Donate
Independent, objective, nonpartisan research

R 998PLR

Authors

R 998PLR

Tagged with:

Publication PDFs

Database

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

View live version

object(Timber\Post)#3711 (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_998PLR.pdf" ["wpmf_size"]=> string(6) "241525" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(181072) "Deep Roots: Local Government Structure in California Paul G. Lewis, 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 Political scientists have long struggled with the question of the “right size” for local government—whether the measure is geography or budgets. Jurisdictions covering many square miles, such as a typical county in California, may have the merit of a diverse and extensive tax base but, at the same time, may be seen as “out of touch” with the needs and desires of the average resident or voter. Larger public budgets may have the merit of flexibility in allocating funds among different services and in setting different priorities over time, but smaller budgets and jurisdictions are seen as more “user friendly” and more easily modified in response to community-level preferences and concerns. For all this understanding of basic budgeting and governance principles, there is no supportable theory for the right-sized budget or jurisdictional boundary. Nevertheless, a substantial literature supports the argument that “excessive” governmental fragmentation may complicate the coordination of services and the planning of land and infrastructure development. In California, Proposition 13 is often viewed as a iii significant incentive for governmental fragmentation: Smaller units of government can be established with the explicit purpose of segregating— and collecting taxes from—high-revenue-producing properties. Research fellow Paul Lewis tackles this topic head-on in this report. He concludes that there is no evidence that Proposition 13 created any unusual movement toward either the creation of special districts or the incorporation of new cities. In fact, the slowing of local government fragmentation brought about by the creation of Local Agency Formation Commissions in 1963 has not fundamentally changed over the 20 years since the passage of Proposition 13. This report is one in a series of PPIC publications stemming from our ongoing program of work in governance and public finance. Two recent publications of direct relevance to the analysis of Proposition 13 include Michael Shires, John Ellwood, and Mary Sprague, Has Proposition 13 Delivered? The Changing Tax Burden in California; and Mark Baldassare, When Government Fails: The Orange County Bankruptcy, University of California Press. The completion of each of our studies results in a better understanding of the changing shape of governance and public finance in California. Although formulating policy options is never easy, the clarity of the issues involved is definitely improving as our work proceeds. David W. Lyon President and CEO Public Policy Institute of California iv Summary Many observers have argued that California’s system of local government is too complex and fragmented. Such complexity, some say, makes government action hard to coordinate, political responsibility hard to focus, and government actions less efficient. Moreover, several commentators have attributed local government fragmentation, in part, to the Proposition 13 tax limitation of 1978, which many expected to lead to a proliferation of new cities and special district governments. More recently, there have been legislative proposals in the early 1990s for regional governance and state planning, extensive recommendations for restructuring local governance from the California Constitution Revision Commission in 1996, and the creation in 1997 of a Commission on Local Governance in the 21st Century. This report aims to provide necessary background and context for these deliberations about local government structure. It summarizes what is known and not known about the causes and consequences of local government fragmentation. The report uses data on government v structure and finances from the State Controller and the U.S. Census Bureau’s Census of Governments (which is taken every five years, with the most recent data from 1992). The data are used to summarize recent historical trends relating to California’s deployment of cities, counties, and special districts—looking in particular at the possible influence of Proposition 13 on the shape of the local public sector. Finally, the report examines variations in the level of local government fragmentation in the different counties of the state and analyzes the factors that underlie these differences. Why Is Local Government Structure Important? Much research suggests that political fragmentation—the degree to which governmental powers are divided geographically and functionally among local governments—has important effects on both the process of policymaking and residents’ quality of life. These effects should be considered in relation to a number of different realms of local governance. For the provision of routine public services, such as trash collection or street cleaning, fragmentation may offer an advantage rather than a disadvantage. Despite reformers’ fears about “duplication” and “overlap,” multiple local governments may be advantageous in providing residents with a range of choices, thus better reflecting their varied tastes for particular levels and types of services. In addition, competition, particularly among municipalities, may restrain the growth and inefficiency of local government bureaucracies. However, the administrative costs of multiple special districts in an area may drive up the price of government. In the realm of land use and infrastructure, fragmentation of local governance is often seen as less benign. It may lead to parochial, self- vi interested land-use decisions and a lack of policy concern for regional challenges such as housing affordability, racial and socioeconomic integration, and the ability of regional infrastructure to accommodate growth patterns. A fragmented public sector may be less effective at promoting regional economic development. And in the realm of equity, fragmentation can separate fiscal resources from needs, as lower-income areas suffer with weaker tax bases. Here, the case for financial assistance from state and national governments is particularly strong. Overall, critics of political fragmentation fault it because areas that are more politically divided are believed to have a more difficult time addressing the “general welfare.” California’s system of local governance—with 58 counties, 471 cities, and nearly 5,000 special districts (many of them in fact governed by cities and counties)—undoubtedly appears complex to most residents. An individual household usually will find itself simultaneously governed by several overlapping jurisdictions. In reality, however, California’s local government structure is less complex than the national average. The state has many times fewer counties and cities per capita and significantly fewer politically independent special districts per capita. And although it may appear disordered, California’s public sector did not evolve randomly but rather in response to the varying needs and demands of homeowners, businesses, and other groups as the state has grown and urbanized. Voters have repeatedly created new cities and special districts, in search of local control of land use, public services, and taxes. vii Did Local Government Fragmentation Increase in the Wake of Proposition 13? Historically, California policymakers have shown a number of waves of interest in issues of local government structure. Over the past two decades, concern has centered in large part on the effects of the Proposition 13 tax limitation upon the “proliferation” of local governments. Many observers pointed with alarm to numerous city incorporations and special district formations in its wake and saw portents of a great increase in governmental fragmentation. The conventional wisdom, found in numerous academic and journalistic accounts, holds that Proposition 13 created incentives for the widespread creation of new cities and special districts. In the case of cities, this property tax limitation removed concerns that a newly formed government would lead to increased property taxes, since Proposition 13 caps ad valorem property tax rates and restricts reassessments. Moreover, with counties struggling fiscally in the post-1978 era of revenue constraints, many areas might be expected to “secede” from county service provision by incorporating as a city, thus capturing some locally generated revenues for local use. In the case of special districts, fiscally starved counties and cities might be expected to encourage the creation of new special-purpose governments. The counties and cities could thereby shed some of their service and revenue-raising burdens onto entities that often are somewhat more flexible in their ability to issue debt or levy fees and service charges. Local officials and developers seeking to finance capital facilities and services for growing populations might also view districts’ relative political insulation and invisibility to taxpayers as an advantage. viii In general, however, this conventional wisdom is not supported by the data. It is true that 54 new cities have incorporated since 1978, but most were in areas of metropolitan Southern California that had extremely rapid population growth and suburbanization and thus were natural candidates for additional governments. Looking at incorporations across the state’s history, we can see that the post-1978 era hardly seems exceptional in its rate of city formation, as Figure S.1 illustrates. Rather, the growth rate in the number of new cities has slowed since the state created Local Agency Formation Commissions (LAFCOs) in each county in 1963 to regulate government boundaries and formations. In fact, the passage of a 1992 state law requiring LAFCOs to approve incorporations only if they are “fiscally neutral” for the affected county appears to have brought incorporations to a virtual halt. Most new population growth in the state continues to be accommodated in existing cities—which regularly expand their boundaries through annexation. Thus the number of cities per 100,000 population has fallen throughout this century, as Figure S.2 shows. Reliable and comparable historical data on special districts are harder to come by, but the evidence indicates that the overall number of special districts has increased only gradually since the 1970s. However, the universe of districts has changed somewhat in character, probably in response to Proposition 13. Districts that exist to finance the construction of capital facilities have increased in number, whereas those engaged in more traditional service provision have declined in number. “Governments created by governments” have become more popular, as policymakers have used the ability of districts to levy enterprise fees and issue debt; meanwhile, politically autonomous districts relying on property tax financing have been a waning presence since Proposition 13. ix 500 450 400 350 300 250 200 150 100 50 0 1850 1870 1890 1910 1930 1950 1970 1990 Figure S.1—Number of Cities in California, 1850–1997 9 8 7 6 5 4 3 2 1 0 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 1997 Figure S.2—Number of Cities per 100,000 Population in California, 1900– 1997 x Overall, the relative share of total local government spending in the state accounted for by politically freestanding special districts, as compared to counties and cities, shifted only marginally between 1972 and 1992, as Figure S.3 indicates. Counts of governments may be somewhat misleading barometers of local government change, so this report also introduces a local political fragmentation index, or PFI. The PFI measures the degree to which spending activity is divided among local governments; it represents the probability that two randomly selected dollars of local expenditure are spent by different local governments. Measured statewide or at the county level, local political fragmentation increased only very slightly in the period 1972 to 1992, as Figure S.4 shows. In short, even as Proposition 13 played havoc with many local governments’ finances, it did not decisively alter overall local government structure. 50 45 40 35 30 25 20 15 10 5 0 1972 Counties Cities Special districts (Census definition) 1977 1982 1987 1992 Figure S.3—Percentage Shares of Total Local Nonschool Expenditures, by Type of Government xi 1.0 0.9 0.8 0.7 0.6 0.5 0.4 1972 Statewide PFI Population-weighted county PFI Average county PFI 1977 1982 1987 1992 Figure S.4—Political Fragmentation Index in California, 1972–1992 What Factors Affect Local Government Structure Across California’s Counties? Examining the political fragmentation index at the county level shows that most California counties also did not experience much variation in their overall local government structure from 1972 to 1992. However, the PFI data do reveal that parts of the state differ quite markedly in their local political structure. For instance, the San Francisco Bay area tends to be more governmentally fragmented than metropolitan Southern California, but the latter area has experienced some convergence toward the higher level of fragmentation in the former area. In general, counties with high levels of political fragmentation tend to have several sizable cities, a low proportion of the population in unincorporated areas, and relatively high usage of special districts. xii Multivariate data analysis was performed to examine the factors underlying the variations among counties in political structure. Although numerous variables were considered, the degree of governmental fragmentation appears to largely be a product of three underlying factors. The most important is the county’s population size, which is positively related to fragmentation. Bigger, more complex counties are likely to have more complex service demands and thus more complicated public sectors. The second main influence on local government structure is intergovernmental aid. Areas that have received a higher share of their revenues from state and federal sources tend to acquire less governmental complexity. This could be because intergovernmental aid can “level up” service levels and infrastructure quality across local areas, and because such aid helps keep existing governments in better fiscal condition. Both of these effects associated with intergovernmental aid tend to reduce the incentives for incorporation of new cities and also reduce the impetus to engage in “creative financing” by forming a special district. A third factor involves the county’s historical pattern of population growth. “Older” counties—those that achieved most of their ultimate population size before 1950—tend to be significantly more fragmented. This effect is consistent with the notion that most of California’s local government development occurred during earlier periods of the state’s history. In particular, residents in earlier eras were quicker to form new cities. What Are the Implications for Policymakers? This portrait of relative stability in California’s system of local governance is a rather remarkable “nonfinding” in an era of very rapid political, economic, and demographic change in the state. One possible xiii reading of this record is that an earlier state policy—the creation of LAFCOs in each county—has succeeded in reining in the fragmentation of government. Although their merits have been widely debated, LAFCOs, which have governing boards representing existing cities and counties, might be expected to try to preserve the viability of existing units by steering population growth into cities and looking warily upon the formation of new units. Later changes to the LAFCO law, requiring them to create “spheres of influence” for cities and providing new instructions that further tightened up incorporation requirements, likely strengthened these tendencies. Would-be reformers should also consider the possibility that existing local governments may be more malleable and responsive to change than is commonly supposed. Throughout the state, cities, counties, and special districts typically have been able to address new issues, empower new constituencies, and respond to economic and population growth without falling into crisis. Does that mean that state and local policymakers should be complacent about California’s local government system? Not necessarily. Local fragmentation continues to separate fiscal resources from social needs and to impede coordination on issues such as land use. Despite California’s home rule tradition, local governments remain “creatures of the state,” subject to fine-tuning from Sacramento. Policymakers should think seriously about ways to enhance the fiscal stability of local governments—keeping in mind the finding that more generous intergovernmental aid appears to reduce the impulse toward local fragmentation. And enhanced mechanisms for regional coordination on growth and infrastructure issues may be necessary. xiv The new Commission on Local Governance for the 21st Century— like the recent California Constitution Revision Commission—presents a forum for considering local government’s strengths and weaknesses and debating possible changes in the state’s rules for local governance. But those who would propose sweeping reforms or large-scale consolidations of local units must confront the reality of a stable and resilient system of local rule. xv Contents Foreword..................................... iii Summary..................................... v Figures ...................................... xxi Tables ....................................... xxiii Acknowledgments ............................... xxv 1. INTRODUCTION: THE CONTROVERSY OVER LOCAL GOVERNMENT STRUCTURE ............ What Difference Does Local Government Structure Make? .. Service Provision............................ Land Use and Economic Development ............. Equity .................................. Representation ............................. Organization of This Report ...................... 1 3 4 5 6 6 8 2. LOCAL GOVERNANCE AND HISTORICAL CHANGE ................................. A Recurring Policy Debate ....................... Creating Local Agency Formation Commissions........ Worrying About the Effects of Proposition 13 ......... Focusing Renewed Attention on Local Government Problems ............................... Frequency of Different Categories of Local Government .... 9 10 10 11 13 15 xvii Counties ................................. Cities ................................... Special Districts ............................ School Districts ............................ Shares of Local Public Spending by Different Types of Government............................. Summary .................................. 3. LOCAL GOVERNMENT FRAGMENTATION IN CALIFORNIA: MEASUREMENT AND TRENDS ..... Problems in Measuring Local Government Structure by Counting Governments ..................... An Index of Local Political Fragmentation ............. Levels of Local Political Fragmentation Among California Counties ............................... Did Proposition 13 Accelerate Fragmentation? .......... Summary .................................. 4. ACCOUNTING FOR VARIATIONS IN LOCAL GOVERNMENT STRUCTURE ACROSS CALIFORNIA .............................. What Factors Might Affect Local Government Structure? ... General Contextual Factors ..................... Intergovernmental Rules and Relationships ........... Socioeconomic Characteristics of the Population ....... Political Influence of Developers or Other Businesses .... Why Fragmentation? Toward a Model of Local Government Structure ............................... Bivariate Relationships ........................ Multivariate Model .......................... Interpretation ............................. Other Variables ............................ Summary .................................. 5. CONCLUSIONS AND POLICY CONSIDERATIONS ... Review of Findings ............................ Relative Stability in an Era of Change—Why? .......... Considerations for State Policymakers ................ 17 18 25 29 30 31 33 33 35 37 41 45 47 47 48 49 52 55 57 60 60 63 66 67 69 70 72 75 xviii Appendix A. Research on the Effects of Local Government Fragmentation ............................... B. Issues Regarding Special District Data in the Census of Governments ............................... C. New Cities Incorporated After Proposition 13 .......... 79 85 91 Bibliography .................................. 95 About the Author ............................... 101 Other PPIC Publications........................... 102 xix Figures S.1. Number of Cities in California, 1850–1997 ......... S.2. Number of Cities per 100,000 Population in California, 1900–1997 .............................. S.3. Percentage Shares of Total Local Nonschool Expenditures, by Type of Government ............ S.4. Political Fragmentation Index in California, 1972– 1992 .................................. 2.1. Number of Cities in California, 1850–1997 ......... 2.2. Number of Cities per 100,000 Population in California, 1900–1997 .............................. 2.3. Number of School Districts in California, 1962–1992... 2.4. Percentage Shares of Total Local Nonschool Expenditures, by Type of Government ............ 3.1. Levels of Political Fragmentation in California Counties, 1992 .................................. 3.2. Political Fragmentation Index in California, 1972– 1992 .................................. 3.3. Political Fragmentation in California’s Major Metropolitan Areas ......................... x x xi xii 20 23 29 31 39 42 44 xxi Tables 2.1. Local Governments per 100,000 Population: California Compared to the Rest of the United States .......... 2.2. Number of New Cities by County, 1978–1997 ....... 2.3. How California Accommodated Population Growth After Proposition 13, 1978–1997 ................ 2.4. Number of Special Districts in California ........... 3.1. Political Fragmentation Index of Local Government Structure in California Counties, 1992 ............ 3.2. Average Percentage Shares of Nonschool Expenditure in High- and Low-PFI Counties .................. 3.3. A Comparison of Local Political Structure in Two Counties, 1992 ........................... 4.1. Variables Studied for Possible Relationships with Local Political Fragmentation ...................... 4.2. Regression Models of Political Fragmentation Indexes for California Counties, 1992 .................... B.1. Reconciling Controller and Census Bureau Counts of Special Districts, 1992 ....................... C.1. California Cities Incorporated Since July 1, 1978 ...... 16 19 23 26 38 40 40 58 62 88 91 xxiii Acknowledgments I wish to thank my research colleagues at PPIC for numerous helpful discussions and critiques of this project. In particular, Fred Silva was of key importance as a sounding board and reality check throughout the project, and Kim Rueben wrote a very useful review of the manuscript. The external reviewers, Alvin Sokolow of the University of California, Davis, and April Manatt, consultant to the California Senate Local Government Committee, also read the manuscript with care and provided helpful comments. Mary Sprague lent early research assistance—and in particular, accomplished the unenviable task of assembling the Census of Governments data into a useable format. Michael Molloy performed a reconciliation of Census and state counts of special districts, discussed in Appendix B. Sun Kim produced the map in Chapter 3. The Demographic Research Unit at the state Department of Finance provided important data. Michael Teitz and Andrew Isserman helped keep the project on track and lent useful substantive advice. Gary Bjork provided expert editorial help. While all of these xxv people (and others) improved the final product, the responsibility for any errors of fact or interpretation rests with me. xxvi 1. Introduction: The Controversy over Local Government Structure In California, as in all American states, local government should be treated as a plural, not a singular noun. Political authority in California is divided among thousands of local jurisdictions. Los Angeles County alone has 88 cities, for instance, and advocates of secession in the San Fernando Valley seek to create another. Local government, though, means more than just cities. An individual household may find itself simultaneously governed by several overlapping jurisdictions: a county, a city, a school district and community college district, along with numerous other specialized and politically independent entities, devoted to purposes ranging from regional mass transit to wastewater treatment to firefighting to mosquito abatement to flood control. Additional “dependent special districts”— technically separate financial entities, but under the political control of county supervisors or city councils—may levy assessments on households 1 or charge fees to customers to provide services or build community facilities for a geographic area. California policymakers have periodically expressed concern over this complexity of local government structure. Although the issue rarely commands newspaper headlines, state legislators and governors have repeatedly struggled with what many have thought of as a “proliferation” of governments. Thus, county-level commissions have been created to oversee annexations and formations of new cities and special districts; regional reorganizations have been debated, as policymakers worried about the effects of highly decentralized land-use powers on growth patterns; and study commissions have been appointed—most recently, in 1997, a Commission on Local Governance for the 21st Century. Lurking in the background over the past two decades has been the concern that the Proposition 13 property tax limitation of 1978 would serve to accelerate the formation of new governments, and thus fundamentally alter California’s local governance structure. Throughout these debates, some have argued that local political “fragmentation” makes government seem more complex and “out of control” to citizens, while also making coordinated government action in response to policy problems more difficult. Other commentators and academic writers have argued that a highly decentralized government structure has a host of underappreciated benefits in terms of efficiency and governmental responsiveness to the public. Rarely, however, have the debates in local and state arenas moved beyond simple catch phrases such as “wasteful duplication” or “local control” to seriously examine the structure of local governance in California. This report, part of a continuing line of research on local government institutions and finances from the Public Policy Institute of 2 California, is intended to remedy this shortcoming. It provides key background to inform the debate over local governance and state/local relations, examining long-term trends in California’s local government structure and considering its determinants. What Difference Does Local Government Structure Make? Why has so much concern been given to local political structure? To some degree, public officials have been uneasy about what was perceived as overly rapid change in the system of counties, cities, and special districts. In general, though, concern for local political structure is hardly abstract but is rather connected to concerns about the efficiency, equity, and coordination of local government policies, services, and revenue patterns. Much research suggests that political fragmentation— the degree to which governmental powers are divided geographically and functionally among local governments—has important effects on the provision of government services and the regulation of land use.1 In addition, many policymakers are concerned that a proliferation of local governments makes policy hard to coordinate in areas such as transportation, housing, and environmental protection, and separates fiscal needs from resources. This topic has been the subject of extensive academic analysis. The literature on the effects of local government arrangements is summarized and cited in Appendix A. In this chapter, I briefly highlight the policy and political implications of such research. The literature is marked by a ____________ 1The term fragmentation, used in this report synonymously with political or governmental complexity, is a descriptive term; it is not intended to connote a negative or positive value judgment. 3 certain amount of confusion and dissensus. One nearly universal problem is a failure to distinguish among the varying roles of local government that might be differentially affected by fragmentation. At least four distinct roles should be considered, touching on different areas of local representation and governance. Service Provision First, local government may be viewed primarily as a provider of fairly routine public services, such as trash collection and fire protection. This role involves most of the everyday duties of local public employees and expenditure of local tax dollars. In this realm—somewhat analogous to the world of business—economic theories of competition among local governments would seem to hold the strongest logic. According to this framework, more fragmentation means that boundaries sort residents into “communities of interest” with similar tastes for services and tax levels, thus allowing a closer match between a government’s offerings and its residents’ desires. The competition associated with fragmentation also may keep local bureaucracies from the dangers of monopolistic service provision. It is possible in some cases, however, that the administrative costs of duplication and overlap can drive up the price of having multiple governments. Research particularly supports the idea that this may be a problem where there are numerous special districts. One particularly important service provided locally is public education. Here too, fragmentation among local public school systems might sharpen competition, and thus increase educators’ attention to educational outcomes. If so, this could boost a region’s labor force quality. Thus, for the provision of some local services, fragmentation may enhance government efficiency. 4 Land Use and Economic Development A second role for cities and counties is to regulate land use, through the so-called “police power” that customarily gives them the power to create zoning ordinances, general plans, rules for subdividing land, and building regulations. Special districts can also affect land-use patterns by building growth-supportive infrastructure, such as sewer lines. This realm of local governance focuses less on the routine servicing of existing development and more on the process of development itself and the shape and character of the community over time. Given the spillover effects of local decisions relating to urban growth on surrounding areas, critics of political fragmentation often have focused on this realm of local government behavior. They allege that localities regulate land use in their narrow self-interest and slight area-wide concerns (for example, by focusing on tax-generating retail development and restricting the construction of affordable multifamily housing). Several studies also have linked increased levels of socioeconomic and racial segregation to self-interested local land-use regulation. A related issue is the local public sector’s contribution to the economic development of its region. Studies of fragmentation have seldom dealt with this realm, although there are reasons to suspect that political structure could have an effect. For example, the segmentation of a region politically might make it more difficult to fund, plan, and construct an effective regional road or transit network, or to site necessary but undesirable public facilities such as solid waste disposal sites and sewage treatment plants. 5 Equity A third area of concern is the equity issue—local government’s role in distributing and redistributing life chances. Fragmentation is often seen as separating fiscal resources from social needs. Partitioning a region into separate taxing-and-spending jurisdictions can lead to a situation in which some residents with disadvantaged backgrounds and high needs find themselves in deteriorating cities with inferior schools and public services, while other jurisdictions blessed with a superior tax base may provide services and facilities that go far beyond what a typical city can offer. Again, however, as Neiman (1982) demonstrates, community social status itself has direct effects on “tastes” or demands for public services that may lead to service disparities. That is, rich communities may simply want and demand more from their schools than poor ones. Thus, only one component of unequal service distributions is the result of government fragmentation per se, whereas another part of the inequality is attributable to different “tastes” for services. In any event, because of the difficulty in rectifying disparities among local governments, and local governments’ natural disincentive to redistribute (Peterson, 1981), typically the state and national governments are expected to step in to emphasize redistributive policies. Representation A central question about local government fragmentation is its significance for effective political representation and accountability. On one hand, greater “local control” through fragmented municipal government is often highly valued, as residents place greater faith in decisionmaking that is less insulated and closer to home. However, important issues touching on a citizen’s role as a resident of an 6 interconnected region—the health of the regional labor market, regional transportation networks, or environmental problems, for example—can lack effective arenas for debate and representation where the public sector is highly divided among numerous cities. In the case of special districts, the fragmentation of governance into different functions may have both positive and negative aspects for representation. If we sort decisionmaking and revenues for certain services into separate decisionmaking venues—for example, a water district or regional transit district—we may make programmatic and budgeting decisions about those services more transparent to citizens, who will “know what their taxes are paying for.” On the other hand, the unique ability of general-purpose governments to engage in more global budgeting—setting an overall taxing and spending level, and making tradeoffs between competing services—is sacrificed if special districts take over specific functions. Thus, a degree of coordination is lost as some revenues are earmarked for particular functions. In addition, some citizens may be baffled by the wide array of local units, some of which remain almost invisible to the public. Thus, the effect of governmental fragmentation may be more complex and nuanced than one might suspect at first. Nevertheless, the structure of local government can be expected to have a potentially very significant effect in all four realms. It is therefore important to focus on the determinants of governmental fragmentation itself. Yet the analysis of the underlying causes of fragmentation has been far less common than descriptions of fragmentation and theorizing about its effects. Why are some areas more politically complex than others? This issue will be a major focus of this report. 7 Organization of This Report Our intent in this introductory chapter has been to summarize for policy audiences what is known about the potential effects and importance of local government fragmentation. The remainder of this report examines California’s particular situation in detail. The report relies on data on government structure and expenditures from the U.S. Census Bureau and the California State Controller, along with standard demographic information from Census and state sources. The distinct governmental realm of school districts is discussed only in passing, as the report emphasizes the choices that Californians have made in providing nonschool-related public goods and services. Chapter 2 presents a descriptive analysis of the state’s local government structure and its changes over time. It also provides an account of the recurring political debate over the “proliferation” of local governments in California. Although many would make a blanket statement that California’s governance is fragmented, it would be more useful to have some relative measure of the complexity of local governmental structure in various parts of the state. This challenge is taken up in Chapter 3. Chapter 4 analyzes the reasons for variations among levels of fragmentation in different counties, and Chapter 5 offers conclusions and policy considerations. 8 2. Local Governance and Historical Change For the typical California resident, the closest and most frequent interactions with government are at the local level. A varied assortment of cities, counties, school districts, and special districts provide services, produce public goods, and demand taxes that collectively help to shape the safety, environment, personal development, and financial burdens of each resident. As discussed in the last chapter, there are good reasons to suspect that the way these governments are organized, and the degree to which authority is divided among local units, significantly affect the governance process and residents’ quality of life. This chapter summarizes California’s policy debates over local government fragmentation and then presents data regarding historical trends in the numbers of various types of local governments in the state. Although simply counting governments can be a somewhat misleading way of discerning the structure of local governance (as will be discussed in Chapter 3), historical analysis of these counts can begin to illuminate 9 the issues that reformers and policymakers sometimes raise. In particular, we can examine whether the Proposition 13 tax limitation had major effects on local government structure. A Recurring Policy Debate Although many scholars and some state policymakers have long been wary of government fragmentation, the popular instinct generally has been one of localism, decentralization, and home rule. Voters have repeatedly created new cities and special districts, in search of local control of land use, public services, and taxes. Typically, the political debate over forming these new governments has been contained inside the county or local area considering the change. Occasionally, however, such decisions have been amplified into a wider, state-level debate about the dimensions of local government structure. In at least three recent periods—in the early 1960s, in the wake of Proposition 13 in 1978, and over the past decade—state policymakers have raised concerns about local government responsibilities and complexity. Creating Local Agency Formation Commissions First, after World War II, as California grew and suburbanized rapidly, new local governments appeared in large numbers. From 1953 to 1963 alone, 75 new cities were formed. Concerned with the escalating number of local units, Governor Edmund “Pat” Brown created a Commission on Metropolitan Area Problems, asking it to consider the questions, “Do we have too many overlapping jurisdictions?” and “What is the danger point in proliferation of local government?” (Teaford, 1997, p. 69). 10 The commission recommended creating a statewide board for reviewing and passing judgment upon proposed local boundary changes. State legislators, reacting to the commission’s report but shying away from this controversial proposal, ultimately passed a compromise provision, the Knox-Nisbet Act of 1963. This law required that each county create a Local Agency Formation Commission (LAFCO). With representatives both from the cities and the county board of supervisors within each county, LAFCOs were given the power to oversee changes in government structure by regulating incorporations, annexations, special district formations, and other structural changes in their respective counties. Given only relatively “loose guidance” by state statutes (Feinbaum, 1987, p. 499), LAFCOs have shown significant variation from county to county in their approaches to jurisdictional arrangements. The overall effectiveness of LAFCOs has been widely debated, but there is no doubt that they have become key players in local intergovernmental politics (Martin, 1984, Chap. 6; Fulton, 1989; Teaford, 1997, pp. 103–107; and Miller, 1981, Chap. 5). In any event, city formation did level off in the later 1960s and into the 1970s, after the LAFCO law took effect. Worrying About the Effects of Proposition 13 Second, a fear arose that the passage of Proposition 13 in 1978 would fundamentally alter or even distort California’s local governance structures. Many observers have argued that Proposition 13 has been an impetus for the formation of new cities and special districts. This section briefly summarizes this “conventional wisdom” as to why Proposition 13 would lead to a proliferation of local governments. 11 Consider the case of cities. Citizens desiring greater “local control,” particularly over land use, always could turn to the possibility of incorporation—that is, creation of a new city. But beginning in 1978, with their property tax rates now frozen by Proposition 13, residents of outlying, unincorporated settlements had an additional incentive to incorporate: Creating a new city would not lead to an increase in their property tax bill. In the past, fears that taxes would go up after city formation allegedly had been a major disincentive to incorporation. But the formulas that the state developed after Proposition 13 for distributing property taxes among local governments within each county meant that the distribution of property tax revenues became a zero-sum game. That is, creating a new city would remove some resources from the county but would not burden property owners in that city with any new ad valorem property taxes, unless their rate had been below the 1 percent maximum rate established by Proposition 13. The LAFCO in each county was given a role in sorting out the redistribution of revenues following an incorporation. Given these formulas and rules, forming a city would allow residents to “capture” some of the tax revenue generated in their area for use in their area. In particular, this consideration could be a major incentive to form a new city if the area contained major retail activity, since a portion of the state sales tax is returned to the jurisdiction in which the sale takes place. Finally, counties were seen as being fiscally weakened by Proposition 13; thus, “escaping” from county service provision by forming a city might make sense for residents seeking higher-quality services and more financially stable government. In the case of special districts, local officials and property developers, rather than residents or homeowners, were expected to be the strategic 12 players after Proposition 13. With general-purpose government resources limited, many cities and counties were seen as less able to engage in new functions or activities, particularly those involving major capital investment. Cities and counties might also be expected to shed some of their existing burdens upon other units, where possible. In any case, creating special districts might be expected to be a convenient response to fiscal limitations. Certain types of special districts are more flexible than counties or cities in charging fees for their services, issuing debt, and localizing costs to the geographic areas expected to benefit. Districts’ relative political insulation and invisibility to taxpayers, as compared to the more prominent counties and cities, might also be viewed as an advantage by local officials and developers seeking capital facilities and services for growing populations. For these reasons, the idea arose that Proposition 13 would inexorably change the calculus of local government structure in California, leading to more formations of cities and districts. This perspective has basically been the received wisdom of those who study the state’s local governments. Scholars and journalistic commentators alike have stressed that Proposition 13 was a jarring change to the local government environment and would have major effects on local political structure (see, for example, Bollens, 1986; Brooks, 1988; Feinbaum, 1987; Fulton, 1989, 1991; Martin and Hogan, 1989; Miller, 1981; Schrag, 1998; and Sokolow et al., 1981). Focusing Renewed Attention on Local Government Problems Several other relevant events have occurred over the past decade. The late 1980s and early 1990s saw several legislative proposals to address the issue of local government complexity, particularly in relation 13 to growth and land use—at the time a dominant issue in the state’s politics. For example, the Assembly Office of Research released a report that concluded that The structure of local government clings to the ideal of self-governing, locallycontrolled communities, reinforces the divisive effects of the state’s fiscal structure and creates a mismatch between the fiscal resources and organizational means devoted to problem-solving and the increasing scale of growth-related problems (Assembly Office of Research, 1989, p. 4). In particular, the report highlighted the purported inadequacies of single-purpose government entities and suggested eliminating them in favor of a Regional Development and Infrastructure Agency in each air basin. These agencies would allocate all property and sales taxes above the current base, on the basis of need. Then-Speaker of the Assembly Willie Brown introduced a bill following up on these recommendations, which would have created seven broadly empowered regional agencies.1 Although that proposal died, debate later centered on the recommendations of the California Constitution Revision Commission, created by the state in 1993 and given the charge of examining state/local relationships. The commission reported that the organization and provision of local government services is far too complex . . . . [T]he confusing array of local government entities makes it difficult for citizens to understand which agency is responsible for providing a particular service and whom they should hold accountable. Additionally, with so many local entities involved in providing government services, redundancies often exist (1996, p. 71). ____________ 1Other reform efforts during this period included bills proposing local revenuesharing and a state planning agency, as well as regionalization proposals from the Los Angeles 2000 committee and the Bay Vision 2020 reform group. 14 Reacting to the commission’s findings, both Governor Wilson and the Legislature have discussed “realigning” the roles of the state, its counties, and its cities in providing certain services. In 1997, the Legislature passed a law in response to a different concern—the continuing desires for independence among many residents of the San Fernando Valley in the city of Los Angeles. AB 62 removes the power of city councils to veto secessions of city territory. (Voters in the affected areas must approve the detachment.) This statute applies not only in Los Angeles but throughout the state.2 And in what could ultimately be a broader action, the same session saw the Legislature pass a law creating a new Commission on Local Governance for the 21st Century. Finding that “nearly 35 years have passed since legislators last conducted a thorough investigation” of local government organization, the statute authorizes the commission to make “a careful study of local agency organization and boundaries,” and to make recommendations concerning the organization of local government and the state laws operative in this area.3 Thus, the Legislature returned once again to the issues occupying it in 1963. Frequency of Different Categories of Local Government If one believes that Proposition 13 led to the creation of many more local governments, one might also expect that California, the first state to ____________ 2AB 62 creates a Special Commission on Los Angeles Boundaries to study detachment proposals for the Valley but places this commission under the control of the county’s LAFCO. See Fulton (1997). 3The Cortese-Knox Local Government Reorganization Act of 1985 is the current basis for municipal organization, district reorganizations, and LAFCO procedures. It consolidated and made consistent the rules in this area, which previously were governed by three separate statutes. See Assembly Committee on Local Government (1996). 15 implement a constitutional tax limitation of this sort, would be among the most politically fragmented states. An initial glance at the numbers regarding local governance in California would probably lead most observers to consider the state’s local government structure to be very complex. Using surveys of numbers of governments as reported by the State Controller and the U.S. Census Bureau, the most recently available counts of local governments include 58 counties, 471 cities, 4,874 special districts (many of which are under the control of counties or cities), and 993 school districts.4 California is the nation’s most populous state, however, and one would therefore expect it to have many governmental units. To make a relevant comparison to other parts of the country, we need a per-capita measure of local governments. In these terms, however, the Golden State does not seem to have a particularly complex local public sector in comparison to other states. Table 2.1 compares the number of local Table 2.1 Local Governments per 100,000 Population: California Compared to the Rest of the United States 1992 California Other 49 states Counties Cities 0.19 1.49 1.33 8.39 Special School Districts Districts 9.05 12.83 3.49 5.95 1977 California Other 49 states 0.26 1.85 9.96 1.51 9.35 12.02 SOURCE: Census of Governments (1977, 1992). 4.96 7.12 ____________ 4The county and city counts are as of June 1998; the school and special district counts date to 1995. 16 governments per 100,000 population in California to those numbers in the remaining 49 U.S. states, using data from the 1977 and 1992 Census of Governments. (The Census of Governments counts only those special districts that are freestanding, politically independent units; see Appendix B for details.) This comparison reveals that California in both years had many times fewer counties and cities per capita, and significantly fewer school districts and politically independent special districts per capita, than the rest of the country. Moreover, the table also shows that California has become less fragmented in relation to the other 49 states since the passage of Proposition 13. For example, California had 17.1 percent fewer special districts per capita than the other states in 1977, but 29 percent fewer in 1992. What can be said about the number and growth of each of the four main types of local governments in the state? Counties California has 58 counties (including San Francisco, which is a joint city-county that is unique in the state). Counties have been the most stable element of the local government architecture. There have been no new counties since 1907, when Imperial County was formed out of San Diego County, although there were several unsuccessful attempts at “county secessions” in the 1970s and 1980s (see Detwiler, 1996b, p. 3). Counties have long been the basic administrative agents of the state in local areas and thus carry out state functions as well as local functions. They are much less susceptible to territorial change than other units of local government. 17 Cities As of 1997, there were 471 incorporated cities in California. A number of observers have pointed out what was portrayed as a flurry of incorporations in the early to mid-1980s, typically seen as a consequence of Proposition 13. Feinbaum (1987, p. 497) wrote that “California cities are being formed more often today than at any time since the mid1960s . . . . [M]ost experts see the trend persisting as the state’s population continues to grow and to shift from established cities and rural areas to rapidly developing suburbs.” Feinbaum viewed the reasons for these incorporations as primarily involving dissatisfaction with county land-use decisions, attempts to capture locally generated tax revenues, defense against annexations, and/or the emergence of community leadership and organizations. Similarly, Martin and Hogan (1989, p. 20) wrote of a “trend toward city incorporation associated with the passage of Proposition 13,” pointing out that 35 new cities formed in the decade after 1978, compared to 18 in the previous decade. “In the environment created by Proposition 13,” they argued, “local control becomes an opportunity for a community to better its position relative to other communities,” citing the creation of redevelopment agencies as well as property and sales tax capture as motivations (p. 20). Fulton (1991, p. 238) concluded that “Since Proposition 13 eliminated the possibility of tax increases, voters have realized that incorporation is ‘free.’ Thus, since 1978, the number of new cities has been on the rise, accelerating the financial problems of counties.” In another study, Musso (1994, p. 38) wrote that “the passage of Proposition 13 strengthened the fiscal incentives for incorporation.” (See also Brooks, 1988; and Schrag, 1998, pp. 180–182.) 18 It is true that 54 new cities have incorporated since Proposition 13, a number that sounds large to many. More than three-fifths of these new cities have been founded in the metropolitan areas of Southern California, in the counties around Los Angeles and San Diego. Table 2.2 lists the number of new cities by county, showing that the rapid-growth Inland Empire county of San Bernardino led the way. Appendix C Table 2.2 Number of New Cities by County, 1978–1997 County San Bernardino Los Angeles Riverside Orange San Diego Contra Costa Santa Barbara Alameda Butte Kings Lake Mono Napa Nevada Placer Sacramento San Joaquin San Luis Obispo San Mateo Shasta Sonoma Ventura Yolo Incorporations 9 8 7 5 4 3 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Total 54 SOURCES: Compiled from California State Controller, Financial Transactions Concerning Cities of California (annual). 19 500 450 400 350 300 250 200 150 100 50 0 1850 1870 1890 1910 1930 SB 1559 passed Prop. 13 passed LAFCOs created Local sales taxes authorized Contract cities pioneered 1950 1970 1990 Figure 2.1—Number of Cities in California, 1850–1997 provides a list of all of the cities in the state that have incorporated since Proposition 13. Is it the case, then, that Proposition 13 led to unprecedented growth in the number of cities in California? Not really: Figure 2.1 shows that the long-term trend toward a greater number of cities did not experience any major or sustained blip upward in its trajectory following Proposition 13. Although there has been a certain amount of variation over time, the number of cities in California has been on a seemingly inexorable upward march since the state’s founding. Rates of increase since 1978 have not been especially high by historical standards. Looked at over this longer historical period, the two major periods for growth in the number of cities in California were from about the turn of the century through 1930 and then again in the 1950s and early 1960s. The first period coincides with a great influx of migrants into California and a time when many parts of the state began to take shape as 20 modern settlements. The state constitution of 1879 had enabled local residents to vote for incorporation of a city, rather than having to receive a special act of the Legislature for incorporation, which was typical earlier in the state’s history. The second era coincides with the push of massive postwar suburbanization in the state. Also in this period, the new City of Lakewood initiated a contracting arrangement for services in 1954, allowing new cities with minimal public employment to form; these “contract cities” arrange with their counties, and other public and private service providers, for most of their services. The “Lakewood plan” proved an attractive way for suburban residents to gain the land-use powers of cityhood without incurring the costs of creating a full-service municipality. Finally, another reason for the growing number of cities at this time was the state’s passage of the Bradley-Burns local sales tax in 1956. This provision opened up an important new source of revenue for local governments, making cityhood more viable in many communities with extensive retailing. Thus, some eras saw a great increase in cities. However, two policy changes at the state level may have caused a moderation in the “proliferation” of cities. As one can see in Figure 2.1, the creation of LAFCOs in 1963 was followed by a noticeable reduction in the rate of growth of municipal formations, despite the continued rapid growth of the state’s population. More recently, passage of SB 1559, a 1992 law requiring that incorporations be “fiscally neutral” for counties and other affected local governments, has been followed by a period in which new incorporations have virtually ceased. SB 1559 was motivated by the continuing challenges that counties faced as new cities incorporated and thereby 21 removed some of the county’s tax revenues. Some cities allegedly formed to capture sales tax revenues of large retail complexes, which previously had been providing sales tax revenues for the county. Although state LAFCO law had always directed the commissions to consider issues of fiscal effects on other local governments, LAFCOs in different counties took varying approaches to this issue. The 1992 law provided a stricter rule, making it far less likely that LAFCOs could approve incorporations resulting in revenue windfalls. Incorporations that are fiscally harmful to the county involved are to be disapproved—unless the city and county can come to an agreement beforehand. Proposed new cities now engage in protracted negotiations with counties and LAFCOs over the future distribution of revenues from property and sales taxes, and sometimes other sources such as vehicle license fees.5 Pro-city activists and legislators claim that the law is vague and makes incorporations virtually impossible. Such disgruntlement has led to recent efforts to amend or repeal the “revenue neutrality” law (Fulton, 1998). Despite the nearly continual increase in the number of cities in the 20th century, it is important to stress that the overall trend since about 1910 has been toward a decreasing number of cities per capita, as illustrated in Figure 2.2. In other words, California has generally accommodated growth through the expansion of existing cities rather than the creation of new ones. Table 2.3 provides another illustration of this phenomenon, focusing on population growth in different types of jurisdictions in the period since Proposition 13. As the table shows, ____________ 5There have been continuing disputes between Citrus Heights and Sacramento County, and between Shasta Lake and Shasta County, over their respective revenue agreements. (These cities represent two of the mere three post-SB 1559 incorporations.) 22 9 8 7 6 5 4 3 2 1 0 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 1997 Figure 2.2—Number of Cities per 100,000 Population in California, 1900–1997 almost three-quarters of the state’s population growth from 1978 to 1997 took place in cities that were already in existence in the former year. By contrast, far smaller proportions of the population increase were accommodated in new cities and in unincorporated areas. In fact, the Table 2.3 How California Accommodated Population Growth After Proposition 13, 1978–1997 Cities that existed before 1978 Cities incorporated after 1978 Areas that remained unincorporated Net Growth in Population 7,285,635 1,743,890 885,265 % of State’s Total Population Increase 73.5 17.6 8.9 SOURCE: Calculated from annual city population estimates of the California Department of Finance, Demographic Research Unit. 23 share of the state’s population living in unincorporated areas dropped from 25 percent in 1978 to 20 percent in 1997. The augmented population of existing cities has been mainly accomplished through annexations. Contemporary annexations draw less attention than some of the more massive city expansions of the past—such as Los Angeles’s annexation of the San Fernando Valley in 1915—but the incremental process of adding to city borders continues to be the main way that expanding and urbanizing populations receive public services. Miller (1993, p. 107), drawing upon the Census Bureau’s Boundary and Annexation Survey, found that California cities had engaged in 5,101 annexation actions between 1980 and 1990, adding 726 square miles and about 283,000 residents to their boundaries. The same source lists 148 detachments of territory from cities, involving only 13 square miles and about 1,000 residents. As Detwiler (1996b, p. 5) points out, the current effort at secession in Los Angeles’s San Fernando Valley runs counter to a hundred years of municipal expansion, and those who advocate dividing cities must contend with a long statutory tradition that favors the status quo. Cities shed small parcels from time to time, but detaching large areas with significant populations is quite rare. There are only two examples of successful secession—Coronado in 1890 and Montebello in 1920—while history records over a dozen former municipalities swallowed by larger cities. The impulse for municipal entropy may be growing but state law still reflects the political power of existing cities. However, this political relationship may have changed with the passage of AB 62 in 1997, which allows areas of cities to detach without the permission of the existing city’s council. This new law was motivated by the desire of some residents of the San Fernando Valley to carve a new city out of the City of Los Angeles. 24 Special Districts Historical data on special districts are spotty by comparison to city data, but their numbers also have drawn attention as a reaction to Proposition 13. For example, Bollens (1986) argued that special districts increased in number rapidly after Proposition 13. He concluded, “It is quite probable that the ability of LAFCOs to decrease district formation has been curtailed by the response to Proposition 13 at the local level” (p. 122). Morgan and Chapman (1994) have studied the dynamics of special district creation and consolidation over time in California. They note that the total number of districts counted by the State Controller, which includes both independent and dependent units, has been relatively constant in the past two decades at about 4,900, increasing by about 6 percent in the period since Proposition 13. Because of changes in methodology, special district counts by the controller before 1975 cannot be reliably compared to those of the subsequent period (Morgan and Chapman, 1994). In 1992, the Controller reported 4,995 special districts, of which 3,139 were considered “independent” and 1,856 were governed by county boards of supervisors or city councils.6 The Controller makes a count of all entities labeled special districts, whereas the Census Bureau uses a more restrictive definition, counting only units that are fiscally and politically freestanding of cities and counties. (Refer to Appendix B for more detail about the variations in ____________ 6There were 233 Mello-Roos districts in 1992—facility financing districts that are all dependent on general-purpose governments. Eighty percent of these were not included in the Controller’s report, but their overall financial activity is relatively insignificant in the context of the California local public sector. There were also 81 transportation planning agencies and 381 community redevelopment agencies. See Shires and Glenn Haber (1997). 25 counting special districts.) The Census Bureau’s count of politically autonomous special districts, taken every five years, increased from 2,228 to 2,797 between 1977 and 1992. However, the Census included a substantial number of districts in these tallies that reported no expenditures, according to its government finance data files.7 If one nets out the zero-expenditure districts to focus on active districts, the tally instead shows a rise from 1,855 in 1977 to 2,250 in 1992. Table 2.4 compares the Controller’s tally of active districts with the more delimited Census enumeration for years in which the Census of Table 2.4 Number of Special Districts in California Year 1977 1982 1987 1992 1995 Controller’s Count 4,587 4,815 4,962 4,857 4,874 Census Bureau’s Count 1,855 2,111 2,549 2,250 N/A NOTE: The Controller’s count is adjusted for consistency in definition of special districts, drawing upon the work of Morgan and Chapman (1994). The Census count is adjusted by deleting districts with reported expenditures of zero. For an explanation of the differing counts and definitions between the Controller’s office and the Census Bureau, refer to Appendix B. SOURCES: California State Controller, Financial Transactions Concerning Special Districts of California (annual); Bureau of the Census, Census of Governments (1977, 1982, 1987, 1992); Morgan and Chapman (1994). ____________ 7This counting of presumably inactive districts is one of the major complaints about Census of Governments data on districts. See Appendix C and Leigland (1990). 26 Governments was taken.8 Both show increases from 1977–1987, with a slight reduction in 1987–1992. The differences between the Census Bureau’s rate of growth in the number of districts, as compared to the somewhat flatter trend reported by the Controller, can be explained in part as a result of the changing types of special districts in the state. “Decreases in many of the traditional categories of special districts have been more than offset by significant increases in the creation of county service areas (CSAs) and joint powers [agencies]. . . . Had these ‘governments created by governments’ been held steady, the total number of special districts would have declined more than 7% since 1978” (Morgan and Chapman, 1994, p. 2-i). The Census of Governments treats most county service areas as dependent entities and thus does not count them. It does, however, count most joint powers agencies as independent because they are not controlled by an individual city or county government, but rather by a separately constituted joint board. This delineation is certainly debatable.9 Bollens (1986, p. 122) also notes the change in type of districts, finding that districts reliant on property taxes were decreasing in number, whereas enterprise districts—those that rely on user fees and charges—were increasing. Even where the governing boards of special districts tend to consist of county or city officials, such “governments created by governments” might be viewed as increasing fragmentation in a certain sense: They ____________ 8The 1997 Census of Governments has not yet been released. 9Some may view joint powers agencies as actually reducing political fragmentation, given that they are cooperative efforts among existing local governments. Others have expressed reservations about the fairness of representation of “member” governments on the governing boards of joint powers agencies, and the lack of political visibility of such units. 27 segregate funds and programs from the overall, general-purpose local budget. This means that decisions about district-funded activities tend to be made separately from overall budgeting decisions, and tradeoffs among public priorities are made less clear. Public officials periodically have expressed interest in consolidating special districts, and the Legislature has passed laws making consolidations potentially easier. Nevertheless, such mergers are relatively infrequent. Sometimes city incorporations lead to the absorption of special districts, particularly in full-service cities with minimal contracting. In this sense, a new city that reduces reliance on independent special districts might actually be viewed as a reduction in fragmentation. Overall, however, once created, districts—like most governmental organizations—have incentives to stay in business and retain their offices. Morgan and Chapman (1994) found that enterprise districts have little incentive to consolidate unless they plan to engage in a significant expansion or capital investment. Some nonenterprise districts, on the other hand, find themselves fiscally strapped by the formulas for property tax distribution promulgated for each county after Proposition 13. Nonenterprise districts in growing areas, with much property development and property turnover, are more likely to have experienced some growth in revenues, whereas such districts in “built-out” areas are less likely to see much growth in property tax revenues. Morgan and Chapman suggested that consolidations between growth-area districts and built-out area districts may be wise in some circumstances, but that districts in growth areas often lack incentives to combine with the less fortunate districts. In any event, the authors found after compiling a number of case studies of consolidations that there is little reason to 28 believe that consolidating districts is a way to rein in the cost of government. School Districts Though not the major focus of this report, the experience of school districts is worth mentioning as a notable counter-example to the trend of growth in local governments. As Figure 2.3 shows, the state experienced a significant drop in the number of school districts between the 1962 and 1992 Census of Governments, most of which occurred in the 1960s. This decline in the number of school districts in California reflects an even steeper decline in the United States as a whole. Throughout the nation, many rural and suburban districts were consolidated into large “modern” school districts at the instigation of the 1,700 1,600 1,500 1,400 1,300 1,200 1,100 1,000 1962 1967 1972 1977 1982 1987 1992 Figure 2.3—Number of School Districts in California, 1962–1992 29 states (Nelson 1990, p. 444); and elementary and high school districts often have merged into unified districts. Such consolidation, recommended by professional educator groups and others who sought to modernize and rationalize public education and curricula, shows that at least in this case, local government structure can in fact be changed relatively quickly when there are powerful interests dedicated to doing so. Shares of Local Public Spending by Different Types of Government Counting governments can illuminate historical changes and comparisons, but other measures of local government structure can also be telling. One such measure is the share of total local spending accounted for by different types of local governments. Figure 2.4 shows the shares of total local nonschool expenditures in the state accounted for by counties, cities, and special districts in the five most recently reported Censuses of Governments. As the figure illustrates, these shares did not show much deviation from 1972 to 1992, although counties surpassed cities in overall volume of expenditure. Counties appear to have declined slightly in budgetary significance between 1977 and 1982, a period during which special districts increased in significance by a similar (small) amount. This may indicate that some activities were turned over from counties to special districts in the immediate wake of Proposition 13. The period after 1982, however, shows very little change. In any event, the passage of Proposition 13 in 1978 once again cannot be charged with any clear and major alteration of governmental responsibilities. This concept—how divided the expenditure “pie” is—moves us toward a different summary measure of local government fragmentation. 30 50 45 40 35 30 25 20 15 10 5 0 1972 Counties Cities Special districts (Census definition) 1977 1982 1987 1992 Figure 2.4—Percentage Shares of Total Local Nonschool Expenditures, by Type of Government It focuses upon the dispersion of responsibilities and activities across governments. In Chapter 3, a summary measure of fragmentation for each county, based on this dispersion concept, will be introduced. Summary California has a profusion of local governments but far fewer for its population size than might be anticipated by comparison to the rest of the nation. City governments and special district governments have increased in number in recent decades, although the trend lines for special districts are relatively flat. County governments have stayed constant in number, whereas school districts have declined. Moreover, the notion that Proposition 13 led to major changes in local government formation patterns appears mistaken. City formation did not depart notably from its long-term trend, and was easily outpaced 31 by population gain, as most population growth was accommodated within existing cities. Some special districts were created, particularly joint powers agencies—which attempt to finance public goods for a larger-than-city area, and county service areas—agencies that provide localized financing for localized services. But these formations were almost matched by a dissolution or consolidation of other districts involved in traditional types of service provision—particularly those dependent on property taxes. In addition, the statewide shares of total local expenditures by type of entity, for purposes other than education, did not see much of a shift in the Proposition 13 years. Counties and cities each accounted for about 40 to 45 percent of such expenditures during this period, with special districts responsible for the rest. 32 3. Local Government Fragmentation in California: Measurement and Trends Chapter 2 provided historical data on numbers of local governments in California. Although useful, raw counts of this type can be somewhat misleading and should not be used as the only indicator of political structure in the state. This chapter presents an alternative measure—the political fragmentation index—and illustrates variations among California’s counties in local political structure using this measure. Problems in Measuring Local Government Structure by Counting Governments A hypothetical example illustrates potential problems with merely counting units of government. First, consider the incorporation of a fullservice, large-scale city, as opposed to the incorporation of a minimal contract city that continues to depend upon special districts and the county for most services. The first would be a major alteration in the 33 government responsibilities in its area—control over police, fire protection, and other key services as well as land-use controls; the second would represent a shift of land-use powers from county to city, but little else. Yet a count measure of local political structure would register both incorporations in the same way—as one more unit of government. Similarly, a small mosquito-control district would be counted equally with a large public transit district. Other researchers have similarly warned observers against an obsession with counting local governments. Morgan and Chapman (1994, p. 2), in a report sponsored by the Association of California Water Agencies, wrote, “critics have periodically called for a reduction in the number of districts in an effort to reduce the amount of government in California. They have confused the concept of too much government with that of too many governments” [emphasis in original]. Indeed, it is possible that creating a new special district to carry out a service, under some circumstances, would lead to increased efficiencies—for example by linking benefits received with willingness of users to pay for a service, or by contracting with a private firm to carry out a service. Another warning came from Foster (1993, p. 530), who wrote, “where ‘more units’ is interpreted as ‘more fragmentation,’ researchers obscure the fact that, depending on its service area, an additional government may integrate rather than fragment . . . .” That is, regional entities such as planning agencies or mass transit authorities can take a hard look at conflictual regional issues that existing local units might prefer to ignore. Many researchers studying local political structure move one step beyond counting governments by using ratio-based measures—for example, local governments per 10,000 population, or local governments 34 per square mile. This approach does provide information about the relative density of local governments. But it still tells us precious little about the division of responsibilities and activities among these units. An Index of Local Political Fragmentation A measure of local political structure should provide interpretable information about the relative responsibilities and expenditures of counties, cities, and special-purpose governments.1 This would provide a better sense of how geographically and functionally divided local public authority is. For this reason, I computed for each California county a political fragmentation index, or PFI, which represents the dispersion of local expenditures among the county, city, and special district governments.2 Data on total direct expenditures (not including intergovernmental expenditures), which provide a sense of the amount of activity engaged in by each unit, were tallied from the Census of Governments. The PFI will range between zero and one for each county. It is interpreted as the probability that any two randomly selected dollars of local public expenditure are spent by different units of government. The PFI is calculated by summing the squared percentages of total expenditure ____________ 1School districts are omitted since they involve a qualitatively different realm of local government activity. Also, there are different bases and incentives for their formation and consolidation. 2This measure is adapted from the political fragmentation index used in Lewis (1996, Chap. 3)—with some changes. Counties rather than metropolitan areas are used as basic units of analysis to maximize the number of observations and cover all of California. Also, the index in the earlier study was used to measure political fragmentation across states and thus was forced to include a weighting scheme reflecting the differing responsibilities of the local and state levels in different states—a complication not necessary in this California-specific study. Finally, the earlier study included only those special districts relevant to land use and development, whereas this study includes all special districts counted by the Census. 35 accounted for by each unit of government (expressed as a decimal) and subtracting this number from one. This is a variant of a widely used social science technique, used to discern how dispersed something is across separate units.3 Counties’ score along the PFI’s zero-to-one range indicates the degree to which they have divided up their expenditure “pie” among different governments. For example, consider the following hypothetical cases: • County A includes a county government that spends 50 percent of total local expenditures, two cities that each spend 20 percent, and a special district that spends 10 percent. In this case, PFI = 0.66. • County B has a more complex governing structure. It has a county government that spends 30 percent of the total, four cities that each spend 15 percent, and two special districts each spending 5 percent. Here, PFI = 0.815. In the first case, there is a 66 percent chance that any two randomly selected dollars are spent by different governments, whereas this probability is 81.5 percent in the case of County B.4 In this study, the county is used as the unit of analysis; within the county, data on county, city, and special district expenditures are included. This provides 58 cases to work with in California—or actually ____________ 3For example, the same type of index has been used to compare nations on the basis of the dispersion of votes across different political parties. Social scientists may recognize the PFI as a Hefindahl index, subtracted from one. 4Note that there are two factors that can lead to a higher PFI value mathematically. If spending shares are assumed to be equal across governments, then raising the number of governments will lead to greater dispersion. Similarly, if we assume that the number of governments is fixed, then PFI increases as spending becomes less concentrated and more dispersed. Thus, both a greater number of units and less concentration of spending generally will tend to increase fragmentation. 36 57, since the unique City and County of San Francisco is omitted from much of the analysis to follow. “Counties are appropriate geographic units because they are typically large enough to permit some jurisdictional choice, but small enough to allow changes in [institutional arrangements] without major changes in workplace location or social patterns” (Zax, 1989, p. 561).5 For the purposes of this study, dependent special districts are not treated as separate units of government. As explained in Appendix B, the Census Bureau includes the expenditure totals of such districts as part of their “parent” county or city. This makes sense because, conceptually, the PFI measure relates to the dispersion of political authority. Dependent districts may have separate financial statements, but they are in effect merely the tools or accounting devices of their parent governments. Levels of Local Political Fragmentation Among California Counties Table 3.1 lists the PFI value for each county in the state as of 1992, the most recent year for which Census of Governments data are available. Figure 3.1 shows the geographical patterns involved. There is clearly a substantial degree of variation in local political structure within the state, ranging from a PFI of 0.05 in Alpine County, a rural county in the Sierra, to a PFI of 0.85 in Alameda County, a largely urbanized county in the San Francisco Bay area. ____________ 5I do not weight the counties by population—rather, in Chapter 4, I include population as an independent variable that may influence political fragmentation. My intent is to generalize about all counties in the state (except San Francisco). 37 Table 3.1 Political Fragmentation Index of Local Government Structure in California Counties, 1992 County Alameda Solano San Mateo San Diego Orange Riverside Los Angeles Contra Costa Monterey Stanislaus Marin Tulare Santa Clara Imperial Sacramento Nevada San Bernardino Mendocino Yolo Ventura San Benito Kings Sonoma Santa Barbara Butte Kern Santa Cruz Sutter Fresno Plumas PFI County PFI 0.85 Inyo 0.63 0.81 Shasta 0.63 0.81 Humboldt 0.63 0.79 San Joaquin 0.63 0.78 El Dorado 0.62 0.77 Madera 0.60 0.76 Glenn 0.60 0.76 San Luis Obispo 0.58 0.75 Napa 0.53 0.75 Lassen 0.52 0.75 Lake 0.51 0.73 Colusa 0.51 0.73 Mono 0.48 0.72 Merced 0.47 0.71 Del Norte 0.46 0.71 Calaveras 0.46 0.71 Siskiyou 0.46 0.70 Amador 0.43 0.69 Tuolumne 0.36 0.68 Yuba 0.35 0.68 Tehama 0.34 0.67 Mariposa 0.32 0.66 Placer 0.31 0.65 Sierra 0.31 0.65 Modoc 0.30 0.65 Trinity 0.21 0.64 Alpine 0.05 0.64 Average 0.59 0.64 Median 0.64 0.63 Standard deviation 0.17 What distinguishes high- and low-fragmentation counties? A detailed analysis of factors associated with fragmentation will be presented in Chapter 4. For now, it is worth looking in more detail at the measurements that constitute the index. Counties with the lowest levels 38 Del Norte Siskiyou Modoc Humboldt Trinity Shasta Lassen Mendocino Tehama Glenn Butte Colusa Lake Sonoma Napa Sacramento Yolo Solano Marin San Francisco Alameda Stanislaus San Mateo Santa Cruz Santa Clara San Benito Political Plumas Fragmentation Index Sierra Yuba Nevada Placer El Dorado Sutter Alpine Amador Calaveras San Joaquin Tuolumne Contra Costa Mono Mariposa Less than 0.5 0.5 to 0.65 0.66 to 0.75 Greater than 0.75 Merced Madera Fresno Inyo Monterey Kings Tulare San Luis Obispo Kern Santa Barbara Los Ventura Angeles San Bernardino Orange Riverside San Diego Imperial Figure 3.1—Levels of Political Fragmentation in California Counties, 1992 of PFI tend to have a county-dominant governmental structure with relatively few sizable cities, and low reliance on special districts. As Table 3.2 illustrates, the 11 counties (or roughly 20 percent) that had the lowest fragmentation scores in 1992 spent 22.5 percent of their local nonschool expenditures through cities and special districts; whereas the 11 highest-fragmentation counties had cities and special districts responsible for 57.7 percent of such expenditures. As one might expect, 39 Table 3.2 Average Percentage Shares of Nonschool Expenditure in High- and Low-PFI Counties County Cities Special districts High-PFI Counties 42.2 34.4 23.3 Low-PFI Counties 77.5 9.8 12.7 57-County Average 58.0 22.3 19.7 SOURCE: Author’s calculations from Census of Governments data (1992). the low-fragmentation counties tend to have a relatively small portion of their population in incorporated cities: They average 77.5 percent unincorporated population, compared to just 16.7 percent unincorporated, on average, in the high-fragmentation counties. For a specific example, Table 3.3 compares two counties, Sonoma and Solano, both located in the northern part of the San Francisco Bay area. Sonoma has a larger population and more units of local government, so one might expect it to have a higher political fragmentation index. In fact, however, its PFI in 1992 was significantly lower than in nearby Solano, which had the second-highest level of Table 3.3 A Comparison of Local Political Structure in Two Counties, 1992 Population % unincorporated County government’s share of total spending, % Number of cities with >2% of total spending in county Number of special districts with >2% of total spending in county Political fragmentation index Sonoma 400,900 41.0 54.7 8 4 54 2 0.661 Solano 358,900 5.5 35.1 7 5 48 3 0.810 40 political fragmentation in the state. As the table shows, more of the cities and special districts in Solano have a substantial budgetary presence, with 2 percent or more of the total local nonschool expenditures in the county. Even more important in accounting for the different levels of PFI, in this case, is that Solano has a very small share of its population living in unincorporated areas, which means that the county government has a limited role as a local service provider. In other words, several units of local government in Solano have a major decisionmaking presence, and the county government is just one government among many, as opposed to the more dominant role the county has in Sonoma. Thus, by this measure, Solano’s local government structure is considered more fragmented, because expenditures are more evenly divided among governments. This tale of two counties illustrates some of the accounting that determines differences in PFI, but it does not show the underlying characteristics of counties that might explain those variations in PFI. That issue will be addressed in the next chapter. Did Proposition 13 Accelerate Fragmentation? One issue to be dealt with before examining the reasons for crosssectional variations in local fragmentation is the longitudinal issue. That is, have levels of PFI increased over time? Was there a noticeable and sustained blip upward in levels of PFI in the wake of Proposition 13? Recall from Chapter 2 that Proposition 13 did not have any obvious effect on the numbers of cities and special districts statewide, though it appeared to have altered the relative frequency of different types of special districts somewhat. 41 The PFI data for 1972 to 1992 are consistent with those findings. Change over time has been very slight, as Figure 3.2 shows. The lower line in the figure shows the average PFI values for the 57 counties in the Census of Governments years. It shows a slight dip in fragmentation in the period just before Proposition 13 and a gradual increase in the period since. However, this line is a bit misleading in that it averages all counties, from tiny Alpine to huge Los Angeles. The middle line solves this problem by presenting a population-weighted measure of PFI for the 57 counties. Finally, the top line of the graph shows PFI for the state as a whole—that is, with the state rather than counties as the unit of analysis. In each case, the lines are quite flat, indicating that fragmentation did not vary much over time. In particular, the period from 1982 to 1992 showed very little change in PFI. 1.0 0.9 Statewide PFI Population-weighted county PFI Average county PFI 0.8 0.7 0.6 0.5 0.4 1972 1977 1982 1987 1992 Figure 3.2—Political Fragmentation Index in California, 1972–1992 42 The lines on the graph support the conclusion that Proposition 13 did not result in a major, jarring shift in the financial responsibilities between counties, cities, and special districts. Nor does it appear that government formations after Proposition 13 led to marked changes in local political structure. Do these statewide measures conceal important differences among counties in their PFI changes over time? There is little evidence to believe so. Most of the counties held quite constant in their PFI indices over this period, generally moving by a few percentage points or less. Moreover, some counties increased slightly and some decreased slightly between each Census of Governments, with later changes in each county often canceling out the earlier changes. Some of the very-smallpopulation counties experienced greater shifts in political fragmentation but these changes were not uniform in direction. There is reason to expect greater shifts in small counties, since with few units of government, any addition or deletion of units leads to bigger relative changes in spending responsibilities than in more complex counties. The one region where there was some minor evidence of a trend was in the greater Los Angeles region, where as seen above the lion’s share of new city incorporations took place. Figure 3.3 compares the overall PFI for the five-county metropolitan Southern California region to the ninecounty San Francisco Bay area.6 It shows that there has been some convergence in the former region toward the high level of fragmentation ____________ 6This figure uses the metropolitan areas as units of analysis. The Southern California region includes Los Angeles, Orange, Riverside, San Bernardino, and Ventura Counties. The Bay area includes San Francisco, San Mateo, Santa Clara, Alameda, Contra Costa, Solano, Napa, Sonoma, and Marin Counties. The higher number of counties per capita in the Bay Area is one major reason why its collective PFI is higher than in Southern California. 43 1.0 0.9 0.8 0.7 0.6 0.5 0.4 1972 1977 Bay area Southern California 1982 1987 1992 Figure 3.3—Political Fragmentation Index in California’s Major Metropolitan Areas in the latter, with a particular increase in the 1977–1982 period. Still, Southern California was growing extremely rapidly in population throughout this period, and its PFI increase hardly seems unprecedented. Moreover, levels of PFI barely budged in the region between 1982 and 1992. As Chapter 4 will show, regression analysis indicates that neither the Bay area nor Southern California is distinctive in its PFI levels, once we have controlled for other relevant factors. Why the lack of change overall? As was shown in Chapter 3, although there have been numerous incorporations in California, most new growth has been accommodated by the gradual growth of existing cities. Thus, annexations increase the relative spending share of some established cities, even as new, generally smaller cities are added. Some special districts have been created, but others have folded or decreased in importance. Finally, counties have retained their major role, with great 44 programmatic responsibilities placed on them by the state and continued service provision in the unincorporated populations (although unincorporated areas are gradually declining as a share of the state’s population). A typical example of these patterns is Santa Barbara County. That county’s population grew steadily from 270,200 in 1972 to 379,400 in 1992, but its PFI values changed only slightly, increasing from 0.634 to 0.648. (The PFI actually peaked in 1982 before decreasing.) The county’s share of total nonschool local expenditures varied between 52 and 58 percent over this period but did not follow a consistent trend upward or downward. Two new cities were incorporated in the postProposition 13 era—Solvang in 1985 and Buellton in 1992, the latter too recent to show up in the 1992 Census of Governments. Both are very small municipalities, however, each with less than 6,000 inhabitants and less than 1 percent of total expenditures in the county. An overemphasis on these new units would obscure the fact that the area’s largest cities—Santa Barbara, Santa Maria, and Lompoc—each grew substantially in population and expenditures, and the county retained its dominant role. Special districts collectively accounted for between 12 and 17 percent of the area’s expenditures over this period, but no single special-purpose entity, with the exception of the Lompoc Hospital District, remained above 2 percent of the area’s expenditures. Thus, at least by this measure, the relative distribution of governmental activities stayed fairly constant over these two decades. Summary This chapter has offered a different and, it is argued, a superior method to measure local political structure as opposed to merely 45 counting governments or governments per capita. The political fragmentation index shows the division of governmental activities among local governments in each county area, using expenditure data as a proxy for each government’s activity level. California’s counties show a fair amount of variation on this PFI measure. High-PFI counties tend to be heavily incorporated, with numerous sizable cities and special districts. However, a look at PFI values over time does not indicate that California has become particularly more fragmented. Shifts in PFI have been gradual, both on the statewide level and within most counties. There is no substantial evidence to support a “Proposition 13 effect” on local government structure. 46 4. Accounting for Variations in Local Government Structure Across California The previous chapter shows that California’s counties differ extensively in their degree of local government fragmentation. This chapter examines the characteristics of counties that underlie these variations in political structure. Drawing on previous studies of local government structure, I identify some basic determinants of these patterns. The results indicate that a few, fairly deep-seated characteristics account for most of the differences among the counties in political structure. What Factors Might Affect Local Government Structure? A number of studies by social scientists have attempted to probe for the determinants of local government structure in the United States. But while growing, this vein of literature is still small and incomplete. As 47 Fisher and Wassmer (1998, p. 445) note, “Considering the theories that exist about local government structure, there has been very little empirical examination of what determines the structure of localities in an area.” The work that does exist typically attempts to explain variations in the number of local governments, across states, metropolitan areas, or counties. But such counting exercises, as noted above, can be misleading, particularly given the profusion of special districts in the Census of Governments that have little—and in some cases no— expenditure activity. This section reviews the literature on government formation and structure, looking for theories than can help us understand patterns of political fragmentation across California’s counties. Most of these studies involve statistical analyses that control for other relevant factors; some caveats are noted below. (Readers who are less interested in the theoretical ideas behind the analysis in this chapter may wish to skip ahead to the section entitled “Why Fragmentation?”). The literature review indicates that several sets of factors have been linked to local political structure. General Contextual Factors Not surprisingly, every relevant study finds that areas with larger populations tend to have more governments. In simple terms, there are more citizens demanding more things from the public sector, and thus the public sector can be expected to be more complex. One might also suspect, then, that areas adding rapidly to their population would experience more government formations. Two studies of city incorporations in California have found that “concerns about growth and 48 development” were among the most important factors in the formation of new cities (Musso, 1994, p. 2; see also Martin, 1984). However, somewhat puzzlingly, population growth is not statistically associated with larger numbers of governments—in fact, several studies find significant negative associations between population growth and government formation. Looking at the historical evidence presented in Chapter 2, we can begin to develop an explanation as to why this might be so. Simply put, the “boom” years for government formation, at least in California, were periods much earlier in the state’s history. Although recent decades have seen high rates of population growth, the number of governments has increased only relatively incrementally, above a base that was largely set decades ago. Thus, it would be wrong to say that rapid population growth “causes” a slower increase in the number of governments. Rather, we might hypothesize that government structure is heavily rooted in the past and reflects earlier periods of population change more than recent growth. Other general contextual factors that studies have included as predictors of local government formation include population density, which is thought to generate more complex public service needs, and/or measures of urbanization. On the thought that “a decentralized government structure contributes to efficiency only if residents are mobile and not restricted in employment choice by job location,” Fisher and Wassmer (1998, p. 452) also include square mileage and highway mileage in their estimations. Intergovernmental Rules and Relationships Another set of factors that have been identified as affecting local government structure might be called intergovernmental variables. 49 These include the rules that states set regarding changes in local government structure, funding and fiscal relationships among levels of government, and political conflicts at the state level. Since this study is limited to California, and most of these characteristics vary only across states, they are of limited applicability here; nevertheless, they are worth reviewing to understand how state governments set parameters for local government. For example, Nelson finds that states with more liberal home-rule authority for local governments have more municipalities and districts in their metropolitan areas, whereas states with minimum population requirements for the formation of new governments tend to have fewer local governments (1990, p. 453). States that make annexation easier experience more incorporations, probably because outlying settlements are more fearful of being annexed in such states (see Fisher and Wassmer, 1998, p. 454). States with easy enabling rules for special districts see more districts formed, as do those with spending and tax limits on general-purpose governments (Bollens, 1986, p. 122). States that permit a broader array of functions to be taken on by special districts also tend to have more districts (Foster, 1997, p. 137). State-level political conflicts, too, may be registered in local government structure. Nice (1983) finds a bivariate relationship between an index of urban/rural political conflict in a state and the number of local governments per capita in the state. He interprets this as an intentional attempt by rural-oriented state legislators to encourage the political fragmentation of major metropolitan areas. However, Marando and Reeves (1988, p. 1002), who examine state provisions for the formation of regional special districts, conclude that “as a rule there is no 50 grand design on the part of state governments to fragment metropolitan areas in order to diffuse political power.” Fiscal rules and conditions associated with a given state may also affect government formation. In particular, special districts may be formed in an attempt to shed the load that burdens municipal and county governments, or to enable local government to take on new activities that are prevented by the fiscal restrictions faced by existing units of government. In an analysis of Southern states and metropolitan areas between 1972 and 1977, MacManus (1981) finds that the creation of property-tax-based special districts is correlated with a lower rate of growth in the property taxation of other local governments.1 She writes, “Perhaps state officials, recognizing the dim likelihood of overturning archaic constitutional limitations on local government revenue sources at the height of tax revolt, ‘compensated’ for the limitations by being less restrictive in their policies toward the creation of special districts” (1981, p. 1214). Similarly, Nelson (1990) finds that metropolitan areas located in states with constitutional limits on taxation by cities have a greater number of special districts. This result was replicated at the state level of analysis by Leigland (1994, p. 540), although Foster (1997, p. 135) comes to the opposite conclusion in an analysis of metropolitan areas. On the issue of intergovernmental financial transfers, several studies have found that intergovernmental aid has a large and significant positive effect on spending levels by local governments (Schneider, 1989, pp. 60, 66). In a related finding by Stein (1987, p. 150), “compensatory state aid has a significant and negative impact on the variation in municipal service bundles within a metropolitan area,” perhaps indicating that ____________ 1The use of simple correlation analysis in MacManus’s study, and the generally weak correlations she finds, make it dangerous to generalize from the findings. 51 intergovernmental money enables areas to “level up” their service provision. Stein also notes that earlier studies found that such aid “has a negative effect on segregation of residential populations” (1987, p. 143), which could lead in turn to less sentiment for new cities, and thus less political fragmentation.2 Working in the opposite direction of these possible negative effects of intergovernmental aid on political fragmentation are the programmatic requirements that have been attached to some categories of intergovernmental money. Foster notes that “since the 1930s, federal funding for mass transportation, housing and urban renewal, natural resources . . . hospitals, and sewage treatment has often privileged specialpurpose over general-purpose governments.” Despite recommendations to the contrary from the Advisory Council on Intergovernmental Relations, “federal policies for provision of air quality, solid waste removal, airports, housing, and health services continue to encourage special-purpose governments” (Foster, 1997, p. 111; see also Bollens, 1986, p. 119). The terms of federal aid for water and irrigation facility construction also have been viewed as favoring special districts over general-purpose governments (Leigland, 1994, p. 537). Socioeconomic Characteristics of the Population One frequently suggested allegation is that new local governments are formed because of an effort by wealthier people to separate themselves into a more fiscally healthy jurisdiction—one with a high property tax or sales tax base and not heavily burdened with the service demands of the less privileged. Lending support to this line of reasoning, ____________ 2The next section will examine possible links between social/racial heterogeneity, segregation, and fragmentation. 52 Martin and Hogan (1989, p. 21), examining incorporations in California after Proposition 13, find that the new cities generally have higher per capita incomes than those of their surrounding counties. This finding was confirmed by Musso (1994, p. 186) in a comparison of unincorporated Census Designated Places in California that had attempted incorporation to those that had not. In case study analysis, Miller (1981) found tax avoidance and capture of locally generated revenues to have motivated incorporations of numerous cities in Los Angeles County even long before Proposition 13. Rather than being centered upon issues of efficient service provision, Miller argues, “the municipal incorporation of the Lakewood Plan cities was a ‘revolt of the rich against the poor’” (1981, p. 9; see also Martin, 1984, pp. 62–63, 94–95). Thus, high income levels, or a large share of high-income households, might be expected to be associated with higher levels of political fragmentation. Perhaps what matters, though, is not the absolute income level of the area but the variations in socioeconomic status—variations that might cause some areas to wish to separate from others. Indeed, economic theory suggests that areas with a great deal of social and economic diversity—or a “heterogeneity of preferences”—should see relatively more governments created. Drawing on Tiebout’s (1956) theory of a competitive market-like structure of local governments, most economists studying this issue expect “that greater variation in the economic characteristics of residents results in greater variation in both the quantity, quality, and mix of services demanded from local governments” (Fisher and Wassmer, 1998, p. 445). In turn, this varied demand is expected to translate into a propensity to create more units of government, other factors being equal. Thus, we might expect that areas 53 with populations that are more heterogeneous in income, race, and even age would be more politically fragmented. An alternative, and less charitable, theory might hold that wealthier or “whiter” communities prefer to exclude the poor and minorities by walling themselves off with political boundaries (see Weiher, 1991). In empirical studies related to this question, Nelson (1990) finds that variability in household incomes in a metropolitan area is associated with more general-purpose and special-purpose governments; he also finds, however, that variation in the age makeup of the population of these areas is associated with fewer special districts. Nelson elected to omit a variable measuring the racial heterogeneity of metropolitan areas, when he found in his initial analysis that racial diversity was associated with fewer units of local government—“a finding that is contrary to expectations . . . .” (1990, p. 452). Given the “lack of a strong theoretical basis” for this result, he chose to exclude the race variable, which may lead one to question the robustness of his income and age results. (Income and age variables were statistically insignificant when the race term was included.) Martinez-Vazquez, Rider, and Walker (1997) find racial heterogeneity associated with higher numbers of school districts—perhaps indicating a preference among voters for keeping schools more racially homogeneous. They find no association between racial heterogeneity and the formation of special districts, however. By way of explanation, they note that special district activities, unlike school districts, generally do not involve interpersonal interactions among residents. Given this seemingly reasonable logic, it is surprising that Fisher and Wassmer (1998)—confirming the result that puzzled Nelson—find that racial variation is associated with fewer general-purpose local 54 governments. They also find, however, that greater variation in the age makeup of metropolitan areas increases the number of school districts and of municipalities and townships.3 In a second paper examining longitudinal changes in numbers of local governments from 1982 to 1992, Wassmer and Fisher (1997, p. 15) find in general that “increased variation in the socioeconomic composition of a metropolitan area leads to more local governments . . . .” Political Influence of Developers or Other Businesses A number of authors have linked the creation of local governments to the influence of developers or other major economic actors in the political process. In particular, “Criticism has been directed at local special districts for their . . . use of public agencies as development tools by private interests” (Bollens, 1986, p. 118). Much of this hypothesis has rested on evidence from case studies. For example, Perrenod (1984) tied the extreme proliferation of water districts in the Houston area to overt efforts of developers to cheaply create infrastructure for each new subdivision. Lewis (1996, pp. 97–100, 142–144, 148) found that large-scale developers in the Denver area viewed the formation of special districts as simply another business deal that often was necessary to carry out their visions for suburban development. Special districts in Colorado have often been the only practical way to build infrastructure and service ____________ 3Fisher and Wassmer merge the counts of municipalities and townships, claiming that both fulfill similar roles in the federal system and thus can be considered collectively (footnote, p. 450). This claim is debatable. Townships are little more than administrative areas in some states, whereas they perform municipal service functions in other states. Treating townships and municipalities as equivalent may thus bias a count measure of political structure. The PFI measure used in this report avoids this problem. In any event, California has no township governments. 55 development in unincorporated areas. They have been turned to increasingly within cities as well, as cities became less willing to underwrite the costs of growth, preferring to offload these costs onto homebuyers. At least two quantitative studies have also attempted to link developers to government formation. Burns (1994) notes that efforts to create new governments are collective action problems that typically require organizational and financial resources, as well as leadership. Particularly in the case of districts, Burns argues, “the most obvious entrepreneurs [working to form governments] . . . are developers who want government to build infrastructure to improve land to profit the developer” (1994, p. 20). Her empirical findings suggest that the number of developers in a county is positively related to special district formations, but only for the decade of the 1960s (1994, p. 99). In her national study of special districts, Foster (1997), noting the case study literature that focused on the role of developers, includes a variable measuring the number of developer-firm equivalents per 10,000 population (based on data from County Business Patterns). Her quantitative results ascribe no major explanatory role to developers, however. Aside from development firms, the role of manufacturers in the founding of cities has sometimes been emphasized. The industrial era in American history provides a number of well-known examples of “company towns”—such as Pullman, Illinois—formed at the behest of commercial interests. Miller (1981, Chap. 2) provides case studies of several cities in Los Angeles County in the 1950s—such as Commerce, Industry, Irwindale, and South El Monte—that were formed under the 56 leadership of local industrialists, who often drew boundaries carefully to include a minimal residential population. These incorporations helped keep taxes low and regulations on industry lax. In her statistical analysis, Burns (1994, Chap. 5) does find that under certain circumstances, the presence of manufacturers in a county leads to a significant growth of municipal governments. She suggests that the ability of new cities to offer a tax haven for business is the explanation of such effects.4 Why Fragmentation? Toward a Model of Local Government Structure As we have seen, existing theories and empirical studies converge on four basic sets of explanations for local political structure: general contextual factors, including the size and maturity of the county; intergovernmental factors, which are not applicable to this California-only analysis, except for intergovernmental aid; socioeconomic characteristics, particularly the variations in the population across income, race, and age categories; and perhaps the political influence of developers or manufacturing firms. These factors involve numerous variables of potential significance, and some difficulties of specification, particularly for the developer/manufacturer concept. Given constraints of data availability, I collected data for California counties on a variety of variables, listed in Table 4.1, to explore their ____________ 4However, Burns’s use of a dummy variable for the presence of a manufacturing establishment in the county seems pointless for contemporary California, in which every county but one includes manufacturers, according County Business Patterns data. Instead, I use a measure of manufacturing establishments per 10,000 population in the analysis below. 57 58 Table 4.1 Variables Studied for Possible Relationships with Local Political Fragmentation Type of Variable General/contextual Intergovernmental Socioeconomic Interest group Variable Population Density Land area Urbanization Historical development Population growth Aridity Geographic diffusion Transportation network Intergovernmental revenues Age Income Homeownership Race/ethnicity Developers Manufacturers How Measured Natural log of 1992 population Natural log of 1992 population per square mile Square mileage of county, 1990 % of population rural, 1990 Metropolitan area (dummy variable) 1920 population as % of 1990 population 1950 population as % of 1990 population % population growth, 1972–1992 Annual precipitation (inches) Average PFI of surrounding counties Southern California metropolis (dummy variable) San Francisco Bay area (dummy variable) State highway miles per square mile, 1995 % of local revenue that was intergovernmental, 1977 % aged 65 or older, 1992, 1977 Index of variation across Census age categories Per capita income, 1989 % of households with income above $75,000, 1989 % households in poverty, 1989 Standard deviation of income across Census categories % of occupied units owner-occupied, 1990 % black population, 1992, 1977 % Hispanic population, 1992, 1977 Index of variation across racial/ethnic categories Development firm equivalents per 10,000 population, 1977 Manufacturing firms per 10,000 population, 1977 Bivariate Association with PFI + + Insignificant – + Insignificant Insignificant Insignificant – + + + + – – Insignificant + + – + – + + + + Insignificant 59 association with the political fragmentation index in 1992.5 All of the data are from standard governmental statistics.6 Most of these variables are suggested by the previous studies in this literature. However, I have also tested the effects of a number of additional variables that I considered to be potentially influential for the development of local political structure in California. For example, I included two dummy variables for county location in the San Francisco Bay area or in metropolitan Southern California to attempt to capture any patterns specific to those areas. Taking this concept of geographic diffusion a step further, I also developed a variable representing the average PFI value for the counties bordering each county, on the thought that a “norm” of governmental structure might have evolved among neighboring areas. Another variable examined was average precipitation in each county; the idea here was that more arid counties might need to develop more involved water-service provision systems, thus leading to more special districts and more governmental fragmentation. ____________ 5Initially I intended to use a pooled sample looking at changes in each county’s PFI level between each issue of Census of Governments. However, as discussed in Chapter 3, changes in these levels for individual counties are so minor and seemingly random across time that I decided to use a simpler cross-sectional design examining PFI in the 57 counties as of 1992. Some of the independent variables use lagged values from 1977, on the assumption that government structure responds only gradually over time to these variables, and to avoid potential problems of endogeneity. 6Population estimates, including ethnic/racial breakdowns, as well as statistics on county land area, are drawn from reports of the California Department of Finance’s Demographic Research Unit. Highway mileage data are from the California Department of Transportation. Precipitation data are from the National Weather Service. Intergovernmental revenue figures are from the Census of Governments. Data on the number of developers and manufacturing firms are from County Business Patterns. My construction of the development-firm index follows Foster (1997, p. 146). All other data are standard Census Bureau demographic variables from the Census of Population and Housing or the County and City Data Book. 59 Bivariate Relationships Because there are many variables of potential interest and only 57 county observations, degrees of freedom limit the number of regressors that may be introduced. Before proceeding to a multivariate analysis where most of these variables must be excluded, it is interesting to examine the simple bivariate relationships between these variables and political fragmentation. For this reason, the last column of Table 4.1 indicates whether each variable has a significant bivariate relationship with PFI, in a positive (+) or negative (–) direction. At this simple level of analysis, many variables appear related to political fragmentation, generally in the anticipated direction. Variables that appear positively associated with the PFI in California counties included measures of county population size and density, urbanization, location in the Bay area and Southern California or proximity to other high-PFI counties, income levels and heterogeneity, racial diversity, and the heavy presence of development firms. By contrast, annual precipitation, intergovernmental revenue, percent elderly population, owner-occupied housing, and poverty populations appear negatively linked with political fragmentation. Multivariate Model I next entered various subsets of these variables into multiple regression equations intended to assess their relative significance and influence upon political fragmentation. Numerous combinations were tried in specifying such models, but ultimately a very simple model with three estimator variables proved remarkably robust. With one exception, none of the other variables listed in Table 4.1 achieve or even approach statistical significance, once these three variables are included. 60 This analysis reveals that population size, intergovernmental revenue share, and historical growth trajectory are the main determinants of political structure in California counties that can be identified—in that order of importance. Population size is positively associated with political fragmentation, as is a growth trajectory in which a county achieved more of its ultimate population size earlier in its history. Counties in which the local public sector received a larger share of its funding from state and national sources were less fragmented politically. Table 4.2 presents the results of this model, which accounts for two thirds of the variation in PFI across counties. It also presents a similar second model in which the racial variation of the county population is introduced in addition to the other three terms. Racial variation here is measured with an index analogous to the PFI; it represents the probability that two randomly selected residents of a county are of different ethno-racial categories.7 This variable, which has a positive sign, was the only other variable to attain statistical significance (at the 0.10 level) once the three variables above were included.8 However, the effect of racial variation on fragmentation is not robust to alternative specifications of the model. When other county characteristics were entered into regressions—variables such as owner-occupancy of housing, rural share of population, age variation, or development-industry size— the racial variation measure was no longer significant. By contrast, the other three determinants of fragmentation identified above—population, ____________ 7The categories used are non-Hispanic white, non-Hispanic black, Hispanic, Asian and Pacific Islander, and Native American. 8The application of stepwise regression analysis—a method rightly criticized because of its atheoretical nature—also results in this four-variable model. 61 Table 4.2 Regression Models of Political Fragmentation Indexes for California Counties, 1992 Variable Population, 1992 (natural log) Model 1 0.081 (9.78) [0.835] Model 2 0.070 (6.92) [0.726] Intergovernmental revenue, 1977, as share of total revenue –0.006 –0.006 (–3.26) (–3.65) [–0.256] [–0.290] Historical growth (1950 population/1990 population) 0.002 (2.27) [0.195] 0.002 (1.86) [0.161] Racial/ethnic variation index (ranges from 0 (low) to 1 (high)) — 0.204 (1.73) [0.169] Adjusted R-squared 0.66 0.67 Number of cases 57 NOTES: Dependent variable is political fragmentation index, 1992. Regression coefficients for each variable are listed first, with tvalues in parentheses and standardized (beta) coefficients in brackets. San Francisco City and County is omitted. 57 intergovernmental revenue, and historical growth—never failed to achieve statistical significance in numerous specifications of the model.9 ____________ 9Removing the five smallest counties in the sample (those with populations under 15,000 in 1992) does not change the general magnitude, direction, or significance of the three main variables, but once again the racial variation variable falls to insignificance. Omitting Los Angeles County from the sample (by far the largest county in the state, and certainly somewhat unique) leads to virtually the same results listed in Table 4.2 but with a slightly higher R-squared. 62 Interpretation What can we make of these results? The finding that larger- population areas are more politically complex is not surprising, and conforms to previous studies. Bigger, more complex counties simply have more complicated public sectors. Larger populations increase the likelihood of additional governmental units; additional units will tend to increase PFI. Turning to intergovernmental revenue, the potential political and policy implications are more interesting. One possible interpretation is that a greater flow of revenues from nonlocal sources means that there is less need for local public sectors to engage in “creative financing” to meet their revenue needs. Thus, fewer special districts may be formed. Similarly, existing county and city governments are likely to be in healthier fiscal situations if there is more intergovernmental funding available (all other things equal). For this reason, residents of wealthier areas within the county may have less of an incentive to incorporate as a separate city, preferring instead to be under county rule (as an unincorporated area) or to be annexed by an existing city. Interpreting the influence of the historical growth trajectory is also interesting. Note that the variable used here is simply the reverse of a measure of population growth. As some scholars have done, we might conclude that areas that have grown faster in recent decades tend to be less fragmented. But there are no persuasive theoretical reasons to draw such an interpretation. Rather, I prefer to think of this measure as representing the maturity or growth history of the county. The variable represents each county’s 63 1950 population as a percentage of its 1990 population.10 The results indicate that for every 10 percent of a county’s 1990 population that it attained before 1950, its PFI would be about 0.02 higher. Although modest, this effect is consistent with the notion that most of California’s local government development occurred during earlier periods in the state’s history. In other words, earlier waves of population growth in the state’s history generated more complex structures of local governance, whereas more recent periods of growth resulted in more streamlined local political development. This is a more theoretically compelling explanation, given the historical evidence on the incorporation of cities in Chapter 2. Other studies have hinted in this direction, noting that municipalities—even small towns—created in earlier periods of history had more full-service responsibility placed in their hands. This would imply more dispersion of government responsibilities among units, and thus a higher PFI. As Schneider wrote, “In the parts of the United States settled earlier, local governments traditionally have more responsibility than in newer parts of the country, but as time passed, the expected role of local government narrowed. The distribution of functional responsibility is thus intimately tied to the historical timing of settlement” (1989, p. 49). Similarly, Burns (1994, p. 9) found that over time, the percentage of U.S. cities offering a range of services, particularly firefighting, water service, and sewerage, has greatly declined. In California, particularly, later-forming cities are more likely to be “contract cities” and thus to take up narrower shares of total local expenditure in the county. Martin and ____________ 10An alternative measure, the county’s 1920 population as a proportion of its 1990 population, yielded very similar results, with a slightly worse overall goodness-of-fit. 64 Hogan (1989, p. 44) found of new cities in the post-Proposition 13 era that “municipal government is limited government: limited in terms of resources and limited in terms of flexibility. New city government appears also to be limited in terms of function.” This account of the history of local government indicates that areas may become “locked in” to early patterns of political development. Fisher and Wassmer (1998, p. 448) wrote, “Despite factors that favor change, a metropolitan area may be unable to alter the set pattern of earlier development [of local government structure].”11 Thus, theories of “path dependence” may provide a useful way of understanding variations in local government structure. “In the case of path-dependence theory, the potential for policies to self-reinforce and become entrenched presents the possibility of policy lock-in . . . . Early events have primary importance,” writes Woodlief (1998, p. 428). Although Woodlief sought to explain differing budgetary choices among large cities, his analysis seems to relate as well to studies of local government organization: “History matters and is cumulative. Early choices can shape later options. Time and the order of events are important contextual variables” (1998, p. 432). ____________ 11However, in their study of U.S. metropolitan areas, Fisher and Wassmer use a measure of age that is a dummy variable signifying whether the central city was incorporated after 1950. They find, contrary to the results here, that “newer” areas, using this measure, have formed more units of general-purpose government. It is unclear how this result may be reconciled with the results presented here. One possibility is that the incorporation date of a central city does not really measure the maturity of an area. It is also possible that metropolitan areas with such late formations of central city governments experienced more municipal formations in earlier periods as “substitutes” for an absent central city government. Or it may indicate that these areas were less urbanized and thus tended to form more small-town governments rather than larger urban and suburban municipalities. 65 Other Variables The racial variation variable in Table 4.2 yields an interesting tentative conclusion: Counties with more racially diverse populations tend to be more fragmented. If true, this result would be consistent with two possible stories about race and government structure. The more charitable account is that a more varied population has more heterogeneous tastes for services and thus creates a more variegated governmental structure. The less comforting interpretation is that residents seek to wall themselves off in separate jurisdictions (armed with land-use powers, in the case of cities) more often in more diverse counties, in search of greater racial and ethnic separation. However, the weakness of the racial variation variable in models— that is, its lack of statistical robustness discussed above—means that we should not make too much of these possible explanations. Simply put, we cannot reject the hypothesis that racial variation has no independent effects on political structure. Also worth noting is the persistent lack of significance of any of the remaining variables, once population, historical growth, and intergovernmental revenue are controlled for. For example, despite much suggestive support in the literature, the variables relating to socioeconomic status and development-industry strength lack explanatory power.12 ____________ 12Various interaction terms were also introduced into regression models, on the thought that two factors might act jointly to affect political fragmentation. For example, percent wealthy can be interacted with percent poverty or percent black to simulate the desire of more advantaged segments of the population to separate themselves politically given the presence of larger disadvantaged populations. Following Burns (1994), the developer index was also interacted with percent nonwhite population. None of these variables yielded interesting results. 66 In this case, the “supply side” of government revenues and service provision—the share of local government funds coming from higher levels of government—appears to play a more important role in affecting local government structure than the citizen “demand side”—the kind of population seeking services. Of course, a larger sample might allow for more complex estimations in which other relationships of interest emerge. Summary This chapter plumbed the small existing literature on the causes of local government structure to propose possible explanations for the varying patterns of fragmentation in California’s counties. The literature review suggested four general sets of possible explanations, including general contextual variables, intergovernmental issues, socioeconomic diversity, and business influence in local politics. Statistical analysis of the California data indicates a relatively simple and powerful account for differences in local political structure around the state. Areas that have larger populations and which developed earlier in the state’s history—when the bulk of the political structure was being established and cities were more likely to be full-service governments— tend to be more fragmented. In addition, areas receiving more of their revenue in the form of intergovernmental aid are less fragmented, perhaps because they have less need to set up new political structures to overcome local revenue shortfalls. There is weaker, more inconsistent evidence that greater racial variation in a county is associated with more political fragmentation. Once we control for these factors, none of the other variables, including those related to socioeconomic status and business influence, appear to matter in a significant way. 67 5. Conclusions and Policy Considerations Local political structure is important because it can have substantial effects on both the process of policymaking and residents’ quality of life. Although the existing research literature is hardly definitive, it tends to suggest several notions about the effects of local government complexity. Fragmented governmental structures (particularly competing municipalities) may promote efficiency in the delivery of some routine public services. But fragmentation also may detract from the ability of areas to manage regional land use and development, to address needs of the regional economy, or to reduce disparities between different communities. This study has not concentrated upon the effects of local government fragmentation, although Chapter 1 and Appendix A summarize the relevant literature and distill its important lessons. Rather, the goal of this report was to portray California’s structure of local government, to 69 document its degree of fragmentation, and to determine whether the local structure has changed markedly over time. These issues are important to policymakers because of the state’s repeated waves of concern over the supposed proliferation of local governments. After the passage of Proposition 13, commentators pointed with alarm to numerous city incorporations and special district formations and saw portents of a great increase in governmental fragmentation. More recently, several reform proposals aimed at rationalizing or regionalizing local government emerged in the late 1980s and into the 1990s. And 1997 saw passage of a bill creating a new study commission to consider options for changing state laws that affect local government structure. Review of Findings Data from both state and national sources reveal that California’s local governmental structure is indeed quite complex. Nevertheless, it is not abnormally fragmented, by U.S. standards, nor is it highly unstable. Looking first at counts of governmental units, we saw that cities experienced their major growth in numbers in the period before World War II, with this growth rate declining in the period since. Proposition 13 did not noticeably alter this trajectory of change, although 54 new cities have sprung up since passage of that initiative in 1978—mostly in rapidly growing Southern California. Growth in the number of special-district governments has been relatively flat since the 1970s, although the universe of special districts in the state has changed somewhat. Districts that exist to finance the construction of capital facilities have increased in number, whereas those providing more traditional public services have declined somewhat in 70 number. “Governments created by governments”—particularly county service areas and joint powers agencies—have increased, which tends to segregate funds for certain government functions from general-fund budgets. Overall, California has far fewer counties, municipalities, and special districts per 100,000 population than the national average. And in the distinctive realm of education policy, the number of school districts has decreased markedly in recent decades. Using the specially constructed political fragmentation index (PFI) in Chapters 3 and 4, we were able to see in a more nuanced fashion the ways in which responsibilities and activities are divided among local government units in each California county. Analysis of the PFI data revealed that counties differ quite markedly in their local government structure. It also confirmed the finding that political structure is only very gradually growing more complex, and does not seem to have been decisively altered by Proposition 13. Data analysis also confirmed the underlying stability of these varied patterns of local government structure throughout the state. The degree of governmental fragmentation appears to largely be a product of two “deep” underlying factors—the county’s population size and its historical pattern of population growth. Simply put, bigger and older areas tend to be more fragmented. A third main influence on local government structure, one more susceptible to policy intervention, is intergovernmental aid. Areas that have received a higher share of their funds from state and federal sources tend to acquire less governmental complexity. There is also some weak and inconsistent evidence that racial diversity is associated with greater fragmentation; otherwise measures of socioeconomic status and heterogeneity do not have 71 significant independent effects on local government structure in these data. This is true even though we might expect a more varied population to lead to more residential sorting and pressures for separate governments. In addition to informing an ongoing policy debate, the findings of this study lay groundwork for additional research. In future studies, scholars may wish to determine the effect of fragmentation on various aspects of social and governmental well being in California counties. For example, do counties that are more politically divided tend to grow more segregated by race or income? The sorting and exclusionary-zoning aspects of suburban fragmentation suggest that this might be an outcome. Are politically fragmented counties more predisposed to fiscal stress? The partitioning of the landscape into separate taxing and spending regimes suggests that some jurisdictions may be more likely to experience budgetary strain. Has political structure affected the ability of areas to integrate their transportation investments with land-use plans? Again, fragmentation may impede such coordination. Researchers may also wish to replicate national studies by seeing whether municipal competition in California has resulted in less public spending or whether special-district reliance boosts overall spending. Relative Stability in an Era of Change—Why? Some might view these findings—that local government structure in California has been relatively stable—as a “nonfinding.” Rather, they are quite remarkable results, given the enormous changes that have occurred in California’s economy and population in recent decades. It is worth asking why local governmental change has been so gradual. 72 One possible reading of this record is that state policy in this area has worked. Recall that the legislature passed a law in 1963 requiring the creation of LAFCOs in each county. LAFCOs have since become important, if not always particularly popular or visible, players in regional politics. They must review proposals for the formation of new cities and special districts and may discourage the creation of new units, particularly where a city formation might create fiscal burdens for existing units. Under rules developed by the state legislature, LAFCOs have designated “spheres of influence” for each existing city. Such an approach makes a presumption in favor of expanding current cities rather than creating new ones. In addition, with their voting board made up of representatives of existing local governments, LAFCOs might be expected to be interested in preserving the viability of existing units, by steering population growth into cities and allowing annexations.1 In a few counties, notably Sacramento, LAFCOs have bucked the statewide trend toward centering growth in cities by instead supporting county policies that have permitted much growth outside of city limits. In any event, the fiscal constraints created for local governments by Proposition 13 may have given LAFCOs all the more reason to take a long, hard look at proposed government formations. Despite the expected presumption of LAFCOs in favor of existing units, one detailed study found that at least through the 1980s, these agencies rarely thwarted incorporation moves overtly. Politicians on ____________ 1LAFCO boards generally have five members: two county supervisors, two representatives selected by the county’s cities, and a fifth member selected by the other four. Several LAFCOs, however, provide for special district representation, and a few have been authorized by the legislature to vary their membership. See Assembly Committee on Local Government (1996), pp. 41–50. 73 LAFCO boards had “much to lose if they denied the apparent will of the voters” (Musso, 1994, p. 91; see also Martin, 1984, Chap. 6). Still, Musso determined that the pattern of incorporations did not follow any logic of fiscal maximization. On the basis of case studies of incorporation proposals in Orange County, she suggested that “many residents viewed additional government—any government—as a nuisance, even if it increased services at no cost to them” (p. 198, emphasis in original). Thus, popular opinion may also be resistant to an increased number of governments, tending to reinforce the gatekeeper role of LAFCOs. Finally, in 1992, the state legislature provided new instructions on city formation to LAFCOs by passing SB 1559, a law seen as tightening up incorporation requirements (Musso, 1994, p. 55). The law, passed in response to county complaints about strategic city incorporations that drew valuable property and sales tax revenues away from counties and into the coffers of craftily drawn cities, instructs LAFCOs to approve incorporations only where they will be fiscally neutral for the affected county and other local governments. In other words, a LAFCO may approve the formation of a new city only where the revenues that the new city can be anticipated to receive are roughly equal to the new city’s expected expenditures on services.2 In the wake of this restriction, there have been only three incorporations in the period since 1992. Beyond the role of LAFCOs, another possible explanation for relative stability in California is that existing local governments may be more malleable and responsive to change than is commonly supposed. ____________ 2The new city may avoid this restriction if it negotiates in advance to provide fiscal concessions to its county or in cases where the county and other local governments are in consensus to allow the incorporation. 74 Throughout the state, cities and counties typically have been able to address new issues, empower new ethnic constituencies, and respond to economic and population growth without falling into crisis. In response to growing suburbanization, annexations have been used to gradually extend the scale of cities to the expanding fringe. Meanwhile, residents of unincorporated areas have numerous formal and informal governance mechanisms available to them to collectively address the future of their communities without incorporating a city government (see Sokolow et al., 1981). Considerations for State Policymakers Overall, then, the division of responsibilities among local governments in California seems relatively stable, and the state has considerably fewer governments per capita than other states. Local boundaries, powers, and governing relationships have not been subject to jarring changes, even as the state and its localities went through wrenching fiscal challenges in the aftermath of Proposition 13 and major recessions. Does that mean that state and local policymakers should be complacent about California’s local government system? Not necessarily. Local fragmentation in California continues to separate fiscal resources from social needs and to impede coordination on regional issues. Nor should this study’s description of a modest number of incorporations since Proposition 13 be read as a call for large-scale increases in new cities. Whereas formations of minimal, “contract” cities may not fragment service provision or financial responsibilities in a county much, they do create new units with the “police powers” to zone land and thereby affect the development of California’s regions. Indeed, land-use 75 motivations have been among the most prevalent reasons for the formation of new cities, in California and throughout the United States (Musso, 1994; Martin, 1984; Gainsborough, 1990; Burns, 1994; Lewis, 1996; and Teaford, 1997). The effects of political fragmentation upon land use have been viewed by many as resulting in uncoordinated growth patterns and a lack of attention to regional problems and needs, such as affordable housing and the transportation demands arising from development. Despite California’s home rule tradition, tinkering with the local political structure ultimately remains the state’s prerogative. The courts have held that “it is well established that the state may create, expand, diminish, or totally abolish municipal corporations with or without the consent of its citizens, or even against their protest” (Scuri v. Board of Supervisors, 1982, quoted in Detwiler, 1996b, p. 6). The state’s recent creation of a Commission on Local Governance for the 21st Century underscores its capabilities in this area. Encouraging the state to review the local system, an earlier study group created by the legislature, the California Constitution Revision Commission (1996), proposed the formation of “citizens’ charter commissions” in each county (or multicounty area). These commissions would be required to study “local government structure, methods of service delivery, and assignment of responsibilities and powers.” After a study period, an adjusted plan of service delivery and financing for the area would be proposed, and put to the voters as a new “home rule community charter.” This plan could include “methods for reducing the number and cost of local government,” a capital improvement program for all the government entities in the area, and provisions for reorganization and boundary changes, if necessary. The state would 76 assist by creating a commission to act as a clearinghouse for ideas on local reforms and to monitor local efforts. Upon approval of the new home rule community charter, all participating local entities would be granted home rule powers and protected from the reallocation of their property tax and other general-purpose local taxes. Although the commission’s idea for countywide review and reform is no guarantee of improved political structure, the plan merits serious consideration. It would compel local governments to reconsider regional arrangements and the area’s future needs, and raise a dialogue in each area concerning local governance. Moreover, its home rule guarantees would provide incentives for cities, counties, and special districts to “buy in” to the process. The commission’s proposal does, however, appear to proceed from a presumption that fewer governments would be better, and cheaper, than the current arrangement. Most of the limited evidence on government reorganization, however, offers little hope that costs will be cut or taxes lowered due to any restructuring. Reorganizations should be premised on the search for a policymaking system more likely to address problems that currently remain unresolved, rather than as part of a search for a Utopia of structural simplification. Where the problems of local governance are primarily fiscal ones, organizational reform is not necessarily the solution. Some would prefer more far-reaching revisions of local government structure. However, fundamental change in this realm—such as through wholesale consolidations of cities or special districts—is unlikely and politically difficult at best. As one national study of local government pointed out, “Talk of the need for sweeping and fundamental reform of state and local governments is common, particularly among those who periodically ‘rediscover’ the states and localities. More experienced and 77 influential reformers are aware of the remarkable resiliency of the basic features of the system in the face of a century of sweeping social, economic, and political change, and they set their sights accordingly” (Danielson, Hershey, and Bayne, 1977, p. 60). This observation accords well with the facts of the California case, where, in spite of major transformations in the state, the local government structure has changed only gradually over a period of decades. To be sure, serious study of the problems of this governance structure is always warranted and reform proposals are worth entertaining. Nevertheless, policymakers who identify problems will be well served working at the margins. They can attempt to alter the incentives and fiscal burdens and resources of California’s localities; more generous intergovernmental aid, as we have seen, appears to reduce the impulse toward fragmentation. State policymakers also can work to strengthen regional political institutions that might repair some of the seeming deficiencies of fragmentation. In short, California’s leaders can work within existing structures to strengthen local governance and public finance rather than proposing to fundamentally change a stable and resilient system. 78 Appendix A Research on the Effects of Local Government Fragmentation In recent decades, scholars of political science, economics, policy analysis, and public administration have given considerable attention to the issue of local government fragmentation—for good reason. Political boundaries separate the landscape into varied taxing and servicing regimes, with implications for both equality among residents and jurisdictions and the efficiency of government operations. Normative essays and empirical research have isolated both potential advantages and disadvantages of local political complexity, with conclusions deriving in part from the analysts’ attachment to different values. For example, the preeminent value that many economists attach to allocative efficiency contrasts with that of political scientists, who are often more concerned with equity and effective political representation. Some scholars, particularly those associated with the public choice school in economics and political science, have emphasized the efficiency 79 gains thought to result from governmental multiplicity. Drawing upon the theory initially set forth by economist Charles Tiebout (1956), advocates of governmentally fragmented areas argue that citizens and firms in such regions can “vote with their feet” to find the jurisdiction that best meets their tastes for service and tax levels. In such a model, competition among jurisdictions helps keep governments leaner and less susceptible to “budget-maximizing bureaucrats.” Overlapping units can each supply specific services to local areas at the most optimal scale economy. Small-scale governments are seen also as less insulated, closer to home, and thus easier for citizens to control. Other analysts paint a less rosy view of political fragmentation. They make the case that areas that are more politically divided often have a more difficult time addressing the “general welfare”—in particular, regional land-use issues such as urban sprawl, housing opportunity, and downtown preservation. Regional strategies for economic development may be sacrificed to beggar-thy-neighbor growth schemes. Politically fragmented areas may also be more segregated by race and income. Moreover, fragmentation is alleged to reduce local government’s ability to aid the needy, since businesses and wealthier residents will find it relatively easy to flee to a nearby jurisdiction that is less generous with their tax money. And public-sector accountability may be hindered because the complexity and multiplicity of government can make it difficult for citizens to identify and hold responsible the appropriate officials. The actual results of empirical studies of the effects of fragmentation are mixed. On the issue of public spending, some studies have found lower per capita costs for certain services in more fragmented areas, whereas many others have come to opposite conclusions; still others find 80 no statistically significant effects. The differing results are traceable in part to the different services examined, the different control variables and multivariate methods used, and differing measures of political fragmentation. (See, for example, Baird and Landon, 1972; Isserman, 1976; Schneider, 1989; Zax, 1989; Forbes and Zampelli, 1989; and Oates, 1989.) There is a reasonable degree of evidence that a greater number of competing municipal governments in an area is associated with less overall public spending by municipalities (Zax, 1989). The one study looking at city incorporations in California in relation to the efficiency issue finds an increase in administrative expenditures after incorporation but potential efficiency gains due to the greater homogeneity of municipal populations (Musso, 1994). By contrast, a profusion of special district governments in an area has been associated with higher local government revenue burdens (Zax, 1989). Regarding spending, the most important recent study of special districts finds relatively convincing evidence that reliance on specialpurpose governments for service provision increases the absolute amounts spent on district-provided services. Such reliance also increases the share of total local expenditure going to district-provided services (Foster, 1997).1 On the other hand, reducing the number of districts that offer a particular service in a region may decrease competition and therefore lower efficiency in that sector. Several instances of consolidation of special districts in California since Proposition 13 were examined by ____________ 1In preliminary statistical analyses of California data, not discussed in this report, I also found evidence supporting the proposition that counties relying heavily on special districts tend to spend more per capita, controlling for income, urbanization, population size, and other relevant factors. 81 Morgan and Chapman (1994), who found that overall revenues and expenditures generally increased after mergers. A complicating factor in all studies of local government taxing or spending is that differing “tastes” for service levels in different regions may affect expenditures per capita. In other words, more spending does not necessarily imply a less “efficient” public sector. More spending may instead mean that governments are responding to local desires for enhanced services. Fiscal coordination in an area may suffer due to a profusion of local governments. Baldassare (1998) found that Orange County’s recent financial crisis was aggravated, and its subsequent recovery greatly complicated, by the array of local governments that had put funds into the county investment pool. These entities guarded their local advantages jealously and were unused to communicating with each other. On the issue of city/suburb equity and disparities, disputes over research methods have rendered the role of political fragmentation unclear. Morgan and Mareschal (1996), in preliminary results, found that fragmentation has few measurable effects on city/suburban inequality measures and may in fact be associated with greater fiscal health of central cities. Earlier, an oft-cited book by Rusk (1993) found fiscally and socially healthier conditions in “elastic” central cities—those able and willing to annex their suburbs regularly, whereas Savitch et al. (1993) found positive connections between central-city well-being and suburban well-being. The methods and theory driving these latter two studies were relatively rudimentary, however, and have been subject to serious criticism and reinterpretation by Blair and Zhang (1994), Blair et al. (1996), and Hill et al. (1995). Looking at who pays for and benefits from government services, Hawkins and Hendrick (1997) found that 82 services provided by counties and by regional special districts—in both cases, governments that included both well-off suburbs and a poorer central city—tended to disproportionately benefit city residents and reduce disparities. One might read this as a defense of fragmentation (since the study shows how overlapping governments can promote equity) but note that the county and regional special districts represent a particular type of area-wide governmental arrangement. Relating equity to land use, Weiher (1991), relying on both case studies and multivariate analysis, concluded that fragmentation exacerbates racial and socioeconomic segregation in metropolitan areas. Migrating households, in Weiher’s model, are able to use boundaries and jurisdictions as information signals in evaluating communities’ residential status, school district quality, and racial makeup. In another study relating to land use, Lewis (1996) found metropolitan political fragmentation associated with weaker dominance of central city downtowns and more of a patchwork pattern of development, with evidence that jobs and housing may be more geographically separated in fragmented areas. Weaker evidence existed for the hypothesis that fragmentation led to greater degrees of sprawl, or residential decentralization, when other relevant factors were held constant. Local political structure has even been studied for its relationship to the quality of political journalism. In a perceptive study, Kaniss (1991) blamed suburban political fragmentation for reducing the capacity of the media to provide meaningful news coverage of the public affairs of a region. Public issues are segmented among so many low-visibility suburban governments that regional newspapers and broadcasters find it difficult and unrewarding to cover regional governmental trends and public issues—especially development issues—outside the central city. 83 This brief review only sketches the major outlines of research on the effects of political fragmentation. Although this literature hardly speaks with one voice, it does illustrate the potentially key importance of local government organization. In Chapter 1, I highlight some of the major policy implications of fragmentation for four broad roles of the local public sector. 84 Appendix B Issues Regarding Special District Data in the Census of Governments The Census of Governments, which is used as the major data source in this report, includes a substantially different enumeration of special district governments in California than does the State Controller. The Census Bureau counts only politically freestanding special districts, not those dependent upon or supervised by a “parent” entity. In short, the bureau recognizes only those districts with “substantial administrative and fiscal independence” (Martinez-Vazquez et al., 1997, p. 289; and Foster, 1997, pp. 10–11). In the bureau’s words, “Where these districts have separate governing bodies, they are counted as governments; when they are governed by county or municipal officials in an ex officio capacity, they are not counted as separate governments” (U.S. Bureau of the Census, 1994, p. A-23). For the purposes of measuring political structure, the Census Bureau’s procedure is quite appropriate. The purpose of this report is 85 not primarily to count the number of units that exist on paper but to measure the dispersion of governmental authority and activity. Comparison to Controller’s Count Comparing the Controller’s data for fiscal year 1991–1992 with the 1992 Census of Governments highlights the differences involved. The Controller counted 4,857 districts as “officially in existence” in that year, whereas the Census Bureau listed 2,797. Of the State Controller’s enumeration, 1,604 districts were governed by a county board of supervisors, and 150 reported to city councils. These are appropriately considered “dependent” and thus would not be counted under the Census Bureau’s rules. Instead, their finances are included as part of the parent government. The types of activities performed by districts that were disproportionately dependent include the following: • County service areas • County waterworks • Highway lighting (all dependent) • Maintenance • Permanent road division • Sewer/sewer maintenance • Storm water The above types of districts are those, then, that will tend to be “undercounted,” in a strictly paper sense, by the Census Bureau. In contrast, the following types of districts are disproportionately “independent,” according to the classification scheme of the State 86 Controller—meaning that most are not directly governed by an individual city council or county board of supervisors: • Cemetery • Community services • Fire protection • Hospital • Irrigation • Joint exercise of powers • Levee • Library • Memorial • Mosquito abatement • Nonprofit corporations • Reclamation • Recreation/park • Sanitary • Public utility • Water However, some of these so-called “independent” districts are in fact joint creations of multiple general-purpose governments and typically are overseen by appointees of each government. In other cases, generalpurpose governments have budgetary review or other discretion over an “independent” district, or appoint its officers. Thus, in cases where a district is largely accountable to, under the oversight of, or completely financed by, an individual general-purpose government, the Census 87 Bureau does not include the district as a separate government. This proviso pertains to 245 nonprofit corporations—which largely construct public facilities such as libraries and city halls, or serve as purchasing entities for equipment. The Census Bureau also removes 11 flood control districts and 10 other districts, from various categories, from its list of autonomous districts. The bureau does count joint-powers agencies as separate governments, where the district’s board is distinct from other individual local governments. There are no clear reasons for the remaining disparities between the Census count and the Controller’s count. However, the disparity is relatively trivial, as Table B.1 shows. Table B.1 Reconciling Controller and Census Bureau Counts of Special Districts, 1992 Total special districts in existence, per Controller Minus: Districts excluded because they are “dependent” on a city or county Districts excluded because they are otherwise accountable to another local government Equals: Ostensible count of autonomous special districts Census of Governments count Difference: 4,836 1,752 266 2,818 2,797 21 (0.7%) Response to Criticism of Census Data on Districts Leigland (1990) has critiqued the Census Bureau’s data on special districts for three main reasons. However, none of these detracts seriously from the usefulness of the Census of Governments data for the current report. First, Leigland argues that the Census’s data on debt is inadequate for special districts. However, the fragmentation index used in this 88 report is based on expenditures, not debt. Second, Leigland claims that the bureau undercounts special districts—particularly those involved in financing public works—by not including dependent units, noting that the creation of districts has been an important creative financing scheme for cities and counties. However, for the reasons discussed above, dependent districts are not politically autonomous units; they are typically little more than accounting devices of general-purpose governments and are counted appropriately by the bureau as part of the finances of the parent units. Finally, Leigland argues that the bureau overcounts other districts by including inactive districts, with zero expenditures, in the Census of Governments listings. The political fragmentation index overcomes this problem, since governments with zero expenditures receive no weight in the index. Again, Leigland’s critique is more damning for those obsessed with counting governments, but it is less important for those interested in the dispersion of political authority. Undoubtedly, some will consider the Census Bureau’s data less than ideal for analysis of special districts in California. However, the bureau constructs its information on special districts in California using, in large part, information obtained from the State Controller’s records—data that have been found to be comprehensive and accurate, at least as regards revenue (Shires and Glenn Haber, 1997). Moreover, the Census Bureau attempts to clarify inconsistencies and questions through consultations with relevant state and local government personnel (Sacks, 1990; and U.S. Bureau of the Census, 1994). In any event, the State Controller’s data were not available for analysis in machine-readable format over the time period of interest. 89 Appendix C New Cities Incorporated After Proposition 13 Table C.1 California Cities Incorporated Since July 1, 1978 (listed by county) County Alameda City Dublin Year of Incorporation 1982 Butte Paradise 1979 Contra Costa Danville San Ramon Orinda 1982 1983 1985 Kings Avenal 1979 Lake Clearlake 1980 Los Angeles La Habra Heights Westlake Village 1978 1981 91 Table C.1 (continued) County City Agoura Hills West Hollywood Santa Clarita Diamond Bar Malibu Calabasas Year of Incorporation 1982 1984 1987 1989 1991 1991 Mono Mammoth Lakes 1984 Napa American Canyon 1992 Nevada Truckee 1993 Orange Mission Viejo Dana Point Laguna Niguel Lake Forest Laguna Hills 1988 1989 1989 1991 1991 Placer Loomis 1984 Riverside Cathedral City La Quinta Moreno Valley Temecula Calimesa Canyon Lake Murrieta 1981 1982 1984 1989 1990 1990 1991 Sacramento Citrus Heights 1997 San Bernardino Grand Terrace Big Bear Lake Twentynine Palms Highland Hesperia Apple Valley Yucaipa Yucca Valley Chino Hills 1978 1980 1987 1987 1988 1988 1989 1991 1991 92 Table C.1 (continued) County San Diego City Poway Santee Solana Beach Encinitas Year of Incorporation 1980 1980 1986 1986 San Joaquin Lathrop 1989 San Luis Obispo Atascadero 1979 San Mateo East Palo Alto 1983 Santa Barbara Solvang Buellton 1985 1992 Shasta Shasta Lake 1993 Sonoma Windsor 1992 Ventura Moorpark 1983 Yolo West Sacramento 1987 93 Bibliography Assembly Committee on Local Government, General Assembly, State of California, Guide to Cortese-Knox Local Government Reorganization Act of 1985, revised ed., Sacramento, California, 1996. Assembly Office of Research, General Assembly, State of California, California 2000: Getting Ahead of the Growth Curve, Sacramento, California, December 1989. Baird, Robert N., and John M. Landon, “Political Fragmentation, Income Distribution, and the Demand for Government Services,” Nebraska Journal of Economics and Business, Vol. 11, No. 4, 1972, pp. 171–184. Baldassare, Mark, When Government Fails: The Orange County Bankruptcy, University of California Press and Public Policy Institute of California, Berkeley, California, 1998. Blair, John P., Samuel R. Staley, and Zhongcai Zhang, “The Central City Elasticity Hypothesis: A Critical Appraisal of Rusk’s Theory of Urban Development,” Journal of the American Planning Association, Vol. 62, No. 3, 1996, pp. 345–353. Blair, John P., and Zhongcai Zhang, “Ties that Bind’ Reexamined,” Economic Development Quarterly, Vol. 8, No. 4, 1994, pp. 373–377. Bollens, Scott A., “Examining the Link between State Policy and the Creation of Local Special Districts,” State and Local Government Review, Fall 1986, pp. 117–124. 95 Brooks, Lea, “The Revenue Chase and Urban Sprawl,” Golden State Report, June 1988, p. 37. Burns, Nancy, The Formation of American Local Governments: Private Values in Public Institutions, Oxford University Press, New York, 1994. California Constitution Revision Commission, Final Report and Recommendations to the Governor and the Legislature, Sacramento, California, 1996. California State Controller, Financial Transactions Concerning Cities of California, Sacramento, California, annual. California State Controller, Financial Transactions Concerning Special Districts of California, Sacramento, California, annual. Danielson, Michael N., Alan M. Hershey, and John M. Bayne, One Nation, So Many Governments, Lexington Books, Lexington, Massachusetts, 1977. Detwiler, Peter M., “Creatures of Statute . . . Children of Trade: The Legal Origins of California Cities,” in California Constitution Revision Commission, History and Perspective, Sacramento, 1996a, pp. 97–104. Detwiler, Peter M., “Suburban Secession: History and Context,” Working Paper 96–12, Graduate Program in Public Policy and Administration, California State University at Sacramento, 1996b. Feinbaum, Robert, “Climate Right for Creating New Cities,” California Journal, October, 1987, pp. 497–499. Fisher, Ronald C., and Robert W. Wassmer, “Economic Influences on the Structure of Local Government in U.S. Metropolitan Areas,” Journal of Urban Economics, Vol. 43, 1988, pp. 444–471. Forbes, Kevin F., and Ernest M. Zampelli, “Is Leviathan a Mythical Beast?” American Economic Review, Vol. 79, No. 3, 1989, pp. 568– 577. Foster, Kathryn A.,“ Exploring the Links between Political Structure and Metropolitan Growth,” Political Geography, Vol. 12, No. 6, 1993, pp. 523–547. Foster, Kathryn A., The Political Economy of Special-Purpose Government. Georgetown University Press, Washington, D.C., 1997. 96 Fulton, William, “Two States Find a Way to Keep the Peace,” Planning, October 1989, pp. 16–17. Fulton, William, Guide to California Planning, Solano Press Books, Point Arena, California, 1991. Fulton, William, “City Secession Bill Moves Forward,” California Planning & Development Report, Vol. 12, No. 9, 1997a, p. 2. Fulton, William, “State to Examine Fiscal Relationship with Locals,” California Planning & Development Report, Vol. 12, No. 10, 1997b, pp. 1, 10. Fulton, William, “Moves Begin to Reform Cityhood Law,” California Planning & Development Report, Vol. 13, No. 4, 1998, pp. 1, 9–10. Gainsborough, Juliette, “Incorporation: Government of the Future? The City of Santa Clarita, A Case Study,” senior thesis, University of California at Berkeley, 1990. Hawkins, Brett W., and Rebecca M. Hendrick, “Do Metropolitan Special Districts Reinforce Sociospatial Inequalities?” Publius: The Journal of Federalism, Vol. 27, No. 1, 1997, pp. 135–143. Hill, Edward W., Harold L. Wolman, and Coit Cook Ford III, “Can Suburbs Survive without Their Central Cities? Examining the Suburban Dependence Hypothesis,” Urban Affairs Review, Vol. 31, No. 2, 1995, pp. 147–174. Isserman, Andrew M., “Interjurisdictional Spillovers, Political Fragmentation and the Level of Local Public Services: A Reexamination,” Urban Studies, Vol. 13, 1976, pp. 1–12. Kaniss, Phyllis, Making Local News, University of Chicago Press, Chicago, 1991. League of California Cities, Conflicts at the City’s Edge: Annexations, Development and Tax Negotiations, League of California Cities, Sacramento, April 1996. Leigland, James, “The Census Bureau’s Role in Research on Special Districts: A Critique,” Western Political Quarterly, Vol. 43, No. 2, 1990a, pp. 367–380. Leigland, James, “In Defense of a Preoccupation with Numbers: A Response,” Western Political Quarterly, Vol. 43, No. 2, 1990b, pp. 385–386. 97 Leigland, James, “Public Authorities and the Determinants of Their Use by State and Local Governments,” Journal of Public Administration Research and Theory, Vol. 4, No. 4, 1994, pp. 521–544. Lewis, Paul G., Shaping Suburbia: How Political Institutions Organize Urban Development, University of Pittsburgh Press, Pittsburgh, Pennsylvania, 1996. MacManus, Susan, “Special District Governments: A Note on Their Use as Property Tax Relief Mechanisms in the 1970s,” Journal of Politics, Vol. 43, 1981, pp. 1207–1214. Marando, Vincent, and Mavis Mann Reeves, “State Responsiveness and Local Government Reorganization,” Social Science Quarterly, Vol. 69, No. 4, 1988, pp. 996–1004. Martin, Linda M., The Politics and Economics of Local Self Government: New Cities in California after Proposition 13, Ph.D. dissertation, University of California at Davis, 1984. Martin, Linda M., and Joan Q. Hogan, “Small Cities, Big Dreams,” Western City, May 1989, pp. 20–23, 44. Martinez-Vazquez, Jorge, Mark Rider, and Mary Beth Walker, “Race and the Structure of School Districts in the United States,” Journal of Urban Economics, Vol. 41, 1997, pp. 281–300. Miller, Gary J., Cities by Contract: The Politics of Municipal Incorporation, MIT Press, Cambridge, Massachusetts, 1981. Miller, Joel, “Annexations and Boundary Changes in the 1980s and 1990–1991,” in The Municipal Year Book 1993, International City/County Management Association, Washington, 1993, pp. 100– 109. Morgan, David R., and Patrice Mareschal, “Central City/Suburban Inequality and Metropolitan Political Fragmentation,” presented at the Annual Meeting of the American Political Science Association, San Francisco, California, 1996. Morgan, Stephen P., and Jeffrey I. Chapman, Special District Consolidations: A Research Study for the Association of California Water Agencies, Los Angeles, California, December 1, 1994. Musso, Juliet Ann, Drawing the Line: The Political Economy of Municipal Incorporation, Ph.D. dissertation, University of California at Berkeley, 1994. 98 Neiman, Max, “An Exploration into Class Clustering and LocalGovernment Inequality,” in Richard C. Rich (ed.), Analyzing UrbanService Distributions, Lexington Books, Lexington, Massachusetts, 1982, chapter 13. Nelson, Michael A., “Decentralization of the Subnational Public Sector: An Empirical Analysis of the Determinants of Local Government Structure in Metropolitan Areas in the U.S.,” Southern Economic Journal, Vol. 57, No. 2, 1990, pp. 443–457. Nice, David C., “An Intergovernmental Perspective on Urban Fragmentation,” Social Science Quarterly, Vol. 64, No. 1, 1983, pp. 111–118. Oates, Wallace E., “Searching for Leviathan: A Reply and Some Further Reflections,” American Economic Review, Vol. 79, No. 3, 1989, pp. 578–583. Ostrom, Elinor, “The Social Stratification—Government Inequality Thesis Explored,” Urban Affairs Quarterly, Vol. 19, No. 1, 1983, pp. 91–112. Perrenod, Virginia, Special Districts, Special Purposes, Texas A&M University Press, College Station, Texas, 1984. Peterson, Paul E., City Limits, University of Chicago Press, Chicago, Illinois, 1981. Rusk, David, Cities Without Suburbs, Johns Hopkins University Press, Baltimore, Maryland, 1993. Sacks, Seymour, “The Census Bureau’s Role in Research on Special Districts: A Critique’: A Necessary Rejoinder,” Western Political Quarterly, Vol. 43, No. 2, 1990, pp. 381–383. Savitch, H. V., David Collins, Daniel Sanders, and John P. Markham, “Ties That Bind: Central Cities, Suburbs, and the New Metropolitan Region,” Economic Development Quarterly, Vol. 7, No. 4, 1993, pp. 341–357. Schneider, Mark, The Competitive City: The Political Economy of Suburbia, University of Pittsburgh Press, Pittsburgh, Pennsylvania, 1989. Schrag, Peter, Paradise Lost: California’s Experience, America’s Future, New Press, New York, 1998. 99 Shires, Michael A., and Melissa Glenn Haber, A Review of Local Government Revenue Data in California, Public Policy Institute of California, San Francisco, 1997. Sokolow, Alvin D., et al., Choices for the Unincorporated Community: A Guide to Local Government Alternatives in California, 2nd. ed., Institute of Governmental Affairs, University of California at Davis, September 1981. Stein, Robert M., “Tiebout’s Sorting Hypothesis,” Urban Affairs Quarterly, Vol. 23, No. 1, 1987, pp. 140–160. Teaford, Jon C., Post-Suburbia: Government and Politics in the Edge Cities, Johns Hopkins University Press, Baltimore, Maryland, 1997. Tiebout, Charles, “A Pure Theory of Local Expenditures,” Journal of Political Economy, Vol. 64, 1956, pp. 416–424. U.S. Bureau of the Census, County Business Patterns 1977—California, Department of Commerce, Washington, D.C., 1979. U.S. Bureau of the Census, 1992 Census of Governments: Volume 1, Government Organization, Washington, D.C., March 1994. U.S. Bureau of the Census, Census of Governments: Finance Summary Statistics [machine-readable data file], Bureau of the Census, Washington, D.C., 1972, 1977, 1982, 1987, and 1992. Wassmer, Robert W., and Ronald C. Fisher, “Time, Tiebout, and Transition in the Structure of Local Government in U.S. Metropolitan Areas,” Working Paper 97-07, Graduate Program in Public Policy and Administration, California State University at Sacramento, 1997. Weiher, Gregory R., The Fractured Metropolis: Political Fragmentation and Metropolitan Segregation, State University of New York Press, Albany, 1991. Woodlief, Anthony, “The Path-Dependent City,” Urban Affairs Review, Vol. 33, No. 3, 1998, pp. 405–437. Zax, Jeffrey, “Is There a Leviathan in Your Neighborhood?” American Economic Review, Vol. 79, No. 3, 1989, pp. 560–567. 100 About the Author PAUL G. LEWIS Paul Lewis is a research fellow at the Public Policy Institute of California. He is a political scientist with research interests in urban and metropolitan problems including transportation, land use, housing, and the structure of local government. His recent publications include Federal Transportation Policy and the Role of Metropolitan Planning Organizations in California (Pubic Policy Institute of California, 1997) and Shaping Suburbia: How Political Institutions Organize Urban Development (University of Pittsburgh Press, 1996). Before joining PPIC, he was an assistant professor of political science at Florida International University. He holds an A.B. in political science from Indiana University and an M.A. and Ph.D. in politics from Princeton. 101 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., 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. Shires, Michael A., John Ellwood, and Mary Sprague (1998), Has Proposition 13 Delivered? The Changing Tax Burden 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. 102" } ["___content":protected]=> string(102) "

R 998PLR

" ["_permalink":protected]=> string(94) "https://www.ppic.org/publication/deep-roots-local-government-structure-in-california/r_998plr/" ["_next":protected]=> array(0) { } ["_prev":protected]=> array(0) { } ["_css_class":protected]=> NULL ["id"]=> int(8136) ["ID"]=> int(8136) ["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(3253) ["post_status"]=> string(7) "inherit" ["post_title"]=> string(8) "R 998PLR" ["post_type"]=> string(10) "attachment" ["slug"]=> string(8) "r_998plr" ["__type":protected]=> NULL ["_wp_attached_file"]=> string(12) "R_998PLR.pdf" ["wpmf_size"]=> string(6) "241525" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(181072) "Deep Roots: Local Government Structure in California Paul G. Lewis, 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 Political scientists have long struggled with the question of the “right size” for local government—whether the measure is geography or budgets. Jurisdictions covering many square miles, such as a typical county in California, may have the merit of a diverse and extensive tax base but, at the same time, may be seen as “out of touch” with the needs and desires of the average resident or voter. Larger public budgets may have the merit of flexibility in allocating funds among different services and in setting different priorities over time, but smaller budgets and jurisdictions are seen as more “user friendly” and more easily modified in response to community-level preferences and concerns. For all this understanding of basic budgeting and governance principles, there is no supportable theory for the right-sized budget or jurisdictional boundary. Nevertheless, a substantial literature supports the argument that “excessive” governmental fragmentation may complicate the coordination of services and the planning of land and infrastructure development. In California, Proposition 13 is often viewed as a iii significant incentive for governmental fragmentation: Smaller units of government can be established with the explicit purpose of segregating— and collecting taxes from—high-revenue-producing properties. Research fellow Paul Lewis tackles this topic head-on in this report. He concludes that there is no evidence that Proposition 13 created any unusual movement toward either the creation of special districts or the incorporation of new cities. In fact, the slowing of local government fragmentation brought about by the creation of Local Agency Formation Commissions in 1963 has not fundamentally changed over the 20 years since the passage of Proposition 13. This report is one in a series of PPIC publications stemming from our ongoing program of work in governance and public finance. Two recent publications of direct relevance to the analysis of Proposition 13 include Michael Shires, John Ellwood, and Mary Sprague, Has Proposition 13 Delivered? The Changing Tax Burden in California; and Mark Baldassare, When Government Fails: The Orange County Bankruptcy, University of California Press. The completion of each of our studies results in a better understanding of the changing shape of governance and public finance in California. Although formulating policy options is never easy, the clarity of the issues involved is definitely improving as our work proceeds. David W. Lyon President and CEO Public Policy Institute of California iv Summary Many observers have argued that California’s system of local government is too complex and fragmented. Such complexity, some say, makes government action hard to coordinate, political responsibility hard to focus, and government actions less efficient. Moreover, several commentators have attributed local government fragmentation, in part, to the Proposition 13 tax limitation of 1978, which many expected to lead to a proliferation of new cities and special district governments. More recently, there have been legislative proposals in the early 1990s for regional governance and state planning, extensive recommendations for restructuring local governance from the California Constitution Revision Commission in 1996, and the creation in 1997 of a Commission on Local Governance in the 21st Century. This report aims to provide necessary background and context for these deliberations about local government structure. It summarizes what is known and not known about the causes and consequences of local government fragmentation. The report uses data on government v structure and finances from the State Controller and the U.S. Census Bureau’s Census of Governments (which is taken every five years, with the most recent data from 1992). The data are used to summarize recent historical trends relating to California’s deployment of cities, counties, and special districts—looking in particular at the possible influence of Proposition 13 on the shape of the local public sector. Finally, the report examines variations in the level of local government fragmentation in the different counties of the state and analyzes the factors that underlie these differences. Why Is Local Government Structure Important? Much research suggests that political fragmentation—the degree to which governmental powers are divided geographically and functionally among local governments—has important effects on both the process of policymaking and residents’ quality of life. These effects should be considered in relation to a number of different realms of local governance. For the provision of routine public services, such as trash collection or street cleaning, fragmentation may offer an advantage rather than a disadvantage. Despite reformers’ fears about “duplication” and “overlap,” multiple local governments may be advantageous in providing residents with a range of choices, thus better reflecting their varied tastes for particular levels and types of services. In addition, competition, particularly among municipalities, may restrain the growth and inefficiency of local government bureaucracies. However, the administrative costs of multiple special districts in an area may drive up the price of government. In the realm of land use and infrastructure, fragmentation of local governance is often seen as less benign. It may lead to parochial, self- vi interested land-use decisions and a lack of policy concern for regional challenges such as housing affordability, racial and socioeconomic integration, and the ability of regional infrastructure to accommodate growth patterns. A fragmented public sector may be less effective at promoting regional economic development. And in the realm of equity, fragmentation can separate fiscal resources from needs, as lower-income areas suffer with weaker tax bases. Here, the case for financial assistance from state and national governments is particularly strong. Overall, critics of political fragmentation fault it because areas that are more politically divided are believed to have a more difficult time addressing the “general welfare.” California’s system of local governance—with 58 counties, 471 cities, and nearly 5,000 special districts (many of them in fact governed by cities and counties)—undoubtedly appears complex to most residents. An individual household usually will find itself simultaneously governed by several overlapping jurisdictions. In reality, however, California’s local government structure is less complex than the national average. The state has many times fewer counties and cities per capita and significantly fewer politically independent special districts per capita. And although it may appear disordered, California’s public sector did not evolve randomly but rather in response to the varying needs and demands of homeowners, businesses, and other groups as the state has grown and urbanized. Voters have repeatedly created new cities and special districts, in search of local control of land use, public services, and taxes. vii Did Local Government Fragmentation Increase in the Wake of Proposition 13? Historically, California policymakers have shown a number of waves of interest in issues of local government structure. Over the past two decades, concern has centered in large part on the effects of the Proposition 13 tax limitation upon the “proliferation” of local governments. Many observers pointed with alarm to numerous city incorporations and special district formations in its wake and saw portents of a great increase in governmental fragmentation. The conventional wisdom, found in numerous academic and journalistic accounts, holds that Proposition 13 created incentives for the widespread creation of new cities and special districts. In the case of cities, this property tax limitation removed concerns that a newly formed government would lead to increased property taxes, since Proposition 13 caps ad valorem property tax rates and restricts reassessments. Moreover, with counties struggling fiscally in the post-1978 era of revenue constraints, many areas might be expected to “secede” from county service provision by incorporating as a city, thus capturing some locally generated revenues for local use. In the case of special districts, fiscally starved counties and cities might be expected to encourage the creation of new special-purpose governments. The counties and cities could thereby shed some of their service and revenue-raising burdens onto entities that often are somewhat more flexible in their ability to issue debt or levy fees and service charges. Local officials and developers seeking to finance capital facilities and services for growing populations might also view districts’ relative political insulation and invisibility to taxpayers as an advantage. viii In general, however, this conventional wisdom is not supported by the data. It is true that 54 new cities have incorporated since 1978, but most were in areas of metropolitan Southern California that had extremely rapid population growth and suburbanization and thus were natural candidates for additional governments. Looking at incorporations across the state’s history, we can see that the post-1978 era hardly seems exceptional in its rate of city formation, as Figure S.1 illustrates. Rather, the growth rate in the number of new cities has slowed since the state created Local Agency Formation Commissions (LAFCOs) in each county in 1963 to regulate government boundaries and formations. In fact, the passage of a 1992 state law requiring LAFCOs to approve incorporations only if they are “fiscally neutral” for the affected county appears to have brought incorporations to a virtual halt. Most new population growth in the state continues to be accommodated in existing cities—which regularly expand their boundaries through annexation. Thus the number of cities per 100,000 population has fallen throughout this century, as Figure S.2 shows. Reliable and comparable historical data on special districts are harder to come by, but the evidence indicates that the overall number of special districts has increased only gradually since the 1970s. However, the universe of districts has changed somewhat in character, probably in response to Proposition 13. Districts that exist to finance the construction of capital facilities have increased in number, whereas those engaged in more traditional service provision have declined in number. “Governments created by governments” have become more popular, as policymakers have used the ability of districts to levy enterprise fees and issue debt; meanwhile, politically autonomous districts relying on property tax financing have been a waning presence since Proposition 13. ix 500 450 400 350 300 250 200 150 100 50 0 1850 1870 1890 1910 1930 1950 1970 1990 Figure S.1—Number of Cities in California, 1850–1997 9 8 7 6 5 4 3 2 1 0 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 1997 Figure S.2—Number of Cities per 100,000 Population in California, 1900– 1997 x Overall, the relative share of total local government spending in the state accounted for by politically freestanding special districts, as compared to counties and cities, shifted only marginally between 1972 and 1992, as Figure S.3 indicates. Counts of governments may be somewhat misleading barometers of local government change, so this report also introduces a local political fragmentation index, or PFI. The PFI measures the degree to which spending activity is divided among local governments; it represents the probability that two randomly selected dollars of local expenditure are spent by different local governments. Measured statewide or at the county level, local political fragmentation increased only very slightly in the period 1972 to 1992, as Figure S.4 shows. In short, even as Proposition 13 played havoc with many local governments’ finances, it did not decisively alter overall local government structure. 50 45 40 35 30 25 20 15 10 5 0 1972 Counties Cities Special districts (Census definition) 1977 1982 1987 1992 Figure S.3—Percentage Shares of Total Local Nonschool Expenditures, by Type of Government xi 1.0 0.9 0.8 0.7 0.6 0.5 0.4 1972 Statewide PFI Population-weighted county PFI Average county PFI 1977 1982 1987 1992 Figure S.4—Political Fragmentation Index in California, 1972–1992 What Factors Affect Local Government Structure Across California’s Counties? Examining the political fragmentation index at the county level shows that most California counties also did not experience much variation in their overall local government structure from 1972 to 1992. However, the PFI data do reveal that parts of the state differ quite markedly in their local political structure. For instance, the San Francisco Bay area tends to be more governmentally fragmented than metropolitan Southern California, but the latter area has experienced some convergence toward the higher level of fragmentation in the former area. In general, counties with high levels of political fragmentation tend to have several sizable cities, a low proportion of the population in unincorporated areas, and relatively high usage of special districts. xii Multivariate data analysis was performed to examine the factors underlying the variations among counties in political structure. Although numerous variables were considered, the degree of governmental fragmentation appears to largely be a product of three underlying factors. The most important is the county’s population size, which is positively related to fragmentation. Bigger, more complex counties are likely to have more complex service demands and thus more complicated public sectors. The second main influence on local government structure is intergovernmental aid. Areas that have received a higher share of their revenues from state and federal sources tend to acquire less governmental complexity. This could be because intergovernmental aid can “level up” service levels and infrastructure quality across local areas, and because such aid helps keep existing governments in better fiscal condition. Both of these effects associated with intergovernmental aid tend to reduce the incentives for incorporation of new cities and also reduce the impetus to engage in “creative financing” by forming a special district. A third factor involves the county’s historical pattern of population growth. “Older” counties—those that achieved most of their ultimate population size before 1950—tend to be significantly more fragmented. This effect is consistent with the notion that most of California’s local government development occurred during earlier periods of the state’s history. In particular, residents in earlier eras were quicker to form new cities. What Are the Implications for Policymakers? This portrait of relative stability in California’s system of local governance is a rather remarkable “nonfinding” in an era of very rapid political, economic, and demographic change in the state. One possible xiii reading of this record is that an earlier state policy—the creation of LAFCOs in each county—has succeeded in reining in the fragmentation of government. Although their merits have been widely debated, LAFCOs, which have governing boards representing existing cities and counties, might be expected to try to preserve the viability of existing units by steering population growth into cities and looking warily upon the formation of new units. Later changes to the LAFCO law, requiring them to create “spheres of influence” for cities and providing new instructions that further tightened up incorporation requirements, likely strengthened these tendencies. Would-be reformers should also consider the possibility that existing local governments may be more malleable and responsive to change than is commonly supposed. Throughout the state, cities, counties, and special districts typically have been able to address new issues, empower new constituencies, and respond to economic and population growth without falling into crisis. Does that mean that state and local policymakers should be complacent about California’s local government system? Not necessarily. Local fragmentation continues to separate fiscal resources from social needs and to impede coordination on issues such as land use. Despite California’s home rule tradition, local governments remain “creatures of the state,” subject to fine-tuning from Sacramento. Policymakers should think seriously about ways to enhance the fiscal stability of local governments—keeping in mind the finding that more generous intergovernmental aid appears to reduce the impulse toward local fragmentation. And enhanced mechanisms for regional coordination on growth and infrastructure issues may be necessary. xiv The new Commission on Local Governance for the 21st Century— like the recent California Constitution Revision Commission—presents a forum for considering local government’s strengths and weaknesses and debating possible changes in the state’s rules for local governance. But those who would propose sweeping reforms or large-scale consolidations of local units must confront the reality of a stable and resilient system of local rule. xv Contents Foreword..................................... iii Summary..................................... v Figures ...................................... xxi Tables ....................................... xxiii Acknowledgments ............................... xxv 1. INTRODUCTION: THE CONTROVERSY OVER LOCAL GOVERNMENT STRUCTURE ............ What Difference Does Local Government Structure Make? .. Service Provision............................ Land Use and Economic Development ............. Equity .................................. Representation ............................. Organization of This Report ...................... 1 3 4 5 6 6 8 2. LOCAL GOVERNANCE AND HISTORICAL CHANGE ................................. A Recurring Policy Debate ....................... Creating Local Agency Formation Commissions........ Worrying About the Effects of Proposition 13 ......... Focusing Renewed Attention on Local Government Problems ............................... Frequency of Different Categories of Local Government .... 9 10 10 11 13 15 xvii Counties ................................. Cities ................................... Special Districts ............................ School Districts ............................ Shares of Local Public Spending by Different Types of Government............................. Summary .................................. 3. LOCAL GOVERNMENT FRAGMENTATION IN CALIFORNIA: MEASUREMENT AND TRENDS ..... Problems in Measuring Local Government Structure by Counting Governments ..................... An Index of Local Political Fragmentation ............. Levels of Local Political Fragmentation Among California Counties ............................... Did Proposition 13 Accelerate Fragmentation? .......... Summary .................................. 4. ACCOUNTING FOR VARIATIONS IN LOCAL GOVERNMENT STRUCTURE ACROSS CALIFORNIA .............................. What Factors Might Affect Local Government Structure? ... General Contextual Factors ..................... Intergovernmental Rules and Relationships ........... Socioeconomic Characteristics of the Population ....... Political Influence of Developers or Other Businesses .... Why Fragmentation? Toward a Model of Local Government Structure ............................... Bivariate Relationships ........................ Multivariate Model .......................... Interpretation ............................. Other Variables ............................ Summary .................................. 5. CONCLUSIONS AND POLICY CONSIDERATIONS ... Review of Findings ............................ Relative Stability in an Era of Change—Why? .......... Considerations for State Policymakers ................ 17 18 25 29 30 31 33 33 35 37 41 45 47 47 48 49 52 55 57 60 60 63 66 67 69 70 72 75 xviii Appendix A. Research on the Effects of Local Government Fragmentation ............................... B. Issues Regarding Special District Data in the Census of Governments ............................... C. New Cities Incorporated After Proposition 13 .......... 79 85 91 Bibliography .................................. 95 About the Author ............................... 101 Other PPIC Publications........................... 102 xix Figures S.1. Number of Cities in California, 1850–1997 ......... S.2. Number of Cities per 100,000 Population in California, 1900–1997 .............................. S.3. Percentage Shares of Total Local Nonschool Expenditures, by Type of Government ............ S.4. Political Fragmentation Index in California, 1972– 1992 .................................. 2.1. Number of Cities in California, 1850–1997 ......... 2.2. Number of Cities per 100,000 Population in California, 1900–1997 .............................. 2.3. Number of School Districts in California, 1962–1992... 2.4. Percentage Shares of Total Local Nonschool Expenditures, by Type of Government ............ 3.1. Levels of Political Fragmentation in California Counties, 1992 .................................. 3.2. Political Fragmentation Index in California, 1972– 1992 .................................. 3.3. Political Fragmentation in California’s Major Metropolitan Areas ......................... x x xi xii 20 23 29 31 39 42 44 xxi Tables 2.1. Local Governments per 100,000 Population: California Compared to the Rest of the United States .......... 2.2. Number of New Cities by County, 1978–1997 ....... 2.3. How California Accommodated Population Growth After Proposition 13, 1978–1997 ................ 2.4. Number of Special Districts in California ........... 3.1. Political Fragmentation Index of Local Government Structure in California Counties, 1992 ............ 3.2. Average Percentage Shares of Nonschool Expenditure in High- and Low-PFI Counties .................. 3.3. A Comparison of Local Political Structure in Two Counties, 1992 ........................... 4.1. Variables Studied for Possible Relationships with Local Political Fragmentation ...................... 4.2. Regression Models of Political Fragmentation Indexes for California Counties, 1992 .................... B.1. Reconciling Controller and Census Bureau Counts of Special Districts, 1992 ....................... C.1. California Cities Incorporated Since July 1, 1978 ...... 16 19 23 26 38 40 40 58 62 88 91 xxiii Acknowledgments I wish to thank my research colleagues at PPIC for numerous helpful discussions and critiques of this project. In particular, Fred Silva was of key importance as a sounding board and reality check throughout the project, and Kim Rueben wrote a very useful review of the manuscript. The external reviewers, Alvin Sokolow of the University of California, Davis, and April Manatt, consultant to the California Senate Local Government Committee, also read the manuscript with care and provided helpful comments. Mary Sprague lent early research assistance—and in particular, accomplished the unenviable task of assembling the Census of Governments data into a useable format. Michael Molloy performed a reconciliation of Census and state counts of special districts, discussed in Appendix B. Sun Kim produced the map in Chapter 3. The Demographic Research Unit at the state Department of Finance provided important data. Michael Teitz and Andrew Isserman helped keep the project on track and lent useful substantive advice. Gary Bjork provided expert editorial help. While all of these xxv people (and others) improved the final product, the responsibility for any errors of fact or interpretation rests with me. xxvi 1. Introduction: The Controversy over Local Government Structure In California, as in all American states, local government should be treated as a plural, not a singular noun. Political authority in California is divided among thousands of local jurisdictions. Los Angeles County alone has 88 cities, for instance, and advocates of secession in the San Fernando Valley seek to create another. Local government, though, means more than just cities. An individual household may find itself simultaneously governed by several overlapping jurisdictions: a county, a city, a school district and community college district, along with numerous other specialized and politically independent entities, devoted to purposes ranging from regional mass transit to wastewater treatment to firefighting to mosquito abatement to flood control. Additional “dependent special districts”— technically separate financial entities, but under the political control of county supervisors or city councils—may levy assessments on households 1 or charge fees to customers to provide services or build community facilities for a geographic area. California policymakers have periodically expressed concern over this complexity of local government structure. Although the issue rarely commands newspaper headlines, state legislators and governors have repeatedly struggled with what many have thought of as a “proliferation” of governments. Thus, county-level commissions have been created to oversee annexations and formations of new cities and special districts; regional reorganizations have been debated, as policymakers worried about the effects of highly decentralized land-use powers on growth patterns; and study commissions have been appointed—most recently, in 1997, a Commission on Local Governance for the 21st Century. Lurking in the background over the past two decades has been the concern that the Proposition 13 property tax limitation of 1978 would serve to accelerate the formation of new governments, and thus fundamentally alter California’s local governance structure. Throughout these debates, some have argued that local political “fragmentation” makes government seem more complex and “out of control” to citizens, while also making coordinated government action in response to policy problems more difficult. Other commentators and academic writers have argued that a highly decentralized government structure has a host of underappreciated benefits in terms of efficiency and governmental responsiveness to the public. Rarely, however, have the debates in local and state arenas moved beyond simple catch phrases such as “wasteful duplication” or “local control” to seriously examine the structure of local governance in California. This report, part of a continuing line of research on local government institutions and finances from the Public Policy Institute of 2 California, is intended to remedy this shortcoming. It provides key background to inform the debate over local governance and state/local relations, examining long-term trends in California’s local government structure and considering its determinants. What Difference Does Local Government Structure Make? Why has so much concern been given to local political structure? To some degree, public officials have been uneasy about what was perceived as overly rapid change in the system of counties, cities, and special districts. In general, though, concern for local political structure is hardly abstract but is rather connected to concerns about the efficiency, equity, and coordination of local government policies, services, and revenue patterns. Much research suggests that political fragmentation— the degree to which governmental powers are divided geographically and functionally among local governments—has important effects on the provision of government services and the regulation of land use.1 In addition, many policymakers are concerned that a proliferation of local governments makes policy hard to coordinate in areas such as transportation, housing, and environmental protection, and separates fiscal needs from resources. This topic has been the subject of extensive academic analysis. The literature on the effects of local government arrangements is summarized and cited in Appendix A. In this chapter, I briefly highlight the policy and political implications of such research. The literature is marked by a ____________ 1The term fragmentation, used in this report synonymously with political or governmental complexity, is a descriptive term; it is not intended to connote a negative or positive value judgment. 3 certain amount of confusion and dissensus. One nearly universal problem is a failure to distinguish among the varying roles of local government that might be differentially affected by fragmentation. At least four distinct roles should be considered, touching on different areas of local representation and governance. Service Provision First, local government may be viewed primarily as a provider of fairly routine public services, such as trash collection and fire protection. This role involves most of the everyday duties of local public employees and expenditure of local tax dollars. In this realm—somewhat analogous to the world of business—economic theories of competition among local governments would seem to hold the strongest logic. According to this framework, more fragmentation means that boundaries sort residents into “communities of interest” with similar tastes for services and tax levels, thus allowing a closer match between a government’s offerings and its residents’ desires. The competition associated with fragmentation also may keep local bureaucracies from the dangers of monopolistic service provision. It is possible in some cases, however, that the administrative costs of duplication and overlap can drive up the price of having multiple governments. Research particularly supports the idea that this may be a problem where there are numerous special districts. One particularly important service provided locally is public education. Here too, fragmentation among local public school systems might sharpen competition, and thus increase educators’ attention to educational outcomes. If so, this could boost a region’s labor force quality. Thus, for the provision of some local services, fragmentation may enhance government efficiency. 4 Land Use and Economic Development A second role for cities and counties is to regulate land use, through the so-called “police power” that customarily gives them the power to create zoning ordinances, general plans, rules for subdividing land, and building regulations. Special districts can also affect land-use patterns by building growth-supportive infrastructure, such as sewer lines. This realm of local governance focuses less on the routine servicing of existing development and more on the process of development itself and the shape and character of the community over time. Given the spillover effects of local decisions relating to urban growth on surrounding areas, critics of political fragmentation often have focused on this realm of local government behavior. They allege that localities regulate land use in their narrow self-interest and slight area-wide concerns (for example, by focusing on tax-generating retail development and restricting the construction of affordable multifamily housing). Several studies also have linked increased levels of socioeconomic and racial segregation to self-interested local land-use regulation. A related issue is the local public sector’s contribution to the economic development of its region. Studies of fragmentation have seldom dealt with this realm, although there are reasons to suspect that political structure could have an effect. For example, the segmentation of a region politically might make it more difficult to fund, plan, and construct an effective regional road or transit network, or to site necessary but undesirable public facilities such as solid waste disposal sites and sewage treatment plants. 5 Equity A third area of concern is the equity issue—local government’s role in distributing and redistributing life chances. Fragmentation is often seen as separating fiscal resources from social needs. Partitioning a region into separate taxing-and-spending jurisdictions can lead to a situation in which some residents with disadvantaged backgrounds and high needs find themselves in deteriorating cities with inferior schools and public services, while other jurisdictions blessed with a superior tax base may provide services and facilities that go far beyond what a typical city can offer. Again, however, as Neiman (1982) demonstrates, community social status itself has direct effects on “tastes” or demands for public services that may lead to service disparities. That is, rich communities may simply want and demand more from their schools than poor ones. Thus, only one component of unequal service distributions is the result of government fragmentation per se, whereas another part of the inequality is attributable to different “tastes” for services. In any event, because of the difficulty in rectifying disparities among local governments, and local governments’ natural disincentive to redistribute (Peterson, 1981), typically the state and national governments are expected to step in to emphasize redistributive policies. Representation A central question about local government fragmentation is its significance for effective political representation and accountability. On one hand, greater “local control” through fragmented municipal government is often highly valued, as residents place greater faith in decisionmaking that is less insulated and closer to home. However, important issues touching on a citizen’s role as a resident of an 6 interconnected region—the health of the regional labor market, regional transportation networks, or environmental problems, for example—can lack effective arenas for debate and representation where the public sector is highly divided among numerous cities. In the case of special districts, the fragmentation of governance into different functions may have both positive and negative aspects for representation. If we sort decisionmaking and revenues for certain services into separate decisionmaking venues—for example, a water district or regional transit district—we may make programmatic and budgeting decisions about those services more transparent to citizens, who will “know what their taxes are paying for.” On the other hand, the unique ability of general-purpose governments to engage in more global budgeting—setting an overall taxing and spending level, and making tradeoffs between competing services—is sacrificed if special districts take over specific functions. Thus, a degree of coordination is lost as some revenues are earmarked for particular functions. In addition, some citizens may be baffled by the wide array of local units, some of which remain almost invisible to the public. Thus, the effect of governmental fragmentation may be more complex and nuanced than one might suspect at first. Nevertheless, the structure of local government can be expected to have a potentially very significant effect in all four realms. It is therefore important to focus on the determinants of governmental fragmentation itself. Yet the analysis of the underlying causes of fragmentation has been far less common than descriptions of fragmentation and theorizing about its effects. Why are some areas more politically complex than others? This issue will be a major focus of this report. 7 Organization of This Report Our intent in this introductory chapter has been to summarize for policy audiences what is known about the potential effects and importance of local government fragmentation. The remainder of this report examines California’s particular situation in detail. The report relies on data on government structure and expenditures from the U.S. Census Bureau and the California State Controller, along with standard demographic information from Census and state sources. The distinct governmental realm of school districts is discussed only in passing, as the report emphasizes the choices that Californians have made in providing nonschool-related public goods and services. Chapter 2 presents a descriptive analysis of the state’s local government structure and its changes over time. It also provides an account of the recurring political debate over the “proliferation” of local governments in California. Although many would make a blanket statement that California’s governance is fragmented, it would be more useful to have some relative measure of the complexity of local governmental structure in various parts of the state. This challenge is taken up in Chapter 3. Chapter 4 analyzes the reasons for variations among levels of fragmentation in different counties, and Chapter 5 offers conclusions and policy considerations. 8 2. Local Governance and Historical Change For the typical California resident, the closest and most frequent interactions with government are at the local level. A varied assortment of cities, counties, school districts, and special districts provide services, produce public goods, and demand taxes that collectively help to shape the safety, environment, personal development, and financial burdens of each resident. As discussed in the last chapter, there are good reasons to suspect that the way these governments are organized, and the degree to which authority is divided among local units, significantly affect the governance process and residents’ quality of life. This chapter summarizes California’s policy debates over local government fragmentation and then presents data regarding historical trends in the numbers of various types of local governments in the state. Although simply counting governments can be a somewhat misleading way of discerning the structure of local governance (as will be discussed in Chapter 3), historical analysis of these counts can begin to illuminate 9 the issues that reformers and policymakers sometimes raise. In particular, we can examine whether the Proposition 13 tax limitation had major effects on local government structure. A Recurring Policy Debate Although many scholars and some state policymakers have long been wary of government fragmentation, the popular instinct generally has been one of localism, decentralization, and home rule. Voters have repeatedly created new cities and special districts, in search of local control of land use, public services, and taxes. Typically, the political debate over forming these new governments has been contained inside the county or local area considering the change. Occasionally, however, such decisions have been amplified into a wider, state-level debate about the dimensions of local government structure. In at least three recent periods—in the early 1960s, in the wake of Proposition 13 in 1978, and over the past decade—state policymakers have raised concerns about local government responsibilities and complexity. Creating Local Agency Formation Commissions First, after World War II, as California grew and suburbanized rapidly, new local governments appeared in large numbers. From 1953 to 1963 alone, 75 new cities were formed. Concerned with the escalating number of local units, Governor Edmund “Pat” Brown created a Commission on Metropolitan Area Problems, asking it to consider the questions, “Do we have too many overlapping jurisdictions?” and “What is the danger point in proliferation of local government?” (Teaford, 1997, p. 69). 10 The commission recommended creating a statewide board for reviewing and passing judgment upon proposed local boundary changes. State legislators, reacting to the commission’s report but shying away from this controversial proposal, ultimately passed a compromise provision, the Knox-Nisbet Act of 1963. This law required that each county create a Local Agency Formation Commission (LAFCO). With representatives both from the cities and the county board of supervisors within each county, LAFCOs were given the power to oversee changes in government structure by regulating incorporations, annexations, special district formations, and other structural changes in their respective counties. Given only relatively “loose guidance” by state statutes (Feinbaum, 1987, p. 499), LAFCOs have shown significant variation from county to county in their approaches to jurisdictional arrangements. The overall effectiveness of LAFCOs has been widely debated, but there is no doubt that they have become key players in local intergovernmental politics (Martin, 1984, Chap. 6; Fulton, 1989; Teaford, 1997, pp. 103–107; and Miller, 1981, Chap. 5). In any event, city formation did level off in the later 1960s and into the 1970s, after the LAFCO law took effect. Worrying About the Effects of Proposition 13 Second, a fear arose that the passage of Proposition 13 in 1978 would fundamentally alter or even distort California’s local governance structures. Many observers have argued that Proposition 13 has been an impetus for the formation of new cities and special districts. This section briefly summarizes this “conventional wisdom” as to why Proposition 13 would lead to a proliferation of local governments. 11 Consider the case of cities. Citizens desiring greater “local control,” particularly over land use, always could turn to the possibility of incorporation—that is, creation of a new city. But beginning in 1978, with their property tax rates now frozen by Proposition 13, residents of outlying, unincorporated settlements had an additional incentive to incorporate: Creating a new city would not lead to an increase in their property tax bill. In the past, fears that taxes would go up after city formation allegedly had been a major disincentive to incorporation. But the formulas that the state developed after Proposition 13 for distributing property taxes among local governments within each county meant that the distribution of property tax revenues became a zero-sum game. That is, creating a new city would remove some resources from the county but would not burden property owners in that city with any new ad valorem property taxes, unless their rate had been below the 1 percent maximum rate established by Proposition 13. The LAFCO in each county was given a role in sorting out the redistribution of revenues following an incorporation. Given these formulas and rules, forming a city would allow residents to “capture” some of the tax revenue generated in their area for use in their area. In particular, this consideration could be a major incentive to form a new city if the area contained major retail activity, since a portion of the state sales tax is returned to the jurisdiction in which the sale takes place. Finally, counties were seen as being fiscally weakened by Proposition 13; thus, “escaping” from county service provision by forming a city might make sense for residents seeking higher-quality services and more financially stable government. In the case of special districts, local officials and property developers, rather than residents or homeowners, were expected to be the strategic 12 players after Proposition 13. With general-purpose government resources limited, many cities and counties were seen as less able to engage in new functions or activities, particularly those involving major capital investment. Cities and counties might also be expected to shed some of their existing burdens upon other units, where possible. In any case, creating special districts might be expected to be a convenient response to fiscal limitations. Certain types of special districts are more flexible than counties or cities in charging fees for their services, issuing debt, and localizing costs to the geographic areas expected to benefit. Districts’ relative political insulation and invisibility to taxpayers, as compared to the more prominent counties and cities, might also be viewed as an advantage by local officials and developers seeking capital facilities and services for growing populations. For these reasons, the idea arose that Proposition 13 would inexorably change the calculus of local government structure in California, leading to more formations of cities and districts. This perspective has basically been the received wisdom of those who study the state’s local governments. Scholars and journalistic commentators alike have stressed that Proposition 13 was a jarring change to the local government environment and would have major effects on local political structure (see, for example, Bollens, 1986; Brooks, 1988; Feinbaum, 1987; Fulton, 1989, 1991; Martin and Hogan, 1989; Miller, 1981; Schrag, 1998; and Sokolow et al., 1981). Focusing Renewed Attention on Local Government Problems Several other relevant events have occurred over the past decade. The late 1980s and early 1990s saw several legislative proposals to address the issue of local government complexity, particularly in relation 13 to growth and land use—at the time a dominant issue in the state’s politics. For example, the Assembly Office of Research released a report that concluded that The structure of local government clings to the ideal of self-governing, locallycontrolled communities, reinforces the divisive effects of the state’s fiscal structure and creates a mismatch between the fiscal resources and organizational means devoted to problem-solving and the increasing scale of growth-related problems (Assembly Office of Research, 1989, p. 4). In particular, the report highlighted the purported inadequacies of single-purpose government entities and suggested eliminating them in favor of a Regional Development and Infrastructure Agency in each air basin. These agencies would allocate all property and sales taxes above the current base, on the basis of need. Then-Speaker of the Assembly Willie Brown introduced a bill following up on these recommendations, which would have created seven broadly empowered regional agencies.1 Although that proposal died, debate later centered on the recommendations of the California Constitution Revision Commission, created by the state in 1993 and given the charge of examining state/local relationships. The commission reported that the organization and provision of local government services is far too complex . . . . [T]he confusing array of local government entities makes it difficult for citizens to understand which agency is responsible for providing a particular service and whom they should hold accountable. Additionally, with so many local entities involved in providing government services, redundancies often exist (1996, p. 71). ____________ 1Other reform efforts during this period included bills proposing local revenuesharing and a state planning agency, as well as regionalization proposals from the Los Angeles 2000 committee and the Bay Vision 2020 reform group. 14 Reacting to the commission’s findings, both Governor Wilson and the Legislature have discussed “realigning” the roles of the state, its counties, and its cities in providing certain services. In 1997, the Legislature passed a law in response to a different concern—the continuing desires for independence among many residents of the San Fernando Valley in the city of Los Angeles. AB 62 removes the power of city councils to veto secessions of city territory. (Voters in the affected areas must approve the detachment.) This statute applies not only in Los Angeles but throughout the state.2 And in what could ultimately be a broader action, the same session saw the Legislature pass a law creating a new Commission on Local Governance for the 21st Century. Finding that “nearly 35 years have passed since legislators last conducted a thorough investigation” of local government organization, the statute authorizes the commission to make “a careful study of local agency organization and boundaries,” and to make recommendations concerning the organization of local government and the state laws operative in this area.3 Thus, the Legislature returned once again to the issues occupying it in 1963. Frequency of Different Categories of Local Government If one believes that Proposition 13 led to the creation of many more local governments, one might also expect that California, the first state to ____________ 2AB 62 creates a Special Commission on Los Angeles Boundaries to study detachment proposals for the Valley but places this commission under the control of the county’s LAFCO. See Fulton (1997). 3The Cortese-Knox Local Government Reorganization Act of 1985 is the current basis for municipal organization, district reorganizations, and LAFCO procedures. It consolidated and made consistent the rules in this area, which previously were governed by three separate statutes. See Assembly Committee on Local Government (1996). 15 implement a constitutional tax limitation of this sort, would be among the most politically fragmented states. An initial glance at the numbers regarding local governance in California would probably lead most observers to consider the state’s local government structure to be very complex. Using surveys of numbers of governments as reported by the State Controller and the U.S. Census Bureau, the most recently available counts of local governments include 58 counties, 471 cities, 4,874 special districts (many of which are under the control of counties or cities), and 993 school districts.4 California is the nation’s most populous state, however, and one would therefore expect it to have many governmental units. To make a relevant comparison to other parts of the country, we need a per-capita measure of local governments. In these terms, however, the Golden State does not seem to have a particularly complex local public sector in comparison to other states. Table 2.1 compares the number of local Table 2.1 Local Governments per 100,000 Population: California Compared to the Rest of the United States 1992 California Other 49 states Counties Cities 0.19 1.49 1.33 8.39 Special School Districts Districts 9.05 12.83 3.49 5.95 1977 California Other 49 states 0.26 1.85 9.96 1.51 9.35 12.02 SOURCE: Census of Governments (1977, 1992). 4.96 7.12 ____________ 4The county and city counts are as of June 1998; the school and special district counts date to 1995. 16 governments per 100,000 population in California to those numbers in the remaining 49 U.S. states, using data from the 1977 and 1992 Census of Governments. (The Census of Governments counts only those special districts that are freestanding, politically independent units; see Appendix B for details.) This comparison reveals that California in both years had many times fewer counties and cities per capita, and significantly fewer school districts and politically independent special districts per capita, than the rest of the country. Moreover, the table also shows that California has become less fragmented in relation to the other 49 states since the passage of Proposition 13. For example, California had 17.1 percent fewer special districts per capita than the other states in 1977, but 29 percent fewer in 1992. What can be said about the number and growth of each of the four main types of local governments in the state? Counties California has 58 counties (including San Francisco, which is a joint city-county that is unique in the state). Counties have been the most stable element of the local government architecture. There have been no new counties since 1907, when Imperial County was formed out of San Diego County, although there were several unsuccessful attempts at “county secessions” in the 1970s and 1980s (see Detwiler, 1996b, p. 3). Counties have long been the basic administrative agents of the state in local areas and thus carry out state functions as well as local functions. They are much less susceptible to territorial change than other units of local government. 17 Cities As of 1997, there were 471 incorporated cities in California. A number of observers have pointed out what was portrayed as a flurry of incorporations in the early to mid-1980s, typically seen as a consequence of Proposition 13. Feinbaum (1987, p. 497) wrote that “California cities are being formed more often today than at any time since the mid1960s . . . . [M]ost experts see the trend persisting as the state’s population continues to grow and to shift from established cities and rural areas to rapidly developing suburbs.” Feinbaum viewed the reasons for these incorporations as primarily involving dissatisfaction with county land-use decisions, attempts to capture locally generated tax revenues, defense against annexations, and/or the emergence of community leadership and organizations. Similarly, Martin and Hogan (1989, p. 20) wrote of a “trend toward city incorporation associated with the passage of Proposition 13,” pointing out that 35 new cities formed in the decade after 1978, compared to 18 in the previous decade. “In the environment created by Proposition 13,” they argued, “local control becomes an opportunity for a community to better its position relative to other communities,” citing the creation of redevelopment agencies as well as property and sales tax capture as motivations (p. 20). Fulton (1991, p. 238) concluded that “Since Proposition 13 eliminated the possibility of tax increases, voters have realized that incorporation is ‘free.’ Thus, since 1978, the number of new cities has been on the rise, accelerating the financial problems of counties.” In another study, Musso (1994, p. 38) wrote that “the passage of Proposition 13 strengthened the fiscal incentives for incorporation.” (See also Brooks, 1988; and Schrag, 1998, pp. 180–182.) 18 It is true that 54 new cities have incorporated since Proposition 13, a number that sounds large to many. More than three-fifths of these new cities have been founded in the metropolitan areas of Southern California, in the counties around Los Angeles and San Diego. Table 2.2 lists the number of new cities by county, showing that the rapid-growth Inland Empire county of San Bernardino led the way. Appendix C Table 2.2 Number of New Cities by County, 1978–1997 County San Bernardino Los Angeles Riverside Orange San Diego Contra Costa Santa Barbara Alameda Butte Kings Lake Mono Napa Nevada Placer Sacramento San Joaquin San Luis Obispo San Mateo Shasta Sonoma Ventura Yolo Incorporations 9 8 7 5 4 3 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 Total 54 SOURCES: Compiled from California State Controller, Financial Transactions Concerning Cities of California (annual). 19 500 450 400 350 300 250 200 150 100 50 0 1850 1870 1890 1910 1930 SB 1559 passed Prop. 13 passed LAFCOs created Local sales taxes authorized Contract cities pioneered 1950 1970 1990 Figure 2.1—Number of Cities in California, 1850–1997 provides a list of all of the cities in the state that have incorporated since Proposition 13. Is it the case, then, that Proposition 13 led to unprecedented growth in the number of cities in California? Not really: Figure 2.1 shows that the long-term trend toward a greater number of cities did not experience any major or sustained blip upward in its trajectory following Proposition 13. Although there has been a certain amount of variation over time, the number of cities in California has been on a seemingly inexorable upward march since the state’s founding. Rates of increase since 1978 have not been especially high by historical standards. Looked at over this longer historical period, the two major periods for growth in the number of cities in California were from about the turn of the century through 1930 and then again in the 1950s and early 1960s. The first period coincides with a great influx of migrants into California and a time when many parts of the state began to take shape as 20 modern settlements. The state constitution of 1879 had enabled local residents to vote for incorporation of a city, rather than having to receive a special act of the Legislature for incorporation, which was typical earlier in the state’s history. The second era coincides with the push of massive postwar suburbanization in the state. Also in this period, the new City of Lakewood initiated a contracting arrangement for services in 1954, allowing new cities with minimal public employment to form; these “contract cities” arrange with their counties, and other public and private service providers, for most of their services. The “Lakewood plan” proved an attractive way for suburban residents to gain the land-use powers of cityhood without incurring the costs of creating a full-service municipality. Finally, another reason for the growing number of cities at this time was the state’s passage of the Bradley-Burns local sales tax in 1956. This provision opened up an important new source of revenue for local governments, making cityhood more viable in many communities with extensive retailing. Thus, some eras saw a great increase in cities. However, two policy changes at the state level may have caused a moderation in the “proliferation” of cities. As one can see in Figure 2.1, the creation of LAFCOs in 1963 was followed by a noticeable reduction in the rate of growth of municipal formations, despite the continued rapid growth of the state’s population. More recently, passage of SB 1559, a 1992 law requiring that incorporations be “fiscally neutral” for counties and other affected local governments, has been followed by a period in which new incorporations have virtually ceased. SB 1559 was motivated by the continuing challenges that counties faced as new cities incorporated and thereby 21 removed some of the county’s tax revenues. Some cities allegedly formed to capture sales tax revenues of large retail complexes, which previously had been providing sales tax revenues for the county. Although state LAFCO law had always directed the commissions to consider issues of fiscal effects on other local governments, LAFCOs in different counties took varying approaches to this issue. The 1992 law provided a stricter rule, making it far less likely that LAFCOs could approve incorporations resulting in revenue windfalls. Incorporations that are fiscally harmful to the county involved are to be disapproved—unless the city and county can come to an agreement beforehand. Proposed new cities now engage in protracted negotiations with counties and LAFCOs over the future distribution of revenues from property and sales taxes, and sometimes other sources such as vehicle license fees.5 Pro-city activists and legislators claim that the law is vague and makes incorporations virtually impossible. Such disgruntlement has led to recent efforts to amend or repeal the “revenue neutrality” law (Fulton, 1998). Despite the nearly continual increase in the number of cities in the 20th century, it is important to stress that the overall trend since about 1910 has been toward a decreasing number of cities per capita, as illustrated in Figure 2.2. In other words, California has generally accommodated growth through the expansion of existing cities rather than the creation of new ones. Table 2.3 provides another illustration of this phenomenon, focusing on population growth in different types of jurisdictions in the period since Proposition 13. As the table shows, ____________ 5There have been continuing disputes between Citrus Heights and Sacramento County, and between Shasta Lake and Shasta County, over their respective revenue agreements. (These cities represent two of the mere three post-SB 1559 incorporations.) 22 9 8 7 6 5 4 3 2 1 0 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 1997 Figure 2.2—Number of Cities per 100,000 Population in California, 1900–1997 almost three-quarters of the state’s population growth from 1978 to 1997 took place in cities that were already in existence in the former year. By contrast, far smaller proportions of the population increase were accommodated in new cities and in unincorporated areas. In fact, the Table 2.3 How California Accommodated Population Growth After Proposition 13, 1978–1997 Cities that existed before 1978 Cities incorporated after 1978 Areas that remained unincorporated Net Growth in Population 7,285,635 1,743,890 885,265 % of State’s Total Population Increase 73.5 17.6 8.9 SOURCE: Calculated from annual city population estimates of the California Department of Finance, Demographic Research Unit. 23 share of the state’s population living in unincorporated areas dropped from 25 percent in 1978 to 20 percent in 1997. The augmented population of existing cities has been mainly accomplished through annexations. Contemporary annexations draw less attention than some of the more massive city expansions of the past—such as Los Angeles’s annexation of the San Fernando Valley in 1915—but the incremental process of adding to city borders continues to be the main way that expanding and urbanizing populations receive public services. Miller (1993, p. 107), drawing upon the Census Bureau’s Boundary and Annexation Survey, found that California cities had engaged in 5,101 annexation actions between 1980 and 1990, adding 726 square miles and about 283,000 residents to their boundaries. The same source lists 148 detachments of territory from cities, involving only 13 square miles and about 1,000 residents. As Detwiler (1996b, p. 5) points out, the current effort at secession in Los Angeles’s San Fernando Valley runs counter to a hundred years of municipal expansion, and those who advocate dividing cities must contend with a long statutory tradition that favors the status quo. Cities shed small parcels from time to time, but detaching large areas with significant populations is quite rare. There are only two examples of successful secession—Coronado in 1890 and Montebello in 1920—while history records over a dozen former municipalities swallowed by larger cities. The impulse for municipal entropy may be growing but state law still reflects the political power of existing cities. However, this political relationship may have changed with the passage of AB 62 in 1997, which allows areas of cities to detach without the permission of the existing city’s council. This new law was motivated by the desire of some residents of the San Fernando Valley to carve a new city out of the City of Los Angeles. 24 Special Districts Historical data on special districts are spotty by comparison to city data, but their numbers also have drawn attention as a reaction to Proposition 13. For example, Bollens (1986) argued that special districts increased in number rapidly after Proposition 13. He concluded, “It is quite probable that the ability of LAFCOs to decrease district formation has been curtailed by the response to Proposition 13 at the local level” (p. 122). Morgan and Chapman (1994) have studied the dynamics of special district creation and consolidation over time in California. They note that the total number of districts counted by the State Controller, which includes both independent and dependent units, has been relatively constant in the past two decades at about 4,900, increasing by about 6 percent in the period since Proposition 13. Because of changes in methodology, special district counts by the controller before 1975 cannot be reliably compared to those of the subsequent period (Morgan and Chapman, 1994). In 1992, the Controller reported 4,995 special districts, of which 3,139 were considered “independent” and 1,856 were governed by county boards of supervisors or city councils.6 The Controller makes a count of all entities labeled special districts, whereas the Census Bureau uses a more restrictive definition, counting only units that are fiscally and politically freestanding of cities and counties. (Refer to Appendix B for more detail about the variations in ____________ 6There were 233 Mello-Roos districts in 1992—facility financing districts that are all dependent on general-purpose governments. Eighty percent of these were not included in the Controller’s report, but their overall financial activity is relatively insignificant in the context of the California local public sector. There were also 81 transportation planning agencies and 381 community redevelopment agencies. See Shires and Glenn Haber (1997). 25 counting special districts.) The Census Bureau’s count of politically autonomous special districts, taken every five years, increased from 2,228 to 2,797 between 1977 and 1992. However, the Census included a substantial number of districts in these tallies that reported no expenditures, according to its government finance data files.7 If one nets out the zero-expenditure districts to focus on active districts, the tally instead shows a rise from 1,855 in 1977 to 2,250 in 1992. Table 2.4 compares the Controller’s tally of active districts with the more delimited Census enumeration for years in which the Census of Table 2.4 Number of Special Districts in California Year 1977 1982 1987 1992 1995 Controller’s Count 4,587 4,815 4,962 4,857 4,874 Census Bureau’s Count 1,855 2,111 2,549 2,250 N/A NOTE: The Controller’s count is adjusted for consistency in definition of special districts, drawing upon the work of Morgan and Chapman (1994). The Census count is adjusted by deleting districts with reported expenditures of zero. For an explanation of the differing counts and definitions between the Controller’s office and the Census Bureau, refer to Appendix B. SOURCES: California State Controller, Financial Transactions Concerning Special Districts of California (annual); Bureau of the Census, Census of Governments (1977, 1982, 1987, 1992); Morgan and Chapman (1994). ____________ 7This counting of presumably inactive districts is one of the major complaints about Census of Governments data on districts. See Appendix C and Leigland (1990). 26 Governments was taken.8 Both show increases from 1977–1987, with a slight reduction in 1987–1992. The differences between the Census Bureau’s rate of growth in the number of districts, as compared to the somewhat flatter trend reported by the Controller, can be explained in part as a result of the changing types of special districts in the state. “Decreases in many of the traditional categories of special districts have been more than offset by significant increases in the creation of county service areas (CSAs) and joint powers [agencies]. . . . Had these ‘governments created by governments’ been held steady, the total number of special districts would have declined more than 7% since 1978” (Morgan and Chapman, 1994, p. 2-i). The Census of Governments treats most county service areas as dependent entities and thus does not count them. It does, however, count most joint powers agencies as independent because they are not controlled by an individual city or county government, but rather by a separately constituted joint board. This delineation is certainly debatable.9 Bollens (1986, p. 122) also notes the change in type of districts, finding that districts reliant on property taxes were decreasing in number, whereas enterprise districts—those that rely on user fees and charges—were increasing. Even where the governing boards of special districts tend to consist of county or city officials, such “governments created by governments” might be viewed as increasing fragmentation in a certain sense: They ____________ 8The 1997 Census of Governments has not yet been released. 9Some may view joint powers agencies as actually reducing political fragmentation, given that they are cooperative efforts among existing local governments. Others have expressed reservations about the fairness of representation of “member” governments on the governing boards of joint powers agencies, and the lack of political visibility of such units. 27 segregate funds and programs from the overall, general-purpose local budget. This means that decisions about district-funded activities tend to be made separately from overall budgeting decisions, and tradeoffs among public priorities are made less clear. Public officials periodically have expressed interest in consolidating special districts, and the Legislature has passed laws making consolidations potentially easier. Nevertheless, such mergers are relatively infrequent. Sometimes city incorporations lead to the absorption of special districts, particularly in full-service cities with minimal contracting. In this sense, a new city that reduces reliance on independent special districts might actually be viewed as a reduction in fragmentation. Overall, however, once created, districts—like most governmental organizations—have incentives to stay in business and retain their offices. Morgan and Chapman (1994) found that enterprise districts have little incentive to consolidate unless they plan to engage in a significant expansion or capital investment. Some nonenterprise districts, on the other hand, find themselves fiscally strapped by the formulas for property tax distribution promulgated for each county after Proposition 13. Nonenterprise districts in growing areas, with much property development and property turnover, are more likely to have experienced some growth in revenues, whereas such districts in “built-out” areas are less likely to see much growth in property tax revenues. Morgan and Chapman suggested that consolidations between growth-area districts and built-out area districts may be wise in some circumstances, but that districts in growth areas often lack incentives to combine with the less fortunate districts. In any event, the authors found after compiling a number of case studies of consolidations that there is little reason to 28 believe that consolidating districts is a way to rein in the cost of government. School Districts Though not the major focus of this report, the experience of school districts is worth mentioning as a notable counter-example to the trend of growth in local governments. As Figure 2.3 shows, the state experienced a significant drop in the number of school districts between the 1962 and 1992 Census of Governments, most of which occurred in the 1960s. This decline in the number of school districts in California reflects an even steeper decline in the United States as a whole. Throughout the nation, many rural and suburban districts were consolidated into large “modern” school districts at the instigation of the 1,700 1,600 1,500 1,400 1,300 1,200 1,100 1,000 1962 1967 1972 1977 1982 1987 1992 Figure 2.3—Number of School Districts in California, 1962–1992 29 states (Nelson 1990, p. 444); and elementary and high school districts often have merged into unified districts. Such consolidation, recommended by professional educator groups and others who sought to modernize and rationalize public education and curricula, shows that at least in this case, local government structure can in fact be changed relatively quickly when there are powerful interests dedicated to doing so. Shares of Local Public Spending by Different Types of Government Counting governments can illuminate historical changes and comparisons, but other measures of local government structure can also be telling. One such measure is the share of total local spending accounted for by different types of local governments. Figure 2.4 shows the shares of total local nonschool expenditures in the state accounted for by counties, cities, and special districts in the five most recently reported Censuses of Governments. As the figure illustrates, these shares did not show much deviation from 1972 to 1992, although counties surpassed cities in overall volume of expenditure. Counties appear to have declined slightly in budgetary significance between 1977 and 1982, a period during which special districts increased in significance by a similar (small) amount. This may indicate that some activities were turned over from counties to special districts in the immediate wake of Proposition 13. The period after 1982, however, shows very little change. In any event, the passage of Proposition 13 in 1978 once again cannot be charged with any clear and major alteration of governmental responsibilities. This concept—how divided the expenditure “pie” is—moves us toward a different summary measure of local government fragmentation. 30 50 45 40 35 30 25 20 15 10 5 0 1972 Counties Cities Special districts (Census definition) 1977 1982 1987 1992 Figure 2.4—Percentage Shares of Total Local Nonschool Expenditures, by Type of Government It focuses upon the dispersion of responsibilities and activities across governments. In Chapter 3, a summary measure of fragmentation for each county, based on this dispersion concept, will be introduced. Summary California has a profusion of local governments but far fewer for its population size than might be anticipated by comparison to the rest of the nation. City governments and special district governments have increased in number in recent decades, although the trend lines for special districts are relatively flat. County governments have stayed constant in number, whereas school districts have declined. Moreover, the notion that Proposition 13 led to major changes in local government formation patterns appears mistaken. City formation did not depart notably from its long-term trend, and was easily outpaced 31 by population gain, as most population growth was accommodated within existing cities. Some special districts were created, particularly joint powers agencies—which attempt to finance public goods for a larger-than-city area, and county service areas—agencies that provide localized financing for localized services. But these formations were almost matched by a dissolution or consolidation of other districts involved in traditional types of service provision—particularly those dependent on property taxes. In addition, the statewide shares of total local expenditures by type of entity, for purposes other than education, did not see much of a shift in the Proposition 13 years. Counties and cities each accounted for about 40 to 45 percent of such expenditures during this period, with special districts responsible for the rest. 32 3. Local Government Fragmentation in California: Measurement and Trends Chapter 2 provided historical data on numbers of local governments in California. Although useful, raw counts of this type can be somewhat misleading and should not be used as the only indicator of political structure in the state. This chapter presents an alternative measure—the political fragmentation index—and illustrates variations among California’s counties in local political structure using this measure. Problems in Measuring Local Government Structure by Counting Governments A hypothetical example illustrates potential problems with merely counting units of government. First, consider the incorporation of a fullservice, large-scale city, as opposed to the incorporation of a minimal contract city that continues to depend upon special districts and the county for most services. The first would be a major alteration in the 33 government responsibilities in its area—control over police, fire protection, and other key services as well as land-use controls; the second would represent a shift of land-use powers from county to city, but little else. Yet a count measure of local political structure would register both incorporations in the same way—as one more unit of government. Similarly, a small mosquito-control district would be counted equally with a large public transit district. Other researchers have similarly warned observers against an obsession with counting local governments. Morgan and Chapman (1994, p. 2), in a report sponsored by the Association of California Water Agencies, wrote, “critics have periodically called for a reduction in the number of districts in an effort to reduce the amount of government in California. They have confused the concept of too much government with that of too many governments” [emphasis in original]. Indeed, it is possible that creating a new special district to carry out a service, under some circumstances, would lead to increased efficiencies—for example by linking benefits received with willingness of users to pay for a service, or by contracting with a private firm to carry out a service. Another warning came from Foster (1993, p. 530), who wrote, “where ‘more units’ is interpreted as ‘more fragmentation,’ researchers obscure the fact that, depending on its service area, an additional government may integrate rather than fragment . . . .” That is, regional entities such as planning agencies or mass transit authorities can take a hard look at conflictual regional issues that existing local units might prefer to ignore. Many researchers studying local political structure move one step beyond counting governments by using ratio-based measures—for example, local governments per 10,000 population, or local governments 34 per square mile. This approach does provide information about the relative density of local governments. But it still tells us precious little about the division of responsibilities and activities among these units. An Index of Local Political Fragmentation A measure of local political structure should provide interpretable information about the relative responsibilities and expenditures of counties, cities, and special-purpose governments.1 This would provide a better sense of how geographically and functionally divided local public authority is. For this reason, I computed for each California county a political fragmentation index, or PFI, which represents the dispersion of local expenditures among the county, city, and special district governments.2 Data on total direct expenditures (not including intergovernmental expenditures), which provide a sense of the amount of activity engaged in by each unit, were tallied from the Census of Governments. The PFI will range between zero and one for each county. It is interpreted as the probability that any two randomly selected dollars of local public expenditure are spent by different units of government. The PFI is calculated by summing the squared percentages of total expenditure ____________ 1School districts are omitted since they involve a qualitatively different realm of local government activity. Also, there are different bases and incentives for their formation and consolidation. 2This measure is adapted from the political fragmentation index used in Lewis (1996, Chap. 3)—with some changes. Counties rather than metropolitan areas are used as basic units of analysis to maximize the number of observations and cover all of California. Also, the index in the earlier study was used to measure political fragmentation across states and thus was forced to include a weighting scheme reflecting the differing responsibilities of the local and state levels in different states—a complication not necessary in this California-specific study. Finally, the earlier study included only those special districts relevant to land use and development, whereas this study includes all special districts counted by the Census. 35 accounted for by each unit of government (expressed as a decimal) and subtracting this number from one. This is a variant of a widely used social science technique, used to discern how dispersed something is across separate units.3 Counties’ score along the PFI’s zero-to-one range indicates the degree to which they have divided up their expenditure “pie” among different governments. For example, consider the following hypothetical cases: • County A includes a county government that spends 50 percent of total local expenditures, two cities that each spend 20 percent, and a special district that spends 10 percent. In this case, PFI = 0.66. • County B has a more complex governing structure. It has a county government that spends 30 percent of the total, four cities that each spend 15 percent, and two special districts each spending 5 percent. Here, PFI = 0.815. In the first case, there is a 66 percent chance that any two randomly selected dollars are spent by different governments, whereas this probability is 81.5 percent in the case of County B.4 In this study, the county is used as the unit of analysis; within the county, data on county, city, and special district expenditures are included. This provides 58 cases to work with in California—or actually ____________ 3For example, the same type of index has been used to compare nations on the basis of the dispersion of votes across different political parties. Social scientists may recognize the PFI as a Hefindahl index, subtracted from one. 4Note that there are two factors that can lead to a higher PFI value mathematically. If spending shares are assumed to be equal across governments, then raising the number of governments will lead to greater dispersion. Similarly, if we assume that the number of governments is fixed, then PFI increases as spending becomes less concentrated and more dispersed. Thus, both a greater number of units and less concentration of spending generally will tend to increase fragmentation. 36 57, since the unique City and County of San Francisco is omitted from much of the analysis to follow. “Counties are appropriate geographic units because they are typically large enough to permit some jurisdictional choice, but small enough to allow changes in [institutional arrangements] without major changes in workplace location or social patterns” (Zax, 1989, p. 561).5 For the purposes of this study, dependent special districts are not treated as separate units of government. As explained in Appendix B, the Census Bureau includes the expenditure totals of such districts as part of their “parent” county or city. This makes sense because, conceptually, the PFI measure relates to the dispersion of political authority. Dependent districts may have separate financial statements, but they are in effect merely the tools or accounting devices of their parent governments. Levels of Local Political Fragmentation Among California Counties Table 3.1 lists the PFI value for each county in the state as of 1992, the most recent year for which Census of Governments data are available. Figure 3.1 shows the geographical patterns involved. There is clearly a substantial degree of variation in local political structure within the state, ranging from a PFI of 0.05 in Alpine County, a rural county in the Sierra, to a PFI of 0.85 in Alameda County, a largely urbanized county in the San Francisco Bay area. ____________ 5I do not weight the counties by population—rather, in Chapter 4, I include population as an independent variable that may influence political fragmentation. My intent is to generalize about all counties in the state (except San Francisco). 37 Table 3.1 Political Fragmentation Index of Local Government Structure in California Counties, 1992 County Alameda Solano San Mateo San Diego Orange Riverside Los Angeles Contra Costa Monterey Stanislaus Marin Tulare Santa Clara Imperial Sacramento Nevada San Bernardino Mendocino Yolo Ventura San Benito Kings Sonoma Santa Barbara Butte Kern Santa Cruz Sutter Fresno Plumas PFI County PFI 0.85 Inyo 0.63 0.81 Shasta 0.63 0.81 Humboldt 0.63 0.79 San Joaquin 0.63 0.78 El Dorado 0.62 0.77 Madera 0.60 0.76 Glenn 0.60 0.76 San Luis Obispo 0.58 0.75 Napa 0.53 0.75 Lassen 0.52 0.75 Lake 0.51 0.73 Colusa 0.51 0.73 Mono 0.48 0.72 Merced 0.47 0.71 Del Norte 0.46 0.71 Calaveras 0.46 0.71 Siskiyou 0.46 0.70 Amador 0.43 0.69 Tuolumne 0.36 0.68 Yuba 0.35 0.68 Tehama 0.34 0.67 Mariposa 0.32 0.66 Placer 0.31 0.65 Sierra 0.31 0.65 Modoc 0.30 0.65 Trinity 0.21 0.64 Alpine 0.05 0.64 Average 0.59 0.64 Median 0.64 0.63 Standard deviation 0.17 What distinguishes high- and low-fragmentation counties? A detailed analysis of factors associated with fragmentation will be presented in Chapter 4. For now, it is worth looking in more detail at the measurements that constitute the index. Counties with the lowest levels 38 Del Norte Siskiyou Modoc Humboldt Trinity Shasta Lassen Mendocino Tehama Glenn Butte Colusa Lake Sonoma Napa Sacramento Yolo Solano Marin San Francisco Alameda Stanislaus San Mateo Santa Cruz Santa Clara San Benito Political Plumas Fragmentation Index Sierra Yuba Nevada Placer El Dorado Sutter Alpine Amador Calaveras San Joaquin Tuolumne Contra Costa Mono Mariposa Less than 0.5 0.5 to 0.65 0.66 to 0.75 Greater than 0.75 Merced Madera Fresno Inyo Monterey Kings Tulare San Luis Obispo Kern Santa Barbara Los Ventura Angeles San Bernardino Orange Riverside San Diego Imperial Figure 3.1—Levels of Political Fragmentation in California Counties, 1992 of PFI tend to have a county-dominant governmental structure with relatively few sizable cities, and low reliance on special districts. As Table 3.2 illustrates, the 11 counties (or roughly 20 percent) that had the lowest fragmentation scores in 1992 spent 22.5 percent of their local nonschool expenditures through cities and special districts; whereas the 11 highest-fragmentation counties had cities and special districts responsible for 57.7 percent of such expenditures. As one might expect, 39 Table 3.2 Average Percentage Shares of Nonschool Expenditure in High- and Low-PFI Counties County Cities Special districts High-PFI Counties 42.2 34.4 23.3 Low-PFI Counties 77.5 9.8 12.7 57-County Average 58.0 22.3 19.7 SOURCE: Author’s calculations from Census of Governments data (1992). the low-fragmentation counties tend to have a relatively small portion of their population in incorporated cities: They average 77.5 percent unincorporated population, compared to just 16.7 percent unincorporated, on average, in the high-fragmentation counties. For a specific example, Table 3.3 compares two counties, Sonoma and Solano, both located in the northern part of the San Francisco Bay area. Sonoma has a larger population and more units of local government, so one might expect it to have a higher political fragmentation index. In fact, however, its PFI in 1992 was significantly lower than in nearby Solano, which had the second-highest level of Table 3.3 A Comparison of Local Political Structure in Two Counties, 1992 Population % unincorporated County government’s share of total spending, % Number of cities with >2% of total spending in county Number of special districts with >2% of total spending in county Political fragmentation index Sonoma 400,900 41.0 54.7 8 4 54 2 0.661 Solano 358,900 5.5 35.1 7 5 48 3 0.810 40 political fragmentation in the state. As the table shows, more of the cities and special districts in Solano have a substantial budgetary presence, with 2 percent or more of the total local nonschool expenditures in the county. Even more important in accounting for the different levels of PFI, in this case, is that Solano has a very small share of its population living in unincorporated areas, which means that the county government has a limited role as a local service provider. In other words, several units of local government in Solano have a major decisionmaking presence, and the county government is just one government among many, as opposed to the more dominant role the county has in Sonoma. Thus, by this measure, Solano’s local government structure is considered more fragmented, because expenditures are more evenly divided among governments. This tale of two counties illustrates some of the accounting that determines differences in PFI, but it does not show the underlying characteristics of counties that might explain those variations in PFI. That issue will be addressed in the next chapter. Did Proposition 13 Accelerate Fragmentation? One issue to be dealt with before examining the reasons for crosssectional variations in local fragmentation is the longitudinal issue. That is, have levels of PFI increased over time? Was there a noticeable and sustained blip upward in levels of PFI in the wake of Proposition 13? Recall from Chapter 2 that Proposition 13 did not have any obvious effect on the numbers of cities and special districts statewide, though it appeared to have altered the relative frequency of different types of special districts somewhat. 41 The PFI data for 1972 to 1992 are consistent with those findings. Change over time has been very slight, as Figure 3.2 shows. The lower line in the figure shows the average PFI values for the 57 counties in the Census of Governments years. It shows a slight dip in fragmentation in the period just before Proposition 13 and a gradual increase in the period since. However, this line is a bit misleading in that it averages all counties, from tiny Alpine to huge Los Angeles. The middle line solves this problem by presenting a population-weighted measure of PFI for the 57 counties. Finally, the top line of the graph shows PFI for the state as a whole—that is, with the state rather than counties as the unit of analysis. In each case, the lines are quite flat, indicating that fragmentation did not vary much over time. In particular, the period from 1982 to 1992 showed very little change in PFI. 1.0 0.9 Statewide PFI Population-weighted county PFI Average county PFI 0.8 0.7 0.6 0.5 0.4 1972 1977 1982 1987 1992 Figure 3.2—Political Fragmentation Index in California, 1972–1992 42 The lines on the graph support the conclusion that Proposition 13 did not result in a major, jarring shift in the financial responsibilities between counties, cities, and special districts. Nor does it appear that government formations after Proposition 13 led to marked changes in local political structure. Do these statewide measures conceal important differences among counties in their PFI changes over time? There is little evidence to believe so. Most of the counties held quite constant in their PFI indices over this period, generally moving by a few percentage points or less. Moreover, some counties increased slightly and some decreased slightly between each Census of Governments, with later changes in each county often canceling out the earlier changes. Some of the very-smallpopulation counties experienced greater shifts in political fragmentation but these changes were not uniform in direction. There is reason to expect greater shifts in small counties, since with few units of government, any addition or deletion of units leads to bigger relative changes in spending responsibilities than in more complex counties. The one region where there was some minor evidence of a trend was in the greater Los Angeles region, where as seen above the lion’s share of new city incorporations took place. Figure 3.3 compares the overall PFI for the five-county metropolitan Southern California region to the ninecounty San Francisco Bay area.6 It shows that there has been some convergence in the former region toward the high level of fragmentation ____________ 6This figure uses the metropolitan areas as units of analysis. The Southern California region includes Los Angeles, Orange, Riverside, San Bernardino, and Ventura Counties. The Bay area includes San Francisco, San Mateo, Santa Clara, Alameda, Contra Costa, Solano, Napa, Sonoma, and Marin Counties. The higher number of counties per capita in the Bay Area is one major reason why its collective PFI is higher than in Southern California. 43 1.0 0.9 0.8 0.7 0.6 0.5 0.4 1972 1977 Bay area Southern California 1982 1987 1992 Figure 3.3—Political Fragmentation Index in California’s Major Metropolitan Areas in the latter, with a particular increase in the 1977–1982 period. Still, Southern California was growing extremely rapidly in population throughout this period, and its PFI increase hardly seems unprecedented. Moreover, levels of PFI barely budged in the region between 1982 and 1992. As Chapter 4 will show, regression analysis indicates that neither the Bay area nor Southern California is distinctive in its PFI levels, once we have controlled for other relevant factors. Why the lack of change overall? As was shown in Chapter 3, although there have been numerous incorporations in California, most new growth has been accommodated by the gradual growth of existing cities. Thus, annexations increase the relative spending share of some established cities, even as new, generally smaller cities are added. Some special districts have been created, but others have folded or decreased in importance. Finally, counties have retained their major role, with great 44 programmatic responsibilities placed on them by the state and continued service provision in the unincorporated populations (although unincorporated areas are gradually declining as a share of the state’s population). A typical example of these patterns is Santa Barbara County. That county’s population grew steadily from 270,200 in 1972 to 379,400 in 1992, but its PFI values changed only slightly, increasing from 0.634 to 0.648. (The PFI actually peaked in 1982 before decreasing.) The county’s share of total nonschool local expenditures varied between 52 and 58 percent over this period but did not follow a consistent trend upward or downward. Two new cities were incorporated in the postProposition 13 era—Solvang in 1985 and Buellton in 1992, the latter too recent to show up in the 1992 Census of Governments. Both are very small municipalities, however, each with less than 6,000 inhabitants and less than 1 percent of total expenditures in the county. An overemphasis on these new units would obscure the fact that the area’s largest cities—Santa Barbara, Santa Maria, and Lompoc—each grew substantially in population and expenditures, and the county retained its dominant role. Special districts collectively accounted for between 12 and 17 percent of the area’s expenditures over this period, but no single special-purpose entity, with the exception of the Lompoc Hospital District, remained above 2 percent of the area’s expenditures. Thus, at least by this measure, the relative distribution of governmental activities stayed fairly constant over these two decades. Summary This chapter has offered a different and, it is argued, a superior method to measure local political structure as opposed to merely 45 counting governments or governments per capita. The political fragmentation index shows the division of governmental activities among local governments in each county area, using expenditure data as a proxy for each government’s activity level. California’s counties show a fair amount of variation on this PFI measure. High-PFI counties tend to be heavily incorporated, with numerous sizable cities and special districts. However, a look at PFI values over time does not indicate that California has become particularly more fragmented. Shifts in PFI have been gradual, both on the statewide level and within most counties. There is no substantial evidence to support a “Proposition 13 effect” on local government structure. 46 4. Accounting for Variations in Local Government Structure Across California The previous chapter shows that California’s counties differ extensively in their degree of local government fragmentation. This chapter examines the characteristics of counties that underlie these variations in political structure. Drawing on previous studies of local government structure, I identify some basic determinants of these patterns. The results indicate that a few, fairly deep-seated characteristics account for most of the differences among the counties in political structure. What Factors Might Affect Local Government Structure? A number of studies by social scientists have attempted to probe for the determinants of local government structure in the United States. But while growing, this vein of literature is still small and incomplete. As 47 Fisher and Wassmer (1998, p. 445) note, “Considering the theories that exist about local government structure, there has been very little empirical examination of what determines the structure of localities in an area.” The work that does exist typically attempts to explain variations in the number of local governments, across states, metropolitan areas, or counties. But such counting exercises, as noted above, can be misleading, particularly given the profusion of special districts in the Census of Governments that have little—and in some cases no— expenditure activity. This section reviews the literature on government formation and structure, looking for theories than can help us understand patterns of political fragmentation across California’s counties. Most of these studies involve statistical analyses that control for other relevant factors; some caveats are noted below. (Readers who are less interested in the theoretical ideas behind the analysis in this chapter may wish to skip ahead to the section entitled “Why Fragmentation?”). The literature review indicates that several sets of factors have been linked to local political structure. General Contextual Factors Not surprisingly, every relevant study finds that areas with larger populations tend to have more governments. In simple terms, there are more citizens demanding more things from the public sector, and thus the public sector can be expected to be more complex. One might also suspect, then, that areas adding rapidly to their population would experience more government formations. Two studies of city incorporations in California have found that “concerns about growth and 48 development” were among the most important factors in the formation of new cities (Musso, 1994, p. 2; see also Martin, 1984). However, somewhat puzzlingly, population growth is not statistically associated with larger numbers of governments—in fact, several studies find significant negative associations between population growth and government formation. Looking at the historical evidence presented in Chapter 2, we can begin to develop an explanation as to why this might be so. Simply put, the “boom” years for government formation, at least in California, were periods much earlier in the state’s history. Although recent decades have seen high rates of population growth, the number of governments has increased only relatively incrementally, above a base that was largely set decades ago. Thus, it would be wrong to say that rapid population growth “causes” a slower increase in the number of governments. Rather, we might hypothesize that government structure is heavily rooted in the past and reflects earlier periods of population change more than recent growth. Other general contextual factors that studies have included as predictors of local government formation include population density, which is thought to generate more complex public service needs, and/or measures of urbanization. On the thought that “a decentralized government structure contributes to efficiency only if residents are mobile and not restricted in employment choice by job location,” Fisher and Wassmer (1998, p. 452) also include square mileage and highway mileage in their estimations. Intergovernmental Rules and Relationships Another set of factors that have been identified as affecting local government structure might be called intergovernmental variables. 49 These include the rules that states set regarding changes in local government structure, funding and fiscal relationships among levels of government, and political conflicts at the state level. Since this study is limited to California, and most of these characteristics vary only across states, they are of limited applicability here; nevertheless, they are worth reviewing to understand how state governments set parameters for local government. For example, Nelson finds that states with more liberal home-rule authority for local governments have more municipalities and districts in their metropolitan areas, whereas states with minimum population requirements for the formation of new governments tend to have fewer local governments (1990, p. 453). States that make annexation easier experience more incorporations, probably because outlying settlements are more fearful of being annexed in such states (see Fisher and Wassmer, 1998, p. 454). States with easy enabling rules for special districts see more districts formed, as do those with spending and tax limits on general-purpose governments (Bollens, 1986, p. 122). States that permit a broader array of functions to be taken on by special districts also tend to have more districts (Foster, 1997, p. 137). State-level political conflicts, too, may be registered in local government structure. Nice (1983) finds a bivariate relationship between an index of urban/rural political conflict in a state and the number of local governments per capita in the state. He interprets this as an intentional attempt by rural-oriented state legislators to encourage the political fragmentation of major metropolitan areas. However, Marando and Reeves (1988, p. 1002), who examine state provisions for the formation of regional special districts, conclude that “as a rule there is no 50 grand design on the part of state governments to fragment metropolitan areas in order to diffuse political power.” Fiscal rules and conditions associated with a given state may also affect government formation. In particular, special districts may be formed in an attempt to shed the load that burdens municipal and county governments, or to enable local government to take on new activities that are prevented by the fiscal restrictions faced by existing units of government. In an analysis of Southern states and metropolitan areas between 1972 and 1977, MacManus (1981) finds that the creation of property-tax-based special districts is correlated with a lower rate of growth in the property taxation of other local governments.1 She writes, “Perhaps state officials, recognizing the dim likelihood of overturning archaic constitutional limitations on local government revenue sources at the height of tax revolt, ‘compensated’ for the limitations by being less restrictive in their policies toward the creation of special districts” (1981, p. 1214). Similarly, Nelson (1990) finds that metropolitan areas located in states with constitutional limits on taxation by cities have a greater number of special districts. This result was replicated at the state level of analysis by Leigland (1994, p. 540), although Foster (1997, p. 135) comes to the opposite conclusion in an analysis of metropolitan areas. On the issue of intergovernmental financial transfers, several studies have found that intergovernmental aid has a large and significant positive effect on spending levels by local governments (Schneider, 1989, pp. 60, 66). In a related finding by Stein (1987, p. 150), “compensatory state aid has a significant and negative impact on the variation in municipal service bundles within a metropolitan area,” perhaps indicating that ____________ 1The use of simple correlation analysis in MacManus’s study, and the generally weak correlations she finds, make it dangerous to generalize from the findings. 51 intergovernmental money enables areas to “level up” their service provision. Stein also notes that earlier studies found that such aid “has a negative effect on segregation of residential populations” (1987, p. 143), which could lead in turn to less sentiment for new cities, and thus less political fragmentation.2 Working in the opposite direction of these possible negative effects of intergovernmental aid on political fragmentation are the programmatic requirements that have been attached to some categories of intergovernmental money. Foster notes that “since the 1930s, federal funding for mass transportation, housing and urban renewal, natural resources . . . hospitals, and sewage treatment has often privileged specialpurpose over general-purpose governments.” Despite recommendations to the contrary from the Advisory Council on Intergovernmental Relations, “federal policies for provision of air quality, solid waste removal, airports, housing, and health services continue to encourage special-purpose governments” (Foster, 1997, p. 111; see also Bollens, 1986, p. 119). The terms of federal aid for water and irrigation facility construction also have been viewed as favoring special districts over general-purpose governments (Leigland, 1994, p. 537). Socioeconomic Characteristics of the Population One frequently suggested allegation is that new local governments are formed because of an effort by wealthier people to separate themselves into a more fiscally healthy jurisdiction—one with a high property tax or sales tax base and not heavily burdened with the service demands of the less privileged. Lending support to this line of reasoning, ____________ 2The next section will examine possible links between social/racial heterogeneity, segregation, and fragmentation. 52 Martin and Hogan (1989, p. 21), examining incorporations in California after Proposition 13, find that the new cities generally have higher per capita incomes than those of their surrounding counties. This finding was confirmed by Musso (1994, p. 186) in a comparison of unincorporated Census Designated Places in California that had attempted incorporation to those that had not. In case study analysis, Miller (1981) found tax avoidance and capture of locally generated revenues to have motivated incorporations of numerous cities in Los Angeles County even long before Proposition 13. Rather than being centered upon issues of efficient service provision, Miller argues, “the municipal incorporation of the Lakewood Plan cities was a ‘revolt of the rich against the poor’” (1981, p. 9; see also Martin, 1984, pp. 62–63, 94–95). Thus, high income levels, or a large share of high-income households, might be expected to be associated with higher levels of political fragmentation. Perhaps what matters, though, is not the absolute income level of the area but the variations in socioeconomic status—variations that might cause some areas to wish to separate from others. Indeed, economic theory suggests that areas with a great deal of social and economic diversity—or a “heterogeneity of preferences”—should see relatively more governments created. Drawing on Tiebout’s (1956) theory of a competitive market-like structure of local governments, most economists studying this issue expect “that greater variation in the economic characteristics of residents results in greater variation in both the quantity, quality, and mix of services demanded from local governments” (Fisher and Wassmer, 1998, p. 445). In turn, this varied demand is expected to translate into a propensity to create more units of government, other factors being equal. Thus, we might expect that areas 53 with populations that are more heterogeneous in income, race, and even age would be more politically fragmented. An alternative, and less charitable, theory might hold that wealthier or “whiter” communities prefer to exclude the poor and minorities by walling themselves off with political boundaries (see Weiher, 1991). In empirical studies related to this question, Nelson (1990) finds that variability in household incomes in a metropolitan area is associated with more general-purpose and special-purpose governments; he also finds, however, that variation in the age makeup of the population of these areas is associated with fewer special districts. Nelson elected to omit a variable measuring the racial heterogeneity of metropolitan areas, when he found in his initial analysis that racial diversity was associated with fewer units of local government—“a finding that is contrary to expectations . . . .” (1990, p. 452). Given the “lack of a strong theoretical basis” for this result, he chose to exclude the race variable, which may lead one to question the robustness of his income and age results. (Income and age variables were statistically insignificant when the race term was included.) Martinez-Vazquez, Rider, and Walker (1997) find racial heterogeneity associated with higher numbers of school districts—perhaps indicating a preference among voters for keeping schools more racially homogeneous. They find no association between racial heterogeneity and the formation of special districts, however. By way of explanation, they note that special district activities, unlike school districts, generally do not involve interpersonal interactions among residents. Given this seemingly reasonable logic, it is surprising that Fisher and Wassmer (1998)—confirming the result that puzzled Nelson—find that racial variation is associated with fewer general-purpose local 54 governments. They also find, however, that greater variation in the age makeup of metropolitan areas increases the number of school districts and of municipalities and townships.3 In a second paper examining longitudinal changes in numbers of local governments from 1982 to 1992, Wassmer and Fisher (1997, p. 15) find in general that “increased variation in the socioeconomic composition of a metropolitan area leads to more local governments . . . .” Political Influence of Developers or Other Businesses A number of authors have linked the creation of local governments to the influence of developers or other major economic actors in the political process. In particular, “Criticism has been directed at local special districts for their . . . use of public agencies as development tools by private interests” (Bollens, 1986, p. 118). Much of this hypothesis has rested on evidence from case studies. For example, Perrenod (1984) tied the extreme proliferation of water districts in the Houston area to overt efforts of developers to cheaply create infrastructure for each new subdivision. Lewis (1996, pp. 97–100, 142–144, 148) found that large-scale developers in the Denver area viewed the formation of special districts as simply another business deal that often was necessary to carry out their visions for suburban development. Special districts in Colorado have often been the only practical way to build infrastructure and service ____________ 3Fisher and Wassmer merge the counts of municipalities and townships, claiming that both fulfill similar roles in the federal system and thus can be considered collectively (footnote, p. 450). This claim is debatable. Townships are little more than administrative areas in some states, whereas they perform municipal service functions in other states. Treating townships and municipalities as equivalent may thus bias a count measure of political structure. The PFI measure used in this report avoids this problem. In any event, California has no township governments. 55 development in unincorporated areas. They have been turned to increasingly within cities as well, as cities became less willing to underwrite the costs of growth, preferring to offload these costs onto homebuyers. At least two quantitative studies have also attempted to link developers to government formation. Burns (1994) notes that efforts to create new governments are collective action problems that typically require organizational and financial resources, as well as leadership. Particularly in the case of districts, Burns argues, “the most obvious entrepreneurs [working to form governments] . . . are developers who want government to build infrastructure to improve land to profit the developer” (1994, p. 20). Her empirical findings suggest that the number of developers in a county is positively related to special district formations, but only for the decade of the 1960s (1994, p. 99). In her national study of special districts, Foster (1997), noting the case study literature that focused on the role of developers, includes a variable measuring the number of developer-firm equivalents per 10,000 population (based on data from County Business Patterns). Her quantitative results ascribe no major explanatory role to developers, however. Aside from development firms, the role of manufacturers in the founding of cities has sometimes been emphasized. The industrial era in American history provides a number of well-known examples of “company towns”—such as Pullman, Illinois—formed at the behest of commercial interests. Miller (1981, Chap. 2) provides case studies of several cities in Los Angeles County in the 1950s—such as Commerce, Industry, Irwindale, and South El Monte—that were formed under the 56 leadership of local industrialists, who often drew boundaries carefully to include a minimal residential population. These incorporations helped keep taxes low and regulations on industry lax. In her statistical analysis, Burns (1994, Chap. 5) does find that under certain circumstances, the presence of manufacturers in a county leads to a significant growth of municipal governments. She suggests that the ability of new cities to offer a tax haven for business is the explanation of such effects.4 Why Fragmentation? Toward a Model of Local Government Structure As we have seen, existing theories and empirical studies converge on four basic sets of explanations for local political structure: general contextual factors, including the size and maturity of the county; intergovernmental factors, which are not applicable to this California-only analysis, except for intergovernmental aid; socioeconomic characteristics, particularly the variations in the population across income, race, and age categories; and perhaps the political influence of developers or manufacturing firms. These factors involve numerous variables of potential significance, and some difficulties of specification, particularly for the developer/manufacturer concept. Given constraints of data availability, I collected data for California counties on a variety of variables, listed in Table 4.1, to explore their ____________ 4However, Burns’s use of a dummy variable for the presence of a manufacturing establishment in the county seems pointless for contemporary California, in which every county but one includes manufacturers, according County Business Patterns data. Instead, I use a measure of manufacturing establishments per 10,000 population in the analysis below. 57 58 Table 4.1 Variables Studied for Possible Relationships with Local Political Fragmentation Type of Variable General/contextual Intergovernmental Socioeconomic Interest group Variable Population Density Land area Urbanization Historical development Population growth Aridity Geographic diffusion Transportation network Intergovernmental revenues Age Income Homeownership Race/ethnicity Developers Manufacturers How Measured Natural log of 1992 population Natural log of 1992 population per square mile Square mileage of county, 1990 % of population rural, 1990 Metropolitan area (dummy variable) 1920 population as % of 1990 population 1950 population as % of 1990 population % population growth, 1972–1992 Annual precipitation (inches) Average PFI of surrounding counties Southern California metropolis (dummy variable) San Francisco Bay area (dummy variable) State highway miles per square mile, 1995 % of local revenue that was intergovernmental, 1977 % aged 65 or older, 1992, 1977 Index of variation across Census age categories Per capita income, 1989 % of households with income above $75,000, 1989 % households in poverty, 1989 Standard deviation of income across Census categories % of occupied units owner-occupied, 1990 % black population, 1992, 1977 % Hispanic population, 1992, 1977 Index of variation across racial/ethnic categories Development firm equivalents per 10,000 population, 1977 Manufacturing firms per 10,000 population, 1977 Bivariate Association with PFI + + Insignificant – + Insignificant Insignificant Insignificant – + + + + – – Insignificant + + – + – + + + + Insignificant 59 association with the political fragmentation index in 1992.5 All of the data are from standard governmental statistics.6 Most of these variables are suggested by the previous studies in this literature. However, I have also tested the effects of a number of additional variables that I considered to be potentially influential for the development of local political structure in California. For example, I included two dummy variables for county location in the San Francisco Bay area or in metropolitan Southern California to attempt to capture any patterns specific to those areas. Taking this concept of geographic diffusion a step further, I also developed a variable representing the average PFI value for the counties bordering each county, on the thought that a “norm” of governmental structure might have evolved among neighboring areas. Another variable examined was average precipitation in each county; the idea here was that more arid counties might need to develop more involved water-service provision systems, thus leading to more special districts and more governmental fragmentation. ____________ 5Initially I intended to use a pooled sample looking at changes in each county’s PFI level between each issue of Census of Governments. However, as discussed in Chapter 3, changes in these levels for individual counties are so minor and seemingly random across time that I decided to use a simpler cross-sectional design examining PFI in the 57 counties as of 1992. Some of the independent variables use lagged values from 1977, on the assumption that government structure responds only gradually over time to these variables, and to avoid potential problems of endogeneity. 6Population estimates, including ethnic/racial breakdowns, as well as statistics on county land area, are drawn from reports of the California Department of Finance’s Demographic Research Unit. Highway mileage data are from the California Department of Transportation. Precipitation data are from the National Weather Service. Intergovernmental revenue figures are from the Census of Governments. Data on the number of developers and manufacturing firms are from County Business Patterns. My construction of the development-firm index follows Foster (1997, p. 146). All other data are standard Census Bureau demographic variables from the Census of Population and Housing or the County and City Data Book. 59 Bivariate Relationships Because there are many variables of potential interest and only 57 county observations, degrees of freedom limit the number of regressors that may be introduced. Before proceeding to a multivariate analysis where most of these variables must be excluded, it is interesting to examine the simple bivariate relationships between these variables and political fragmentation. For this reason, the last column of Table 4.1 indicates whether each variable has a significant bivariate relationship with PFI, in a positive (+) or negative (–) direction. At this simple level of analysis, many variables appear related to political fragmentation, generally in the anticipated direction. Variables that appear positively associated with the PFI in California counties included measures of county population size and density, urbanization, location in the Bay area and Southern California or proximity to other high-PFI counties, income levels and heterogeneity, racial diversity, and the heavy presence of development firms. By contrast, annual precipitation, intergovernmental revenue, percent elderly population, owner-occupied housing, and poverty populations appear negatively linked with political fragmentation. Multivariate Model I next entered various subsets of these variables into multiple regression equations intended to assess their relative significance and influence upon political fragmentation. Numerous combinations were tried in specifying such models, but ultimately a very simple model with three estimator variables proved remarkably robust. With one exception, none of the other variables listed in Table 4.1 achieve or even approach statistical significance, once these three variables are included. 60 This analysis reveals that population size, intergovernmental revenue share, and historical growth trajectory are the main determinants of political structure in California counties that can be identified—in that order of importance. Population size is positively associated with political fragmentation, as is a growth trajectory in which a county achieved more of its ultimate population size earlier in its history. Counties in which the local public sector received a larger share of its funding from state and national sources were less fragmented politically. Table 4.2 presents the results of this model, which accounts for two thirds of the variation in PFI across counties. It also presents a similar second model in which the racial variation of the county population is introduced in addition to the other three terms. Racial variation here is measured with an index analogous to the PFI; it represents the probability that two randomly selected residents of a county are of different ethno-racial categories.7 This variable, which has a positive sign, was the only other variable to attain statistical significance (at the 0.10 level) once the three variables above were included.8 However, the effect of racial variation on fragmentation is not robust to alternative specifications of the model. When other county characteristics were entered into regressions—variables such as owner-occupancy of housing, rural share of population, age variation, or development-industry size— the racial variation measure was no longer significant. By contrast, the other three determinants of fragmentation identified above—population, ____________ 7The categories used are non-Hispanic white, non-Hispanic black, Hispanic, Asian and Pacific Islander, and Native American. 8The application of stepwise regression analysis—a method rightly criticized because of its atheoretical nature—also results in this four-variable model. 61 Table 4.2 Regression Models of Political Fragmentation Indexes for California Counties, 1992 Variable Population, 1992 (natural log) Model 1 0.081 (9.78) [0.835] Model 2 0.070 (6.92) [0.726] Intergovernmental revenue, 1977, as share of total revenue –0.006 –0.006 (–3.26) (–3.65) [–0.256] [–0.290] Historical growth (1950 population/1990 population) 0.002 (2.27) [0.195] 0.002 (1.86) [0.161] Racial/ethnic variation index (ranges from 0 (low) to 1 (high)) — 0.204 (1.73) [0.169] Adjusted R-squared 0.66 0.67 Number of cases 57 NOTES: Dependent variable is political fragmentation index, 1992. Regression coefficients for each variable are listed first, with tvalues in parentheses and standardized (beta) coefficients in brackets. San Francisco City and County is omitted. 57 intergovernmental revenue, and historical growth—never failed to achieve statistical significance in numerous specifications of the model.9 ____________ 9Removing the five smallest counties in the sample (those with populations under 15,000 in 1992) does not change the general magnitude, direction, or significance of the three main variables, but once again the racial variation variable falls to insignificance. Omitting Los Angeles County from the sample (by far the largest county in the state, and certainly somewhat unique) leads to virtually the same results listed in Table 4.2 but with a slightly higher R-squared. 62 Interpretation What can we make of these results? The finding that larger- population areas are more politically complex is not surprising, and conforms to previous studies. Bigger, more complex counties simply have more complicated public sectors. Larger populations increase the likelihood of additional governmental units; additional units will tend to increase PFI. Turning to intergovernmental revenue, the potential political and policy implications are more interesting. One possible interpretation is that a greater flow of revenues from nonlocal sources means that there is less need for local public sectors to engage in “creative financing” to meet their revenue needs. Thus, fewer special districts may be formed. Similarly, existing county and city governments are likely to be in healthier fiscal situations if there is more intergovernmental funding available (all other things equal). For this reason, residents of wealthier areas within the county may have less of an incentive to incorporate as a separate city, preferring instead to be under county rule (as an unincorporated area) or to be annexed by an existing city. Interpreting the influence of the historical growth trajectory is also interesting. Note that the variable used here is simply the reverse of a measure of population growth. As some scholars have done, we might conclude that areas that have grown faster in recent decades tend to be less fragmented. But there are no persuasive theoretical reasons to draw such an interpretation. Rather, I prefer to think of this measure as representing the maturity or growth history of the county. The variable represents each county’s 63 1950 population as a percentage of its 1990 population.10 The results indicate that for every 10 percent of a county’s 1990 population that it attained before 1950, its PFI would be about 0.02 higher. Although modest, this effect is consistent with the notion that most of California’s local government development occurred during earlier periods in the state’s history. In other words, earlier waves of population growth in the state’s history generated more complex structures of local governance, whereas more recent periods of growth resulted in more streamlined local political development. This is a more theoretically compelling explanation, given the historical evidence on the incorporation of cities in Chapter 2. Other studies have hinted in this direction, noting that municipalities—even small towns—created in earlier periods of history had more full-service responsibility placed in their hands. This would imply more dispersion of government responsibilities among units, and thus a higher PFI. As Schneider wrote, “In the parts of the United States settled earlier, local governments traditionally have more responsibility than in newer parts of the country, but as time passed, the expected role of local government narrowed. The distribution of functional responsibility is thus intimately tied to the historical timing of settlement” (1989, p. 49). Similarly, Burns (1994, p. 9) found that over time, the percentage of U.S. cities offering a range of services, particularly firefighting, water service, and sewerage, has greatly declined. In California, particularly, later-forming cities are more likely to be “contract cities” and thus to take up narrower shares of total local expenditure in the county. Martin and ____________ 10An alternative measure, the county’s 1920 population as a proportion of its 1990 population, yielded very similar results, with a slightly worse overall goodness-of-fit. 64 Hogan (1989, p. 44) found of new cities in the post-Proposition 13 era that “municipal government is limited government: limited in terms of resources and limited in terms of flexibility. New city government appears also to be limited in terms of function.” This account of the history of local government indicates that areas may become “locked in” to early patterns of political development. Fisher and Wassmer (1998, p. 448) wrote, “Despite factors that favor change, a metropolitan area may be unable to alter the set pattern of earlier development [of local government structure].”11 Thus, theories of “path dependence” may provide a useful way of understanding variations in local government structure. “In the case of path-dependence theory, the potential for policies to self-reinforce and become entrenched presents the possibility of policy lock-in . . . . Early events have primary importance,” writes Woodlief (1998, p. 428). Although Woodlief sought to explain differing budgetary choices among large cities, his analysis seems to relate as well to studies of local government organization: “History matters and is cumulative. Early choices can shape later options. Time and the order of events are important contextual variables” (1998, p. 432). ____________ 11However, in their study of U.S. metropolitan areas, Fisher and Wassmer use a measure of age that is a dummy variable signifying whether the central city was incorporated after 1950. They find, contrary to the results here, that “newer” areas, using this measure, have formed more units of general-purpose government. It is unclear how this result may be reconciled with the results presented here. One possibility is that the incorporation date of a central city does not really measure the maturity of an area. It is also possible that metropolitan areas with such late formations of central city governments experienced more municipal formations in earlier periods as “substitutes” for an absent central city government. Or it may indicate that these areas were less urbanized and thus tended to form more small-town governments rather than larger urban and suburban municipalities. 65 Other Variables The racial variation variable in Table 4.2 yields an interesting tentative conclusion: Counties with more racially diverse populations tend to be more fragmented. If true, this result would be consistent with two possible stories about race and government structure. The more charitable account is that a more varied population has more heterogeneous tastes for services and thus creates a more variegated governmental structure. The less comforting interpretation is that residents seek to wall themselves off in separate jurisdictions (armed with land-use powers, in the case of cities) more often in more diverse counties, in search of greater racial and ethnic separation. However, the weakness of the racial variation variable in models— that is, its lack of statistical robustness discussed above—means that we should not make too much of these possible explanations. Simply put, we cannot reject the hypothesis that racial variation has no independent effects on political structure. Also worth noting is the persistent lack of significance of any of the remaining variables, once population, historical growth, and intergovernmental revenue are controlled for. For example, despite much suggestive support in the literature, the variables relating to socioeconomic status and development-industry strength lack explanatory power.12 ____________ 12Various interaction terms were also introduced into regression models, on the thought that two factors might act jointly to affect political fragmentation. For example, percent wealthy can be interacted with percent poverty or percent black to simulate the desire of more advantaged segments of the population to separate themselves politically given the presence of larger disadvantaged populations. Following Burns (1994), the developer index was also interacted with percent nonwhite population. None of these variables yielded interesting results. 66 In this case, the “supply side” of government revenues and service provision—the share of local government funds coming from higher levels of government—appears to play a more important role in affecting local government structure than the citizen “demand side”—the kind of population seeking services. Of course, a larger sample might allow for more complex estimations in which other relationships of interest emerge. Summary This chapter plumbed the small existing literature on the causes of local government structure to propose possible explanations for the varying patterns of fragmentation in California’s counties. The literature review suggested four general sets of possible explanations, including general contextual variables, intergovernmental issues, socioeconomic diversity, and business influence in local politics. Statistical analysis of the California data indicates a relatively simple and powerful account for differences in local political structure around the state. Areas that have larger populations and which developed earlier in the state’s history—when the bulk of the political structure was being established and cities were more likely to be full-service governments— tend to be more fragmented. In addition, areas receiving more of their revenue in the form of intergovernmental aid are less fragmented, perhaps because they have less need to set up new political structures to overcome local revenue shortfalls. There is weaker, more inconsistent evidence that greater racial variation in a county is associated with more political fragmentation. Once we control for these factors, none of the other variables, including those related to socioeconomic status and business influence, appear to matter in a significant way. 67 5. Conclusions and Policy Considerations Local political structure is important because it can have substantial effects on both the process of policymaking and residents’ quality of life. Although the existing research literature is hardly definitive, it tends to suggest several notions about the effects of local government complexity. Fragmented governmental structures (particularly competing municipalities) may promote efficiency in the delivery of some routine public services. But fragmentation also may detract from the ability of areas to manage regional land use and development, to address needs of the regional economy, or to reduce disparities between different communities. This study has not concentrated upon the effects of local government fragmentation, although Chapter 1 and Appendix A summarize the relevant literature and distill its important lessons. Rather, the goal of this report was to portray California’s structure of local government, to 69 document its degree of fragmentation, and to determine whether the local structure has changed markedly over time. These issues are important to policymakers because of the state’s repeated waves of concern over the supposed proliferation of local governments. After the passage of Proposition 13, commentators pointed with alarm to numerous city incorporations and special district formations and saw portents of a great increase in governmental fragmentation. More recently, several reform proposals aimed at rationalizing or regionalizing local government emerged in the late 1980s and into the 1990s. And 1997 saw passage of a bill creating a new study commission to consider options for changing state laws that affect local government structure. Review of Findings Data from both state and national sources reveal that California’s local governmental structure is indeed quite complex. Nevertheless, it is not abnormally fragmented, by U.S. standards, nor is it highly unstable. Looking first at counts of governmental units, we saw that cities experienced their major growth in numbers in the period before World War II, with this growth rate declining in the period since. Proposition 13 did not noticeably alter this trajectory of change, although 54 new cities have sprung up since passage of that initiative in 1978—mostly in rapidly growing Southern California. Growth in the number of special-district governments has been relatively flat since the 1970s, although the universe of special districts in the state has changed somewhat. Districts that exist to finance the construction of capital facilities have increased in number, whereas those providing more traditional public services have declined somewhat in 70 number. “Governments created by governments”—particularly county service areas and joint powers agencies—have increased, which tends to segregate funds for certain government functions from general-fund budgets. Overall, California has far fewer counties, municipalities, and special districts per 100,000 population than the national average. And in the distinctive realm of education policy, the number of school districts has decreased markedly in recent decades. Using the specially constructed political fragmentation index (PFI) in Chapters 3 and 4, we were able to see in a more nuanced fashion the ways in which responsibilities and activities are divided among local government units in each California county. Analysis of the PFI data revealed that counties differ quite markedly in their local government structure. It also confirmed the finding that political structure is only very gradually growing more complex, and does not seem to have been decisively altered by Proposition 13. Data analysis also confirmed the underlying stability of these varied patterns of local government structure throughout the state. The degree of governmental fragmentation appears to largely be a product of two “deep” underlying factors—the county’s population size and its historical pattern of population growth. Simply put, bigger and older areas tend to be more fragmented. A third main influence on local government structure, one more susceptible to policy intervention, is intergovernmental aid. Areas that have received a higher share of their funds from state and federal sources tend to acquire less governmental complexity. There is also some weak and inconsistent evidence that racial diversity is associated with greater fragmentation; otherwise measures of socioeconomic status and heterogeneity do not have 71 significant independent effects on local government structure in these data. This is true even though we might expect a more varied population to lead to more residential sorting and pressures for separate governments. In addition to informing an ongoing policy debate, the findings of this study lay groundwork for additional research. In future studies, scholars may wish to determine the effect of fragmentation on various aspects of social and governmental well being in California counties. For example, do counties that are more politically divided tend to grow more segregated by race or income? The sorting and exclusionary-zoning aspects of suburban fragmentation suggest that this might be an outcome. Are politically fragmented counties more predisposed to fiscal stress? The partitioning of the landscape into separate taxing and spending regimes suggests that some jurisdictions may be more likely to experience budgetary strain. Has political structure affected the ability of areas to integrate their transportation investments with land-use plans? Again, fragmentation may impede such coordination. Researchers may also wish to replicate national studies by seeing whether municipal competition in California has resulted in less public spending or whether special-district reliance boosts overall spending. Relative Stability in an Era of Change—Why? Some might view these findings—that local government structure in California has been relatively stable—as a “nonfinding.” Rather, they are quite remarkable results, given the enormous changes that have occurred in California’s economy and population in recent decades. It is worth asking why local governmental change has been so gradual. 72 One possible reading of this record is that state policy in this area has worked. Recall that the legislature passed a law in 1963 requiring the creation of LAFCOs in each county. LAFCOs have since become important, if not always particularly popular or visible, players in regional politics. They must review proposals for the formation of new cities and special districts and may discourage the creation of new units, particularly where a city formation might create fiscal burdens for existing units. Under rules developed by the state legislature, LAFCOs have designated “spheres of influence” for each existing city. Such an approach makes a presumption in favor of expanding current cities rather than creating new ones. In addition, with their voting board made up of representatives of existing local governments, LAFCOs might be expected to be interested in preserving the viability of existing units, by steering population growth into cities and allowing annexations.1 In a few counties, notably Sacramento, LAFCOs have bucked the statewide trend toward centering growth in cities by instead supporting county policies that have permitted much growth outside of city limits. In any event, the fiscal constraints created for local governments by Proposition 13 may have given LAFCOs all the more reason to take a long, hard look at proposed government formations. Despite the expected presumption of LAFCOs in favor of existing units, one detailed study found that at least through the 1980s, these agencies rarely thwarted incorporation moves overtly. Politicians on ____________ 1LAFCO boards generally have five members: two county supervisors, two representatives selected by the county’s cities, and a fifth member selected by the other four. Several LAFCOs, however, provide for special district representation, and a few have been authorized by the legislature to vary their membership. See Assembly Committee on Local Government (1996), pp. 41–50. 73 LAFCO boards had “much to lose if they denied the apparent will of the voters” (Musso, 1994, p. 91; see also Martin, 1984, Chap. 6). Still, Musso determined that the pattern of incorporations did not follow any logic of fiscal maximization. On the basis of case studies of incorporation proposals in Orange County, she suggested that “many residents viewed additional government—any government—as a nuisance, even if it increased services at no cost to them” (p. 198, emphasis in original). Thus, popular opinion may also be resistant to an increased number of governments, tending to reinforce the gatekeeper role of LAFCOs. Finally, in 1992, the state legislature provided new instructions on city formation to LAFCOs by passing SB 1559, a law seen as tightening up incorporation requirements (Musso, 1994, p. 55). The law, passed in response to county complaints about strategic city incorporations that drew valuable property and sales tax revenues away from counties and into the coffers of craftily drawn cities, instructs LAFCOs to approve incorporations only where they will be fiscally neutral for the affected county and other local governments. In other words, a LAFCO may approve the formation of a new city only where the revenues that the new city can be anticipated to receive are roughly equal to the new city’s expected expenditures on services.2 In the wake of this restriction, there have been only three incorporations in the period since 1992. Beyond the role of LAFCOs, another possible explanation for relative stability in California is that existing local governments may be more malleable and responsive to change than is commonly supposed. ____________ 2The new city may avoid this restriction if it negotiates in advance to provide fiscal concessions to its county or in cases where the county and other local governments are in consensus to allow the incorporation. 74 Throughout the state, cities and counties typically have been able to address new issues, empower new ethnic constituencies, and respond to economic and population growth without falling into crisis. In response to growing suburbanization, annexations have been used to gradually extend the scale of cities to the expanding fringe. Meanwhile, residents of unincorporated areas have numerous formal and informal governance mechanisms available to them to collectively address the future of their communities without incorporating a city government (see Sokolow et al., 1981). Considerations for State Policymakers Overall, then, the division of responsibilities among local governments in California seems relatively stable, and the state has considerably fewer governments per capita than other states. Local boundaries, powers, and governing relationships have not been subject to jarring changes, even as the state and its localities went through wrenching fiscal challenges in the aftermath of Proposition 13 and major recessions. Does that mean that state and local policymakers should be complacent about California’s local government system? Not necessarily. Local fragmentation in California continues to separate fiscal resources from social needs and to impede coordination on regional issues. Nor should this study’s description of a modest number of incorporations since Proposition 13 be read as a call for large-scale increases in new cities. Whereas formations of minimal, “contract” cities may not fragment service provision or financial responsibilities in a county much, they do create new units with the “police powers” to zone land and thereby affect the development of California’s regions. Indeed, land-use 75 motivations have been among the most prevalent reasons for the formation of new cities, in California and throughout the United States (Musso, 1994; Martin, 1984; Gainsborough, 1990; Burns, 1994; Lewis, 1996; and Teaford, 1997). The effects of political fragmentation upon land use have been viewed by many as resulting in uncoordinated growth patterns and a lack of attention to regional problems and needs, such as affordable housing and the transportation demands arising from development. Despite California’s home rule tradition, tinkering with the local political structure ultimately remains the state’s prerogative. The courts have held that “it is well established that the state may create, expand, diminish, or totally abolish municipal corporations with or without the consent of its citizens, or even against their protest” (Scuri v. Board of Supervisors, 1982, quoted in Detwiler, 1996b, p. 6). The state’s recent creation of a Commission on Local Governance for the 21st Century underscores its capabilities in this area. Encouraging the state to review the local system, an earlier study group created by the legislature, the California Constitution Revision Commission (1996), proposed the formation of “citizens’ charter commissions” in each county (or multicounty area). These commissions would be required to study “local government structure, methods of service delivery, and assignment of responsibilities and powers.” After a study period, an adjusted plan of service delivery and financing for the area would be proposed, and put to the voters as a new “home rule community charter.” This plan could include “methods for reducing the number and cost of local government,” a capital improvement program for all the government entities in the area, and provisions for reorganization and boundary changes, if necessary. The state would 76 assist by creating a commission to act as a clearinghouse for ideas on local reforms and to monitor local efforts. Upon approval of the new home rule community charter, all participating local entities would be granted home rule powers and protected from the reallocation of their property tax and other general-purpose local taxes. Although the commission’s idea for countywide review and reform is no guarantee of improved political structure, the plan merits serious consideration. It would compel local governments to reconsider regional arrangements and the area’s future needs, and raise a dialogue in each area concerning local governance. Moreover, its home rule guarantees would provide incentives for cities, counties, and special districts to “buy in” to the process. The commission’s proposal does, however, appear to proceed from a presumption that fewer governments would be better, and cheaper, than the current arrangement. Most of the limited evidence on government reorganization, however, offers little hope that costs will be cut or taxes lowered due to any restructuring. Reorganizations should be premised on the search for a policymaking system more likely to address problems that currently remain unresolved, rather than as part of a search for a Utopia of structural simplification. Where the problems of local governance are primarily fiscal ones, organizational reform is not necessarily the solution. Some would prefer more far-reaching revisions of local government structure. However, fundamental change in this realm—such as through wholesale consolidations of cities or special districts—is unlikely and politically difficult at best. As one national study of local government pointed out, “Talk of the need for sweeping and fundamental reform of state and local governments is common, particularly among those who periodically ‘rediscover’ the states and localities. More experienced and 77 influential reformers are aware of the remarkable resiliency of the basic features of the system in the face of a century of sweeping social, economic, and political change, and they set their sights accordingly” (Danielson, Hershey, and Bayne, 1977, p. 60). This observation accords well with the facts of the California case, where, in spite of major transformations in the state, the local government structure has changed only gradually over a period of decades. To be sure, serious study of the problems of this governance structure is always warranted and reform proposals are worth entertaining. Nevertheless, policymakers who identify problems will be well served working at the margins. They can attempt to alter the incentives and fiscal burdens and resources of California’s localities; more generous intergovernmental aid, as we have seen, appears to reduce the impulse toward fragmentation. State policymakers also can work to strengthen regional political institutions that might repair some of the seeming deficiencies of fragmentation. In short, California’s leaders can work within existing structures to strengthen local governance and public finance rather than proposing to fundamentally change a stable and resilient system. 78 Appendix A Research on the Effects of Local Government Fragmentation In recent decades, scholars of political science, economics, policy analysis, and public administration have given considerable attention to the issue of local government fragmentation—for good reason. Political boundaries separate the landscape into varied taxing and servicing regimes, with implications for both equality among residents and jurisdictions and the efficiency of government operations. Normative essays and empirical research have isolated both potential advantages and disadvantages of local political complexity, with conclusions deriving in part from the analysts’ attachment to different values. For example, the preeminent value that many economists attach to allocative efficiency contrasts with that of political scientists, who are often more concerned with equity and effective political representation. Some scholars, particularly those associated with the public choice school in economics and political science, have emphasized the efficiency 79 gains thought to result from governmental multiplicity. Drawing upon the theory initially set forth by economist Charles Tiebout (1956), advocates of governmentally fragmented areas argue that citizens and firms in such regions can “vote with their feet” to find the jurisdiction that best meets their tastes for service and tax levels. In such a model, competition among jurisdictions helps keep governments leaner and less susceptible to “budget-maximizing bureaucrats.” Overlapping units can each supply specific services to local areas at the most optimal scale economy. Small-scale governments are seen also as less insulated, closer to home, and thus easier for citizens to control. Other analysts paint a less rosy view of political fragmentation. They make the case that areas that are more politically divided often have a more difficult time addressing the “general welfare”—in particular, regional land-use issues such as urban sprawl, housing opportunity, and downtown preservation. Regional strategies for economic development may be sacrificed to beggar-thy-neighbor growth schemes. Politically fragmented areas may also be more segregated by race and income. Moreover, fragmentation is alleged to reduce local government’s ability to aid the needy, since businesses and wealthier residents will find it relatively easy to flee to a nearby jurisdiction that is less generous with their tax money. And public-sector accountability may be hindered because the complexity and multiplicity of government can make it difficult for citizens to identify and hold responsible the appropriate officials. The actual results of empirical studies of the effects of fragmentation are mixed. On the issue of public spending, some studies have found lower per capita costs for certain services in more fragmented areas, whereas many others have come to opposite conclusions; still others find 80 no statistically significant effects. The differing results are traceable in part to the different services examined, the different control variables and multivariate methods used, and differing measures of political fragmentation. (See, for example, Baird and Landon, 1972; Isserman, 1976; Schneider, 1989; Zax, 1989; Forbes and Zampelli, 1989; and Oates, 1989.) There is a reasonable degree of evidence that a greater number of competing municipal governments in an area is associated with less overall public spending by municipalities (Zax, 1989). The one study looking at city incorporations in California in relation to the efficiency issue finds an increase in administrative expenditures after incorporation but potential efficiency gains due to the greater homogeneity of municipal populations (Musso, 1994). By contrast, a profusion of special district governments in an area has been associated with higher local government revenue burdens (Zax, 1989). Regarding spending, the most important recent study of special districts finds relatively convincing evidence that reliance on specialpurpose governments for service provision increases the absolute amounts spent on district-provided services. Such reliance also increases the share of total local expenditure going to district-provided services (Foster, 1997).1 On the other hand, reducing the number of districts that offer a particular service in a region may decrease competition and therefore lower efficiency in that sector. Several instances of consolidation of special districts in California since Proposition 13 were examined by ____________ 1In preliminary statistical analyses of California data, not discussed in this report, I also found evidence supporting the proposition that counties relying heavily on special districts tend to spend more per capita, controlling for income, urbanization, population size, and other relevant factors. 81 Morgan and Chapman (1994), who found that overall revenues and expenditures generally increased after mergers. A complicating factor in all studies of local government taxing or spending is that differing “tastes” for service levels in different regions may affect expenditures per capita. In other words, more spending does not necessarily imply a less “efficient” public sector. More spending may instead mean that governments are responding to local desires for enhanced services. Fiscal coordination in an area may suffer due to a profusion of local governments. Baldassare (1998) found that Orange County’s recent financial crisis was aggravated, and its subsequent recovery greatly complicated, by the array of local governments that had put funds into the county investment pool. These entities guarded their local advantages jealously and were unused to communicating with each other. On the issue of city/suburb equity and disparities, disputes over research methods have rendered the role of political fragmentation unclear. Morgan and Mareschal (1996), in preliminary results, found that fragmentation has few measurable effects on city/suburban inequality measures and may in fact be associated with greater fiscal health of central cities. Earlier, an oft-cited book by Rusk (1993) found fiscally and socially healthier conditions in “elastic” central cities—those able and willing to annex their suburbs regularly, whereas Savitch et al. (1993) found positive connections between central-city well-being and suburban well-being. The methods and theory driving these latter two studies were relatively rudimentary, however, and have been subject to serious criticism and reinterpretation by Blair and Zhang (1994), Blair et al. (1996), and Hill et al. (1995). Looking at who pays for and benefits from government services, Hawkins and Hendrick (1997) found that 82 services provided by counties and by regional special districts—in both cases, governments that included both well-off suburbs and a poorer central city—tended to disproportionately benefit city residents and reduce disparities. One might read this as a defense of fragmentation (since the study shows how overlapping governments can promote equity) but note that the county and regional special districts represent a particular type of area-wide governmental arrangement. Relating equity to land use, Weiher (1991), relying on both case studies and multivariate analysis, concluded that fragmentation exacerbates racial and socioeconomic segregation in metropolitan areas. Migrating households, in Weiher’s model, are able to use boundaries and jurisdictions as information signals in evaluating communities’ residential status, school district quality, and racial makeup. In another study relating to land use, Lewis (1996) found metropolitan political fragmentation associated with weaker dominance of central city downtowns and more of a patchwork pattern of development, with evidence that jobs and housing may be more geographically separated in fragmented areas. Weaker evidence existed for the hypothesis that fragmentation led to greater degrees of sprawl, or residential decentralization, when other relevant factors were held constant. Local political structure has even been studied for its relationship to the quality of political journalism. In a perceptive study, Kaniss (1991) blamed suburban political fragmentation for reducing the capacity of the media to provide meaningful news coverage of the public affairs of a region. Public issues are segmented among so many low-visibility suburban governments that regional newspapers and broadcasters find it difficult and unrewarding to cover regional governmental trends and public issues—especially development issues—outside the central city. 83 This brief review only sketches the major outlines of research on the effects of political fragmentation. Although this literature hardly speaks with one voice, it does illustrate the potentially key importance of local government organization. In Chapter 1, I highlight some of the major policy implications of fragmentation for four broad roles of the local public sector. 84 Appendix B Issues Regarding Special District Data in the Census of Governments The Census of Governments, which is used as the major data source in this report, includes a substantially different enumeration of special district governments in California than does the State Controller. The Census Bureau counts only politically freestanding special districts, not those dependent upon or supervised by a “parent” entity. In short, the bureau recognizes only those districts with “substantial administrative and fiscal independence” (Martinez-Vazquez et al., 1997, p. 289; and Foster, 1997, pp. 10–11). In the bureau’s words, “Where these districts have separate governing bodies, they are counted as governments; when they are governed by county or municipal officials in an ex officio capacity, they are not counted as separate governments” (U.S. Bureau of the Census, 1994, p. A-23). For the purposes of measuring political structure, the Census Bureau’s procedure is quite appropriate. The purpose of this report is 85 not primarily to count the number of units that exist on paper but to measure the dispersion of governmental authority and activity. Comparison to Controller’s Count Comparing the Controller’s data for fiscal year 1991–1992 with the 1992 Census of Governments highlights the differences involved. The Controller counted 4,857 districts as “officially in existence” in that year, whereas the Census Bureau listed 2,797. Of the State Controller’s enumeration, 1,604 districts were governed by a county board of supervisors, and 150 reported to city councils. These are appropriately considered “dependent” and thus would not be counted under the Census Bureau’s rules. Instead, their finances are included as part of the parent government. The types of activities performed by districts that were disproportionately dependent include the following: • County service areas • County waterworks • Highway lighting (all dependent) • Maintenance • Permanent road division • Sewer/sewer maintenance • Storm water The above types of districts are those, then, that will tend to be “undercounted,” in a strictly paper sense, by the Census Bureau. In contrast, the following types of districts are disproportionately “independent,” according to the classification scheme of the State 86 Controller—meaning that most are not directly governed by an individual city council or county board of supervisors: • Cemetery • Community services • Fire protection • Hospital • Irrigation • Joint exercise of powers • Levee • Library • Memorial • Mosquito abatement • Nonprofit corporations • Reclamation • Recreation/park • Sanitary • Public utility • Water However, some of these so-called “independent” districts are in fact joint creations of multiple general-purpose governments and typically are overseen by appointees of each government. In other cases, generalpurpose governments have budgetary review or other discretion over an “independent” district, or appoint its officers. Thus, in cases where a district is largely accountable to, under the oversight of, or completely financed by, an individual general-purpose government, the Census 87 Bureau does not include the district as a separate government. This proviso pertains to 245 nonprofit corporations—which largely construct public facilities such as libraries and city halls, or serve as purchasing entities for equipment. The Census Bureau also removes 11 flood control districts and 10 other districts, from various categories, from its list of autonomous districts. The bureau does count joint-powers agencies as separate governments, where the district’s board is distinct from other individual local governments. There are no clear reasons for the remaining disparities between the Census count and the Controller’s count. However, the disparity is relatively trivial, as Table B.1 shows. Table B.1 Reconciling Controller and Census Bureau Counts of Special Districts, 1992 Total special districts in existence, per Controller Minus: Districts excluded because they are “dependent” on a city or county Districts excluded because they are otherwise accountable to another local government Equals: Ostensible count of autonomous special districts Census of Governments count Difference: 4,836 1,752 266 2,818 2,797 21 (0.7%) Response to Criticism of Census Data on Districts Leigland (1990) has critiqued the Census Bureau’s data on special districts for three main reasons. However, none of these detracts seriously from the usefulness of the Census of Governments data for the current report. First, Leigland argues that the Census’s data on debt is inadequate for special districts. However, the fragmentation index used in this 88 report is based on expenditures, not debt. Second, Leigland claims that the bureau undercounts special districts—particularly those involved in financing public works—by not including dependent units, noting that the creation of districts has been an important creative financing scheme for cities and counties. However, for the reasons discussed above, dependent districts are not politically autonomous units; they are typically little more than accounting devices of general-purpose governments and are counted appropriately by the bureau as part of the finances of the parent units. Finally, Leigland argues that the bureau overcounts other districts by including inactive districts, with zero expenditures, in the Census of Governments listings. The political fragmentation index overcomes this problem, since governments with zero expenditures receive no weight in the index. Again, Leigland’s critique is more damning for those obsessed with counting governments, but it is less important for those interested in the dispersion of political authority. Undoubtedly, some will consider the Census Bureau’s data less than ideal for analysis of special districts in California. However, the bureau constructs its information on special districts in California using, in large part, information obtained from the State Controller’s records—data that have been found to be comprehensive and accurate, at least as regards revenue (Shires and Glenn Haber, 1997). Moreover, the Census Bureau attempts to clarify inconsistencies and questions through consultations with relevant state and local government personnel (Sacks, 1990; and U.S. Bureau of the Census, 1994). In any event, the State Controller’s data were not available for analysis in machine-readable format over the time period of interest. 89 Appendix C New Cities Incorporated After Proposition 13 Table C.1 California Cities Incorporated Since July 1, 1978 (listed by county) County Alameda City Dublin Year of Incorporation 1982 Butte Paradise 1979 Contra Costa Danville San Ramon Orinda 1982 1983 1985 Kings Avenal 1979 Lake Clearlake 1980 Los Angeles La Habra Heights Westlake Village 1978 1981 91 Table C.1 (continued) County City Agoura Hills West Hollywood Santa Clarita Diamond Bar Malibu Calabasas Year of Incorporation 1982 1984 1987 1989 1991 1991 Mono Mammoth Lakes 1984 Napa American Canyon 1992 Nevada Truckee 1993 Orange Mission Viejo Dana Point Laguna Niguel Lake Forest Laguna Hills 1988 1989 1989 1991 1991 Placer Loomis 1984 Riverside Cathedral City La Quinta Moreno Valley Temecula Calimesa Canyon Lake Murrieta 1981 1982 1984 1989 1990 1990 1991 Sacramento Citrus Heights 1997 San Bernardino Grand Terrace Big Bear Lake Twentynine Palms Highland Hesperia Apple Valley Yucaipa Yucca Valley Chino Hills 1978 1980 1987 1987 1988 1988 1989 1991 1991 92 Table C.1 (continued) County San Diego City Poway Santee Solana Beach Encinitas Year of Incorporation 1980 1980 1986 1986 San Joaquin Lathrop 1989 San Luis Obispo Atascadero 1979 San Mateo East Palo Alto 1983 Santa Barbara Solvang Buellton 1985 1992 Shasta Shasta Lake 1993 Sonoma Windsor 1992 Ventura Moorpark 1983 Yolo West Sacramento 1987 93 Bibliography Assembly Committee on Local Government, General Assembly, State of California, Guide to Cortese-Knox Local Government Reorganization Act of 1985, revised ed., Sacramento, California, 1996. Assembly Office of Research, General Assembly, State of California, California 2000: Getting Ahead of the Growth Curve, Sacramento, California, December 1989. Baird, Robert N., and John M. Landon, “Political Fragmentation, Income Distribution, and the Demand for Government Services,” Nebraska Journal of Economics and Business, Vol. 11, No. 4, 1972, pp. 171–184. Baldassare, Mark, When Government Fails: The Orange County Bankruptcy, University of California Press and Public Policy Institute of California, Berkeley, California, 1998. Blair, John P., Samuel R. Staley, and Zhongcai Zhang, “The Central City Elasticity Hypothesis: A Critical Appraisal of Rusk’s Theory of Urban Development,” Journal of the American Planning Association, Vol. 62, No. 3, 1996, pp. 345–353. Blair, John P., and Zhongcai Zhang, “Ties that Bind’ Reexamined,” Economic Development Quarterly, Vol. 8, No. 4, 1994, pp. 373–377. Bollens, Scott A., “Examining the Link between State Policy and the Creation of Local Special Districts,” State and Local Government Review, Fall 1986, pp. 117–124. 95 Brooks, Lea, “The Revenue Chase and Urban Sprawl,” Golden State Report, June 1988, p. 37. Burns, Nancy, The Formation of American Local Governments: Private Values in Public Institutions, Oxford University Press, New York, 1994. California Constitution Revision Commission, Final Report and Recommendations to the Governor and the Legislature, Sacramento, California, 1996. California State Controller, Financial Transactions Concerning Cities of California, Sacramento, California, annual. California State Controller, Financial Transactions Concerning Special Districts of California, Sacramento, California, annual. Danielson, Michael N., Alan M. Hershey, and John M. Bayne, One Nation, So Many Governments, Lexington Books, Lexington, Massachusetts, 1977. Detwiler, Peter M., “Creatures of Statute . . . Children of Trade: The Legal Origins of California Cities,” in California Constitution Revision Commission, History and Perspective, Sacramento, 1996a, pp. 97–104. Detwiler, Peter M., “Suburban Secession: History and Context,” Working Paper 96–12, Graduate Program in Public Policy and Administration, California State University at Sacramento, 1996b. Feinbaum, Robert, “Climate Right for Creating New Cities,” California Journal, October, 1987, pp. 497–499. Fisher, Ronald C., and Robert W. Wassmer, “Economic Influences on the Structure of Local Government in U.S. Metropolitan Areas,” Journal of Urban Economics, Vol. 43, 1988, pp. 444–471. Forbes, Kevin F., and Ernest M. Zampelli, “Is Leviathan a Mythical Beast?” American Economic Review, Vol. 79, No. 3, 1989, pp. 568– 577. Foster, Kathryn A.,“ Exploring the Links between Political Structure and Metropolitan Growth,” Political Geography, Vol. 12, No. 6, 1993, pp. 523–547. Foster, Kathryn A., The Political Economy of Special-Purpose Government. Georgetown University Press, Washington, D.C., 1997. 96 Fulton, William, “Two States Find a Way to Keep the Peace,” Planning, October 1989, pp. 16–17. Fulton, William, Guide to California Planning, Solano Press Books, Point Arena, California, 1991. Fulton, William, “City Secession Bill Moves Forward,” California Planning & Development Report, Vol. 12, No. 9, 1997a, p. 2. Fulton, William, “State to Examine Fiscal Relationship with Locals,” California Planning & Development Report, Vol. 12, No. 10, 1997b, pp. 1, 10. Fulton, William, “Moves Begin to Reform Cityhood Law,” California Planning & Development Report, Vol. 13, No. 4, 1998, pp. 1, 9–10. Gainsborough, Juliette, “Incorporation: Government of the Future? The City of Santa Clarita, A Case Study,” senior thesis, University of California at Berkeley, 1990. Hawkins, Brett W., and Rebecca M. Hendrick, “Do Metropolitan Special Districts Reinforce Sociospatial Inequalities?” Publius: The Journal of Federalism, Vol. 27, No. 1, 1997, pp. 135–143. Hill, Edward W., Harold L. Wolman, and Coit Cook Ford III, “Can Suburbs Survive without Their Central Cities? Examining the Suburban Dependence Hypothesis,” Urban Affairs Review, Vol. 31, No. 2, 1995, pp. 147–174. Isserman, Andrew M., “Interjurisdictional Spillovers, Political Fragmentation and the Level of Local Public Services: A Reexamination,” Urban Studies, Vol. 13, 1976, pp. 1–12. Kaniss, Phyllis, Making Local News, University of Chicago Press, Chicago, 1991. League of California Cities, Conflicts at the City’s Edge: Annexations, Development and Tax Negotiations, League of California Cities, Sacramento, April 1996. Leigland, James, “The Census Bureau’s Role in Research on Special Districts: A Critique,” Western Political Quarterly, Vol. 43, No. 2, 1990a, pp. 367–380. Leigland, James, “In Defense of a Preoccupation with Numbers: A Response,” Western Political Quarterly, Vol. 43, No. 2, 1990b, pp. 385–386. 97 Leigland, James, “Public Authorities and the Determinants of Their Use by State and Local Governments,” Journal of Public Administration Research and Theory, Vol. 4, No. 4, 1994, pp. 521–544. Lewis, Paul G., Shaping Suburbia: How Political Institutions Organize Urban Development, University of Pittsburgh Press, Pittsburgh, Pennsylvania, 1996. MacManus, Susan, “Special District Governments: A Note on Their Use as Property Tax Relief Mechanisms in the 1970s,” Journal of Politics, Vol. 43, 1981, pp. 1207–1214. Marando, Vincent, and Mavis Mann Reeves, “State Responsiveness and Local Government Reorganization,” Social Science Quarterly, Vol. 69, No. 4, 1988, pp. 996–1004. Martin, Linda M., The Politics and Economics of Local Self Government: New Cities in California after Proposition 13, Ph.D. dissertation, University of California at Davis, 1984. Martin, Linda M., and Joan Q. Hogan, “Small Cities, Big Dreams,” Western City, May 1989, pp. 20–23, 44. Martinez-Vazquez, Jorge, Mark Rider, and Mary Beth Walker, “Race and the Structure of School Districts in the United States,” Journal of Urban Economics, Vol. 41, 1997, pp. 281–300. Miller, Gary J., Cities by Contract: The Politics of Municipal Incorporation, MIT Press, Cambridge, Massachusetts, 1981. Miller, Joel, “Annexations and Boundary Changes in the 1980s and 1990–1991,” in The Municipal Year Book 1993, International City/County Management Association, Washington, 1993, pp. 100– 109. Morgan, David R., and Patrice Mareschal, “Central City/Suburban Inequality and Metropolitan Political Fragmentation,” presented at the Annual Meeting of the American Political Science Association, San Francisco, California, 1996. Morgan, Stephen P., and Jeffrey I. Chapman, Special District Consolidations: A Research Study for the Association of California Water Agencies, Los Angeles, California, December 1, 1994. Musso, Juliet Ann, Drawing the Line: The Political Economy of Municipal Incorporation, Ph.D. dissertation, University of California at Berkeley, 1994. 98 Neiman, Max, “An Exploration into Class Clustering and LocalGovernment Inequality,” in Richard C. Rich (ed.), Analyzing UrbanService Distributions, Lexington Books, Lexington, Massachusetts, 1982, chapter 13. Nelson, Michael A., “Decentralization of the Subnational Public Sector: An Empirical Analysis of the Determinants of Local Government Structure in Metropolitan Areas in the U.S.,” Southern Economic Journal, Vol. 57, No. 2, 1990, pp. 443–457. Nice, David C., “An Intergovernmental Perspective on Urban Fragmentation,” Social Science Quarterly, Vol. 64, No. 1, 1983, pp. 111–118. Oates, Wallace E., “Searching for Leviathan: A Reply and Some Further Reflections,” American Economic Review, Vol. 79, No. 3, 1989, pp. 578–583. Ostrom, Elinor, “The Social Stratification—Government Inequality Thesis Explored,” Urban Affairs Quarterly, Vol. 19, No. 1, 1983, pp. 91–112. Perrenod, Virginia, Special Districts, Special Purposes, Texas A&M University Press, College Station, Texas, 1984. Peterson, Paul E., City Limits, University of Chicago Press, Chicago, Illinois, 1981. Rusk, David, Cities Without Suburbs, Johns Hopkins University Press, Baltimore, Maryland, 1993. Sacks, Seymour, “The Census Bureau’s Role in Research on Special Districts: A Critique’: A Necessary Rejoinder,” Western Political Quarterly, Vol. 43, No. 2, 1990, pp. 381–383. Savitch, H. V., David Collins, Daniel Sanders, and John P. Markham, “Ties That Bind: Central Cities, Suburbs, and the New Metropolitan Region,” Economic Development Quarterly, Vol. 7, No. 4, 1993, pp. 341–357. Schneider, Mark, The Competitive City: The Political Economy of Suburbia, University of Pittsburgh Press, Pittsburgh, Pennsylvania, 1989. Schrag, Peter, Paradise Lost: California’s Experience, America’s Future, New Press, New York, 1998. 99 Shires, Michael A., and Melissa Glenn Haber, A Review of Local Government Revenue Data in California, Public Policy Institute of California, San Francisco, 1997. Sokolow, Alvin D., et al., Choices for the Unincorporated Community: A Guide to Local Government Alternatives in California, 2nd. ed., Institute of Governmental Affairs, University of California at Davis, September 1981. Stein, Robert M., “Tiebout’s Sorting Hypothesis,” Urban Affairs Quarterly, Vol. 23, No. 1, 1987, pp. 140–160. Teaford, Jon C., Post-Suburbia: Government and Politics in the Edge Cities, Johns Hopkins University Press, Baltimore, Maryland, 1997. Tiebout, Charles, “A Pure Theory of Local Expenditures,” Journal of Political Economy, Vol. 64, 1956, pp. 416–424. U.S. Bureau of the Census, County Business Patterns 1977—California, Department of Commerce, Washington, D.C., 1979. U.S. Bureau of the Census, 1992 Census of Governments: Volume 1, Government Organization, Washington, D.C., March 1994. U.S. Bureau of the Census, Census of Governments: Finance Summary Statistics [machine-readable data file], Bureau of the Census, Washington, D.C., 1972, 1977, 1982, 1987, and 1992. Wassmer, Robert W., and Ronald C. Fisher, “Time, Tiebout, and Transition in the Structure of Local Government in U.S. Metropolitan Areas,” Working Paper 97-07, Graduate Program in Public Policy and Administration, California State University at Sacramento, 1997. Weiher, Gregory R., The Fractured Metropolis: Political Fragmentation and Metropolitan Segregation, State University of New York Press, Albany, 1991. Woodlief, Anthony, “The Path-Dependent City,” Urban Affairs Review, Vol. 33, No. 3, 1998, pp. 405–437. Zax, Jeffrey, “Is There a Leviathan in Your Neighborhood?” American Economic Review, Vol. 79, No. 3, 1989, pp. 560–567. 100 About the Author PAUL G. LEWIS Paul Lewis is a research fellow at the Public Policy Institute of California. He is a political scientist with research interests in urban and metropolitan problems including transportation, land use, housing, and the structure of local government. His recent publications include Federal Transportation Policy and the Role of Metropolitan Planning Organizations in California (Pubic Policy Institute of California, 1997) and Shaping Suburbia: How Political Institutions Organize Urban Development (University of Pittsburgh Press, 1996). Before joining PPIC, he was an assistant professor of political science at Florida International University. He holds an A.B. in political science from Indiana University and an M.A. and Ph.D. in politics from Princeton. 101 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., 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. Shires, Michael A., John Ellwood, and Mary Sprague (1998), Has Proposition 13 Delivered? The Changing Tax Burden 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. 102" ["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_998plr" ["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_998PLR.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) }