5,661,550,731 6.4 1995 75,540,377,001 81.3 634,712,424 0.7 5,497,714,659 5.9 1,080,762,038 1.2 3,751,894,129 4.0 1,468,815,789 1.6 4,918,470,192 5.3 Total 30,276,236,909 34,151,118,437 64,477,508,200 88,437,568,268 92,892,746,232 100.0 100.0 100.0 100.0 100.0 find some specific reporting variations within the specific revenue categories that could affect the findings reported here, however. Specifically, we found that city and county intergovernmental revenues appeared to be underreported in the State Controller’s report.14 Because the overall totals were consistently quite accurate, we surmised that these underreported amounts were simply miscategorized in the final reports. Such a miscategorization would bias our reported revenue burden upward—that is, it would overreport public revenues—because we did not deduct enough revenues when removing intergovernmental revenues. A similar finding was made in the opposite direction for current services—a category that includes public enterprise revenues and the general service revenues listed above.15 These underreportings and overreportings appear to roughly offset each other and, in any case, are minor relative to the overall magnitude of the revenues involved in this analysis. As a result, we believe that the data we use here are of high quality. To conclude, Table 2.4 presents what we believe should be included in public policy discussion about the size of the revenue burden imposed by state and local governments in California. Although the total size of government under this estimate is significantly smaller than we previously thought, it is important to remember that the $205 billion total presented in Table 2.2 includes many revenues that were counted twice, as well as significant revenues that arose from enterprise activities. ____________ 14The study, which used 1992 data, found that city intergovernmental revenues were underreported by about 15.2 percent and county intergovernmental revenues by about 0.1 percent. Much of this underreporting was linked with federal housing and transportation subsidies. 15These revenues were overreported, most likely due to classification errors, by 4 percent in counties and 3 percent in cities. 42 Nonetheless, the $93 billion generated by state and local governments represents a lot of revenue. The next chapter, which discusses how best to measure changes in the public revenue burden in California, will place these revenues in context. 43 3. Measuring Changes in the State and Local Revenue Burden in California As we discussed in Chapter 1, there are two parts to our research question. Chapter 2 answered the question, What constitutes the public revenue burden? Now we turn to the second part of the question, How do we measure the changes that have occurred? In this chapter, we will discuss the different ways that the revenue burden is measured and the advantages and shortcomings of each. To do this, we will generate estimates of the revenue burden over time using two dominant methodologies and discuss briefly the meaning and limitations of each— a discussion that will be enhanced and expanded in the final chapter of this study. Implicit in our main policy question is the further question, How has the revenue burden changed “relative to what?” The obvious answer is relative to what residents paid before, or relative to time. However, measuring the revenue burden over time introduces a significant level of 45 complexity to the problem because California has changed dramatically over the past 20 years. Most notably, the state’s population has grown 41 percent, by more than nine million people. Inflation has driven overall consumer prices up nearly 150 percent, and the overall income of residents—measured by annual personal income—has more than tripled, growing by more than one-half trillion dollars. If we simply ask the question, Are we paying more to the state and local government today than yesterday, the answer is an unequivocal Yes! As Table 2.4 showed, the public revenue burden rose from $30 billion in 1978 to $93 billion in 1995—more than tripling! But this finding does not take into consideration the profound changes the state has undergone over the past 20 years. We need to consider the public revenue burden relative to the many changes that have reshaped the state and its fiscal landscape, especially its economy and its population. Because of the importance of income and population to the state’s fiscal options, the two most common approaches to measuring the size of the public revenue burden focus on gauging it relative to these two elements. Measures that consider revenues relative to income ask the question, How much of our income are we paying?—much the same as asking, How much of my salary do I pay on rent? Measures that focus on population ask the question, How much do people pay on average?— similar to asking, How much was the average rent in my building this month compared to last month? We discuss each of these measures below. 46 Income-Based Measures of the Public Revenue Burden The rationale for using income-based measures is that they measure an individual’s or society’s ability to pay for services. For example, paying $1,000 a month for rent is a much more difficult proposition for someone whose annual income is $15,000 than for someone whose income is $100,000. Similarly, citizens with low incomes would have a much smaller ability to pay their state a given amount of taxes than citizens with higher incomes.1 And, just as an individual’s income—and hence his ability to pay a given level of rent—changes over time, so the income of a state’s citizens changes over time, which affects its population’s ability to pay a given level of revenues. California’s per capita personal income, for example, rose from $8,951 in 1978 to $23,279 in 1995.2 When we consider the statewide income of California, the cleanest measure available is personal income. This measure, which is produced by the Regional Economics Information System, Bureau of Economic Analysis, United States Department of Commerce, measures the overall income of individuals in the state. Other possible candidates for measuring the wealth of Californians include the Gross State Product (GSP), which corresponds to the most commonly used measure of national income—the Gross Domestic Product (GDP), and total ____________ 1Also implicit in this discussion is the question of who is paying the taxes. For example, a wealthy state’s average taxes might be lower than another state’s, but that burden might be disproportionately concentrated in either higher- or lower-income taxpayers. The issues of taxpayer equity that this concentration introduces, however, are beyond the scope of this report. 2These amounts are calculated from California Department of Finance reports of California personal income and population. 47 household income. Because GSP and state personal income are closely related, the GSP measure should produce results that closely parallel those we present below using personal income. Since the measurement of GSP, however, is not as precise as the measurement of state personal income,3 we will use personal income in our analysis. We do not use household income because it fails to include the business-sector portion of the state’s income capacity. To provide an income-based measure of the public revenue burden, we simply divide the public revenue burden by the measure of wealth— in this case, personal income. We have done this in Table 3.1. As this table shows, the public revenue burden has varied significantly over the 17 years in our study. In 1978, when the voters approved Proposition 13, the public paid 15 percent of its income to state and local governments. By 1981, state and local governments’ share of personal income fell dramatically to 11.4 percent. Although a significant portion of this decline is probably attributable to Proposition 13, a statewide recession and significant state-level tax cuts also directly contributed. During the 1980s and early 1990s, however, the public revenue burden grew, rising to more than 13 percent of total personal income in 1992 when, in response to the recession, state and local policymakers increased revenues through fees and tax increases.4 The public revenue burden has since declined to approximately the same level it was in 1988, dropping to 12.5 percent of personal income. ____________ 3Estimates of GSP are typically generated by roughly allocating the national GDP estimates across 63 industries to states. State personal income, on the other hand, is measured by compiling the relevant income information for each state. 4Because we have included only five years in our analysis, it is not possible to ascertain when the public revenue burden peaked as a share of personal income. 48 Table 3.1 Public Revenues in California as a Share of Personal Income for All Levels of State and Local Government 49 Revenue Type Public revenues California personal income Percentage of personal income 1978 30,284,118,492 202,241,530,500 15.0 1981 34,151,118,437 298,482,837,000 11.4 1988 64,477,508,200 527,589,456,000 12.2 1992 88,437,568,268 671,172,174,000 13.2 1995 92,892,746,232 743,218,326,000 12.5 If one believes that Proposition 13 was exclusively about tax reform, and that increases in other revenues such as interest, fines, and user fees are not critical, we could take a similar look at only tax revenues in California. In Table 3.2, we see how the tax burden has changed during this same period, looking at taxes as a share of state personal income. Since taxes are the main component of state and local revenues, we find that the tax trends correspond closely to those of the public revenue burden overall. There is a significant decrease in tax revenues as a share of personal income in the period immediately after Proposition 13, again attributable not only to Proposition 13 but also to statewide tax reforms and a statewide recession. This decrease is then followed by a gradual increase through the 1980s and early 1990s, with a final decrease back to approximately 1988 levels by 1995. This measure corresponds most closely to that used by Gold in his analysis, which we discussed in Chapter 1, and our results are consistent with his. One might, however, want to take a broader view of the public revenue burden in California. Some participants in the debate over the size of state and local government argue that governments have found creative ways around the constraints imposed by Proposition 13, primarily through increases in regulatory fees and assessments. To illuminate this issue, we present a different measure of the revenue burden in Table 3.3, which includes three revenue streams in question: taxes, assessments, and regulatory fees. This measure examines the whole family of revenues that are taxes by nature, if not in their specific 50 Table 3.2 Tax Revenues in California as a Share of Personal Income for All Levels of State and Local Government Revenue Type 1978 1981 1988 1992 1995 Tax revenues 26,409,666,294 28,392,188,943 52,504,319,843 71,586,395,964 75,540,377,001 Percentage of personal income 13.1 9.5 10.0 10.7 10.2 51 Table 3.3 TAR Revenues in California as a Share of Personal Income for All Levels of State and Local Government Revenue Type 1978 1981 1988 TAR revenues 27,832,754,660 29,893,010,645 56,155,384,750 Percentage of personal income 13.8 10.0 10.6 1992 76,732,365,772 11.4 1995 81,672,804,084 11.0 implementation.5 For our discussion, we will call this group of revenues TAR revenues (taxes, assessments, and regulatory fees). The findings are similar to those presented in Tables 3.1 and 3.2 because TAR revenues account for a significant portion of overall public revenues (82 percent of public revenues in 1995). One interesting difference, however, is that TAR revenues do appear to fall proportionately less than tax revenues between 1992 and 1995, indicating the persistence of assessments and regulatory fees as a mechanism for financing government in this period. Population-Based Measures of the Public Revenue Burden Population-based measures of changes in the public revenue burden tell us how the revenue burden has changed for the average citizen over time. It tells us how much the average person gives to state and local government, not as a share of income, but in strict dollar terms. Proponents of this measure would argue that it provides a real estimate of the “price of government.” They would also argue that demands for public services are driven more by the number of people using the services than by changes in the population’s income. Hence, populationbased measures would provide a fairer representation of the resources available to government. These population-based measures, also described as per capita measures, are customarily generated by dividing public revenues by the ____________ 5In Chapter 2, we made distinctions between general taxes (taxes on broad and general activities), assessments (taxes levied to fund specific activities and purposes and applied to a limited group of taxpayers), and regulatory fees and assessments (charges for participating in or undertaking specific voluntary activities). In this measure, we are generally including all monies levied by the government (the broadest definition of a tax). 52 state population in a given year.6 We have done this in Table 3.4 for the three revenue streams introduced in Tables 3.1, 3.2, and 3.3: public revenues, tax revenues, and TAR revenues. This table describes the total tax revenues per person in the state. Since this estimate also includes children, typical taxpayers actually see a multiple of this number, as they pay the taxes for their entire household. As we see here, the taxpayer sees an increasing amount each year going to state and local governments as all three revenue series increase over the 17 years in our study—even between 1978 and 1981 when the implementation of Proposition 13 coincided with state-level tax reforms and a statewide recession. This table explains, at least in part, the ongoing series of initiatives designed to “reel in” the expansion of government. Taxpayers and voters have a sense that government is evergrowing. Although this table may accurately characterize what the average Californian sees from year to year, it does not necessarily provide a good population-based measure of the revenue burden in California. This is Table 3.4 Per Capita Public Revenues, Tax Revenues, and TAR Revenues in California for All Levels of State and Local Government (Current Dollars) Revenue Type Per capita public revenues Per capita tax revenues Per capita TAR revenues 1978 1,340 1,169 1,232 1981 1,421 1,182 1,244 1988 2,298 1,871 2,002 1992 2,864 2,334 2,485 1995 2,910 2,366 2,558 ____________ 6For this analysis, we have generated an estimate of the population in a given fiscal year by averaging the California Department of Finance’s July 1 estimates of the state’s population for the beginning of the fiscal year and for the following year. Hence, our fiscal year 1978 population estimate is an average of the Department of Finance’s estimates for July 1, 1977, and July 1, 1978. 53 because it fails to take into account inflation, the year-to-year increase in prices that both individuals and governments encounter when purchasing the resources necessary to produce their services. Food, rents, gas, and other commodities increase in price over time. Governments and businesses must pay higher salaries over time to allow their employees to keep up with these rising prices. This gradual increase in costs means that a dollar today does not buy nearly as much as a dollar did 17 years ago. If we examine the public revenue burden over time without taking into account the reduced buying power of more current dollars, then we will overestimate the amount of services that governments can provide in a given year. The most common way to account for inflation is to use price indices, which are tracked by the federal government, to adjust prior-year dollars for changes in the cost of living. A price index allows one to convert the cost of something in a former year into what it would cost this year. For example, suppose a loaf of bread cost 40 cents in 1978 and $1.00 today. We could describe the price of a loaf of bread either in 1978 dollars (40 cents) or in today’s dollars ($1.00). If we assumed that the cost of other things grew at approximately the same rate, then a dollar’s worth of goods in 1978 would be worth approximately $2.50 today and, alternatively, $2.50 in today’s dollars would buy goods today that would have cost $1.00 in 1978. When one makes these corrections for the effects of inflation, the resulting revenues are called real revenues and the amounts are reported in constant 19xx dollars, where 19xx is the year into whose dollars the other year’s dollar amounts are converted. The use of these price indices allows us to make year-to-year comparisons of the value of money. One widely used index is the 54 Consumer Price Index (CPI).7 The federal government conducts annual surveys of prices and prepares estimates of the annual change in prices. Conveniently for us, the federal government even computes the changes in California prices separately. We use this version of the CPI to adjust the per capita burden amounts in Table 3.3 for inflation converting all the other years into equivalent 1995 dollars.8 The results are shown in Table 3.5.9 As we can see from this table, the earlier years’ revenues increased, reflecting the increased buying power of a dollar in the earlier years. The trends we see in this table are quite different from those we saw in Table 3.4 and actually correspond quite closely to those we saw in Tables 3.2 and 3.3 when we were measuring the public revenue burden as a share of personal income. The real per capita revenue burden declines significantly between 1978 and 1981 and then rises over the 1980s and early 1990s before falling back to 1988 levels in 1995. ____________ 7Other price indices are possible candidates for use here. One strong candidate would be the national CPI deflator for government goods and services. Using this deflator introduces a conceptual problem, however. Inasmuch as the availability of public funds defines the resources available to purchase services (more resources often lead to larger increases in expenditures for the same units), it becomes problematic to use the amounts actually paid for those services to discount their costs. Because of this concern, we have chosen to use the broader inflation measure, the CPI. 8This conversion was done by dividing each year’s index by the fiscal year 1995 index. Note that the resulting dollars are not precisely the same as 1995 dollars because a true recalibration of the index would require reweighting the bundle of goods included in the index by their 1995 proportions instead of those already in use in the index (1982–84 in this case). Because the federal government does not provide annual recalibrations of these amounts, this approach is a close approximation and it is commonly used in public policy analyses. 9It is important to note that the use of another index may produce different findings than we show here. However, it would take a 13.5 percent increase in the total price change over the period to reverse the direction of the change in real per capita public revenues from a decrease to an increase over this time. This level of error in the price index is possible, but unlikely. 55 Table 3.5 Real per Capita Public Revenues, Tax Revenues, and TAR Revenues in California for All Levels of State and Local Government (Constant 1995 Dollars) Revenue Type Real per capita public revenues Real per capita tax revenues Real per capita TAR revenues 1978 3,305 2,882 3,037 1981 2,498 2,077 2,187 1988 2,944 2,397 2,564 1992 3,057 2,492 2,653 1995 2,910 2,366 2,558 We have uncovered an interesting story in this chapter. Although the amounts that taxpayers pay to government appear to have risen dramatically over the 17 post-Proposition 13 years in this study, a different picture develops when income and inflation are factored into the equation. In fact, both common measures of the magnitude of the public revenue burden, both income- and population-based, tell a remarkably similar story. Revenues seem to decline dramatically after the implementation of Proposition 13 and then rise gradually through the 1980s. For the years we studied, they peak in 1992 and, by 1995, fall back to levels comparable to those in 1988. In our final chapter, we will discuss in more detail the implications of these patterns for public policy and the policy debate surrounding the size of the public revenue burden in California. 56 4. Our Findings The goal of this study was to present a comprehensive and informative overview of the public revenue burden in California and to provide a definitive answer on how it has changed over time. We believe that the definition of the public revenue burden we presented in Chapter 2 and the results we presented in Chapter 3 from the application of that definition have accomplished both of these goals. What We Have Found Figure 4.1 shows how the public revenue burden changed over time. It fell dramatically in the years after Proposition 13 (to 76 percent of preProposition 13 levels) and then rose over the 1980s. By 1992, the public revenue burden had returned to about 90 percent of pre-Proposition 13 levels, and it then declined to about 85 percent of pre-Proposition 13 levels in 1995. 57 Percentage 100 90 80 70 60 10 0 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 Figure 4.1—Composite Measure of Overall Public Revenues in California The trendlines are similar for both income-based and population- based measures. The composite measure used in Figure 4.1 is simply an average of the two series relative to their values in 1978. The actual changes in the two measures—revenue burden as a percentage of personal income and real per capita revenues—track quite closely, as can be seen in Figure 4.2. Although the real per capita revenue burden rises a bit faster than the revenue burden as a share of personal income in the 1980s, the changes relative to their levels in 1978 are almost identical. Both measures seem to tell the same story. Proposition 13 contributed significantly, in combination with other policy and economic factors, to a 24 percent decrease in the size of state and local 58 Percent of personal income Real per capita income (dollars) 16 3,500 14 Real per capita revenue burden 3,000 12 Percentage of 2,500 personal income 10 2,000 8 1,500 6 4 1,000 2 500 00 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 Figure 4.2—Two Measures of Public Revenues in California government1—to the delight of some and the chagrin of others. During the 1980s, however, the losses/gains were eroded as the size of government grew by 12 percent between 1981 and 1988. The growth in the size of government continued until, and was likely accelerated by, the recession of the early 1990s. Although we do not know for certain when the size of post-Proposition 13 government peaked,2 it reached its highest point in our series in 1992 at 90 percent of its pre-Proposition 13 level—an increase of 19 percent since 1981. By 1995, the size of state ____________ 1Note that not all of this decrease can be attributed to Proposition 13. The combination of state tax reductions and a statewide recession also contributed to the decrease in state and local revenues in 1981. 2Since we analyzed only five years in this study, we have no way of knowing whether 1992 was the peak. This was the first year of the recession and was a year marked by significant expansions of several types of state revenues to make up for the losses in revenues generated by the recession. Other scholars’ data indicate that the peak may have occurred earlier, in 1990. 59 and local government had returned to approximately the same size it was in 1988. We can also look at the changing revenue burden in terms of the two measures described above. Measured as a share of personal income, the public revenue burden declined 17 percent from 1978 to 1995 but rose 9 percent from 1981 to 1995. Similarly, real per capita public revenues declined 12 percent from 1978 to 1995 but rose 16 percent from 1981 to 1995. The average annual change in the public revenue burden was –8.6 percent from 1978 to 1981 and then averaged +1.3 percent from 1981 to 1992 before declining 1.8 percent a year from 1992 to 1995. Some would argue that the changes we describe above are directly attributable to changes in the economic and business cycles in California—that is to say that they are simply the result of the changing fortunes in our economy—and as state personal income grew during the 1980s the state’s progressive tax system stepped in to produce the growth in the public revenue burden noted in our study.3 However, as we noted in Chapter 1, we chose only five years in our study to address just this issue. We refer the reader to Appendix B for the full discussion, but the general patterns that we have identified over the past two decades hold, even when we compare across comparable points in the business cycle. In November 1996, voters chose to significantly expand the constraints on their government, passing Proposition 218. If revenue burden levels had declined to prerecessionary levels, how then do our ____________ 3Note that the data underlying our analysis indicate that this is not the case. If we were to report only the change in the portion of the revenue burden raised by state government, we find that it remains flat at 7.1 percent of personal income between 1981 and 1988 and then rises to 7.7 percent of personal income in 1992 before returning to 7.2 percent in 1995. Most of the growth over this entire period occurs in cities, counties, and school districts. 60 numbers explain this occurrence? There are many likely explanations. However, a contributing factor was probably the 12 percent growth in state and local government revenues that we identified during the 1980s. As we discuss in Chapter 2, much of the growth in this period was in the form of assessments, increased local taxes, increased regulatory fees, and increased service charges. These high-profile revenues would be likely to create a direct response in the voting public. An additional issue that our study has brought to light is shown in Figure 4.3. This figure shows the per capita public revenue burden in both current and real terms. As we can see from this diagram, the average resident in the state sees an increasing number of his dollars going to state and local government. Dollars 3,500 3,000 2,500 Current 2,000 1,500 1,000 Real 500 0 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 Figure 4.3—Per Capita Public Revenues, in Real (1978) and Current Dollars 61 Even though real revenues4 declined slightly over this period, public revenues in noninflation adjusted dollars have risen continuously and dramatically, even in the period immediately following Proposition 13. As we discussed in Chapter 3, taxpayers and voters think in terms of current dollars. For example, we think about how our property tax bill went up 2 percent this year. We do not usually consider the fact that, although our property tax bill went up 2 percent, our wages went up 3 percent and the price of gasoline or bread may have risen by 3 percent as well. Thus, although real per capita public revenues are certainly a legitimate measure to use in the policy debate, their message is often lost on voters. The bottom line of our study is that, after removing the appropriate revenue types,5 both measures of the changes in the public revenue burden seem to tell the same story, and that story lends support to both of the major camps in the public policy debate. The fact is that Proposition 13 did immediately contribute to a significant reduction in the revenues available to state and local governments in California. It is also true that these public revenues have never returned to pre-Proposition 13 levels. It is also true, however, that state and local governments have been able to significantly mitigate the full effects of Proposition 13 and that ____________ 4To make it easier to follow, we have used 1978 dollars in this figure. 5Some will find it controversial to remove the three types of revenues that we have eliminated (intergovernmental, competitively provided service revenues, and public enterprise revenues), in large part because this is not the way it has been historically done. We believe, however, that the arguments for their removal are strong and robust and allow for a more accurate understanding of what has happened to the public revenue burden in California over the past two decades. 62 the public revenue burden has risen substantially and progressively since Proposition 13’s implementation in the late 1970s and early 1980s.6 What Our Findings Suggest to Current and Future Policy in California Our findings have several implications for public policy in California. First, the pervasiveness of the study’s findings across measures and time shows that how to measure the changing revenue burden should not be the focal point of the debate—our analysis showed similar results with both measures. Furthermore, even if alternative measures of the revenue burden are used, as in Appendix C, the findings are consistent. We hope that this finding will clarify the confusion introduced by several previous studies on the topic (see Table 1.3) and allow the debate to focus instead on the critical issue of the appropriate size of state and local governments. Beyond this direct contribution, we believe that this study speaks to the future prospects for the public sector in California. Perhaps in response to the growth in the size of the state and local sectors in the years (in absolute terms) since the implementation of Proposition 13 (as shown in Figure 4.3), the voters of California have placed significant additional constraints on the public sector through the passage of Proposition 218. The constraints imposed by this proposition will significantly increase the hurdles that local governments will have to clear to expand their revenues through several of the mechanisms that were the mainstays of revenue growth during the late 1980s and early 1990s. ____________ 6We should also note that, as other PPIC research has shown, state and local governments can transfer the costs of government to private citizens and taxpayers in other ways (see, for example, Dresch and Sheffrin, 1997). Unfortunately, because of data limitations, our study could not fully incorporate those costs into this analysis. 63 Interestingly enough, this initiative seems to be more of a response to “general government growth” than to specific local needs, because voters seem to be passing many of the measures that have been brought to the ballot as a result of Proposition 218. In light of the requirements imposed by Proposition 218 and their attendant logistical, political, and fiscal costs, it is likely that the growth of new local revenue streams will slow in the future. In light of the flat rate of growth in real revenues identified in this study, it follows that the resources to fund expansions in the level of services provided at the state and local levels will grow, at best, slowly. There does seem to be some hope for specific programs and initiatives, as voters have in recent years been much more receptive to local bonds and finance measures. Because a large share of state and local revenues is derived from taxes, it is unlikely that state and local governments will experience funding shortfalls during periods of high economic growth, when tax coffers swell. In recessionary periods, however, the ability to raise additional revenues through increased license fees, service charges, and user fees— state and local governments’ response during the last recession—will be constrained by both Proposition 218 and their extensive use during the last recession. This combination could leave California’s state and local budgets sensitive to economic shocks and could result in reductions in public support for discretionary programs, such as higher education, with the onset of a future recession. This study has shown that Proposition 13 has contributed to a significant rollback of the public revenue burden. The effects of this rollback continue today, but its full effect has been partially offset by growth in the public revenue burden in the intervening years. It is 64 possible and even probable, however, that Proposition 218 will limit state and local governments’ ability to continue this growth and that the future size of the public revenue burden will remain at its current levels into the future. 65 Appendix A Detailed Comparisons with Other Studies In the introduction to this report, we briefly discuss four studies that examined the magnitude of the state and local revenue burden in California. Before beginning our study, we felt we needed to understand precisely what each of these scholars and analysts did in theirs and how their assumptions and choices affected their findings. We examined each study in detail and sought to replicate their findings with our data. Only in this way could we be certain that any findings of ours that differed from theirs were truly the result of differences in our assumptions and not data-driven. We were able to replicate each study’s revenue burden estimates fairly closely. The most variation occurred between our numbers and Gold’s 1977–78 estimates. Our estimate was about 5 percent lower than his. We discuss below our replication of each study’s estimates. 67 Steven Gold’s Estimates Steven Gold’s revenue burden estimates were the lowest among the studies, because he counted only “tax revenue.” He gave no source for his revenue numbers, so we used State Controller data to recreate his figures. For 1990–91, we limited the categories we included in the revenue burden to the following “tax” categories from the State Controller data: taxes, special benefit assessments, and licenses and permits for cities and counties; property tax revenue for schools; taxes and assessments for nonenterprise special districts and redevelopment agencies; and local transportation funds and other locally funded sales tax for transportation planning agencies. For enterprise special districts, the categories were assessments, county allocation, special district augmentation fund, property tax, sales tax, and debt service taxes. State revenues for our estimates included major taxes and licenses, regulatory taxes and licenses, revenue from local agencies, and services to the public. We tried to capture the same sources of revenue in the 1977–78 revenue burden estimate as we did in the 1990–91 figure. Although there were fewer categories delineated in the State Controller’s books in 1977–78, it is reasonable to assume that revenue sources from most of the newer categories were included previously in one of the other broader categories. In 1977–78, enterprise special districts had only two relevant categories—taxes, and taxes and assessments—instead of six; city and county classifications were the same as in 1990–91 with the exception of the lack of the special benefit assessment category. The categories for nonenterprise special districts and schools were unchanged. There were no transportation planning agency revenues recorded in 1977–78. In addition to major taxes and licenses, state revenues for 1977–78 included those from the “other revenues” category, with the exception of interest 68 on investments, oil and gas revenues, penalties and interest on unemployment contributions, proceeds under unclaimed property tax, and sales of state property. As shown in Table A.1, both our estimate of the 1990–91 state revenue burden and Gold’s are the same. Local estimates for that year differ by less than 2 percent, or approximately $300 million ($25.2 billion compared to $24.9 billion). In 1977–78, our estimate is below Gold’s by about 5 percent, or $1.4 billion ($27.8 billion compared to $29.2 billion). Table A.1 Comparison of Estimated Revenues Using Steven Gold’s Methodology, as a Percentage of Personal Income Study Gold Our study Difference 1977–78 Total 14.62 13.92 0.70 1990–91 Local 3.97 4.02 –0.05 1990–91 State 7.28 7.28 0.00 California Taxpayers’ Association’s Estimates The California Taxpayers’ Association (Cal-Tax) cites the Bureau of the Census as the source for its revenue figures. Thus, to replicate its numbers, it was necessary to choose a year in which the Census published government financial transaction data. Since detailed Census data were not published in both 1977–78 and 1990–91,1 the most recent year that ____________ 1The Bureau of the Census publishes estimates of state and local revenues, by type of government for various revenue categories each year, but the actual detailed information necessary for us to reconcile our estimates to those of Cal-Tax is available only in the specific full Census years that occur in years ending in 7 and 2, hence 1986–87 and 1991–92. Note that the 1991–92 data were not available in time for us to include that year in our reconciliation. 69 Cal-Tax published the revenue burden, we used the most current Census year in which the Cal-Tax estimate coincided with the detailed Census publications—1986–87—to validate our methodology. For our calculations, own-source revenues were derived by subtracting intergovernmental revenue from total general revenue. The resulting estimates exclude utility revenues (water supply, electric power, gas supply, and transit) and employee retirement revenue. Cal-Tax’s own-source revenue estimates for cities, counties, and special districts in 1986–87 equaled our figures from the Census data. Differences between the sets of numbers within the two remaining classifications were minimal. Our own-source figure for the state was $43.25 billion whereas Cal-Tax’s was $43.19 billion—a $60 million difference. The school revenue calculations differed by $40 million, with our total as $5.22 billion and Cal-Tax’s as $5.26 billion. Since the discrepancies are tiny, as shown in Table A.2, the overall revenue burden estimate is the same for us and Cal-Tax at 15.71 percent. As discussed in Chapter 1, there are some significant methodological issues associated with using the Bureau of the Census’ reported revenue amounts. We will present a detailed discussion of these differences in a future study. Table A.2 Comparison of Estimated Revenues Using the Cal-Tax Methodology, 1986–87 (Public Revenues in Billions of Dollars) Study California Taxpayers’ Association Our study Local Revenues 31.45 31.49 State Revenues 43.25 43.19 Total Revenues 74.70 74.68 Difference –0.04 0.06 0.02 70 Legislative Analyst’s Office’s Estimates The Legislative Analyst’s Office document containing the revenue burden estimate indicates that theirs was obtained directly from the U.S. Department of Commerce. The replication approach was quite similar to the one we pursued with the California Taxpayers’ Association estimate above. Using 1991–92 Census numbers for own-source local and state revenues and personal income, we calculated the revenue burden for 1991–92 to be 16.66 percent. The variation between the two revenue estimates is negligible. It is possible that the difference is due to different personal income numbers in the denominator, because such numbers may have been revised since these original estimates were made. John Kirlin et al. Estimates John Kirlin et al. provide the most detailed breakdown of the revenues that constitute their revenue burden estimates. Since the State Controller data are listed as the source for their revenue burden estimates, we were able to evaluate their numbers in each category. A description of our evaluation for each category is given in Table A.3. As shown in the table, we were able to determine how Kirlin and his colleagues calculated most of their revenue numbers. The primary differences between their estimates and ours center on transit revenues that were reported by the State Board of Equalization but were difficult to trace in the State Controller’s reports. We were unable to ascertain with certainty that these revenues were actually unreported and not miscategorized in the State Controller’s reports and we therefore did not add them to our estimates of the public revenue burden in California. Otherwise the two estimates correspond quite well. 71 Table A.3 Comparison of Estimated Revenues Using Kirlin et al.’s Methodology, 1986–87 (Public Revenues in Millions of Dollars) Study 1991–92: Kirlin et al. Our study Special School and Cities and Districts Transit Counties and RDAsa Districts State 54,172 16,134 54,164 16,133 30,736 30,358 61,887 61,986 Total 162,929 162,644 Difference 8 1 378 –99 288 1977–78: Kirlin et al. Our study 16,800 5,955 16,787 5,955 8,446 8,187 20,223 20,201 51,424 51,130 Difference 13 0 259 22 294 NOTES: Our replication of the 1991–92 revenue estimate for counties does not include other financing sources for San Francisco ($155 million) because we know that Kirlin et al. exclude these revenues. We also do not include in the replication revenues from current service charges for self-insurance districts ($973 million) in our 1991–92 nonenterprise special district total, because Kirlin et al. exclude them. The sum of the 1977–78 revenues for each of Kirlin et al.’s entities is $51,424, but they list it as $51,167, in part because they omitted transit sales revenues in their total sales tax figure. aRedevelopment agencies. 72 Appendix B The Years Included in This Report Because of the complications associated with the way special district revenues are reported in California, it was necessary to keypunch a considerable amount of data from the State Controller’s Annual Report on Financial Transactions Concerning Special Districts for each of the years we wished to study. As a result of this significant cost, we selected only five years for our analysis: 1977–78, 1980–81, 1987–88, 1991–92, and 1994–95. The first fiscal year was chosen because it was the year that Proposition 13 passed. 1991–92 was chosen because several other studies reported revenue burdens for this fiscal year. 1994–95 was chosen because it was the most recent fiscal year for which the data were available. The two remaining years, 1980–81 and 1987–88, were selected because they represented points in the business cycle that corresponded to 1991–92 and 1977–78, respectively, as shown in Figure B.1. 73 Percentage 6 5 4 3 2• 1 0 • • –1 •–2 –3 –4 1975 1977 1979 1981 1983 1985 1987 1989 • 1991 1993 1995 Figure B.1—Annual Growth in California Real per Capita Personal Income As the figure shows, the state was in the depths of recession in both 1981 and 1992, and 1978 and 1988 were comparable points in periods of economic growth. This aspect of comparison may be particularly important when using the income-oriented measures of the public revenue burden—such as percentage of personal income—which are closely associated with changes in the business cycle. 1981 has the additional benefit of coming soon after Proposition 13 was fully implemented1 and before a series of accounting changes were instituted by state and local governments.2 ____________ 1The three-year delay after 1978, when Proposition 13 was passed by the voters, is actually ideal timing. Proposition 13 was initially implemented in fiscal year 1979, but significant revisions were subsequently instituted in fiscal year 1980. As a result, 1981 actually represents the first year after the full transition to the post-Proposition 13 world. 2Fiscal year 1981 falls before a series of accounting changes that were implemented in 1982 and 1983. Although these changes will not affect the longer term because the year-to-year variation largely nets out, there were noticeable effects on reported revenues during the transition years. 74 The most recent year for which data were available was 1995, but the changes in the public revenue burden from 1992 to 1995 may provide some insights into the state of the world when Proposition 218 was passed by the voters in November 1996. To compare the findings across the business cycle, Figure B.1 plots the progression of the California business cycle over time by presenting the annual growth in real per capita personal income in California. To help us understand the changes over time, it is useful to compare our composite indices for two pairs of years, the first pair corresponding to plateaus after economic peaks (1978 and 1988), and the second pair to a bottoming out of recessions (1981 and 1992). In the first case—the two economic midpoint years 1978 and 1988—the findings are consistent with our overall conclusions. The overall public revenue burden, as measured by our composite index, declines from 100 to 85 percent of 1978 levels. Although we cannot reach any conclusions about changes over the intervening years by using these two points, the findings regarding the two recession years add weight to our overall conclusions. The findings in the second comparison are more compelling. In the comparison of the two recession years, we see an increase from 76 percent of the 1978 level to 90 percent—again pointing to our finding that the revenue burden increased over this time, but to a level less than where state and local government revenues were before Proposition 13. Perhaps the most interesting aspect of this finding is the seeming inconsistency of outcomes. If we assume that a recession has the same repressive effect on state tax revenues, then the increases in the 1992 recession are so much the more remarkable. The state’s fiscal status was quite different during the two recessions, however. During the recession 75 in 1981, not only was Proposition 13 implemented—creating a significant demand for state resources among local governments—but the state instituted several tax cuts. During the 1992 recession, the state had to institute significant tax increases to keep the state solvent. This difference in state action goes a long way toward explaining the seeming inconsistency of the two findings, although the broader story of a reduced, but growing public revenue burden remains intact. 76 Appendix C Revenue Burden Measures Under Alternative Definitions In this appendix, we will reproduce the findings presented in Chapter 3 for the public revenue burden under three alternative scenarios: (1) restoring general service revenues to our definition of public revenues, (2) restoring enterprise revenues, and (3) restoring both. We will present summary statistics for each of three measures for each scenario: (1) the public revenue burden as a share of personal income, (2) the revenue burden adjusted for changes in the population, measured by per capita revenues, and (3) the revenue burden adjusted changes in both the population and inflation, measured by real per capita revenues. Restoring General Services Restoring those public revenues for which the public sector has direct private sector competition produces the results shown in Table C.1. 77 Table C.1 Measures of California’s Public Revenues if General Services Are Included Measure Public revenues ($ billions) Percentage share of personal income Per capita public revenues ($) Real per capita public revenues (1995 $) 1978 31.97 15.8 1,415 1981 37.28 12.5 1,552 3,489 2,727 1988 70.05 13.3 2,497 3,198 1992 1995 95.62 101.04 14.2 13.6 3,097 3,165 3,306 3,165 This formulation most closely corresponds to the own-source definition often used for convenience in policy debates. Here, the story remains much the same as we found with our primary definition of the public revenue burden. Public revenues decline immediately after Proposition 13 and then regain much of the decline through the early 1990s before declining once again in the mid-1990s. In real per capita terms, the growth is slightly higher in this formulation than it is in our primary definition. Real per capita revenues rise to nearly 95 percent of pre-Proposition 13 levels by 1992 before dropping off to 90 percent in 1995. This is because state and local governments significantly increased their prices for products they produced in competition with the private sector during the recession.1 Restoring Enterprise Revenues Some would argue that, although we should not include general service revenues, we should include enterprise revenues that result from ____________ 1For example, student fees at the University of California rose dramatically from $2,486 per year in 1992 to $4,111 in 1995 (see CPEC, 1996, for further details). 78 policies and pricing strategies instituted by public organizations.2 Restoring public revenues that arise from enterprise activities produces the results shown in Table C.2. The general pattern of the story is still consistent with our main findings in this study, although the thresholds are much higher. Because of their dependence on service charges, enterprise revenues declined only to 83 percent of pre-Proposition 13 levels by 1981 and then rose back to 93 percent of those levels in 1992 before dropping back to 90 percent in 1990. This step-up in the levels of change over time tells a story about the increasing cost of services provided by these entities and further reinforces the validity of our decision to exclude them from our estimate of the public revenue burden in California. Table C.2 Measures of California’s Public Revenues if Enterprise Revenues Are Included Measure Public revenues (billions) Percentage share of personal income Per capita public revenues ($) Real per capita public revenues (1995 $) 1978 1981 1988 1992 1995 34.75 41.88 78.19 107.30 114.96 17.2 14.0 14.8 16.0 15.5 1,538 1,743 2,787 3,475 3,601 3,793 3,064 3,570 3,709 3,601 Restoring Both General Services and Enterprise Revenues Finally, Table C.3 presents the changes in public revenues if one were to include both those public revenues for which the public sector ____________ 2Note that under this definition, one may wish to include, at the very least, the higher education enterprise, which is included in the general services category. We do not do so here for the sake of consistency. 79 Table C.3 Measures of California’s Public Revenues if Both General Service Revenues and Enterprise Revenues Are Included Measure Public revenues ($ billions) Percentage share of personal income Per capita public revenues ($) Real per capita public revenues (1995 $) 1978 1981 1988 1992 36.44 45.02 83.76 114.47 18.0 15.1 15.9 17.1 1,613 1,873 2,986 3,708 1995 123.11 16.6 3,856 3,976 3,293 3,824 3,958 3,856 has direct private sector competition and those that arise from enterprise activities. As one would expect, this has the effect of significantly expanding the estimate of the revenue burden to the point that we are now 96 percent of the way back to pre-Proposition 13 levels, as measured by real per capita income. We believe, however, that this measure is too inclusive and disguises key distinctions in revenues that should not be included in estimates of the public revenue burden in California. 80 References Baldassare, Mark, When Government Fails: The Orange County Bankruptcy, Public Policy Institute of California, San Francisco, California, 1998. California Postsecondary Education Commission (CPEC), Fiscal Profiles 1996, Sacramento, California, 1996. California State Controller, Annual Report on Financial Transactions Concerning Cities, Sacramento, California, 1978, 1981, 1988, 1992, and 1995. California State Controller, Annual Report on Financial Transactions Concerning Counties, Sacramento, California, 1978, 1981, 1988, 1992, and 1995. California State Controller, Annual Report on Financial Transactions Concerning Special Districts, Sacramento, California, 1978, 1981, 1988, 1992, and 1995. California Taxpayers’ Association, “California State and Local Revenue Growth: Effects of Proposition 13 Are Almost Gone,” Cal-Tax News, April 1, 1994. Dresch, Marla, and Steven M. Sheffrin, Who Pays for Development Fees and Exactions? Public Policy Institute of California, San Francisco, California, 1997. 81 Dunstan, Roger, Local Government Revenue and Expenditures Since Proposition 13: A Historical Primer, California Research Bureau, report CRB-IS-93-006, Sacramento, California, August 1993. Gold, Steven D., “California’s Tax Burden from a National Perspective,” Symposium on State Taxation of Business Activities, University of California at Davis, March 26, 1993. Kirlin, John J., Jeffrey I. Chapman, Peter Asmus, and Roy Thompson, “Fiscal Reform in California,” in California Fiscal Reform: A Plan for Action, California Business—Higher Education Forum, Oakland, California, June 1994. Legislative Analyst’s Office, “State and Local Finance—Fiscal Overview,” Cal Guide: A Profile of State Programs and Finances, Sacramento, California, January 1995. Leigland, James, “The Census Bureau’s Role in Research on Special Districts: A Critique,” Western Political Quarterly, Vol. 43, No. 2, June 1990, pp. 367–380. Sheffrin, Steven, and Marla Dresch, Estimating the Tax Burden in California, Technical Assistance Program Report, California Policy Seminar, University of California at Berkeley, 1995. Shires, Michael A., Patterns in California Government Revenues Since Proposition 13, Public Policy Institute of California, San Francisco, California, forthcoming 1998. Shires, Michael A., and Melissa Glenn Haber, A Review of Local Government Revenue Data in California, Public Policy Institute of California, San Francisco, California, 1997. 82 About the Authors MICHAEL A. SHIRES Michael A. Shires is a research fellow at the Public Policy Institute of California. Before joining PPIC, he was a doctoral fellow at RAND's Graduate School of Policy Studies, concentrating on domestic education policy, California fiscal policy, and international trade policy. He is the author of several publications on U.S. trade relations, higher education, and California state finance and education. He has also been active as a consultant to private companies on strategic planning, marketing research, and tax planning and preparation. He holds a B.A. in economics from the University of California, Los Angeles; an M.B.A. from UCLA’s Anderson Graduate School of Management; and a Ph.D. in public policy analysis from the RAND Graduate School. JOHN ELLWOOD John Ellwood is an adjunct fellow at the Public Policy Institute of California and a professor in the Richard and Rhoda Goldman School of Public Policy at the University of California, Berkeley. He served as PPIC's research director from 1996–1997. He has held academic positions at the University of Minnesota, Dartmouth College, and Princeton University. He also has extensive experience in government, including service as the special assistant to the director of the Congressional Budget Office from 1976 to 1980. He received an A.B. in political science from the Franklin and Marshall College and an M.A. and Ph.D. in political science from the Johns Hopkins University. MARY SPRAGUE Mary Sprague is a Ph.D. student in political science at the University of California, Berkeley. During 1996–1997 she worked as a research assistant at the Public Policy Institute of California. Previously, she spent a year as a Javits Fellow on the U.S. Senate Finance Committee. She holds an A.B. in public policy from Stanford University and an M.P.P. from the University of California, Berkeley. 85 Other PPIC Publications Dardia, Michael (1998), Subsidizing Redevelopment in California. Dresch, Marla, and Steven M. Sheffrin (1997), Who Pays for Development Fees and Exactions? Johnson, Hans P. (1996), Undocumented Immigration to California: 1980–1993. Lewis, Paul G. (1998), Deep Roots: Local Government Structure in California. Lewis, Paul G., and Mary Sprague (1997), Federal Transportation Policy and the Role of Metropolitan Planning Organizations in California. MaCurdy, Thomas, and Margaret O’Brien-Strain (1997), Who Will Be Affected by Welfare Reform in California? MaCurdy, Thomas, and Margaret O’Brien-Strain (1998), Reform Reversed? The Restoration of Welfare Benefits to Immigrants in California. Reed, Deborah, Melissa Glenn Haber, and Laura Mameesh (1996), The Distribution of Income in California. Reyes, Belinda I. (1997), Dynamics of Immigration: Return Migration to Western Mexico. Shires, Michael A., and Melissa Glenn Haber (1997), A Review of Local Government Revenue Data in California. Smolensky, Eugene, Eirik Evenhouse, and Siobhán Reilly (1997), Welfare Reform: A Primer in 12 Questions. Spetz, Joanne (1996), Nursing Staff Trends in California Hospitals: 1977 through 1995. Copies of PPIC publications may be ordered by e-mail (order@ppic.org) or phone (415) 291-4400. Full text versions are also available at: www.ppic.org. 86" ["post_date_gmt"]=> string(19) "2017-05-20 09:35:09" ["comment_status"]=> string(4) "open" ["ping_status"]=> string(6) "closed" ["post_password"]=> string(0) "" ["post_name"]=> string(8) "r_998msr" ["to_ping"]=> string(0) "" ["pinged"]=> string(0) "" ["post_modified"]=> string(19) "2017-05-20 02:35:09" ["post_modified_gmt"]=> string(19) "2017-05-20 09:35:09" ["post_content_filtered"]=> string(0) "" ["guid"]=> string(50) "http://148.62.4.17/wp-content/uploads/R_998MSR.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) }