Donate
Independent, objective, nonpartisan research

R 102PLR

Authors

R 102PLR

Tagged with:

Publication PDFs

Database

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

View live version

object(Timber\Post)#3742 (44) { ["ImageClass"]=> string(12) "Timber\Image" ["PostClass"]=> string(11) "Timber\Post" ["TermClass"]=> string(11) "Timber\Term" ["object_type"]=> string(4) "post" ["custom"]=> array(5) { ["_wp_attached_file"]=> string(12) "R_102PLR.pdf" ["wpmf_size"]=> string(7) "1009450" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(302447) "Cities Under Pressure: Local Growth Controls and Residential Development Policy ••• Paul G. Lewis Max Neiman 2002 PUBLIC POLICY INSTITUTE OF CALIFORNIA Library of Congress Cataloging-in-Publication Data Lewis, Paul George, 1966– Cities under pressure : local growth controls and residential development policy / Paul G. Lewis, Max Neiman p. cm. Includes bibliographical references (p. ). ISBN 1-58213-040-X 1. Housing policy—California. 2. Housing development— Government policy—California. 3. Local government— California. I. Neiman, Max. II. Title. HD7303.C2 L49 2002 333.77'15'09794—dc21 2001048636 Copyright © 2002 by Public Policy Institute of California All rights reserved San Francisco, CA Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source and the above copyright notice is included. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. Foreword One of the most frustrating and contentious issues in California in recent years has been the shortage of housing. Even as 10 million people have been added to the state’s population over the last two decades, housing production has fallen short of demand, particularly during the recent period of economic growth. This shortfall has resulted in wildly escalated housing prices in many urban centers and heated markets in urban ring communities as far as 100 miles from employment centers. Many carefully crafted arguments have sought to explain this mismatch between supply and demand. They range from the inability of any state to absorb such huge numbers of new residents to the view that planners, conservationists, and community activists have made it devilishly difficult to build any new housing. Some researchers have also argued that state tax policies make residential growth economically unattractive for revenue-starved suburban communities. Yet another line of argument has focused on local growth controls, which are frequently placed near the top of the list of reasons for the shortage of affordable housing in California. According to this argument, a lack of enthusiasm for housing in California cities—and in some cases, outright hostility to multifamily or “affordable” developments—goes a long way toward explaining the state’s lagging housing production. Paul Lewis and Max Neiman address this argument head-on by studying which communities enact growth restrictions and why. They find that relatively few cities across the state have passed strict controls on housing development and that cities’ motives in managing residential growth do not appear to reflect simple elitism or civic selfishness. The authors conclude that growth controls may constrain some homebuilding but that broader market forces and state policies probably do more to explain California’s housing costs and slow rate of production. iii Why are these findings important? Over the last 30 years, California has seen the power of the public purse shift from local government to the state. From school finance to long-term debt for municipal facilities, the state government has played a more important role with each passing decade. As a result, local governments have sought to increase their revenue leverage with the state and to limit their exposure to the most costly types of development. In the last five years, various PPIC studies have examined the strategies and tactics cities have used in this regard. These studies have one consistent theme: Local governments are only marginally capable of influencing the total amount of growth in the state. Some shifting of growth and revenue probably occurs as a result of local policy; but for the most part, large retailers are located more or less where the companies want them, new industrial growth tends to cluster where similar activities exist, and new housing is built where the demand is greatest and large tracts of developable land still exist. Some local governments are more adept at managing growth than others, but their actions occur within a policy environment created largely at the state level. This most recent PPIC study confirms that general impression built over time from previous studies. Although the authors maintain that residential growth controls do not play a large role in the state’s housing problems, they do not downplay the problems that cities face in managing growth. Rather, they offer several precepts for managing conflicts over growth before they become permanent and unhelpful features of city politics. More important, perhaps, they urge state policymakers to consider ways to support local officials who are willing to accommodate growth but do not wish to burden their constituents or treasuries unduly. David W. Lyon President and CEO Public Policy Institute of California iv Summary California faces a number of daunting growth challenges. One of the most pressing is meeting the demand for housing for its increasingly diverse population. The state’s difficulties with housing production and affordability have received a great deal of attention since California entered a period of sustained and rapid economic and population growth in the mid-1990s. In examining the statewide challenge of residential policy, however, one is quickly confronted with the fact that many of the most relevant governmental decisions—regarding planning, zoning, permitting, or the siting of affordable housing projects, for instance—are made by local governments. Many critics of local regulation and advocates for the need to produce more housing point to local residential growth controls as a major cause of housing shortfalls. This report examines residential development policy in California municipalities (cities), where most of the state’s new growth is accommodated. It places particular emphasis on efforts by cities to control or manage the pace or form of new housing development. Which communities adopt growth controls, and why? Attention is also directed toward the local politics of residential growth, to the role of local conditions and controversies in affecting land-use policies, and to the overall posture of city governments toward accommodating new housing. The authors draw much of their evidence from a detailed mail questionnaire of city planning officials responsible for implementing local residential policies in the major regions of the state—including metropolitan Southern California, the San Francisco Bay Area, and the Central Valley—conducted in 1998–1999. Data from the survey, which drew a 76 percent response rate, are used in combination with a broad variety of community-level statistics drawn from the U.S. Census and state sources. v The Dilemma of Local Growth Policy On one level, it seems odd to suggest that California’s local governments are not doing enough to accommodate new housing and population growth. During the 1990s, growth swept all major regions of the state, and the state’s population grew by over 4 million. This increment alone is larger than the populations of 25 states. Nearly 95 percent of this growth was absorbed by municipalities. Cities have been able to accommodate the state’s increasing demographic and economic diversity, generally without becoming paralyzed by social or political crises at the local level. From another standpoint, some argue that local governments have not done enough to accommodate the pressures for new housing development in the state. California constitutes about 12 percent of the national population, but its share of the nation’s housing production fell from 14.7 percent in 1990 to 8.3 percent in 1999. Furthermore, in comparison with other parts of the country, the state’s housing growth has not kept pace with its job creation. Multifamily housing, in particular, declined precipitously after the late 1980s. Median home prices and rents escalated in the recent economic expansion, and affordability rates fell. According to one view, then, cities’ lack of enthusiasm for housing— and in some cases, outright hostility to multifamily or “affordable” developments—go a long way toward explaining California’s lagging housing production. Critics see local residential growth controls and “ballot-box planning” as ubiquitous and accuse local governments of having both the goal and effect of stifling residential development. Actual research on this point has been less persuasive, however, as investigators have come to mixed results about whether growth controls have been effective in restricting development. Furthermore, few researchers have examined motivations for local growth management, outside of limited and unrepresentative case studies. It may be more reasonable to regard city government as an arena for the attempted balancing of pro- and antigrowth interests active in local politics, with outcomes that vary widely among cities depending on various local conditions and the capacity of the locality to accommodate more growth. vi Local Politics and the Adoption of GrowthManagement Policies The 1998–1999 mail survey of 297 city planning officials indicates that there is considerable variety among California cities regarding residential policies. Respondents were asked about 16 types of local policies that seek to influence either the pace, shape, or form of new housing. As Table S.1 indicates, adoption of most of these policies has been the exception rather than the rule. Furthermore, 76 percent of cities employed three or fewer policies, of the 16 policy tools examined. Cities were most likely to have adopted policies to “shape” residential development—such as design review requirements or affordable housing set-asides—that some see as relatively innocuous from the standpoint of housing production. Policies designed to overtly restrict the amount of housing developed in a city, or to condition housing construction on infrastructure capacity, were less common. There are important regional differences, however, with San Francisco Bay Area municipalities significantly more likely to adopt Table S.1 Adoption of Residential Growth-Management Policies by Cities in California’s Three Major Regions Policy Design review standards Projects must include affordable housing component City has experienced a moratorium Encourage growth in built-up areas only Satisfy traffic standards before allowing development Use capital improvements to control rate or location of growth Official population ceiling Annual limit on housing units constructed Annual limit on building permits Restrict growth to built-up areas only Formula for allowable annual growth Ranking of proposed residential projects Annual limit on multifamily dwellings Annual limit on water connections Popular vote required for sewer capacity increase Recent reduction in residential zoning Source: Lewis and Neiman (2000). % of Cities with Policy 83 31 30 28 27 14 13 9 6 6 5 5 4 4 3 2 vii growth-management policies. The typical Bay Area city adopted 3.5 policies out of the 16, compared to 2.5 among Central Valley cities, 2.1 in Los Angeles County cities, and 2.4 among other Southern California cities. These regional differences do not seem to originate from greater opposition to growth among city officials in the Bay Area. In fact, more than two-thirds of the planners responding to the survey from each region characterized their city council majority as either encouraging residential growth or taking a neutral attitude toward it. And even in the Bay Area, a large majority of respondents indicate that their city’s review process for residential development has either stayed the same or gotten shorter in the last five years. Rather, grassroots-level citizen opposition to growth appears to be much higher in the Bay Area. Half of the respondents from that region report that residential growth issues are often or almost always controversial in their city, compared to about onequarter of respondents in Los Angeles County and the rest of Southern California, and one-eighth in the Central Valley. Bay Area respondents are also much more likely to report that growth issues have been influential in affecting the outcomes of mayoral or council elections. Local citizen unease is important because it is strongly linked to the number of growth-control policies in the city. This relationship is also visible in regard to local voter initiatives. Although only 16 percent of respondents indicate that voter initiatives have been a major source of policies to slow residential development in the city, the level of growthregulating activity is considerably higher in jurisdictions that have had, or expect to have, growth-related initiatives (Figure S.1). In short, local officials appear to formulate residential policies, in part, in reaction to or anticipation of residents’ opposition to new development. Growth controls appear to emerge in large part because of the dynamics of local politics rather than dispassionate analysis of local or regional housing trends or environmental conditions. Nevertheless, pitched battles over residential development seem to be limited to a relatively small proportion of California cities, whereas growth is a more routine process for most communities. viii Average number of policies 5 Yes 4 No 3 2 1 0 Had citizen growth initiative? Expect citizen growth initiative? Initiative status Figure S.1—Citizen Initiatives and City Growth-Management Policies The fact that cities in the region that grew at the slowest rate—the Bay Area—tend to adopt the highest number of growth-management policies indicates that the sheer rate of growth itself is not necessarily the best determinant of local residential policy. A series of statistical analyses, taking account of various city characteristics, show a more complex relationship between local conditions and residential policies. Taken as a whole, the results indicate that local growth controls and residential policies are generally motivated by something more than simply the self-interest of existing homeowners. Rather, day-to-day local conditions in a given city, such as long commute times or an excess of housing to local jobs, seem to motivate citizen opposition to residential growth. In relatively established or “mature” local communities without other overriding issues (such as unemployment or crime), such citizen opposition to growth contributes to the passage of policies that attempt to increase public oversight and control over the rate and character of housing development. ix The Broader Context of Local Residential Policies There seems little doubt that housing supply has not kept pace with the vast levels of population increase in California. However, how much of the supply problem can be attributed to the actions of localities is an exceptionally difficult matter to resolve. Although local growthmanagement policies draw media attention and are often singled out for criticism, most cities have few such policies—and particularly, few overt limitations on housing development. Indeed, many localities welcome residential growth, and many community officials resist serious controls on residential development. Consequently, the main effects of local residential development policies might be to redistribute the location of housing development among communities within a housing market. Overall, the macro-level effects of growth-management policies on the state's housing might be less important than is believed, and it thus may be premature to impose via state action constraints on how localities may regulate and manage residential development. The authors suggest that the debate should be refocused away from local growth controls, toward factors that are perhaps less obvious in California’s development process. Other factors likely play a larger role than local growth-management policies in contributing to California’s housing policy challenges. Some factors appear to limit housing supply, including the initial low-density zoning of some communities, restrictive building codes, changes in the federal tax treatment of commercial real estate, and litigation and liability insurance problems facing condominium builders. Housing advocates would do well to focus as much attention on these “invisible policies” as on local growth-control policies. At the same time, demand for housing in many areas has been particularly intense in recent years, because of rapid population growth, the continuing desirability of the coastal areas of the state, and the wealth gains—particularly capital gains—that have occurred for many California households. Furthermore, statewide institutions and policies merit attention. Under California’s system of public finance, cities may find that accepting housing works against their fiscal self-interest. The California Environmental Quality Act and the initiative process are additional statewide factors that can provide procedural tools for housing opponents x and introduce uncertainty both for builders and local policymakers. Meanwhile, the state has fallen behind in providing and maintaining its infrastructure, including transportation facilities, which means that growth is likely to create additional strains and controversies in many communities. Looking at the factors arrayed against housing production, local growth management appears to be a relatively small part of the picture, although further research is necessary to clarify its role. What does this mean for state policymakers? The finding that long commute times make communities less amenable to housing construction provides a strong hint that California must improve its infrastructure deficiencies if it is to convince its residents and local officials that new housing will not reduce existing residents’ quality of life. The state’s system for financing local governments is similarly a central fact of life for city policymakers, as it defines the costs and rewards of various types of growth and development. Most important, to avoid potentially paralyzing controversies over growth—and the punitive antigrowth ballot initiatives that sometimes result from such controversies—cities and counties need to find ways to manage community conflict before it erupts. Local governments must convince their residents that they do have sound plans to accommodate future growth sensitively. Meanwhile, state policymakers might consider policies directed at increasing the amount of information available to those in the residential development process and addressing some of the unpredictability of that process. xi Contents Foreword ..................................... Summary..................................... Figures ...................................... Tables ....................................... Acknowledgments ............................... iii v xvii xix xxi 1. EXAMINING CITY GROWTH POLICIES........... Why Study Local Growth Policies?.................. The Focus: Cities and Growth-Management Policy....... What Evidence Do We Use?...................... Summing Up ............................... 2. CALIFORNIA CITIES AND RESIDENTIAL DEVELOPMENT ............................ Two Sides of the Local Growth Dilemma ............. Cities Awash in Growth ....................... The California Housing Crunch .................. Two Perspectives on Local Governments and the Housing Market ................................ The Indictment: Cities as Housing Villains .......... The Indictment’s Hidden Assumptions ............. The Defense: Cities Managing Cross-Pressures ........ Research on the Effects of Growth Management on Housing Production and Cost ....................... Summing Up ............................... 3. PATTERNS OF GROWTH-MANAGEMENT POLICY ADOPTION ............................... Three Approaches to Managing Growth .............. Restricting Residential Growth................... Linking Growth to Infrastructure ................. Shaping Housing Development .................. Overall Levels of Growth-Management Activity ......... Summing Up ............................... 1 2 5 6 10 11 11 11 13 15 15 17 19 21 25 27 27 28 30 32 35 38 xiii 4. POLITICAL CONTROVERSY AND GROWTH MANAGEMENT ............................ Is Residential Growth Always Controversial? ........... What Role Do Local Citizen Initiatives Play in Growth Management? ............................ What Is the Attitude of City Councils Toward Residential Growth? ............................... Summing Up ............................... Interpreting the Evidence Thus Far ................. 5. FACTORS SHAPING CITIES’ WILLINGNESS TO ACCEPT RESIDENTIAL GROWTH............... Considering Potential Explanatory Factors ............. Local Political Factors ........................ Demographic Makeup of the City ................ City Size, Urbanization, and Location .............. Local Conditions Relating to Growth .............. Controversy and Growth Management: A Two-Way Street ................................. Explaining Housing and Growth-Control Policy Outcomes .. Explaining City Government Orientations Toward Housing and Growth ............................. Explaining Levels of Public Controversy over Residential Development ............................ Summing Up ............................... 6. THE BROADER CONTEXT OF LOCAL RESIDENTIAL POLICIES ................................. Is Local Growth Management at the Root of the Housing Problem? ............................... The Timing of California’s Housing Production Slump .... Less Visible Factors Retarding Housing Supply .......... Initial Zoning ............................. Building Codes............................. Tax Treatment ............................. Construction Defect Litigation................... Factors Fueling Housing Demand and Prices in California .. Desirability of Coastal Areas .................... The Capital Gains Boom ...................... 41 41 44 47 49 49 51 51 51 53 55 56 57 58 63 66 69 73 73 76 77 78 79 80 80 81 81 82 xiv Residential Development and California’s Local Fiscal System ................................ Other “Macro” State Policies Affecting Local Decisionmaking .......................... Environmental Review ........................ The Initiative Process......................... State Infrastructure Shortfalls .................... A Broader Policy Dilemma ....................... Summing Up ............................... 83 84 84 85 86 87 88 7. REFOCUSING THE GROWTH DEBATE ........... 91 Why Do Conflicts over Local Growth Management Have Such Visibility? ........................... 93 The Challenge for Cities: Successfully Managing Conflicts over Growth............................. 94 Two State Policy Challenges: Local Government Finance and Infrastructure Shortfalls ................... 95 Policy Options to Put Housing on a More Equal Footing ... 98 Full “Planning Disclosure” for Homebuyers .......... 99 Land Supply Monitoring ...................... 101 Addressing Condominium Litigation............... 101 CEQA Streamlining and One-Stop Permitting ........ 102 Local Government Finance Reform ................ 103 More Support for Urban Infill Projects ............. 104 Density Bonuses, Inclusionary Housing Requirements, and the “Builder’s Remedy”...................... 105 Housing Action Coalitions ..................... 108 A Final Thought ............................. 110 Appendix: Multivariate Data Analyses: Variables and Results ... 111 Bibliography .................................. 129 About the Authors ............................... 137 Other Related PPIC Publications ..................... 139 xv Figures S.1. Citizen Initiatives and City Growth-Management Policies................................. 3.1. Frequency Distribution of the Number of Residential Growth-Management Policies Adopted ............ 3.2. Average Number of Growth-Management Policies, by Region ................................. 4.1. Number of Growth-Management Policies, by Level of Controversy ............................. 4.2. Citizen Initiatives and City Growth-Management Policies................................. 4.3. Reported Attitude of City Council Majority Toward Residential Growth ......................... 6.1. California’s Housing and Economic Performance, Relative to That of the Nation .................. ix 36 37 44 46 48 76 xvii Tables S.1. Adoption of Residential Growth-Management Policies by Cities in California’s Three Major Regions ........ vii 3.1. Use of Restrictive Residential Policies ............. 29 3.2. Use of Policies Linking Housing Development to Infrastructure............................. 31 3.3. Use of Policies to Shape Residential Growth ......... 34 3.4. Comparing “Nonadopters” of Growth-Management Policies to “Heavy Adopters” ................... 36 4.1. Prevalence of Controversy Regarding Residential Growth Issues .................................. 42 4.2. Effects of Growth Issues on Local Elections ......... 43 4.3. Prevalence of Citizen Antigrowth Initiatives ......... 45 A.1. Descriptive Statistics for Variables in the Multivariate Models................................. 115 A.2. Number of Growth Management Policies Adopted (Regression Model) ......................... 118 A.3. Planner Assessments of City Policies (Ordered Logit) ... 119 A.4. Adoption of Affordable Housing Requirement (Probit Model)................................. 120 A.5. Planner Assessments of City Government Orientations to Residential Growth (Ordered Logit) ............ 122 A.6. City Manager Ratings of Desirability of Multifamily Development (Regression Model) ................ 123 A.7. Planner Assessments of Growth-Related Political Controversy (Ordered Logit) ................... 125 A.8. Citizen Antigrowth Initiatives (Probit Model) ........ 126 xix Acknowledgments This project would not have been possible without the efforts of hundreds of city planning directors, city managers, and other officials who responded to our questionnaires. We thank them for their time and candor. Very useful and thorough reviews of the manuscript came from Michael Dardia, Steven Erie, Mark Schneider, and Fred Silva. Mark Baldassare provided helpful guidance throughout the project. Hugh Louch, Elisa Barbour, Lisa Pitch, and Maria Anna Gonzales assisted in the research in a variety of ways. Peter Richardson provided editorial expertise. Although all of these people improved the final product, the authors are solely responsible for any errors of fact or interpretation. xxi 1. Examining City Growth Policies As California entered the 2000s, the challenge of local growth reappeared as an issue of the moment. By one estimate, the November 2000 general election in California featured 60 measures on city or county ballots that attempted to change local growth and land-use policies. Many of these ballot questions proposed to regulate particular developments or to slow growth, whereas others could be classified as progrowth. In about two-thirds of these measures, voters decided in favor of the slow-growth option (Fulton and Shigley, 2000).1 Historical data on local ballot measures indicate that 2000 was only the most recent peak for local opposition to growth. The number of such ballot measures tends to wane during periods of recession and shoot upward during periods when, as in recent years, population and economic growth have been rapid (Fulton et al., 2000; Glickfeld, Graymer, and Morrison, 1987). Local voter initiatives, which are the most visible manifestation of growth-management policy in California, receive a great deal of media attention. However, they are not the most common or typical method of affecting the amount, pace, or form of local land development. The vast majority of local land-use changes proceed under little scrutiny from the voting public; they are regulated by city ordinances, plans, and review procedures that are less visible to casual observers. Because local governments in California have a substantial degree of discretion in regulating land use and shaping development, serious attention should be given to their role in affecting the production of new ____________ 1A national survey indicated that there were 553 growth-related state and local ballot measures throughout the United States on the November 2000 ballot (Myers and Puentes, 2001). However, that survey defined “growth-related” very broadly to include, for example, parks and recreation bonds, transportation, and school construction issues. 1 housing and other development in the state. In particular, with many observers believing that the state is in a crisis of housing supply and affordability, the role of local governments in allegedly constraining and slowing the approval of proposals for new housing has come under renewed scrutiny. Although numerous journalistic accounts and case studies of local housing policy have been written, there have been relatively few large-sample, empirical analyses of local policies toward housing. In particular, there is a need to set local residential policies, and indeed the state’s housing production issue, in a broader context, and to suggest explanations as to why some city governments take a fairly restrictive stance on new residential development, whereas others are very accommodating. In this report, we take up this challenge, studying the nature and extent of growth-management policies among municipal (city) governments in three major economic regions of California: Southern California, the San Francisco Bay Area, and the greater Central Valley. Why Study Local Growth Policies? Anyone who wants to study growth policy statewide is immediately confronted with a problem: For the most part, local governments make development decisions. In the absence of a clear state policy regarding growth, California’s posture toward residential development is largely an amalgamation of the separate policies of its 475 cities and 58 counties. Through the so-called “police power” delegated to them by the state’s constitution, these local governments have considerable discretion over land use, in pursuit of the overall health, safety, and welfare of their residents. Subject only to conformity with state and federal law and constitutional principles, local governments’ discretion under the police power has been held by the state courts to be as broad as that of the state legislature itself. Land-use regulation may be exercised under the police power not only to protect against obvious public risks (for example, a smoky factory in a residential area) but also in pursuit of such relatively intangible qualities as community character and aesthetic improvement (see Curtin, 2000, pp. 1–4). Furthermore, California legal traditions have given a broad grant of authority to local governments under the 2 police power. For example, the California courts consider development to be a privilege of landowners, whereas in some other states it is considered a right (Curtin, 2000, pp. 228–229).2 Thus, cities and counties have broad authority over land development, including allowing new housing. Although all local general plans must heed a fairly elaborate state law regarding the preparation of the “housing element,” the housing goals and policies announced in these plans are hardly self-implementing. Specific local decisions by planning commissions and local councils typically are necessary to approve new housing subdivisions, even if they appear to conform with the housing element—meaning that local discretion remains a key factor in housing construction. In short, California’s planning and development laws “give local governments wide latitude regarding how they wish to structure the regulatory process” (Department of Housing and Community Development, 2000a, p. 78). Currently, state officials and various interest groups are very concerned about local growth policies, in large part because of the state’s lagging housing production and high housing costs, which have been viewed with alarm in several recent policy reports (California Senate Office of Research, 1999; Department of Housing and Community Development, 2000a; California Budget Project, 2000; Housing Crisis Task Force, 2000; Williams, 2000). Some fear that the recent run-up in California housing prices and the lack of housing production are serious threats to the state’s economic expansion and prosperity (Center for the Continuing Study of the California Economy, 1999). Industry groups and civic associations have mentioned housing costs and availability in such areas as the Santa Clara Valley as impediments to firm location and expansion. Because of fears that local government actions are slowing or preventing housing production, there has been a call for reform in ____________ 2Landowners are generally considered to have a right to develop in California only if they have an approved development agreement with the city or county or an approved vesting tentative map. Local government changes in land-use regulations, such as general plan amendments and citizen initiatives, can sometimes interrupt even developments that have already been approved. The California Supreme Court has held that to have a vested right to complete construction, a landowner must have a building permit, performed substantial work, and incurred liabilities in accordance with the permit (Curtin, 2000, p. 177). 3 everything from permitting procedures to environmental regulations to local government structure. Local policy merits attention also because it has been at the center of recurring waves of antigrowth activism by local residents and interest groups in various parts of the state. As noted, the fairly large number of local voter initiatives on growth indicates this arousal of controversy. Below the level of ballot wars, local controversies simmer in various political arenas, including mayoral and city council campaigns, planning commission hearings, neighborhood-level conflicts over rapid change, and debates over city policy changes on such topics as general plans, housing elements, housing subsidies, and annexation. At a broader level of public opinion, six in ten adults responding to a recent statewide survey felt that their own community had been growing rapidly in the past few years, and 53 percent rated the performance of their city government in handling growth issues as only “fair” or “poor.” A slight majority said that they would favor a (hypothetical) local initiative to slow development in their community, even if it meant having less economic growth (Baldassare, 2001). Local growth policies are also important by virtue of their close relationship to other major policy debates. For example, the debate over suburban sprawl and “smart growth” relates in part to the ability and willingness of older communities, which are usually closer to the traditional urban cores, to accommodate additional housing. If newer, low-density localities at the edge of the metropolis favor housing development more than older cities do, then urban development is likely to be more decentralized and auto-dependent. Such development may also threaten metropolitan areas with a loss of open space and farmland, although there are spirited debates about the nature and extent of these problems (Gordon and Richardson, 1997; Ewing, 1997). Finally, local growth policies are related to California’s continuing debate over its system of public finance. Numerous commentators argue that the state’s system of local taxation and revenue-raising creates poor incentives for balanced land development (Chapman, 1998; Center for the Continuing Study of the California Economy, 1999). 4 The Focus: Cities and Growth-Management Policy Our focus in this study is on city governments and their actions toward residential development. We concentrate on cities for two reasons: their preeminence in regulating residential development and the availability of more useful data for cities than for county governments. Through annexation (spreading out), infill (filling in), or redevelopment (replacing old land uses with new ones), cities accommodate the vast majority of new growth in the state. Of California’s population increase between 1990 and 2000, 94.6 percent occurred in cities.3 In unincorporated areas (those lands outside city boundaries), county governments control land use and development policies. Some development takes place initially in county territory and is later annexed by a city, or incorporates as a city. Moreover, although many counties have a stated policy to direct new urban growth toward existing cities, other counties, such as Sacramento, have actively competed with their cities for new growth. Thus, counties in many parts of the state are also important arenas of residential growth politics. Unfortunately for purposes of analysis, counties present problems. The detailed Census and other data we use later in this report are available for municipalities (cities) and for counties as entire units but not for specific unincorporated portions of counties. Many counties experience growth pressures in several different portions of their territory, and development goals and actions may differ widely in these different areas. Thus, we concentrate on cities because of their predominant role and their coherence as units of analysis, and our empirical conclusions relate only to cities. Much of our focus is on cities’ growth-management policies, the most restrictive of which are sometimes called growth controls. We define local growth management as a class of land-use policies and planning ____________ 3A comparison of city population figures from the 2000 Census to the 1990 Census shows that 82.4 percent of population growth was accommodated by cities that existed as of 1990, and an additional 12.2 percent by cities that incorporated during the 1990s. Only 5.4 percent of population growth was in unincorporated areas. Obviously, however, the newly incorporated cities previously were unincorporated areas, and similarly, some new growth that occurred initially in unincorporated areas was subsequently annexed into cities. 5 guidelines that have been developed for use by local governments in the period since about 1970. Ranging beyond traditional zoning or general plans—although they may be incorporated as part of a general plan— growth-management policies are geared at regulating the pace, form, or location of development within the community. Adopted by local governments (or through local voter initiatives), the policies range from growth boundaries to requirements that traffic standards be met before housing is built to annual caps on the number of building permits issued. For purposes of this report, we do not include under the category of growth-management policies such planning tools or regulations as zoning or general plans, which are more traditional frameworks for development decisions. Indeed, there is no feasible way to measure the comparative restrictiveness of local land-use zones.4 Nor do we include policy decisions relating to park or open-space acquisition, or fees or assessments on new development. Such policies are typically undertaken for reasons not directly geared at controlling growth per se (though they may have consequences for the pace, form, or location of growth). Statewide policies, such as environmental impact review under the California Environment Quality Act, or review by the state’s Coastal Commission, also do not fall under the rubric of local-option growth management devices. However, we will address some of these related policy issues in passing. What Evidence Do We Use? In the ensuing analysis, we draw heavily upon original surveys of top city officials regarding local development strategies and growthmanagement policies. Surveys allow us to ask important local decisionmakers from various parts of the state a range of questions, in a uniform fashion. They also allow us to quantify and compare city ____________ 4Not only are there no regularly updated local zoning maps for localities in California, but there are very difficult issues to resolve in assessing how “restrictive” local land-use zoning actually is. Even in a community with very high-density residential zoning, it might be the case that the large size of commercial or other land-use zones winds up limiting residential development. Zoning policies among new communities and new developments, moreover, tend to be associated with the original social conditions of these new places, often reflecting even the marketing considerations of the developers (Neiman, 1980). 6 policies. Another approach might have been extensive case studies of a large number of municipalities, but this would have proven prohibitive both in cost and in time, and still would not guarantee that the events occurring in the case study communities were broadly representative. In 1998 and 1999, we sent a mail survey regarding local residential policies to the planning directors (or others identified by the planning director as being most knowledgeable about local residential policies) of each city in three major regions of the state: Southern California, the San Francisco Bay Area, and the Central Valley.5 These regions, which together constitute 94.2 percent of California’s population, were defined very broadly for purposes of this study.6 Given the distinctive evolution of the inner Los Angeles area compared to the more diffuse and suburban counties surrounding it, Los Angeles County cities were broken out from the other areas of Southern California. The survey questionnaire asked respondents to indicate whether their city had enacted various local growth policies or techniques, ranging from design review standards to annual caps on the number of building permits issued. Other topics included the local process for reviewing development applications, which factors were most important in slowing residential development, local political conflicts over growth, and the perceived effects of local policies. We also asked for background information on the respondents. After contacting each intended respondent up to four times, we reached a response rate of 76 percent, gathering usable surveys from 297 cities.7 All of the pertinent results are tabulated and briefly discussed in a previous paper (Lewis and Neiman, ____________ 5The Southern California survey was conducted in late 1998, the other two regions in 1999. We do not expect the several-month gap between the surveys to create any significant problem in comparing the responses across regions. 6Southern California is defined to include seven counties: Los Angeles, Orange, Riverside, San Bernardino, San Diego, Santa Barbara, and Ventura, with a total population of approximately 20.1 million. The Bay Area includes nine counties: Alameda, Contra Costa, Marin, Napa, San Francisco, San Mateo, Santa Clara, Solano, and Sonoma, which collectively have 6.9 million residents. The Central Valley is defined as the following 18 counties: Butte, Colusa, Fresno, Glenn, Kern, Kings, Madera, Merced, Placer, Sacramento, San Joaquin, Shasta, Stanislaus, Sutter, Tehama, Tulare, Yolo, and Yuba. The Central Valley’s population is about 5.6 million. 7One additional city from Southern California responded to only a portion of the survey. It is included where possible in the analysis. 7 2000),8 but this report focuses in a more systematic and thematic way on the relevant survey results. To a lesser extent, we also draw upon another survey, entitled “Development Strategies in California Cities,” which was mailed in 1998 to the top administrative official of every city in California—generally the city manager or city administrator. Seventy percent of the cities then in existence in the state returned a usable questionnaire, meaning that 330 were received. This questionnaire was less detailed than the one discussed above, as it asked broadly about the desirability of various potential land uses (including single- and multifamily housing and various commercial and industrial categories) to each city’s administration, both in areas of new development on vacant land and in redevelopment zones. Other topics included an assessment of the importance of various factors (such as job creation, affordable housing provision, and general plan consistency) on city development strategies and some analogous questions about city strategies for annexation. A tabulation of all of the results is presented in Barbour and Lewis (1998).9 Although any survey carries with it some risk of measurement error, several factors affirm the validity of the data we have collected. As noted above, response rates were quite high, at 76 and 70 percent. The cities responding to the planner survey have a collective population of 19.4 million residents, and those responding to the city manager survey 22.9 million, according to population statistics from the 2000 Census. These numbers represent 57.3 percent and 67.6 percent of the statewide population, respectively. Furthermore, response rates were generally quite similar across regions and across city population-size categories, although very small jurisdictions responded at a lower rate. Another aspect of validity is whether responses to the survey questions can be taken as accurate representations of local policies. Given the potential sensitivity of several of the questions, respondents were promised anonymity, and we do not reveal the specific responses of ____________ 8The full text of the survey results is available at http://www.ppic.org/publications/ occasional/lewis.neiman.pdf. 9Other publications on various growth-related topics that have made use of these survey data include Lewis and Barbour (1999), Lewis (2001), and Lewis (forthcoming). 8 any individual cities. Therefore, there was little motivation for respondents to “fib” in answering the survey questions. As expected, the cities that scored high in the number of growth-management policies were often mentioned as strict growth regulators in the media or in other literature on the topic. More solid support for the validity of our survey data derives from our comparison of the results of our survey to that of an earlier survey of local growth management policies by Glickfeld and Levine (1992, survey taken in 1988). For the 248 cities that responded to both the Glickfeld and Levine survey and to our survey ten years later, the correlation in the total number of policies adopted was 0.34. We find this relatively solid degree of association across the two surveys to be a heartening validity measure, especially given the elapsed time, the differences in survey methodology, and the variety of policies that respondents were asked about. There is also a healthy correlation between responses to two somewhat comparable questions about local affordable housing policy, one of which appeared in our city manager survey and one of which in our planner survey.10 Finally, although the survey data are two to three years old—and therefore, some individual cities will have modified their policies—there is little reason to believe that the major differences among cities and among regions, which we will highlight in this report, have changed to any great extent. Other data on community characteristics are drawn from the Census—from 2000, where data are available, but otherwise from the 1990.11 We also use state data regarding local demographics, fiscal ____________ 10A binary item on the planner survey asked whether the city had a policy that new residential development must include an affordable housing component. A Likert-scale item on the city manager survey asked respondents to assess the importance of “meeting affordable housing needs” when assessing development and redevelopment proposals. The correlation between the two sets of responses, for cities responding to both surveys, is 0.25. Each variable is a positive and significant predictor of the other in a regression model, even when controlling for numerous other city characteristics. 11At this writing, only a portion of data on residents and housing units measured by the 2000 Census has been released for California cities. For our purposes, these include population counts by race or Hispanic origin and measures of homeownership and housing type. We must rely on the 1990 Census for a handful of other variables. However, this need not be a disabling problem, as community characteristics and relative differences tend to persist over time. For example, the percentage of whites, blacks, 9 characteristics, and political party registration.12 Finally, we draw somewhat on news coverage of local growth and housing issues around the state. Summing Up Local governments have a large degree of authority in the development of land for new housing, thus putting them in a pivotal position for shaping the nature and amount of housing production in the state. Cities, in particular, control the land-use policies relating to most new residential growth. California’s controversies over local antigrowth activism, housing affordability, sprawl, and other topics demand that serious attention be paid to local residential policies. In particular, we will highlight the extent of, and motivations for, city growthmanagement policies that attempt to regulate the pace, form, or location of new development. To do so, we rely heavily on recent surveys of planning directors and city managers, which drew solid response rates. Chapter 2 introduces the debate regarding cities’ role in the California housing market and is followed in Chapter 3 by a description of the amount and types of city growth-management policy in each region. Chapter 4 examines some of the political dimensions of growthmanagement policy adoption, looking in particular at the role of local political controversy, citizen initiatives, and the city council. In Chapter 5, we assess possible explanations as to why cities choose particular growth orientations, with some cities favoring further housing development and others acting to restrict growth. Chapter 6 sets these city policies, and the debate over housing production in general, within a broader context of statewide factors that constrain housing supply, increase housing costs, and diminish cities’ enthusiasm for residential development. Finally, Chapter 7 suggests a number of policy implications of our analysis and sets out some specific policy options worth consideration. ________________________________________________________ Asians, and Hispanics in the 456 California cities existing as of 1990 is correlated at very high levels (r > 0.9) with the shares of those groups in these cities as of 2000. 12In a few instances, we also refer to whether the city council is elected by districts or at-large. For this information, we draw upon a questionnaire sent by PPIC researchers in 2000 to all city clerks in the state. (The response rate on that question was 82 percent.) 10 2. California Cities and Residential Development Are local growth-management policies the villain in California’s housing woes? This chapter describes the state’s housing dilemmas and sets out city governments’ purported contribution to the problems of undersupply and affordability. In contrast to the sinister portrayal of cities’ roles sometimes discussed, we propose a more nuanced account that focuses on city governments as arenas for resolving difficult political cross-pressures regarding growth. Two Sides of the Local Growth Dilemma The character of the debate over residential development in California depends on one’s vantage point. From the standpoint of many cities and their residents, growth is a powerful wave washing toward their community providing opportunities and responsibilities, but also threatening the character and traditional scale of their community. From the standpoint of many state officials, business leaders, housing advocates, homebuilders, and would-be homeowners, housing development is a numbers game that the state is losing. Underproduction of new residential units, it is said, threatens the state’s vision of itself as a place of opportunity, where homeownership is a part of the California Dream. Others emphasize that rising housing costs for workers become reflected in the costs of doing business in the state, thereby undermining the state’s desirability as a place to conduct business. This section sets out the two sides of the growth dilemma. Cities Awash in Growth To some, it seems odd to suggest that the state’s local governments are not doing enough to accommodate new housing and population growth. According to the U.S. Census, California’s population grew by 11 4.1 million during the 1990s; this increment alone is a greater number of people than the populations of 25 states in 2000. Growth swept all major regions of the state, with both metropolitan Southern California and the San Francisco Bay Area increasing by 12.6 percent and the Central Valley by a prodigious 19.6 percent. Among California cities, the average rate of growth over the 1990s was 14.4 percent (15.8 percent in those cities responding to our growth management survey). Equally notable was the increased diversity of populations and land uses across communities. According to the 2000 Census, 55 percent of all the cities in the state had populations that were at least 20 percent Hispanic; a quarter of cities were at least 10 percent Asian. At the same time, jobs and commerce were also spreading beyond traditional centers. California’s “postsuburban” form is much remarked upon, as outlying centers join central cities as important employment centers. Although some have decried this loss of urban centrality, it is worth pointing out that the movement of jobs toward the suburbs goes a certain way toward improving the job/housing balance in the state’s metropolitan areas, bringing employment to where most workers live. According to commuting data from the 1990 Census, 52 suburbs had more than 1.19 jobs per resident worker, which was the average job/worker ratio in the state’s 49 Census-designated central cities.1 Our main points are that growth has continued to be rapid despite the deep recession in the early part of the 1990s, and that growth’s changes profoundly affected most corners of the state. Cities have been able to accommodate the state’s increasing demographic and economic diversity, generally without becoming paralyzed by social or political crises at the local level. Compared to protracted and bitter debate at the state level over issues such as immigration and affirmative action in the past decade, cities mainly appear to have been in the business of adapting to, planning for, and servicing new residents and businesses. When growth seems overly rapid, however, the “equilibrium” conditions of local politics and community life can be upset. Traffic ____________ 1Authors’ calculations from the Census Transportation Planning Package— California Element. (Data are available only for cities of at least 2,500 population.) Data for 2000 are not yet available. 12 levels may increase to the point that local roadways are overburdened. Local school systems may be forced into year-round sessions or the use of portable classrooms. Open spaces that are treasured by many members of the community may be lost. Rapid growth may fundamentally change the character of some communities. In some cases, the sheer pace of development itself may be threatening to some local residents, who are used to growth occurring, but at a much slower pace. The California Housing Crunch From another standpoint, local governments have not done enough to accommodate the pressures for new housing development in the state, and one must go no further than the numbers to be able to demonstrate this. For example, California’s share of the nation’s housing production fell precipitously for much of the decade. Even in the midst of fairly solid population growth in the 1990s, the state’s share of permitted residential units in the United States fell from 14.7 percent in 1990 to 8.3 percent in 1999. Although there are numerous reasons why this could be so, the decline seems at first blush to be fairly dramatic, and local government growth controls have been prominently mentioned as important contributors to the problem. According to this point of view, cities’ lack of enthusiasm for housing—and in some cases, outright hostility to new residential development—goes a long way toward explaining California’s lagging housing production.2 Other indicators of housing problems in the state are easy to find. The median price of an existing, detached, single-family home in California was $267,810 in July 2001—and $481,280 in the San Francisco Bay Area. The percentage of California households owning their own homes (57 percent as of the 2000 Census) remains significantly lower than the overall national rate (66 percent). More stark ____________ 2In reality (as we shall note again in Chapter 6), it is difficult to draw very profound conclusions from California’s lower share of national housing production. Household formation in California rose by 10.8 percent between 1990 and 2000, according to the 2000 Census—a lower rate than the 14.7 percent national increase in household formation. The United States added 1.01 net new housing units for every additional household, whereas California added 0.92 net new housing units per additional household. 13 is the difference between the so-called “affordability ratio” in California as opposed to the overall U.S. ratio. The affordability ratio is one measure analysts use to examine housing costs in relation to the ability of people to pay. It represents the proportion of households in an area that could afford to purchase the median-priced single-family home for sale at a given point in time, using normal assumptions about the percentage of income that can go toward home payments. The affordability ratio was 32 percent in California as of August 2001, compared to 54 percent in the nation at large.3 For large numbers of families, particularly in the younger age brackets and lower income levels, renting an apartment is a more relevant option than homeownership. Here too, however, California’s housing numbers are discouraging. In fact, multifamily housing was the sector of the residential market that deteriorated most rapidly in production in the 1990s (Landis, 2000, p. 33; California Budget Project, 2000, p. 38). As almost any renter in the Los Angeles region, San Diego, or the Bay Area would quickly confirm, rents rose very rapidly during the recent economic upswing and most notably in the coastal regions. Average household size—that is, the number of persons per dwelling unit—has been on the increase in California—from 2.75 in 1980 to 2.87 in 2000, according to Census data. Some attribute these increases in household size to declines in housing affordability (through doubling-up or taking in boarders), although most of California’s upswing in household size is likely attributable to the rising numbers of immigrants, who traditionally reside in units with larger numbers of residents. Another way planners and social scientists examine housing production is by comparing the number of new housing units in an area to the number of new jobs. In the United States as a whole in the late 1990s, approximately one new housing unit was being produced per two new jobs; in California (between 1994 and 1998), the ratio was one new housing unit per 3.9 new jobs. In some areas of rapid employment growth, such as the Silicon Valley, these ratios were far worse, with more than eight jobs for each new housing unit, by some estimates (California ____________ 3The data in this paragraph are from the website of the California Association of Realtors (www.car.org). 14 Budget Project, 2000, p. 39). These figures are one measure of an imbalance between jobs and housing. California went through a similar economic cycle in the 1980s, with a recession early in the decade and rapid growth in the latter portion of the decade. Nevertheless, the state’s ratio of net new jobs per net new housing units in that decade was a relatively healthy 1.6 (based on a net growth of 1.9 million dwelling units and 3.0 million jobs statewide in that decade, according to the Census). Clearly, something was different in the 1990s that did not allow housing production to meet the need to as great a degree.4 Local government actions have been viewed as one possible culprit. Two Perspectives on Local Governments and the Housing Market Assessing the role of local policies in California’s housing problems requires that we to move beyond the rhetoric that marks much of the debate and consider the full context of city actions and the nature of their policies. That is the major task of the rest of this report. Here we discuss one account that sees municipal governments as a major element of the housing production problem, followed by another perspective that views city governments as agents of compromise in a multisided political debate about the costs, benefits, and consequences of population growth. The Indictment: Cities as Housing Villains According to strong critics of local government growth-management policies, both scholars and journalists, local policies have a particularly harmful effect on housing opportunity for the average state resident. Perhaps most often cited of these critics is Bernard J. Frieden, whose 1979 book, The Environmental Protection Hustle, offered a stinging indictment of local growth controls that had begun to take shape in the 1970s. Frieden’s book and its broad-brush condemnation of local ____________ 4It is quite possible that there was overbuilding in residential real estate in the 1980s, particularly in Southern California, and therefore California’s low production in the 1990s might be partially attributed to its excess inventory. This argument is less persuasive for the Bay Area, however. (We thank Michael Dardia for making this observation.) 15 development controls continue to be cited approvingly by critics who see local growth policies as too stringent (Taylor and VanDoren, 2000). Even though Frieden’s study of Bay Area residential development restrictions is over 20 years old, his view reflects a long-standing interpretation of local development controls: that they are inspired by elitist, snobbish populations who disguise their exclusionary motivations with references to quality-of-life amenities and environmental protection. In this view, socially advantaged current residents are pulling up the ladder, so to speak, to prevent others from coming to their communities. Many studies of local housing policies, especially of suburban jurisdictions, have critiqued so-called snob zoning, especially large-lot and exclusionary zoning. Usually accompanied by elaborate building codes and subdivision regulations, such zoning has long been associated with elite suburban aversion to low-income and higher-density housing, particularly in the East and Midwest (Adang, 1964; Babcock and Bosselman, 1973). Indeed, by the time Frieden concluded that some communities were gaming environmental law to prevent housing for the less affluent, he was merely dramatizing and perpetuating this suspicion of local policies onto the California scene, where growth-management policies rather than zoning were at issue. Frieden’s work was an important intellectual watershed in the debate over the effects of local land-use and development policy on housing supply and prices, and on the geographical distribution of low-income and minority families, and it continues to be central to contemporary research and policy debates over zoning, land use, growth control, or “smart growth” (Bogart, 1993; Bollick, 2000; Hartnett, 1993; Shen, 1996). Because recent commentators have largely echoed the critical emphasis Frieden placed on local growth controls, his arguments are worth a closer examination. First, critics of local policy make the argument that growth controls are ubiquitous, an increasingly common feature of the local regulatory scene. Although Frieden’s case studies, and those of many others, tend to be drawn from Bay Area communities such as Marin County with famously finicky attitudes toward growth, growth controls are sometimes held to be sweeping the state and nation (see also Dowall, 1984). A second allegation is that local growth-management policies have the overt aim of stifling residential development. According to the critics, local 16 growth controls are often couched in terms of environmental goals, but these function largely as a smokescreen for baser motivations. Some of these critics allege that existing property owners mainly wish to limit new housing development, thereby inflating the prices of their own homes (or the rents of their apartments); in short, they influence local governments to pass restrictive policies to creative an artificial shortage of housing (Brueckner, 1995). Other critics argue that the motives behind growth management are more social than economic. From this point of view, restrictive local regulations on residential development—particularly those limiting multifamily or “affordable” housing construction—may have their basis in simple snobbery, or a fear of newcomers, perhaps with a touch of racism or ethnocentrism underlying the restrictions. Indeed, it is not difficult to find anecdotes and case studies that support the snobbery hypothesis (Frieden, 1979; Danielson, 1976). Finally, the condemnatory account of local growth controls holds that they are a major cause of housing shortfalls, seriously reducing the number of buildable lots in metropolitan communities. In short, because of their ubiquity and their growth-stifling intent, local government growth restrictions are said to shut the poor and middle class alike out of numerous communities, particularly in the prosperous suburbs. The Indictment’s Hidden Assumptions Several dubious assumptions underlie the indictment. Such assumptions include the following: • Growth-management policies are effective in slowing or restricting growth at the local level. The mere passage of a growthmanagement measure—either at the ballot box, as an ordinance by the city council, or as an administrative implementing procedure in the planning department—does not ensure that it will manage growth effectively. In some cases, such policies can be symbolic efforts that create few binding constraints on local development; in other jurisdictions, growth controls may be weakened over time or ignored by those responsible for implementation (Warner and Molotch, 2000; Logan and Zhou, 17 1989). Moreover, the threat of future controls might even stimulate current growth. In fact, as we will discuss below, evidence on the actual effects of local growth regulation is quite mixed. • No legitimate public policy motives underlie local growth management. Although the indictment focuses on the purported flimsiness of environmental protection as a reason for growth control, other defensible motivations are certainly possible. Cities may experience a decline in public services such as public schools or recreation because of rapid residential growth. Infrastructure, such as streets, school buildings, water, sewerage, and municipal utility systems, may be overburdened or at capacity. There may be an insufficient number of jobs available nearby for new residents, necessitating long commutes for residents. Finally, the community may be interested in taking a step back from the intensity of new growth to assess its plans for the future and consider how its amenities—including the natural environment—might best be preserved. Chapter 5 explores some of these motives in more depth. • Any local government restrictions that reduce the number of units built below what the builder has proposed results in a net reduction in the local housing supply. This assumption overlooks the “law of anticipated reactions”: If builders expect to have difficulty getting all their units approved, they are likely to propose more units than they otherwise would. Under such conditions, builders may be likely to propose a higher-than-optimal number of units in their new developments, knowing that this number is likely to be reduced. Furthermore, builders are not obligated to build the number of units approved, and they occasionally postpone or reduce construction for reasons having little to do with local regulations, such as housing market conditions or financing difficulties. • Builders have few choices if they cannot build in the more restrictive communities. If builders shy away from some communities because of restrictive political conditions, or if their housing proposals are rejected, does this leave them without options? 18 Not if there are other, comparable communities in the region available to accept similar projects. California contains hundreds of jurisdictions, presenting housing developers with a wide portfolio of options for proposing housing projects. The actions, or regulatory reputations, of individual cities may lead to housing production being “moved around” within a region, as opposed to a net reduction in the number of units produced. Of course, this may raise the likelihood of sprawl, or decentralized development patterns, if the growth-accepting jurisdictions are located farther from central job areas. But that is not the same result as thwarting housing production entirely.5 • Local governments are the agents of slow-growth interests or are willing participants in the restriction of housing. Are city governments merely puppets of antigrowth activists or homeowner cartels? Even a casual acquaintance with a handful of local governments makes this difficult to believe. Moreover, there is a body of academic literature on urban political economy that tends to conclude that rather than cultivating a slow-growth environment, a number of forces create a progrowth orientation among most local governments. These forces include the desire to enhance the importance of the locality through growth, and the urge to attract new businesses to provide local jobs and investments by developing the local customer base with population growth. One must also consider the political importance of major campaign contributors, which notably include real estate interests, local retailers, and other elements of the so-called “growth machine” (Logan and Molotch, 1987; Stone, 1989). The Defense: Cities Managing Cross-Pressures It is likely that local political arenas are more pluralistic than is typically admitted either by the progrowth indictment of local ____________ 5Moreover, it is not clear that an “unregulated” system of land use, in which all communities wanted and welcomed growth on the developers’ terms, would produce less “sprawl” than the current arrangement. 19 government or by the antigrowth critics’ portrayal of local policymakers’ “caving in” to developers at every turn. Rather, cities in many rapidgrowth regions are in the challenging position of feeling pressure from both sides. On the one hand, cities have the need, responsibility, and incentive to allow for new housing; on the other, there is frequently pressure from local residents and certain interest groups to prevent overly rapid growth, or specific housing projects, from potentially harming the community’s amenities and perceived quality of life. In a state as large as California, there are no doubt some jurisdictions in which either housing developers or antigrowth activists have been able to dominate the local political system. Perhaps the more typical situation, however, puts city elected and appointed officials in the position of attempting to manage the conflicts and controversies engendered by growth. Compromise, after all, is inherent in democratic politics. Antigrowth activism has certainly developed into a powerful grassroots movement in a variety of localities, as the number of restrictive local citizen initiatives on land-use topics indicates. There are also many communities, however, where growth management fails to reach the policy agenda, or is defeated. There are even a few localities in which growth is nearly always welcome and where accommodating housing is viewed as a duty. In our experience, most city officials hardly seem to be slow-growth extremists. The need for housing and the many equity problems that result from excessive regulation of local residential development are central features of housing officials’ professional training. Housing officials, moreover, are at the receiving end of much information regarding the “crisis” of housing. The need for housing, and housing affordability problems, furthermore, are experienced by local residents in jurisdictions throughout the state; many of these residents, and their sympathetic family members, also vote and become active in politics. With evidence accumulating rapidly from news media and policy reports that the state faces serious housing shortfalls, many mayors, councilmembers, city managers, and planners surely feel a responsibility to do what they can to accommodate what they believe is needed additional housing. Pressure from the business community to keep the local cost of living from escalating too far for their employees adds to the 20 pro-housing forces likely to affect local officials. In addition, one should not underestimate the influence of progrowth coalitions in many jurisdictions. Of course, local policymakers do not make decisions about housing and land use in a vacuum. There are constraints on what they can accomplish, based on the land area of the city and its capacity for siting new housing, the local and regional infrastructure, the system of public finance that funds services for new residents, and the nature of the housing market and the development industry in their area, among other factors. Later in the report, we will highlight the importance of many such contextual factors and show how they may be important influences on housing production in themselves. Research on the Effects of Growth Management on Housing Production and Cost Before taking a closer empirical look at the types of residential policies California cities employ, it is worth reviewing what is known about the effectiveness and effects of local growth-management policies. Although our mail survey on local policies is too recent to allow us to assess their effects on housing production or price, some studies have begun to explore this issue. Both critics and advocates of local growth controls frequently seem to assume that such policies are effective in holding down growth rates. Criticisms of such controls further maintain that such efforts to restrict supply necessarily lead to increased housing prices (Advisory Commission on Regulatory Barriers to Affordable Housing, 1991). However, our review of empirical studies does not allow any such confident and certain conclusions. If anything, the relevant literature may lean, on balance, toward finding that growth control lacks significant effects on population growth or housing prices—but the results are so mixed that one would be hard-pressed to offer this as a definitive conclusion. Below are some examples of major published studies—conducted by economists, urban planners, sociologists, and political scientists, and using a variety of analytic methods. 21 • An examination of 387 suburbs nationwide, focusing on environmental-protection-oriented growth restrictions, found “only modest effects on subsequent change in local population, median family income, median rent, and black percentage. We argue that formal policy tools or legislation cannot be accepted as indicating that their stated objectives will be realized” (Logan and Zhou, 1989, p. 461). • By contrast, a study of 63 municipalities in Ohio found “consistent and robust evidence that subjecting rezoning decisions to public referenda created a housing unit ‘growth penalty’ for cities” (Staley, 2001, p. 25). Note that this result focuses on the popular referendum requirement per se, not the broader class of growth-management devices. The argument is that referendum requirements introduce a large degree of uncertainty for developers, which hinders investment. • A study of 97 Northern California cities through the 1970s found that population “growth rates are not influenced by growth controls or the social variables associated with antigrowth policies” (Baldassare and Protash, 1982, p. 339). • A study of 490 California cities and counties found that local growth management policies “significantly displaced new construction, particularly rental housing, possibly exacerbating the expansion of the metropolitan areas into the interiors of the state.” However, not all growth controls are created equal: “Measures which limited available land or which downsized existing zoning had stronger effects” (Levine, 1999, p. 2047). • An earlier study by the same author and a colleague, however, concluded that “Growth measures do not appear to have reduced construction activity significantly at the state, the county, or the metropolitan level” in California (Glickfeld and Levine, 1992, p. xii). • An examination of the region of California with perhaps the strictest growth controls—the South Coast of Santa Barbara County, in the 1974–1979 period—concluded that a shortfall of housing production resulting from water hookup moratoriums 22 and zoning changes accounted for about one-quarter of the change in real house prices (Mercer and Morgan, 1982).6 • An examination of Southern California suburbs during the 1980s, which used many of the same measures of growthmanagement policies used in the current report, concluded that “Restrictive growth-controlling cities do not appear to become richer or poorer; they appear to become less black” (Donovan and Neiman, 1995, p. 790). No significant effects on local population growth were detected. • Warner and Molotch (2000, Appendix A) used a pooled crosssectional design to study 11 Southern California localities across the period from 1971 to 1990, looking for the effects of newly adopted growth controls (with a two-year lag) on annual percentage changes in housing units. Once proper statistical controls were introduced, they found “scant evidence that controls had much of an effect, particularly on the supply of new housing” (p. 52). • A particularly relevant analysis by Landis (1992) compared seven midsized California municipalities that had restrictive growth controls to six well-matched local governments without growth controls. Examined over a decade-long period, the author found that the local policies were “largely irrelevant to the management of urban growth”; there were no major differences between the growth-control cities and their matched pairs in terms of population growth rates, housing production shortfalls, or home price increases. Why were effects not detectable? Landis suggests that the control policies were “porous” and often fairly generous, that there were likely spillover opportunities for housing development nearby, and that price effects of local policies were likely overwhelmed by regionwide factors—a point we shall return to in Chapter 6. ____________ 6However, the study suffers from a case of extreme collinearity (r = 0.97) between the variable for housing-unit shortfalls and a measure of countywide employment, which may render its conclusions tenuous. (See Mercer and Morgan, 1982, p. 215, fn 12). 23 Overall, it seems safe to say that there is simply no consensus among the experts, as of yet, on the effects of local growth controls. Some critics of local growth management approvingly cite the sophisticated review of other studies by William Fischel (1990), entitled Do Growth Controls Matter? In the first paragraph of the review, Fischel states, “The answer to the title’s question is yes. . . . The effects are evident in land values and housing prices” (p. 1). However, most of the studies Fischel cites that lead him to this conclusion are studies of local zoning, not contemporary growth controls; he “decline[s] to make a sharp distinction between growth management and traditional zoning” because both derive from the local police power (p. 3 fn). In another context, he notes that the most exclusionary local governments do not typically adopt modern growth controls because “the elitist communities long ago adopted strict zoning regulations, so that additional growth controls are unnecessary” (p. 33 fn).7 Perhaps, then, it is zoning—a much older and more established variety of local land-use regulation— that should receive more scrutiny from policymakers, the media, and scholars. Even so, modern growth management, not zoning, draws most current-day attention to local residential policies in California. On this topic, Fischel is more equivocal in summarizing the results of econometric studies. He notes the lack of evidence in growth-control studies for decreased housing supply (p. 33). And although several studies have found growth management to be associated with significantly higher housing costs, he notes that such increased prices may simply be the result of the increased amenity levels brought about by growth control policies: We know . . . that community and neighborhood amenities raise housing values. Growth controls may create residential amenities, or, for rapidly growing communities, they may prevent impending disamenities. Hence the ____________ 7Logan and Zhou (1989, p. 464 fn) similarly find that suburbs in their sample that appear to exclude multifamily housing through zoning were less likely to adopt growth controls. 24 higher housing prices could be taken as evidence that growth controls do what they are intended to do (p. 33).8 On balance, Fischel is inclined to believe that growth controls probably do have more social costs than social benefits. However, he is tentative on this point “because only a few studies have addressed it in a persuasive framework” (p. 53). Clearly, there is a need for additional scholarship on local growth-management’s effects, especially in the California context. Before researchers proceed to that difficult task, however, it is important to characterize the actual extent of growth controls and to learn something of the motivations behind them. Those issues are the main focus of the current report. Summing Up California is once again growing very rapidly, as it has for nearly all of its history. The absolute increase in the number of residents can seem staggering—and no doubt is staggering to residents of many individual communities. Still, despite receiving millions of new residents and building hundreds of thousands of new housing units, California is widely thought to have systemic problems in building the amount of housing necessary to accommodate the state’s burgeoning population. The state’s housing production has been well below what one would expect, given the rapid increase in jobs and households. Do local growth-management policies, such as annual housing-unit caps, greenbelts, or moratoriums on development, play a major role in the state’s housing problems? One line of argument holds that they do, because growth controls allegedly are ubiquitous and very often have the overt aim of limiting housing production. We have suggested a number of possible shortcomings in this argument, because of its simplistic assumptions about the behavior of local governments and the nature of ____________ 8Similarly, early studies that found housing price increases resulting from strict controls in Petaluma and Davis are difficult to generalize from because of the potential uniqueness of those communities. “Petaluma was a famous test case, and Davis is a university town known for its attention to environmental amenities. Notoriety itself might have sent housing prices up by signaling to house buyers the existence of an exclusive community” (Fischel, 1990, pp. 31–32). 25 local residential policies. Moreover, the research literature on growth controls lacks any real consensus on the issue of whether such policies have much effect on housing production or prices. Instead, we have suggested that city politics is likely to be an arena for the balancing of pro- and antigrowth interests, with policy choices that vary widely among cities depending on local conditions and the capacity of the locality to accommodate more growth. The next step, then, is to examine the patterns of adoption of various types of local growth-management policies. 26 3. Patterns of GrowthManagement Policy Adoption This chapter describes three major categories of growth-regulating policies, and draws upon the results of our 1998–1999 survey of 297 city planning directors to report on the use of these policies by cities in the major regions of California. Because the aggregate data tend to mask major distinctions among different areas of the state, we view them by region. Three Approaches to Managing Growth In our survey, we asked city planners questions about cities’ adoption of 16 specific types of growth-management policies. Upon closer examination, these policies can be classified fairly readily into three major groupings, based on the overall purpose of the policy. First, some policies represent more or less overt attempts to restrict residential development—that is, to slow down its pace by reducing the number or types of residential units that can be developed. These are the most overt growth controls. The second group of policies includes those that link residential growth to the infrastructure capacity of the community—for example, by requiring that roads or other public facilities be deemed adequate to meet the demands of an increased population. The third and perhaps least-restrictive type of growthmanagement policy has the goal of shaping the form, style, or location of new housing within the community. The threefold categorization of growth-regulating policies is intuitively sensible and reflects the major substantive approaches to managing local growth. Although it is possible to design a number of alternative strategies to measure local policy (e.g., Glickfeld and Levine, 1992), the first test is to determine whether the simplest, most straightforward approach produces useful findings. 27 We did not ask overall questions about the city’s general plan or zoning. Although these are extremely important elements of the local regulatory apparatus, they are not normally considered under the rubric of “growth management.” All California cities and counties are required to have a general plan (including a housing element) and zoning consistent with that plan, whereas the growth-management policies we examine are adopted at the discretion of each community.1 Partially as a result, the policies we examine tend to be the most visible and controversial elements of local land-use regulation. However, we do ask respondents to express their views regarding the overall effect of their land-use policies on their communities, which includes zoning policy as well. As we shall discuss later in this report, initial zoning may be worthy of much closer scrutiny in future research. Restricting Residential Growth We now discuss cities’ use of each of the three major groups of growth-management policies. Under the category of “restrictions” on housing, we included the following seven city actions: • Annual limits on total building permits, • Annual limits on residential units authorized, • Annual limits on multifamily dwelling units built, • Recent substantial reductions in land zoned for residential use, • Policies linking local residential growth rates to some formula or external growth rate, such as that in the county, region, or state, • Formal policies prescribing a local population ceiling, and • Moratoriums on building permits, water connections, or sewer hook-ups. ____________ 1Under California law, specific ordinances and planning and subdivision regulations must, under nearly all circumstances, demonstrate “consistency” with the local general plan, which is held to be something of an overall constitution for local development decisionmaking. Therefore, as some growth-management policies are adopted, whether by the city council or in a local initiative election, they often contain language that modifies the general plan to ensure consistency. Although the state’s intention was to ensure that local policies were consistent with the local plan, in reality the plan may also be amended to reflect new policies. However, some cities have adopted general plans that show a strong growth-management flavor throughout. 28 Although some might choose to include the last policy on the list with the infrastructure linkage policies, to be discussed in the next section, such moratoriums share with the other policies listed above an overall character of overt restriction. Each policy either limits the number of units that may be produced or levels of population increase, or takes land for residential purposes out of circulation. How common are these most restrictive of growth-management policies? Our survey responses indicate that most of these policies are uncommon in all regions, as Table 3.1 indicates. Annual caps on building permits, residential units, or multifamily dwellings are employed by less than a tenth of all cities responding. Recent reductions in the amount of land zoned for residential use are even less common, and virtually nonexistent outside the San Francisco Bay Area. Somewhat more common are official population ceilings and temporary moratoriums on Table 3.1 Use of Restrictive Residential Policies Policy Annual limit on building permits Annual limit on housing units constructed Annual limit on multifamily dwellings Recent reduction in residential zoning Formula for allowable annual growth Official population ceiling City has had a moratorium Mean number of these policies used per city L. A. County (N = 56) 2 2 5 2 5 7 33 0.57 Percentage of Cities Using Policy Other So. S. F. Bay Central California Area Valley Total (N = 91) (N = 77) (N = 73) (N = 297) 4 10 5 6 9 18 7 9 1 57 4 1 50 2 2 84 5 16 12 12 13 30 32 25 30 0.65 0.90 0.60 0.69 Notes: Percentages represent the number of respondents indicating use of the policy as a proportion of all those providing an answer to the question, including “Don’t know” or “N/A” responses. For more detailed tabulation, see Lewis and Neiman (2000, pp. 5–6). 29 growth. The former exists in about one in eight cities, and the latter in about three in ten cities.2 Although moratoria and population ceilings can sound somewhat severe, their actual long-term effect on housing production is uncertain. Some cities have established generous or optimistic growth ceilings, and such an overall “population cap” may prove more symbolic than binding in the long run. As for moratoria, they are by definition temporary, and typically must be based on some infrastructure or planning deficiency that needs to be overcome (often depleted sewer capacity), at which point residential permits will be processed again. Our question did not ask how recently the moratorium had occurred in the city.3 Four of these seven restrictive policies are most commonly adopted in the Bay Area. As the last line in the table indicates, the average Bay Area city has about 0.9 restrictive policies of the seven possible, compared to about 0.6 in the other three areas. Bay Area communities are substantially more likely to adopt some overtly restrictive policies, such as annual limits on building permits and housing construction. Nevertheless, across cities in all the regions studied, 166 (55.9 percent) indicated that they employed none of these policies; only 40 (13.5 percent) have more than one such policy. Linking Growth to Infrastructure Under the category of “linking” growth-management policies are the following four city requirements: • Policies that require traffic standards to be satisfied before new development occurs, ____________ 2Logan and Zhou (1989, p. 464), using data from a 1973 survey, found that 20 percent of suburbs in a national sample had imposed growth moratoriums. 3The exact wording of the question was, “To the best of your knowledge, have there been any MORATORIA on building permits, water connections, or sewer hook-ups in your city?” Although it is possible that planners who have been employed longer are more likely to be aware of past moratoria, our data indicate that there is no correlation between the number of years of experience reported by respondents and their response as to whether a moratorium had occurred (r = –0.02). 30 • The use of capital improvements and public works projects (such as street widening or sewer capacity) to control the rate or location of residential growth, • Requiring a popular vote to increase sewer capacity, and • Limiting the number of annual water connections. Each of these relates proposals for new development to some concern over the city’s physical infrastructure. The last two policies share some commonalities with the more restrictive policies discussed above, particularly the annual limit on water connections, which can function as a numerical cap for annual housing construction. As Table 3.2 indicates, however, the percentage of cities embracing an annual limit on water connections (4 percent of all cities responding) and the popular vote requirement for sewer capacity increases (3 percent) is extremely low, as was the case for the more overt caps on housing units discussed above. In other words, few cities report the use of Table 3.2 Use of Policies Linking Housing Development to Infrastructure Percentage of Cities Using Policy L. A. Other So. S. F. Bay Central County California Area Valley Total Policy (N = 56) (N = 91) (N = 77) (N = 73) (N = 297) Satisfy traffic standards before allowing development 23 26 34 21 27 Use capital improvements to control rate or location of growth 7 13 13 21 14 Popular vote required for sewer capacity increase 02 36 3 Annual limit on water connections 41 55 4 Mean number of these policies used per city 0.34 0.42 0.55 0.52 0.46 Notes: Percentages represent the number of respondents indicating use of the policy as a proportion of all those providing an answer to the question, including “Don’t know” or “N/A” responses. For more detailed tabulation, see Lewis and Neiman (2000, pp. 7–8). 31 infrastructure policies to create yearly restrictions on housing development. The other two types of infrastructure policies are more common, although still used by only a small minority of communities. About one city in seven uses capital improvements to shape residential development, whereas just over one-quarter of cities require that traffic standards be satisfied before new development occurs. The latter action belongs to a class of policies often termed “adequate public facilities ordinances,” in which localities attempt to legislate the desired levels of service that must be attained in the community before new development can be approved. Since traffic problems are one of the most visible and irritating examples of public facilities that fall below standards, it is not surprising that a fair number of cities single these out for scrutiny in the development process. In the case of these infrastructure-related policies, the differences among regions are more muted than in the case of the overt restrictions. It is worth noting, however, that Central Valley cities are among the most likely to use three of the policies (all except traffic standards). This may indicate a higher perceived level of infrastructure limitations in that region, which grew at the most rapid rate during the 1990s. It may also reflect difficulties in paying for infrastructure, given that the Central Valley is the poorest of the regions. Finally, it might also reflect Central Valley concerns regarding the loss of agricultural land to urban uses, where the “premature” extension of development infrastructure might stimulate the spread of urban pressures onto agricultural land. Shaping Housing Development The final category, for policies that “shape” growth, is something of a residual category for policies that are not explicitly infrastructurerelated and that do not embrace overt restrictions on housing development. Included are the following five types of city actions: • Ranking proposed residential projects through a point system, • A requirement that residential developments include “affordable” housing, 32 • A policy that restricts new residential growth to areas that are already developed, • A policy that encourages residential growth to occur in alreadydeveloped areas, and • Design review standards. These policies fall in a class that generally seems more benign than overt restrictions. Nevertheless, each arguably may add to the cost of housing or limit production. For example, policies that target new development to already-developed areas have the goal of increasing infill and reducing sprawl. (This class of policies includes growth boundaries and urban limit lines.) But by limiting the array of land choices available to developers, and focusing on properties that may be smaller or more difficult or expensive to develop, these policies can result in more expensive housing. Policies that require affordable housing as part of larger developments do take heed of the needs of low-income households. But some argue that affordable housing quotas may shift costs to other homebuyers; that is, lower-priced units may have their costs subsidized by the market-rate units, which are then priced higher. There is a great deal of controversy over this argument, with some analysts concluding that under many market conditions, the costs will come from developer or landowner profits (Calavita and Grimes, 1998, p. 152). Even design review standards, which are geared toward improving the aesthetic appearance of the community, can easily increase housing costs, since they may result in the creation of a quality “floor” of materials or standards that builders must meet. Efforts to create stylistic unity in architectural schemes in local communities may result in a smaller array of choices of housing types available to consumers. Design review may encompass building setback requirements, which can affect the number of units that can feasibly be included in the project. This is not meant as a criticism of these policies, but as a way of setting them within the general category of local government regulatory activity affecting housing. 33 Table 3.3 indicates that only one of these policies, design review, is in very widespread use, with 83 percent of cities employing it.4 By contrast, the most restrictive policies (point systems and restriction of development to built-up areas) are each used by only about one in 20 cities. The “carrot” approach to urban growth boundaries—encouraging (rather than requiring) builders to direct housing to built-up areas—is used by 28 percent of responding cities. A slightly higher percentage embrace affordable housing set-asides. As in the case of the overt restrictions, Bay Area cities show a higher level of regulatory activity to “shape” growth, for all five activities. Design review standards are particularly ubiquitous, with only two out of Table 3.3 Use of Policies to Shape Residential Growth Policy Rank proposed residential projects Projects must include affordable housing Restrict growth to built-up areas only Encourage growth in built-up areas only Design review standards Mean number of these policies used per city Percentage of Cities Using Policy L. A. Other So. S. F. Bay Central County California Area Valley Total (N = 57) (N = 91) (N = 77) (N = 73) (N = 298) 44 6 45 25 26 49 21 31 5 2 14 4 6 16 21 75 82 39 36 28 97 73 83 1.25 1.34 2.07 1.37 1.52 Notes: Percentages represent the number of respondents indicating use of the policy as a proportion of all those providing an answer to the question, including “Don’t know” or “N/A” responses. For more detailed tabulation, see Lewis and Neiman (2000, pp. 9–10). ____________ 4The prevalence of design review might disguise the fact that the stringency of these reviews can vary extensively. Unfortunately, our study did not measure the full array of topics that might be encompassed in design review, ranging from paint colors to roofing materials and appearance. Some of these design review items might result in substantial delays and increases in housing costs. 34 77 respondents from that region (less than 3 percent) indicating that their cities do not engage in design review. Bay Area cities are also significantly more likely to have affordable housing set-aside requirements, with about half of the municipalities in that region doing so, compared with about one-quarter of cities in Los Angeles County and the rest of Southern California, and about one-fifth in the Central Valley. Although it is not possible to ascertain at this time whether such policies actually can or will increase the supply of affordable housing, it is reasonable to assume that such policies reflect concern about the worsening problems of affordability in the Bay Area in recent years. The regions outside the Bay Area generally are fairly similar to one another in the degree of adoption of these policies, except that Central Valley cities are much more likely to encourage infill than their counterparts in Southern California. The notable rate of respondents indicating that their cities encourage growth in built-up areas might be additional indirect evidence of the Central Valley's increasing concern about how the spread of urbanization affects the maintenance of agriculture in the region. Overall Levels of Growth-Management Activity We have seen thus far that each of the growth-management policies has been adopted by only a minority of cities, with the exception of one relatively uncontroversial policy, design review standards. More important, perhaps, is the overall effect of such policies upon builders operating in California jurisdictions. Among all the regions surveyed, the average city has adopted about 0.7 overt restrictions, 0.5 policies to link growth to infrastructure, and 1.5 policies to shape new housing development, for a total of about 2.7 policies out of a possible 16. Figure 3.1 shows the entire distribution of the number of growth-management policies adopted. Of the 297 respondents, only 24 (or 8 percent) adopted six or more policies, whereas 18 (6 percent) adopted none. To get a sense of which types of cities might be more likely to engage in such activities, Table 3.4 shows demographic and other characteristics for these “heavy adopters” and “nonadopters” of growth management. As the table indicates, there 35 Number of cities 90 80 70 60 50 40 30 20 10 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Number of policies adopted Figure 3.1—Frequency Distribution of the Number of Residential Growth-Management Policies Adopted Table 3.4 Comparing “Nonadopters” of Growth-Management Policies to “Heavy Adopters” City Characteristic Per capita income, 1990, $ Residents aged 25+ who are college graduates, 1990, % Owner-occupancy of housing, 2000, % Average unemployment rate, 1990a, % Population non-Hispanic white, 2000a, % Population growth rate in city, 1990–1998, % Population growth rate in county, 1990–1998, % City incorporation datea Democrats, of major-party registered voters, 1999, % Average for Cities with No Policies 17,615 17.7 61.7 8.7 49.5 14.9 14.3 1939 54.5 Average for Cities with > 5 Policies 17,994 23.4 64.4 5.5 65.1 14.9 11.9 1915 53.0 Sources: 2000 Census, 1990 Census, California Department of Finance Demographic Research Unit, California Secretary of State, author calculations. aStatistically significant difference in this characteristic (p < 0.05). 36 are some social differences between these extreme ends of the spectrum, with heavy adopters tending to have significantly higher proportions of non-Hispanic whites in the population and higher proportions of college graduates. Heavy-adopting cities also tend to be older than nonadopters (i.e., they were incorporated earlier). Nevertheless, some characteristics that one might assume would be indicators of slow-growth policy adoption, such as high income and homeownership levels, or rapid growth rates, are not very different at all between the heavy adopters and nonadopters. Clearly, a more careful analysis of why cities adopt growthmanagement policies is warranted, and we will provide this in Chapter 5. For now, it is worth noting once again the major differences in cities’ policy adoption across the three regions. As Figure 3.2 shows, cities in the Bay Area average about 3.5 growth-management policies, compared to 2.5 in the Central Valley, 2.1 in Los Angeles County, and 2.4 elsewhere in Southern California. Thus, studies of slow-growth policies that draw upon case studies from the Bay Area, such as Frieden (1979), are probably unrepresentative of growth policies in most of California. 4.0 3.5 Shape housing development Link to infrastructure 3.0 Restrict residential growth Number of policies 2.5 2.0 1.5 1.0 0.5 0 Los Angeles Other Southern San Francisco Central County California Bay Area Valley Figure 3.2—Average Number of Growth-Management Policies, by Region 37 The Bay Area has a somewhat unique set of attributes that may make it more self-conscious regarding issues of regional growth and quality of life (see also Baldassare, 1994) and thus more likely to be amenable to growth management policies. These attributes include environmental features such as the dominant presence of the Bay itself, the ubiquity of ridge lines and steep slopes that make development potentially difficult or unsafe, the proximity of high-value viticulture in the wine country of Napa and Sonoma Counties, and a focus on preserving the centrality and historic amenities of the city of San Francisco. Also notable are the particularly high levels of income, education, and traffic congestion by comparison to other U.S. metropolitan areas (the latter condition compounded by the area’s unusual topography) and the regional economy’s relatively strong reliance on amenity-oriented tourism. In addition, it may well be the case that the early adoption of growthmanagement policies in the Bay Area, and the more liberal and environmentalist ideology characteristic of residents of that region, led to a diffusion of policy adoptions among cities there. Also notable is the slightly higher average number of residential regulations in the Central Valley compared to Southern California. This is the opposite result of what one might expect, given the more rural nature of many Central Valley communities, the more recent spread of urbanization there, and its somewhat more conservative political culture. Infrastructure limitations and the desire to protect high-value farmland from development may help account for the Central Valley’s growthmanagement policies. Summing Up In reviewing the patterns of adoption of growth-management policies in the major regions of California, perhaps the most notable fact is the small number of such adoptions, with the average city in our dataset having only about 2.7 out of the 16 policies. This finding seems consistent with an earlier survey of growth-management policy adoption by Glickfeld and Levine (1992), in which the average California city they surveyed (in 1988) had enacted 1.9 out of a possible total of 14 policies. It appears that some observers have tended to generalize about local 38 growth management based on the actions of high-profile but unrepresentative communities. In our survey, cities were most likely to have adopted policies to shape residential development, such as design review requirements or affordable housing set-asides. Policies designed to overtly restrict the total amount of housing developed in the city, or to condition housing construction on infrastructure capacity, were less common. Insofar as there were “heavy” adopters of growth-management policies (which we defined as six or more policies out of 16), they tended to have populations with a smaller proportion of minorities and a higher share of college graduates than cities adopting no policies. In other aspects, however, such as population growth rates, per capita income, and political party registration, these two groups of communities were remarkably similar. Regional differences, however, are readily apparent, with cities in the San Francisco Bay Area enacting policies more frequently than those in the Central Valley or in Southern California. This pattern was particularly true of policies to restrict the development of housing, or to shape the type of housing that was developed. Nevertheless, even Bay Area cities tended to enact only a few of the 16 policies we asked about. 39 4. Political Controversy and Growth Management In this chapter, we begin to probe more closely for the local factors that may influence whether cities will adopt growth-management policies. In particular, we are interested in the local politics of residential growth. Frieden (1979), using case studies, portrayed local governments—alongside homeowners, neighborhood groups, and environmentalists—as active and often very willing participants in the movement to restrain housing development in California. Is this perspective borne out by our survey data from cities throughout the three regions? Are local officials leaders in the movement to restrict or manage growth? Are they willing followers of public opinion? Or are they more progrowth in orientation and only loosely bound by local antigrowth sentiment? Is there evidence for elitist and exclusionary motivations for the use of residential development controls? In examining the data presented in the following pages, we conclude that a major factor influencing the adoption of residential development controls is the eruption of controversy and conflict over growth. It is possible, we contend, that once the conventional patterns of local residential development politics are disrupted, subsequent policies might contain more restrictive provisions. Is Residential Growth Always Controversial? Although growth issues are sometimes portrayed as a source of constant controversy in contemporary California, this may be a reflection of media coverage. That is, it may be the case that most residential development is a relatively quiet and routine process, but the topic only becomes “newsworthy” when a specific controversy with overt conflicts develops. We asked the planners responding to our mail survey, “How controversial would you say residential growth issues are in your city?” 41 As Table 4.1 indicates, controversy appears to be somewhat sporadic in many cities, with 52 percent stating that residential growth issues were “sometimes” controversial. A fifth of respondents said that such issues simply were not controversial, whereas 28 percent found them often or always controversial. Once again, there are very clear regional distinctions in the data. Fully half of the respondents from the Bay Area (the setting for Frieden’s case studies) call residential growth often or always controversial, compared to about one-quarter in cities in Los Angeles County and the rest of Southern California, and about one-eighth of Central Valley respondents. Similar findings emerge in response to a question about the influence of residential development issues on city elections. As Table 4.2 shows, 39 percent of respondents indicate that growth issues hardly ever affect council or mayoral elections, 31 percent say that election outcomes have been affected a few times, and 22 percent feel that growth issues have often been influential in affecting elections. Again, growth does not appear to be a burning issue in many local elections, although it is clearly a prominent issue at some times in some places. As for which places, the table illustrates the regional pattern; 41 percent of Bay Area respondents said that local growth issues have been influential in affecting local elections, as compared to 14 percent in the Central Valley, 11 percent in Los Angeles County, and 18 percent in the other counties of the Southland. Table 4.1 Prevalence of Controversy Regarding Residential Growth Issues “In considering the past and more recent periods, how controversial would you say residential growth issues are in your city?” L. A. County Other So. S. F. Bay Central Cities California Area Valley Total Response (N = 55) (N = 91) (N = 76) (N = 72) (N = 294) Not at all controversial 22% 22% 4% 28% 19% Sometimes controversial 55 51 43 60 52 Often controversial 18 21 25 8 18 Almost always controversial 4 7 25 4 10 Don’t know 2 0 30 1 42 Table 4.2 Effects of Growth Issues on Local Elections “Which of the following do you believe best describes the influence of residential development issues on your city’s elections?” L. A. County Other So. S. F. Bay Central Cities California Area Valley Total Response (N = 56) (N = 92) (N = 76) (N = 72) (N = 296) Growth issues hardly ever affect council or mayoral elections 50% 37% 21% 51% 39% There have been a few times when growth issues have affected the outcomes of local council or mayoral elections 29 33 34 28 31 Local growth issues have often been influential in affecting the outcomes of local council or mayoral elections 11 18 41 14 22 Don’t know 11 12 4 7 8 The next question is whether such controversy accelerates the adoption of growth-management policies. Indeed, controversy levels do appear to be related to the overall adoption of growth-management policies. Recall from the last chapter that the average city in our sample had adopted 2.7 growth-management policies out of the 16 mentioned in the survey. As Figure 4.1 shows, there is a distinct connection between levels of political controversy and the adoption of growthmanagement policies. In cities where respondents report growth issues to be almost always controversial, the average number of adopted policies is 4.1, compared to 2.0 policies in cities reporting no controversy.1 Thus, local growth controversies appear to be a precipitating factor in the adoption of growth management policies. This conclusion is ____________ 1The simple correlation between respondents’ answers to the “general controversy” question and the total number of growth management policies is a strong 0.34. 43 5 4 Number of policies 3 2 1 0 Almost always Often Sometimes Not at all controversial controversial controversial controversial Figure 4.1—Number of Growth-Management Policies, by Level of Controversy supported by an earlier study of growth controls among Southern California municipalities (Donovan and Neiman, 1992). In a related finding, Landis et al. (1995, p. xxiii) conclude that localities with greater levels of growth controversy tend to require Environmental Impact Reports for a higher percentage of proposed projects. What Role Do Local Citizen Initiatives Play in Growth Management? The adoption of growth-limiting or growth-managing policies tends to receive most attention from journalists and scholars in cases where the voting public itself enacts such measures, through so-called ballot-box planning. Judging by our survey evidence, however, citizen initiatives do not appear to be a direct source of many residential policies in the large majority of cities (Table 4.3). Fewer than one in six respondents indicated that initiatives have been a major source of slow-growth policies in their city; fewer than one in eight felt that a future initiative to slow growth would be likely in their city. Once again, these responses 44 Table 4.3 Prevalence of Citizen Antigrowth Initiatives “In your city, have initiatives been a major source of policies to slow residential development? Don’t include referenda placed on the ballot by the council.” L. A. County Other So. S. F. Bay Central Cities California Area Valley Total (N = 52) (N = 87) (N = 76) (N = 72) (N = 287) Yes 10% 15% 33% 4% 16% No 90 85 67 96 84 “Is there a good chance that an initiative measure to slow residential development will occur in your city?” L. A. County Other So. S. F. Bay Central Cities California Area Valley Total (N = 50) (N = 79) (N = 75) (N = 67) (N = 271) Yes 4% 10% 24% 7% 12% No 96 90 76 93 88 were quite different in the Bay Area, which showed by far the highest likelihood that initiatives are playing or will play a major role. In the Central Valley, by contrast, antigrowth initiatives have been quite rare. It is also worth noting that respondents in cities that have had citizen slow-growth initiatives in the past are more likely to expect additional ones in the future. Among respondents answering both questions, 19 of the 40 who indicated past experience with citizen initiatives said that another initiative was likely in the future.2 In part, this finding indicates that growth issues occupy a persistent place on the political agenda of a handful of municipalities in the state. Another consideration is that some initiative measures have language requiring future citizen votes on proposed projects of certain types, such as residential projects greater ____________ 2Among the 297 respondents, ten failed to answer the question about past initiatives and 26 did not answer the question about future initiatives. The drop-off in responses for the latter question can probably be attributed to an unwillingness to speculate about the future. 45 Average number of policies than a certain number of units or projects that require zoning changes. In this way, one initiative can beget future initiatives. Our next step was to examine whether the number of growthmanagement policies is noticeably higher in cities that have experienced, or expect to experience, citizen initiatives. Figure 4.2 shows that this is indeed the case. The average number of growth management policies is 3.9 in cities that have had initiatives compared to 2.5 in those that have not. To some degree, however, this finding is tautological, since cities indicating that initiatives “have been a major source of policies to slow residential development” will, by definition, have such policies. Perhaps more interesting, then, is the pattern in regard to expected future citizen initiatives. In those cities whose respondents expected such an initiative, the average number of policies was 4.2, compared to 2.5 in cities that do not anticipate an initiative.3 The significantly higher number of policies in cities having had or expecting initiatives is also evident within each of 5 Yes 4 No 3 2 1 0 Had citizen growth initiative? Expect citizen growth initiative? Initiative status Figure 4.2—Citizen Initiatives and City Growth-Management Policies ____________ 3This positive and highly significant relationship between the respondent’s expectation that an antigrowth initiative is likely and the number of growth-management policies is apparent even after controlling for past initiatives. 46 the three categories of residential growth policies (policies to restrict, link, or shape). The relationship between expected future initiatives and the number of policies may indicate that city governments pass residential policies in anticipation of citizen mobilization on the growth issue. That is, local governments may engage in growth management in an effort to head off or co-opt citizen antigrowth movements.4 Citizen initiatives are often more restrictive and inflexible in their requirements than policies passed by city councils, and so city policymakers may attempt to craft more flexible or generous policies that assuage residents’ demands for action on the growth issue. Of course, this presumes that city councils are, in fact, more progrowth in orientation—more accommodating to new housing—than the local antigrowth activists in such communities. What Is the Attitude of City Councils Toward Residential Growth? Do local policymakers share in the discontent over residential growth, or are they more interested in accommodating additional housing? To help answer this question, we asked the planners responding to our survey to describe “the general attitude of the majority of your city council toward residential growth.” As Figure 4.3 indicates, the most common response was that the council encourages residential development, and the next most common reply was that the council is “mostly neutral.” Only 13 percent of the 294 responses to this question indicated that the city council “occasionally slows the rate of residential growth when growth issues become controversial,” and 6 percent said the council “generally tries to slow growth and often proposes limitations on residential development.” (Four percent of respondents chose the “don’t know” response.) Because most residential policies are set by city councils, these responses are important barometers of local orientations toward growth. ____________ 4Similarly, growth-control proponents may pursue initiatives even when their chances for winning the initiative election are not good. In the 1980s, “San Francisco, for example, [was] host to a string of failed initiatives. Yet proponents of those initiatives saw these measures as “pushing” the local government to enact its own growth controls” (Glickfeld, Graymer, and Morrison, 1987, p. 136). 47 Percentage of cities 100 90 Neutral 80 Encouraging 70 60 50 40 30 20 10 0 Los Angeles Other Southern San Francisco Central County California Bay Area Valley Figure 4.3—Reported Attitude of City Council Majority Toward Residential Growth In fact, when we asked what the respondents’ perceptions were of the origins of their cities’ policies that affect residential growth rates, 58 percent said that such policies were “enacted by the council, without much neighborhood pressure.” Only 1 percent of the respondents said that these policies were “enacted pretty much exclusively as a result of the initiative process.”5 Thus, most city councils in these regions can be thought of as either generally supportive of new housing or at least neutral. This is true even in the San Francisco Bay Area, where 42 percent of councils are seen as encouraging and 25 percent as neutral. Indeed, Bay Area councils are actually seen as much more likely to encourage housing than city councils in Los Angeles County, where the respondents place most of their councils into the “neutral” category. Despite the high levels of activism and controversy in the Bay Area, city councils may be more ____________ 5The other results were that 25 percent saw such policies as “enacted by the council, but mainly as a result of neighborhood pressure,” whereas 15 percent concluded they were “enacted through city council action and through the initiative process.” 48 sensitive to the increasingly severe housing problems facing the region. City officials, consequently, might be struggling with or recognizing their responsibilities to accommodate residential development in the region. Many politically active residents of cities in the Bay Area, however, are apparently less supportive. Summing Up The primary finding of this chapter is that local controversy over growth, and the reality or expectation of local citizen initiatives relating to growth, are strongly associated with the presence of a greater number of growth-management policies. Thus, growth-management policies appear to emerge in large part as a result of the dynamics of local politics and residents’ activism, rather than through dispassionate analysis of local or regional housing trends or environmental conditions. Nevertheless, only a minority of cities seem to be regularly rocked by residential growth conflicts, and the vast majority have had little experience with slow-growth citizen initiatives. In short, pitched battles over residential development seem to be limited to a relatively few localities, whereas growth is a more routine process in most communities. City councils, which play the strongest role in setting policies that affect residential growth rates, typically are viewed as progrowth or neutral in each of the regions studied. Interpreting the Evidence Thus Far The results above lead us to propose the following scenario regarding the emergence of local growth management: Occasionally, local conflicts erupt over residential proposals. These may occur because of deteriorating local conditions that are blamed on growth, or rapid changes in community character, or the emergence of an antigrowth “political entrepreneur” in local politics (Schneider and Teske, 1993). If city officials do not act sufficiently rapidly or strongly to assuage resident concerns or persuade residents that the city has taken steps to accommodate the new housing gracefully, the result may be sustained controversy over residential growth—perhaps enough to affect local 49 mayoral or council elections. The result is perhaps the formation of local groups favoring slower, managed, “smart,” or no growth at all. Although most cities do not reach this stage of political turmoil over development, those with an aroused and dissatisfied populace are likely to see some growth-management measures passed, generally by the council but in a smaller number of cases by citizen initiative. At this point, the growth issue may go into hibernation in some communities. In others, the passage of growth-management measures may in turn create further controversy—either by alienating progrowth interests among the citizenry, or by setting unrealistic expectations that growth will slow down, or by requiring that future decisions about residential proposals be subject to popular vote. If growth controversies occur in a setting where local officials are decidedly progrowth, then even with a short-term victory, citizens concerned with slowing or stopping growth are likely to persist because they distrust their officials. Moreover, local battles, when they result in restrictive development actions, also are likely to produce the actuality or threat of litigation. Finally, in some cases, the threat to slow or stop growth might actually stimulate the programming of considerable future growth, by causing developers to rush the processing of permits before limits take effect. As a result, when growth controversies become heated, they are likely to stay that way for a considerable period. If the foregoing account is true, it evokes an image of grassroots citizen activism leading the antigrowth movement in California, with city governments acting more as followers—and perhaps often, reluctant followers. Although most city councils appear to feel a responsibility to accommodate a reasonable share of their region’s residential growth pressures, it is politically perilous for them act unilaterally and discount residents’ fears and concerns about growth. In the next chapter, we will consider a wider variety of local characteristics that may help explain why some cities are more prone than others to growth disputes and growthlimiting policies. 50 5. Factors Shaping Cities’ Willingness to Accept Residential Growth What factors produce conditions that are particularly conducive to the passage of local residential growth controls? In this chapter, we summarize the results of a series of statistical analyses designed to answer this question. The chapter begins with a discussion of various city characteristics that might possibly influence local residential policies and some of the data analysis issues involved. The analysis then proceeds in three parts: First, we examine explanations for various housing policy choices of cities, including the number of growth-management policies and local requirements for affordable housing set-asides. Second, we look at city governments’ overall orientation toward housing development—that is, how enthusiastic or wary officials are reported to be regarding residential growth in their cities. The third topic explored is local political controversy over residential development, which is a major determinant of growth-management policy adoption. This chapter is meant to point out the most interesting and statistically reliable results of our various multivariate models. The appendix of this report provides results of the models and further information on variables, data sources, and methodology. Considering Potential Explanatory Factors Local Political Factors One set of potentially important factors for explaining city actions is political. We have already seen the important relationship between citizen controversy over growth and the number of growth-management policies pursued. Other political factors of interest might include the ideological or partisan leanings of the local populace, the number of 51 planning staff available to develop and enforce growth rules and ordinances, and the fiscal climate of the city. The strength of local interest groups concerned with growth, the method for electing city officials, and the presence of competing important issues on the local political agenda might also be of importance. Although we could not locate suitable measures for all potential political and institutional variables, we rely upon the following variables to measure political characteristics in each city: • A measure of citizen opposition to growth (to be discussed below), • The percentage of major-party registered voters who are Democrats, • The number of planning staff per 1,000 local residents, • The dollar amount of own-source city government revenues, per capita, and • The estimated percentage share of all property taxes paid by city residents that go toward the city government—as opposed to the county, school district, or special districts.1 Some of the statistical models also include some ancillary measures (where such data seem important to the question at hand): • The importance of neighborhood groups and business groups in local development policymaking, as judged by the city manager, • Whether city council members are elected by districts or at-large, • How active the city is in redevelopment policy (according to the city manager), and • The crime rate of the city—an issue that may compete for time and attention from local policymakers. How would each characteristic affect policy? One might expect cities with higher levels of citizen opposition to growth to have city governments that do more to limit residential development. The ____________ 1This is sometimes referred to as the city’s “split” of the property tax. Some observers of local government in California argue that cities with a low share of the property tax may be more inclined to engage in the “fiscalization of land use,” seeking out development projects that provide more sales tax or fee revenues. 52 influence of Democratic party affiliation is less clear, since labor and minority groups (often pro-housing development) and environmentalists (sometimes antigrowth) are all important parts of the Democratic coalition. Cities with high levels of local revenue, or a larger percentage share of the property tax, might possibly be more willing to accept housing, which is often seen as fiscally costly. Diaz and Green (2001) found, however, that municipalities in Wisconsin with greater fiscal capacity engaged in more growth-management efforts. It is unclear whether an increase in planning staff resources would increase cities’ ability to regulate growth or lead to more support for housing development. Planning staffs might provide the institutional capacity to draft local development restrictions, but the presence of larger number of planners might also contribute to a more balanced, professional norm of including reasonable levels of housing or concern for the potential sprawl-inducing effects of excluding housing. Local business groups, such as the Chamber of Commerce, might be expected to approve of local population growth and more housing, whereas neighborhood groups are often reputed to oppose such growth. Districtbased elections of the city council could lead to an emphasis on “not in my backyard” politics, and thus less enthusiasm for residential development. Finally, cities with high crime rates might be less likely to become convulsed in controversy over residential issues, since the crime issue may tend to dominate the local policy scene or high-crime communities may also feel that residential growth is the economic stimulus that might reverse their crisis conditions. Demographic Makeup of the City We are also interested in the demographic background and socioeconomic status of city residents because local residential policy may possibly respond to or reflect differences in the wealth or income, status, and racial or ethnic background of the population. One problem with using such measures to characterize California cities is that many are closely related, which makes it difficult to distinguish the independent effects of each variable. For example, cities in California with higher percentages of Hispanics in the population also tend to have higher unemployment rates and more children in the population. Cities with 53 high household incomes tend to have high proportions of homeowners and low unemployment rates, and so on.2 To resolve this difficulty, we rely on one measure of status and two measures of ethnic background that are often associated with socioeconomic status. Each seems potentially quite important in the formation of local development policy, but the variables have the advantage of not being closely correlated to one another. Each is measured using 2000 Census data for the cities in our sample: • The percentage of housing units that are owner-occupied, • The percentage of the city’s population that is Hispanic, and • The percentage of the population that is African-American, or black.3 Here, the expectation is that communities with higher levels of owner-occupancy might have more exclusionary, status-based motivations for local housing policy (Danielson, 1976). The influence of local Hispanic and black populations is less clear, although minority groups in the United States are often thought to be less likely to embrace antigrowth political movements (Neiman and Loveridge, 1981; Carman, 1998; DeLeon, 1992). Cities with higher minority group populations may also be more likely to support requirements for affordable housing or be more concerned with the overall need for economic development, including the anticipated stimulus of residential construction. On the other hand, it is also possible that officials in cities with higher concentrations of minority group members may emphasize market-rate housing or even pursue limitations on multifamily development, in an effort to avoid becoming a segregated, low-income minority community. ____________ 2Each of the above relationships has a correlation of 0.6 or greater across the cities responding to the survey. To avoid multicollinearity, the variables selected for inclusion in the model all have correlations below 0.5. 3In the 2000 Census, respondents were able to identify with more than one race; moreover, Hispanic is not a racial category. However, in an effort to create nonoverlapping categories for this statistical analysis, we calculated for each city the percentage Hispanic (of any race) and the percentage non-Hispanic black (who did not also identify with another race). 54 City Size, Urbanization, and Location Larger communities, and central cities in particular, may be somewhat distinctive in their orientations toward growth (Lewis, 2001; Neiman, Andronovich, and Fernandez, 2000). Larger communities tend to have a greater number of active political groups, and larger city governments, which may lead to the development of more policies. Diaz and Green (2001) find that in Wisconsin, municipal population is positively associated with the adoption of growth management tools. On the other hand, progrowth organizations, such as downtown businesses, construction unions, and media organizations, are often very politically active in central cities and other large communities, which may represent a countervailing force against regulating growth. Furthermore, central city political leaders may be more interested in capturing growth, since they are likely to be held somewhat responsible for local economic vitality, particularly if they are surrounded by growing, competitive suburban communities. Smaller rural and suburban communities may face distinctive types of growth challenges. Regional differences may also help account for city policies. As we have seen, the Bay Area appears to have much higher levels of growth management than other parts of the state. It remains to be seen whether its distinctiveness remains after we take account statistically of its major differences from the rest of the state (for example, in socioeconomic status and Democratic party affiliation). Some may anticipate that the fast-growing, largely suburban regions of Southern California outside of Los Angeles County may have a distinctive growth politics as well. We thus include the following variables in this category: • The local population size, • An indication of whether the city is a central city, a suburb, or a rural community,4 • An indication of whether the city is in the San Francisco Bay Area, and ____________ 4We rely on Census Bureau designations of central cities. Municipalities that are urbanized and located within metropolitan areas, but are not central cities, are counted as suburbs. This is the most numerous category in California. The remainder are classified as rural. See Lewis (2001) for further discussion of this typology. 55 • An indication of whether the city is in Southern California but outside Los Angeles County.5 Local Conditions Relating to Growth The willingness of city residents and elected officials to accommodate housing may be shaped powerfully by the capability of the city’s infrastructure to handle new growth and other conditions in the city (or its surrounding area) that might be thought of as the outcomes of previous growth. City officials are likely to ask themselves, “What kind of city is this? How are we geared to handle new housing, and what might be its costs to our existing character?” For example, a city rich in employment and with abundant, uncrowded transportation infrastructure may be far more willing to accept multifamily development than a city with an excess of housing units and long commutes to jobs in the local area. Cities with problems of housing affordability may be more accepting of new housing development. Regions that are growing quickly may create more pressure to accommodate growth locally, but such growth may also arouse more citizen controversy. And communities with high degrees of transiency, or population turnover, may have different reactions to housing proposals than more settled communities, or than resort communities with many part-time residents. Accordingly, we draw upon the following variables in explaining local policies: • Recent population growth rates: For questions relating to city government policies, we use the county growth rate (between 1990–1998), since city policymakers likely consider local policies in the light of growth pressures in the surrounding region.6 For other questions, such as those relating to citizen controversy, we use the city’s population growth rate over the same period. ____________ 5This includes cities in San Diego, Orange, Riverside, San Bernardino, Ventura, and Santa Barbara Counties. 6Glickfeld and Levine (1992) argue that local growth controls may in large part be a reaction to regional growth. 56 • The percentage of residents in 1990 who had lived in the same home five years earlier: a measure of residential stability, from the Census.7 • The average one-way commuting time among city residents, which may measure local roadway conditions as well as distance from homes to work. • The percentage of local housing units not connected to a modern sewer: Where developers must rely on septic systems rather than a public sewer system, it will be difficult if not impossible to build housing at high densities. This factor may therefore make city policymakers warier about adding housing. But at the same time, it may make overt local growth controls unnecessary, insofar as the lack of infrastructure itself helps stop growth. • The ratio of jobs within the city to workers who live in the city. • The percentage of local housing units that are seasonal or recreational: an indication of whether the city may have environmental amenities important to its character. (This information is available from the 2000 Census.) • The affordability of the local housing stock as of 1990: This is measured as the ratio of the median house value to the median income. This variable is used only for one statistical model relating to local affordable housing policy. Controversy and Growth Management: A TwoWay Street As we have seen, the level of local citizen controversy over growth is powerfully related to the number of growth-management policies. However, it is also realistic to expect that the number of local policies ____________ 7Unfortunately, for some variables, the 1990 Census is the most recent comparable data available at the time of this writing. (Thus far, city-level data from the 2000 Census are available only for racial, ethnic, and age groups, as well as household arrangements and homeownership.) However, since many residential policies and growth controls date back quite a few years, this reliance on 1990 data need not be a disabling data problem. Moreover, we think of the accumulated set of local policies in any given community as in some important way a response to prior community conditions. 57 may affect the level of controversy. This expectation is borne out by the fact that a tally of local growth-management policies as of 1988 (taken from Glickfeld and Levine, 1992)—a decade before our survey—is a significant predictor of controversy levels at the time of our survey (see the appendix for details). In short, it appears that causality runs both ways: Controversy promotes the adoption of growth-management policies, but controversy also grows as more policies are adopted—perhaps because the passage of policies motivates opposition or elevates the growth issue on the public agenda. Moreover, as controversy persists, the various organizations that emerge around the issue of growth become more sustained, even institutionalized perhaps, thereby moving controversy to more intense levels. This relationship is interesting, but it also creates statistical problems because the variables may simultaneously determine one another. Although there are various ways to address this issue, we have chosen to avoid the problem in the data analyses by substituting another measure of local citizen behavior for our controversy measure. The measure we use instead, also derived from the planning director survey, asked respondents to assess the importance of “citizen opposition to growth” in constraining residential development in their cities. In this case, the causality appears more straightforward. Citizen opposition to growth should generate more growth-control policies, but it seems unlikely that the number of policies itself would increase levels of citizen opposition to growth. Explaining Housing and Growth-Control Policy Outcomes We use multivariate statistical techniques that are designed to identify the effect of each city characteristic on local policies, while holding constant the other characteristics. For example, if we find that high homeownership rates are associated with a greater number of city growth-management policies, this indicates that homeownership has this relationship with housing policy independent of the racial makeup of the city, its population size, and so on. (See the appendix for more details on methodology.) 58 We first examine the sum of local growth-management policies—the total number of policies that restrict or shape growth or link housing development to infrastructure requirements. As noted in the last chapter, this total may range from 0 to 16. Based on the multivariate results, several factors appear to be significantly related to the number of growthmanagement policies adopted: • Not surprisingly, citizen opposition to growth is associated with a larger number of policies. • Cities with higher levels of homeownership tend to have a greater number of policies, but so do communities with higher shares of Hispanics in the local population. • Cities with a more stable population locally (less transient populations from 1985 to 1990) and slower population growth at the county level tend to have a greater number of policies. • Cities with higher percentages of unsewered housing units tend to have fewer policies. • Central cities and rural communities tend to have more growth management policies than suburbs. Both the Bay Area and the portions of Southern California outside Los Angeles County have a higher incidence of growth management, even when controlling for city characteristics. There are a number of interesting aspects to these results. First, although demographic factors are hardly the only explanation for growth management, upper-status communities (judging by the homeownership rate) do appear to engage in more of such policies, consistent with a social-status or exclusionary explanation. At the same time, however, communities with a higher proportion of Hispanics also have more growth-management policies. As Hispanics are the fastest growing segment of the California population, it is possible that governments of cities with high proportions of Latinos fear becoming overwhelmed by growth, although that is only speculation. The main point is that one cannot assume that upper-income white communities are the main bastions of growth control. 59 Second, mere population pressure itself does not generate such policies. If anything, population growth and turnover appear to create conditions that are somewhat less amenable to the passage of growth controls. This finding is consistent with Pendall’s (1999) work on housing protest in the Bay Area, which points out that cities with more recent population influxes tended to experience fewer citizen protests against new housing developments. Pendall attributes this result to the inability of newer and transient populations to effectively mobilize within local political systems, whereas residents of more settled, slow-growing communities may have greater political resources. Moreover, such cities may have an established pace and rhythm, along with longer-settled residents who recoil at the prospect of rapid growth and react by supporting development controls. The fact that a lack of a modern sewerage system decreases the number of policies is also revealing. It indicates that this infrastructure deficiency itself may lessen the pressure of rapid housing growth, since dense developments are not generally possible without public sewerage facilities. What we cannot determine from this is whether city leaders are consciously substituting a lack of sewerage facilities for growth-control policies, or whether the infrastructure deficiency occurs for other reasons and thus simply functions as a technological barrier to growth. Respondents’ answers to a number of other survey questions help cast light on city residential policies. In one case, they were asked to assess the importance of “density restrictions on residential land” (on a scale of 1 to 5) in constraining or slowing residential development. Their responses indirectly help us assess the role of zoning restrictions and subdivision requirements that may limit the density of housing built in cities. Factors that are significantly related to a high score on this question—meaning that density restrictions are important in slowing growth—include the following: • The most important correlate of density restrictions appears to be citizen opposition to growth. But citizen opposition is not solely a feature of high-status communities. In this case, cities 60 with high homeownership levels are no more likely to view density restrictions as important.8 • Findings on population pressures are analogous to those reported above. Cities with low population turnover and low levels of countywide population growth are more likely to find density restrictions important. This is again consistent with Pendall’s (1999) arguments about housing protest. • Central city respondents were significantly less likely to view density restrictions as important in constraining housing. In a related question, planning directors were asked to characterize their cities’ review process for residential development, in comparison to that of other cities in their area. In their answers, they labeled their own city as either less strict, equally strict, somewhat more strict, or much more strict than nearby communities. The following factors help predict the stringency of local residential review processes: • Cities with higher levels of citizen opposition to growth, and more homeownership, tend to be stricter. • Central cities and municipalities with more planning staff per 1,000 residents likewise tend to be stricter. • Higher rates of population growth at the county level are associated with a less stringent review process at the local level. • Cities with higher local revenues (per capita) are viewed as stricter, despite what one might presume to be their greater budgetary flexibility. In another survey question, we asked planners to speculate about the overall effect of their towns’ residential policies on the local population. The possible responses are that the city’s residential policies have led to a population somewhat lower in social status than it would otherwise be; the same in social status; somewhat higher in status; or much more affluent than it would otherwise be. Although admittedly somewhat subjective (which caused a relatively high 28 percent of respondents to ____________ 8Note that there is little or no relationship between homeownership levels and the degree of citizen opposition; the correlation between these two variables is only 0.03. 61 say “don’t know”), this question nevertheless gets to the heart of the issue: the perceived effect of local policies on city demographics. The respondents were not defensive, as some might expect. They readily pointed out the connection that is likely to exist between greater numbers of local restrictions and higher levels of social status for local residents. In short, they acknowledged the role that such restrictions might play in limiting access to a community. From our multivariate analysis, we make the following inferences: • Cities with more citizen opposition to growth and higher homeownership levels are seen as having policies that have led to more affluent populations. • Cities with larger populations are seen as having residential policies that are less likely to increase the population’s social status. • Cities with a higher share of unsewered housing units are also seen as having less socioeconomically restrictive policies. This perception indicates that sewerage deficiencies are not generally part of an intentional program of growth restriction but rather an unintended infrastructure limitation that may hinder development of all types.9 • By contrast, cities with long commute times indicate that city residential policies have been socioeconomically restrictive. Finally, we examine responses to a question that asked whether cities “currently have a policy to require residential development to include affordable housing, however that is defined in your community.” Our statistical analysis assesses factors that affect the probability of respondents’ answering yes or no to this question. In this model, we also included the variable examining the affordability of homes in the community. Interestingly, in this case citizen opposition to growth was ____________ 9The issue is important in evaluating local policies, because observers have sometimes claimed that water shortages in some parts of the state (Marin County, Santa Barbara) have basically been intentionally courted by antigrowth forces, seeking an excuse for disapproving housing development. (On “man-made drought,” see Frieden, 1979, pp. 43–48.) 62 not significantly related to the local policy outcome. Factors that were related to an affordable-housing set-aside include: • Cities with more “unaffordable” housing prices (relative to local incomes), as of 1990, are more likely to have an affordability requirement. This is possibly a reassuring result, indicating that cities with the worst affordability problems are taking actions to ensure production of affordable units—although it might also indicate desperation among respondents whose communities have already suffered deteriorations in affordability. • Cities with more stable, less transient populations are less likely to have the policy. • Perhaps surprisingly, cities with high homeownership levels and smaller populations are more likely to have affordability requirements for developers. It is possible that such cities have trouble meeting their affordable housing goals if they do not overtly require that such units be built. • Cities with a higher share of Hispanics, and cities in the generally more liberal Bay Area, are more likely to embrace affordability set-asides. Explaining City Government Orientation Toward Housing and Growth The next set of analyses deals with the city’s overall orientation or policy position toward residential development. Survey questions on this topic called for somewhat subjective responses but are relatively direct in asking respondents to assess the “friendliness” of the city government toward housing. The first two questions are drawn from the planning director survey. We asked the planners to assess “the general attitude of the majority of your city council toward residential growth.” Respondents could indicate that the council generally encourages residential growth, that it is mostly neutral, that it occasionally acts to slow the rate of growth when it becomes controversial, or that their council often proposes limitations on residential development. Which factors help explain cities’ positions on this four-point scale? 63 • Citizen opposition to growth is by far the most significant variable in increasing council antigrowth orientations. One would expect, in a democratic system, that council activity would reflect voters’ strong preferences on such an issue, to some degree. • Cities with longer commute times are much more likely to have councils wary of residential development, as are cities with stable (nontransient) populations. • Cities with high shares of unsewered housing units tend to have city councils that are less likely to take actions to slow growth. Again, a lack of sewer infrastructure probably makes such actions a moot point. • Cities with larger populations tend to have councils that are more progrowth. • Perhaps surprisingly, cities with high homeownership levels and more Democratic registrants also tend to have progrowth councils.10 We also asked the planners about their views regarding their cities’ development policies, with particular attention to whether commercial development is treated more favorably than housing. They could answer that their city “encourages all sorts” of residential and commercial growth, that it encourages commercial growth but is less receptive to multifamily and affordable housing projects, that it encourages most commercial growth but makes residential development more difficult, or that their city makes it difficult for both commercial and residential development. For simplicity, we will use the shorthand continuum of “permissive” to “restrictive” in characterizing respondents’ answers about their cities. Findings are relatively consistent with those discussed above: • Citizen opposition to growth is associated with greater restrictiveness, as is local population stability. ____________ 10When we included the variable measuring whether city councils are elected by district or at-large, it was not statistically significant. However, when controlling for district elections, the share of Hispanics in the population was positively related to council restrictiveness. 64 • Long commute times are again associated with greater restrictiveness. • Cities with more Democrats report more permissive development orientations, whereas the percentage of Hispanic residents in the city is associated with restrictiveness. • When information on city council election methods is included (reducing the number of cases), it appears that cities with district-based elections are more likely to have restrictive policies, perhaps indicating that district-based council members respond more to antigrowth sentiments in the neighborhoods. We can use some of the responses to the survey of city managers to cast further light on city orientations toward housing. Recall that the city manager survey was a statewide analysis, whereas the planning director survey was limited to the three major economic regions of the state. We are forced to do without some variables that were derived from the planner survey, such as the measure of citizen opposition to growth and the number of planning staff. However, we can turn instead to useful data from the city manager survey, in which respondents were asked to rate the importance of neighborhoods and business groups in land-use decisionmaking. The dependent variables—that is, the things we are trying to explain—are answers to two questions relating to multifamily development. First, regarding new development projects on vacant land, we asked, “Given your city’s overall strategies and plans for land use and future development, how desirable to your city administration would multifamily housing be?” The second item was an analogous question, about the receptivity of the city administration to multifamily development in redevelopment areas.11 For the most part, findings are consistent across the new development and redevelopment questions. • Cities with longer commute times are significantly less enthusiastic about multifamily housing development. ____________ 11Responses to the “new development” question are limited to those cities indicating that they had vacant land available. Responses to the redevelopment question are limited to cities indicating that the city was engaged in redevelopment. 65 • Where neighborhood interests are influential, city governments are more accepting of this type of housing, according to the respondents. • Respondents in the Bay Area indicate more interest in multifamily housing among their city administrations. In Southern California (outside Los Angeles County), however, multifamily development is less desirable for redevelopment areas. • There is at least modest evidence that cities growing at a faster rate in the 1990s are, in fact, more willing to accommodate multifamily housing. • Finally, cities with a “very active” redevelopment policy effort are much more interested in multifamily housing for their redevelopment areas. Explaining Levels of Public Controversy over Residential Development As we have seen, the public’s opposition to growth is a major motivator of certain growth-limiting policies and orientations on the part of local government. It is therefore important to ask which factors help generate public controversy over residential development in the first place and which local characteristics help predict whether citizens will bring the growth debate to the ballot box. We return to the planning director survey to assess these issues, examining questionnaire responses about the influence of growth issues in local politics. Note that we hardly claim to have a definitive explanation of local growth controversy. The level of political controversy in a city is likely to be affected by the vagaries of local history and politics, which we can capture only roughly in such an aggregate analysis. For example, a degradation of local public services, a sense of changing community character, or a widespread sense among residents that the community has lost control of its destiny might raise antigrowth controversy. Although we cannot measure such community patterns directly, the multivariate 66 model does help identify factors that probably make controversy more or less likely.12 For the first analysis, we attempt to account for the respondent’s answer to a question about the level of public controversy over residential growth issues in the city. Respondents replied using a four-point scale, ranging from “not at all controversial” to “almost always controversial.” The results indicate that several contributing factors stand out: • Community ethnicity and socioeconomic status appear to matter, but in a relatively complex fashion. All other things equal, cities with high proportions of Hispanics and blacks in the population experience less controversy over growth—but so do cities with high homeownership rates.13 It is possible that high homeownership in a city equates to low conflict because such cities are homogenous enough that they already have a degree of consensus regarding what types of growth to pursue. For example, such cities may be zoned overwhelmingly for single-family housing, which gives developers few opportunities to propose (potentially controversial) apartment projects. • Not surprisingly, cities with longer commute times have higher levels of public controversy. • Central cities of metropolitan areas, as well as rural cities, show more controversy than suburban cities. Despite a number of highly publicized growth battles in suburbia, the result may indicate that central cities, which often have a wide variety of land uses, neighborhoods, and income levels, are prone to ____________ 12In these models, we drop the variables relating to planning staff size, recreational homes, and own-source revenue because of their persistent lack of anything near statistical significance. Local population change is used in place of county-level population change, on the assumption that local increases are most visible and immediate to local residents, who may be less concerned with the dynamics of the entire regional housing market. 13Percentage Hispanic is quite correlated with the percentage of the city population that was unemployed (as of 1990), a variable not included in the model. If percentage unemployed is used in the model in place of percentage Hispanic, the unemployment variable also shows a strongly negative relationship to controversy. Thus, the results discussed here may mainly indicate community need or deprivation, rather than any aspects of ethnicity per se. 67 experience more conflicts over growth than the more homogeneous suburbs. Rural towns often experience the pressures of urbanization and farmland conversion, which could render growth issues more controversial, all other things equal. • San Francisco Bay Area communities report higher levels of controversy, even controlling for these other factors. • When we add local crime rates to the model (which slightly reduces the number of observations, because of missing data), it appears that high crime rates are associated with less controversy over residential growth. Crime, as a competing issue, may well displace growth from the local “controversy” agenda, by giving residents something more frightening to worry about. Highcrime communities also tend to have other problems attracting development, and, as such, might welcome development of any kind to provide local economic stimulus. Note that the other variables discussed above retain their effects when crime is included in the model. Another question asked the planners to describe the influence of residential development issues on their city’s mayor and council elections. They could respond that growth issues “hardly ever” affect local elections, that growth has done so “a few times,” or that growth issues are “often influential” in affecting election outcomes. • Here again, a higher share of Hispanics in the local population is associated with fewer electoral battles centered on the growth issue. • The same is true of high crime rates, if crime is entered into the model. • Cities that are more “job-heavy” are less likely to experience these housing-related electoral controversies (although the inclusion of crime rates appears to eliminate this relationship statistically). • Bay Area cities and central cities are considerably more likely to experience the electoral controversies—as are rural communities. Suburbs, it seems are less likely to have growth issues enter the electoral sphere, all other things equal. 68 Finally, we examine the issue of citizen initiatives regarding growth. Two survey questions asked whether “initiatives on the ballot [have] been a major source of policies to slow residential development,” and whether there is “a good chance that an initiative measure to slow residential development will occur” in the city. Focusing on the significant findings that are consistent across these two questions: • Cities in the Bay Area show higher probabilities of initiatives, both in the past and the expected future. Once again, suburbs appear to show less of this controversy than central cities and rural towns. • Cities with greater proportions of Hispanics and blacks in the population have a lower probability of initiative use.14 • “Job-heavy” cities also appear to be less likely to bring housing growth issues to the ballot box, as do cities with a higher proportion of unsewered housing units. • Interestingly, planners from cities that had faster rates of population growth in the 1990s are more likely to indicate that a future antigrowth initiative is probable. Summing Up In this chapter, we first identified four sets of factors that might be thought to affect local housing and growth policies: local political and fiscal factors, the demographic makeup of the city, city size and location, and local infrastructure and growth-related conditions. We then reviewed a wide variety of statistical analyses designed to cast light on the motivations behind cities’ use of growth management and their orientations toward residential growth. We report numerous multivariate models, each of which approaches the problem in a somewhat different way. Our main goal has been to detect overall patterns of findings, rather than to stress any individual result. Taking the empirical results as a whole, we can derive a tentative ____________ 14Here again, percentage Hispanic may serve as a proxy for percentage unemployed or some other measure of economic deprivation. Populations with less economic wherewithal tend to have a more difficult time mobilizing the resources necessary to carry out a political campaign, such as an initiative effort. 69 but plausible account of local growth management and housing policy that seems consistent with the evidence. As the previous chapter suggested, the level of citizen opposition to growth is a fairly consistent determinant of local residential policies, as we would expect in a democratic political system. Citizen controversy over growth, in turn, is more likely to arise under certain types of conditions. For example, growth-related controversies appear to be more common in cities with knotty commuting problems and in those where “bread and butter” issues, such as crime or unemployment, are not major factors. Cities with an “excess” amount of housing, given the number of jobs within the city, are also more likely to see voters bring their concerns over additional housing to the ballot box (both via antigrowth initiatives and in mayor and council elections). A high growth rate and high socioeconomic status do not, themselves, automatically translate into antigrowth activism. A number of local factors combine with citizen antigrowth movements to make city government growth-management policies, or restrictive orientations toward housing among policymakers, more likely. Again, local growth-related conditions play a major role, as one would expect from local officials’ assessing the capacity of their locality to handle growth. Long commute times render city councils more restrictive toward housing and make multifamily housing appear unattractive to officials. Communities with longer commute times might be experiencing substantial traffic congestion, since it is not clear whether the longer times are due strictly to greater distance or more traffic. Obviously, higher levels of traffic are often connected to less enthusiasm for growth. A rapidly growing county-level population, on the other hand, creates pressures to accommodate housing growth. A lack of local sewerage facilities appears to make widespread urban growth less possible, thereby reducing the likelihood that growth-management policies will be needed or that the council will have to take a hard line on growth. At the same time, cities with “unaffordable” local housing appear to pursue affordable housing set-asides in their planning policies. In short, conditions “on the ground” in California communities play a major role in shaping city residential policies. 70 The characteristics of the local population matter too. Cities with transient populations appear more likely to have progrowth councils and to pass fewer growth-management policies, perhaps because the community is less “settled” and has less of a distinctive identity. Local ethnicity and socioeconomic status affect housing policy in multiple ways that are not always easily predictable by those who see growth management as the tool of affluent whites. Cities with high homeownership rates, for example, do have more growth-management policies and a stricter review process of residential proposals but also appear more likely to have affordable housing set-asides and city councils that are neutral or progrowth, judging by the survey responses.15 Cities with more Hispanics experience less controversy over residential growth but tend to have a higher number of growth management policies, all other things equal. Despite our best attempts to institute controls for conditions that differ across various parts of California, San Francisco Bay Area communities appear distinctive in many of these analyses. Compared to cities in other regions, those in the Bay Area experience more controversy, are more likely to pass antigrowth citizen initiatives, and enact a greater number of growth-management policies—even as the region’s city officials embrace affordability set-asides in new developments and view multifamily housing as more attractive than their counterparts elsewhere in the state. By contrast, once we account for local conditions and demographic characteristics, suburban municipalities actually appear somewhat less troubled by growth controversies and appear less likely to pass growth-management policies.16 This evidence suggests that local growth controls and residential policies seem to be motivated by something more than simply the self____________ 15Of course, it is also possible that in the communities with very wealthy households, general market conditions might make it exceptionally difficult for more moderate- or lower-income families to find housing, no matter what the local inclusionary effort is. 16Even without controls, suburbs tend to have a smaller number of growthmanagement policies (2.5 on average) than central cities (3.0) or rural cities (2.9), although only the difference between central cities and suburbs is statistically significant. 71 interest of wealthy suburbanites, despite what many critics contend. Day-to-day local conditions, such as traffic congestion or an excess of housing to local jobs, seem to motivate citizen resentment of growth. In other words, it is the real consequences of growth “on the ground,” rather than merely local snobbery, that provoke citizen opposition. In relatively stable and settled local communities without other overriding issues, this citizen unease contributes to the passage of policies that attempt to increase public control of the rate and character of housing development. 72 6. The Broader Context of Local Residential Policies Even as the economy has cooled down, housing supply and affordability remain significant problems in California. Judging by previous cycles, even as it is possible that growth will abate for a while, population increases and urban development will accelerate again, and concern for local residential development controls will also resurface. For now, there is not as much housing—particularly multifamily housing— being produced as would seem to be warranted by the high prices and demand levels in many areas of the state. This feature of California growth has gotten worse in the previous decade. Thus, policy observers have sought the causes of California’s housing dilemma. As in previous public discussions regarding growth, local growth management is often assessed a significant portion of the blame for the affordable housing deficit. Our purpose in this chapter is to discuss the rather limited role that local growth-management policies probably play in affecting residential supply and price levels and to highlight broader contextual features of the housing market public policies in the state that may be of greater importance. Is Local Growth Management at the Root of the Housing Problem? For the most part, local growth-management policies of the type we have described in this report—and particularly, those that set overt limits on the amount of housing that can be developed—do little or nothing to further the goal of housing production. Nevertheless, they are unlikely to be one of the most significant factors constraining housing. Given the small numbers of jurisdictions pursuing serious growth controls, the reputed progrowth or neutral attitudes of most city councils, and the 73 reasonable motivations that appear to underlie cities’ growthmanagement policies, we have come to doubt that these local policies are a major root cause of housing affordability or supply problems. And although we do not have direct evidence regarding the relative factors that constrain housing, it appears likely that broader features of California’s political economy, to be discussed in this chapter, are more important than local growth management. The average city in the three regions has adopted fewer than three of the 16 commonly discussed policies that we asked about in the survey— and fewer than one of the seven policies we termed overt restrictions. In fact, nearly a quarter of the respondent cities (24.2 percent) had adopted either no growth-management policies at all, or else only design review standards—a fairly ubiquitous and frequently benign land-use regulation. Three quarters (75.8 percent) adopted fewer than four policies. Although clearly not all growth-management policies are created equal in terms of their ability to limit growth, active growth-management efforts are concentrated in relatively few cities. Of equal importance is the fact that many cities appear to welcome population growth. Of the city managers reporting in our 1998 statewide survey that their cities have vacant land available for development, 60 percent rated single-family housing as a somewhat desirable or very desirable type of development to the local administration; 26 percent even reported that multifamily housing— generally thought to be least advantageous from the standpoint of city treasuries—was seen as somewhat desirable or very desirable for their city. Among those cities pursuing redevelopment, 37 percent of city managers reported that single-family development was desirable for redevelopment areas, and 41 percent said that multifamily housing would be desired in such areas. These favorable orientations to housing in some jurisdictions may be accompanied by an easing of regulatory hurdles. In our survey of planning directors in the three regions, 58 percent indicated that their city had “made it easier (quicker) for residential development projects to be reviewed.” In response to a similar question, 81 percent said that in the past five years, the time required to complete the review of residential projects in the city had either stayed the same or gotten shorter. 74 Therefore, given that some cities welcome housing and a relatively small number shun it, it is quite possible that city growth-management policies may shift the production of housing units in any given area rather than limit it. As observers have sometimes suggested, local efforts to slow regional growth may be akin to pushing down on an inflated balloon in one place: It simply pops up in another. A further consideration is that local growth-management policies are not a new trend. Such policies date back at least to the early 1970s, when environmentalism and antigrowth sentiment first became a major social force (Frieden, 1979). Nevertheless, the passage of these earlier growth management policies did not prevent the state from experiencing major increases in population and housing in more recent decades. Some areas of the state were considered to have actually overbuilt housing during the 1980s. When the early 1990s recession took hold, property values dipped and substantial numbers of homes went on sale in foreclosure. Finally, it is worth noting that local growth-management policies often seem to be passed in reaction to rapid development in a city, rather than in anticipation of it. In these cases, growth management follows a population boom, rather than preventing it. Furthermore, some growthmanagement policies appear to be largely symbolic, aimed at pacifying the local antigrowth movement rather than actually slowing down development (Warner and Molotch, 2000). For example, some city councils have passed rather generous “caps” on annual development that have little chance of being exceeded by the number of units proposed by developers. It is even possible that local efforts to slow growth might generate efforts to hurry the processing of permits to avoid new regulations. The result can be the programming of substantial amounts of new development in a city. Construction that is “in the pipeline” when a growth-management measure passes is generally allowed to be built under a grandfather clause. Thus, some cities continue to see increases in housing development after passing a growth limitation. For this reason, we have not attempted to use the results of our recent survey of local growth controls to try to explain differences in city growth rates. Clearly, a long-term perspective is necessary in evaluating the systematic effect of growth-management policies on local growth. As 75 indicated in Chapter 2, the academic literature has not yet provided definitive conclusions about the effects of local policies. The Timing of California’s Housing Production Slump In Chapter 2, we noted that California’s share of national housing production fell into a deep decline in the 1990s, with only recent and modest recovery. Rather than affixing responsibility for this decline on the relatively small share of cities that have passed growth-management measures in recent years, one might be better advised to examine California’s relative economic performance in this period. In Figure 6.1, we show the state’s unemployment rate (as compared to the nation’s) over the same time period as the state’s housing production (relative to the nation’s). For both ratios, as the line slopes downward, it means that California is doing relatively worse, in terms of either unemployment or housing production. Although the two lines in the figure have different slopes— California’s worsening unemployment did not deteriorate to the same 16 1.0 14 Unemployment 0.9 measure 0.8 12 0.7 10 0.6 8 Housing measure 6 4 2 0.5 0.4 0.3 0.2 0.1 00 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Figure 6.1—California’s Housing and Economic Performance, Relative to That of the Nation California’s percentage of U.S. housing permits Ratio of U.S. to California unemployment rate 76 extent as its housing production—the direction of change from year to year is very similar across the two series.1 The graph reflects the central fact that housing production is very sensitive to broader economic trends—in this case, the state’s recession of the early to mid-1990s, which was longer, deeper, and later than the national recession. With California’s unemployment rate very near the national average in 1990, the state was creating nearly 15 percent of U.S. housing starts. The state’s economic fortunes rapidly deteriorated, however, such that by 1994, the national unemployment rate was 6.1 percent compared to the state’s 8.6 percent. This was the low point of the state’s performance during the decade, and, following a year’s lag, the state’s share of housing production bottomed out as well in 1995, with California producing just 6.3 percent of U.S. housing. What followed was a broad, though uneven, recovery in the latter part of the decade, both in employment and in housing. On neither measure, however, had the state regained its strong position, relative to the nation at large, by 1999. Given the notoriously cyclical and volatile character of housing markets, it is not surprising that the dip in housing production was much steeper than that in unemployment. Although we certainly do not claim that California’s relative economic performance was the only factor affecting its declining residential production in the 1990s, the connection seems close enough to presume that the depth of the state’s recession, from which California has only more recently recovered, greatly affected the housing market. Moreover, the restructuring of the state’s economy and the distribution of different clusters of population in the state might have affected the housing markets in ways that are not fully understood. Less Visible Factors Retarding Housing Supply Few would dispute that macroeconomic factors are central to understanding trends in housing production. However, at least four less visible factors are also likely to play a significant role in depressing residential production: initial low-density zoning, restrictive building codes, the tax treatment of real estate, and condominium defect ____________ 1The correlation between the two data series is 0.96. 77 litigation. Housing advocates would do well to focus as much attention on these “invisible policies” as on local growth-control policies. Initial Zoning When voters or city councils choose to “downzone,” or change local general plans so as to require lower housing densities, attention from the media and interest groups is often intense. However, many communities have evolved in a low-density, large-lot fashion from their inception, without ever instituting any overt growth controls of the type discussed in this report. The existing zoning requirements of such cities, often affluent suburbs, are such that higher-density (and therefore, less expensive) housing construction is not permitted to begin with. In short, some communities have never experienced serious growth-control movements in recent decades, because their zoning traditions (often many decades old) make it prohibitive or impossible to develop large quantities of housing. In this way, local growth seems more incremental, and local streets and services are less likely to be negatively affected by growth. Therefore, growth controversy is less likely to occur. As Warner and Molotch argue, “Following the rules already in place is a powerful form of growth regulation that is not represented on lists of growth controls” (2000, p. 82). Zoning codes that restrict residential development to large lots— regulations that in some cases have been in place since the incorporation of these communities as separate municipalities—are less visible and controversial than pitched political battles over current-day growth controversies. But such zoning may well have had a greater effect on residential building rates than the growth-management policies sometimes passed in desperation after a community has begun experiencing growth-related conflicts. As one scholar states: While new and complex growth management programs have garnered much attention from the planning profession and environmental groups, it is still by placing undeveloped land into large-lot zones that communities most commonly control population growth and land conversion. This practice ensures that land either cannot be developed economically, or can be used only for the most expensive type of structures (Dowall, 1984, pp. 185–186). 78 Indeed, Pendall (2000) finds that municipalities with low-density zoning patterns experience lower levels of rental housing development, and with fewer entry-level housing options add fewer minorities to their populations than otherwise similar cities without such zoning. Neiman (1980) and Farley (1964) present evidence indicating that, over time, communities tend to retain their initial status differences vis-à-vis other localities, which may well be a function of initial zoning. Exclusionary or large-lot zoning appears less common in California than in eastern states. Still, local zoning deserves further study to determine the degree to which it limits housing construction in California or serves to segregate residents by socioeconomic status. It should be recalled, however, that we have as yet no accurate compendium of how zoning policies affect the supply of housing for residential development. It is not only large-lot, low-density zoning that might reduce the supply of land for residential development. If localities over-zone for other purposes, including commercial and retail uses, there is also a potential zoning effect that adversely affects housing supply. Finally, in many cases such zoning ordinances are established early on and are an aspect of the developers’ marketing campaign for communities located in highly desirable areas. Even in the absence of zoning ordinances, such places could accomplish most of the purposes of zoning through privately enacted deed restrictions. If the highly restrictive zoning ordinances of some of California’s elite communities had not been enacted, it is not certain that there would have been in their place more socially diverse communities. Building Codes Just as large-lot zoning can be used to create a “floor” for the price or quality level of housing in a community, restrictive rules regarding the materials or standards used in housing construction can also put upward pressure on the cost of new housing. Local and state building codes and requirements can address everything from the materials to be used in pipes, to seismic standards, to concerns about the thickness of walls, to fire-sprinkler systems. Although such standards typically have very 79 worthy goals relating to the public’s health, comfort, and safety, on the margin these regulations do reduce the ability of builders to produce lowcost housing units by specifying minimum quality standards. The tradeoffs inherent in such regulation are also worth consideration by those who wish to maximize housing production. Tax Treatment In the United States, single-family housing investments are given a powerful incentive: the deductibility of mortgage interest from income for the purpose of calculating income tax liability. In the 1986 federal tax reform, however, incentives for investment in apartment developments were reduced, as Congress elected to simplify the tax code by ending most deductions for commercial real estate. Thereafter, the permitting of multifamily units in California—which had been fairly robust in the mid-1980s—slowed to a relative trickle, by historical standards, and has not yet recovered its previously high levels (Landis, 2000, p. 33; Dunstan and Swenson, 1999, p. 2). Construction Defect Litigation Another factor that has disproportionately affected multifamily construction—the types of units most likely to meet the needs of Californians in the lower and middle portions of the income distribution—concerns litigation over condominium projects. Observers of housing markets in California have argued that (until very recently) judicial decisions seemed to promote lawsuits as the way to resolve even minor disputes between condominium purchasers and builders. The courts have dealt with housing sales on a strict liability basis, as opposed to a caveat emptor (let the buyer beware) basis (Dunstan and Swenson, 1999, p. 4). Purchasers of a new home or condo were allowed to bring lawsuits if construction defects were discovered at any time within ten years of purchase. Lawsuits were particularly common among condo owners because of the presence of organized owner associations (Sanchez, 2000). Many builders claimed that a profusion of frivolous lawsuits brought by condo associations and their attorneys raised builders’ costs and made liability insurance more expensive or difficult to secure. Many plaintiffs’ 80 attorneys, by contrast, argued that the large number of lawsuits was primarily a result of shoddy construction, particularly in the housing boom period of the 1980s (Dunstan and Swenson, 1999; Rojas, 2001). Since numerous parties (subcontractors, insurance companies) can be brought into these suits, they have often been protracted and difficult to resolve (although most are eventually settled out of court). Whatever the roots of the problem, liability insurance has become significantly more expensive for contractors to obtain, and most insurance firms that previously provided this type of underwriting appear to have left the California market (Dunstan and Swenson, 1999, p. 6; O’Toole, 2000). Builders of downtown San Diego condos report paying roughly $20,000 in insurance costs per unit in late 2000 (“Condo Scarcity,” 2000). This litigious situation may have been changed by a decision of the California Supreme Court in December 2000, holding that homeowners (and homeowners’ associations) may not recover negligence damages for construction defects if the alleged defects have not actually caused property damage or injury (Aas v. Superior Court of San Diego, 2000 Cal. LEXIS 9048). However, some attorneys and builders predict that similar lawsuits may continue to be brought under other legal options, and insurance costs will remain high (Sanchez, 2000). Therefore, it is unclear whether this constraint on the supply of multifamily housing will be eased in the foreseeable future. Builders’ groups are seeking legislation that would move toward an arbitration system for resolving claims. Factors Fueling Housing Demand and Prices in California The factors discussed above serve to limit housing supplies. But also important, in the context of contemporary California, are factors that fuel demand. Intense demand and limited supplies work together to generate rapid price increases. Desirability of Coastal Areas California’s affordability problems are concentrated in the coastal counties that are part of the state’s major metropolitan complexes, ranging from Sonoma to Santa Cruz Counties in the Bay Area, and from 81 Santa Barbara to San Diego Counties on the south coast. This geographic pattern is no surprise, given the economic vibrancy of these coastal areas, their rapid levels of employment growth in the latter 1990s, and the longtime aspirations of many Americans to live near the amenity-rich Pacific shore. Although it is true that localities in the coastal counties are more likely to institute growth controls, housing in these areas has long commanded a price premium compared to other areas of the state. During a period of rapid economic expansion, such as that during the latter 1990s, the desire of wealthy, mobile households to live in coastal areas bids up residential prices. The Capital Gains Boom The wealth of many households in California escalated rapidly in the mid- to late 1990s because of the long expansion and high rates of return from the stock market and other securities. Through instruments such as mutual funds and employee stock-ownership plans, participation in the stock market has broadened widely, and many investors experienced unprecedented gains in their portfolios in this period. These gains were readily available for use as down payments on homes, another factor that bid up prices and led to rapid cost escalation in some parts of the state. Researchers at the Federal Reserve Bank of San Francisco estimated the relationship between changes in the stock prices of San Francisco Bay Area high-technology firms and housing price changes in that region, noting, “economic theory would suggest that households experiencing unexpected changes to wealth adjust their consumption of durable goods such as housing” (Krainer and Furlong, 2000, p. 1). They found that stock price changes did help forecast housing price changes, above and beyond standard predictions focusing on employment growth and lagged values of economic indicators. The findings indicate a distinct “wealth effect” on Bay Area housing prices, relating to the boom in hightechnology stocks of firms based in the region. This wealth effect “has helped propel land prices to soaring levels and has created a new class of buyers that are driving a big trend toward luxury production homes with big square footages and long lists of near-custom options” (O’Toole, 2000). 82 Residential Development and California’s Local Fiscal System Another key feature that may affect residential development in California is the state’s convoluted system of local public finance. The property tax in California accounts for a rather low share of overall local revenues. For example, in fiscal year 1997–1998, only 6.5 percent of total city revenues came from the property tax (California State Controller, 2000, p. vii). Furthermore, local governments in California are not empowered to set their own property tax rates. Rather, the state legislature must allocate property tax revenues among the various local governments serving any particular property. Cities often receive a very small slice of the property tax pie; the statewide average share for cities is about 11 percent (Legislative Analyst, 2000, p. 14).2 This situation may have a detrimental effect on the development of housing in cities, although this point has never been empirically demonstrated. Assume that a builder proposes to construct a set of houses, each valued at $250,000, in a city that receives a 15 percent share of the property tax (a figure that is actually above average). With property tax rates limited to 1 percent (assuming no special assessments), the homeowner would pay $2,500 per year, but the city would receive only $375 of that amount. This figure would most likely result in the city “losing money” on the development, because the cost of the services it would provide to each home (police, fire, sanitation, infrastructure, etc.) almost surely would exceed the amount collected in property taxes. Because the city controls land-use decisions within its boundaries, it has a powerful disincentive to accept “too much” housing. Of course, cities have other revenue tools at their disposal. One of the most common, a revenue source of growing importance in many ____________ 2Proposition 13, the property tax limitation passed by voters in 1978, limited property tax rates to 1 percent and annual reassessments to 2 percent (until the property is sold). In addition, the share of the property tax directed to cities and counties declined beginning in 1992–1993, when the state redirected funds to the newly created Education Revenue Augmentation Funds (ERAF) in each county. This was a way for the state to reduce its own general-fund obligations to school districts when it was experiencing a fiscal emergency. Although often criticized, and partially mitigated by other transfers, the ERAF shift remains in effect. 83 parts of the state, is the developer fee or exaction, by which builders provide fees, land, or in-kind donations to help mitigate the public-sector costs of providing for the new development. The fees on a new housing unit are often many times the amount of annual property tax revenues generated by the unit. In a tight housing market, however, builders may be able to pass all or most of these costs on to homebuyers (Dresch and Sheffrin, 1997). Thus, developer fees also work to increase housing costs. Many cities have also attempted to raise their local sales tax revenues by promoting retail development within the community. But this emphasis on retail may also have negative effects on residential development if retail projects are preferred to those that include a residential component, or if land suitable for residential construction is zoned for retail use in the hopes of attracting commercial growth. PPIC’s 1998 survey of 330 city managers in California, regarding their cities’ development strategies, indicated that retail was the most preferred land use for both new development and redevelopment areas, with single-family and multifamily development rated near the bottom of the list of land-use categories (Lewis and Barbour, 1999). Therefore, the state’s system for local public finance, in the post–Proposition 13 era, adds to the burdens faced by residential development. Other “Macro” State Policies Affecting Local Decisionmaking Other statewide policies and rules also affect the decision calculus facing local governments, developers, and slow-growth activists. Here we briefly mention three important examples. Environmental Review The California Environmental Quality Act (CEQA), passed in 1970, requires that local governments ascertain whether development projects will have a “significant effect” on a variety of environmental values—and if they do, prepare an Environmental Impact Report (EIR).3 Originally ____________ 3Completion of a negative declaration (i.e., a finding of no significant environmental impact) may take up to 180 days, and the completion of a full EIR may 84 thought to apply solely to public-sector projects, the California courts have interpreted CEQA as applicable to private development projects as well, when those projects receive a discretionary decision by a governmental entity to award an entitlement (such as a permit). Although a full review of CEQA and its implications for development are beyond the scope of this study,4 our major point is that CEQA, which is mandated by state law, can be an unpredictable element of the residential development process and can introduce a great deal of uncertainty into homebuilding. Although the act was arguably intended to force governments to take account of significant environmental considerations in their planning and decisionmaking, CEQA has become an additional lever outside the traditional planning process for opponents of residential projects. In particular, they can sometimes use provisions of the EIR process to slow down, downsize, or beat back proposed projects, often through litigation. By adding to the time and uncertainty involved in the development process in the state, CEQA may have the effect of reducing housing development.5 The Initiative Process Direct democracy is another feature of the governmental landscape that can powerfully affect the development process. Like CEQA, the initiative process is a statewide rule imposed on local governments. To a greater extent than in other states, California law permits aggrieved ________________________________________________________ take one year from the time the application for the project is complete. (There are provisions for extensions under certain conditions.) As of 1990, about 20 projects received negative declarations (or mitigated negative declarations) for every one that required an EIR. However, California planners subject a much higher proportion of projects to initial studies than in most other states with environmental-review requirements (Landis et al., 1995, pp. xx–xxi). 4For useful overviews of the law, see Curtin (2000, Chap. 6), and Landis et al. (1995). 5Although its procedures are mandated by state law, CEQA permits local governments a fair amount of discretion in deciding what constitutes a reviewable project and what the local criteria and objectives for environmental review should be. By using their ability to issue a mitigated negative declaration, in which conditions are attached to a project before it is declared to lack a significant environmental impact, local governments have been able to push developers into providing certain public benefits for the community or building the types of projects favored by the locality. See Warner and Molotch (2000, pp. 95–100, 133–134). 85 individuals or groups to bring land-use matters before local voters for resolution—including decisions on individual projects as well as broader planning matters. This fact of political life ups the ante on both local government officials and developers as they engage in discussions or negotiations over housing proposals. The possibility that an initiative could overturn an otherwise “approved” residential project increases the degree of risk facing the developer. Thus, the potential threat of a local voter initiative could limit the enthusiasm of policymakers or developers for certain types of projects. For example, larger, denser, “affordable,” or mixed-use projects may offend residents’ sensibilities and increase the probability of a land-use initiative. These initiatives are often broad enough to affect a wide range of future planning and land-use options, meaning that public officials may be just as wary as developers of voter involvement. It is therefore no surprise that, in our planning director survey, cities reporting that future initiatives are likely tend to have adopted a higher number of growthmanagement policies. The idea may be to engage in enough growth control now to attempt to ward off the blunt instrument of land-use initiatives in the future. State Infrastructure Shortfalls There is an emerging consensus that the state has failed to invest in the maintenance and construction of public infrastructure in recent decades and faces an enormous backlog of unmet infrastructure needs. From school construction and modernization to power generation to bus systems, examples are rife of public goods that are underprovided or strained to capacity (Dowall and Whittington, forthcoming; Dickerson and Silverstein, 2001). A particularly relevant example concerns roads and transportation. We saw in the last chapter that cities with longer commute times tend to be less favorable to housing development, and more prone to controversy over residential growth. The state’s ranking at or near the bottom of the 50 states in transportation expenditures per capita over the 1990s has done little to inspire the confidence of local residents and officials that transportation improvements will be forthcoming. As Landis and Kroll (1989, pp. 157–158) concluded from their study of “growth wars” in 86 Southern California, “More than any other single factor, traffic congestion served to mobilize anti-growth sentiment,” although they point out that voters themselves have often been opposed to special transportation taxes and freeway-construction programs. Nevertheless, transportation and related infrastructure problems, such as school overcrowding, likely reinforce the wariness over growth among local interests. Recent data from the Inland Empire Annual Survey of residents in fast-growing Riverside and San Bernardino Counties indicate that respondents who give low ratings to the quality of local public services, and those who think traffic conditions are bad, are more likely to believe that growth has been bad for the region. For example, among Inland Empire respondents who strongly agreed that population growth was good for the region, 42 percent indicated that traffic was a problem, whereas 69 percent of the respondents who strongly disagreed that population growth was good for the region felt that traffic was a problem (Bockman, Neiman, and Sirotnik, 2001). A Broader Policy Dilemma In assessing the effect of public policies on housing markets, there is a kind of meta-question that stands astride this entire issue. What is the optimal level of housing in California? And from what standard do we derive our answer—the needs of the poor, the tastes of those preferring higher density, big city living, or the partialities of individuals inclined toward standard suburban tract living? Or should we sympathize with the objectives of those concerned about housing the state’s workers in suitable and healthy environments? Should the housing demand of perhaps millions of undocumented immigrants also be included in projections of future housing needs? Is there any optimal level of housing production about which analysts and policymakers, much less citizens, are likely to agree? On the one hand, the values of those who fret about the need for further growth find housing costs too high and supply too low. Those who are alarmed at the huge California population and its continuing growth, along with the variety of service, resource, and ecological strains facing the state, feel that housing construction is already excessive and poorly planned. Freemarket devotees insist that local regulations distort housing markets and 87 deprive entrepreneurs and property owners of their rights. “Smart growth” and environmental advocates assert that housing markets are artificially stoked by an array of policies that contrive to escalate housing demand. In some ways, owner-occupied housing is among the most lavishly subsidized of all consumer choices, given the tax deductibility of mortgage interest, federal programs to insure mortgages, and other policies. Some growth critics argue further that local policies fail to account for the externalities imposed by the collective effects of typical large-scale residential development projects. In any case, California continues to have a powerful attraction to people throughout the world. In flush economic periods, the state is almost always likely to find that growth will stress both the public and private sectors’ capacity to satisfy demand. In short, there is no consensus regarding the substantive dimensions, causal dynamics, or remedies of the problem of housing cost and supply. Consequently, it is very difficult to rigorously determine the role of residential development controls in the overall mix of factors that affect housing markets. Summing Up Local growth-management policies are often singled out for criticism for contributing to California’s housing problems. Our intent in this chapter has been to illustrate that such policies probably play a relatively small role in the context of the overall housing market—which is not to say that these policies are always benign or never have an effect. Most cities have few growth-management policies, few cities set overt limitations on housing development, and many welcome residential growth. Other, broader factors are likely to play a larger role in contributing to the challenges facing housing policy in California. Some factors appear to limit housing supply, including low-density zoning, restrictive building codes, the current tax treatment of commercial real estate, and litigation and liability insurance problems facing the condominium market. At the same time, demand for housing has been intense because of the continuing desirability of coastal areas of the state, and the wealth gains, particularly capital gains, that occurred for many California households in recent years. 88 Under the state’s fiscal rules, California cities may find that accepting housing works against their fiscal self-interest. CEQA and the initiative process are two other statewide factors that provide tools for housing opponents and that introduce uncertainty for builders and local policymakers. Meanwhile, the state has fallen behind in providing and maintaining its infrastructure, including transportation facilities, which means that growth is likely to create additional strains and controversies in many communities. Looking at the factors arrayed against housing production, local growth management appears to be a relatively small part of the picture, although future research is necessary to clarify its role. In some respects, it is remarkable that California localities accommodate as much new housing as they do. 89 7. Refocusing the Growth Debate In considering California’s posture toward housing and other types of growth, one must first acknowledge that the state’s “growth policy” is really the amalgamation of hundreds of local growth policies. These local policies, in turn, are devised in the face of the very real conditions— social, economic, fiscal, and environmental—faced by local policymakers and residents. One particular class of local policy actions—growth controls— retains a particular notoriety as a potential drag on housing development in the state. Our survey of city planners indicates, though, that residential growth-management policies are relatively uncommon among California municipalities. Moreover, the most restrictive options, such as annual caps on building permits, are embraced by only a small handful of communities, whereas certain policies to link housing development to infrastructure capacity, or to shape the form or location of growth within a city, are somewhat more common. The most notable determinant of the level of local growth-management activity is the level of public opposition to growth in the city. In addition, local conditions, such as commuting times and jobs/housing balances, help determine cities’ orientations toward housing and growth. Even taking account of all these factors, there appears to be a regional ethos in the San Francisco Bay Area supportive of a somewhat higher level of growth management, perhaps reflecting a diffusion of the growth-control concept in a region that was among the first to experiment with such policies. However, even in the Bay Area many communities seemingly do little to overtly “control” growth. Additionally, even if some of the localities with higher development controls reduced their number or stringency, it is not clear that these places would suddenly become havens for moderate- or low-income 91 housing. Some of these communities are likely to remain high-cost exclusive places, because the market demand for access to them would produce very high land rents in any case. Residential growth controls are merely a subset of the policies and other factors that affect homebuilding rates and, in our opinion, have received disproportionate attention.1 Other local policies, particularly zoning and general plan standards, are more fundamental in determining the amount of growth that can be accommodated. Statewide factors, such as the system for financing local governments, the presence of the initiative process, CEQA review mandates, and frequent litigation over condominium construction, also should not be ignored as factors that potentially hamper housing development. Ongoing market factors, such as the desirability of coastal communities and the escalation of household capital-gains wealth, have probably played a major role in fueling the rise of housing prices in many parts of the state. Finally, although we do not endorse or condemn growth management, it is important to note that fostering vibrant local governments, which have important functions and responsibilities in the context of a democratic society, is also a worthwhile policy objective. Providing local populations with the ability to make important decisions about what sorts of public policies they wish to promulgate to shape the physical development of their area is a critical feature of any serious commitment to meaningful local governance. Of course, this poses serious puzzles regarding how to balance localism with regional, state, and even national objectives. A high regard for local government surely must be weighed against high regard for equity, economic growth, and ecological harmony. ____________ 1The recent Department of Housing and Community Development (DHCD) report on housing constraints in California, Raising the Roof (2000a), is similarly somewhat equivocal regarding the likely role of local restrictive policies in slowing housing production. Although the authors of the report note that the entitlement process for new housing can be long and unpredictable in California, much of the uncertainty in the process rests with interpretation of CEQA, a state-mandated law. In the 46 case studies of development projects examined in the DHCD report, the approvals process, though complicated, “[did] not, as a rule, result in interminable approval delays or in ad hoc unit and density reductions ” (p. 109). 92 Yet, as we have seen, much of the public discourse and perhaps even the research on these matters occur in the absence of agreed-upon principles. For example, how do we assess the tension between local and regional housing objectives? What sorts of norms should guide decisions in such an arena? What is meant by adequate housing production? What does a free market in housing really mean, given the many public policy choices that fuel and intensify housing demand? Why Do Conflicts over Local Growth Management Have Such Visibility? Given our assessment that growth controls, per se, are probably not a major cause of California's housing difficulties, why is there so much attention directed at this topic? On the one hand, there are some fascinating legal issues associated with the question and there is an enormous body of legal research and litigation associated with local efforts to manage growth. In addition, conflicts over development projects or ballot fights over growth-control initiatives are interesting fodder for journalists covering local government—“sexy” stories within an often mundane world of local council meetings and planning commission hearings. Norms of news reporting may sometimes lead to distorted coverage of the development issue, however (Kaniss, 1991). Beat reporters generally define a story in terms of a current conflict (e.g., a developer and neighbors of a project arguing at a planning commission hearing). Broader trends, structures, or institutions affecting regional development are less often seen as timely news (e.g., a city’s low-density zoning or setback requirements and the effects of these on the economic viability of projects, the evolving balance between employment and housing in an area, the long-term effects of a sewer-system extension on growth patterns). For many observers, moreover, the entire issue of managing development is symbolic. On one side, environmentalists, anti-sprawl advocates, historic preservationists, and resource conservationists view typical suburban residential development as a source of waste, environmental degradation, and increased levels of separation by race, economic class, age, and lifestyle. On the other side, there are the 93 defenders of property rights, trade unions, building contractors, and realtors who see local development controls as anti-working class, hostile to traditional American values, and excessive government intrusion into free markets. Additionally, liberals concerned about the need to provide adequate housing for the poor and moderate-income households are often critical of local development controls that hinder multifamily and higher-density housing development. In short, for many critics and advocates of local growth controls, such policies are emblematic of concern for larger issues than those occurring in any given community. Finally, there are the short-term, immediate financial stakes that housing developers and property owners have in their individual projects or land values. Thus, real estate interests naturally and sincerely attach great significance to such controls and help elevate the topic and contribute to its salience on the policy agenda. The Challenge for Cities: Successfully Managing Conflicts over Growth Our conclusion is that local growth-management policies are by and large understandable responses to the cross-pressures city policymakers feel—not, for the most part, narrowly self-interested actions motivated by animosity to newcomers or by property owners’ collusion. In addition, broader forces in the housing market and the state policy environment probably have a more central role in explaining California’s high housing costs and slow production—which is not to say that local growth control policies never hinder homebuilding. Findings in this study indicate that conflict over growth, if allowed to escalate, can produce even higher levels of local restrictions on residential development. Higher levels of conflict associated with such events as city council turnover, rancor in public meetings, recall elections, and local ballot initiatives are often connected to the formation of local groups, which then fight it out over growth issues. If unresolved, these conflicts are likely to intensify and become a persistent feature of local politics. Moreover, the kinds of policies that derive from highly charged, strident decisionmaking environments may be more restrictive than would otherwise be the case. In short, our various indicators of 94 local political controversy over growth are connected in important ways to the number of growth controls adopted. Local controversy over growth issues is also strongly related to the existence of local citizen initiatives, which are typically blunt policy instruments. Emergency controls over development, such as growth moratoriums or ballot-box zoning, are poor substitutes for careful, realistic, and visionary long-range planning. To avoid potentially paralyzing controversy over growth, cities and counties need to find ways to manage community conflict before it erupts or, if it does occur, to have local officials take the lead in arbitrating competing demands if at all possible. Conducting thoughtful general planning efforts and revisions—including “specific plans” for important or sensitive parts of the community—and engaging a broad segment of the public throughout this process seem key to achieving more local consensus. A public that feels shut out from the planning process or is uninformed about its details is more likely to react to development proposals in ways that potentially hamstring builders and limit the discretion of local officials. In cities and counties where officials frequently amend or ignore their general plans when convenient for fiscal or political reasons, the patience of local residents can quickly wear down. In short, local governments must convince their residents that they do have a sound plan to sensitively accommodate future growth—and there is room for the state government to assist in these efforts. If the management of conflict is carefully designed, and if there is consistency in implementing local policy, it is much easier to assert that citizens have obligations to be fair and to permit development that is in accord with local policy. Two State Policy Challenges: Local Government Finance and Infrastructure Shortfalls Thoughtful, skillful, and consistent local officials can make major inroads into the management of growth conflicts. But some problems that inflame negative perceptions about housing development cannot be dealt with adequately at the local government level. It is critical to have some sense of the overall context in which local growth-management efforts occur in California. 95 The city manager survey provides substantial indirect evidence of how fiscal motivations apparently affect local land-use policies. Local governments operate within a fiscal system that helps define the costs and rewards of various types of growth and development. The state’s fiscal system for local governments does little to soothe local anxiety about residential development. This is a central fact of life for city policymakers, who must focus to a large degree on ensuring that resources are sufficient to provide necessary local services to sustain the community. Local governments’ impulse to actively curtail residential development should be reduced if residential development is less fiscally problematic or if the state actually makes residential development a fiscally positive outcome. California’s system for funding local governments currently skews incentives for land development. The limited role of the local property tax and the lack of local control over its rate mean that local officials often must seek other ways to financially support the services required for new development. Escalating developer fees and special assessments are one option, increasingly in use since the passage of Proposition 13, but these might have disproportionate effects on new housing and contribute to higher home costs. Moreover, reliance on today’s new-home fees to pay for services for last year’s development is a potentially risky financial strategy if an economic slowdown hits, as a number of cities have discovered (Arax, 1999). A major alternative strategy for raising revenue—pursuit of local sales tax revenues—tends to favor retail development, possibly at the expense of housing development. The skittishness about residential development is in some cases exacerbated when localities find themselves in competition for retail development and begin to allocate more local resources, whether fiscal or in the form of providing greater amounts of lower-cost land, to prospective commercial developers. Although many city officials responding to our survey indicate a willingness and sense of responsibility to accommodate new housing, one can expect them to go only so far in the face of the fiscal disincentives facing residential development. Unless the state increases the share of the property tax directed toward general-purpose local governments, many 96 cities may remain understandably wary about major increases in residential development. Second, insofar as there are shortfalls in the maintenance or expansion of infrastructure, citizens are likely to regard population growth, and thus housing construction, as a threat to local services or quality of life. It is no surprise that local policy reflects these concerns. Our findings indicate that cities whose residents experience long commute times, for example, are less interested in further residential development and more prone to resident and city council opposition to growth. Such local conditions, rather than the rate of recent population growth itself, appear to have a dampening effect on local government enthusiasm for housing. This finding is a strong hint that California must address its infrastructure problems if it is to convince its residents and local officials that new housing (and the associated new residents) will not reduce existing residents’ quality of life. Major shortfalls in the state’s infrastructure exist not just in the electrical power system but in transportation, school facilities, parks and recreation, and water and sewerage provision. Improving these systems, while still maintaining the state’s environmental quality, which is also highly prized by residents, is perhaps the major challenge facing state policymakers in the effort to accommodate new populations. If traffic congestion and existing population density are not to be hindrances to housing development, not only must infrastructure be improved but homebuilders and policymakers must find ways to integrate new housing more gracefully into the fabric of built-up communities. This goal requires attention to aesthetic and environmental sensitivity, support of cities’ commercial functions (such as in downtown mixed-use areas), as well as transportation linkages such as accessibility to public transit and pedestrian amenities. Successful examples of medium- to high-density housing developments that do not disrupt the local quality of life—and perhaps make a net improvement to the community—provide the most powerful ammunition to supporters of new housing in existing urban environments. 97 Policy Options to Put Housing on a More Equal Footing The central focus of this study has been on the nature of, and motivations for, local growth-management policies, not housing policy per se. Of course, the reason we concern ourselves with local residential development controls is because these policies are so often linked to various housing problems and are sometimes the targets of reform and legal attack. However, we have argued that among the panoply of factors and other policies that affect housing supply and price, local development controls are probably not near the top in importance. Nevertheless, having noted that many factors in the state create potential problems for the production of housing, especially affordable housing, what other policy options or leverage might the state have regarding its housing-production woes? The possibilities we discuss in this section might reduce pressures among localities to be hostile or unfriendly to residential development. Our discussion, which is necessarily suggestive and brief, is guided by two underlying principles. One is that the “problem” of growth controls is better managed by focusing on other features of the housing market that probably adversely influence construction of more housing. The second is that if policymakers are to curtail the authority of local governments to take on important issues in their communities—such as defining a vision for community development and remediating the sideeffects of growth—then there should be a substantial public benefit produced. One seemingly straightforward policy option is to increase the state and local funding available for affordable housing programs, including funds for construction and rehabilitation, down payment assistance for households of limited means, and rent subsidies. Without directly addressing the structural constraints to residential supply that we identified in Chapter 6, however, such expenditures do not fundamentally alter the nature of the problem, however ameliorative the expenditures might be for those households receiving benefits from them. For example, recently developed policies that provide special housing assistance to schoolteachers and public safety employees, in California 98 and nationwide, are a special variant of this approach; they provide useful assistance for a relatively small number of workers in these favored professions, but do nothing to address the fundamental gap between demand and supply for California housing. It should be pointed out that the state has passed several important laws in recent years geared at increasing housing production and confronting affordability obstacles. These are reviewed in detail elsewhere (California Senate Office of Research, 1999; Department of Housing and Community Development, 2000b), but among the most significant were laws to • Allocate $100 million in incentive grants to local governments that increase housing supplies, with an emphasis on balancing jobs and housing, • Increase the California Housing Finance Agency’s revenue bond capacity, • Continue the state’s tax credit for development of low-income housing, • Streamline local approvals for affordable housing projects, and • Increase funding for farm worker housing. Such legislation—and the formation of a Housing Task Force by Governor Davis—indicates that state leaders have a clear interest in the state’s housing market and its affordability problems. In the same spirit, we list below some additional policy options, related to the issues raised in this report. These are presented not as recommendations but rather as considerations worthy of study and analysis by state and local lawmakers and others in the policy community. Our focus is on policy alternatives that could increase the amount of information available to those in the development process, address some of the “unpredictability” of that process, or reduce or mitigate local growth conflicts or controversies. Full “Planning Disclosure” for Homebuyers Under current law, a person selling a home must disclose known defects and hazards to the purchaser. However, homebuyers sometimes make the largest investment of their lives with little knowledge about the local plans for, or potential future development of, their neighborhoods. 99 This information shortfall can lead to protests, litigation, and political controversies when builders propose constructing a new development adjacent to an existing one—when residents thought they were buying into a project that adjoined open space. Individual cities and counties, or the state legislature, could pass legislation requiring that any residential real estate transaction include information about the land uses permitted within a certain distance of the home (perhaps a quarter mile), the zoning map for that area, and the permitted maximum future densities and numbers of units. (Some cities apparently already require that some of this information be disclosed to purchasers, but that is not uniformly the case.) The local generalpurpose government could prepare standardized information for homebuyers, to be distributed by real estate agents handling the sale. By signing this disclosure, homebuyers would signify that they know and acknowledge the plans and possible future development scenarios of the neighborhood they are buying into.2 This proposal flows directly from our finding that local controversy and conflict over growth are likely to generate housing restrictions, and that such restrictions might very well be excessive. One should not underestimate the important symbolic effect of a formal notification to a homeowner of the fact that adjacent land is planned for particular uses. When residents oppose something that they formally accepted at the time they purchased their homes, this prior acceptance would undermine their subsequent opposition. Naturally, such a policy should not deprive citizens of their right to petition for changes in policy, but this low-cost requirement might have some important benefits. In a similar vein, nearly 100 cities and counties currently have “right-to-farm” ordinances, in which home purchasers in agricultural areas are informed that they may be subject to the effects of nearby farming operations, such as spraying, noise, and odors. Although right-to-farm ordinances do not prevent a property owner from suing a nearby farm owner, they do serve ____________ 2Section 1102.6 of the California Civil Code already requires that a great many facts about the property be disclosed to the purchaser. Sections 1102.6a and 6b specifically authorize cities and counties to require additional disclosure statements, even providing as an example, “Adjacent land is zoned for timber production which may be subject to harvest” (California Civil Code, 2001, p. 170). 100 an important informational and educational rule (Wacker, Sokolow, and Elkins, 2001). Land Supply Monitoring Some observers have made the case that the supply of buildable land that can be provided with urban services appears to be dwindling in California’s major coastal metropolitan areas (Southern California Studies Center, 2001). Other analysts have found that “California has more than enough raw land to accommodate projected housing growth (at current densities)” for decades, with only Los Angeles, Orange, and San Francisco Counties lacking land capacity to meet projected demands (Department of Housing and Community Development, 2000a, Chap. 3). If land supply is constrained, and local governments pursue growth boundary policies that set additional land off-limits for development, land costs may become an ever-larger obstacle that could send housing prices spiraling higher. At present, however, there is insufficient unbiased information about the amount of privately held, residentially zoned land that has or has not been subdivided and could reasonably support housing. The science of “land supply monitoring” is still in its infancy. Nevertheless, if regional planning agencies could work with local governments to systematically track the amount of such land over time, policymakers and researchers would gain a much better sense of the potential inventory of residential land (including infill development) and the challenges that this constraint does or does not pose in various markets. Currently, the state Department of Conservation tracks the conversion of farmland to other uses. One could certainly make the case that state policymakers should consider providing support for agencies to track the supply of serviceable urban land as well, perhaps on a pilot basis at first. Addressing Condominium Litigation As noted in Chapter 6, litigation over alleged construction defects in condominiums is one factor widely believed to increase costs and decrease production of this important type of multifamily development. Although there are no accepted data on the exact degree to which such litigation has contributed to the construction shortfall in California, 101 attached multifamily developments are the housing sector that has declined the most in production since the 1980s. The recent California Supreme Court decision that limits liability awards in cases not involving injury or property damage may be one step toward making such projects more insurable.3 Others at the state level have been involved in negotiations to attempt to craft a longer-run solution. One approach proposes ten-year warranties for purchasers of new homes, with provisions for arbitration (rather than litigation) in cases of liability disputes. Increasing the predictability and dependability of condominium and townhouse transactions promises to benefit a number of important parties, including consumers, builders, insurers, and community associations. CEQA Streamlining and One-Stop Permitting Although this is not the place for a thorough investigation of the California Environmental Quality Act, it is worth reiterating that CEQA review can be one of the most unpredictable elements of the residential development process. Those interested in reform in this area might focus on streamlining CEQA to remove from the review process “environmental” impacts that can be regulated effectively through a city’s general plan or ordinances, such as parking, noise, public safety, and fiscal effects. On a broader level, more attention might be given to preparing thorough master environmental impact reviews at the time of general plan revisions or as part of the creation of “specific plans” for specified parts of the city. Residential and other projects that are consistent with those plans could then be “tiered” off the master EIR. Some jurisdictions have reported success in moving toward a one-stop permitting process, under which CEQA and other reviews are consolidated into a single process, based on conditions and mitigation measures specified in city ordinances or the general plan. See Landis et al. (1995, Chaps 5–6) for examples and reform options. ____________ 3Senate Bill 355, introduced in 2001, would have overturned this decision and widened the basis for defect lawsuits. The bill did not pass for enactment in 2001, but remains active. 102 Local Government Finance Reform The issue of the incentives created by the state’s system of financing of local governments, including the “fiscalization of land use,” has been discussed and editorialized upon so often that we do not wish to recapitulate the entire argument here (but see Lewis and Barbour, 1999, Chap. 6). In our survey of city managers, the desire for retail development, the lower degree of interest in housing, and the importance of augmenting sales tax revenue were vividly apparent. Numerous parties, ranging from groups convened by the state controller, to the Legislative Analyst, to the statewide associations of cities and counties, have proposed specific options for altering the local revenue-raising system, with some more complex than others (see Senate Local Government Committee, 2001); the legislature has also convened a number of groups to study the issue and propose remedies. Nearly all of these are based upon principles including reducing local reliance on a sales tax based on local retail transactions (which is seen as encouraging retail development at the expense of housing and industry) and increasing the share of property taxes directed toward cities and counties (thereby, presumably, making housing development more fiscally rewarding). Some have also suggested restoring some degree of authority over local property tax rates, which was lost as a result of Proposition 13 in 1978 (Legislative Analyst, 2000). Although such proposals have yet to result in major, long-term reform legislation, the subject remains an open one in Sacramento. We hasten to point out that our statistical results in this report offer no specific support for the idea that cities with a larger share of the property tax will take a more housing-friendly policy position. Indeed, the variable relating to the city’s percentage share of the property tax was not a significant predictor of any of the policy outcomes or orientations we investigated.4 Nevertheless, it is widely believed that low property tax rewards for housing hinder its chances in the development process. This topic merits further study. ____________ 4In fact, cities with one or more of the most restrictive growth-management policies have essentially the same estimated share of the property tax (13.2 percent on average) as cities without restrictions (13.4 percent). 103 More Support for Urban Infill Projects Many observers point to the potential of urban infill projects in established neighborhoods as potential win-win propositions. They can help improve and bolster older neighborhoods while providing additional housing units close to employment centers and making use of existing infrastructure. However, such projects face a number of constraints, which may involve environmental cleanup, land site assembly, high land costs, or complex financing for mixed-use projects. As a consequence, many builders without the resources necessary for such projects may choose to develop less-complex projects on vacant land at the fringes of metropolitan areas. Redevelopment policy is one major tool cities can use to address the issue of land costs and site assembly. Most infill housing in California in recent decades has received assistance from redevelopment agencies, and nearly one in 20 housing permits statewide went to redevelopmentfunded projects between 1988 and 1996 (Department of Housing and Community Development, 2000a, p. 69). Improved or increased housing facilities are often an important element of successful revitalization of a blighted area and can have synergistic effects for retail or commercial facilities developed as part of the project. The state’s Department of Housing and Community Development, or the Office of Planning and Research, could commission a study and prepare publications on best practices, illustrating and publicizing how mixed-use projects have been gracefully integrated into existing neighborhoods and downtowns. Recently, some regional agencies have begun providing financial rewards and technical assistance to cities and counties where transitoriented housing developments are built. Such programs can help address housing shortages and transportation problems simultaneously, by increasing the supply of new units in proximity to mass transit stops. For example, the Bay Area’s Metropolitan Transportation Commission provides funds to local governments in which multifamily developments of at least 25 units per acre are being planned within a one-third mile walk of a bus or rail route that provides frequent service. The funds, provided on a per-bedroom basis, may be used for relevant transportation 104 projects and upgrades in the area of the project.5 To provide a more powerful incentive for such developments, the level of funding may need to be increased far beyond current levels (currently, a one-time $1,000 to $2,000 award per bedroom) and the allowed uses broadened beyond transportation. Similarly, the state legislature passed a law in 2000 allocating $5 million in “predevelopment loans” to local governments or developers for proposed projects near a transit station (Department of Housing and Community Development 2000b). If this small program provides useful results, additional state legislation could provide expanded funding for a program of larger scale, perhaps to be administered by regional planning agencies and councils of government. It is worth noting, too, that in some parts of the state, the weakening office market and the tight housing market are giving developers a financial incentive to add highdensity housing to their commercial project in urban areas (Sanchez, 2001). Density Bonuses, Inclusionary Housing Requirements, and the “Builder’s Remedy” Under California’s Housing Element Law, originally passed in 1969 and modified several times since, regional councils of governments must periodically undertake a housing needs assessment and determine housing “allocations” for each city and unincorporated county area, including allocations of very low, low, and moderate-income housing.6 Furthermore, the housing element of each jurisdiction’s general plan must specify goals, policies, and programs for housing development and conservation, for all income segments (Curtin, 2000, pp. 10–12; Calavita and Grimes, 1998). Despite these requirements, there have been few if any penalties for lack of implementation of the housing goals or failure to achieve them. In short, these “fair share” requirements of housingrelated planning have been more process- than results-oriented. In 2001, ____________ 5This program was modeled on one developed by the council of governments in San Mateo County. For information, see http://www.mtc.dst.ca.us/projects/livable_ communities/lchip.htm. 6The local allocations must add up to the regional allocation imposed on the region by DHCD. 105 bills pending in the state legislature seek to provide legal ammunition for builders or others seeking redress from local governments that restrict affordable housing developments.7 In seeking to create more workable mechanisms for the production of affordable housing, state and local policymakers may wish to devote renewed attention and analysis to inclusionary housing programs and density bonuses. Inclusionary housing programs are local policies requiring that developers of larger residential projects provide a specified share of “affordable” units within the project. (In many cases, the builders are given the alternative of paying per-unit fees into an affordable housing trust fund.) According to Calavita and Grimes (1998, p. 158), 75 local governments in the state had inclusionary housing programs as of early 1996. Their interviews with city planners in the San Diego area indicated that many communities with inclusionary housing requirements adopted them to demonstrate tangible results regarding their housing elements and thereby avoid lawsuits from builders or housing advocates or conflicts with state government. However, critics of inclusionary housing requirements argue that they do nothing to increase the net number of housing units—and thus, do little to battle California’s undersupply of housing—unless the developer is permitted to build more units than would have otherwise been the case (Gruen, 2001). Otherwise, reducing the cost of some units in a development may merely shift the costs of subsidizing those units to other purchasers in the complex, whose purchase prices may be raised accordingly. A density bonus is the mechanism that allows additional units to be constructed. A density bonus is a concession granted to a builder who agrees to build a development in which a certain threshold share of units meets affordability requirements. The bonus allows the construction of additional market-rate units over and above the number otherwise ____________ 7For example, AB 369 (Dutra) authorizes courts to order local governments to pay plaintiff’s attorneys’ fees in cases where affordable housing developments are denied without required written findings. It was passed and signed into law. SB 910 (Dunn) would authorize courts to order financial penalties for localities failing to comply with the state’s housing element law, with the ultimate penalty being a reduction of certain state subventions to the jurisdiction. The Senate passed this bill but it failed to emerge from the Assembly in 2001. 106 permitted under local plans and zoning. Existing California law requires that localities provide 25 percent density bonuses (and an additional development incentive) to builders of projects that reserve 20 percent of units for lower-income residents, or 10 percent for very low-income residents, or 50 percent for senior citizens (California Government Code, Section 65915). Nevertheless, a developer generally must initiate the decision to set aside affordable units and apply for a density bonus, and actual use of the density bonus approach is uneven across California. According to a 1999 survey by the Office of Planning and Research, at least 215 cities and 30 counties in California have ordinances that “encourage” developers to use the state’s density bonus law (Governor’s Office of Planning and Research, 2000, p. 51). One policy option would be to create a more routine program of density bonuses and inclusionary housing requirements, with uniform thresholds of affordable units, for jurisdictions that show particularly high levels of unmet needs for affordable units. The state could require inclusionary housing programs with accompanying density bonuses for such communities or for communities whose housing elements are considered to be out of compliance. To ensure that the net stock of housing units actually increases, the density bonuses would be applied to all such projects if the affordable units were built on-site (rather than paid for in-lieu). Provisions could be applied to guard against too-rapid overbuilding in any given community—for example, a “trigger” would be reached when the number of housing units receiving entitlements to be built reached a certain percentage of the community’s initial housing units. At that point, the inclusionary housing and density bonus requirements would be suspended. Drawing upon New Jersey’s experience with its statewide mandated fair-share affordable housing program (Haar, 1996), developers could be permitted to sue recalcitrant communities whose housing elements are out of compliance or that have made little progress in producing affordable housing. The trial court would be authorized to impose a “builder’s remedy” for the provision of affordable units, mandating bonus market-rate units to the developer in exchange for the developer’s provision of a certain share of affordable units. Marrying the public interest goal of affordable housing to the profit motive of builders could 107 have a more systematic effect than simply requiring that localities write plans for housing (Haar, 1996). In Massachusetts, developers who propose a housing project in which at least 25 percent of the units are reserved for affordable housing are exempted from local zoning laws if the community has not met its affordable housing production requirements under state law. We are not completely persuaded that this mandated density bonus is the best approach for California to take, but merely put it forth as meriting further study. The proposal raises major issues of local autonomy and of reimbursable state mandates. However, housing element requirements and the provision of housing have long been held to be matters of overriding “statewide interest.” The litigation option is less flexible and more adversarial than we might like; as a deterrent, however, it could encourage pro-housing actions and planning changes by cities and counties—including jurisdictions that have engaged in such low-density zoning that lower socioeconomic groups effectively cannot gain entry to the community.8 However, this approach requires substantial confidence that there is an accepted method of making equitable and rational measurements of what a community's share of affordable housing ought to be, an assumption about which there is considerable controversy. Housing Action Coalitions Our final policy discussion relates as much to the private and nonprofit sectors as to government and represents a possible means to moderate local growth-related conflicts. In recent years, several San Francisco Bay Area communities have seen the formation of so-called housing action coalitions that lobby for appropriate housing developments. At local planning commission or city council hearings, ____________ 8The New Jersey experience shows that use of the builder’s remedy might require the creation of a set of specialized courts to handle land-use and affordable housing matters—an option that has been considered in prior reform efforts concerning land-use regulation in California—with the possible appointment of “special masters” trained in planning and real estate issues. Courts would need to consider whether the property held by the builder was appropriate for the project with the affordable-housing component (accessible to services, transportation, and employment, for example). 108 where project opponents often appear in full force and project proponents include only the developer’s staff, these pro-housing groups attempt to make their voices heard. Housing action coalitions, which often include affordable housing advocates, members of the clergy, architects, and members of the business community, provide a process for developers to “nominate” their projects for the group’s support. The coalition reviews the proposal in detail, ideally at an early stage, examining the proposed project’s relationship to neighborhood design, public transportation options, and affordable housing provision, among other criteria. The coalition may work with the developer and neighborhood groups to suggest modifications to the project that might allay neighbors’ concerns, without compromising housing production goals. If the project wins the group’s endorsement, and the group has credibility with local policymakers, the endorsement may be treated as something like the Good Housekeeping seal of approval. In communities with particularly contentious approval processes and high degrees of citizen activism in the development process, this is a promising avenue to improve project design and community consensus. Foundations, business associations, and the like could provide seed money to help start up and support incipient pro-housing interest groups. At the state level, relevant agencies concerned with housing, such as the Department of Housing and Community Development, the Office of Planning and Research, the Treasurer’s office, and others, could convene a continuing working group in collaboration with representatives of statewide housing groups, the development industry, and local government associations. This group would function as a kind of meta-level housing action coalition. The governor’s creation of a housing task force could be the basis of a more permanent working team in Sacramento. The group could identify obstacles and opportunities in state-level programs and administrative procedures that affect housing production (particularly multifamily and affordable) and suggest changes to streamline or assist in housing development. Recognizing and providing resources for such an effort would depend on a continuing, 109 articulated commitment to the housing goal on the part of the statewide elected officials. A Final Thought As this is being written, the state’s booming economy of the late 1990s has slowed down, and housing price inflation has begun to moderate. If indeed the state slips into recession, or if growth and housing issues no longer seem as pressing, some may turn their attention to other issues of the moment. However, a period of slower economic growth is not the time to turn away from the growth and housing dilemma, as there is every reason to believe that if there is a lull, that is all it will be. Issues of growth and conflicts surrounding residential development will inevitably appear again. Rather, when the real estate market cools—perhaps along with housing policy conflicts—seems a particularly propitious time to address the knotty problems of housing affordability and production, as well as local planning to avoid future growth controversies. Experience shows that when economic growth resumes, the political pressure cooker in which these issues are decided can make long-range thinking and policy action difficult. 110 Appendix Multivariate Data Analyses: Variables and Results This appendix provides detailed results of multivariate data analyses to support the discussion in Chapter 5. That chapter interprets the results, whereas this appendix is intended for readers who want a more detailed sense of the modeling and statistical estimations. After an initial section reporting data sources, the analyses in the appendix proceed in the same order as they are discussed in Chapter 5. Data Sources Each dependent variable in these analyses is derived from the answers to survey questions by respondents of our 1999 planning director survey or, in a few cases, to our 1998 city manager survey. More detail on question wording will be provided as we address each model. A full tabulation of the results of these surveys is available in Lewis and Neiman (2000) and Barbour and Lewis (1998), respectively. Regarding the independent (predictor) variables, the following variables were derived or calculated for each municipality using newly released data from the 2000 Census: • Percentage of housing that is owner-occupied, • Percentage of housing units that are recreational or seasonal, • Percentage Hispanic (of any race), and • Percentage black (not Hispanic).1 ____________ 1In the 2000 Census, respondents were permitted to check more than one category identifying their race. For the calculations in this report, we elected to use the “low estimate” of persons who checked the racial category in question (in this case, black) and no other categories. Note that racial-category percentages calculated in this manner are very closely correlated (r > 0.9) with city racial percentages from the 1990 Census, in which respondents could check only one racial category 111 Other Census variables are unfortunately not yet available for 2000, and therefore were derived from the 1990 Census: • Average commute time (one-way, in minutes), • Percentage of housing units unsewered, including those units identified as using septic systems as well as “all other methods” other than public sewerage, • Residential stability: the percentage of households living at the same address in 1990 as in 1985, and • Housing “unaffordability”: the ratio of the median house value in the city to the median household income. In addition, information to construct several other variables was gathered from a variety of sources, as discussed below: • Citizen opposition to growth: This variable is based on responses to the planning director survey. Respondents were asked, “Please review the following list of factors and circle the number that you feel best describes how important each factor is in constraining or slowing residential development in your city . . . Citizen opposition to growth.” Respondents chose a value between 1 (“not at all important”) and 5 (“very important”). • Percentage Democrat: This is the percentage of two-party registrants (i.e., Republican and Democratic) who are Democrats. Data are from the report of registration by the California Secretary of State, February 10, 1999. • Planning staff per 1,000 residents: Our survey of planning directors asked respondents to indicate the number of personnel employed in the city’s planning department—or, if the planning staff is contained within another department, to indicate the number of individuals involved in planning activities. The number of staff was then normalized to the 1998 population (Department of Finance estimate). • City share of the property tax: This is an estimate of the percentage of property tax revenues generated within each city that go to the city government (as opposed to the county government, school districts, special districts, etc.). This is 112 calculated from fiscal data for 1997–1998, made available as a database by the Legislative Analyst’s Office.2 To construct this estimation, it is assumed that total property taxes generated in a city are equal to 1 percent of the city’s net assessed valuation. The city government’s nondebt property tax revenue is then compared with this estimate of property taxes paid to calculate the “city’s share.” • Logarithm of per capita, own-source revenues: This is calculated from California Controller reports on city financial transactions for the 1992–1993 fiscal year. Own-source revenues are calculated as total revenues minus intergovernmental subventions and minus city enterprise revenues. Own-source revenues are calculated on a per-capita basis using 1993 population estimates by the Department of Finance. The log of this number is then taken because of the extreme skewness of the distribution. • County population change, 1990–1998: This is the percentage increase in population at the county level, based on population estimates by the Department of Finance. • City population change, 1990–1998: In some models, we substitute city population change for county population change. • Jobs-to-workers ratio: This is the logarithm of the ratio of the number of jobs within the city to the number of employed residents in the city. It is calculated from Bureau of Transportation Statistics data (1994). • Logarithm of 1998 city population: For this we used Department of Finance estimates. • Bay Area: This is a binary variable taking the value of 1 if the city is located in the nine-county San Francisco Bay Area. • Southern California, not Los Angeles County: This is a binary variable taking the value of 1 if the city is located in one of the following counties: Orange, Riverside, San Bernardino, Ventura, Santa Barbara, or San Diego. ____________ 2Data are derived from the Local Government Database at http://www.lao.ca.gov/ lao_menu_local_govt.asp. 113 • Central city: This is a binary variable taking the value of 1 if the city is designated as one of California’s 49 central cities in the 1990 Census. • Rural: This is a binary variable taking the value of 1 if the city is located outside a Metropolitan Statistical Area (MSA) or is not in an urbanized portion of an MSA. In addition, the following variables are used in a few models: • District council elections: This is a binary variable taking the value of 1 if all or some members of the city council are elected by districts rather than at-large. Data are from a Public Policy Institute of California statewide mail survey of city clerks in 2000, relating to local electoral institutions and voter turnout. (The response rate to this question was 82 percent.) • Local crime rate: This is the logarithm of the city’s average crime rate for the years 1990, 1991, and 1992, as reported by the California Department of Justice. • Business/Chamber of Commerce importance: In the city manager survey, respondents were asked, “Generally speaking, how important are the following considerations to your city administration’s strategies in attracting new development and responding to development proposals? Please review each item below and indicate how important it is [on a scale of 1 to 7]. . . . Support of Chamber of Commerce or other local business interests for project.” • Neighborhood importance: As part of the same question, respondents evaluated the importance of “acceptability of proposal to nearby neighborhoods.” Table A.1 provides summary information about the mean, high value, low value, and data source of each of the independent variables (except binary variables) and all of the dependent variables used in the various models. 114 Table A.1 Descriptive Statistics for Variables in the Multivariate Models Low High Mean Value Value Source Independent Variable Citizen opposition to growth 2.7 1.0 5.0 Planner survey % Democrats 56.0 18.8 95.0 California Secretary of State Planning staff per 1,000 0.2 0.0 1.9 Planner survey City's % share of property tax Own-source per capita revenue (log) Importance of neighborhoods 13.2 0.0 65.9 Calculated from LAO database; see text 6.1 4.0 12.7 Calculated from California Controller reports 5.7 1.0 7.0 City manager survey Importance of business interests 4.8 1.0 7.0 City manager survey Crime rate (log) 8.6 6.8 14.0 California Dept. of Justice % owner occupancy 61.0 16.0 97.1 2000 Census % Hispanic 30.2 2.2 98.3 2000 Census % black 3.8 0.0 46.4 2000 Census Commute time 22.8 6.3 40.5 1990 Census % unsewered units 5.9 0.0 99.0 1990 Census Residential stability 45.1 11.4 72.2 1990 Census County population change City population change Housing unaffordability 12.3 0.1 30.8 California Department of Finance 13.6 –31.8 171.1 California Department of Finance 4.2 1.5 11.6 Calculated from 1990 Census % recreational units 1.9 0.0 63.5 1990 Census Job/worker ratio (log) Population (log) –0.1 –1.8 2.4 Census Transportation Planning Package 10.0 4.4 15.1 California Department of Finance Dependent Variable Number of growth management 2.7 0.0 10.0 Planner survey policies Importance of density restrictions 2.8 1.0 5.0 Planner survey Strictness of review process 2.1 1.0 4.0 Planner survey Policies’ effect on social status 2.5 1.0 4.0 Planner survey Adoption of affordable housing requirement Council limits residential development 0.3 0.0 1.0 Planner survey 1.8 1.0 4.0 Planner survey 115 Table A.1 (continued) City makes development more difficult Multifamily housing desirability (new development) Multifamily housing desirability (redevelopment) Controversy over residential growth Residential issues affect elections Initiatives: major source of antigrowth policies Good chance of slow-growth initiative Low High Mean Value Value Source 1.7 1.0 4.0 Planner survey 3.6 1.0 7.0 City manager survey 3.8 1.0 7.0 City manager survey 2.2 1.0 4.0 Planner survey 2.0 1.0 4.0 Planner survey 0.2 0.0 1.0 Planner survey 0.1 0.0 1.0 Planner survey Controversy and Growth-Management Policies City planning directors were asked, “How controversial would you say residential growth issues are in your city?” They responded using a four-point scale that ranged from “not at all controversial” to “almost always controversial.” The relationship of local controversy levels to the number of growth-management policies is strong, as the following simple bivariate regression reveals: Number of policies = 1.05 + 0.73 (controversy) + e (t-score of controversy = 6.02; adjusted R-sq. = 0.11; N = 289) However, as discussed in Chapter 5, it is also reasonable to suppose that the number of growth-management policies may influence the level of controversy. We can help evaluate this question by relying on the number of growth-management policies identified in each city in a 1988 survey by Glickfeld and Levine (1992) to see whether that number is associated with the level of controversy in 1998. Indeed, it is, as the following bivariate model illustrates. We use ordered logit rather than 116 regression because of the nature of the dependent variable.3 The number of observations is smaller because some new cities incorporated after 1988 and because some cities responded to one survey but not the other: Controversy level = 0.33 (number of 1988 policies) + e (z-value of number of policies = 5.13; log likelihood = –274.59; probability > chi-sq. = 0.000; N = 243) Because of the apparent endogeneity between controversy and the number of policies, we do not use controversy level as an independent variable in subsequent models. Although there are various ways to deal with this problem (e.g., instrumental variables), we chose to substitute for controversy our survey-based measure of “citizen opposition to growth,” which is unlikely to be similarly increased by the number of growth-management policies. Models Relating to Housing and Growth-Control Policies A multivariate regression model of the number of growthmanagement policies identified by our survey is shown in Table A.2. As noted in the report, the number of policies may range, in theory, from 0 to 16, although the average is 2.7. In modeling the determinants of the number of policies, we use the rate of population growth at the county level, rather than in the city, because policymakers are likely to be concerned with the overall level of residential growth pressures in their region. (In any event, city-level population growth proved insignificant in these models.) ____________ 3Ordered logit models are preferred when the dependent variable takes the form of a small number of ordered categories, as in this case, where controversy ranges along a fourpoint scale from “not at all” to “almost always” controversial. Throughout this appendix, we use ordered logit rather than ordinary least squares (OLS) regression in these circumstances. Results using OLS are generally extremely similar in terms of the signs and significance levels of the variables. Respondents choosing the “don’t know” answer are dropped from the analysis. 117 Table A.2 Number of Growth Management Policies Adopted (Regression Model) Independent Variables Citizen opposition to growth % Democrats Planning staff per 1,000 City’s % share of property tax Own-source per capita revenue (log) % owner occupancy % Hispanic % black Commute time % unsewered units Residential stability County population change % recreational units Job/worker ratio (log) Population (log) Bay Area Southern California, not Los Angeles County Central city Rural B0 No. of cities Probability > F Adjusted R-squared 0.205 (2.39)** –0.010 (0.65) –0.167 (0.31) –0.001 (0.09) 0.229 (0.88) 0.036 (2.67)*** 0.021 (2.42)** –0.024 (1.17) 0.010 (0.32) –0.020 (3.20)*** –0.058 (3.69)*** –0.153 (4.97)*** 0.045 (2.02)** 0.070 (0.25) –0.118 (0.73) 1.783 (4.35)*** 0.652 (2.08)** 0.903 (2.30)** 1.171 (2.54)** 3.117 (1.10) 259 0.000 0.27 Notes: **p < 0.05, ***p < 0.01. Cell entries are unstandardized regression coefficients, with absolute values of T-values listed in parentheses, calculated using robust standard errors. Table A.3 uses the same model, this time in ordered-logit form, to examine city planning directors’ answers to three survey questions relating to local residential policies. In the first column of results, the dependent variable is the respondents’ assessments of the importance of “density restrictions on residential land” in slowing or constraining residential development in their cities. They ranked the importance of density restrictions on a five-point scale, from “not at all important” to “very important.” 118 Table A.3 Planner Assessments of City Policies (Ordered Logit) Independent Variables Citizen opposition to growth % Democrats Planning staff per 1,000 City’s % share of property tax Own-source per capita revenue (log) % owner occupancy % Hispanic % black Commute time % unsewered units Residential stability County population change % recreational units Job/worker ratio (log) Population (log) Bay Area Southern California, not Los Angeles County Central city Rural No. of cities Log likelihood Probability > chi-squared Pseudo R-squared Importance of Density Restrictions 0.677 (6.02)*** –0.029 (1.78)* 0.544 (0.94) 0.004 (0.19) –0.021 (0.06) –0.020 (1.42) 0.016 (1.57) 0.021 (0.78) 0.029 (0.72) 0.009 (1.01) 0.049 (2.59)** –0.104 (3.01)*** –0.004 (0.13) –0.038 (0.11) 0.275 (1.62) 0.077 (0.17) 0.226 (0.65) –1.324 (2.90)*** –0.225 (0.52) 258 –353.77 0.000 0.12 Strictness of Policies’ Effect Review Process on Social Status 0.273 (2.41)** 0.255 (1.92)* –0.018 (1.12) –0.019 (0.89) 2.069 (2.24)** –0.407 (0.57) –0.012 (0.58) –0.010 (0.39) 1.018 (2.50)** 0.137 (0.35) 0.036 (2.20)** 0.046 (2.33)** –0.004 (0.42) –0.012 (0.99) 0.021 (0.79) –0.024 (0.60) 0.032 (0.82) 0.079 (1.75)* 0.005 (0.62) –0.026 (2.01)** 0.008 (0.38) –0.008 (0.32) –0.083 (1.97)** –0.100 (1.59) 0.019 (0.54) 0.028 (0.57) –0.373 (1.05) 0.439 (1.01) –0.001 (0.01) –0.437 (1.97)** –0.495 (1.14) 0.226 (0.45) –0.227 (0.65) –0.195 (0.39) 0.939 (2.28)** 0.609 (1.04) –0.395 (0.92) –0.161 (0.29) 252 189 –251.36 –179.89 0.000 0.000 0.15 0.16 Notes: *p < 0.1, **p < 0.05, ***p < 0.01. Cell entries are ordered logit coefficients, with absolute values of z-values listed in parentheses, calculated using robust standard errors. In the second column of results, the dependent variable is the respondents’ assessment of the review process for residential projects in their city, in comparison to other cities in the area. Respondents used a four-point scale of less strict, equally strict, somewhat more strict, or much more strict. For the results shown in the last column of Table A.3, the dependent variable concerns the respondent’s assessment of “the overall effect of your city’s residential development policies on who lives there.” Answers were on a four-point scale, ranging from “the city population is somewhat lower in social status than it would otherwise be, because of 119 [the] city’s residential development policies,” to “the city population is much more affluent than it would otherwise be.” Table A.4 concerns another type of local policy outcome. “Does your city currently have a policy to require residential development to include affordable housing, however that is defined in your community?” Because the dependent variable is bivariate (1 = yes, 0 = no), we use a Table A.4 Adoption of Affordable Housing Requirement (Probit Model) Independent Variables Citizen opposition to growth % Democrats Planning staff per 1,000 City’s % share of property tax Own-source per capita revenue (log) % owner occupancy % Hispanic % black Commute time % unsewered units Residential stability City population change Housing unaffordability Job/worker ratio (log) Population (log) Bay Area Southern California, not Los Angeles County Central city Rural B0 No. of cities Log likelihood Probability > chi-squared % correctly predicted Pseudo R-squared 0.045 (0.56) 0.009 (0.77) –0.271 (0.46) 0.007 (0.50) 0.225 (0.93) 0.040 (2.86)*** 0.012 (1.65)* –0.021 (1.10) –0.033 (1.20) –0.001 (0.18) –0.059 (3.77)*** –0.021 (2.48)** 0.352 (3.45)*** –0.258 (1.00) –0.347 (2.34)** 0.707 (2.54)** 0.167 (0.65) 0.111 (0.35) –0.119 (0.38) 0.253 (0.12) 252 –129.53 0.000 74.6 0.19 Notes: *p < 0.1, **p < 0.05, ***p < 0.01. Cell entries are probit estimates, with absolute values of z-values listed in parentheses, calculated using robust standard errors. 120 probit model. We also include the variable measuring the “unaffordability” of local housing as of 1990. Models Relating to City Government Orientations Toward Growth Table A.5 reports ordered logit results regarding city government orientations toward residential development. The first column of results is an analysis in which the dependent variable is the planner’s assessment of “the general attitude of the majority of [the] city council toward residential growth.” Responses were along a four-point scale, depending on whether the council was seen as mostly encouraging residential growth, being neutral toward it, occasionally slowing the rate of growth, or often proposing limitations on residential development. The second column of results relates to the same dependent variable but includes among the independent variables a measure of whether the council is elected by districts. This results in the loss of some observations, because of missing data. The third and fourth columns of results in Table A.5 relate to the planner’s response to a question about the relative restrictiveness of city policies: “Which of the following comes closest to your view of the policies of your city regarding development?” Respondents had four options, which we coded from 1 to 4, as follows: • “My city encourages all sorts of residential and commercial growth.” • “My city encourages most commercial growth, although it is less receptive to multifamily or “affordable” housing projects.” • “My city encourages most commercial growth, but it makes all residential development more difficult.” • “My city makes it more difficult for both commercial and residential development.” Again, in the fourth column, the variable denoting district council elections is included. 121 Table A.5 Planner Assessments of City Government Orientations to Residential Growth (Ordered Logit) 122 Independent Variables Citizen opposition to growth District elections % Democrats Planning staff per 1,000 City’s % share of prop. tax Own-source per capita revenue (log) % owner occupancy % Hispanic % black Commute time % unsewered units Residential stability City population change % recreational units Job/worker ratio (log) Population (log) Bay Area Southern California, not Los Angeles County Central city Rural Council Limits Residential Development Model 1 Model 2 0.706 (5.43)*** 0.786 (5.03)*** — 1.003 (1.46) –0.040 (1.98)** –0.050 (2.18)** 0.257 (0.35) –0.051 (0.07) 0.005 (0.15) 0.002 (0.05) 0.409 (1.25) 0.572 (1.45) –0.034 (2.27)** –0.022 (1.39) 0.018 (1.61) 0.027 (2.04)** 0.019 (0.50) 0.028 (0.62) 0.162 (3.41)*** 0.154 (3.05)*** –0.024 (2.38)** –0.038 (2.84)*** 0.058 (2.35)** 0.067 (2.40)** –0.000 (0.02) –0.009 (0.65) 0.038 (0.75) 0.064 (1.79)* –0.037 (0.09) –0.254 (0.56) –0.464 (2.38)** –0.634 (3.01)*** –0.013 (0.02) –0.119 (0.20) –0.396 (1.04) –0.486 (1.27) 0.592 (1.15) 0.668 (1.16) –0.186 (0.33) –0.315 (0.55) City Makes Development More Difficult Model 1 Model 2 0.597 (3.97)*** 0.566 (3.55)*** — 0.993 (1.78)* –0.047 (2.19)** –0.058 (2.37)** 0.652 (0.42) 1.530 (0.83) 0.019 (0.74) 0.011 (0.41) 0.411 (1.32) 0.395 (1.06) –0.000 (0.02) 0.001 (0.03) 0.024 (1.94)* 0.021 (1.76)* 0.039 (1.13) 0.045 (1.10) 0.132 (2.65)*** 0.154 (2.83)*** 0.008 (0.81) –0.007 (0.52) 0.061 (2.02)** 0.052 (1.55) –0.025 (0.97) –0.025 (0.91) –0.032 (0.86) –0.043 (0.97) 0.607 (1.47) 0.439 (0.93) –0.403 (1.58) –0.552 (2.08)** 1.251 (2.43)** 1.302 (2.32)** 0.485 (1.32) 0.642 (1.61) 0.137 (0.25) 0.302 (0.48) 0.118 (0.23) –0.087 (0.15) No. of cities Log likelihood Probability > chi-squared Pseudo R-squared 251 –257.45 0.000 0.13 208 –208.25 0.000 0.16 240 –212.12 0.000 0.15 199 –173.74 0.000 0.17 Notes: *p < 0.1, **p < 0.05, ***p < 0.01. Cell entries are ordered logit coefficients, with absolute values of z-values listed in parentheses, calculated using robust standard errors. In Table A.6, dependent variables are drawn from the city manager survey on local development strategies. City managers were asked to rate the desirability to their city administration of various types of land uses, using a scale of 1 to 7 from “very undesirable” to “very desirable.” Here we focus on the respondents’ desirability score for multifamily residential development. The first column of results pertains to multifamily Table A.6 City Manager Ratings of Desirability of Multifamily Development (Regression Model) Independent Variables Importance of neighborhoods Importance of business interests % Democrats Very active redevelopment policy City’s % share of property tax Own-source per capita revenue (log) % owner occupancy % Hispanic % black Commute time % unsewered units Residential stability City population change % recreational units Job/worker ratio (log) Population (log) Bay Area Southern California, not Los Angeles County Central city Rural B0 No. of cities Probability > F Adjusted R-squared Multifamily Rating Multifamily Rating (New Development) (Redevelopment) 0.239 (1.90)* 0.370 (2.83)*** 0.111 (1.05) –0.090 (0.76) 0.023 (1.58) –0.014 (0.83) — –0.010 (0.44) 0.739 (2.74)*** –0.006 (0.27) –0.189 (0.66) 0.341 (1.17) 0.027 (1.80)* 0.002 (0.11) –0.013 (1.79)* –0.009 (1.10) –0.011 (0.37) –0.002 (0.09) –0.140 (4.62)*** –0.098 (2.93)*** –0.005 (0.65) –0.011 (0.96) –0.025 (1.51) –0.024 (1.25) 0.013 (1.65) 0.012 (1.96)* 0.005 (0.14) 0.026 (1.61) 0.258 (0.92) –0.379 (1.27) 0.547 (3.30)*** 0.284 (1.37) 1.338 (3.31)*** 1.117 (2.74)*** 0.155 (0.54) –0.064 (0.16) 0.816 (2.32)** –1.742 (0.59) 191 0.000 0.23 –0.774 (2.41)** –0.088 (0.20) 0.611 (1.51) 0.608 (0.18) 221 0.000 0.22 Notes: *p < 0.1, **p < 0.05, ***p < 0.01. Cell entries are unstandardized regression coefficients, with absolute values of T-values listed in parentheses, calculated using robust standard errors. 123 development on vacant land (in “new development” areas), whereas the second column concerns ratings for multifamily development in the city’s redevelopment areas. Responses were gathered only from those city managers indicating that their cities had at least some vacant land available (for the question on new development) or indicating that their cities engage in redevelopment policy (for the question on redevelopment). Included among the independent variables are the importance scores that the city managers gave for the influence of business interests and neighborhood interests in the development process. For the redevelopment regression, we also include a variable indicating cities in which the city manager said the local redevelopment effort was “very active.” The variable measuring citizen opposition to growth cannot be used because it is not derived from the city manager survey, and its use would result in the deletion of many cases.4 Models Relating to Citizen Controversy and Initiatives In this set of analyses, we address the degree of citizen controversy over residential issues. For the modeling in this section, the variable relating to citizen opposition to growth is dropped, since it may be one manifestation of controversy. Variables regarding planning staff size, recreational homes, own-source revenue, and the city’s share of the property tax were also dropped, as these proved very insignificant and are probably less relevant to the generation of popular controversy. Results in Table A.7 relate to two questions in the planning director survey. The first is our overall measure of citizen controversy over residential issues, discussed above. In the second column of results, the variable measuring the local crime rate is introduced (which unfortunately results in the loss of some observations because of missing data). The intuition behind the inclusion of crime rates, which is supported by the results, is that another major local issue, in this case public safety, may displace residential growth issues from a position of controversy. ____________ 4The variable measuring district council elections was insignificant when entered, and is omitted here. 124 Table A.7 Planner Assessments of Growth-Related Political Controversy (Ordered Logit) 125 Independent Variables Crime rate % Democrats % owner occupancy % Hispanic % black Commute time % unsewered units Residential stability City population change Job/worker ratio (log) Population (log) Bay Area Southern California, not Los Angeles County Central city Rural No. of cities Log likelihood Probability > chi-squared Pseudo R-squared Controversy over Residential Growth Model 1 Model 2 — –1.415 (3.15)*** 0.007 (0.42) 0.016 (0.93) –0.031 (1.98)** –0.037 (2.31)** –0.031 (3.38)*** –0.030 (3.13)*** –0.078 (2.65)*** –0.052 (1.82)* 0.115 (3.32)*** 0.134 (3.70)*** –0.003 (0.32) –0.006 (0.61) 0.031 (1.44) 0.020 (0.80) 0.006 (0.96) 0.011 (1.39) –0.391 (1.31) 0.109 (0.32) 0.140 (0.90) 0.160 (0.97) 1.086 (2.42)** 0.903 (1.97)* 0.328 (0.98) 1.035 (2.35)** 0.973 (2.16)** 274 –291.43 0.000 0.11 0.452 (1.31) 1.334 (2.93)*** 1.331 (2.83)*** 258 –267.83 0.000 0.13 Residential Issues Affect Elections Model 1 Model 2 — –1.482 (3.12)*** –0.012 (0.72) 0.002 (0.11) –0.010 (0.64) –0.022 (1.32) –0.030 (3.31)*** –0.031 (3.43)*** –0.072 (1.52) –0.056 (1.23) –0.012 (0.29) 0.013 (0.29) –0.015 (1.10) –0.016 (1.16) 0.009 (0.46) 0.014 (0.70) –0.004 (0.75) 0.004 (0.62) –0.698 (2.20)** –0.263 (0.76) 0.168 (0.96) 0.192 (1.02) 1.411 (3.44)*** 1.188 (2.83)*** 0.454 (1.43) 1.081 (2.31)** 0.694 (1.48) 257 –236.70 0.000 0.14 0.664 (1.97)** 1.473 (2.98)*** 1.072 (2.13)** 241 –216.03 0.000 0.16 Notes: *p < 0.1, **p < 0.05, ***p < 0.01. Cell entries are ordered logit coefficients, with absolute values of z-values listed in parentheses, calculated using robust standard errors. The next two columns of results in the table relate to planners’ responses to a question about the influence of residential development issues on city council or mayoral elections. There are three possible responses, ranging from growth issues “hardly ever” affecting elections to being “often influential” in affecting the outcomes of elections. Again, the crime rate variable is entered into the model in the fourth column. Finally, Table A.8 reports probit estimates relating to local antigrowth initiatives. The first column reflects the probability that planners answer in the affirmative to the following question: “In your city, have initiatives on the ballot been a major source of policies to slow Table A.8 Citizen Antigrowth Initiatives (Probit Model) Independent Variables Initiatives Have Been Good Chance Slow- a Major Source of Growth Initiative Antigrowth Policies Will Occur % Democrats % owner occupancy % Hispanic % black Commute time % unsewered units Residential stability City population change Job/worker ratio (log) Population (log) Bay Area Southern California, not Los Angeles County Central city Rural B0 No. of cities Log likelihood Probability > chi-squared % correctly predicted Pseudo R-squared –0.015 (1.05) –0.020 (1.43) –0.013 (1.76)* –0.053 (2.25)** 0.052 (1.45) –0.057 (2.08)** –0.001 (0.06) –0.010 (1.01) –0.496 (1.84)* 0.182 (1.19) 1.078 (3.13)*** 0.366 (1.20) 0.846 (2.24)** 0.610 (1.46) –2.082 (0.98) 270 –91.92 0.000 81.5 0.23 –0.007 (0.47) 0.003 (0.25) –0.017 (2.03)** –0.053 (1.62) 0.023 (0.74) –0.033 (2.10)** –0.024 (1.45) 0.010 (1.75)* –0.728 (2.43)** 0.105 (0.66) 0.911 (2.06)** 0.253 (0.70) 0.961 (2.29)** 0.866 (1.93)* –1.853 (0.81) 255 –73.57 0.002 87.8 0.22 Notes: *p < 0.1, **p < 0.05, ***p < 0.01. Cell entries are probit coefficients, with absolute values of z-values listed in parentheses, calculated using robust standard errors. 126 residential development? Don’t include referenda placed on the ballot by the council.” The second column relates to the question, “Is there a good chance that an initiative measure to slow residential development will occur in your city?” 127 Bibliography Aas, et al. v. The Superior Court of San Diego County, 2000 Cal. LEXIS 9048. Adang, Peter, “Snob Zoning: A Look at the Economic and Social Impact of Low Density Zoning,” Syracuse Law Review, Vol. 15, Spring 1964, pp. 507–537. Advisory Commission on Regulatory Barriers to Affordable Housing, “Not In My Back Yard”: Removing Barriers to Affordable Housing, Report to President Bush and Secretary Kemp, Department of Housing and Urban Development, Washington, D.C., 1991. Arax, Mark, “Putting the Brakes on Growth: Bypassed by the State’s Economic Rebound, the Central Valley Finds That Increased Development Just Brings More Municipal Debt,” Los Angeles Times, October 6, 1999. Babcock, Richard, and Fred Bosselman, Exclusionary Zoning: Land Use Regulation and Housing in the 1970s, Praeger, New York, 1973. Baldassare, Mark, “Regional Variations in Support for Regional Governance,” Urban Affairs Quarterly, Vol. 30, No. 2, 1994, pp. 275– 284. Baldassare, Mark, PPIC Statewide Survey: Special Survey on Growth, Public Policy Institute of California, San Francisco, California, May 2001. Baldassare, Mark, and William Protash, “Growth Controls, Population Growth, and Community Satisfaction,” American Sociological Review, Vol. 47, June, 1982, pp. 339–346. Barbour, Elisa, and Paul Lewis, “Development Priorities in California Cities: Results from a PPIC Survey,” Occasional Paper, Public Policy 129 Institute of California, San Francisco, California, 1998, available at http://www.ppic.org/publications/occasional/lewis.pdf. Bockman, Shel, Max Neiman, and Barbara Sirotnik, Riverside County Zones Report: The 2000–2001 Annual Survey, Inland Empire Research Consortium, Center for Social and Behavioral Science Research, University of California, Riverside, 2001. Bogart, William, “‘What Big Teeth You Have!’ Identifying the Motivations for Exclusionary Zoning,” Urban Studies, Vol. 30, No. 10, 1993, pp. 1669–1681. Bollick, Clint, “Subverting the American Dream: Government-Dictated ‘Smart Growth’ Is Unwise and Unconstitutional,” University of Pennsylvania Law Review, Vol. 148, No. 3, 2000, pp. 859–872. Brueckner, Jan, “Strategic Control of Growth in a System of Cities,” Journal of Political Economy, Vol. 57, 1995, pp. 393–416. Bureau of Transportation Statistics, 1990 Census Transportation Planning Package—California Statewide Element, data file, U.S. Department of Transportation, Washington, D.C., 1994. Calavita, Nico, and Kenneth Grimes, “Inclusionary Housing in California: The Experience of Two Decades,” Journal of the American Planning Association, Vol. 64, No. 2, 1998, pp. 150–169. California Budget Project, Locked Out! California’s Affordable Housing Crisis, California Budget Project, Sacramento, California, 2000. California Senate Office of Research, The Right Home in the Right Place at the Right Price, October 1999, available at http://www. sen.ca.gov/sor/housing/Righthome.html. California State Controller, Cities Annual Report: Fiscal Year 1997–98, Sacramento, California, 2000. Carman, Christopher, “Dimensions of Environmental Policy Support in the United States,” Social Science Quarterly, Vol. 79, No. 4, 1998, pp. 717–733. Center for the Continuing Study of the California Economy, Land Use and the California Economy, Palo Alto, California 1999. 130 Chapman, Jeffrey, “Proposition 13: Some Unintended Consequences,” Occasional Paper, Public Policy Institute of California, San Francisco, California, 1998. “Condo Scarcity: Construction Litigation Must Be Reined In,” San Diego Union-Tribune, December 10, 2000, p. G2. Curtin, Daniel, Jr., Curtin’s California Land Use and Planning Law, 20th ed., Solano Press Books, Point Arena, California, 2000. Danielson, Michael, The Politics of Exclusion, Columbia University Press, New York, 1976. DeLeon, Richard E., Left Coast City: Progressive Politics in San Francisco, University Press of Kansas, Lawrence, Kansas, 1992. Department of Housing and Community Development (State of California), Raising the Roof: California's Housing Development Projections and Constraints, 1997–2020, Department of Housing and Community Development, Sacramento, California, 2000a. Department of Housing and Community Development, Division of Housing Policy Development, New Funds Available for Housing, December 19, 2000b, available at http://www.hcd.ca.gov/hpd/fund_ hpd.html. Diaz, Daniel, and Gary Paul Green, “Fiscal Stress and Growth Management Effort in Wisconsin Cities, Villages, and Towns,” State & Local Government Review, Vol. 33, No. 1, 2001, pp. 7–22. Dickerson, Marla, and Stuart Silverstein, “New Crises Loom in State’s Aging Infrastructure,” Los Angeles Times, February 18, 2001, p. A1. Donovan, Todd, and Max Neiman, “Citizen Mobilization and the Adoption of Local Growth Control,” Western Political Quarterly, Vol. 45, No. 3, 1992, pp. 651–675. Donovan, Todd, and Max Neiman, “Local Growth Control Policy and Changes in Community Characteristics,” Social Science Quarterly, Vol. 76, No. 4, 1995, pp. 780–793. 131 Dowall, David, The Suburban Squeeze: Land Conversion and Regulation in the San Francisco Bay Area, University of California Press, Berkeley, California, 1984. Dowall, David E., and Jan Whittington, Making Room for the Future: Rebuilding California’s Infrastructure, Public Policy Institute of California, San Francisco, California, forthcoming. Dresch, Marla, and Steven Sheffrin, Who Pays for Development Fees and Exactions? Public Policy Institute of California, San Francisco, California, 1997. Dunstan, Roger, and Jennifer Swenson, Construction Defect Litigation and the Condominium Market, California Research Bureau, Sacramento, California, November 1999. Ewing, Reid, “Is Los Angeles-Style Sprawl Desirable?” Journal of the American Planning Association, Winter 1997, pp. 107–126. Farley, Reynolds, “Suburban Persistence,” American Sociological Review, Vol. 29, February 1964, pp. 38–47. Fischel, William, Do Growth Controls Matter? A Review of Empirical Evidence on the Effectiveness and Efficiency of Local Government Land Use Regulation, Lincoln Institute of Land Policy, Cambridge, Massachusetts, 1990. Frieden, Bernard J., The Environmental Protection Hustle, MIT Press, Cambridge, Massachusetts, 1979. Fulton, William, and Paul Shigley, “Election Analysis: Big Day for SlowGrowth Advocates,” California Planning & Development Report, November 2000. Fulton, William, Paul Shigley, Alicia Harrison, and Peter Sezzi, Trends in Local Land Use Ballot Measures, 1986–2000, Solimar Research Group, Inc., Ventura, California, 2000. Glickfeld, Madelyn, LeRoy Graymer, and Kerry Morrison, “Trends in Local Growth Control Ballot Measures in California,” UCLA Journal of Environmental Law and Policy, Vol. 6, No. 2, 1987, pp. 111–158. 132 Glickfeld, Madelyn, and Ned Levine, Regional Growth, Local Reaction: The Enactment and Effects of Local Growth Control and Management Measures in California, Lincoln Institute of Land Policy, Cambridge, Massachusetts, 1992. Gordon, Peter, and Harry Richardson, “Are Compact Cities a Desirable Planning Goal?” Journal of the American Planning Association, Winter 1997, pp. 95–106. Governor’s Office of Planning and Research, The California Planners’ Book of Lists, 2000, Governor’s Office of Planning and Research, Sacramento, California, 2000. Gruen, Claude, “Forced Donations Counterproductive,” Gruen Gruen & Associates, Trends, July 2001, available at http://www.ggassoc. com/trends.htm#trends22. Haar, Charles, Suburbs Under Siege: Race, Space, and Audacious Judges, Princeton University Press, Princeton, New Jersey, 1996. Hartnett, James, “Affordable Housing, Exclusionary Zoning, and American Apartheid,” New York University Law Review, Vol. 68, No. 1, 1993, pp. 89–135. Housing Crisis Task Force, City of Los Angeles, In Short Supply: Report of the Housing Crisis Task Force, report to the Los Angeles City Council, 2000, available at http://housingcrisisla.ucla.edu/. Kaniss, Phyliss, Making Local News, University of Chicago Press, Chicago, Illinois, 1991. Krainer, John, and Fred Furlong, “Tech Stocks and House Prices in California,” FRBSF Economic Letter, Federal Reserve Bank of San Francisco, Number 2000–27, September 15, 2000. Landis, John, “Do Growth Controls Work? A New Assessment,” Journal of the American Planning Association, Vol. 58, 1992, pp. 489–506. Landis, John, “Growth as Destiny: Understanding California’s Postwar Growth Patterns and Trends,” in Metropolitan Development Patterns: Annual Roundtable 2000, Lincoln Institute of Land Policy, Cambridge, Massachusetts, 2000. 133 Landis, John, and Cynthia Kroll, “The Southern California Growth War,” in John Kirlin & D. Winkler, eds., California Policy Choices, Vol. 5, University of Southern California, Los Angeles, California, 1989, pp. 123–163. Landis, John, Rolf Pendall, Robert Olshansky, and William Huang, Fixing CEQA: Options and Opportunities for Reforming the California Environmental Quality Act, California Policy Seminar, University of California, Berkeley, California, 1995. Legislative Analyst, State of California, Reconsidering AB 8: Exploring Alternative Ways to Allocate Property Taxes, Legislative Analyst’s Office, Sacramento, California, February 3, 2000. Levine, Ned, “The Effect of Local Growth Controls on Regional Housing Production and Population Redistribution in California,” Urban Studies, Vol. 36, No. 12, 1999, pp. 2047–2068. Lewis, Paul, “Looking Outward or Turning Inward? Motivations for Development Decisions in California Central Cities and Suburbs,” Urban Affairs Review, Vol. 36, No. 5, 2001, pp. 696–720. Lewis, Paul, “Offering Incentives for New Development: The Role of City Social Status, Politics, and Local Growth Experiences,” Journal of Urban Affairs, forthcoming. Lewis, Paul, and Elisa Barbour, California Cities and the Local Sales Tax, Public Policy Institute of California, San Francisco, California, 1999. Lewis, Paul, and Max Neiman, “Residential Development and Growth Control Policies: Survey Results from Cities in Three California Regions,” Occasional Paper, Public Policy Institute of California, San Francisco, California, 2000, available at http://www.ppic.org/ publications/occasional/lewis.neiman.pdf. Logan, John, and Harvey Molotch, Urban Fortunes: The Political Economy of Place, University of California Press, Berkeley, 1987. Logan, John, and Min Zhou, “Do Suburban Growth Controls Control Growth?” American Sociological Review, Vol. 54, June 1989, pp. 461– 471. 134 Mercer, Lloyd, and W. Douglas Morgan, “An Estimate of Residential Growth Controls’ Impact on House Prices,” in M. Bruce Johnson, ed., Resolving the Housing Crisis, Ballinger Publishing Company, Cambridge, Massachusetts, 1982, pp. 189–215. Myers, Phyllis, and Robert Puentes, “Growth at the Ballot Box: Electing the Shape of Communities in November 2000,” Discussion Paper, Brookings Institution, Center on Urban and Metropolitan Policy, Washington, D.C., 2001. Neiman, Max, “Zoning Policy, Income Clustering, and Suburban Change,” Social Science Quarterly, Vol. 61, 1980, pp. 666–675. Neiman, Max, Gregory Andranovich, and Kenneth Fernandez, Local Economic Development in Southern California’s Suburbs: 1990–1997, Public Policy Institute of California, San Francisco, 2000. Neiman, Max, and Ronald O. Loveridge, “Environmentalism and Local Growth Control: A Probe into the Class Bias Thesis,” Environment and Behavior, Vol. 13, 1981, pp. 759–772. O’Toole, Patrick, “California Tectonics: California’s Housing Industry,” Professional Builder, Vol. 65, No. 7, 2000, p. 58. Pendall, Rolf, “Opposition to Housing: NIMBY and Beyond,” Urban Affairs Review, Vol. 35, No. 1, 1999, pp. 112–136. Pendall, Rolf, “Local Land Use Regulation and the Chain of Exclusion,” Journal of the American Planning Association, Vol. 66, No. 2, 2000, pp. 125–142. Rojas, Aurelio, “Builders vs. Lawyers at Capitol,” Sacramento Bee, January 2, 2001, p. A1. Sanchez, Jesus, “Ruling by Court Unlikely to Spur Condo Construction,” Los Angeles Times, December 9, 2000, p. C1. Sanchez, Jesus, “Taking Up Residence in Office Space,” Los Angeles Times, March 13, 2001. Senate Local Government Committee, Tension and Ambiguity: A Legislative Guide to Recent Efforts to Reform California’s State-Local 135 Fiscal Relationship, California State Legislature, Sacramento, California, March 2001. Schneider, Mark, and Paul Teske, “The Antigrowth Entrepreneur: Challenging the ‘Equilibrium’ of the Growth Machine,” Journal of Politics, Vol. 55, 1993, pp. 720–736. Shen, Q., “Spatial Impacts of Locally Enacted Growth Controls: The San Francisco Bay Region in the 1980s,” Environment and Planning B, Vol. 23, No. 1, 1996, pp. 61–91. Southern California Studies Center, Sprawl Hits the Wall: Confronting the Realities of Metropolitan Los Angeles, University of Southern California, Los Angeles, California, 2001. Staley, Samuel, “Ballot-Box Zoning, Transaction Costs, and Urban Growth,” Journal of the American Planning Association, Vol. 67, No. 1, 2001, pp. 25–37. Stone, Clarence, Regime Politics: Governing Atlanta, 1946–1988, University Press of Kansas, Lawrence, Kansas, 1989. Taylor, Jerry, and Peter VanDoren, “‘Smart Growth’ Is Good for Everyone Else,” San Diego Union-Tribune, March 15, 2000, p. B-9. Wacker, Matthew, Alvin D. Sokolow, and Rachel Elkins, “County Right-to-Farm Ordinances in California: An Assessment of Impact and Effectiveness,” AIC Issues Brief, Agricultural Issues Center, University of California, Davis, California, May 2001. Warner, Kee, and Harvey Molotch, Growth Control: Inner Workings and External Effects, California Policy Seminar, University of California, Berkeley, California, 1992. Warner, Kee, and Harvey Molotch, Building Rules: How Local Controls Shape Community Environments and Economies, Westview Press, Boulder, Colorado, 2000. Williams, A. Kevin, The Long Wait: The Critical Shortage of Housing in California, Corporation for Supportive Housing and Housing California, Sacramento, California, 2000. 136 About the Authors PAUL G. LEWIS Paul Lewis is a research fellow at the Public Policy Institute of California, where he studies local government and urban policy issues. He has authored or co-authored publications on local government structure, local sales taxes, urban development, transportation policy, and regional governance, and has made presentations on such topics to a variety of governmental bodies and civic organizations. He holds an A.B. in political science from Indiana University and an M.A. and Ph.D. in politics from Princeton. MAX NEIMAN Max Neiman is Director of the Center for Social and Behavioral Sciences Research and Professor of Political Science at the University of California, Riverside. He is the author of five books and numerous articles on such topics as energy conservation, evaluating government performance, and the development of California’s initiative industry. His research interests include the politics of regulation in residential and economic development. He is a co-producer of the Annual Inland Empire Survey and the recipient of grants from the National Science Foundation and the John Randolph and Dora Haynes Foundation. He received his B.A. degree at the University of Illinois, Chicago, and his doctorate at the University of Wisconsin. 137 Other Related PPIC Publications Subsidizing Redevelopment in California Michael Dardia Who Pays for Development Fees and Exactions? Marla Dresch, Steven M. Sheffrin California Cities and the Local Sales Tax Paul G. Lewis, Elisa Barbour Local Economic Development in Southern California’s Suburbs: 1990–1997 Max Neiman, Gregory Andranovich, Kenneth Fernandez PPIC publications may be ordered by phone or from our website (800) 232-5343 [mainland U.S.] (415) 291-4400 [Canada, Hawaii, overseas] www.ppic.org 139" } ["___content":protected]=> string(102) "

R 102PLR

" ["_permalink":protected]=> string(121) "https://www.ppic.org/publication/cities-under-pressure-local-growth-controls-and-residential-development-policy/r_102plr/" ["_next":protected]=> array(0) { } ["_prev":protected]=> array(0) { } ["_css_class":protected]=> NULL ["id"]=> int(8158) ["ID"]=> int(8158) ["post_author"]=> string(1) "1" ["post_content"]=> string(0) "" ["post_date"]=> string(19) "2017-05-20 02:35:21" ["post_excerpt"]=> string(0) "" ["post_parent"]=> int(3286) ["post_status"]=> string(7) "inherit" ["post_title"]=> string(8) "R 102PLR" ["post_type"]=> string(10) "attachment" ["slug"]=> string(8) "r_102plr" ["__type":protected]=> NULL ["_wp_attached_file"]=> string(12) "R_102PLR.pdf" ["wpmf_size"]=> string(7) "1009450" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(302447) "Cities Under Pressure: Local Growth Controls and Residential Development Policy ••• Paul G. Lewis Max Neiman 2002 PUBLIC POLICY INSTITUTE OF CALIFORNIA Library of Congress Cataloging-in-Publication Data Lewis, Paul George, 1966– Cities under pressure : local growth controls and residential development policy / Paul G. Lewis, Max Neiman p. cm. Includes bibliographical references (p. ). ISBN 1-58213-040-X 1. Housing policy—California. 2. Housing development— Government policy—California. 3. Local government— California. I. Neiman, Max. II. Title. HD7303.C2 L49 2002 333.77'15'09794—dc21 2001048636 Copyright © 2002 by Public Policy Institute of California All rights reserved San Francisco, CA Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source and the above copyright notice is included. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. Foreword One of the most frustrating and contentious issues in California in recent years has been the shortage of housing. Even as 10 million people have been added to the state’s population over the last two decades, housing production has fallen short of demand, particularly during the recent period of economic growth. This shortfall has resulted in wildly escalated housing prices in many urban centers and heated markets in urban ring communities as far as 100 miles from employment centers. Many carefully crafted arguments have sought to explain this mismatch between supply and demand. They range from the inability of any state to absorb such huge numbers of new residents to the view that planners, conservationists, and community activists have made it devilishly difficult to build any new housing. Some researchers have also argued that state tax policies make residential growth economically unattractive for revenue-starved suburban communities. Yet another line of argument has focused on local growth controls, which are frequently placed near the top of the list of reasons for the shortage of affordable housing in California. According to this argument, a lack of enthusiasm for housing in California cities—and in some cases, outright hostility to multifamily or “affordable” developments—goes a long way toward explaining the state’s lagging housing production. Paul Lewis and Max Neiman address this argument head-on by studying which communities enact growth restrictions and why. They find that relatively few cities across the state have passed strict controls on housing development and that cities’ motives in managing residential growth do not appear to reflect simple elitism or civic selfishness. The authors conclude that growth controls may constrain some homebuilding but that broader market forces and state policies probably do more to explain California’s housing costs and slow rate of production. iii Why are these findings important? Over the last 30 years, California has seen the power of the public purse shift from local government to the state. From school finance to long-term debt for municipal facilities, the state government has played a more important role with each passing decade. As a result, local governments have sought to increase their revenue leverage with the state and to limit their exposure to the most costly types of development. In the last five years, various PPIC studies have examined the strategies and tactics cities have used in this regard. These studies have one consistent theme: Local governments are only marginally capable of influencing the total amount of growth in the state. Some shifting of growth and revenue probably occurs as a result of local policy; but for the most part, large retailers are located more or less where the companies want them, new industrial growth tends to cluster where similar activities exist, and new housing is built where the demand is greatest and large tracts of developable land still exist. Some local governments are more adept at managing growth than others, but their actions occur within a policy environment created largely at the state level. This most recent PPIC study confirms that general impression built over time from previous studies. Although the authors maintain that residential growth controls do not play a large role in the state’s housing problems, they do not downplay the problems that cities face in managing growth. Rather, they offer several precepts for managing conflicts over growth before they become permanent and unhelpful features of city politics. More important, perhaps, they urge state policymakers to consider ways to support local officials who are willing to accommodate growth but do not wish to burden their constituents or treasuries unduly. David W. Lyon President and CEO Public Policy Institute of California iv Summary California faces a number of daunting growth challenges. One of the most pressing is meeting the demand for housing for its increasingly diverse population. The state’s difficulties with housing production and affordability have received a great deal of attention since California entered a period of sustained and rapid economic and population growth in the mid-1990s. In examining the statewide challenge of residential policy, however, one is quickly confronted with the fact that many of the most relevant governmental decisions—regarding planning, zoning, permitting, or the siting of affordable housing projects, for instance—are made by local governments. Many critics of local regulation and advocates for the need to produce more housing point to local residential growth controls as a major cause of housing shortfalls. This report examines residential development policy in California municipalities (cities), where most of the state’s new growth is accommodated. It places particular emphasis on efforts by cities to control or manage the pace or form of new housing development. Which communities adopt growth controls, and why? Attention is also directed toward the local politics of residential growth, to the role of local conditions and controversies in affecting land-use policies, and to the overall posture of city governments toward accommodating new housing. The authors draw much of their evidence from a detailed mail questionnaire of city planning officials responsible for implementing local residential policies in the major regions of the state—including metropolitan Southern California, the San Francisco Bay Area, and the Central Valley—conducted in 1998–1999. Data from the survey, which drew a 76 percent response rate, are used in combination with a broad variety of community-level statistics drawn from the U.S. Census and state sources. v The Dilemma of Local Growth Policy On one level, it seems odd to suggest that California’s local governments are not doing enough to accommodate new housing and population growth. During the 1990s, growth swept all major regions of the state, and the state’s population grew by over 4 million. This increment alone is larger than the populations of 25 states. Nearly 95 percent of this growth was absorbed by municipalities. Cities have been able to accommodate the state’s increasing demographic and economic diversity, generally without becoming paralyzed by social or political crises at the local level. From another standpoint, some argue that local governments have not done enough to accommodate the pressures for new housing development in the state. California constitutes about 12 percent of the national population, but its share of the nation’s housing production fell from 14.7 percent in 1990 to 8.3 percent in 1999. Furthermore, in comparison with other parts of the country, the state’s housing growth has not kept pace with its job creation. Multifamily housing, in particular, declined precipitously after the late 1980s. Median home prices and rents escalated in the recent economic expansion, and affordability rates fell. According to one view, then, cities’ lack of enthusiasm for housing— and in some cases, outright hostility to multifamily or “affordable” developments—go a long way toward explaining California’s lagging housing production. Critics see local residential growth controls and “ballot-box planning” as ubiquitous and accuse local governments of having both the goal and effect of stifling residential development. Actual research on this point has been less persuasive, however, as investigators have come to mixed results about whether growth controls have been effective in restricting development. Furthermore, few researchers have examined motivations for local growth management, outside of limited and unrepresentative case studies. It may be more reasonable to regard city government as an arena for the attempted balancing of pro- and antigrowth interests active in local politics, with outcomes that vary widely among cities depending on various local conditions and the capacity of the locality to accommodate more growth. vi Local Politics and the Adoption of GrowthManagement Policies The 1998–1999 mail survey of 297 city planning officials indicates that there is considerable variety among California cities regarding residential policies. Respondents were asked about 16 types of local policies that seek to influence either the pace, shape, or form of new housing. As Table S.1 indicates, adoption of most of these policies has been the exception rather than the rule. Furthermore, 76 percent of cities employed three or fewer policies, of the 16 policy tools examined. Cities were most likely to have adopted policies to “shape” residential development—such as design review requirements or affordable housing set-asides—that some see as relatively innocuous from the standpoint of housing production. Policies designed to overtly restrict the amount of housing developed in a city, or to condition housing construction on infrastructure capacity, were less common. There are important regional differences, however, with San Francisco Bay Area municipalities significantly more likely to adopt Table S.1 Adoption of Residential Growth-Management Policies by Cities in California’s Three Major Regions Policy Design review standards Projects must include affordable housing component City has experienced a moratorium Encourage growth in built-up areas only Satisfy traffic standards before allowing development Use capital improvements to control rate or location of growth Official population ceiling Annual limit on housing units constructed Annual limit on building permits Restrict growth to built-up areas only Formula for allowable annual growth Ranking of proposed residential projects Annual limit on multifamily dwellings Annual limit on water connections Popular vote required for sewer capacity increase Recent reduction in residential zoning Source: Lewis and Neiman (2000). % of Cities with Policy 83 31 30 28 27 14 13 9 6 6 5 5 4 4 3 2 vii growth-management policies. The typical Bay Area city adopted 3.5 policies out of the 16, compared to 2.5 among Central Valley cities, 2.1 in Los Angeles County cities, and 2.4 among other Southern California cities. These regional differences do not seem to originate from greater opposition to growth among city officials in the Bay Area. In fact, more than two-thirds of the planners responding to the survey from each region characterized their city council majority as either encouraging residential growth or taking a neutral attitude toward it. And even in the Bay Area, a large majority of respondents indicate that their city’s review process for residential development has either stayed the same or gotten shorter in the last five years. Rather, grassroots-level citizen opposition to growth appears to be much higher in the Bay Area. Half of the respondents from that region report that residential growth issues are often or almost always controversial in their city, compared to about onequarter of respondents in Los Angeles County and the rest of Southern California, and one-eighth in the Central Valley. Bay Area respondents are also much more likely to report that growth issues have been influential in affecting the outcomes of mayoral or council elections. Local citizen unease is important because it is strongly linked to the number of growth-control policies in the city. This relationship is also visible in regard to local voter initiatives. Although only 16 percent of respondents indicate that voter initiatives have been a major source of policies to slow residential development in the city, the level of growthregulating activity is considerably higher in jurisdictions that have had, or expect to have, growth-related initiatives (Figure S.1). In short, local officials appear to formulate residential policies, in part, in reaction to or anticipation of residents’ opposition to new development. Growth controls appear to emerge in large part because of the dynamics of local politics rather than dispassionate analysis of local or regional housing trends or environmental conditions. Nevertheless, pitched battles over residential development seem to be limited to a relatively small proportion of California cities, whereas growth is a more routine process for most communities. viii Average number of policies 5 Yes 4 No 3 2 1 0 Had citizen growth initiative? Expect citizen growth initiative? Initiative status Figure S.1—Citizen Initiatives and City Growth-Management Policies The fact that cities in the region that grew at the slowest rate—the Bay Area—tend to adopt the highest number of growth-management policies indicates that the sheer rate of growth itself is not necessarily the best determinant of local residential policy. A series of statistical analyses, taking account of various city characteristics, show a more complex relationship between local conditions and residential policies. Taken as a whole, the results indicate that local growth controls and residential policies are generally motivated by something more than simply the self-interest of existing homeowners. Rather, day-to-day local conditions in a given city, such as long commute times or an excess of housing to local jobs, seem to motivate citizen opposition to residential growth. In relatively established or “mature” local communities without other overriding issues (such as unemployment or crime), such citizen opposition to growth contributes to the passage of policies that attempt to increase public oversight and control over the rate and character of housing development. ix The Broader Context of Local Residential Policies There seems little doubt that housing supply has not kept pace with the vast levels of population increase in California. However, how much of the supply problem can be attributed to the actions of localities is an exceptionally difficult matter to resolve. Although local growthmanagement policies draw media attention and are often singled out for criticism, most cities have few such policies—and particularly, few overt limitations on housing development. Indeed, many localities welcome residential growth, and many community officials resist serious controls on residential development. Consequently, the main effects of local residential development policies might be to redistribute the location of housing development among communities within a housing market. Overall, the macro-level effects of growth-management policies on the state's housing might be less important than is believed, and it thus may be premature to impose via state action constraints on how localities may regulate and manage residential development. The authors suggest that the debate should be refocused away from local growth controls, toward factors that are perhaps less obvious in California’s development process. Other factors likely play a larger role than local growth-management policies in contributing to California’s housing policy challenges. Some factors appear to limit housing supply, including the initial low-density zoning of some communities, restrictive building codes, changes in the federal tax treatment of commercial real estate, and litigation and liability insurance problems facing condominium builders. Housing advocates would do well to focus as much attention on these “invisible policies” as on local growth-control policies. At the same time, demand for housing in many areas has been particularly intense in recent years, because of rapid population growth, the continuing desirability of the coastal areas of the state, and the wealth gains—particularly capital gains—that have occurred for many California households. Furthermore, statewide institutions and policies merit attention. Under California’s system of public finance, cities may find that accepting housing works against their fiscal self-interest. The California Environmental Quality Act and the initiative process are additional statewide factors that can provide procedural tools for housing opponents x and introduce uncertainty both for builders and local policymakers. Meanwhile, the state has fallen behind in providing and maintaining its infrastructure, including transportation facilities, which means that growth is likely to create additional strains and controversies in many communities. Looking at the factors arrayed against housing production, local growth management appears to be a relatively small part of the picture, although further research is necessary to clarify its role. What does this mean for state policymakers? The finding that long commute times make communities less amenable to housing construction provides a strong hint that California must improve its infrastructure deficiencies if it is to convince its residents and local officials that new housing will not reduce existing residents’ quality of life. The state’s system for financing local governments is similarly a central fact of life for city policymakers, as it defines the costs and rewards of various types of growth and development. Most important, to avoid potentially paralyzing controversies over growth—and the punitive antigrowth ballot initiatives that sometimes result from such controversies—cities and counties need to find ways to manage community conflict before it erupts. Local governments must convince their residents that they do have sound plans to accommodate future growth sensitively. Meanwhile, state policymakers might consider policies directed at increasing the amount of information available to those in the residential development process and addressing some of the unpredictability of that process. xi Contents Foreword ..................................... Summary..................................... Figures ...................................... Tables ....................................... Acknowledgments ............................... iii v xvii xix xxi 1. EXAMINING CITY GROWTH POLICIES........... Why Study Local Growth Policies?.................. The Focus: Cities and Growth-Management Policy....... What Evidence Do We Use?...................... Summing Up ............................... 2. CALIFORNIA CITIES AND RESIDENTIAL DEVELOPMENT ............................ Two Sides of the Local Growth Dilemma ............. Cities Awash in Growth ....................... The California Housing Crunch .................. Two Perspectives on Local Governments and the Housing Market ................................ The Indictment: Cities as Housing Villains .......... The Indictment’s Hidden Assumptions ............. The Defense: Cities Managing Cross-Pressures ........ Research on the Effects of Growth Management on Housing Production and Cost ....................... Summing Up ............................... 3. PATTERNS OF GROWTH-MANAGEMENT POLICY ADOPTION ............................... Three Approaches to Managing Growth .............. Restricting Residential Growth................... Linking Growth to Infrastructure ................. Shaping Housing Development .................. Overall Levels of Growth-Management Activity ......... Summing Up ............................... 1 2 5 6 10 11 11 11 13 15 15 17 19 21 25 27 27 28 30 32 35 38 xiii 4. POLITICAL CONTROVERSY AND GROWTH MANAGEMENT ............................ Is Residential Growth Always Controversial? ........... What Role Do Local Citizen Initiatives Play in Growth Management? ............................ What Is the Attitude of City Councils Toward Residential Growth? ............................... Summing Up ............................... Interpreting the Evidence Thus Far ................. 5. FACTORS SHAPING CITIES’ WILLINGNESS TO ACCEPT RESIDENTIAL GROWTH............... Considering Potential Explanatory Factors ............. Local Political Factors ........................ Demographic Makeup of the City ................ City Size, Urbanization, and Location .............. Local Conditions Relating to Growth .............. Controversy and Growth Management: A Two-Way Street ................................. Explaining Housing and Growth-Control Policy Outcomes .. Explaining City Government Orientations Toward Housing and Growth ............................. Explaining Levels of Public Controversy over Residential Development ............................ Summing Up ............................... 6. THE BROADER CONTEXT OF LOCAL RESIDENTIAL POLICIES ................................. Is Local Growth Management at the Root of the Housing Problem? ............................... The Timing of California’s Housing Production Slump .... Less Visible Factors Retarding Housing Supply .......... Initial Zoning ............................. Building Codes............................. Tax Treatment ............................. Construction Defect Litigation................... Factors Fueling Housing Demand and Prices in California .. Desirability of Coastal Areas .................... The Capital Gains Boom ...................... 41 41 44 47 49 49 51 51 51 53 55 56 57 58 63 66 69 73 73 76 77 78 79 80 80 81 81 82 xiv Residential Development and California’s Local Fiscal System ................................ Other “Macro” State Policies Affecting Local Decisionmaking .......................... Environmental Review ........................ The Initiative Process......................... State Infrastructure Shortfalls .................... A Broader Policy Dilemma ....................... Summing Up ............................... 83 84 84 85 86 87 88 7. REFOCUSING THE GROWTH DEBATE ........... 91 Why Do Conflicts over Local Growth Management Have Such Visibility? ........................... 93 The Challenge for Cities: Successfully Managing Conflicts over Growth............................. 94 Two State Policy Challenges: Local Government Finance and Infrastructure Shortfalls ................... 95 Policy Options to Put Housing on a More Equal Footing ... 98 Full “Planning Disclosure” for Homebuyers .......... 99 Land Supply Monitoring ...................... 101 Addressing Condominium Litigation............... 101 CEQA Streamlining and One-Stop Permitting ........ 102 Local Government Finance Reform ................ 103 More Support for Urban Infill Projects ............. 104 Density Bonuses, Inclusionary Housing Requirements, and the “Builder’s Remedy”...................... 105 Housing Action Coalitions ..................... 108 A Final Thought ............................. 110 Appendix: Multivariate Data Analyses: Variables and Results ... 111 Bibliography .................................. 129 About the Authors ............................... 137 Other Related PPIC Publications ..................... 139 xv Figures S.1. Citizen Initiatives and City Growth-Management Policies................................. 3.1. Frequency Distribution of the Number of Residential Growth-Management Policies Adopted ............ 3.2. Average Number of Growth-Management Policies, by Region ................................. 4.1. Number of Growth-Management Policies, by Level of Controversy ............................. 4.2. Citizen Initiatives and City Growth-Management Policies................................. 4.3. Reported Attitude of City Council Majority Toward Residential Growth ......................... 6.1. California’s Housing and Economic Performance, Relative to That of the Nation .................. ix 36 37 44 46 48 76 xvii Tables S.1. Adoption of Residential Growth-Management Policies by Cities in California’s Three Major Regions ........ vii 3.1. Use of Restrictive Residential Policies ............. 29 3.2. Use of Policies Linking Housing Development to Infrastructure............................. 31 3.3. Use of Policies to Shape Residential Growth ......... 34 3.4. Comparing “Nonadopters” of Growth-Management Policies to “Heavy Adopters” ................... 36 4.1. Prevalence of Controversy Regarding Residential Growth Issues .................................. 42 4.2. Effects of Growth Issues on Local Elections ......... 43 4.3. Prevalence of Citizen Antigrowth Initiatives ......... 45 A.1. Descriptive Statistics for Variables in the Multivariate Models................................. 115 A.2. Number of Growth Management Policies Adopted (Regression Model) ......................... 118 A.3. Planner Assessments of City Policies (Ordered Logit) ... 119 A.4. Adoption of Affordable Housing Requirement (Probit Model)................................. 120 A.5. Planner Assessments of City Government Orientations to Residential Growth (Ordered Logit) ............ 122 A.6. City Manager Ratings of Desirability of Multifamily Development (Regression Model) ................ 123 A.7. Planner Assessments of Growth-Related Political Controversy (Ordered Logit) ................... 125 A.8. Citizen Antigrowth Initiatives (Probit Model) ........ 126 xix Acknowledgments This project would not have been possible without the efforts of hundreds of city planning directors, city managers, and other officials who responded to our questionnaires. We thank them for their time and candor. Very useful and thorough reviews of the manuscript came from Michael Dardia, Steven Erie, Mark Schneider, and Fred Silva. Mark Baldassare provided helpful guidance throughout the project. Hugh Louch, Elisa Barbour, Lisa Pitch, and Maria Anna Gonzales assisted in the research in a variety of ways. Peter Richardson provided editorial expertise. Although all of these people improved the final product, the authors are solely responsible for any errors of fact or interpretation. xxi 1. Examining City Growth Policies As California entered the 2000s, the challenge of local growth reappeared as an issue of the moment. By one estimate, the November 2000 general election in California featured 60 measures on city or county ballots that attempted to change local growth and land-use policies. Many of these ballot questions proposed to regulate particular developments or to slow growth, whereas others could be classified as progrowth. In about two-thirds of these measures, voters decided in favor of the slow-growth option (Fulton and Shigley, 2000).1 Historical data on local ballot measures indicate that 2000 was only the most recent peak for local opposition to growth. The number of such ballot measures tends to wane during periods of recession and shoot upward during periods when, as in recent years, population and economic growth have been rapid (Fulton et al., 2000; Glickfeld, Graymer, and Morrison, 1987). Local voter initiatives, which are the most visible manifestation of growth-management policy in California, receive a great deal of media attention. However, they are not the most common or typical method of affecting the amount, pace, or form of local land development. The vast majority of local land-use changes proceed under little scrutiny from the voting public; they are regulated by city ordinances, plans, and review procedures that are less visible to casual observers. Because local governments in California have a substantial degree of discretion in regulating land use and shaping development, serious attention should be given to their role in affecting the production of new ____________ 1A national survey indicated that there were 553 growth-related state and local ballot measures throughout the United States on the November 2000 ballot (Myers and Puentes, 2001). However, that survey defined “growth-related” very broadly to include, for example, parks and recreation bonds, transportation, and school construction issues. 1 housing and other development in the state. In particular, with many observers believing that the state is in a crisis of housing supply and affordability, the role of local governments in allegedly constraining and slowing the approval of proposals for new housing has come under renewed scrutiny. Although numerous journalistic accounts and case studies of local housing policy have been written, there have been relatively few large-sample, empirical analyses of local policies toward housing. In particular, there is a need to set local residential policies, and indeed the state’s housing production issue, in a broader context, and to suggest explanations as to why some city governments take a fairly restrictive stance on new residential development, whereas others are very accommodating. In this report, we take up this challenge, studying the nature and extent of growth-management policies among municipal (city) governments in three major economic regions of California: Southern California, the San Francisco Bay Area, and the greater Central Valley. Why Study Local Growth Policies? Anyone who wants to study growth policy statewide is immediately confronted with a problem: For the most part, local governments make development decisions. In the absence of a clear state policy regarding growth, California’s posture toward residential development is largely an amalgamation of the separate policies of its 475 cities and 58 counties. Through the so-called “police power” delegated to them by the state’s constitution, these local governments have considerable discretion over land use, in pursuit of the overall health, safety, and welfare of their residents. Subject only to conformity with state and federal law and constitutional principles, local governments’ discretion under the police power has been held by the state courts to be as broad as that of the state legislature itself. Land-use regulation may be exercised under the police power not only to protect against obvious public risks (for example, a smoky factory in a residential area) but also in pursuit of such relatively intangible qualities as community character and aesthetic improvement (see Curtin, 2000, pp. 1–4). Furthermore, California legal traditions have given a broad grant of authority to local governments under the 2 police power. For example, the California courts consider development to be a privilege of landowners, whereas in some other states it is considered a right (Curtin, 2000, pp. 228–229).2 Thus, cities and counties have broad authority over land development, including allowing new housing. Although all local general plans must heed a fairly elaborate state law regarding the preparation of the “housing element,” the housing goals and policies announced in these plans are hardly self-implementing. Specific local decisions by planning commissions and local councils typically are necessary to approve new housing subdivisions, even if they appear to conform with the housing element—meaning that local discretion remains a key factor in housing construction. In short, California’s planning and development laws “give local governments wide latitude regarding how they wish to structure the regulatory process” (Department of Housing and Community Development, 2000a, p. 78). Currently, state officials and various interest groups are very concerned about local growth policies, in large part because of the state’s lagging housing production and high housing costs, which have been viewed with alarm in several recent policy reports (California Senate Office of Research, 1999; Department of Housing and Community Development, 2000a; California Budget Project, 2000; Housing Crisis Task Force, 2000; Williams, 2000). Some fear that the recent run-up in California housing prices and the lack of housing production are serious threats to the state’s economic expansion and prosperity (Center for the Continuing Study of the California Economy, 1999). Industry groups and civic associations have mentioned housing costs and availability in such areas as the Santa Clara Valley as impediments to firm location and expansion. Because of fears that local government actions are slowing or preventing housing production, there has been a call for reform in ____________ 2Landowners are generally considered to have a right to develop in California only if they have an approved development agreement with the city or county or an approved vesting tentative map. Local government changes in land-use regulations, such as general plan amendments and citizen initiatives, can sometimes interrupt even developments that have already been approved. The California Supreme Court has held that to have a vested right to complete construction, a landowner must have a building permit, performed substantial work, and incurred liabilities in accordance with the permit (Curtin, 2000, p. 177). 3 everything from permitting procedures to environmental regulations to local government structure. Local policy merits attention also because it has been at the center of recurring waves of antigrowth activism by local residents and interest groups in various parts of the state. As noted, the fairly large number of local voter initiatives on growth indicates this arousal of controversy. Below the level of ballot wars, local controversies simmer in various political arenas, including mayoral and city council campaigns, planning commission hearings, neighborhood-level conflicts over rapid change, and debates over city policy changes on such topics as general plans, housing elements, housing subsidies, and annexation. At a broader level of public opinion, six in ten adults responding to a recent statewide survey felt that their own community had been growing rapidly in the past few years, and 53 percent rated the performance of their city government in handling growth issues as only “fair” or “poor.” A slight majority said that they would favor a (hypothetical) local initiative to slow development in their community, even if it meant having less economic growth (Baldassare, 2001). Local growth policies are also important by virtue of their close relationship to other major policy debates. For example, the debate over suburban sprawl and “smart growth” relates in part to the ability and willingness of older communities, which are usually closer to the traditional urban cores, to accommodate additional housing. If newer, low-density localities at the edge of the metropolis favor housing development more than older cities do, then urban development is likely to be more decentralized and auto-dependent. Such development may also threaten metropolitan areas with a loss of open space and farmland, although there are spirited debates about the nature and extent of these problems (Gordon and Richardson, 1997; Ewing, 1997). Finally, local growth policies are related to California’s continuing debate over its system of public finance. Numerous commentators argue that the state’s system of local taxation and revenue-raising creates poor incentives for balanced land development (Chapman, 1998; Center for the Continuing Study of the California Economy, 1999). 4 The Focus: Cities and Growth-Management Policy Our focus in this study is on city governments and their actions toward residential development. We concentrate on cities for two reasons: their preeminence in regulating residential development and the availability of more useful data for cities than for county governments. Through annexation (spreading out), infill (filling in), or redevelopment (replacing old land uses with new ones), cities accommodate the vast majority of new growth in the state. Of California’s population increase between 1990 and 2000, 94.6 percent occurred in cities.3 In unincorporated areas (those lands outside city boundaries), county governments control land use and development policies. Some development takes place initially in county territory and is later annexed by a city, or incorporates as a city. Moreover, although many counties have a stated policy to direct new urban growth toward existing cities, other counties, such as Sacramento, have actively competed with their cities for new growth. Thus, counties in many parts of the state are also important arenas of residential growth politics. Unfortunately for purposes of analysis, counties present problems. The detailed Census and other data we use later in this report are available for municipalities (cities) and for counties as entire units but not for specific unincorporated portions of counties. Many counties experience growth pressures in several different portions of their territory, and development goals and actions may differ widely in these different areas. Thus, we concentrate on cities because of their predominant role and their coherence as units of analysis, and our empirical conclusions relate only to cities. Much of our focus is on cities’ growth-management policies, the most restrictive of which are sometimes called growth controls. We define local growth management as a class of land-use policies and planning ____________ 3A comparison of city population figures from the 2000 Census to the 1990 Census shows that 82.4 percent of population growth was accommodated by cities that existed as of 1990, and an additional 12.2 percent by cities that incorporated during the 1990s. Only 5.4 percent of population growth was in unincorporated areas. Obviously, however, the newly incorporated cities previously were unincorporated areas, and similarly, some new growth that occurred initially in unincorporated areas was subsequently annexed into cities. 5 guidelines that have been developed for use by local governments in the period since about 1970. Ranging beyond traditional zoning or general plans—although they may be incorporated as part of a general plan— growth-management policies are geared at regulating the pace, form, or location of development within the community. Adopted by local governments (or through local voter initiatives), the policies range from growth boundaries to requirements that traffic standards be met before housing is built to annual caps on the number of building permits issued. For purposes of this report, we do not include under the category of growth-management policies such planning tools or regulations as zoning or general plans, which are more traditional frameworks for development decisions. Indeed, there is no feasible way to measure the comparative restrictiveness of local land-use zones.4 Nor do we include policy decisions relating to park or open-space acquisition, or fees or assessments on new development. Such policies are typically undertaken for reasons not directly geared at controlling growth per se (though they may have consequences for the pace, form, or location of growth). Statewide policies, such as environmental impact review under the California Environment Quality Act, or review by the state’s Coastal Commission, also do not fall under the rubric of local-option growth management devices. However, we will address some of these related policy issues in passing. What Evidence Do We Use? In the ensuing analysis, we draw heavily upon original surveys of top city officials regarding local development strategies and growthmanagement policies. Surveys allow us to ask important local decisionmakers from various parts of the state a range of questions, in a uniform fashion. They also allow us to quantify and compare city ____________ 4Not only are there no regularly updated local zoning maps for localities in California, but there are very difficult issues to resolve in assessing how “restrictive” local land-use zoning actually is. Even in a community with very high-density residential zoning, it might be the case that the large size of commercial or other land-use zones winds up limiting residential development. Zoning policies among new communities and new developments, moreover, tend to be associated with the original social conditions of these new places, often reflecting even the marketing considerations of the developers (Neiman, 1980). 6 policies. Another approach might have been extensive case studies of a large number of municipalities, but this would have proven prohibitive both in cost and in time, and still would not guarantee that the events occurring in the case study communities were broadly representative. In 1998 and 1999, we sent a mail survey regarding local residential policies to the planning directors (or others identified by the planning director as being most knowledgeable about local residential policies) of each city in three major regions of the state: Southern California, the San Francisco Bay Area, and the Central Valley.5 These regions, which together constitute 94.2 percent of California’s population, were defined very broadly for purposes of this study.6 Given the distinctive evolution of the inner Los Angeles area compared to the more diffuse and suburban counties surrounding it, Los Angeles County cities were broken out from the other areas of Southern California. The survey questionnaire asked respondents to indicate whether their city had enacted various local growth policies or techniques, ranging from design review standards to annual caps on the number of building permits issued. Other topics included the local process for reviewing development applications, which factors were most important in slowing residential development, local political conflicts over growth, and the perceived effects of local policies. We also asked for background information on the respondents. After contacting each intended respondent up to four times, we reached a response rate of 76 percent, gathering usable surveys from 297 cities.7 All of the pertinent results are tabulated and briefly discussed in a previous paper (Lewis and Neiman, ____________ 5The Southern California survey was conducted in late 1998, the other two regions in 1999. We do not expect the several-month gap between the surveys to create any significant problem in comparing the responses across regions. 6Southern California is defined to include seven counties: Los Angeles, Orange, Riverside, San Bernardino, San Diego, Santa Barbara, and Ventura, with a total population of approximately 20.1 million. The Bay Area includes nine counties: Alameda, Contra Costa, Marin, Napa, San Francisco, San Mateo, Santa Clara, Solano, and Sonoma, which collectively have 6.9 million residents. The Central Valley is defined as the following 18 counties: Butte, Colusa, Fresno, Glenn, Kern, Kings, Madera, Merced, Placer, Sacramento, San Joaquin, Shasta, Stanislaus, Sutter, Tehama, Tulare, Yolo, and Yuba. The Central Valley’s population is about 5.6 million. 7One additional city from Southern California responded to only a portion of the survey. It is included where possible in the analysis. 7 2000),8 but this report focuses in a more systematic and thematic way on the relevant survey results. To a lesser extent, we also draw upon another survey, entitled “Development Strategies in California Cities,” which was mailed in 1998 to the top administrative official of every city in California—generally the city manager or city administrator. Seventy percent of the cities then in existence in the state returned a usable questionnaire, meaning that 330 were received. This questionnaire was less detailed than the one discussed above, as it asked broadly about the desirability of various potential land uses (including single- and multifamily housing and various commercial and industrial categories) to each city’s administration, both in areas of new development on vacant land and in redevelopment zones. Other topics included an assessment of the importance of various factors (such as job creation, affordable housing provision, and general plan consistency) on city development strategies and some analogous questions about city strategies for annexation. A tabulation of all of the results is presented in Barbour and Lewis (1998).9 Although any survey carries with it some risk of measurement error, several factors affirm the validity of the data we have collected. As noted above, response rates were quite high, at 76 and 70 percent. The cities responding to the planner survey have a collective population of 19.4 million residents, and those responding to the city manager survey 22.9 million, according to population statistics from the 2000 Census. These numbers represent 57.3 percent and 67.6 percent of the statewide population, respectively. Furthermore, response rates were generally quite similar across regions and across city population-size categories, although very small jurisdictions responded at a lower rate. Another aspect of validity is whether responses to the survey questions can be taken as accurate representations of local policies. Given the potential sensitivity of several of the questions, respondents were promised anonymity, and we do not reveal the specific responses of ____________ 8The full text of the survey results is available at http://www.ppic.org/publications/ occasional/lewis.neiman.pdf. 9Other publications on various growth-related topics that have made use of these survey data include Lewis and Barbour (1999), Lewis (2001), and Lewis (forthcoming). 8 any individual cities. Therefore, there was little motivation for respondents to “fib” in answering the survey questions. As expected, the cities that scored high in the number of growth-management policies were often mentioned as strict growth regulators in the media or in other literature on the topic. More solid support for the validity of our survey data derives from our comparison of the results of our survey to that of an earlier survey of local growth management policies by Glickfeld and Levine (1992, survey taken in 1988). For the 248 cities that responded to both the Glickfeld and Levine survey and to our survey ten years later, the correlation in the total number of policies adopted was 0.34. We find this relatively solid degree of association across the two surveys to be a heartening validity measure, especially given the elapsed time, the differences in survey methodology, and the variety of policies that respondents were asked about. There is also a healthy correlation between responses to two somewhat comparable questions about local affordable housing policy, one of which appeared in our city manager survey and one of which in our planner survey.10 Finally, although the survey data are two to three years old—and therefore, some individual cities will have modified their policies—there is little reason to believe that the major differences among cities and among regions, which we will highlight in this report, have changed to any great extent. Other data on community characteristics are drawn from the Census—from 2000, where data are available, but otherwise from the 1990.11 We also use state data regarding local demographics, fiscal ____________ 10A binary item on the planner survey asked whether the city had a policy that new residential development must include an affordable housing component. A Likert-scale item on the city manager survey asked respondents to assess the importance of “meeting affordable housing needs” when assessing development and redevelopment proposals. The correlation between the two sets of responses, for cities responding to both surveys, is 0.25. Each variable is a positive and significant predictor of the other in a regression model, even when controlling for numerous other city characteristics. 11At this writing, only a portion of data on residents and housing units measured by the 2000 Census has been released for California cities. For our purposes, these include population counts by race or Hispanic origin and measures of homeownership and housing type. We must rely on the 1990 Census for a handful of other variables. However, this need not be a disabling problem, as community characteristics and relative differences tend to persist over time. For example, the percentage of whites, blacks, 9 characteristics, and political party registration.12 Finally, we draw somewhat on news coverage of local growth and housing issues around the state. Summing Up Local governments have a large degree of authority in the development of land for new housing, thus putting them in a pivotal position for shaping the nature and amount of housing production in the state. Cities, in particular, control the land-use policies relating to most new residential growth. California’s controversies over local antigrowth activism, housing affordability, sprawl, and other topics demand that serious attention be paid to local residential policies. In particular, we will highlight the extent of, and motivations for, city growthmanagement policies that attempt to regulate the pace, form, or location of new development. To do so, we rely heavily on recent surveys of planning directors and city managers, which drew solid response rates. Chapter 2 introduces the debate regarding cities’ role in the California housing market and is followed in Chapter 3 by a description of the amount and types of city growth-management policy in each region. Chapter 4 examines some of the political dimensions of growthmanagement policy adoption, looking in particular at the role of local political controversy, citizen initiatives, and the city council. In Chapter 5, we assess possible explanations as to why cities choose particular growth orientations, with some cities favoring further housing development and others acting to restrict growth. Chapter 6 sets these city policies, and the debate over housing production in general, within a broader context of statewide factors that constrain housing supply, increase housing costs, and diminish cities’ enthusiasm for residential development. Finally, Chapter 7 suggests a number of policy implications of our analysis and sets out some specific policy options worth consideration. ________________________________________________________ Asians, and Hispanics in the 456 California cities existing as of 1990 is correlated at very high levels (r > 0.9) with the shares of those groups in these cities as of 2000. 12In a few instances, we also refer to whether the city council is elected by districts or at-large. For this information, we draw upon a questionnaire sent by PPIC researchers in 2000 to all city clerks in the state. (The response rate on that question was 82 percent.) 10 2. California Cities and Residential Development Are local growth-management policies the villain in California’s housing woes? This chapter describes the state’s housing dilemmas and sets out city governments’ purported contribution to the problems of undersupply and affordability. In contrast to the sinister portrayal of cities’ roles sometimes discussed, we propose a more nuanced account that focuses on city governments as arenas for resolving difficult political cross-pressures regarding growth. Two Sides of the Local Growth Dilemma The character of the debate over residential development in California depends on one’s vantage point. From the standpoint of many cities and their residents, growth is a powerful wave washing toward their community providing opportunities and responsibilities, but also threatening the character and traditional scale of their community. From the standpoint of many state officials, business leaders, housing advocates, homebuilders, and would-be homeowners, housing development is a numbers game that the state is losing. Underproduction of new residential units, it is said, threatens the state’s vision of itself as a place of opportunity, where homeownership is a part of the California Dream. Others emphasize that rising housing costs for workers become reflected in the costs of doing business in the state, thereby undermining the state’s desirability as a place to conduct business. This section sets out the two sides of the growth dilemma. Cities Awash in Growth To some, it seems odd to suggest that the state’s local governments are not doing enough to accommodate new housing and population growth. According to the U.S. Census, California’s population grew by 11 4.1 million during the 1990s; this increment alone is a greater number of people than the populations of 25 states in 2000. Growth swept all major regions of the state, with both metropolitan Southern California and the San Francisco Bay Area increasing by 12.6 percent and the Central Valley by a prodigious 19.6 percent. Among California cities, the average rate of growth over the 1990s was 14.4 percent (15.8 percent in those cities responding to our growth management survey). Equally notable was the increased diversity of populations and land uses across communities. According to the 2000 Census, 55 percent of all the cities in the state had populations that were at least 20 percent Hispanic; a quarter of cities were at least 10 percent Asian. At the same time, jobs and commerce were also spreading beyond traditional centers. California’s “postsuburban” form is much remarked upon, as outlying centers join central cities as important employment centers. Although some have decried this loss of urban centrality, it is worth pointing out that the movement of jobs toward the suburbs goes a certain way toward improving the job/housing balance in the state’s metropolitan areas, bringing employment to where most workers live. According to commuting data from the 1990 Census, 52 suburbs had more than 1.19 jobs per resident worker, which was the average job/worker ratio in the state’s 49 Census-designated central cities.1 Our main points are that growth has continued to be rapid despite the deep recession in the early part of the 1990s, and that growth’s changes profoundly affected most corners of the state. Cities have been able to accommodate the state’s increasing demographic and economic diversity, generally without becoming paralyzed by social or political crises at the local level. Compared to protracted and bitter debate at the state level over issues such as immigration and affirmative action in the past decade, cities mainly appear to have been in the business of adapting to, planning for, and servicing new residents and businesses. When growth seems overly rapid, however, the “equilibrium” conditions of local politics and community life can be upset. Traffic ____________ 1Authors’ calculations from the Census Transportation Planning Package— California Element. (Data are available only for cities of at least 2,500 population.) Data for 2000 are not yet available. 12 levels may increase to the point that local roadways are overburdened. Local school systems may be forced into year-round sessions or the use of portable classrooms. Open spaces that are treasured by many members of the community may be lost. Rapid growth may fundamentally change the character of some communities. In some cases, the sheer pace of development itself may be threatening to some local residents, who are used to growth occurring, but at a much slower pace. The California Housing Crunch From another standpoint, local governments have not done enough to accommodate the pressures for new housing development in the state, and one must go no further than the numbers to be able to demonstrate this. For example, California’s share of the nation’s housing production fell precipitously for much of the decade. Even in the midst of fairly solid population growth in the 1990s, the state’s share of permitted residential units in the United States fell from 14.7 percent in 1990 to 8.3 percent in 1999. Although there are numerous reasons why this could be so, the decline seems at first blush to be fairly dramatic, and local government growth controls have been prominently mentioned as important contributors to the problem. According to this point of view, cities’ lack of enthusiasm for housing—and in some cases, outright hostility to new residential development—goes a long way toward explaining California’s lagging housing production.2 Other indicators of housing problems in the state are easy to find. The median price of an existing, detached, single-family home in California was $267,810 in July 2001—and $481,280 in the San Francisco Bay Area. The percentage of California households owning their own homes (57 percent as of the 2000 Census) remains significantly lower than the overall national rate (66 percent). More stark ____________ 2In reality (as we shall note again in Chapter 6), it is difficult to draw very profound conclusions from California’s lower share of national housing production. Household formation in California rose by 10.8 percent between 1990 and 2000, according to the 2000 Census—a lower rate than the 14.7 percent national increase in household formation. The United States added 1.01 net new housing units for every additional household, whereas California added 0.92 net new housing units per additional household. 13 is the difference between the so-called “affordability ratio” in California as opposed to the overall U.S. ratio. The affordability ratio is one measure analysts use to examine housing costs in relation to the ability of people to pay. It represents the proportion of households in an area that could afford to purchase the median-priced single-family home for sale at a given point in time, using normal assumptions about the percentage of income that can go toward home payments. The affordability ratio was 32 percent in California as of August 2001, compared to 54 percent in the nation at large.3 For large numbers of families, particularly in the younger age brackets and lower income levels, renting an apartment is a more relevant option than homeownership. Here too, however, California’s housing numbers are discouraging. In fact, multifamily housing was the sector of the residential market that deteriorated most rapidly in production in the 1990s (Landis, 2000, p. 33; California Budget Project, 2000, p. 38). As almost any renter in the Los Angeles region, San Diego, or the Bay Area would quickly confirm, rents rose very rapidly during the recent economic upswing and most notably in the coastal regions. Average household size—that is, the number of persons per dwelling unit—has been on the increase in California—from 2.75 in 1980 to 2.87 in 2000, according to Census data. Some attribute these increases in household size to declines in housing affordability (through doubling-up or taking in boarders), although most of California’s upswing in household size is likely attributable to the rising numbers of immigrants, who traditionally reside in units with larger numbers of residents. Another way planners and social scientists examine housing production is by comparing the number of new housing units in an area to the number of new jobs. In the United States as a whole in the late 1990s, approximately one new housing unit was being produced per two new jobs; in California (between 1994 and 1998), the ratio was one new housing unit per 3.9 new jobs. In some areas of rapid employment growth, such as the Silicon Valley, these ratios were far worse, with more than eight jobs for each new housing unit, by some estimates (California ____________ 3The data in this paragraph are from the website of the California Association of Realtors (www.car.org). 14 Budget Project, 2000, p. 39). These figures are one measure of an imbalance between jobs and housing. California went through a similar economic cycle in the 1980s, with a recession early in the decade and rapid growth in the latter portion of the decade. Nevertheless, the state’s ratio of net new jobs per net new housing units in that decade was a relatively healthy 1.6 (based on a net growth of 1.9 million dwelling units and 3.0 million jobs statewide in that decade, according to the Census). Clearly, something was different in the 1990s that did not allow housing production to meet the need to as great a degree.4 Local government actions have been viewed as one possible culprit. Two Perspectives on Local Governments and the Housing Market Assessing the role of local policies in California’s housing problems requires that we to move beyond the rhetoric that marks much of the debate and consider the full context of city actions and the nature of their policies. That is the major task of the rest of this report. Here we discuss one account that sees municipal governments as a major element of the housing production problem, followed by another perspective that views city governments as agents of compromise in a multisided political debate about the costs, benefits, and consequences of population growth. The Indictment: Cities as Housing Villains According to strong critics of local government growth-management policies, both scholars and journalists, local policies have a particularly harmful effect on housing opportunity for the average state resident. Perhaps most often cited of these critics is Bernard J. Frieden, whose 1979 book, The Environmental Protection Hustle, offered a stinging indictment of local growth controls that had begun to take shape in the 1970s. Frieden’s book and its broad-brush condemnation of local ____________ 4It is quite possible that there was overbuilding in residential real estate in the 1980s, particularly in Southern California, and therefore California’s low production in the 1990s might be partially attributed to its excess inventory. This argument is less persuasive for the Bay Area, however. (We thank Michael Dardia for making this observation.) 15 development controls continue to be cited approvingly by critics who see local growth policies as too stringent (Taylor and VanDoren, 2000). Even though Frieden’s study of Bay Area residential development restrictions is over 20 years old, his view reflects a long-standing interpretation of local development controls: that they are inspired by elitist, snobbish populations who disguise their exclusionary motivations with references to quality-of-life amenities and environmental protection. In this view, socially advantaged current residents are pulling up the ladder, so to speak, to prevent others from coming to their communities. Many studies of local housing policies, especially of suburban jurisdictions, have critiqued so-called snob zoning, especially large-lot and exclusionary zoning. Usually accompanied by elaborate building codes and subdivision regulations, such zoning has long been associated with elite suburban aversion to low-income and higher-density housing, particularly in the East and Midwest (Adang, 1964; Babcock and Bosselman, 1973). Indeed, by the time Frieden concluded that some communities were gaming environmental law to prevent housing for the less affluent, he was merely dramatizing and perpetuating this suspicion of local policies onto the California scene, where growth-management policies rather than zoning were at issue. Frieden’s work was an important intellectual watershed in the debate over the effects of local land-use and development policy on housing supply and prices, and on the geographical distribution of low-income and minority families, and it continues to be central to contemporary research and policy debates over zoning, land use, growth control, or “smart growth” (Bogart, 1993; Bollick, 2000; Hartnett, 1993; Shen, 1996). Because recent commentators have largely echoed the critical emphasis Frieden placed on local growth controls, his arguments are worth a closer examination. First, critics of local policy make the argument that growth controls are ubiquitous, an increasingly common feature of the local regulatory scene. Although Frieden’s case studies, and those of many others, tend to be drawn from Bay Area communities such as Marin County with famously finicky attitudes toward growth, growth controls are sometimes held to be sweeping the state and nation (see also Dowall, 1984). A second allegation is that local growth-management policies have the overt aim of stifling residential development. According to the critics, local 16 growth controls are often couched in terms of environmental goals, but these function largely as a smokescreen for baser motivations. Some of these critics allege that existing property owners mainly wish to limit new housing development, thereby inflating the prices of their own homes (or the rents of their apartments); in short, they influence local governments to pass restrictive policies to creative an artificial shortage of housing (Brueckner, 1995). Other critics argue that the motives behind growth management are more social than economic. From this point of view, restrictive local regulations on residential development—particularly those limiting multifamily or “affordable” housing construction—may have their basis in simple snobbery, or a fear of newcomers, perhaps with a touch of racism or ethnocentrism underlying the restrictions. Indeed, it is not difficult to find anecdotes and case studies that support the snobbery hypothesis (Frieden, 1979; Danielson, 1976). Finally, the condemnatory account of local growth controls holds that they are a major cause of housing shortfalls, seriously reducing the number of buildable lots in metropolitan communities. In short, because of their ubiquity and their growth-stifling intent, local government growth restrictions are said to shut the poor and middle class alike out of numerous communities, particularly in the prosperous suburbs. The Indictment’s Hidden Assumptions Several dubious assumptions underlie the indictment. Such assumptions include the following: • Growth-management policies are effective in slowing or restricting growth at the local level. The mere passage of a growthmanagement measure—either at the ballot box, as an ordinance by the city council, or as an administrative implementing procedure in the planning department—does not ensure that it will manage growth effectively. In some cases, such policies can be symbolic efforts that create few binding constraints on local development; in other jurisdictions, growth controls may be weakened over time or ignored by those responsible for implementation (Warner and Molotch, 2000; Logan and Zhou, 17 1989). Moreover, the threat of future controls might even stimulate current growth. In fact, as we will discuss below, evidence on the actual effects of local growth regulation is quite mixed. • No legitimate public policy motives underlie local growth management. Although the indictment focuses on the purported flimsiness of environmental protection as a reason for growth control, other defensible motivations are certainly possible. Cities may experience a decline in public services such as public schools or recreation because of rapid residential growth. Infrastructure, such as streets, school buildings, water, sewerage, and municipal utility systems, may be overburdened or at capacity. There may be an insufficient number of jobs available nearby for new residents, necessitating long commutes for residents. Finally, the community may be interested in taking a step back from the intensity of new growth to assess its plans for the future and consider how its amenities—including the natural environment—might best be preserved. Chapter 5 explores some of these motives in more depth. • Any local government restrictions that reduce the number of units built below what the builder has proposed results in a net reduction in the local housing supply. This assumption overlooks the “law of anticipated reactions”: If builders expect to have difficulty getting all their units approved, they are likely to propose more units than they otherwise would. Under such conditions, builders may be likely to propose a higher-than-optimal number of units in their new developments, knowing that this number is likely to be reduced. Furthermore, builders are not obligated to build the number of units approved, and they occasionally postpone or reduce construction for reasons having little to do with local regulations, such as housing market conditions or financing difficulties. • Builders have few choices if they cannot build in the more restrictive communities. If builders shy away from some communities because of restrictive political conditions, or if their housing proposals are rejected, does this leave them without options? 18 Not if there are other, comparable communities in the region available to accept similar projects. California contains hundreds of jurisdictions, presenting housing developers with a wide portfolio of options for proposing housing projects. The actions, or regulatory reputations, of individual cities may lead to housing production being “moved around” within a region, as opposed to a net reduction in the number of units produced. Of course, this may raise the likelihood of sprawl, or decentralized development patterns, if the growth-accepting jurisdictions are located farther from central job areas. But that is not the same result as thwarting housing production entirely.5 • Local governments are the agents of slow-growth interests or are willing participants in the restriction of housing. Are city governments merely puppets of antigrowth activists or homeowner cartels? Even a casual acquaintance with a handful of local governments makes this difficult to believe. Moreover, there is a body of academic literature on urban political economy that tends to conclude that rather than cultivating a slow-growth environment, a number of forces create a progrowth orientation among most local governments. These forces include the desire to enhance the importance of the locality through growth, and the urge to attract new businesses to provide local jobs and investments by developing the local customer base with population growth. One must also consider the political importance of major campaign contributors, which notably include real estate interests, local retailers, and other elements of the so-called “growth machine” (Logan and Molotch, 1987; Stone, 1989). The Defense: Cities Managing Cross-Pressures It is likely that local political arenas are more pluralistic than is typically admitted either by the progrowth indictment of local ____________ 5Moreover, it is not clear that an “unregulated” system of land use, in which all communities wanted and welcomed growth on the developers’ terms, would produce less “sprawl” than the current arrangement. 19 government or by the antigrowth critics’ portrayal of local policymakers’ “caving in” to developers at every turn. Rather, cities in many rapidgrowth regions are in the challenging position of feeling pressure from both sides. On the one hand, cities have the need, responsibility, and incentive to allow for new housing; on the other, there is frequently pressure from local residents and certain interest groups to prevent overly rapid growth, or specific housing projects, from potentially harming the community’s amenities and perceived quality of life. In a state as large as California, there are no doubt some jurisdictions in which either housing developers or antigrowth activists have been able to dominate the local political system. Perhaps the more typical situation, however, puts city elected and appointed officials in the position of attempting to manage the conflicts and controversies engendered by growth. Compromise, after all, is inherent in democratic politics. Antigrowth activism has certainly developed into a powerful grassroots movement in a variety of localities, as the number of restrictive local citizen initiatives on land-use topics indicates. There are also many communities, however, where growth management fails to reach the policy agenda, or is defeated. There are even a few localities in which growth is nearly always welcome and where accommodating housing is viewed as a duty. In our experience, most city officials hardly seem to be slow-growth extremists. The need for housing and the many equity problems that result from excessive regulation of local residential development are central features of housing officials’ professional training. Housing officials, moreover, are at the receiving end of much information regarding the “crisis” of housing. The need for housing, and housing affordability problems, furthermore, are experienced by local residents in jurisdictions throughout the state; many of these residents, and their sympathetic family members, also vote and become active in politics. With evidence accumulating rapidly from news media and policy reports that the state faces serious housing shortfalls, many mayors, councilmembers, city managers, and planners surely feel a responsibility to do what they can to accommodate what they believe is needed additional housing. Pressure from the business community to keep the local cost of living from escalating too far for their employees adds to the 20 pro-housing forces likely to affect local officials. In addition, one should not underestimate the influence of progrowth coalitions in many jurisdictions. Of course, local policymakers do not make decisions about housing and land use in a vacuum. There are constraints on what they can accomplish, based on the land area of the city and its capacity for siting new housing, the local and regional infrastructure, the system of public finance that funds services for new residents, and the nature of the housing market and the development industry in their area, among other factors. Later in the report, we will highlight the importance of many such contextual factors and show how they may be important influences on housing production in themselves. Research on the Effects of Growth Management on Housing Production and Cost Before taking a closer empirical look at the types of residential policies California cities employ, it is worth reviewing what is known about the effectiveness and effects of local growth-management policies. Although our mail survey on local policies is too recent to allow us to assess their effects on housing production or price, some studies have begun to explore this issue. Both critics and advocates of local growth controls frequently seem to assume that such policies are effective in holding down growth rates. Criticisms of such controls further maintain that such efforts to restrict supply necessarily lead to increased housing prices (Advisory Commission on Regulatory Barriers to Affordable Housing, 1991). However, our review of empirical studies does not allow any such confident and certain conclusions. If anything, the relevant literature may lean, on balance, toward finding that growth control lacks significant effects on population growth or housing prices—but the results are so mixed that one would be hard-pressed to offer this as a definitive conclusion. Below are some examples of major published studies—conducted by economists, urban planners, sociologists, and political scientists, and using a variety of analytic methods. 21 • An examination of 387 suburbs nationwide, focusing on environmental-protection-oriented growth restrictions, found “only modest effects on subsequent change in local population, median family income, median rent, and black percentage. We argue that formal policy tools or legislation cannot be accepted as indicating that their stated objectives will be realized” (Logan and Zhou, 1989, p. 461). • By contrast, a study of 63 municipalities in Ohio found “consistent and robust evidence that subjecting rezoning decisions to public referenda created a housing unit ‘growth penalty’ for cities” (Staley, 2001, p. 25). Note that this result focuses on the popular referendum requirement per se, not the broader class of growth-management devices. The argument is that referendum requirements introduce a large degree of uncertainty for developers, which hinders investment. • A study of 97 Northern California cities through the 1970s found that population “growth rates are not influenced by growth controls or the social variables associated with antigrowth policies” (Baldassare and Protash, 1982, p. 339). • A study of 490 California cities and counties found that local growth management policies “significantly displaced new construction, particularly rental housing, possibly exacerbating the expansion of the metropolitan areas into the interiors of the state.” However, not all growth controls are created equal: “Measures which limited available land or which downsized existing zoning had stronger effects” (Levine, 1999, p. 2047). • An earlier study by the same author and a colleague, however, concluded that “Growth measures do not appear to have reduced construction activity significantly at the state, the county, or the metropolitan level” in California (Glickfeld and Levine, 1992, p. xii). • An examination of the region of California with perhaps the strictest growth controls—the South Coast of Santa Barbara County, in the 1974–1979 period—concluded that a shortfall of housing production resulting from water hookup moratoriums 22 and zoning changes accounted for about one-quarter of the change in real house prices (Mercer and Morgan, 1982).6 • An examination of Southern California suburbs during the 1980s, which used many of the same measures of growthmanagement policies used in the current report, concluded that “Restrictive growth-controlling cities do not appear to become richer or poorer; they appear to become less black” (Donovan and Neiman, 1995, p. 790). No significant effects on local population growth were detected. • Warner and Molotch (2000, Appendix A) used a pooled crosssectional design to study 11 Southern California localities across the period from 1971 to 1990, looking for the effects of newly adopted growth controls (with a two-year lag) on annual percentage changes in housing units. Once proper statistical controls were introduced, they found “scant evidence that controls had much of an effect, particularly on the supply of new housing” (p. 52). • A particularly relevant analysis by Landis (1992) compared seven midsized California municipalities that had restrictive growth controls to six well-matched local governments without growth controls. Examined over a decade-long period, the author found that the local policies were “largely irrelevant to the management of urban growth”; there were no major differences between the growth-control cities and their matched pairs in terms of population growth rates, housing production shortfalls, or home price increases. Why were effects not detectable? Landis suggests that the control policies were “porous” and often fairly generous, that there were likely spillover opportunities for housing development nearby, and that price effects of local policies were likely overwhelmed by regionwide factors—a point we shall return to in Chapter 6. ____________ 6However, the study suffers from a case of extreme collinearity (r = 0.97) between the variable for housing-unit shortfalls and a measure of countywide employment, which may render its conclusions tenuous. (See Mercer and Morgan, 1982, p. 215, fn 12). 23 Overall, it seems safe to say that there is simply no consensus among the experts, as of yet, on the effects of local growth controls. Some critics of local growth management approvingly cite the sophisticated review of other studies by William Fischel (1990), entitled Do Growth Controls Matter? In the first paragraph of the review, Fischel states, “The answer to the title’s question is yes. . . . The effects are evident in land values and housing prices” (p. 1). However, most of the studies Fischel cites that lead him to this conclusion are studies of local zoning, not contemporary growth controls; he “decline[s] to make a sharp distinction between growth management and traditional zoning” because both derive from the local police power (p. 3 fn). In another context, he notes that the most exclusionary local governments do not typically adopt modern growth controls because “the elitist communities long ago adopted strict zoning regulations, so that additional growth controls are unnecessary” (p. 33 fn).7 Perhaps, then, it is zoning—a much older and more established variety of local land-use regulation— that should receive more scrutiny from policymakers, the media, and scholars. Even so, modern growth management, not zoning, draws most current-day attention to local residential policies in California. On this topic, Fischel is more equivocal in summarizing the results of econometric studies. He notes the lack of evidence in growth-control studies for decreased housing supply (p. 33). And although several studies have found growth management to be associated with significantly higher housing costs, he notes that such increased prices may simply be the result of the increased amenity levels brought about by growth control policies: We know . . . that community and neighborhood amenities raise housing values. Growth controls may create residential amenities, or, for rapidly growing communities, they may prevent impending disamenities. Hence the ____________ 7Logan and Zhou (1989, p. 464 fn) similarly find that suburbs in their sample that appear to exclude multifamily housing through zoning were less likely to adopt growth controls. 24 higher housing prices could be taken as evidence that growth controls do what they are intended to do (p. 33).8 On balance, Fischel is inclined to believe that growth controls probably do have more social costs than social benefits. However, he is tentative on this point “because only a few studies have addressed it in a persuasive framework” (p. 53). Clearly, there is a need for additional scholarship on local growth-management’s effects, especially in the California context. Before researchers proceed to that difficult task, however, it is important to characterize the actual extent of growth controls and to learn something of the motivations behind them. Those issues are the main focus of the current report. Summing Up California is once again growing very rapidly, as it has for nearly all of its history. The absolute increase in the number of residents can seem staggering—and no doubt is staggering to residents of many individual communities. Still, despite receiving millions of new residents and building hundreds of thousands of new housing units, California is widely thought to have systemic problems in building the amount of housing necessary to accommodate the state’s burgeoning population. The state’s housing production has been well below what one would expect, given the rapid increase in jobs and households. Do local growth-management policies, such as annual housing-unit caps, greenbelts, or moratoriums on development, play a major role in the state’s housing problems? One line of argument holds that they do, because growth controls allegedly are ubiquitous and very often have the overt aim of limiting housing production. We have suggested a number of possible shortcomings in this argument, because of its simplistic assumptions about the behavior of local governments and the nature of ____________ 8Similarly, early studies that found housing price increases resulting from strict controls in Petaluma and Davis are difficult to generalize from because of the potential uniqueness of those communities. “Petaluma was a famous test case, and Davis is a university town known for its attention to environmental amenities. Notoriety itself might have sent housing prices up by signaling to house buyers the existence of an exclusive community” (Fischel, 1990, pp. 31–32). 25 local residential policies. Moreover, the research literature on growth controls lacks any real consensus on the issue of whether such policies have much effect on housing production or prices. Instead, we have suggested that city politics is likely to be an arena for the balancing of pro- and antigrowth interests, with policy choices that vary widely among cities depending on local conditions and the capacity of the locality to accommodate more growth. The next step, then, is to examine the patterns of adoption of various types of local growth-management policies. 26 3. Patterns of GrowthManagement Policy Adoption This chapter describes three major categories of growth-regulating policies, and draws upon the results of our 1998–1999 survey of 297 city planning directors to report on the use of these policies by cities in the major regions of California. Because the aggregate data tend to mask major distinctions among different areas of the state, we view them by region. Three Approaches to Managing Growth In our survey, we asked city planners questions about cities’ adoption of 16 specific types of growth-management policies. Upon closer examination, these policies can be classified fairly readily into three major groupings, based on the overall purpose of the policy. First, some policies represent more or less overt attempts to restrict residential development—that is, to slow down its pace by reducing the number or types of residential units that can be developed. These are the most overt growth controls. The second group of policies includes those that link residential growth to the infrastructure capacity of the community—for example, by requiring that roads or other public facilities be deemed adequate to meet the demands of an increased population. The third and perhaps least-restrictive type of growthmanagement policy has the goal of shaping the form, style, or location of new housing within the community. The threefold categorization of growth-regulating policies is intuitively sensible and reflects the major substantive approaches to managing local growth. Although it is possible to design a number of alternative strategies to measure local policy (e.g., Glickfeld and Levine, 1992), the first test is to determine whether the simplest, most straightforward approach produces useful findings. 27 We did not ask overall questions about the city’s general plan or zoning. Although these are extremely important elements of the local regulatory apparatus, they are not normally considered under the rubric of “growth management.” All California cities and counties are required to have a general plan (including a housing element) and zoning consistent with that plan, whereas the growth-management policies we examine are adopted at the discretion of each community.1 Partially as a result, the policies we examine tend to be the most visible and controversial elements of local land-use regulation. However, we do ask respondents to express their views regarding the overall effect of their land-use policies on their communities, which includes zoning policy as well. As we shall discuss later in this report, initial zoning may be worthy of much closer scrutiny in future research. Restricting Residential Growth We now discuss cities’ use of each of the three major groups of growth-management policies. Under the category of “restrictions” on housing, we included the following seven city actions: • Annual limits on total building permits, • Annual limits on residential units authorized, • Annual limits on multifamily dwelling units built, • Recent substantial reductions in land zoned for residential use, • Policies linking local residential growth rates to some formula or external growth rate, such as that in the county, region, or state, • Formal policies prescribing a local population ceiling, and • Moratoriums on building permits, water connections, or sewer hook-ups. ____________ 1Under California law, specific ordinances and planning and subdivision regulations must, under nearly all circumstances, demonstrate “consistency” with the local general plan, which is held to be something of an overall constitution for local development decisionmaking. Therefore, as some growth-management policies are adopted, whether by the city council or in a local initiative election, they often contain language that modifies the general plan to ensure consistency. Although the state’s intention was to ensure that local policies were consistent with the local plan, in reality the plan may also be amended to reflect new policies. However, some cities have adopted general plans that show a strong growth-management flavor throughout. 28 Although some might choose to include the last policy on the list with the infrastructure linkage policies, to be discussed in the next section, such moratoriums share with the other policies listed above an overall character of overt restriction. Each policy either limits the number of units that may be produced or levels of population increase, or takes land for residential purposes out of circulation. How common are these most restrictive of growth-management policies? Our survey responses indicate that most of these policies are uncommon in all regions, as Table 3.1 indicates. Annual caps on building permits, residential units, or multifamily dwellings are employed by less than a tenth of all cities responding. Recent reductions in the amount of land zoned for residential use are even less common, and virtually nonexistent outside the San Francisco Bay Area. Somewhat more common are official population ceilings and temporary moratoriums on Table 3.1 Use of Restrictive Residential Policies Policy Annual limit on building permits Annual limit on housing units constructed Annual limit on multifamily dwellings Recent reduction in residential zoning Formula for allowable annual growth Official population ceiling City has had a moratorium Mean number of these policies used per city L. A. County (N = 56) 2 2 5 2 5 7 33 0.57 Percentage of Cities Using Policy Other So. S. F. Bay Central California Area Valley Total (N = 91) (N = 77) (N = 73) (N = 297) 4 10 5 6 9 18 7 9 1 57 4 1 50 2 2 84 5 16 12 12 13 30 32 25 30 0.65 0.90 0.60 0.69 Notes: Percentages represent the number of respondents indicating use of the policy as a proportion of all those providing an answer to the question, including “Don’t know” or “N/A” responses. For more detailed tabulation, see Lewis and Neiman (2000, pp. 5–6). 29 growth. The former exists in about one in eight cities, and the latter in about three in ten cities.2 Although moratoria and population ceilings can sound somewhat severe, their actual long-term effect on housing production is uncertain. Some cities have established generous or optimistic growth ceilings, and such an overall “population cap” may prove more symbolic than binding in the long run. As for moratoria, they are by definition temporary, and typically must be based on some infrastructure or planning deficiency that needs to be overcome (often depleted sewer capacity), at which point residential permits will be processed again. Our question did not ask how recently the moratorium had occurred in the city.3 Four of these seven restrictive policies are most commonly adopted in the Bay Area. As the last line in the table indicates, the average Bay Area city has about 0.9 restrictive policies of the seven possible, compared to about 0.6 in the other three areas. Bay Area communities are substantially more likely to adopt some overtly restrictive policies, such as annual limits on building permits and housing construction. Nevertheless, across cities in all the regions studied, 166 (55.9 percent) indicated that they employed none of these policies; only 40 (13.5 percent) have more than one such policy. Linking Growth to Infrastructure Under the category of “linking” growth-management policies are the following four city requirements: • Policies that require traffic standards to be satisfied before new development occurs, ____________ 2Logan and Zhou (1989, p. 464), using data from a 1973 survey, found that 20 percent of suburbs in a national sample had imposed growth moratoriums. 3The exact wording of the question was, “To the best of your knowledge, have there been any MORATORIA on building permits, water connections, or sewer hook-ups in your city?” Although it is possible that planners who have been employed longer are more likely to be aware of past moratoria, our data indicate that there is no correlation between the number of years of experience reported by respondents and their response as to whether a moratorium had occurred (r = –0.02). 30 • The use of capital improvements and public works projects (such as street widening or sewer capacity) to control the rate or location of residential growth, • Requiring a popular vote to increase sewer capacity, and • Limiting the number of annual water connections. Each of these relates proposals for new development to some concern over the city’s physical infrastructure. The last two policies share some commonalities with the more restrictive policies discussed above, particularly the annual limit on water connections, which can function as a numerical cap for annual housing construction. As Table 3.2 indicates, however, the percentage of cities embracing an annual limit on water connections (4 percent of all cities responding) and the popular vote requirement for sewer capacity increases (3 percent) is extremely low, as was the case for the more overt caps on housing units discussed above. In other words, few cities report the use of Table 3.2 Use of Policies Linking Housing Development to Infrastructure Percentage of Cities Using Policy L. A. Other So. S. F. Bay Central County California Area Valley Total Policy (N = 56) (N = 91) (N = 77) (N = 73) (N = 297) Satisfy traffic standards before allowing development 23 26 34 21 27 Use capital improvements to control rate or location of growth 7 13 13 21 14 Popular vote required for sewer capacity increase 02 36 3 Annual limit on water connections 41 55 4 Mean number of these policies used per city 0.34 0.42 0.55 0.52 0.46 Notes: Percentages represent the number of respondents indicating use of the policy as a proportion of all those providing an answer to the question, including “Don’t know” or “N/A” responses. For more detailed tabulation, see Lewis and Neiman (2000, pp. 7–8). 31 infrastructure policies to create yearly restrictions on housing development. The other two types of infrastructure policies are more common, although still used by only a small minority of communities. About one city in seven uses capital improvements to shape residential development, whereas just over one-quarter of cities require that traffic standards be satisfied before new development occurs. The latter action belongs to a class of policies often termed “adequate public facilities ordinances,” in which localities attempt to legislate the desired levels of service that must be attained in the community before new development can be approved. Since traffic problems are one of the most visible and irritating examples of public facilities that fall below standards, it is not surprising that a fair number of cities single these out for scrutiny in the development process. In the case of these infrastructure-related policies, the differences among regions are more muted than in the case of the overt restrictions. It is worth noting, however, that Central Valley cities are among the most likely to use three of the policies (all except traffic standards). This may indicate a higher perceived level of infrastructure limitations in that region, which grew at the most rapid rate during the 1990s. It may also reflect difficulties in paying for infrastructure, given that the Central Valley is the poorest of the regions. Finally, it might also reflect Central Valley concerns regarding the loss of agricultural land to urban uses, where the “premature” extension of development infrastructure might stimulate the spread of urban pressures onto agricultural land. Shaping Housing Development The final category, for policies that “shape” growth, is something of a residual category for policies that are not explicitly infrastructurerelated and that do not embrace overt restrictions on housing development. Included are the following five types of city actions: • Ranking proposed residential projects through a point system, • A requirement that residential developments include “affordable” housing, 32 • A policy that restricts new residential growth to areas that are already developed, • A policy that encourages residential growth to occur in alreadydeveloped areas, and • Design review standards. These policies fall in a class that generally seems more benign than overt restrictions. Nevertheless, each arguably may add to the cost of housing or limit production. For example, policies that target new development to already-developed areas have the goal of increasing infill and reducing sprawl. (This class of policies includes growth boundaries and urban limit lines.) But by limiting the array of land choices available to developers, and focusing on properties that may be smaller or more difficult or expensive to develop, these policies can result in more expensive housing. Policies that require affordable housing as part of larger developments do take heed of the needs of low-income households. But some argue that affordable housing quotas may shift costs to other homebuyers; that is, lower-priced units may have their costs subsidized by the market-rate units, which are then priced higher. There is a great deal of controversy over this argument, with some analysts concluding that under many market conditions, the costs will come from developer or landowner profits (Calavita and Grimes, 1998, p. 152). Even design review standards, which are geared toward improving the aesthetic appearance of the community, can easily increase housing costs, since they may result in the creation of a quality “floor” of materials or standards that builders must meet. Efforts to create stylistic unity in architectural schemes in local communities may result in a smaller array of choices of housing types available to consumers. Design review may encompass building setback requirements, which can affect the number of units that can feasibly be included in the project. This is not meant as a criticism of these policies, but as a way of setting them within the general category of local government regulatory activity affecting housing. 33 Table 3.3 indicates that only one of these policies, design review, is in very widespread use, with 83 percent of cities employing it.4 By contrast, the most restrictive policies (point systems and restriction of development to built-up areas) are each used by only about one in 20 cities. The “carrot” approach to urban growth boundaries—encouraging (rather than requiring) builders to direct housing to built-up areas—is used by 28 percent of responding cities. A slightly higher percentage embrace affordable housing set-asides. As in the case of the overt restrictions, Bay Area cities show a higher level of regulatory activity to “shape” growth, for all five activities. Design review standards are particularly ubiquitous, with only two out of Table 3.3 Use of Policies to Shape Residential Growth Policy Rank proposed residential projects Projects must include affordable housing Restrict growth to built-up areas only Encourage growth in built-up areas only Design review standards Mean number of these policies used per city Percentage of Cities Using Policy L. A. Other So. S. F. Bay Central County California Area Valley Total (N = 57) (N = 91) (N = 77) (N = 73) (N = 298) 44 6 45 25 26 49 21 31 5 2 14 4 6 16 21 75 82 39 36 28 97 73 83 1.25 1.34 2.07 1.37 1.52 Notes: Percentages represent the number of respondents indicating use of the policy as a proportion of all those providing an answer to the question, including “Don’t know” or “N/A” responses. For more detailed tabulation, see Lewis and Neiman (2000, pp. 9–10). ____________ 4The prevalence of design review might disguise the fact that the stringency of these reviews can vary extensively. Unfortunately, our study did not measure the full array of topics that might be encompassed in design review, ranging from paint colors to roofing materials and appearance. Some of these design review items might result in substantial delays and increases in housing costs. 34 77 respondents from that region (less than 3 percent) indicating that their cities do not engage in design review. Bay Area cities are also significantly more likely to have affordable housing set-aside requirements, with about half of the municipalities in that region doing so, compared with about one-quarter of cities in Los Angeles County and the rest of Southern California, and about one-fifth in the Central Valley. Although it is not possible to ascertain at this time whether such policies actually can or will increase the supply of affordable housing, it is reasonable to assume that such policies reflect concern about the worsening problems of affordability in the Bay Area in recent years. The regions outside the Bay Area generally are fairly similar to one another in the degree of adoption of these policies, except that Central Valley cities are much more likely to encourage infill than their counterparts in Southern California. The notable rate of respondents indicating that their cities encourage growth in built-up areas might be additional indirect evidence of the Central Valley's increasing concern about how the spread of urbanization affects the maintenance of agriculture in the region. Overall Levels of Growth-Management Activity We have seen thus far that each of the growth-management policies has been adopted by only a minority of cities, with the exception of one relatively uncontroversial policy, design review standards. More important, perhaps, is the overall effect of such policies upon builders operating in California jurisdictions. Among all the regions surveyed, the average city has adopted about 0.7 overt restrictions, 0.5 policies to link growth to infrastructure, and 1.5 policies to shape new housing development, for a total of about 2.7 policies out of a possible 16. Figure 3.1 shows the entire distribution of the number of growth-management policies adopted. Of the 297 respondents, only 24 (or 8 percent) adopted six or more policies, whereas 18 (6 percent) adopted none. To get a sense of which types of cities might be more likely to engage in such activities, Table 3.4 shows demographic and other characteristics for these “heavy adopters” and “nonadopters” of growth management. As the table indicates, there 35 Number of cities 90 80 70 60 50 40 30 20 10 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Number of policies adopted Figure 3.1—Frequency Distribution of the Number of Residential Growth-Management Policies Adopted Table 3.4 Comparing “Nonadopters” of Growth-Management Policies to “Heavy Adopters” City Characteristic Per capita income, 1990, $ Residents aged 25+ who are college graduates, 1990, % Owner-occupancy of housing, 2000, % Average unemployment rate, 1990a, % Population non-Hispanic white, 2000a, % Population growth rate in city, 1990–1998, % Population growth rate in county, 1990–1998, % City incorporation datea Democrats, of major-party registered voters, 1999, % Average for Cities with No Policies 17,615 17.7 61.7 8.7 49.5 14.9 14.3 1939 54.5 Average for Cities with > 5 Policies 17,994 23.4 64.4 5.5 65.1 14.9 11.9 1915 53.0 Sources: 2000 Census, 1990 Census, California Department of Finance Demographic Research Unit, California Secretary of State, author calculations. aStatistically significant difference in this characteristic (p < 0.05). 36 are some social differences between these extreme ends of the spectrum, with heavy adopters tending to have significantly higher proportions of non-Hispanic whites in the population and higher proportions of college graduates. Heavy-adopting cities also tend to be older than nonadopters (i.e., they were incorporated earlier). Nevertheless, some characteristics that one might assume would be indicators of slow-growth policy adoption, such as high income and homeownership levels, or rapid growth rates, are not very different at all between the heavy adopters and nonadopters. Clearly, a more careful analysis of why cities adopt growthmanagement policies is warranted, and we will provide this in Chapter 5. For now, it is worth noting once again the major differences in cities’ policy adoption across the three regions. As Figure 3.2 shows, cities in the Bay Area average about 3.5 growth-management policies, compared to 2.5 in the Central Valley, 2.1 in Los Angeles County, and 2.4 elsewhere in Southern California. Thus, studies of slow-growth policies that draw upon case studies from the Bay Area, such as Frieden (1979), are probably unrepresentative of growth policies in most of California. 4.0 3.5 Shape housing development Link to infrastructure 3.0 Restrict residential growth Number of policies 2.5 2.0 1.5 1.0 0.5 0 Los Angeles Other Southern San Francisco Central County California Bay Area Valley Figure 3.2—Average Number of Growth-Management Policies, by Region 37 The Bay Area has a somewhat unique set of attributes that may make it more self-conscious regarding issues of regional growth and quality of life (see also Baldassare, 1994) and thus more likely to be amenable to growth management policies. These attributes include environmental features such as the dominant presence of the Bay itself, the ubiquity of ridge lines and steep slopes that make development potentially difficult or unsafe, the proximity of high-value viticulture in the wine country of Napa and Sonoma Counties, and a focus on preserving the centrality and historic amenities of the city of San Francisco. Also notable are the particularly high levels of income, education, and traffic congestion by comparison to other U.S. metropolitan areas (the latter condition compounded by the area’s unusual topography) and the regional economy’s relatively strong reliance on amenity-oriented tourism. In addition, it may well be the case that the early adoption of growthmanagement policies in the Bay Area, and the more liberal and environmentalist ideology characteristic of residents of that region, led to a diffusion of policy adoptions among cities there. Also notable is the slightly higher average number of residential regulations in the Central Valley compared to Southern California. This is the opposite result of what one might expect, given the more rural nature of many Central Valley communities, the more recent spread of urbanization there, and its somewhat more conservative political culture. Infrastructure limitations and the desire to protect high-value farmland from development may help account for the Central Valley’s growthmanagement policies. Summing Up In reviewing the patterns of adoption of growth-management policies in the major regions of California, perhaps the most notable fact is the small number of such adoptions, with the average city in our dataset having only about 2.7 out of the 16 policies. This finding seems consistent with an earlier survey of growth-management policy adoption by Glickfeld and Levine (1992), in which the average California city they surveyed (in 1988) had enacted 1.9 out of a possible total of 14 policies. It appears that some observers have tended to generalize about local 38 growth management based on the actions of high-profile but unrepresentative communities. In our survey, cities were most likely to have adopted policies to shape residential development, such as design review requirements or affordable housing set-asides. Policies designed to overtly restrict the total amount of housing developed in the city, or to condition housing construction on infrastructure capacity, were less common. Insofar as there were “heavy” adopters of growth-management policies (which we defined as six or more policies out of 16), they tended to have populations with a smaller proportion of minorities and a higher share of college graduates than cities adopting no policies. In other aspects, however, such as population growth rates, per capita income, and political party registration, these two groups of communities were remarkably similar. Regional differences, however, are readily apparent, with cities in the San Francisco Bay Area enacting policies more frequently than those in the Central Valley or in Southern California. This pattern was particularly true of policies to restrict the development of housing, or to shape the type of housing that was developed. Nevertheless, even Bay Area cities tended to enact only a few of the 16 policies we asked about. 39 4. Political Controversy and Growth Management In this chapter, we begin to probe more closely for the local factors that may influence whether cities will adopt growth-management policies. In particular, we are interested in the local politics of residential growth. Frieden (1979), using case studies, portrayed local governments—alongside homeowners, neighborhood groups, and environmentalists—as active and often very willing participants in the movement to restrain housing development in California. Is this perspective borne out by our survey data from cities throughout the three regions? Are local officials leaders in the movement to restrict or manage growth? Are they willing followers of public opinion? Or are they more progrowth in orientation and only loosely bound by local antigrowth sentiment? Is there evidence for elitist and exclusionary motivations for the use of residential development controls? In examining the data presented in the following pages, we conclude that a major factor influencing the adoption of residential development controls is the eruption of controversy and conflict over growth. It is possible, we contend, that once the conventional patterns of local residential development politics are disrupted, subsequent policies might contain more restrictive provisions. Is Residential Growth Always Controversial? Although growth issues are sometimes portrayed as a source of constant controversy in contemporary California, this may be a reflection of media coverage. That is, it may be the case that most residential development is a relatively quiet and routine process, but the topic only becomes “newsworthy” when a specific controversy with overt conflicts develops. We asked the planners responding to our mail survey, “How controversial would you say residential growth issues are in your city?” 41 As Table 4.1 indicates, controversy appears to be somewhat sporadic in many cities, with 52 percent stating that residential growth issues were “sometimes” controversial. A fifth of respondents said that such issues simply were not controversial, whereas 28 percent found them often or always controversial. Once again, there are very clear regional distinctions in the data. Fully half of the respondents from the Bay Area (the setting for Frieden’s case studies) call residential growth often or always controversial, compared to about one-quarter in cities in Los Angeles County and the rest of Southern California, and about one-eighth of Central Valley respondents. Similar findings emerge in response to a question about the influence of residential development issues on city elections. As Table 4.2 shows, 39 percent of respondents indicate that growth issues hardly ever affect council or mayoral elections, 31 percent say that election outcomes have been affected a few times, and 22 percent feel that growth issues have often been influential in affecting elections. Again, growth does not appear to be a burning issue in many local elections, although it is clearly a prominent issue at some times in some places. As for which places, the table illustrates the regional pattern; 41 percent of Bay Area respondents said that local growth issues have been influential in affecting local elections, as compared to 14 percent in the Central Valley, 11 percent in Los Angeles County, and 18 percent in the other counties of the Southland. Table 4.1 Prevalence of Controversy Regarding Residential Growth Issues “In considering the past and more recent periods, how controversial would you say residential growth issues are in your city?” L. A. County Other So. S. F. Bay Central Cities California Area Valley Total Response (N = 55) (N = 91) (N = 76) (N = 72) (N = 294) Not at all controversial 22% 22% 4% 28% 19% Sometimes controversial 55 51 43 60 52 Often controversial 18 21 25 8 18 Almost always controversial 4 7 25 4 10 Don’t know 2 0 30 1 42 Table 4.2 Effects of Growth Issues on Local Elections “Which of the following do you believe best describes the influence of residential development issues on your city’s elections?” L. A. County Other So. S. F. Bay Central Cities California Area Valley Total Response (N = 56) (N = 92) (N = 76) (N = 72) (N = 296) Growth issues hardly ever affect council or mayoral elections 50% 37% 21% 51% 39% There have been a few times when growth issues have affected the outcomes of local council or mayoral elections 29 33 34 28 31 Local growth issues have often been influential in affecting the outcomes of local council or mayoral elections 11 18 41 14 22 Don’t know 11 12 4 7 8 The next question is whether such controversy accelerates the adoption of growth-management policies. Indeed, controversy levels do appear to be related to the overall adoption of growth-management policies. Recall from the last chapter that the average city in our sample had adopted 2.7 growth-management policies out of the 16 mentioned in the survey. As Figure 4.1 shows, there is a distinct connection between levels of political controversy and the adoption of growthmanagement policies. In cities where respondents report growth issues to be almost always controversial, the average number of adopted policies is 4.1, compared to 2.0 policies in cities reporting no controversy.1 Thus, local growth controversies appear to be a precipitating factor in the adoption of growth management policies. This conclusion is ____________ 1The simple correlation between respondents’ answers to the “general controversy” question and the total number of growth management policies is a strong 0.34. 43 5 4 Number of policies 3 2 1 0 Almost always Often Sometimes Not at all controversial controversial controversial controversial Figure 4.1—Number of Growth-Management Policies, by Level of Controversy supported by an earlier study of growth controls among Southern California municipalities (Donovan and Neiman, 1992). In a related finding, Landis et al. (1995, p. xxiii) conclude that localities with greater levels of growth controversy tend to require Environmental Impact Reports for a higher percentage of proposed projects. What Role Do Local Citizen Initiatives Play in Growth Management? The adoption of growth-limiting or growth-managing policies tends to receive most attention from journalists and scholars in cases where the voting public itself enacts such measures, through so-called ballot-box planning. Judging by our survey evidence, however, citizen initiatives do not appear to be a direct source of many residential policies in the large majority of cities (Table 4.3). Fewer than one in six respondents indicated that initiatives have been a major source of slow-growth policies in their city; fewer than one in eight felt that a future initiative to slow growth would be likely in their city. Once again, these responses 44 Table 4.3 Prevalence of Citizen Antigrowth Initiatives “In your city, have initiatives been a major source of policies to slow residential development? Don’t include referenda placed on the ballot by the council.” L. A. County Other So. S. F. Bay Central Cities California Area Valley Total (N = 52) (N = 87) (N = 76) (N = 72) (N = 287) Yes 10% 15% 33% 4% 16% No 90 85 67 96 84 “Is there a good chance that an initiative measure to slow residential development will occur in your city?” L. A. County Other So. S. F. Bay Central Cities California Area Valley Total (N = 50) (N = 79) (N = 75) (N = 67) (N = 271) Yes 4% 10% 24% 7% 12% No 96 90 76 93 88 were quite different in the Bay Area, which showed by far the highest likelihood that initiatives are playing or will play a major role. In the Central Valley, by contrast, antigrowth initiatives have been quite rare. It is also worth noting that respondents in cities that have had citizen slow-growth initiatives in the past are more likely to expect additional ones in the future. Among respondents answering both questions, 19 of the 40 who indicated past experience with citizen initiatives said that another initiative was likely in the future.2 In part, this finding indicates that growth issues occupy a persistent place on the political agenda of a handful of municipalities in the state. Another consideration is that some initiative measures have language requiring future citizen votes on proposed projects of certain types, such as residential projects greater ____________ 2Among the 297 respondents, ten failed to answer the question about past initiatives and 26 did not answer the question about future initiatives. The drop-off in responses for the latter question can probably be attributed to an unwillingness to speculate about the future. 45 Average number of policies than a certain number of units or projects that require zoning changes. In this way, one initiative can beget future initiatives. Our next step was to examine whether the number of growthmanagement policies is noticeably higher in cities that have experienced, or expect to experience, citizen initiatives. Figure 4.2 shows that this is indeed the case. The average number of growth management policies is 3.9 in cities that have had initiatives compared to 2.5 in those that have not. To some degree, however, this finding is tautological, since cities indicating that initiatives “have been a major source of policies to slow residential development” will, by definition, have such policies. Perhaps more interesting, then, is the pattern in regard to expected future citizen initiatives. In those cities whose respondents expected such an initiative, the average number of policies was 4.2, compared to 2.5 in cities that do not anticipate an initiative.3 The significantly higher number of policies in cities having had or expecting initiatives is also evident within each of 5 Yes 4 No 3 2 1 0 Had citizen growth initiative? Expect citizen growth initiative? Initiative status Figure 4.2—Citizen Initiatives and City Growth-Management Policies ____________ 3This positive and highly significant relationship between the respondent’s expectation that an antigrowth initiative is likely and the number of growth-management policies is apparent even after controlling for past initiatives. 46 the three categories of residential growth policies (policies to restrict, link, or shape). The relationship between expected future initiatives and the number of policies may indicate that city governments pass residential policies in anticipation of citizen mobilization on the growth issue. That is, local governments may engage in growth management in an effort to head off or co-opt citizen antigrowth movements.4 Citizen initiatives are often more restrictive and inflexible in their requirements than policies passed by city councils, and so city policymakers may attempt to craft more flexible or generous policies that assuage residents’ demands for action on the growth issue. Of course, this presumes that city councils are, in fact, more progrowth in orientation—more accommodating to new housing—than the local antigrowth activists in such communities. What Is the Attitude of City Councils Toward Residential Growth? Do local policymakers share in the discontent over residential growth, or are they more interested in accommodating additional housing? To help answer this question, we asked the planners responding to our survey to describe “the general attitude of the majority of your city council toward residential growth.” As Figure 4.3 indicates, the most common response was that the council encourages residential development, and the next most common reply was that the council is “mostly neutral.” Only 13 percent of the 294 responses to this question indicated that the city council “occasionally slows the rate of residential growth when growth issues become controversial,” and 6 percent said the council “generally tries to slow growth and often proposes limitations on residential development.” (Four percent of respondents chose the “don’t know” response.) Because most residential policies are set by city councils, these responses are important barometers of local orientations toward growth. ____________ 4Similarly, growth-control proponents may pursue initiatives even when their chances for winning the initiative election are not good. In the 1980s, “San Francisco, for example, [was] host to a string of failed initiatives. Yet proponents of those initiatives saw these measures as “pushing” the local government to enact its own growth controls” (Glickfeld, Graymer, and Morrison, 1987, p. 136). 47 Percentage of cities 100 90 Neutral 80 Encouraging 70 60 50 40 30 20 10 0 Los Angeles Other Southern San Francisco Central County California Bay Area Valley Figure 4.3—Reported Attitude of City Council Majority Toward Residential Growth In fact, when we asked what the respondents’ perceptions were of the origins of their cities’ policies that affect residential growth rates, 58 percent said that such policies were “enacted by the council, without much neighborhood pressure.” Only 1 percent of the respondents said that these policies were “enacted pretty much exclusively as a result of the initiative process.”5 Thus, most city councils in these regions can be thought of as either generally supportive of new housing or at least neutral. This is true even in the San Francisco Bay Area, where 42 percent of councils are seen as encouraging and 25 percent as neutral. Indeed, Bay Area councils are actually seen as much more likely to encourage housing than city councils in Los Angeles County, where the respondents place most of their councils into the “neutral” category. Despite the high levels of activism and controversy in the Bay Area, city councils may be more ____________ 5The other results were that 25 percent saw such policies as “enacted by the council, but mainly as a result of neighborhood pressure,” whereas 15 percent concluded they were “enacted through city council action and through the initiative process.” 48 sensitive to the increasingly severe housing problems facing the region. City officials, consequently, might be struggling with or recognizing their responsibilities to accommodate residential development in the region. Many politically active residents of cities in the Bay Area, however, are apparently less supportive. Summing Up The primary finding of this chapter is that local controversy over growth, and the reality or expectation of local citizen initiatives relating to growth, are strongly associated with the presence of a greater number of growth-management policies. Thus, growth-management policies appear to emerge in large part as a result of the dynamics of local politics and residents’ activism, rather than through dispassionate analysis of local or regional housing trends or environmental conditions. Nevertheless, only a minority of cities seem to be regularly rocked by residential growth conflicts, and the vast majority have had little experience with slow-growth citizen initiatives. In short, pitched battles over residential development seem to be limited to a relatively few localities, whereas growth is a more routine process in most communities. City councils, which play the strongest role in setting policies that affect residential growth rates, typically are viewed as progrowth or neutral in each of the regions studied. Interpreting the Evidence Thus Far The results above lead us to propose the following scenario regarding the emergence of local growth management: Occasionally, local conflicts erupt over residential proposals. These may occur because of deteriorating local conditions that are blamed on growth, or rapid changes in community character, or the emergence of an antigrowth “political entrepreneur” in local politics (Schneider and Teske, 1993). If city officials do not act sufficiently rapidly or strongly to assuage resident concerns or persuade residents that the city has taken steps to accommodate the new housing gracefully, the result may be sustained controversy over residential growth—perhaps enough to affect local 49 mayoral or council elections. The result is perhaps the formation of local groups favoring slower, managed, “smart,” or no growth at all. Although most cities do not reach this stage of political turmoil over development, those with an aroused and dissatisfied populace are likely to see some growth-management measures passed, generally by the council but in a smaller number of cases by citizen initiative. At this point, the growth issue may go into hibernation in some communities. In others, the passage of growth-management measures may in turn create further controversy—either by alienating progrowth interests among the citizenry, or by setting unrealistic expectations that growth will slow down, or by requiring that future decisions about residential proposals be subject to popular vote. If growth controversies occur in a setting where local officials are decidedly progrowth, then even with a short-term victory, citizens concerned with slowing or stopping growth are likely to persist because they distrust their officials. Moreover, local battles, when they result in restrictive development actions, also are likely to produce the actuality or threat of litigation. Finally, in some cases, the threat to slow or stop growth might actually stimulate the programming of considerable future growth, by causing developers to rush the processing of permits before limits take effect. As a result, when growth controversies become heated, they are likely to stay that way for a considerable period. If the foregoing account is true, it evokes an image of grassroots citizen activism leading the antigrowth movement in California, with city governments acting more as followers—and perhaps often, reluctant followers. Although most city councils appear to feel a responsibility to accommodate a reasonable share of their region’s residential growth pressures, it is politically perilous for them act unilaterally and discount residents’ fears and concerns about growth. In the next chapter, we will consider a wider variety of local characteristics that may help explain why some cities are more prone than others to growth disputes and growthlimiting policies. 50 5. Factors Shaping Cities’ Willingness to Accept Residential Growth What factors produce conditions that are particularly conducive to the passage of local residential growth controls? In this chapter, we summarize the results of a series of statistical analyses designed to answer this question. The chapter begins with a discussion of various city characteristics that might possibly influence local residential policies and some of the data analysis issues involved. The analysis then proceeds in three parts: First, we examine explanations for various housing policy choices of cities, including the number of growth-management policies and local requirements for affordable housing set-asides. Second, we look at city governments’ overall orientation toward housing development—that is, how enthusiastic or wary officials are reported to be regarding residential growth in their cities. The third topic explored is local political controversy over residential development, which is a major determinant of growth-management policy adoption. This chapter is meant to point out the most interesting and statistically reliable results of our various multivariate models. The appendix of this report provides results of the models and further information on variables, data sources, and methodology. Considering Potential Explanatory Factors Local Political Factors One set of potentially important factors for explaining city actions is political. We have already seen the important relationship between citizen controversy over growth and the number of growth-management policies pursued. Other political factors of interest might include the ideological or partisan leanings of the local populace, the number of 51 planning staff available to develop and enforce growth rules and ordinances, and the fiscal climate of the city. The strength of local interest groups concerned with growth, the method for electing city officials, and the presence of competing important issues on the local political agenda might also be of importance. Although we could not locate suitable measures for all potential political and institutional variables, we rely upon the following variables to measure political characteristics in each city: • A measure of citizen opposition to growth (to be discussed below), • The percentage of major-party registered voters who are Democrats, • The number of planning staff per 1,000 local residents, • The dollar amount of own-source city government revenues, per capita, and • The estimated percentage share of all property taxes paid by city residents that go toward the city government—as opposed to the county, school district, or special districts.1 Some of the statistical models also include some ancillary measures (where such data seem important to the question at hand): • The importance of neighborhood groups and business groups in local development policymaking, as judged by the city manager, • Whether city council members are elected by districts or at-large, • How active the city is in redevelopment policy (according to the city manager), and • The crime rate of the city—an issue that may compete for time and attention from local policymakers. How would each characteristic affect policy? One might expect cities with higher levels of citizen opposition to growth to have city governments that do more to limit residential development. The ____________ 1This is sometimes referred to as the city’s “split” of the property tax. Some observers of local government in California argue that cities with a low share of the property tax may be more inclined to engage in the “fiscalization of land use,” seeking out development projects that provide more sales tax or fee revenues. 52 influence of Democratic party affiliation is less clear, since labor and minority groups (often pro-housing development) and environmentalists (sometimes antigrowth) are all important parts of the Democratic coalition. Cities with high levels of local revenue, or a larger percentage share of the property tax, might possibly be more willing to accept housing, which is often seen as fiscally costly. Diaz and Green (2001) found, however, that municipalities in Wisconsin with greater fiscal capacity engaged in more growth-management efforts. It is unclear whether an increase in planning staff resources would increase cities’ ability to regulate growth or lead to more support for housing development. Planning staffs might provide the institutional capacity to draft local development restrictions, but the presence of larger number of planners might also contribute to a more balanced, professional norm of including reasonable levels of housing or concern for the potential sprawl-inducing effects of excluding housing. Local business groups, such as the Chamber of Commerce, might be expected to approve of local population growth and more housing, whereas neighborhood groups are often reputed to oppose such growth. Districtbased elections of the city council could lead to an emphasis on “not in my backyard” politics, and thus less enthusiasm for residential development. Finally, cities with high crime rates might be less likely to become convulsed in controversy over residential issues, since the crime issue may tend to dominate the local policy scene or high-crime communities may also feel that residential growth is the economic stimulus that might reverse their crisis conditions. Demographic Makeup of the City We are also interested in the demographic background and socioeconomic status of city residents because local residential policy may possibly respond to or reflect differences in the wealth or income, status, and racial or ethnic background of the population. One problem with using such measures to characterize California cities is that many are closely related, which makes it difficult to distinguish the independent effects of each variable. For example, cities in California with higher percentages of Hispanics in the population also tend to have higher unemployment rates and more children in the population. Cities with 53 high household incomes tend to have high proportions of homeowners and low unemployment rates, and so on.2 To resolve this difficulty, we rely on one measure of status and two measures of ethnic background that are often associated with socioeconomic status. Each seems potentially quite important in the formation of local development policy, but the variables have the advantage of not being closely correlated to one another. Each is measured using 2000 Census data for the cities in our sample: • The percentage of housing units that are owner-occupied, • The percentage of the city’s population that is Hispanic, and • The percentage of the population that is African-American, or black.3 Here, the expectation is that communities with higher levels of owner-occupancy might have more exclusionary, status-based motivations for local housing policy (Danielson, 1976). The influence of local Hispanic and black populations is less clear, although minority groups in the United States are often thought to be less likely to embrace antigrowth political movements (Neiman and Loveridge, 1981; Carman, 1998; DeLeon, 1992). Cities with higher minority group populations may also be more likely to support requirements for affordable housing or be more concerned with the overall need for economic development, including the anticipated stimulus of residential construction. On the other hand, it is also possible that officials in cities with higher concentrations of minority group members may emphasize market-rate housing or even pursue limitations on multifamily development, in an effort to avoid becoming a segregated, low-income minority community. ____________ 2Each of the above relationships has a correlation of 0.6 or greater across the cities responding to the survey. To avoid multicollinearity, the variables selected for inclusion in the model all have correlations below 0.5. 3In the 2000 Census, respondents were able to identify with more than one race; moreover, Hispanic is not a racial category. However, in an effort to create nonoverlapping categories for this statistical analysis, we calculated for each city the percentage Hispanic (of any race) and the percentage non-Hispanic black (who did not also identify with another race). 54 City Size, Urbanization, and Location Larger communities, and central cities in particular, may be somewhat distinctive in their orientations toward growth (Lewis, 2001; Neiman, Andronovich, and Fernandez, 2000). Larger communities tend to have a greater number of active political groups, and larger city governments, which may lead to the development of more policies. Diaz and Green (2001) find that in Wisconsin, municipal population is positively associated with the adoption of growth management tools. On the other hand, progrowth organizations, such as downtown businesses, construction unions, and media organizations, are often very politically active in central cities and other large communities, which may represent a countervailing force against regulating growth. Furthermore, central city political leaders may be more interested in capturing growth, since they are likely to be held somewhat responsible for local economic vitality, particularly if they are surrounded by growing, competitive suburban communities. Smaller rural and suburban communities may face distinctive types of growth challenges. Regional differences may also help account for city policies. As we have seen, the Bay Area appears to have much higher levels of growth management than other parts of the state. It remains to be seen whether its distinctiveness remains after we take account statistically of its major differences from the rest of the state (for example, in socioeconomic status and Democratic party affiliation). Some may anticipate that the fast-growing, largely suburban regions of Southern California outside of Los Angeles County may have a distinctive growth politics as well. We thus include the following variables in this category: • The local population size, • An indication of whether the city is a central city, a suburb, or a rural community,4 • An indication of whether the city is in the San Francisco Bay Area, and ____________ 4We rely on Census Bureau designations of central cities. Municipalities that are urbanized and located within metropolitan areas, but are not central cities, are counted as suburbs. This is the most numerous category in California. The remainder are classified as rural. See Lewis (2001) for further discussion of this typology. 55 • An indication of whether the city is in Southern California but outside Los Angeles County.5 Local Conditions Relating to Growth The willingness of city residents and elected officials to accommodate housing may be shaped powerfully by the capability of the city’s infrastructure to handle new growth and other conditions in the city (or its surrounding area) that might be thought of as the outcomes of previous growth. City officials are likely to ask themselves, “What kind of city is this? How are we geared to handle new housing, and what might be its costs to our existing character?” For example, a city rich in employment and with abundant, uncrowded transportation infrastructure may be far more willing to accept multifamily development than a city with an excess of housing units and long commutes to jobs in the local area. Cities with problems of housing affordability may be more accepting of new housing development. Regions that are growing quickly may create more pressure to accommodate growth locally, but such growth may also arouse more citizen controversy. And communities with high degrees of transiency, or population turnover, may have different reactions to housing proposals than more settled communities, or than resort communities with many part-time residents. Accordingly, we draw upon the following variables in explaining local policies: • Recent population growth rates: For questions relating to city government policies, we use the county growth rate (between 1990–1998), since city policymakers likely consider local policies in the light of growth pressures in the surrounding region.6 For other questions, such as those relating to citizen controversy, we use the city’s population growth rate over the same period. ____________ 5This includes cities in San Diego, Orange, Riverside, San Bernardino, Ventura, and Santa Barbara Counties. 6Glickfeld and Levine (1992) argue that local growth controls may in large part be a reaction to regional growth. 56 • The percentage of residents in 1990 who had lived in the same home five years earlier: a measure of residential stability, from the Census.7 • The average one-way commuting time among city residents, which may measure local roadway conditions as well as distance from homes to work. • The percentage of local housing units not connected to a modern sewer: Where developers must rely on septic systems rather than a public sewer system, it will be difficult if not impossible to build housing at high densities. This factor may therefore make city policymakers warier about adding housing. But at the same time, it may make overt local growth controls unnecessary, insofar as the lack of infrastructure itself helps stop growth. • The ratio of jobs within the city to workers who live in the city. • The percentage of local housing units that are seasonal or recreational: an indication of whether the city may have environmental amenities important to its character. (This information is available from the 2000 Census.) • The affordability of the local housing stock as of 1990: This is measured as the ratio of the median house value to the median income. This variable is used only for one statistical model relating to local affordable housing policy. Controversy and Growth Management: A TwoWay Street As we have seen, the level of local citizen controversy over growth is powerfully related to the number of growth-management policies. However, it is also realistic to expect that the number of local policies ____________ 7Unfortunately, for some variables, the 1990 Census is the most recent comparable data available at the time of this writing. (Thus far, city-level data from the 2000 Census are available only for racial, ethnic, and age groups, as well as household arrangements and homeownership.) However, since many residential policies and growth controls date back quite a few years, this reliance on 1990 data need not be a disabling data problem. Moreover, we think of the accumulated set of local policies in any given community as in some important way a response to prior community conditions. 57 may affect the level of controversy. This expectation is borne out by the fact that a tally of local growth-management policies as of 1988 (taken from Glickfeld and Levine, 1992)—a decade before our survey—is a significant predictor of controversy levels at the time of our survey (see the appendix for details). In short, it appears that causality runs both ways: Controversy promotes the adoption of growth-management policies, but controversy also grows as more policies are adopted—perhaps because the passage of policies motivates opposition or elevates the growth issue on the public agenda. Moreover, as controversy persists, the various organizations that emerge around the issue of growth become more sustained, even institutionalized perhaps, thereby moving controversy to more intense levels. This relationship is interesting, but it also creates statistical problems because the variables may simultaneously determine one another. Although there are various ways to address this issue, we have chosen to avoid the problem in the data analyses by substituting another measure of local citizen behavior for our controversy measure. The measure we use instead, also derived from the planning director survey, asked respondents to assess the importance of “citizen opposition to growth” in constraining residential development in their cities. In this case, the causality appears more straightforward. Citizen opposition to growth should generate more growth-control policies, but it seems unlikely that the number of policies itself would increase levels of citizen opposition to growth. Explaining Housing and Growth-Control Policy Outcomes We use multivariate statistical techniques that are designed to identify the effect of each city characteristic on local policies, while holding constant the other characteristics. For example, if we find that high homeownership rates are associated with a greater number of city growth-management policies, this indicates that homeownership has this relationship with housing policy independent of the racial makeup of the city, its population size, and so on. (See the appendix for more details on methodology.) 58 We first examine the sum of local growth-management policies—the total number of policies that restrict or shape growth or link housing development to infrastructure requirements. As noted in the last chapter, this total may range from 0 to 16. Based on the multivariate results, several factors appear to be significantly related to the number of growthmanagement policies adopted: • Not surprisingly, citizen opposition to growth is associated with a larger number of policies. • Cities with higher levels of homeownership tend to have a greater number of policies, but so do communities with higher shares of Hispanics in the local population. • Cities with a more stable population locally (less transient populations from 1985 to 1990) and slower population growth at the county level tend to have a greater number of policies. • Cities with higher percentages of unsewered housing units tend to have fewer policies. • Central cities and rural communities tend to have more growth management policies than suburbs. Both the Bay Area and the portions of Southern California outside Los Angeles County have a higher incidence of growth management, even when controlling for city characteristics. There are a number of interesting aspects to these results. First, although demographic factors are hardly the only explanation for growth management, upper-status communities (judging by the homeownership rate) do appear to engage in more of such policies, consistent with a social-status or exclusionary explanation. At the same time, however, communities with a higher proportion of Hispanics also have more growth-management policies. As Hispanics are the fastest growing segment of the California population, it is possible that governments of cities with high proportions of Latinos fear becoming overwhelmed by growth, although that is only speculation. The main point is that one cannot assume that upper-income white communities are the main bastions of growth control. 59 Second, mere population pressure itself does not generate such policies. If anything, population growth and turnover appear to create conditions that are somewhat less amenable to the passage of growth controls. This finding is consistent with Pendall’s (1999) work on housing protest in the Bay Area, which points out that cities with more recent population influxes tended to experience fewer citizen protests against new housing developments. Pendall attributes this result to the inability of newer and transient populations to effectively mobilize within local political systems, whereas residents of more settled, slow-growing communities may have greater political resources. Moreover, such cities may have an established pace and rhythm, along with longer-settled residents who recoil at the prospect of rapid growth and react by supporting development controls. The fact that a lack of a modern sewerage system decreases the number of policies is also revealing. It indicates that this infrastructure deficiency itself may lessen the pressure of rapid housing growth, since dense developments are not generally possible without public sewerage facilities. What we cannot determine from this is whether city leaders are consciously substituting a lack of sewerage facilities for growth-control policies, or whether the infrastructure deficiency occurs for other reasons and thus simply functions as a technological barrier to growth. Respondents’ answers to a number of other survey questions help cast light on city residential policies. In one case, they were asked to assess the importance of “density restrictions on residential land” (on a scale of 1 to 5) in constraining or slowing residential development. Their responses indirectly help us assess the role of zoning restrictions and subdivision requirements that may limit the density of housing built in cities. Factors that are significantly related to a high score on this question—meaning that density restrictions are important in slowing growth—include the following: • The most important correlate of density restrictions appears to be citizen opposition to growth. But citizen opposition is not solely a feature of high-status communities. In this case, cities 60 with high homeownership levels are no more likely to view density restrictions as important.8 • Findings on population pressures are analogous to those reported above. Cities with low population turnover and low levels of countywide population growth are more likely to find density restrictions important. This is again consistent with Pendall’s (1999) arguments about housing protest. • Central city respondents were significantly less likely to view density restrictions as important in constraining housing. In a related question, planning directors were asked to characterize their cities’ review process for residential development, in comparison to that of other cities in their area. In their answers, they labeled their own city as either less strict, equally strict, somewhat more strict, or much more strict than nearby communities. The following factors help predict the stringency of local residential review processes: • Cities with higher levels of citizen opposition to growth, and more homeownership, tend to be stricter. • Central cities and municipalities with more planning staff per 1,000 residents likewise tend to be stricter. • Higher rates of population growth at the county level are associated with a less stringent review process at the local level. • Cities with higher local revenues (per capita) are viewed as stricter, despite what one might presume to be their greater budgetary flexibility. In another survey question, we asked planners to speculate about the overall effect of their towns’ residential policies on the local population. The possible responses are that the city’s residential policies have led to a population somewhat lower in social status than it would otherwise be; the same in social status; somewhat higher in status; or much more affluent than it would otherwise be. Although admittedly somewhat subjective (which caused a relatively high 28 percent of respondents to ____________ 8Note that there is little or no relationship between homeownership levels and the degree of citizen opposition; the correlation between these two variables is only 0.03. 61 say “don’t know”), this question nevertheless gets to the heart of the issue: the perceived effect of local policies on city demographics. The respondents were not defensive, as some might expect. They readily pointed out the connection that is likely to exist between greater numbers of local restrictions and higher levels of social status for local residents. In short, they acknowledged the role that such restrictions might play in limiting access to a community. From our multivariate analysis, we make the following inferences: • Cities with more citizen opposition to growth and higher homeownership levels are seen as having policies that have led to more affluent populations. • Cities with larger populations are seen as having residential policies that are less likely to increase the population’s social status. • Cities with a higher share of unsewered housing units are also seen as having less socioeconomically restrictive policies. This perception indicates that sewerage deficiencies are not generally part of an intentional program of growth restriction but rather an unintended infrastructure limitation that may hinder development of all types.9 • By contrast, cities with long commute times indicate that city residential policies have been socioeconomically restrictive. Finally, we examine responses to a question that asked whether cities “currently have a policy to require residential development to include affordable housing, however that is defined in your community.” Our statistical analysis assesses factors that affect the probability of respondents’ answering yes or no to this question. In this model, we also included the variable examining the affordability of homes in the community. Interestingly, in this case citizen opposition to growth was ____________ 9The issue is important in evaluating local policies, because observers have sometimes claimed that water shortages in some parts of the state (Marin County, Santa Barbara) have basically been intentionally courted by antigrowth forces, seeking an excuse for disapproving housing development. (On “man-made drought,” see Frieden, 1979, pp. 43–48.) 62 not significantly related to the local policy outcome. Factors that were related to an affordable-housing set-aside include: • Cities with more “unaffordable” housing prices (relative to local incomes), as of 1990, are more likely to have an affordability requirement. This is possibly a reassuring result, indicating that cities with the worst affordability problems are taking actions to ensure production of affordable units—although it might also indicate desperation among respondents whose communities have already suffered deteriorations in affordability. • Cities with more stable, less transient populations are less likely to have the policy. • Perhaps surprisingly, cities with high homeownership levels and smaller populations are more likely to have affordability requirements for developers. It is possible that such cities have trouble meeting their affordable housing goals if they do not overtly require that such units be built. • Cities with a higher share of Hispanics, and cities in the generally more liberal Bay Area, are more likely to embrace affordability set-asides. Explaining City Government Orientation Toward Housing and Growth The next set of analyses deals with the city’s overall orientation or policy position toward residential development. Survey questions on this topic called for somewhat subjective responses but are relatively direct in asking respondents to assess the “friendliness” of the city government toward housing. The first two questions are drawn from the planning director survey. We asked the planners to assess “the general attitude of the majority of your city council toward residential growth.” Respondents could indicate that the council generally encourages residential growth, that it is mostly neutral, that it occasionally acts to slow the rate of growth when it becomes controversial, or that their council often proposes limitations on residential development. Which factors help explain cities’ positions on this four-point scale? 63 • Citizen opposition to growth is by far the most significant variable in increasing council antigrowth orientations. One would expect, in a democratic system, that council activity would reflect voters’ strong preferences on such an issue, to some degree. • Cities with longer commute times are much more likely to have councils wary of residential development, as are cities with stable (nontransient) populations. • Cities with high shares of unsewered housing units tend to have city councils that are less likely to take actions to slow growth. Again, a lack of sewer infrastructure probably makes such actions a moot point. • Cities with larger populations tend to have councils that are more progrowth. • Perhaps surprisingly, cities with high homeownership levels and more Democratic registrants also tend to have progrowth councils.10 We also asked the planners about their views regarding their cities’ development policies, with particular attention to whether commercial development is treated more favorably than housing. They could answer that their city “encourages all sorts” of residential and commercial growth, that it encourages commercial growth but is less receptive to multifamily and affordable housing projects, that it encourages most commercial growth but makes residential development more difficult, or that their city makes it difficult for both commercial and residential development. For simplicity, we will use the shorthand continuum of “permissive” to “restrictive” in characterizing respondents’ answers about their cities. Findings are relatively consistent with those discussed above: • Citizen opposition to growth is associated with greater restrictiveness, as is local population stability. ____________ 10When we included the variable measuring whether city councils are elected by district or at-large, it was not statistically significant. However, when controlling for district elections, the share of Hispanics in the population was positively related to council restrictiveness. 64 • Long commute times are again associated with greater restrictiveness. • Cities with more Democrats report more permissive development orientations, whereas the percentage of Hispanic residents in the city is associated with restrictiveness. • When information on city council election methods is included (reducing the number of cases), it appears that cities with district-based elections are more likely to have restrictive policies, perhaps indicating that district-based council members respond more to antigrowth sentiments in the neighborhoods. We can use some of the responses to the survey of city managers to cast further light on city orientations toward housing. Recall that the city manager survey was a statewide analysis, whereas the planning director survey was limited to the three major economic regions of the state. We are forced to do without some variables that were derived from the planner survey, such as the measure of citizen opposition to growth and the number of planning staff. However, we can turn instead to useful data from the city manager survey, in which respondents were asked to rate the importance of neighborhoods and business groups in land-use decisionmaking. The dependent variables—that is, the things we are trying to explain—are answers to two questions relating to multifamily development. First, regarding new development projects on vacant land, we asked, “Given your city’s overall strategies and plans for land use and future development, how desirable to your city administration would multifamily housing be?” The second item was an analogous question, about the receptivity of the city administration to multifamily development in redevelopment areas.11 For the most part, findings are consistent across the new development and redevelopment questions. • Cities with longer commute times are significantly less enthusiastic about multifamily housing development. ____________ 11Responses to the “new development” question are limited to those cities indicating that they had vacant land available. Responses to the redevelopment question are limited to cities indicating that the city was engaged in redevelopment. 65 • Where neighborhood interests are influential, city governments are more accepting of this type of housing, according to the respondents. • Respondents in the Bay Area indicate more interest in multifamily housing among their city administrations. In Southern California (outside Los Angeles County), however, multifamily development is less desirable for redevelopment areas. • There is at least modest evidence that cities growing at a faster rate in the 1990s are, in fact, more willing to accommodate multifamily housing. • Finally, cities with a “very active” redevelopment policy effort are much more interested in multifamily housing for their redevelopment areas. Explaining Levels of Public Controversy over Residential Development As we have seen, the public’s opposition to growth is a major motivator of certain growth-limiting policies and orientations on the part of local government. It is therefore important to ask which factors help generate public controversy over residential development in the first place and which local characteristics help predict whether citizens will bring the growth debate to the ballot box. We return to the planning director survey to assess these issues, examining questionnaire responses about the influence of growth issues in local politics. Note that we hardly claim to have a definitive explanation of local growth controversy. The level of political controversy in a city is likely to be affected by the vagaries of local history and politics, which we can capture only roughly in such an aggregate analysis. For example, a degradation of local public services, a sense of changing community character, or a widespread sense among residents that the community has lost control of its destiny might raise antigrowth controversy. Although we cannot measure such community patterns directly, the multivariate 66 model does help identify factors that probably make controversy more or less likely.12 For the first analysis, we attempt to account for the respondent’s answer to a question about the level of public controversy over residential growth issues in the city. Respondents replied using a four-point scale, ranging from “not at all controversial” to “almost always controversial.” The results indicate that several contributing factors stand out: • Community ethnicity and socioeconomic status appear to matter, but in a relatively complex fashion. All other things equal, cities with high proportions of Hispanics and blacks in the population experience less controversy over growth—but so do cities with high homeownership rates.13 It is possible that high homeownership in a city equates to low conflict because such cities are homogenous enough that they already have a degree of consensus regarding what types of growth to pursue. For example, such cities may be zoned overwhelmingly for single-family housing, which gives developers few opportunities to propose (potentially controversial) apartment projects. • Not surprisingly, cities with longer commute times have higher levels of public controversy. • Central cities of metropolitan areas, as well as rural cities, show more controversy than suburban cities. Despite a number of highly publicized growth battles in suburbia, the result may indicate that central cities, which often have a wide variety of land uses, neighborhoods, and income levels, are prone to ____________ 12In these models, we drop the variables relating to planning staff size, recreational homes, and own-source revenue because of their persistent lack of anything near statistical significance. Local population change is used in place of county-level population change, on the assumption that local increases are most visible and immediate to local residents, who may be less concerned with the dynamics of the entire regional housing market. 13Percentage Hispanic is quite correlated with the percentage of the city population that was unemployed (as of 1990), a variable not included in the model. If percentage unemployed is used in the model in place of percentage Hispanic, the unemployment variable also shows a strongly negative relationship to controversy. Thus, the results discussed here may mainly indicate community need or deprivation, rather than any aspects of ethnicity per se. 67 experience more conflicts over growth than the more homogeneous suburbs. Rural towns often experience the pressures of urbanization and farmland conversion, which could render growth issues more controversial, all other things equal. • San Francisco Bay Area communities report higher levels of controversy, even controlling for these other factors. • When we add local crime rates to the model (which slightly reduces the number of observations, because of missing data), it appears that high crime rates are associated with less controversy over residential growth. Crime, as a competing issue, may well displace growth from the local “controversy” agenda, by giving residents something more frightening to worry about. Highcrime communities also tend to have other problems attracting development, and, as such, might welcome development of any kind to provide local economic stimulus. Note that the other variables discussed above retain their effects when crime is included in the model. Another question asked the planners to describe the influence of residential development issues on their city’s mayor and council elections. They could respond that growth issues “hardly ever” affect local elections, that growth has done so “a few times,” or that growth issues are “often influential” in affecting election outcomes. • Here again, a higher share of Hispanics in the local population is associated with fewer electoral battles centered on the growth issue. • The same is true of high crime rates, if crime is entered into the model. • Cities that are more “job-heavy” are less likely to experience these housing-related electoral controversies (although the inclusion of crime rates appears to eliminate this relationship statistically). • Bay Area cities and central cities are considerably more likely to experience the electoral controversies—as are rural communities. Suburbs, it seems are less likely to have growth issues enter the electoral sphere, all other things equal. 68 Finally, we examine the issue of citizen initiatives regarding growth. Two survey questions asked whether “initiatives on the ballot [have] been a major source of policies to slow residential development,” and whether there is “a good chance that an initiative measure to slow residential development will occur” in the city. Focusing on the significant findings that are consistent across these two questions: • Cities in the Bay Area show higher probabilities of initiatives, both in the past and the expected future. Once again, suburbs appear to show less of this controversy than central cities and rural towns. • Cities with greater proportions of Hispanics and blacks in the population have a lower probability of initiative use.14 • “Job-heavy” cities also appear to be less likely to bring housing growth issues to the ballot box, as do cities with a higher proportion of unsewered housing units. • Interestingly, planners from cities that had faster rates of population growth in the 1990s are more likely to indicate that a future antigrowth initiative is probable. Summing Up In this chapter, we first identified four sets of factors that might be thought to affect local housing and growth policies: local political and fiscal factors, the demographic makeup of the city, city size and location, and local infrastructure and growth-related conditions. We then reviewed a wide variety of statistical analyses designed to cast light on the motivations behind cities’ use of growth management and their orientations toward residential growth. We report numerous multivariate models, each of which approaches the problem in a somewhat different way. Our main goal has been to detect overall patterns of findings, rather than to stress any individual result. Taking the empirical results as a whole, we can derive a tentative ____________ 14Here again, percentage Hispanic may serve as a proxy for percentage unemployed or some other measure of economic deprivation. Populations with less economic wherewithal tend to have a more difficult time mobilizing the resources necessary to carry out a political campaign, such as an initiative effort. 69 but plausible account of local growth management and housing policy that seems consistent with the evidence. As the previous chapter suggested, the level of citizen opposition to growth is a fairly consistent determinant of local residential policies, as we would expect in a democratic political system. Citizen controversy over growth, in turn, is more likely to arise under certain types of conditions. For example, growth-related controversies appear to be more common in cities with knotty commuting problems and in those where “bread and butter” issues, such as crime or unemployment, are not major factors. Cities with an “excess” amount of housing, given the number of jobs within the city, are also more likely to see voters bring their concerns over additional housing to the ballot box (both via antigrowth initiatives and in mayor and council elections). A high growth rate and high socioeconomic status do not, themselves, automatically translate into antigrowth activism. A number of local factors combine with citizen antigrowth movements to make city government growth-management policies, or restrictive orientations toward housing among policymakers, more likely. Again, local growth-related conditions play a major role, as one would expect from local officials’ assessing the capacity of their locality to handle growth. Long commute times render city councils more restrictive toward housing and make multifamily housing appear unattractive to officials. Communities with longer commute times might be experiencing substantial traffic congestion, since it is not clear whether the longer times are due strictly to greater distance or more traffic. Obviously, higher levels of traffic are often connected to less enthusiasm for growth. A rapidly growing county-level population, on the other hand, creates pressures to accommodate housing growth. A lack of local sewerage facilities appears to make widespread urban growth less possible, thereby reducing the likelihood that growth-management policies will be needed or that the council will have to take a hard line on growth. At the same time, cities with “unaffordable” local housing appear to pursue affordable housing set-asides in their planning policies. In short, conditions “on the ground” in California communities play a major role in shaping city residential policies. 70 The characteristics of the local population matter too. Cities with transient populations appear more likely to have progrowth councils and to pass fewer growth-management policies, perhaps because the community is less “settled” and has less of a distinctive identity. Local ethnicity and socioeconomic status affect housing policy in multiple ways that are not always easily predictable by those who see growth management as the tool of affluent whites. Cities with high homeownership rates, for example, do have more growth-management policies and a stricter review process of residential proposals but also appear more likely to have affordable housing set-asides and city councils that are neutral or progrowth, judging by the survey responses.15 Cities with more Hispanics experience less controversy over residential growth but tend to have a higher number of growth management policies, all other things equal. Despite our best attempts to institute controls for conditions that differ across various parts of California, San Francisco Bay Area communities appear distinctive in many of these analyses. Compared to cities in other regions, those in the Bay Area experience more controversy, are more likely to pass antigrowth citizen initiatives, and enact a greater number of growth-management policies—even as the region’s city officials embrace affordability set-asides in new developments and view multifamily housing as more attractive than their counterparts elsewhere in the state. By contrast, once we account for local conditions and demographic characteristics, suburban municipalities actually appear somewhat less troubled by growth controversies and appear less likely to pass growth-management policies.16 This evidence suggests that local growth controls and residential policies seem to be motivated by something more than simply the self____________ 15Of course, it is also possible that in the communities with very wealthy households, general market conditions might make it exceptionally difficult for more moderate- or lower-income families to find housing, no matter what the local inclusionary effort is. 16Even without controls, suburbs tend to have a smaller number of growthmanagement policies (2.5 on average) than central cities (3.0) or rural cities (2.9), although only the difference between central cities and suburbs is statistically significant. 71 interest of wealthy suburbanites, despite what many critics contend. Day-to-day local conditions, such as traffic congestion or an excess of housing to local jobs, seem to motivate citizen resentment of growth. In other words, it is the real consequences of growth “on the ground,” rather than merely local snobbery, that provoke citizen opposition. In relatively stable and settled local communities without other overriding issues, this citizen unease contributes to the passage of policies that attempt to increase public control of the rate and character of housing development. 72 6. The Broader Context of Local Residential Policies Even as the economy has cooled down, housing supply and affordability remain significant problems in California. Judging by previous cycles, even as it is possible that growth will abate for a while, population increases and urban development will accelerate again, and concern for local residential development controls will also resurface. For now, there is not as much housing—particularly multifamily housing— being produced as would seem to be warranted by the high prices and demand levels in many areas of the state. This feature of California growth has gotten worse in the previous decade. Thus, policy observers have sought the causes of California’s housing dilemma. As in previous public discussions regarding growth, local growth management is often assessed a significant portion of the blame for the affordable housing deficit. Our purpose in this chapter is to discuss the rather limited role that local growth-management policies probably play in affecting residential supply and price levels and to highlight broader contextual features of the housing market public policies in the state that may be of greater importance. Is Local Growth Management at the Root of the Housing Problem? For the most part, local growth-management policies of the type we have described in this report—and particularly, those that set overt limits on the amount of housing that can be developed—do little or nothing to further the goal of housing production. Nevertheless, they are unlikely to be one of the most significant factors constraining housing. Given the small numbers of jurisdictions pursuing serious growth controls, the reputed progrowth or neutral attitudes of most city councils, and the 73 reasonable motivations that appear to underlie cities’ growthmanagement policies, we have come to doubt that these local policies are a major root cause of housing affordability or supply problems. And although we do not have direct evidence regarding the relative factors that constrain housing, it appears likely that broader features of California’s political economy, to be discussed in this chapter, are more important than local growth management. The average city in the three regions has adopted fewer than three of the 16 commonly discussed policies that we asked about in the survey— and fewer than one of the seven policies we termed overt restrictions. In fact, nearly a quarter of the respondent cities (24.2 percent) had adopted either no growth-management policies at all, or else only design review standards—a fairly ubiquitous and frequently benign land-use regulation. Three quarters (75.8 percent) adopted fewer than four policies. Although clearly not all growth-management policies are created equal in terms of their ability to limit growth, active growth-management efforts are concentrated in relatively few cities. Of equal importance is the fact that many cities appear to welcome population growth. Of the city managers reporting in our 1998 statewide survey that their cities have vacant land available for development, 60 percent rated single-family housing as a somewhat desirable or very desirable type of development to the local administration; 26 percent even reported that multifamily housing— generally thought to be least advantageous from the standpoint of city treasuries—was seen as somewhat desirable or very desirable for their city. Among those cities pursuing redevelopment, 37 percent of city managers reported that single-family development was desirable for redevelopment areas, and 41 percent said that multifamily housing would be desired in such areas. These favorable orientations to housing in some jurisdictions may be accompanied by an easing of regulatory hurdles. In our survey of planning directors in the three regions, 58 percent indicated that their city had “made it easier (quicker) for residential development projects to be reviewed.” In response to a similar question, 81 percent said that in the past five years, the time required to complete the review of residential projects in the city had either stayed the same or gotten shorter. 74 Therefore, given that some cities welcome housing and a relatively small number shun it, it is quite possible that city growth-management policies may shift the production of housing units in any given area rather than limit it. As observers have sometimes suggested, local efforts to slow regional growth may be akin to pushing down on an inflated balloon in one place: It simply pops up in another. A further consideration is that local growth-management policies are not a new trend. Such policies date back at least to the early 1970s, when environmentalism and antigrowth sentiment first became a major social force (Frieden, 1979). Nevertheless, the passage of these earlier growth management policies did not prevent the state from experiencing major increases in population and housing in more recent decades. Some areas of the state were considered to have actually overbuilt housing during the 1980s. When the early 1990s recession took hold, property values dipped and substantial numbers of homes went on sale in foreclosure. Finally, it is worth noting that local growth-management policies often seem to be passed in reaction to rapid development in a city, rather than in anticipation of it. In these cases, growth management follows a population boom, rather than preventing it. Furthermore, some growthmanagement policies appear to be largely symbolic, aimed at pacifying the local antigrowth movement rather than actually slowing down development (Warner and Molotch, 2000). For example, some city councils have passed rather generous “caps” on annual development that have little chance of being exceeded by the number of units proposed by developers. It is even possible that local efforts to slow growth might generate efforts to hurry the processing of permits to avoid new regulations. The result can be the programming of substantial amounts of new development in a city. Construction that is “in the pipeline” when a growth-management measure passes is generally allowed to be built under a grandfather clause. Thus, some cities continue to see increases in housing development after passing a growth limitation. For this reason, we have not attempted to use the results of our recent survey of local growth controls to try to explain differences in city growth rates. Clearly, a long-term perspective is necessary in evaluating the systematic effect of growth-management policies on local growth. As 75 indicated in Chapter 2, the academic literature has not yet provided definitive conclusions about the effects of local policies. The Timing of California’s Housing Production Slump In Chapter 2, we noted that California’s share of national housing production fell into a deep decline in the 1990s, with only recent and modest recovery. Rather than affixing responsibility for this decline on the relatively small share of cities that have passed growth-management measures in recent years, one might be better advised to examine California’s relative economic performance in this period. In Figure 6.1, we show the state’s unemployment rate (as compared to the nation’s) over the same time period as the state’s housing production (relative to the nation’s). For both ratios, as the line slopes downward, it means that California is doing relatively worse, in terms of either unemployment or housing production. Although the two lines in the figure have different slopes— California’s worsening unemployment did not deteriorate to the same 16 1.0 14 Unemployment 0.9 measure 0.8 12 0.7 10 0.6 8 Housing measure 6 4 2 0.5 0.4 0.3 0.2 0.1 00 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Figure 6.1—California’s Housing and Economic Performance, Relative to That of the Nation California’s percentage of U.S. housing permits Ratio of U.S. to California unemployment rate 76 extent as its housing production—the direction of change from year to year is very similar across the two series.1 The graph reflects the central fact that housing production is very sensitive to broader economic trends—in this case, the state’s recession of the early to mid-1990s, which was longer, deeper, and later than the national recession. With California’s unemployment rate very near the national average in 1990, the state was creating nearly 15 percent of U.S. housing starts. The state’s economic fortunes rapidly deteriorated, however, such that by 1994, the national unemployment rate was 6.1 percent compared to the state’s 8.6 percent. This was the low point of the state’s performance during the decade, and, following a year’s lag, the state’s share of housing production bottomed out as well in 1995, with California producing just 6.3 percent of U.S. housing. What followed was a broad, though uneven, recovery in the latter part of the decade, both in employment and in housing. On neither measure, however, had the state regained its strong position, relative to the nation at large, by 1999. Given the notoriously cyclical and volatile character of housing markets, it is not surprising that the dip in housing production was much steeper than that in unemployment. Although we certainly do not claim that California’s relative economic performance was the only factor affecting its declining residential production in the 1990s, the connection seems close enough to presume that the depth of the state’s recession, from which California has only more recently recovered, greatly affected the housing market. Moreover, the restructuring of the state’s economy and the distribution of different clusters of population in the state might have affected the housing markets in ways that are not fully understood. Less Visible Factors Retarding Housing Supply Few would dispute that macroeconomic factors are central to understanding trends in housing production. However, at least four less visible factors are also likely to play a significant role in depressing residential production: initial low-density zoning, restrictive building codes, the tax treatment of real estate, and condominium defect ____________ 1The correlation between the two data series is 0.96. 77 litigation. Housing advocates would do well to focus as much attention on these “invisible policies” as on local growth-control policies. Initial Zoning When voters or city councils choose to “downzone,” or change local general plans so as to require lower housing densities, attention from the media and interest groups is often intense. However, many communities have evolved in a low-density, large-lot fashion from their inception, without ever instituting any overt growth controls of the type discussed in this report. The existing zoning requirements of such cities, often affluent suburbs, are such that higher-density (and therefore, less expensive) housing construction is not permitted to begin with. In short, some communities have never experienced serious growth-control movements in recent decades, because their zoning traditions (often many decades old) make it prohibitive or impossible to develop large quantities of housing. In this way, local growth seems more incremental, and local streets and services are less likely to be negatively affected by growth. Therefore, growth controversy is less likely to occur. As Warner and Molotch argue, “Following the rules already in place is a powerful form of growth regulation that is not represented on lists of growth controls” (2000, p. 82). Zoning codes that restrict residential development to large lots— regulations that in some cases have been in place since the incorporation of these communities as separate municipalities—are less visible and controversial than pitched political battles over current-day growth controversies. But such zoning may well have had a greater effect on residential building rates than the growth-management policies sometimes passed in desperation after a community has begun experiencing growth-related conflicts. As one scholar states: While new and complex growth management programs have garnered much attention from the planning profession and environmental groups, it is still by placing undeveloped land into large-lot zones that communities most commonly control population growth and land conversion. This practice ensures that land either cannot be developed economically, or can be used only for the most expensive type of structures (Dowall, 1984, pp. 185–186). 78 Indeed, Pendall (2000) finds that municipalities with low-density zoning patterns experience lower levels of rental housing development, and with fewer entry-level housing options add fewer minorities to their populations than otherwise similar cities without such zoning. Neiman (1980) and Farley (1964) present evidence indicating that, over time, communities tend to retain their initial status differences vis-à-vis other localities, which may well be a function of initial zoning. Exclusionary or large-lot zoning appears less common in California than in eastern states. Still, local zoning deserves further study to determine the degree to which it limits housing construction in California or serves to segregate residents by socioeconomic status. It should be recalled, however, that we have as yet no accurate compendium of how zoning policies affect the supply of housing for residential development. It is not only large-lot, low-density zoning that might reduce the supply of land for residential development. If localities over-zone for other purposes, including commercial and retail uses, there is also a potential zoning effect that adversely affects housing supply. Finally, in many cases such zoning ordinances are established early on and are an aspect of the developers’ marketing campaign for communities located in highly desirable areas. Even in the absence of zoning ordinances, such places could accomplish most of the purposes of zoning through privately enacted deed restrictions. If the highly restrictive zoning ordinances of some of California’s elite communities had not been enacted, it is not certain that there would have been in their place more socially diverse communities. Building Codes Just as large-lot zoning can be used to create a “floor” for the price or quality level of housing in a community, restrictive rules regarding the materials or standards used in housing construction can also put upward pressure on the cost of new housing. Local and state building codes and requirements can address everything from the materials to be used in pipes, to seismic standards, to concerns about the thickness of walls, to fire-sprinkler systems. Although such standards typically have very 79 worthy goals relating to the public’s health, comfort, and safety, on the margin these regulations do reduce the ability of builders to produce lowcost housing units by specifying minimum quality standards. The tradeoffs inherent in such regulation are also worth consideration by those who wish to maximize housing production. Tax Treatment In the United States, single-family housing investments are given a powerful incentive: the deductibility of mortgage interest from income for the purpose of calculating income tax liability. In the 1986 federal tax reform, however, incentives for investment in apartment developments were reduced, as Congress elected to simplify the tax code by ending most deductions for commercial real estate. Thereafter, the permitting of multifamily units in California—which had been fairly robust in the mid-1980s—slowed to a relative trickle, by historical standards, and has not yet recovered its previously high levels (Landis, 2000, p. 33; Dunstan and Swenson, 1999, p. 2). Construction Defect Litigation Another factor that has disproportionately affected multifamily construction—the types of units most likely to meet the needs of Californians in the lower and middle portions of the income distribution—concerns litigation over condominium projects. Observers of housing markets in California have argued that (until very recently) judicial decisions seemed to promote lawsuits as the way to resolve even minor disputes between condominium purchasers and builders. The courts have dealt with housing sales on a strict liability basis, as opposed to a caveat emptor (let the buyer beware) basis (Dunstan and Swenson, 1999, p. 4). Purchasers of a new home or condo were allowed to bring lawsuits if construction defects were discovered at any time within ten years of purchase. Lawsuits were particularly common among condo owners because of the presence of organized owner associations (Sanchez, 2000). Many builders claimed that a profusion of frivolous lawsuits brought by condo associations and their attorneys raised builders’ costs and made liability insurance more expensive or difficult to secure. Many plaintiffs’ 80 attorneys, by contrast, argued that the large number of lawsuits was primarily a result of shoddy construction, particularly in the housing boom period of the 1980s (Dunstan and Swenson, 1999; Rojas, 2001). Since numerous parties (subcontractors, insurance companies) can be brought into these suits, they have often been protracted and difficult to resolve (although most are eventually settled out of court). Whatever the roots of the problem, liability insurance has become significantly more expensive for contractors to obtain, and most insurance firms that previously provided this type of underwriting appear to have left the California market (Dunstan and Swenson, 1999, p. 6; O’Toole, 2000). Builders of downtown San Diego condos report paying roughly $20,000 in insurance costs per unit in late 2000 (“Condo Scarcity,” 2000). This litigious situation may have been changed by a decision of the California Supreme Court in December 2000, holding that homeowners (and homeowners’ associations) may not recover negligence damages for construction defects if the alleged defects have not actually caused property damage or injury (Aas v. Superior Court of San Diego, 2000 Cal. LEXIS 9048). However, some attorneys and builders predict that similar lawsuits may continue to be brought under other legal options, and insurance costs will remain high (Sanchez, 2000). Therefore, it is unclear whether this constraint on the supply of multifamily housing will be eased in the foreseeable future. Builders’ groups are seeking legislation that would move toward an arbitration system for resolving claims. Factors Fueling Housing Demand and Prices in California The factors discussed above serve to limit housing supplies. But also important, in the context of contemporary California, are factors that fuel demand. Intense demand and limited supplies work together to generate rapid price increases. Desirability of Coastal Areas California’s affordability problems are concentrated in the coastal counties that are part of the state’s major metropolitan complexes, ranging from Sonoma to Santa Cruz Counties in the Bay Area, and from 81 Santa Barbara to San Diego Counties on the south coast. This geographic pattern is no surprise, given the economic vibrancy of these coastal areas, their rapid levels of employment growth in the latter 1990s, and the longtime aspirations of many Americans to live near the amenity-rich Pacific shore. Although it is true that localities in the coastal counties are more likely to institute growth controls, housing in these areas has long commanded a price premium compared to other areas of the state. During a period of rapid economic expansion, such as that during the latter 1990s, the desire of wealthy, mobile households to live in coastal areas bids up residential prices. The Capital Gains Boom The wealth of many households in California escalated rapidly in the mid- to late 1990s because of the long expansion and high rates of return from the stock market and other securities. Through instruments such as mutual funds and employee stock-ownership plans, participation in the stock market has broadened widely, and many investors experienced unprecedented gains in their portfolios in this period. These gains were readily available for use as down payments on homes, another factor that bid up prices and led to rapid cost escalation in some parts of the state. Researchers at the Federal Reserve Bank of San Francisco estimated the relationship between changes in the stock prices of San Francisco Bay Area high-technology firms and housing price changes in that region, noting, “economic theory would suggest that households experiencing unexpected changes to wealth adjust their consumption of durable goods such as housing” (Krainer and Furlong, 2000, p. 1). They found that stock price changes did help forecast housing price changes, above and beyond standard predictions focusing on employment growth and lagged values of economic indicators. The findings indicate a distinct “wealth effect” on Bay Area housing prices, relating to the boom in hightechnology stocks of firms based in the region. This wealth effect “has helped propel land prices to soaring levels and has created a new class of buyers that are driving a big trend toward luxury production homes with big square footages and long lists of near-custom options” (O’Toole, 2000). 82 Residential Development and California’s Local Fiscal System Another key feature that may affect residential development in California is the state’s convoluted system of local public finance. The property tax in California accounts for a rather low share of overall local revenues. For example, in fiscal year 1997–1998, only 6.5 percent of total city revenues came from the property tax (California State Controller, 2000, p. vii). Furthermore, local governments in California are not empowered to set their own property tax rates. Rather, the state legislature must allocate property tax revenues among the various local governments serving any particular property. Cities often receive a very small slice of the property tax pie; the statewide average share for cities is about 11 percent (Legislative Analyst, 2000, p. 14).2 This situation may have a detrimental effect on the development of housing in cities, although this point has never been empirically demonstrated. Assume that a builder proposes to construct a set of houses, each valued at $250,000, in a city that receives a 15 percent share of the property tax (a figure that is actually above average). With property tax rates limited to 1 percent (assuming no special assessments), the homeowner would pay $2,500 per year, but the city would receive only $375 of that amount. This figure would most likely result in the city “losing money” on the development, because the cost of the services it would provide to each home (police, fire, sanitation, infrastructure, etc.) almost surely would exceed the amount collected in property taxes. Because the city controls land-use decisions within its boundaries, it has a powerful disincentive to accept “too much” housing. Of course, cities have other revenue tools at their disposal. One of the most common, a revenue source of growing importance in many ____________ 2Proposition 13, the property tax limitation passed by voters in 1978, limited property tax rates to 1 percent and annual reassessments to 2 percent (until the property is sold). In addition, the share of the property tax directed to cities and counties declined beginning in 1992–1993, when the state redirected funds to the newly created Education Revenue Augmentation Funds (ERAF) in each county. This was a way for the state to reduce its own general-fund obligations to school districts when it was experiencing a fiscal emergency. Although often criticized, and partially mitigated by other transfers, the ERAF shift remains in effect. 83 parts of the state, is the developer fee or exaction, by which builders provide fees, land, or in-kind donations to help mitigate the public-sector costs of providing for the new development. The fees on a new housing unit are often many times the amount of annual property tax revenues generated by the unit. In a tight housing market, however, builders may be able to pass all or most of these costs on to homebuyers (Dresch and Sheffrin, 1997). Thus, developer fees also work to increase housing costs. Many cities have also attempted to raise their local sales tax revenues by promoting retail development within the community. But this emphasis on retail may also have negative effects on residential development if retail projects are preferred to those that include a residential component, or if land suitable for residential construction is zoned for retail use in the hopes of attracting commercial growth. PPIC’s 1998 survey of 330 city managers in California, regarding their cities’ development strategies, indicated that retail was the most preferred land use for both new development and redevelopment areas, with single-family and multifamily development rated near the bottom of the list of land-use categories (Lewis and Barbour, 1999). Therefore, the state’s system for local public finance, in the post–Proposition 13 era, adds to the burdens faced by residential development. Other “Macro” State Policies Affecting Local Decisionmaking Other statewide policies and rules also affect the decision calculus facing local governments, developers, and slow-growth activists. Here we briefly mention three important examples. Environmental Review The California Environmental Quality Act (CEQA), passed in 1970, requires that local governments ascertain whether development projects will have a “significant effect” on a variety of environmental values—and if they do, prepare an Environmental Impact Report (EIR).3 Originally ____________ 3Completion of a negative declaration (i.e., a finding of no significant environmental impact) may take up to 180 days, and the completion of a full EIR may 84 thought to apply solely to public-sector projects, the California courts have interpreted CEQA as applicable to private development projects as well, when those projects receive a discretionary decision by a governmental entity to award an entitlement (such as a permit). Although a full review of CEQA and its implications for development are beyond the scope of this study,4 our major point is that CEQA, which is mandated by state law, can be an unpredictable element of the residential development process and can introduce a great deal of uncertainty into homebuilding. Although the act was arguably intended to force governments to take account of significant environmental considerations in their planning and decisionmaking, CEQA has become an additional lever outside the traditional planning process for opponents of residential projects. In particular, they can sometimes use provisions of the EIR process to slow down, downsize, or beat back proposed projects, often through litigation. By adding to the time and uncertainty involved in the development process in the state, CEQA may have the effect of reducing housing development.5 The Initiative Process Direct democracy is another feature of the governmental landscape that can powerfully affect the development process. Like CEQA, the initiative process is a statewide rule imposed on local governments. To a greater extent than in other states, California law permits aggrieved ________________________________________________________ take one year from the time the application for the project is complete. (There are provisions for extensions under certain conditions.) As of 1990, about 20 projects received negative declarations (or mitigated negative declarations) for every one that required an EIR. However, California planners subject a much higher proportion of projects to initial studies than in most other states with environmental-review requirements (Landis et al., 1995, pp. xx–xxi). 4For useful overviews of the law, see Curtin (2000, Chap. 6), and Landis et al. (1995). 5Although its procedures are mandated by state law, CEQA permits local governments a fair amount of discretion in deciding what constitutes a reviewable project and what the local criteria and objectives for environmental review should be. By using their ability to issue a mitigated negative declaration, in which conditions are attached to a project before it is declared to lack a significant environmental impact, local governments have been able to push developers into providing certain public benefits for the community or building the types of projects favored by the locality. See Warner and Molotch (2000, pp. 95–100, 133–134). 85 individuals or groups to bring land-use matters before local voters for resolution—including decisions on individual projects as well as broader planning matters. This fact of political life ups the ante on both local government officials and developers as they engage in discussions or negotiations over housing proposals. The possibility that an initiative could overturn an otherwise “approved” residential project increases the degree of risk facing the developer. Thus, the potential threat of a local voter initiative could limit the enthusiasm of policymakers or developers for certain types of projects. For example, larger, denser, “affordable,” or mixed-use projects may offend residents’ sensibilities and increase the probability of a land-use initiative. These initiatives are often broad enough to affect a wide range of future planning and land-use options, meaning that public officials may be just as wary as developers of voter involvement. It is therefore no surprise that, in our planning director survey, cities reporting that future initiatives are likely tend to have adopted a higher number of growthmanagement policies. The idea may be to engage in enough growth control now to attempt to ward off the blunt instrument of land-use initiatives in the future. State Infrastructure Shortfalls There is an emerging consensus that the state has failed to invest in the maintenance and construction of public infrastructure in recent decades and faces an enormous backlog of unmet infrastructure needs. From school construction and modernization to power generation to bus systems, examples are rife of public goods that are underprovided or strained to capacity (Dowall and Whittington, forthcoming; Dickerson and Silverstein, 2001). A particularly relevant example concerns roads and transportation. We saw in the last chapter that cities with longer commute times tend to be less favorable to housing development, and more prone to controversy over residential growth. The state’s ranking at or near the bottom of the 50 states in transportation expenditures per capita over the 1990s has done little to inspire the confidence of local residents and officials that transportation improvements will be forthcoming. As Landis and Kroll (1989, pp. 157–158) concluded from their study of “growth wars” in 86 Southern California, “More than any other single factor, traffic congestion served to mobilize anti-growth sentiment,” although they point out that voters themselves have often been opposed to special transportation taxes and freeway-construction programs. Nevertheless, transportation and related infrastructure problems, such as school overcrowding, likely reinforce the wariness over growth among local interests. Recent data from the Inland Empire Annual Survey of residents in fast-growing Riverside and San Bernardino Counties indicate that respondents who give low ratings to the quality of local public services, and those who think traffic conditions are bad, are more likely to believe that growth has been bad for the region. For example, among Inland Empire respondents who strongly agreed that population growth was good for the region, 42 percent indicated that traffic was a problem, whereas 69 percent of the respondents who strongly disagreed that population growth was good for the region felt that traffic was a problem (Bockman, Neiman, and Sirotnik, 2001). A Broader Policy Dilemma In assessing the effect of public policies on housing markets, there is a kind of meta-question that stands astride this entire issue. What is the optimal level of housing in California? And from what standard do we derive our answer—the needs of the poor, the tastes of those preferring higher density, big city living, or the partialities of individuals inclined toward standard suburban tract living? Or should we sympathize with the objectives of those concerned about housing the state’s workers in suitable and healthy environments? Should the housing demand of perhaps millions of undocumented immigrants also be included in projections of future housing needs? Is there any optimal level of housing production about which analysts and policymakers, much less citizens, are likely to agree? On the one hand, the values of those who fret about the need for further growth find housing costs too high and supply too low. Those who are alarmed at the huge California population and its continuing growth, along with the variety of service, resource, and ecological strains facing the state, feel that housing construction is already excessive and poorly planned. Freemarket devotees insist that local regulations distort housing markets and 87 deprive entrepreneurs and property owners of their rights. “Smart growth” and environmental advocates assert that housing markets are artificially stoked by an array of policies that contrive to escalate housing demand. In some ways, owner-occupied housing is among the most lavishly subsidized of all consumer choices, given the tax deductibility of mortgage interest, federal programs to insure mortgages, and other policies. Some growth critics argue further that local policies fail to account for the externalities imposed by the collective effects of typical large-scale residential development projects. In any case, California continues to have a powerful attraction to people throughout the world. In flush economic periods, the state is almost always likely to find that growth will stress both the public and private sectors’ capacity to satisfy demand. In short, there is no consensus regarding the substantive dimensions, causal dynamics, or remedies of the problem of housing cost and supply. Consequently, it is very difficult to rigorously determine the role of residential development controls in the overall mix of factors that affect housing markets. Summing Up Local growth-management policies are often singled out for criticism for contributing to California’s housing problems. Our intent in this chapter has been to illustrate that such policies probably play a relatively small role in the context of the overall housing market—which is not to say that these policies are always benign or never have an effect. Most cities have few growth-management policies, few cities set overt limitations on housing development, and many welcome residential growth. Other, broader factors are likely to play a larger role in contributing to the challenges facing housing policy in California. Some factors appear to limit housing supply, including low-density zoning, restrictive building codes, the current tax treatment of commercial real estate, and litigation and liability insurance problems facing the condominium market. At the same time, demand for housing has been intense because of the continuing desirability of coastal areas of the state, and the wealth gains, particularly capital gains, that occurred for many California households in recent years. 88 Under the state’s fiscal rules, California cities may find that accepting housing works against their fiscal self-interest. CEQA and the initiative process are two other statewide factors that provide tools for housing opponents and that introduce uncertainty for builders and local policymakers. Meanwhile, the state has fallen behind in providing and maintaining its infrastructure, including transportation facilities, which means that growth is likely to create additional strains and controversies in many communities. Looking at the factors arrayed against housing production, local growth management appears to be a relatively small part of the picture, although future research is necessary to clarify its role. In some respects, it is remarkable that California localities accommodate as much new housing as they do. 89 7. Refocusing the Growth Debate In considering California’s posture toward housing and other types of growth, one must first acknowledge that the state’s “growth policy” is really the amalgamation of hundreds of local growth policies. These local policies, in turn, are devised in the face of the very real conditions— social, economic, fiscal, and environmental—faced by local policymakers and residents. One particular class of local policy actions—growth controls— retains a particular notoriety as a potential drag on housing development in the state. Our survey of city planners indicates, though, that residential growth-management policies are relatively uncommon among California municipalities. Moreover, the most restrictive options, such as annual caps on building permits, are embraced by only a small handful of communities, whereas certain policies to link housing development to infrastructure capacity, or to shape the form or location of growth within a city, are somewhat more common. The most notable determinant of the level of local growth-management activity is the level of public opposition to growth in the city. In addition, local conditions, such as commuting times and jobs/housing balances, help determine cities’ orientations toward housing and growth. Even taking account of all these factors, there appears to be a regional ethos in the San Francisco Bay Area supportive of a somewhat higher level of growth management, perhaps reflecting a diffusion of the growth-control concept in a region that was among the first to experiment with such policies. However, even in the Bay Area many communities seemingly do little to overtly “control” growth. Additionally, even if some of the localities with higher development controls reduced their number or stringency, it is not clear that these places would suddenly become havens for moderate- or low-income 91 housing. Some of these communities are likely to remain high-cost exclusive places, because the market demand for access to them would produce very high land rents in any case. Residential growth controls are merely a subset of the policies and other factors that affect homebuilding rates and, in our opinion, have received disproportionate attention.1 Other local policies, particularly zoning and general plan standards, are more fundamental in determining the amount of growth that can be accommodated. Statewide factors, such as the system for financing local governments, the presence of the initiative process, CEQA review mandates, and frequent litigation over condominium construction, also should not be ignored as factors that potentially hamper housing development. Ongoing market factors, such as the desirability of coastal communities and the escalation of household capital-gains wealth, have probably played a major role in fueling the rise of housing prices in many parts of the state. Finally, although we do not endorse or condemn growth management, it is important to note that fostering vibrant local governments, which have important functions and responsibilities in the context of a democratic society, is also a worthwhile policy objective. Providing local populations with the ability to make important decisions about what sorts of public policies they wish to promulgate to shape the physical development of their area is a critical feature of any serious commitment to meaningful local governance. Of course, this poses serious puzzles regarding how to balance localism with regional, state, and even national objectives. A high regard for local government surely must be weighed against high regard for equity, economic growth, and ecological harmony. ____________ 1The recent Department of Housing and Community Development (DHCD) report on housing constraints in California, Raising the Roof (2000a), is similarly somewhat equivocal regarding the likely role of local restrictive policies in slowing housing production. Although the authors of the report note that the entitlement process for new housing can be long and unpredictable in California, much of the uncertainty in the process rests with interpretation of CEQA, a state-mandated law. In the 46 case studies of development projects examined in the DHCD report, the approvals process, though complicated, “[did] not, as a rule, result in interminable approval delays or in ad hoc unit and density reductions ” (p. 109). 92 Yet, as we have seen, much of the public discourse and perhaps even the research on these matters occur in the absence of agreed-upon principles. For example, how do we assess the tension between local and regional housing objectives? What sorts of norms should guide decisions in such an arena? What is meant by adequate housing production? What does a free market in housing really mean, given the many public policy choices that fuel and intensify housing demand? Why Do Conflicts over Local Growth Management Have Such Visibility? Given our assessment that growth controls, per se, are probably not a major cause of California's housing difficulties, why is there so much attention directed at this topic? On the one hand, there are some fascinating legal issues associated with the question and there is an enormous body of legal research and litigation associated with local efforts to manage growth. In addition, conflicts over development projects or ballot fights over growth-control initiatives are interesting fodder for journalists covering local government—“sexy” stories within an often mundane world of local council meetings and planning commission hearings. Norms of news reporting may sometimes lead to distorted coverage of the development issue, however (Kaniss, 1991). Beat reporters generally define a story in terms of a current conflict (e.g., a developer and neighbors of a project arguing at a planning commission hearing). Broader trends, structures, or institutions affecting regional development are less often seen as timely news (e.g., a city’s low-density zoning or setback requirements and the effects of these on the economic viability of projects, the evolving balance between employment and housing in an area, the long-term effects of a sewer-system extension on growth patterns). For many observers, moreover, the entire issue of managing development is symbolic. On one side, environmentalists, anti-sprawl advocates, historic preservationists, and resource conservationists view typical suburban residential development as a source of waste, environmental degradation, and increased levels of separation by race, economic class, age, and lifestyle. On the other side, there are the 93 defenders of property rights, trade unions, building contractors, and realtors who see local development controls as anti-working class, hostile to traditional American values, and excessive government intrusion into free markets. Additionally, liberals concerned about the need to provide adequate housing for the poor and moderate-income households are often critical of local development controls that hinder multifamily and higher-density housing development. In short, for many critics and advocates of local growth controls, such policies are emblematic of concern for larger issues than those occurring in any given community. Finally, there are the short-term, immediate financial stakes that housing developers and property owners have in their individual projects or land values. Thus, real estate interests naturally and sincerely attach great significance to such controls and help elevate the topic and contribute to its salience on the policy agenda. The Challenge for Cities: Successfully Managing Conflicts over Growth Our conclusion is that local growth-management policies are by and large understandable responses to the cross-pressures city policymakers feel—not, for the most part, narrowly self-interested actions motivated by animosity to newcomers or by property owners’ collusion. In addition, broader forces in the housing market and the state policy environment probably have a more central role in explaining California’s high housing costs and slow production—which is not to say that local growth control policies never hinder homebuilding. Findings in this study indicate that conflict over growth, if allowed to escalate, can produce even higher levels of local restrictions on residential development. Higher levels of conflict associated with such events as city council turnover, rancor in public meetings, recall elections, and local ballot initiatives are often connected to the formation of local groups, which then fight it out over growth issues. If unresolved, these conflicts are likely to intensify and become a persistent feature of local politics. Moreover, the kinds of policies that derive from highly charged, strident decisionmaking environments may be more restrictive than would otherwise be the case. In short, our various indicators of 94 local political controversy over growth are connected in important ways to the number of growth controls adopted. Local controversy over growth issues is also strongly related to the existence of local citizen initiatives, which are typically blunt policy instruments. Emergency controls over development, such as growth moratoriums or ballot-box zoning, are poor substitutes for careful, realistic, and visionary long-range planning. To avoid potentially paralyzing controversy over growth, cities and counties need to find ways to manage community conflict before it erupts or, if it does occur, to have local officials take the lead in arbitrating competing demands if at all possible. Conducting thoughtful general planning efforts and revisions—including “specific plans” for important or sensitive parts of the community—and engaging a broad segment of the public throughout this process seem key to achieving more local consensus. A public that feels shut out from the planning process or is uninformed about its details is more likely to react to development proposals in ways that potentially hamstring builders and limit the discretion of local officials. In cities and counties where officials frequently amend or ignore their general plans when convenient for fiscal or political reasons, the patience of local residents can quickly wear down. In short, local governments must convince their residents that they do have a sound plan to sensitively accommodate future growth—and there is room for the state government to assist in these efforts. If the management of conflict is carefully designed, and if there is consistency in implementing local policy, it is much easier to assert that citizens have obligations to be fair and to permit development that is in accord with local policy. Two State Policy Challenges: Local Government Finance and Infrastructure Shortfalls Thoughtful, skillful, and consistent local officials can make major inroads into the management of growth conflicts. But some problems that inflame negative perceptions about housing development cannot be dealt with adequately at the local government level. It is critical to have some sense of the overall context in which local growth-management efforts occur in California. 95 The city manager survey provides substantial indirect evidence of how fiscal motivations apparently affect local land-use policies. Local governments operate within a fiscal system that helps define the costs and rewards of various types of growth and development. The state’s fiscal system for local governments does little to soothe local anxiety about residential development. This is a central fact of life for city policymakers, who must focus to a large degree on ensuring that resources are sufficient to provide necessary local services to sustain the community. Local governments’ impulse to actively curtail residential development should be reduced if residential development is less fiscally problematic or if the state actually makes residential development a fiscally positive outcome. California’s system for funding local governments currently skews incentives for land development. The limited role of the local property tax and the lack of local control over its rate mean that local officials often must seek other ways to financially support the services required for new development. Escalating developer fees and special assessments are one option, increasingly in use since the passage of Proposition 13, but these might have disproportionate effects on new housing and contribute to higher home costs. Moreover, reliance on today’s new-home fees to pay for services for last year’s development is a potentially risky financial strategy if an economic slowdown hits, as a number of cities have discovered (Arax, 1999). A major alternative strategy for raising revenue—pursuit of local sales tax revenues—tends to favor retail development, possibly at the expense of housing development. The skittishness about residential development is in some cases exacerbated when localities find themselves in competition for retail development and begin to allocate more local resources, whether fiscal or in the form of providing greater amounts of lower-cost land, to prospective commercial developers. Although many city officials responding to our survey indicate a willingness and sense of responsibility to accommodate new housing, one can expect them to go only so far in the face of the fiscal disincentives facing residential development. Unless the state increases the share of the property tax directed toward general-purpose local governments, many 96 cities may remain understandably wary about major increases in residential development. Second, insofar as there are shortfalls in the maintenance or expansion of infrastructure, citizens are likely to regard population growth, and thus housing construction, as a threat to local services or quality of life. It is no surprise that local policy reflects these concerns. Our findings indicate that cities whose residents experience long commute times, for example, are less interested in further residential development and more prone to resident and city council opposition to growth. Such local conditions, rather than the rate of recent population growth itself, appear to have a dampening effect on local government enthusiasm for housing. This finding is a strong hint that California must address its infrastructure problems if it is to convince its residents and local officials that new housing (and the associated new residents) will not reduce existing residents’ quality of life. Major shortfalls in the state’s infrastructure exist not just in the electrical power system but in transportation, school facilities, parks and recreation, and water and sewerage provision. Improving these systems, while still maintaining the state’s environmental quality, which is also highly prized by residents, is perhaps the major challenge facing state policymakers in the effort to accommodate new populations. If traffic congestion and existing population density are not to be hindrances to housing development, not only must infrastructure be improved but homebuilders and policymakers must find ways to integrate new housing more gracefully into the fabric of built-up communities. This goal requires attention to aesthetic and environmental sensitivity, support of cities’ commercial functions (such as in downtown mixed-use areas), as well as transportation linkages such as accessibility to public transit and pedestrian amenities. Successful examples of medium- to high-density housing developments that do not disrupt the local quality of life—and perhaps make a net improvement to the community—provide the most powerful ammunition to supporters of new housing in existing urban environments. 97 Policy Options to Put Housing on a More Equal Footing The central focus of this study has been on the nature of, and motivations for, local growth-management policies, not housing policy per se. Of course, the reason we concern ourselves with local residential development controls is because these policies are so often linked to various housing problems and are sometimes the targets of reform and legal attack. However, we have argued that among the panoply of factors and other policies that affect housing supply and price, local development controls are probably not near the top in importance. Nevertheless, having noted that many factors in the state create potential problems for the production of housing, especially affordable housing, what other policy options or leverage might the state have regarding its housing-production woes? The possibilities we discuss in this section might reduce pressures among localities to be hostile or unfriendly to residential development. Our discussion, which is necessarily suggestive and brief, is guided by two underlying principles. One is that the “problem” of growth controls is better managed by focusing on other features of the housing market that probably adversely influence construction of more housing. The second is that if policymakers are to curtail the authority of local governments to take on important issues in their communities—such as defining a vision for community development and remediating the sideeffects of growth—then there should be a substantial public benefit produced. One seemingly straightforward policy option is to increase the state and local funding available for affordable housing programs, including funds for construction and rehabilitation, down payment assistance for households of limited means, and rent subsidies. Without directly addressing the structural constraints to residential supply that we identified in Chapter 6, however, such expenditures do not fundamentally alter the nature of the problem, however ameliorative the expenditures might be for those households receiving benefits from them. For example, recently developed policies that provide special housing assistance to schoolteachers and public safety employees, in California 98 and nationwide, are a special variant of this approach; they provide useful assistance for a relatively small number of workers in these favored professions, but do nothing to address the fundamental gap between demand and supply for California housing. It should be pointed out that the state has passed several important laws in recent years geared at increasing housing production and confronting affordability obstacles. These are reviewed in detail elsewhere (California Senate Office of Research, 1999; Department of Housing and Community Development, 2000b), but among the most significant were laws to • Allocate $100 million in incentive grants to local governments that increase housing supplies, with an emphasis on balancing jobs and housing, • Increase the California Housing Finance Agency’s revenue bond capacity, • Continue the state’s tax credit for development of low-income housing, • Streamline local approvals for affordable housing projects, and • Increase funding for farm worker housing. Such legislation—and the formation of a Housing Task Force by Governor Davis—indicates that state leaders have a clear interest in the state’s housing market and its affordability problems. In the same spirit, we list below some additional policy options, related to the issues raised in this report. These are presented not as recommendations but rather as considerations worthy of study and analysis by state and local lawmakers and others in the policy community. Our focus is on policy alternatives that could increase the amount of information available to those in the development process, address some of the “unpredictability” of that process, or reduce or mitigate local growth conflicts or controversies. Full “Planning Disclosure” for Homebuyers Under current law, a person selling a home must disclose known defects and hazards to the purchaser. However, homebuyers sometimes make the largest investment of their lives with little knowledge about the local plans for, or potential future development of, their neighborhoods. 99 This information shortfall can lead to protests, litigation, and political controversies when builders propose constructing a new development adjacent to an existing one—when residents thought they were buying into a project that adjoined open space. Individual cities and counties, or the state legislature, could pass legislation requiring that any residential real estate transaction include information about the land uses permitted within a certain distance of the home (perhaps a quarter mile), the zoning map for that area, and the permitted maximum future densities and numbers of units. (Some cities apparently already require that some of this information be disclosed to purchasers, but that is not uniformly the case.) The local generalpurpose government could prepare standardized information for homebuyers, to be distributed by real estate agents handling the sale. By signing this disclosure, homebuyers would signify that they know and acknowledge the plans and possible future development scenarios of the neighborhood they are buying into.2 This proposal flows directly from our finding that local controversy and conflict over growth are likely to generate housing restrictions, and that such restrictions might very well be excessive. One should not underestimate the important symbolic effect of a formal notification to a homeowner of the fact that adjacent land is planned for particular uses. When residents oppose something that they formally accepted at the time they purchased their homes, this prior acceptance would undermine their subsequent opposition. Naturally, such a policy should not deprive citizens of their right to petition for changes in policy, but this low-cost requirement might have some important benefits. In a similar vein, nearly 100 cities and counties currently have “right-to-farm” ordinances, in which home purchasers in agricultural areas are informed that they may be subject to the effects of nearby farming operations, such as spraying, noise, and odors. Although right-to-farm ordinances do not prevent a property owner from suing a nearby farm owner, they do serve ____________ 2Section 1102.6 of the California Civil Code already requires that a great many facts about the property be disclosed to the purchaser. Sections 1102.6a and 6b specifically authorize cities and counties to require additional disclosure statements, even providing as an example, “Adjacent land is zoned for timber production which may be subject to harvest” (California Civil Code, 2001, p. 170). 100 an important informational and educational rule (Wacker, Sokolow, and Elkins, 2001). Land Supply Monitoring Some observers have made the case that the supply of buildable land that can be provided with urban services appears to be dwindling in California’s major coastal metropolitan areas (Southern California Studies Center, 2001). Other analysts have found that “California has more than enough raw land to accommodate projected housing growth (at current densities)” for decades, with only Los Angeles, Orange, and San Francisco Counties lacking land capacity to meet projected demands (Department of Housing and Community Development, 2000a, Chap. 3). If land supply is constrained, and local governments pursue growth boundary policies that set additional land off-limits for development, land costs may become an ever-larger obstacle that could send housing prices spiraling higher. At present, however, there is insufficient unbiased information about the amount of privately held, residentially zoned land that has or has not been subdivided and could reasonably support housing. The science of “land supply monitoring” is still in its infancy. Nevertheless, if regional planning agencies could work with local governments to systematically track the amount of such land over time, policymakers and researchers would gain a much better sense of the potential inventory of residential land (including infill development) and the challenges that this constraint does or does not pose in various markets. Currently, the state Department of Conservation tracks the conversion of farmland to other uses. One could certainly make the case that state policymakers should consider providing support for agencies to track the supply of serviceable urban land as well, perhaps on a pilot basis at first. Addressing Condominium Litigation As noted in Chapter 6, litigation over alleged construction defects in condominiums is one factor widely believed to increase costs and decrease production of this important type of multifamily development. Although there are no accepted data on the exact degree to which such litigation has contributed to the construction shortfall in California, 101 attached multifamily developments are the housing sector that has declined the most in production since the 1980s. The recent California Supreme Court decision that limits liability awards in cases not involving injury or property damage may be one step toward making such projects more insurable.3 Others at the state level have been involved in negotiations to attempt to craft a longer-run solution. One approach proposes ten-year warranties for purchasers of new homes, with provisions for arbitration (rather than litigation) in cases of liability disputes. Increasing the predictability and dependability of condominium and townhouse transactions promises to benefit a number of important parties, including consumers, builders, insurers, and community associations. CEQA Streamlining and One-Stop Permitting Although this is not the place for a thorough investigation of the California Environmental Quality Act, it is worth reiterating that CEQA review can be one of the most unpredictable elements of the residential development process. Those interested in reform in this area might focus on streamlining CEQA to remove from the review process “environmental” impacts that can be regulated effectively through a city’s general plan or ordinances, such as parking, noise, public safety, and fiscal effects. On a broader level, more attention might be given to preparing thorough master environmental impact reviews at the time of general plan revisions or as part of the creation of “specific plans” for specified parts of the city. Residential and other projects that are consistent with those plans could then be “tiered” off the master EIR. Some jurisdictions have reported success in moving toward a one-stop permitting process, under which CEQA and other reviews are consolidated into a single process, based on conditions and mitigation measures specified in city ordinances or the general plan. See Landis et al. (1995, Chaps 5–6) for examples and reform options. ____________ 3Senate Bill 355, introduced in 2001, would have overturned this decision and widened the basis for defect lawsuits. The bill did not pass for enactment in 2001, but remains active. 102 Local Government Finance Reform The issue of the incentives created by the state’s system of financing of local governments, including the “fiscalization of land use,” has been discussed and editorialized upon so often that we do not wish to recapitulate the entire argument here (but see Lewis and Barbour, 1999, Chap. 6). In our survey of city managers, the desire for retail development, the lower degree of interest in housing, and the importance of augmenting sales tax revenue were vividly apparent. Numerous parties, ranging from groups convened by the state controller, to the Legislative Analyst, to the statewide associations of cities and counties, have proposed specific options for altering the local revenue-raising system, with some more complex than others (see Senate Local Government Committee, 2001); the legislature has also convened a number of groups to study the issue and propose remedies. Nearly all of these are based upon principles including reducing local reliance on a sales tax based on local retail transactions (which is seen as encouraging retail development at the expense of housing and industry) and increasing the share of property taxes directed toward cities and counties (thereby, presumably, making housing development more fiscally rewarding). Some have also suggested restoring some degree of authority over local property tax rates, which was lost as a result of Proposition 13 in 1978 (Legislative Analyst, 2000). Although such proposals have yet to result in major, long-term reform legislation, the subject remains an open one in Sacramento. We hasten to point out that our statistical results in this report offer no specific support for the idea that cities with a larger share of the property tax will take a more housing-friendly policy position. Indeed, the variable relating to the city’s percentage share of the property tax was not a significant predictor of any of the policy outcomes or orientations we investigated.4 Nevertheless, it is widely believed that low property tax rewards for housing hinder its chances in the development process. This topic merits further study. ____________ 4In fact, cities with one or more of the most restrictive growth-management policies have essentially the same estimated share of the property tax (13.2 percent on average) as cities without restrictions (13.4 percent). 103 More Support for Urban Infill Projects Many observers point to the potential of urban infill projects in established neighborhoods as potential win-win propositions. They can help improve and bolster older neighborhoods while providing additional housing units close to employment centers and making use of existing infrastructure. However, such projects face a number of constraints, which may involve environmental cleanup, land site assembly, high land costs, or complex financing for mixed-use projects. As a consequence, many builders without the resources necessary for such projects may choose to develop less-complex projects on vacant land at the fringes of metropolitan areas. Redevelopment policy is one major tool cities can use to address the issue of land costs and site assembly. Most infill housing in California in recent decades has received assistance from redevelopment agencies, and nearly one in 20 housing permits statewide went to redevelopmentfunded projects between 1988 and 1996 (Department of Housing and Community Development, 2000a, p. 69). Improved or increased housing facilities are often an important element of successful revitalization of a blighted area and can have synergistic effects for retail or commercial facilities developed as part of the project. The state’s Department of Housing and Community Development, or the Office of Planning and Research, could commission a study and prepare publications on best practices, illustrating and publicizing how mixed-use projects have been gracefully integrated into existing neighborhoods and downtowns. Recently, some regional agencies have begun providing financial rewards and technical assistance to cities and counties where transitoriented housing developments are built. Such programs can help address housing shortages and transportation problems simultaneously, by increasing the supply of new units in proximity to mass transit stops. For example, the Bay Area’s Metropolitan Transportation Commission provides funds to local governments in which multifamily developments of at least 25 units per acre are being planned within a one-third mile walk of a bus or rail route that provides frequent service. The funds, provided on a per-bedroom basis, may be used for relevant transportation 104 projects and upgrades in the area of the project.5 To provide a more powerful incentive for such developments, the level of funding may need to be increased far beyond current levels (currently, a one-time $1,000 to $2,000 award per bedroom) and the allowed uses broadened beyond transportation. Similarly, the state legislature passed a law in 2000 allocating $5 million in “predevelopment loans” to local governments or developers for proposed projects near a transit station (Department of Housing and Community Development 2000b). If this small program provides useful results, additional state legislation could provide expanded funding for a program of larger scale, perhaps to be administered by regional planning agencies and councils of government. It is worth noting, too, that in some parts of the state, the weakening office market and the tight housing market are giving developers a financial incentive to add highdensity housing to their commercial project in urban areas (Sanchez, 2001). Density Bonuses, Inclusionary Housing Requirements, and the “Builder’s Remedy” Under California’s Housing Element Law, originally passed in 1969 and modified several times since, regional councils of governments must periodically undertake a housing needs assessment and determine housing “allocations” for each city and unincorporated county area, including allocations of very low, low, and moderate-income housing.6 Furthermore, the housing element of each jurisdiction’s general plan must specify goals, policies, and programs for housing development and conservation, for all income segments (Curtin, 2000, pp. 10–12; Calavita and Grimes, 1998). Despite these requirements, there have been few if any penalties for lack of implementation of the housing goals or failure to achieve them. In short, these “fair share” requirements of housingrelated planning have been more process- than results-oriented. In 2001, ____________ 5This program was modeled on one developed by the council of governments in San Mateo County. For information, see http://www.mtc.dst.ca.us/projects/livable_ communities/lchip.htm. 6The local allocations must add up to the regional allocation imposed on the region by DHCD. 105 bills pending in the state legislature seek to provide legal ammunition for builders or others seeking redress from local governments that restrict affordable housing developments.7 In seeking to create more workable mechanisms for the production of affordable housing, state and local policymakers may wish to devote renewed attention and analysis to inclusionary housing programs and density bonuses. Inclusionary housing programs are local policies requiring that developers of larger residential projects provide a specified share of “affordable” units within the project. (In many cases, the builders are given the alternative of paying per-unit fees into an affordable housing trust fund.) According to Calavita and Grimes (1998, p. 158), 75 local governments in the state had inclusionary housing programs as of early 1996. Their interviews with city planners in the San Diego area indicated that many communities with inclusionary housing requirements adopted them to demonstrate tangible results regarding their housing elements and thereby avoid lawsuits from builders or housing advocates or conflicts with state government. However, critics of inclusionary housing requirements argue that they do nothing to increase the net number of housing units—and thus, do little to battle California’s undersupply of housing—unless the developer is permitted to build more units than would have otherwise been the case (Gruen, 2001). Otherwise, reducing the cost of some units in a development may merely shift the costs of subsidizing those units to other purchasers in the complex, whose purchase prices may be raised accordingly. A density bonus is the mechanism that allows additional units to be constructed. A density bonus is a concession granted to a builder who agrees to build a development in which a certain threshold share of units meets affordability requirements. The bonus allows the construction of additional market-rate units over and above the number otherwise ____________ 7For example, AB 369 (Dutra) authorizes courts to order local governments to pay plaintiff’s attorneys’ fees in cases where affordable housing developments are denied without required written findings. It was passed and signed into law. SB 910 (Dunn) would authorize courts to order financial penalties for localities failing to comply with the state’s housing element law, with the ultimate penalty being a reduction of certain state subventions to the jurisdiction. The Senate passed this bill but it failed to emerge from the Assembly in 2001. 106 permitted under local plans and zoning. Existing California law requires that localities provide 25 percent density bonuses (and an additional development incentive) to builders of projects that reserve 20 percent of units for lower-income residents, or 10 percent for very low-income residents, or 50 percent for senior citizens (California Government Code, Section 65915). Nevertheless, a developer generally must initiate the decision to set aside affordable units and apply for a density bonus, and actual use of the density bonus approach is uneven across California. According to a 1999 survey by the Office of Planning and Research, at least 215 cities and 30 counties in California have ordinances that “encourage” developers to use the state’s density bonus law (Governor’s Office of Planning and Research, 2000, p. 51). One policy option would be to create a more routine program of density bonuses and inclusionary housing requirements, with uniform thresholds of affordable units, for jurisdictions that show particularly high levels of unmet needs for affordable units. The state could require inclusionary housing programs with accompanying density bonuses for such communities or for communities whose housing elements are considered to be out of compliance. To ensure that the net stock of housing units actually increases, the density bonuses would be applied to all such projects if the affordable units were built on-site (rather than paid for in-lieu). Provisions could be applied to guard against too-rapid overbuilding in any given community—for example, a “trigger” would be reached when the number of housing units receiving entitlements to be built reached a certain percentage of the community’s initial housing units. At that point, the inclusionary housing and density bonus requirements would be suspended. Drawing upon New Jersey’s experience with its statewide mandated fair-share affordable housing program (Haar, 1996), developers could be permitted to sue recalcitrant communities whose housing elements are out of compliance or that have made little progress in producing affordable housing. The trial court would be authorized to impose a “builder’s remedy” for the provision of affordable units, mandating bonus market-rate units to the developer in exchange for the developer’s provision of a certain share of affordable units. Marrying the public interest goal of affordable housing to the profit motive of builders could 107 have a more systematic effect than simply requiring that localities write plans for housing (Haar, 1996). In Massachusetts, developers who propose a housing project in which at least 25 percent of the units are reserved for affordable housing are exempted from local zoning laws if the community has not met its affordable housing production requirements under state law. We are not completely persuaded that this mandated density bonus is the best approach for California to take, but merely put it forth as meriting further study. The proposal raises major issues of local autonomy and of reimbursable state mandates. However, housing element requirements and the provision of housing have long been held to be matters of overriding “statewide interest.” The litigation option is less flexible and more adversarial than we might like; as a deterrent, however, it could encourage pro-housing actions and planning changes by cities and counties—including jurisdictions that have engaged in such low-density zoning that lower socioeconomic groups effectively cannot gain entry to the community.8 However, this approach requires substantial confidence that there is an accepted method of making equitable and rational measurements of what a community's share of affordable housing ought to be, an assumption about which there is considerable controversy. Housing Action Coalitions Our final policy discussion relates as much to the private and nonprofit sectors as to government and represents a possible means to moderate local growth-related conflicts. In recent years, several San Francisco Bay Area communities have seen the formation of so-called housing action coalitions that lobby for appropriate housing developments. At local planning commission or city council hearings, ____________ 8The New Jersey experience shows that use of the builder’s remedy might require the creation of a set of specialized courts to handle land-use and affordable housing matters—an option that has been considered in prior reform efforts concerning land-use regulation in California—with the possible appointment of “special masters” trained in planning and real estate issues. Courts would need to consider whether the property held by the builder was appropriate for the project with the affordable-housing component (accessible to services, transportation, and employment, for example). 108 where project opponents often appear in full force and project proponents include only the developer’s staff, these pro-housing groups attempt to make their voices heard. Housing action coalitions, which often include affordable housing advocates, members of the clergy, architects, and members of the business community, provide a process for developers to “nominate” their projects for the group’s support. The coalition reviews the proposal in detail, ideally at an early stage, examining the proposed project’s relationship to neighborhood design, public transportation options, and affordable housing provision, among other criteria. The coalition may work with the developer and neighborhood groups to suggest modifications to the project that might allay neighbors’ concerns, without compromising housing production goals. If the project wins the group’s endorsement, and the group has credibility with local policymakers, the endorsement may be treated as something like the Good Housekeeping seal of approval. In communities with particularly contentious approval processes and high degrees of citizen activism in the development process, this is a promising avenue to improve project design and community consensus. Foundations, business associations, and the like could provide seed money to help start up and support incipient pro-housing interest groups. At the state level, relevant agencies concerned with housing, such as the Department of Housing and Community Development, the Office of Planning and Research, the Treasurer’s office, and others, could convene a continuing working group in collaboration with representatives of statewide housing groups, the development industry, and local government associations. This group would function as a kind of meta-level housing action coalition. The governor’s creation of a housing task force could be the basis of a more permanent working team in Sacramento. The group could identify obstacles and opportunities in state-level programs and administrative procedures that affect housing production (particularly multifamily and affordable) and suggest changes to streamline or assist in housing development. Recognizing and providing resources for such an effort would depend on a continuing, 109 articulated commitment to the housing goal on the part of the statewide elected officials. A Final Thought As this is being written, the state’s booming economy of the late 1990s has slowed down, and housing price inflation has begun to moderate. If indeed the state slips into recession, or if growth and housing issues no longer seem as pressing, some may turn their attention to other issues of the moment. However, a period of slower economic growth is not the time to turn away from the growth and housing dilemma, as there is every reason to believe that if there is a lull, that is all it will be. Issues of growth and conflicts surrounding residential development will inevitably appear again. Rather, when the real estate market cools—perhaps along with housing policy conflicts—seems a particularly propitious time to address the knotty problems of housing affordability and production, as well as local planning to avoid future growth controversies. Experience shows that when economic growth resumes, the political pressure cooker in which these issues are decided can make long-range thinking and policy action difficult. 110 Appendix Multivariate Data Analyses: Variables and Results This appendix provides detailed results of multivariate data analyses to support the discussion in Chapter 5. That chapter interprets the results, whereas this appendix is intended for readers who want a more detailed sense of the modeling and statistical estimations. After an initial section reporting data sources, the analyses in the appendix proceed in the same order as they are discussed in Chapter 5. Data Sources Each dependent variable in these analyses is derived from the answers to survey questions by respondents of our 1999 planning director survey or, in a few cases, to our 1998 city manager survey. More detail on question wording will be provided as we address each model. A full tabulation of the results of these surveys is available in Lewis and Neiman (2000) and Barbour and Lewis (1998), respectively. Regarding the independent (predictor) variables, the following variables were derived or calculated for each municipality using newly released data from the 2000 Census: • Percentage of housing that is owner-occupied, • Percentage of housing units that are recreational or seasonal, • Percentage Hispanic (of any race), and • Percentage black (not Hispanic).1 ____________ 1In the 2000 Census, respondents were permitted to check more than one category identifying their race. For the calculations in this report, we elected to use the “low estimate” of persons who checked the racial category in question (in this case, black) and no other categories. Note that racial-category percentages calculated in this manner are very closely correlated (r > 0.9) with city racial percentages from the 1990 Census, in which respondents could check only one racial category 111 Other Census variables are unfortunately not yet available for 2000, and therefore were derived from the 1990 Census: • Average commute time (one-way, in minutes), • Percentage of housing units unsewered, including those units identified as using septic systems as well as “all other methods” other than public sewerage, • Residential stability: the percentage of households living at the same address in 1990 as in 1985, and • Housing “unaffordability”: the ratio of the median house value in the city to the median household income. In addition, information to construct several other variables was gathered from a variety of sources, as discussed below: • Citizen opposition to growth: This variable is based on responses to the planning director survey. Respondents were asked, “Please review the following list of factors and circle the number that you feel best describes how important each factor is in constraining or slowing residential development in your city . . . Citizen opposition to growth.” Respondents chose a value between 1 (“not at all important”) and 5 (“very important”). • Percentage Democrat: This is the percentage of two-party registrants (i.e., Republican and Democratic) who are Democrats. Data are from the report of registration by the California Secretary of State, February 10, 1999. • Planning staff per 1,000 residents: Our survey of planning directors asked respondents to indicate the number of personnel employed in the city’s planning department—or, if the planning staff is contained within another department, to indicate the number of individuals involved in planning activities. The number of staff was then normalized to the 1998 population (Department of Finance estimate). • City share of the property tax: This is an estimate of the percentage of property tax revenues generated within each city that go to the city government (as opposed to the county government, school districts, special districts, etc.). This is 112 calculated from fiscal data for 1997–1998, made available as a database by the Legislative Analyst’s Office.2 To construct this estimation, it is assumed that total property taxes generated in a city are equal to 1 percent of the city’s net assessed valuation. The city government’s nondebt property tax revenue is then compared with this estimate of property taxes paid to calculate the “city’s share.” • Logarithm of per capita, own-source revenues: This is calculated from California Controller reports on city financial transactions for the 1992–1993 fiscal year. Own-source revenues are calculated as total revenues minus intergovernmental subventions and minus city enterprise revenues. Own-source revenues are calculated on a per-capita basis using 1993 population estimates by the Department of Finance. The log of this number is then taken because of the extreme skewness of the distribution. • County population change, 1990–1998: This is the percentage increase in population at the county level, based on population estimates by the Department of Finance. • City population change, 1990–1998: In some models, we substitute city population change for county population change. • Jobs-to-workers ratio: This is the logarithm of the ratio of the number of jobs within the city to the number of employed residents in the city. It is calculated from Bureau of Transportation Statistics data (1994). • Logarithm of 1998 city population: For this we used Department of Finance estimates. • Bay Area: This is a binary variable taking the value of 1 if the city is located in the nine-county San Francisco Bay Area. • Southern California, not Los Angeles County: This is a binary variable taking the value of 1 if the city is located in one of the following counties: Orange, Riverside, San Bernardino, Ventura, Santa Barbara, or San Diego. ____________ 2Data are derived from the Local Government Database at http://www.lao.ca.gov/ lao_menu_local_govt.asp. 113 • Central city: This is a binary variable taking the value of 1 if the city is designated as one of California’s 49 central cities in the 1990 Census. • Rural: This is a binary variable taking the value of 1 if the city is located outside a Metropolitan Statistical Area (MSA) or is not in an urbanized portion of an MSA. In addition, the following variables are used in a few models: • District council elections: This is a binary variable taking the value of 1 if all or some members of the city council are elected by districts rather than at-large. Data are from a Public Policy Institute of California statewide mail survey of city clerks in 2000, relating to local electoral institutions and voter turnout. (The response rate to this question was 82 percent.) • Local crime rate: This is the logarithm of the city’s average crime rate for the years 1990, 1991, and 1992, as reported by the California Department of Justice. • Business/Chamber of Commerce importance: In the city manager survey, respondents were asked, “Generally speaking, how important are the following considerations to your city administration’s strategies in attracting new development and responding to development proposals? Please review each item below and indicate how important it is [on a scale of 1 to 7]. . . . Support of Chamber of Commerce or other local business interests for project.” • Neighborhood importance: As part of the same question, respondents evaluated the importance of “acceptability of proposal to nearby neighborhoods.” Table A.1 provides summary information about the mean, high value, low value, and data source of each of the independent variables (except binary variables) and all of the dependent variables used in the various models. 114 Table A.1 Descriptive Statistics for Variables in the Multivariate Models Low High Mean Value Value Source Independent Variable Citizen opposition to growth 2.7 1.0 5.0 Planner survey % Democrats 56.0 18.8 95.0 California Secretary of State Planning staff per 1,000 0.2 0.0 1.9 Planner survey City's % share of property tax Own-source per capita revenue (log) Importance of neighborhoods 13.2 0.0 65.9 Calculated from LAO database; see text 6.1 4.0 12.7 Calculated from California Controller reports 5.7 1.0 7.0 City manager survey Importance of business interests 4.8 1.0 7.0 City manager survey Crime rate (log) 8.6 6.8 14.0 California Dept. of Justice % owner occupancy 61.0 16.0 97.1 2000 Census % Hispanic 30.2 2.2 98.3 2000 Census % black 3.8 0.0 46.4 2000 Census Commute time 22.8 6.3 40.5 1990 Census % unsewered units 5.9 0.0 99.0 1990 Census Residential stability 45.1 11.4 72.2 1990 Census County population change City population change Housing unaffordability 12.3 0.1 30.8 California Department of Finance 13.6 –31.8 171.1 California Department of Finance 4.2 1.5 11.6 Calculated from 1990 Census % recreational units 1.9 0.0 63.5 1990 Census Job/worker ratio (log) Population (log) –0.1 –1.8 2.4 Census Transportation Planning Package 10.0 4.4 15.1 California Department of Finance Dependent Variable Number of growth management 2.7 0.0 10.0 Planner survey policies Importance of density restrictions 2.8 1.0 5.0 Planner survey Strictness of review process 2.1 1.0 4.0 Planner survey Policies’ effect on social status 2.5 1.0 4.0 Planner survey Adoption of affordable housing requirement Council limits residential development 0.3 0.0 1.0 Planner survey 1.8 1.0 4.0 Planner survey 115 Table A.1 (continued) City makes development more difficult Multifamily housing desirability (new development) Multifamily housing desirability (redevelopment) Controversy over residential growth Residential issues affect elections Initiatives: major source of antigrowth policies Good chance of slow-growth initiative Low High Mean Value Value Source 1.7 1.0 4.0 Planner survey 3.6 1.0 7.0 City manager survey 3.8 1.0 7.0 City manager survey 2.2 1.0 4.0 Planner survey 2.0 1.0 4.0 Planner survey 0.2 0.0 1.0 Planner survey 0.1 0.0 1.0 Planner survey Controversy and Growth-Management Policies City planning directors were asked, “How controversial would you say residential growth issues are in your city?” They responded using a four-point scale that ranged from “not at all controversial” to “almost always controversial.” The relationship of local controversy levels to the number of growth-management policies is strong, as the following simple bivariate regression reveals: Number of policies = 1.05 + 0.73 (controversy) + e (t-score of controversy = 6.02; adjusted R-sq. = 0.11; N = 289) However, as discussed in Chapter 5, it is also reasonable to suppose that the number of growth-management policies may influence the level of controversy. We can help evaluate this question by relying on the number of growth-management policies identified in each city in a 1988 survey by Glickfeld and Levine (1992) to see whether that number is associated with the level of controversy in 1998. Indeed, it is, as the following bivariate model illustrates. We use ordered logit rather than 116 regression because of the nature of the dependent variable.3 The number of observations is smaller because some new cities incorporated after 1988 and because some cities responded to one survey but not the other: Controversy level = 0.33 (number of 1988 policies) + e (z-value of number of policies = 5.13; log likelihood = –274.59; probability > chi-sq. = 0.000; N = 243) Because of the apparent endogeneity between controversy and the number of policies, we do not use controversy level as an independent variable in subsequent models. Although there are various ways to deal with this problem (e.g., instrumental variables), we chose to substitute for controversy our survey-based measure of “citizen opposition to growth,” which is unlikely to be similarly increased by the number of growth-management policies. Models Relating to Housing and Growth-Control Policies A multivariate regression model of the number of growthmanagement policies identified by our survey is shown in Table A.2. As noted in the report, the number of policies may range, in theory, from 0 to 16, although the average is 2.7. In modeling the determinants of the number of policies, we use the rate of population growth at the county level, rather than in the city, because policymakers are likely to be concerned with the overall level of residential growth pressures in their region. (In any event, city-level population growth proved insignificant in these models.) ____________ 3Ordered logit models are preferred when the dependent variable takes the form of a small number of ordered categories, as in this case, where controversy ranges along a fourpoint scale from “not at all” to “almost always” controversial. Throughout this appendix, we use ordered logit rather than ordinary least squares (OLS) regression in these circumstances. Results using OLS are generally extremely similar in terms of the signs and significance levels of the variables. Respondents choosing the “don’t know” answer are dropped from the analysis. 117 Table A.2 Number of Growth Management Policies Adopted (Regression Model) Independent Variables Citizen opposition to growth % Democrats Planning staff per 1,000 City’s % share of property tax Own-source per capita revenue (log) % owner occupancy % Hispanic % black Commute time % unsewered units Residential stability County population change % recreational units Job/worker ratio (log) Population (log) Bay Area Southern California, not Los Angeles County Central city Rural B0 No. of cities Probability > F Adjusted R-squared 0.205 (2.39)** –0.010 (0.65) –0.167 (0.31) –0.001 (0.09) 0.229 (0.88) 0.036 (2.67)*** 0.021 (2.42)** –0.024 (1.17) 0.010 (0.32) –0.020 (3.20)*** –0.058 (3.69)*** –0.153 (4.97)*** 0.045 (2.02)** 0.070 (0.25) –0.118 (0.73) 1.783 (4.35)*** 0.652 (2.08)** 0.903 (2.30)** 1.171 (2.54)** 3.117 (1.10) 259 0.000 0.27 Notes: **p < 0.05, ***p < 0.01. Cell entries are unstandardized regression coefficients, with absolute values of T-values listed in parentheses, calculated using robust standard errors. Table A.3 uses the same model, this time in ordered-logit form, to examine city planning directors’ answers to three survey questions relating to local residential policies. In the first column of results, the dependent variable is the respondents’ assessments of the importance of “density restrictions on residential land” in slowing or constraining residential development in their cities. They ranked the importance of density restrictions on a five-point scale, from “not at all important” to “very important.” 118 Table A.3 Planner Assessments of City Policies (Ordered Logit) Independent Variables Citizen opposition to growth % Democrats Planning staff per 1,000 City’s % share of property tax Own-source per capita revenue (log) % owner occupancy % Hispanic % black Commute time % unsewered units Residential stability County population change % recreational units Job/worker ratio (log) Population (log) Bay Area Southern California, not Los Angeles County Central city Rural No. of cities Log likelihood Probability > chi-squared Pseudo R-squared Importance of Density Restrictions 0.677 (6.02)*** –0.029 (1.78)* 0.544 (0.94) 0.004 (0.19) –0.021 (0.06) –0.020 (1.42) 0.016 (1.57) 0.021 (0.78) 0.029 (0.72) 0.009 (1.01) 0.049 (2.59)** –0.104 (3.01)*** –0.004 (0.13) –0.038 (0.11) 0.275 (1.62) 0.077 (0.17) 0.226 (0.65) –1.324 (2.90)*** –0.225 (0.52) 258 –353.77 0.000 0.12 Strictness of Policies’ Effect Review Process on Social Status 0.273 (2.41)** 0.255 (1.92)* –0.018 (1.12) –0.019 (0.89) 2.069 (2.24)** –0.407 (0.57) –0.012 (0.58) –0.010 (0.39) 1.018 (2.50)** 0.137 (0.35) 0.036 (2.20)** 0.046 (2.33)** –0.004 (0.42) –0.012 (0.99) 0.021 (0.79) –0.024 (0.60) 0.032 (0.82) 0.079 (1.75)* 0.005 (0.62) –0.026 (2.01)** 0.008 (0.38) –0.008 (0.32) –0.083 (1.97)** –0.100 (1.59) 0.019 (0.54) 0.028 (0.57) –0.373 (1.05) 0.439 (1.01) –0.001 (0.01) –0.437 (1.97)** –0.495 (1.14) 0.226 (0.45) –0.227 (0.65) –0.195 (0.39) 0.939 (2.28)** 0.609 (1.04) –0.395 (0.92) –0.161 (0.29) 252 189 –251.36 –179.89 0.000 0.000 0.15 0.16 Notes: *p < 0.1, **p < 0.05, ***p < 0.01. Cell entries are ordered logit coefficients, with absolute values of z-values listed in parentheses, calculated using robust standard errors. In the second column of results, the dependent variable is the respondents’ assessment of the review process for residential projects in their city, in comparison to other cities in the area. Respondents used a four-point scale of less strict, equally strict, somewhat more strict, or much more strict. For the results shown in the last column of Table A.3, the dependent variable concerns the respondent’s assessment of “the overall effect of your city’s residential development policies on who lives there.” Answers were on a four-point scale, ranging from “the city population is somewhat lower in social status than it would otherwise be, because of 119 [the] city’s residential development policies,” to “the city population is much more affluent than it would otherwise be.” Table A.4 concerns another type of local policy outcome. “Does your city currently have a policy to require residential development to include affordable housing, however that is defined in your community?” Because the dependent variable is bivariate (1 = yes, 0 = no), we use a Table A.4 Adoption of Affordable Housing Requirement (Probit Model) Independent Variables Citizen opposition to growth % Democrats Planning staff per 1,000 City’s % share of property tax Own-source per capita revenue (log) % owner occupancy % Hispanic % black Commute time % unsewered units Residential stability City population change Housing unaffordability Job/worker ratio (log) Population (log) Bay Area Southern California, not Los Angeles County Central city Rural B0 No. of cities Log likelihood Probability > chi-squared % correctly predicted Pseudo R-squared 0.045 (0.56) 0.009 (0.77) –0.271 (0.46) 0.007 (0.50) 0.225 (0.93) 0.040 (2.86)*** 0.012 (1.65)* –0.021 (1.10) –0.033 (1.20) –0.001 (0.18) –0.059 (3.77)*** –0.021 (2.48)** 0.352 (3.45)*** –0.258 (1.00) –0.347 (2.34)** 0.707 (2.54)** 0.167 (0.65) 0.111 (0.35) –0.119 (0.38) 0.253 (0.12) 252 –129.53 0.000 74.6 0.19 Notes: *p < 0.1, **p < 0.05, ***p < 0.01. Cell entries are probit estimates, with absolute values of z-values listed in parentheses, calculated using robust standard errors. 120 probit model. We also include the variable measuring the “unaffordability” of local housing as of 1990. Models Relating to City Government Orientations Toward Growth Table A.5 reports ordered logit results regarding city government orientations toward residential development. The first column of results is an analysis in which the dependent variable is the planner’s assessment of “the general attitude of the majority of [the] city council toward residential growth.” Responses were along a four-point scale, depending on whether the council was seen as mostly encouraging residential growth, being neutral toward it, occasionally slowing the rate of growth, or often proposing limitations on residential development. The second column of results relates to the same dependent variable but includes among the independent variables a measure of whether the council is elected by districts. This results in the loss of some observations, because of missing data. The third and fourth columns of results in Table A.5 relate to the planner’s response to a question about the relative restrictiveness of city policies: “Which of the following comes closest to your view of the policies of your city regarding development?” Respondents had four options, which we coded from 1 to 4, as follows: • “My city encourages all sorts of residential and commercial growth.” • “My city encourages most commercial growth, although it is less receptive to multifamily or “affordable” housing projects.” • “My city encourages most commercial growth, but it makes all residential development more difficult.” • “My city makes it more difficult for both commercial and residential development.” Again, in the fourth column, the variable denoting district council elections is included. 121 Table A.5 Planner Assessments of City Government Orientations to Residential Growth (Ordered Logit) 122 Independent Variables Citizen opposition to growth District elections % Democrats Planning staff per 1,000 City’s % share of prop. tax Own-source per capita revenue (log) % owner occupancy % Hispanic % black Commute time % unsewered units Residential stability City population change % recreational units Job/worker ratio (log) Population (log) Bay Area Southern California, not Los Angeles County Central city Rural Council Limits Residential Development Model 1 Model 2 0.706 (5.43)*** 0.786 (5.03)*** — 1.003 (1.46) –0.040 (1.98)** –0.050 (2.18)** 0.257 (0.35) –0.051 (0.07) 0.005 (0.15) 0.002 (0.05) 0.409 (1.25) 0.572 (1.45) –0.034 (2.27)** –0.022 (1.39) 0.018 (1.61) 0.027 (2.04)** 0.019 (0.50) 0.028 (0.62) 0.162 (3.41)*** 0.154 (3.05)*** –0.024 (2.38)** –0.038 (2.84)*** 0.058 (2.35)** 0.067 (2.40)** –0.000 (0.02) –0.009 (0.65) 0.038 (0.75) 0.064 (1.79)* –0.037 (0.09) –0.254 (0.56) –0.464 (2.38)** –0.634 (3.01)*** –0.013 (0.02) –0.119 (0.20) –0.396 (1.04) –0.486 (1.27) 0.592 (1.15) 0.668 (1.16) –0.186 (0.33) –0.315 (0.55) City Makes Development More Difficult Model 1 Model 2 0.597 (3.97)*** 0.566 (3.55)*** — 0.993 (1.78)* –0.047 (2.19)** –0.058 (2.37)** 0.652 (0.42) 1.530 (0.83) 0.019 (0.74) 0.011 (0.41) 0.411 (1.32) 0.395 (1.06) –0.000 (0.02) 0.001 (0.03) 0.024 (1.94)* 0.021 (1.76)* 0.039 (1.13) 0.045 (1.10) 0.132 (2.65)*** 0.154 (2.83)*** 0.008 (0.81) –0.007 (0.52) 0.061 (2.02)** 0.052 (1.55) –0.025 (0.97) –0.025 (0.91) –0.032 (0.86) –0.043 (0.97) 0.607 (1.47) 0.439 (0.93) –0.403 (1.58) –0.552 (2.08)** 1.251 (2.43)** 1.302 (2.32)** 0.485 (1.32) 0.642 (1.61) 0.137 (0.25) 0.302 (0.48) 0.118 (0.23) –0.087 (0.15) No. of cities Log likelihood Probability > chi-squared Pseudo R-squared 251 –257.45 0.000 0.13 208 –208.25 0.000 0.16 240 –212.12 0.000 0.15 199 –173.74 0.000 0.17 Notes: *p < 0.1, **p < 0.05, ***p < 0.01. Cell entries are ordered logit coefficients, with absolute values of z-values listed in parentheses, calculated using robust standard errors. In Table A.6, dependent variables are drawn from the city manager survey on local development strategies. City managers were asked to rate the desirability to their city administration of various types of land uses, using a scale of 1 to 7 from “very undesirable” to “very desirable.” Here we focus on the respondents’ desirability score for multifamily residential development. The first column of results pertains to multifamily Table A.6 City Manager Ratings of Desirability of Multifamily Development (Regression Model) Independent Variables Importance of neighborhoods Importance of business interests % Democrats Very active redevelopment policy City’s % share of property tax Own-source per capita revenue (log) % owner occupancy % Hispanic % black Commute time % unsewered units Residential stability City population change % recreational units Job/worker ratio (log) Population (log) Bay Area Southern California, not Los Angeles County Central city Rural B0 No. of cities Probability > F Adjusted R-squared Multifamily Rating Multifamily Rating (New Development) (Redevelopment) 0.239 (1.90)* 0.370 (2.83)*** 0.111 (1.05) –0.090 (0.76) 0.023 (1.58) –0.014 (0.83) — –0.010 (0.44) 0.739 (2.74)*** –0.006 (0.27) –0.189 (0.66) 0.341 (1.17) 0.027 (1.80)* 0.002 (0.11) –0.013 (1.79)* –0.009 (1.10) –0.011 (0.37) –0.002 (0.09) –0.140 (4.62)*** –0.098 (2.93)*** –0.005 (0.65) –0.011 (0.96) –0.025 (1.51) –0.024 (1.25) 0.013 (1.65) 0.012 (1.96)* 0.005 (0.14) 0.026 (1.61) 0.258 (0.92) –0.379 (1.27) 0.547 (3.30)*** 0.284 (1.37) 1.338 (3.31)*** 1.117 (2.74)*** 0.155 (0.54) –0.064 (0.16) 0.816 (2.32)** –1.742 (0.59) 191 0.000 0.23 –0.774 (2.41)** –0.088 (0.20) 0.611 (1.51) 0.608 (0.18) 221 0.000 0.22 Notes: *p < 0.1, **p < 0.05, ***p < 0.01. Cell entries are unstandardized regression coefficients, with absolute values of T-values listed in parentheses, calculated using robust standard errors. 123 development on vacant land (in “new development” areas), whereas the second column concerns ratings for multifamily development in the city’s redevelopment areas. Responses were gathered only from those city managers indicating that their cities had at least some vacant land available (for the question on new development) or indicating that their cities engage in redevelopment policy (for the question on redevelopment). Included among the independent variables are the importance scores that the city managers gave for the influence of business interests and neighborhood interests in the development process. For the redevelopment regression, we also include a variable indicating cities in which the city manager said the local redevelopment effort was “very active.” The variable measuring citizen opposition to growth cannot be used because it is not derived from the city manager survey, and its use would result in the deletion of many cases.4 Models Relating to Citizen Controversy and Initiatives In this set of analyses, we address the degree of citizen controversy over residential issues. For the modeling in this section, the variable relating to citizen opposition to growth is dropped, since it may be one manifestation of controversy. Variables regarding planning staff size, recreational homes, own-source revenue, and the city’s share of the property tax were also dropped, as these proved very insignificant and are probably less relevant to the generation of popular controversy. Results in Table A.7 relate to two questions in the planning director survey. The first is our overall measure of citizen controversy over residential issues, discussed above. In the second column of results, the variable measuring the local crime rate is introduced (which unfortunately results in the loss of some observations because of missing data). The intuition behind the inclusion of crime rates, which is supported by the results, is that another major local issue, in this case public safety, may displace residential growth issues from a position of controversy. ____________ 4The variable measuring district council elections was insignificant when entered, and is omitted here. 124 Table A.7 Planner Assessments of Growth-Related Political Controversy (Ordered Logit) 125 Independent Variables Crime rate % Democrats % owner occupancy % Hispanic % black Commute time % unsewered units Residential stability City population change Job/worker ratio (log) Population (log) Bay Area Southern California, not Los Angeles County Central city Rural No. of cities Log likelihood Probability > chi-squared Pseudo R-squared Controversy over Residential Growth Model 1 Model 2 — –1.415 (3.15)*** 0.007 (0.42) 0.016 (0.93) –0.031 (1.98)** –0.037 (2.31)** –0.031 (3.38)*** –0.030 (3.13)*** –0.078 (2.65)*** –0.052 (1.82)* 0.115 (3.32)*** 0.134 (3.70)*** –0.003 (0.32) –0.006 (0.61) 0.031 (1.44) 0.020 (0.80) 0.006 (0.96) 0.011 (1.39) –0.391 (1.31) 0.109 (0.32) 0.140 (0.90) 0.160 (0.97) 1.086 (2.42)** 0.903 (1.97)* 0.328 (0.98) 1.035 (2.35)** 0.973 (2.16)** 274 –291.43 0.000 0.11 0.452 (1.31) 1.334 (2.93)*** 1.331 (2.83)*** 258 –267.83 0.000 0.13 Residential Issues Affect Elections Model 1 Model 2 — –1.482 (3.12)*** –0.012 (0.72) 0.002 (0.11) –0.010 (0.64) –0.022 (1.32) –0.030 (3.31)*** –0.031 (3.43)*** –0.072 (1.52) –0.056 (1.23) –0.012 (0.29) 0.013 (0.29) –0.015 (1.10) –0.016 (1.16) 0.009 (0.46) 0.014 (0.70) –0.004 (0.75) 0.004 (0.62) –0.698 (2.20)** –0.263 (0.76) 0.168 (0.96) 0.192 (1.02) 1.411 (3.44)*** 1.188 (2.83)*** 0.454 (1.43) 1.081 (2.31)** 0.694 (1.48) 257 –236.70 0.000 0.14 0.664 (1.97)** 1.473 (2.98)*** 1.072 (2.13)** 241 –216.03 0.000 0.16 Notes: *p < 0.1, **p < 0.05, ***p < 0.01. Cell entries are ordered logit coefficients, with absolute values of z-values listed in parentheses, calculated using robust standard errors. The next two columns of results in the table relate to planners’ responses to a question about the influence of residential development issues on city council or mayoral elections. There are three possible responses, ranging from growth issues “hardly ever” affecting elections to being “often influential” in affecting the outcomes of elections. Again, the crime rate variable is entered into the model in the fourth column. Finally, Table A.8 reports probit estimates relating to local antigrowth initiatives. The first column reflects the probability that planners answer in the affirmative to the following question: “In your city, have initiatives on the ballot been a major source of policies to slow Table A.8 Citizen Antigrowth Initiatives (Probit Model) Independent Variables Initiatives Have Been Good Chance Slow- a Major Source of Growth Initiative Antigrowth Policies Will Occur % Democrats % owner occupancy % Hispanic % black Commute time % unsewered units Residential stability City population change Job/worker ratio (log) Population (log) Bay Area Southern California, not Los Angeles County Central city Rural B0 No. of cities Log likelihood Probability > chi-squared % correctly predicted Pseudo R-squared –0.015 (1.05) –0.020 (1.43) –0.013 (1.76)* –0.053 (2.25)** 0.052 (1.45) –0.057 (2.08)** –0.001 (0.06) –0.010 (1.01) –0.496 (1.84)* 0.182 (1.19) 1.078 (3.13)*** 0.366 (1.20) 0.846 (2.24)** 0.610 (1.46) –2.082 (0.98) 270 –91.92 0.000 81.5 0.23 –0.007 (0.47) 0.003 (0.25) –0.017 (2.03)** –0.053 (1.62) 0.023 (0.74) –0.033 (2.10)** –0.024 (1.45) 0.010 (1.75)* –0.728 (2.43)** 0.105 (0.66) 0.911 (2.06)** 0.253 (0.70) 0.961 (2.29)** 0.866 (1.93)* –1.853 (0.81) 255 –73.57 0.002 87.8 0.22 Notes: *p < 0.1, **p < 0.05, ***p < 0.01. Cell entries are probit coefficients, with absolute values of z-values listed in parentheses, calculated using robust standard errors. 126 residential development? Don’t include referenda placed on the ballot by the council.” The second column relates to the question, “Is there a good chance that an initiative measure to slow residential development will occur in your city?” 127 Bibliography Aas, et al. v. The Superior Court of San Diego County, 2000 Cal. LEXIS 9048. Adang, Peter, “Snob Zoning: A Look at the Economic and Social Impact of Low Density Zoning,” Syracuse Law Review, Vol. 15, Spring 1964, pp. 507–537. Advisory Commission on Regulatory Barriers to Affordable Housing, “Not In My Back Yard”: Removing Barriers to Affordable Housing, Report to President Bush and Secretary Kemp, Department of Housing and Urban Development, Washington, D.C., 1991. Arax, Mark, “Putting the Brakes on Growth: Bypassed by the State’s Economic Rebound, the Central Valley Finds That Increased Development Just Brings More Municipal Debt,” Los Angeles Times, October 6, 1999. Babcock, Richard, and Fred Bosselman, Exclusionary Zoning: Land Use Regulation and Housing in the 1970s, Praeger, New York, 1973. Baldassare, Mark, “Regional Variations in Support for Regional Governance,” Urban Affairs Quarterly, Vol. 30, No. 2, 1994, pp. 275– 284. Baldassare, Mark, PPIC Statewide Survey: Special Survey on Growth, Public Policy Institute of California, San Francisco, California, May 2001. Baldassare, Mark, and William Protash, “Growth Controls, Population Growth, and Community Satisfaction,” American Sociological Review, Vol. 47, June, 1982, pp. 339–346. Barbour, Elisa, and Paul Lewis, “Development Priorities in California Cities: Results from a PPIC Survey,” Occasional Paper, Public Policy 129 Institute of California, San Francisco, California, 1998, available at http://www.ppic.org/publications/occasional/lewis.pdf. Bockman, Shel, Max Neiman, and Barbara Sirotnik, Riverside County Zones Report: The 2000–2001 Annual Survey, Inland Empire Research Consortium, Center for Social and Behavioral Science Research, University of California, Riverside, 2001. Bogart, William, “‘What Big Teeth You Have!’ Identifying the Motivations for Exclusionary Zoning,” Urban Studies, Vol. 30, No. 10, 1993, pp. 1669–1681. Bollick, Clint, “Subverting the American Dream: Government-Dictated ‘Smart Growth’ Is Unwise and Unconstitutional,” University of Pennsylvania Law Review, Vol. 148, No. 3, 2000, pp. 859–872. Brueckner, Jan, “Strategic Control of Growth in a System of Cities,” Journal of Political Economy, Vol. 57, 1995, pp. 393–416. Bureau of Transportation Statistics, 1990 Census Transportation Planning Package—California Statewide Element, data file, U.S. Department of Transportation, Washington, D.C., 1994. Calavita, Nico, and Kenneth Grimes, “Inclusionary Housing in California: The Experience of Two Decades,” Journal of the American Planning Association, Vol. 64, No. 2, 1998, pp. 150–169. California Budget Project, Locked Out! California’s Affordable Housing Crisis, California Budget Project, Sacramento, California, 2000. California Senate Office of Research, The Right Home in the Right Place at the Right Price, October 1999, available at http://www. sen.ca.gov/sor/housing/Righthome.html. California State Controller, Cities Annual Report: Fiscal Year 1997–98, Sacramento, California, 2000. Carman, Christopher, “Dimensions of Environmental Policy Support in the United States,” Social Science Quarterly, Vol. 79, No. 4, 1998, pp. 717–733. Center for the Continuing Study of the California Economy, Land Use and the California Economy, Palo Alto, California 1999. 130 Chapman, Jeffrey, “Proposition 13: Some Unintended Consequences,” Occasional Paper, Public Policy Institute of California, San Francisco, California, 1998. “Condo Scarcity: Construction Litigation Must Be Reined In,” San Diego Union-Tribune, December 10, 2000, p. G2. Curtin, Daniel, Jr., Curtin’s California Land Use and Planning Law, 20th ed., Solano Press Books, Point Arena, California, 2000. Danielson, Michael, The Politics of Exclusion, Columbia University Press, New York, 1976. DeLeon, Richard E., Left Coast City: Progressive Politics in San Francisco, University Press of Kansas, Lawrence, Kansas, 1992. Department of Housing and Community Development (State of California), Raising the Roof: California's Housing Development Projections and Constraints, 1997–2020, Department of Housing and Community Development, Sacramento, California, 2000a. Department of Housing and Community Development, Division of Housing Policy Development, New Funds Available for Housing, December 19, 2000b, available at http://www.hcd.ca.gov/hpd/fund_ hpd.html. Diaz, Daniel, and Gary Paul Green, “Fiscal Stress and Growth Management Effort in Wisconsin Cities, Villages, and Towns,” State & Local Government Review, Vol. 33, No. 1, 2001, pp. 7–22. Dickerson, Marla, and Stuart Silverstein, “New Crises Loom in State’s Aging Infrastructure,” Los Angeles Times, February 18, 2001, p. A1. Donovan, Todd, and Max Neiman, “Citizen Mobilization and the Adoption of Local Growth Control,” Western Political Quarterly, Vol. 45, No. 3, 1992, pp. 651–675. Donovan, Todd, and Max Neiman, “Local Growth Control Policy and Changes in Community Characteristics,” Social Science Quarterly, Vol. 76, No. 4, 1995, pp. 780–793. 131 Dowall, David, The Suburban Squeeze: Land Conversion and Regulation in the San Francisco Bay Area, University of California Press, Berkeley, California, 1984. Dowall, David E., and Jan Whittington, Making Room for the Future: Rebuilding California’s Infrastructure, Public Policy Institute of California, San Francisco, California, forthcoming. Dresch, Marla, and Steven Sheffrin, Who Pays for Development Fees and Exactions? Public Policy Institute of California, San Francisco, California, 1997. Dunstan, Roger, and Jennifer Swenson, Construction Defect Litigation and the Condominium Market, California Research Bureau, Sacramento, California, November 1999. Ewing, Reid, “Is Los Angeles-Style Sprawl Desirable?” Journal of the American Planning Association, Winter 1997, pp. 107–126. Farley, Reynolds, “Suburban Persistence,” American Sociological Review, Vol. 29, February 1964, pp. 38–47. Fischel, William, Do Growth Controls Matter? A Review of Empirical Evidence on the Effectiveness and Efficiency of Local Government Land Use Regulation, Lincoln Institute of Land Policy, Cambridge, Massachusetts, 1990. Frieden, Bernard J., The Environmental Protection Hustle, MIT Press, Cambridge, Massachusetts, 1979. Fulton, William, and Paul Shigley, “Election Analysis: Big Day for SlowGrowth Advocates,” California Planning & Development Report, November 2000. Fulton, William, Paul Shigley, Alicia Harrison, and Peter Sezzi, Trends in Local Land Use Ballot Measures, 1986–2000, Solimar Research Group, Inc., Ventura, California, 2000. Glickfeld, Madelyn, LeRoy Graymer, and Kerry Morrison, “Trends in Local Growth Control Ballot Measures in California,” UCLA Journal of Environmental Law and Policy, Vol. 6, No. 2, 1987, pp. 111–158. 132 Glickfeld, Madelyn, and Ned Levine, Regional Growth, Local Reaction: The Enactment and Effects of Local Growth Control and Management Measures in California, Lincoln Institute of Land Policy, Cambridge, Massachusetts, 1992. Gordon, Peter, and Harry Richardson, “Are Compact Cities a Desirable Planning Goal?” Journal of the American Planning Association, Winter 1997, pp. 95–106. Governor’s Office of Planning and Research, The California Planners’ Book of Lists, 2000, Governor’s Office of Planning and Research, Sacramento, California, 2000. Gruen, Claude, “Forced Donations Counterproductive,” Gruen Gruen & Associates, Trends, July 2001, available at http://www.ggassoc. com/trends.htm#trends22. Haar, Charles, Suburbs Under Siege: Race, Space, and Audacious Judges, Princeton University Press, Princeton, New Jersey, 1996. Hartnett, James, “Affordable Housing, Exclusionary Zoning, and American Apartheid,” New York University Law Review, Vol. 68, No. 1, 1993, pp. 89–135. Housing Crisis Task Force, City of Los Angeles, In Short Supply: Report of the Housing Crisis Task Force, report to the Los Angeles City Council, 2000, available at http://housingcrisisla.ucla.edu/. Kaniss, Phyliss, Making Local News, University of Chicago Press, Chicago, Illinois, 1991. Krainer, John, and Fred Furlong, “Tech Stocks and House Prices in California,” FRBSF Economic Letter, Federal Reserve Bank of San Francisco, Number 2000–27, September 15, 2000. Landis, John, “Do Growth Controls Work? A New Assessment,” Journal of the American Planning Association, Vol. 58, 1992, pp. 489–506. Landis, John, “Growth as Destiny: Understanding California’s Postwar Growth Patterns and Trends,” in Metropolitan Development Patterns: Annual Roundtable 2000, Lincoln Institute of Land Policy, Cambridge, Massachusetts, 2000. 133 Landis, John, and Cynthia Kroll, “The Southern California Growth War,” in John Kirlin & D. Winkler, eds., California Policy Choices, Vol. 5, University of Southern California, Los Angeles, California, 1989, pp. 123–163. Landis, John, Rolf Pendall, Robert Olshansky, and William Huang, Fixing CEQA: Options and Opportunities for Reforming the California Environmental Quality Act, California Policy Seminar, University of California, Berkeley, California, 1995. Legislative Analyst, State of California, Reconsidering AB 8: Exploring Alternative Ways to Allocate Property Taxes, Legislative Analyst’s Office, Sacramento, California, February 3, 2000. Levine, Ned, “The Effect of Local Growth Controls on Regional Housing Production and Population Redistribution in California,” Urban Studies, Vol. 36, No. 12, 1999, pp. 2047–2068. Lewis, Paul, “Looking Outward or Turning Inward? Motivations for Development Decisions in California Central Cities and Suburbs,” Urban Affairs Review, Vol. 36, No. 5, 2001, pp. 696–720. Lewis, Paul, “Offering Incentives for New Development: The Role of City Social Status, Politics, and Local Growth Experiences,” Journal of Urban Affairs, forthcoming. Lewis, Paul, and Elisa Barbour, California Cities and the Local Sales Tax, Public Policy Institute of California, San Francisco, California, 1999. Lewis, Paul, and Max Neiman, “Residential Development and Growth Control Policies: Survey Results from Cities in Three California Regions,” Occasional Paper, Public Policy Institute of California, San Francisco, California, 2000, available at http://www.ppic.org/ publications/occasional/lewis.neiman.pdf. Logan, John, and Harvey Molotch, Urban Fortunes: The Political Economy of Place, University of California Press, Berkeley, 1987. Logan, John, and Min Zhou, “Do Suburban Growth Controls Control Growth?” American Sociological Review, Vol. 54, June 1989, pp. 461– 471. 134 Mercer, Lloyd, and W. Douglas Morgan, “An Estimate of Residential Growth Controls’ Impact on House Prices,” in M. Bruce Johnson, ed., Resolving the Housing Crisis, Ballinger Publishing Company, Cambridge, Massachusetts, 1982, pp. 189–215. Myers, Phyllis, and Robert Puentes, “Growth at the Ballot Box: Electing the Shape of Communities in November 2000,” Discussion Paper, Brookings Institution, Center on Urban and Metropolitan Policy, Washington, D.C., 2001. Neiman, Max, “Zoning Policy, Income Clustering, and Suburban Change,” Social Science Quarterly, Vol. 61, 1980, pp. 666–675. Neiman, Max, Gregory Andranovich, and Kenneth Fernandez, Local Economic Development in Southern California’s Suburbs: 1990–1997, Public Policy Institute of California, San Francisco, 2000. Neiman, Max, and Ronald O. Loveridge, “Environmentalism and Local Growth Control: A Probe into the Class Bias Thesis,” Environment and Behavior, Vol. 13, 1981, pp. 759–772. O’Toole, Patrick, “California Tectonics: California’s Housing Industry,” Professional Builder, Vol. 65, No. 7, 2000, p. 58. Pendall, Rolf, “Opposition to Housing: NIMBY and Beyond,” Urban Affairs Review, Vol. 35, No. 1, 1999, pp. 112–136. Pendall, Rolf, “Local Land Use Regulation and the Chain of Exclusion,” Journal of the American Planning Association, Vol. 66, No. 2, 2000, pp. 125–142. Rojas, Aurelio, “Builders vs. Lawyers at Capitol,” Sacramento Bee, January 2, 2001, p. A1. Sanchez, Jesus, “Ruling by Court Unlikely to Spur Condo Construction,” Los Angeles Times, December 9, 2000, p. C1. Sanchez, Jesus, “Taking Up Residence in Office Space,” Los Angeles Times, March 13, 2001. Senate Local Government Committee, Tension and Ambiguity: A Legislative Guide to Recent Efforts to Reform California’s State-Local 135 Fiscal Relationship, California State Legislature, Sacramento, California, March 2001. Schneider, Mark, and Paul Teske, “The Antigrowth Entrepreneur: Challenging the ‘Equilibrium’ of the Growth Machine,” Journal of Politics, Vol. 55, 1993, pp. 720–736. Shen, Q., “Spatial Impacts of Locally Enacted Growth Controls: The San Francisco Bay Region in the 1980s,” Environment and Planning B, Vol. 23, No. 1, 1996, pp. 61–91. Southern California Studies Center, Sprawl Hits the Wall: Confronting the Realities of Metropolitan Los Angeles, University of Southern California, Los Angeles, California, 2001. Staley, Samuel, “Ballot-Box Zoning, Transaction Costs, and Urban Growth,” Journal of the American Planning Association, Vol. 67, No. 1, 2001, pp. 25–37. Stone, Clarence, Regime Politics: Governing Atlanta, 1946–1988, University Press of Kansas, Lawrence, Kansas, 1989. Taylor, Jerry, and Peter VanDoren, “‘Smart Growth’ Is Good for Everyone Else,” San Diego Union-Tribune, March 15, 2000, p. B-9. Wacker, Matthew, Alvin D. Sokolow, and Rachel Elkins, “County Right-to-Farm Ordinances in California: An Assessment of Impact and Effectiveness,” AIC Issues Brief, Agricultural Issues Center, University of California, Davis, California, May 2001. Warner, Kee, and Harvey Molotch, Growth Control: Inner Workings and External Effects, California Policy Seminar, University of California, Berkeley, California, 1992. Warner, Kee, and Harvey Molotch, Building Rules: How Local Controls Shape Community Environments and Economies, Westview Press, Boulder, Colorado, 2000. Williams, A. Kevin, The Long Wait: The Critical Shortage of Housing in California, Corporation for Supportive Housing and Housing California, Sacramento, California, 2000. 136 About the Authors PAUL G. LEWIS Paul Lewis is a research fellow at the Public Policy Institute of California, where he studies local government and urban policy issues. He has authored or co-authored publications on local government structure, local sales taxes, urban development, transportation policy, and regional governance, and has made presentations on such topics to a variety of governmental bodies and civic organizations. He holds an A.B. in political science from Indiana University and an M.A. and Ph.D. in politics from Princeton. MAX NEIMAN Max Neiman is Director of the Center for Social and Behavioral Sciences Research and Professor of Political Science at the University of California, Riverside. He is the author of five books and numerous articles on such topics as energy conservation, evaluating government performance, and the development of California’s initiative industry. His research interests include the politics of regulation in residential and economic development. He is a co-producer of the Annual Inland Empire Survey and the recipient of grants from the National Science Foundation and the John Randolph and Dora Haynes Foundation. He received his B.A. degree at the University of Illinois, Chicago, and his doctorate at the University of Wisconsin. 137 Other Related PPIC Publications Subsidizing Redevelopment in California Michael Dardia Who Pays for Development Fees and Exactions? Marla Dresch, Steven M. Sheffrin California Cities and the Local Sales Tax Paul G. Lewis, Elisa Barbour Local Economic Development in Southern California’s Suburbs: 1990–1997 Max Neiman, Gregory Andranovich, Kenneth Fernandez PPIC publications may be ordered by phone or from our website (800) 232-5343 [mainland U.S.] (415) 291-4400 [Canada, Hawaii, overseas] www.ppic.org 139" ["post_date_gmt"]=> string(19) "2017-05-20 09:35:21" ["comment_status"]=> string(4) "open" ["ping_status"]=> string(6) "closed" ["post_password"]=> string(0) "" ["post_name"]=> string(8) "r_102plr" ["to_ping"]=> string(0) "" ["pinged"]=> string(0) "" ["post_modified"]=> string(19) "2017-05-20 02:35:21" ["post_modified_gmt"]=> string(19) "2017-05-20 09:35:21" ["post_content_filtered"]=> string(0) "" ["guid"]=> string(50) "http://148.62.4.17/wp-content/uploads/R_102PLR.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) }