The question of whether California will be able to meet its projected housing needs may ultimately comes down to a matter of local regulation. As demonstrated in the previous chapter, California has ample supplies of environmentally appropriate land to accommodate projected housing growth. Whether developers and homebuilders will be allowed to access that land will depend on whether adequate amounts of it are designated for residential development, and how the development approvals process is administered in California's 530 local governments.
California's development-approvals process, or land entitlements process as it is also known, is the most detailed and complicated in the nation. Project sponsors must go through more reviews, submit more documentation, and face greater uncertainty over project approvals in California than almost anywhere else. As complicated and frustrating as the process can be, it also produces some benefits. Despite having added more than 15 million new residents since 1970, California has protected more privately owned, environmentally sensitive land from development than any other state. The current debate over sprawl notwithstanding, California's urban development patterns are also less sprawling than those of other fast-growing states (Pendall 1998).
The development approvals process in California is really three processes. The first is the process as set forth in state planning and environmental laws and local codes. It consists of formal requirements, procedures, and steps for reviewing, evaluating, and approving (or denying) project applications. The second is the process as actually administered. It is highly discretionary, is subject to enforcement only through the courts, and provides local governments with tremendous bargaining power over project sponsors. The third process is one that citizens establish for themselves and their communities through the initiative process. The California Supreme Court has been extremely accommodating to citizen-sponsored initiatives in all areas, but especially in the area of local land use controls. Overlaid on top of each other, these three processes form a planning and development review process that is fundamentally different in every municipality in which it is applied. Depending on the municipality, it may also be different for every project. Except in rural areas, and even then, rarely, the concept of "as of right" development no longer exists in California.
Local land use discretion is not infinite, however. The California Legislature has repeatedly stipulated—ùand the courts have confirmed—that housing is an issue of statewide policy concern. For more than 25 years, governmental entities at the local, regional, and state levels have been required by statute to plan for the state's housing needs. Cities and counties, for example, have been required to prepare and adopt general plan housing elements since 1969. Permit processing and the efficacy of the development review process are among the housing-related issues the State has determined to be of policy concern, via adoption and amendment of the Permits Streamlining Act, and processing time lines pursuant to CEQA. 1*
This chapter explores the ways in which the development-approvals process constrains housing production in California. It begins with an outline of the processes' major regulatory elements. Next, based on the results of a survey of local governments, it explores the evolving use of local growth control and growth management techniques. Next, to understand how such procedures are actually applied and how they affect housing production, it examines the development review histories of 46 single- and multi-family housing projects from across California. Finally, it considers whether and how the development approvals process is likely to constrain future housing production.
No one, save perhaps a few land use attorneys, completely understands California's development approvals process. State planning law requires that California municipalities adopt a variety of planning and development ordinances, including a general plan, a zoning map, subdivision ordinances, and building codes. Additionally, California municipalities are required to administer the California Environmental Quality Act, or CEQA. Local governments may also impose their own planning and development requirements via ordinance, referendum, or voter-sponsored initiative.
To the uninitiated, California planning and development laws must seem highly schizophrenic. They are on the one hand, remarkably detailed and explicit regarding the types of regulatory reviews local jurisdictions must conduct. Yet on the other hand, they are remarkably vague and inexplicit regarding how reviews are to be conducted. As a consequence, they provide local elected officials—and by extension, their planning commissions and departments—with almost unlimited discretion in the approval or denial of particular projects. They also give local governments wide latitude regarding how they wish to structure the regulatory process. Thus, while the form of the regulatory process is similar across all jurisdictions, its substance can be far different.
Exhibit 22 and the following discussion present the seven basic elements of the residential regulatory approvals process in California:
Annexations in California may occur at the request of landowners or may be undertaken unilaterally by local governments. Annexation decisions are generally made by local legislative bodies (i.e., local city councils), although in some municipalities—pursuant to the passage of an initiative or referendum—annexation decisions require voter approval.
SOI boundaries are intended to guide annexation decisions, not force them, so as a consequence, cities have wide discretion when deciding whether or not to annex particular parcels. When making those decisions, city officials rely both on existing documentation (e.g., local general plans, special studies, and LAFCO-generated SOI maps) and on findings regarding whether the annexed site can reasonably and economically be provided with essential city services. Because they are typically located on larger sites or at the municipal fringe, annexation reviews are much more commonly required of single-family projects than multi-family and condominium developments.
Annexation requests may be turned down for any reason (regardless of findings) and are not generally subject to legal appeal. As a result, annexation law provides municipalities with tremendous leverage over landowners and developers, particularly in those counties where unincorporated development is discouraged.
Zoning and general plan amendment reviews are inherently discretionary. When undertaking a rezoning/general plan amendment, local legislative bodies (planning commissions, city councils, and county boards of supervisors) are required only to find that the proposed action is not inconsistent with adopted policies and plans, usually as articulated in the general plan. California state law allows planning commissions to approve zoning changes but requires that general plan amendments be approved by a city council or county boards of supervisors. This has the effect of subjecting most major land use changes to a two-tiered review. State law further requires that each review include an opportunity for public comment. To further increase local discretion, many California cities have adopted extremely detailed zoning ordinances which require project sponsors to obtain a conditional use permit for projects that do not exactly conform to the stipulated requirements.
Many local zoning codes include Planned Use Development (PUD) or Planned District (PD) designations. Usually applied to large parcels, PUD and PD designations allow project sponsors to sidestep certain subdivision approval requirements and instead obtain a single master plan approval. PUDs and PDs give developers more flexibility in planning and designing their projects, all the while subjecting them to a greater degree of public and regulatory scrutiny.
California State law does not require municipalities to adopt or publish standards for use in rezonings and general plan reviews. Other than the general plan itself, applicants for rezonings and general plan amendments often have little to go by when submitting proposals. From the applicant's perspective, this makes the rezoning/general plan review process highly uncertain, both with respect to any conditions of approval and the ultimate outcome.
As a way of promoting more specific planning outcomes, some California jurisdictions have come to favor specific plans over zoning maps and/or comprehensively updating their general plans. Applied to neighborhoods and groups of parcels (rather than to entire municipalities), specific plans combine zoning and subdivision requirements, site and lot plans, public facilities and infrastructure plans, open-space plans, and capital improvements financing/impact fees into a single planning document. Most specific plans are accompanied by Environmental Impact Reports. Many are adopted as general plan amendments. Once a specific plan and its EIR have been approved, and depending on what they include, all subsequent developments which are fully consistent with the adopted specific plan are also approved—thus allowing the project sponsor to skip directly to building permit review.4* This process of an earlier comprehensive review voiding the need for later discretionary reviews is known as tiering. Specific plans offer project sponsors a greater degree of downstream certainty, but once adopted, can be extremely inflexible. As a result, adopted specific plans are frequently amended to respond to changing market conditions.
Many CEQA reviews result in neither a negative declaration nor an EIR, but are instead processed as a mitigated negative declaration (MND). Issuance of an MND typically means that the project sponsor agrees to pro-actively mitigate any adverse impacts (as identified in the initial study) in exchange for not having to undertake an EIR. MNDs are much more circumscribed than EIRs, involve much less in the way of detailed analysis and public comment, and do not require certification.
If an EIR is required, it must ultimately be certified by the local legislative body, typically the city council or board of supervisors, for the project to go forth. EIRs are typically produced in draft form, then made available for public comment. Final EIR documents must incorporate responses to comments on the draft document. Depending on the project and community, EIRs may be undertaken by the local planning department, by private consultants, or by a combination of the two.
CEQA provides review agencies with almost unlimited discretion. When undertaking initial studies, communities are not required to make use of standardized review criteria or thresholds; and most jurisdictions do not do so. It does require them when undertaking an EIR, to pre-identify areas or issues requiring further study or analysis, a procedure known as "scoping."
CEQA requires project sponsors to mitigate all reasonably mitigatable impacts (as identified in the initial study or EIR), but also allows local decision-makers to approve projects which generate unmitigated impacts. This last action is termed "certification with a finding of overriding consideration." Determinations regarding what constitutes adequate impact mitigation are left to local decision-makers. Citizens may sue to overturn a certified EIR on procedural grounds, or on grounds of inadequate analysis, but not on substantive grounds. The tremendous amount of discretion built into CEQA notwithstanding, most CEQA reviews occur in a timely and straightforward fashion (Landis, et al., Fixing CEQA1995). Many projects are tiered off more comprehensive documents known as Program EIRs, or, in the case of many large suburban housing projects, specific plan EIRs. Even when tiering is not possible, many review agencies prefer the use of MNDs to EIRs.
Many growth measures impose additional project review requirements. Housing cap measures, for example, typically include project rating systems which are used to pare down the number of units to required levels. Jurisdictions with adequate public-facilities ordinances review projects to insure that the community has sufficient service capacity. Sponsors can increase the likelihood that their projects will be approved by providing or paying for needed infrastructure capacity; or else by designing them in accordance with the ratings criteria. To avoid "double jeopardy," some communities combine growth management reviews with rezoning, general plan, or tentative subdivision map reviews.
The application of local development review procedures does not occur in a political vacuum. From its inception, the State of California was settled, developed, and sold by people committed to the idea that growth in all of its manifestations was good. Indeed, until 1978, most California cities and counties continued to welcome growth. Californians saw growth as an indicator of prosperity, and development as a mechanism for increasing property values and tax revenues. Housing was especially welcome, for where it went, jobs and commercial development were sure to follow.
The positive and benign view of residential development began to change in the aftermath of Proposition 13. Enacted by initiative in 1978, Proposition 13 limited the ability of cities and counties to access their existing tax base to pay for needed services. Henceforth, the incremental costs of public services would be paid for by new development in the form of increased fees and higher assessments. With residential development considered too service-intensive (and thus too costly) compared with retail and commercial development, many jurisdictions began to actively discourage new residential developmentùparticularly family and middle-income housing. Housing also came under fire during the 1980s, as did most forms of new development, because of its perceived contribution to increased traffic congestion.
California municipalities are required by State general plan law to accommodate their respective "fair shares" of regional housing growth, so they are not to use their general plans and zoning ordinances to directly or overtly discourage housing production. However, some jurisdictions use discretionary review procedures and new growth management techniques to limit housing construction. These techniques all tie permissible levels of development, usually residential development, to particular infrastructure, fiscal, or environmental impact thresholds. To obtain approvals, residential developers must either reduce the sizes and densities of their projects (thereby reducing immediate project impacts), and/or compensate cities for potential negative impacts through the payment of impact fees and exactions.
The problem with this system is that it doesn't work. The amount of money which must be exacted from new development to ameliorate all impacts, including the costs of upgrading services for existing residents, is usually well beyond the ability of most individual projects—certainly almost all housing projects—to support. While it may be reasonable to require new development to pay its own way, it is not reasonable, as the current system too often does, to ask new development to pay everybody's way.
The inability of growth to benefit everyone, or at least to compensate everyone, coupled with increasing levels of growth, led many Californians to begin questioning whether growth was a good thing after all. The California Environmental Quality Act (CEQA), enacted in 1970, provided Californians concerned over the impacts of growth with the opportunity to voice their concerns at the local level. As was its intent, CEQA "democratized" the local planning and development review process by requiring that developers disclose the full range of their project's impacts, and by requiring that local legislative bodies disclose their reasons for approving or rejecting projects. More important, CEQA allowed residents and citizens to contest such disclosures.
By the early 1980s, growth and development issues had become too important to leave to "business as usual," and concerned residents were increasingly turning to the ballot box to enact more immediate and specific growth management ordinances. Most such ordinances were directed toward the most land- and service-intensive manifestation of growth: housing.
Looking back, voter use of the initiative process to directly address growth issues—and to force local elected officials to pay greater attention to growth issues—can be seen to occur in waves. Initiative activity increases when the state is growing, as was the case during the late 1970s, late 1980s, and now, late 1990s; and then wanes during periods of economic contraction.
Each new wave of initiative activity brings its own approaches and wrinkles. During the growth period of the late-1970s, for example, many communities followed the leads of Petaluma and Livermore in adopting population and housing growth caps. Ten years later, during the late-1980s, local growth initiatives focused on making new development "pay for itself" through the adoption of APFOs and general plan growth management elements. Current growth management initiatives emphasize the adoption of urban growth boundaries and tighter zoning controls.
What does the growth management landscape in California look like today? How might it constrain future housing production? The last comprehensive survey of local growth management and control regulations in California was undertaken in 1993 by Madelyn Glickfeld and Ned Levine. Thanks to their diligent follow-up efforts, Glickfeld and Levine received responses from 410 of 466 cities and from 55 of 57 counties. A summary of their results is presented in Exhibit 23.
Just over 12 percent of the 410 cities surveyed by Glickfeld and Levine had adopted some form of residential growth cap as of 1992. Nine percent had adopted numerical limits on population and/or residential growth. Commercial development limits were much less popular, having been adopted by 4.7 percent of responding cities. Caps on industrial space had been adopted by three percent of responding cites, while 2.7 percent had adopted caps on subdivision activity. Northern California cities were more likely to have adopted population, housing, or commercial growth caps than cities in other parts of the state. Commercial and industrial growth caps were more likely to have been adopted by medium and large-sized cities. Population and residential caps were more frequently adopted by smaller cities hoping to stay that way. Housing and subdivision caps were slightly more popular with counties than with cities, while population, commercial, and industrial growth caps were somewhat less popular.
Among cities, the most popular growth management approaches were FAR reductions (28 percent), residential down-zoning (26.6 percent), and the adoption of residential APFOs (23.6 percent). As of 1992, Glickfeld and Levine found, fully a quarter of California cities had adopted one or more of these approaches. Between 10 and 20 percent of California cities had enacted some form of control over commercial development, either APFOs (18.6 percent), height reductions (16.6 percent), or down-zoning (13.4 percent). Just over ten percent of the cities responding to Glickfeld and Levine's survey had adopted a general plan growth management element. Eight percent had adopted growth phasing plans, and five percent had adopted or tightened an urban limit line.
Different growth management approaches were favored in different regions. Southern California cities were more likely to try to manage growth through down-zoning, FAR reductions, and zoning changes. Northern California cities favored urban limit lines, APFOs, and adding a growth management element to their general plans. Phasing, infrastructure controls, and urban limit lines were popular among Central Valley cities. Sacramento area and central coast cities favored a mixture of approaches. The larger a city, the more likely it was to have adopted one or more growth management measures.
More than half of California counties had enacted an APFO measure by 1992, and about a third had acted to reduce development densities, either by downzoning or through FAR reductions. A quarter of California counties as of 1992 had enacted growth phasing plans, and one in ten had either tightened or adopted an urban limit line.
Overall, Glickfeld and Levine concluded that regional, not local growth was the primary precurser for the enactment of local growth measures, and that while there were many local economic, community, and institutional correlates with the enactment of local growth measures, none was as important as regional growth. In terms of their effects, local growth measures did not reduce overall construction levels, but did seem to encourage growth to "leapfrog" outward towards the urban fringe.
To determine how the growth management landscape had changed since 1992, the University of California at Berkeley and the California Department of Housing and Community Development jointly conducted a two-page mail survey during the summer of 1998 of all California cities and counties. Surveys were mailed to all local planning departments; respondents were asked to mail or fax-back the completed surveys. A follow-up survey of non-respondents was mailed out in late August.
Respondents were asked whether and how they had adopted or substantially altered the following growth management measures since January 1995:
Respondents were also asked about pending measures, measures scheduled for the November 1998 ballot, measures that had been lifted or allowed to expire, and recent annexation activity. To the extent possible, the UCB/CHCD survey was designed to be backwards-compatible with the 1993 Glickfeld/Levine survey. The UCB/HCD survey form and a frequency distribution of the survey results are included as Appendices J and K. Completed responses were received from 48 counties and 322 cities, yielding an overall response rate of 65 percent. Exhibit 24summarizes the UCB/HCD survey results by city and county
As did Glickfeld and Levine, we differentiated between growth controls, which limit the amount of development, and growth management measures, which regulate development's quality, location, and impacts. Relatively few California cities and counties adopted new growth control measures between 1995 and July 1998. Seven of the 322 responding cities adopted limits on residential construction; three adopted limits on commercial development, and three more imposed annexation restrictions. Of the 46 responding counties, only one imposed a new cap on residential development. A second county imposed new annexation restrictions.
New growth management measures were slightly more popular. Twenty-five of the responding 322 cities indicated that they had undertaken significant residential up zoning since 1995. Another 24 cities had downzoned significant land area. Twenty-four cities had rezoned significant areas from one use to another, and 18 cities had reduced allowable building heights and floor-area ratios. Fifteen of the 322 responding cities adopted an urban growth boundary or urban limit line, or substantially changed the one they had. Twelve cities enacted or changed their residential APFOs; 11 cities enacted a new growth management element, and eight cities adopted a commercial APFO. County adoption of these growth management measures followed a similar distribution.
More than a quarter of the responding cities reported having annexed significant new lands since 1995. Five percent significantly increased their residential development fees. Ordinances mandating voter approval of zoning and general plan changes were enacted in five cities. A slightly higher proportion of cities than counties increased residential development fees.
Exhibit 25 organizes the UCB/HCD survey responses by region, city size, 1985-95 population growth, and percent population growth. Among regions, cities in Northern California were slightly more likely to have enacted annexation restrictions, urban growth boundaries, residential APFOs, and general plan growth management elements than cities in other California regions. Hoping to attract rather than repel growth, cities in the Central Valley were significantly more likely to have engaged in up zoning or to have annexed land for development; they were also more likely to have increased residential development fees.
City size mattered little in the adoption of different growth measures between 1995 and July 1998. Small cities (those with populations between 10,000 and 25,000) were slightly more likely than large cities to have up-zoned land, or increased residential development fees. Large cities with populations above 100,000 were more likely to have downzoned residential land, rezoned land from one use to another, or enacted a new or revised growth management element. Except for new annexations, population growth between 1985 and 1995 explained relatively little of the variation in post-1995 growth management activity. Percentage population growth explained more. Cities that grew by 10 percent or more between 1985 and 1995 were somewhat more likely to have engaged in either significant up- or downzoning than slower-growing cities. Slower-growing cities were somewhat more likely than faster-growing cities to have imposed restrictions on annexation, to have adopted a residential APFO or growth management element, or to have enacted an ordinance mandating voter approval of major zoning and planning changes.
Exhibits 26 and 27add the UCB/HCD survey results to those of Glickfeld and Levine to produce a series of composite trend lines (Note that neither survey covers the 1993-94 period). Both exhibits confirm Glickfeld and Levine's argument regarding the inherently defensive nature of local growth measures.
The biggest run-up in the use of residential growth caps by California cities and counties occurred during the mid- and late 1980s (Exhibit 26). Forty-five California cities and counties had adopted a population or residential building permit cap as of 1985. By 1992, that number had increased to over 100. Since then, 10 additional jurisdictions have adopted such caps. The late 1980s and early 1990s were also the heyday of caps on commercial development, although fewer than two-dozen jurisdictions ever adopted such limits.
The late 1980s were an even more active period for local growth management (Exhibit 27). Some growth management approaches were more popular than others. According to Glickfeld and Levine, the four most popular growth management approaches were: (i) downward adjustments to local building height and FAR limits; (ii) the adoption of APFOs as applied to residential development, (iii) the adoption of commercial APFOs; and (iv) extensive, multi-parcel downzoning. As Exhibit 27 shows:
When growth subsided in the early 1990s, so did the impetus to adopt new regulations. Only a handful of jurisdictions adopted new growth restrictions between 1992 and 1995.
Not until 1997—two years after the end of the recession—did the pace of growth regulation begin to pick up. Although it is still too early to be definitive, so far, this most recent generation of growth regulations seems qualitatively different from the previous one. Rather than calling for entirely new programs, many communities are refining their existing programs and approaches. Growth management and growth control are out, general plans revisions and specific plans are in. Neighborhood efforts to stop individual projects are also prevalent. Ownership housing, especially if it pays its own way fiscally, is acceptable; apartments projects are not.
Nineteen ninety eight was also a breakthrough year for local urban growth boundaries (UGBs). More than 20 local UGBs have been adopted statewide since 1995, mostly in Northern California. Modeled after Portland's long-standing regional UGB, UGBs are being pushed by environmental organizations throughout California as a comprehensive tool for preserving open space and farmland, and encouraging infill and higher-density development.
There are some important differences between Portland's UGB and those adopted so far in California. Portland's is a regional UGB covering all Oregon jurisdictions in the metropolitan area. Unable to flee Portland's UGB, builders—at least those on the Portland side of the Columbia River—have had to learn how to work within it. In California, UGBs have been adopted individually, by freestanding cities. This puts greater development pressure on those nearby cities that do not have a UGB. Developers operating inside Portland's UGB are also provided with greater approval certainty than so far has been the case in California.
Whatever their specifics, UGBs have become the 1990s version of a growth management "magic bullet." While there is little question that UGBs may in theory contribute to more-organized and less land-extensive growth patterns, whether the specific UGB programs adopted in California will do so remains open to debate. The fact that most UGBs in California have been adopted by initiative, and have not been based on any serious analysis of their prospective impacts, is particularly troubling.
The UCB/HCD growth management survey was undertaken during the summer months of 1998. It included a question regarding growth management measures which were either pending or scheduled for the November 1998 ballot. The tabulated results to that question, presented as Exhibit 28, indicate the current diversity of local approaches to growth management in California.
The relationship between the development approvals process as it exists in law and on paper, and the process as it actually gets played out, is rarely straightforward. Knowledgeable developers have a mantra that applies to California, "Every community and every project is different." To see how different, the regulatory case histories of a representative sample of 46 contemporary housing development projects were examined. This section presents our findings.
Thousands of housing projects consisting of tens of thousands of housing units are proposed each year in California, so finding case studies is not a problem. Finding representative and appropriate case studies, however, takes a little more thought. As a first step, we identified 50-plus cities and counties in which there had been significant residential development activity between 1995 and 1997, and which were projected to continue growing. Our intent was to identify jurisdictions with a strong demand for housing, that were not avowedly anti-growth, and that had experience processing large numbers of project applications. Our preliminary list of communities was selected to insure geographic balance between regions of the state, as well as between cities and suburbs. Where possible, we identified jurisdictions with high levels of both single- and multi-family construction activity. Santa Barbara County was included not because of its high level of construction activity, but to represent Central Coast jurisdictions facing strong development pressures.
We next contacted the planning and building departments in each of the preliminary case study jurisdictions to assess the current availability of completed project files. Where such files were available, local planners were asked to identify "typical" residential development projects for the 1995-1997 period as prospective project case studies. We specified typical projects as:
We specifically asked to avoid projects which were unrepresentative in terms of review times and requirements. 6* These included projects which had either been "fast-tracked" or which had taken an unusually long time to process.
The process of assembling individual project case histories involved four steps. First, the researchers traveled to each case study community and reviewed firsthand the complete case files for each project. To supplement the written material, in-person interviews were conducted with knowledgeable planners when feasible. Second, a draft case history was prepared. Third, each case history was mailed or faxed to the project sponsor or developer for corrections and clarifications. Last, the updated case histories were mailed back to the corresponding planner or review officer for further corrections. The complete case histories are included as Appendix K (single-family projects) and Appendix L (multi-family projects).
At the end of the selection process we had identified 24 single-family projects and 22 multi-family projects in 31 cities and counties for more detailed analysis (Exhibit 29). To facilitate comparisons of the development review process between housing types, where possible, we selected single- and multi-family case studies in the same jurisdiction. Four Northern California cities—Fairfield, Fremont, Salinas, and San Jose—include both single-family and multi-family projects. Four Central Valley jurisdictions—Bakersfield, Fresno, Redding, and Sacramento—include both project forms, as do four Southern California jurisdictions—Carlsbad, Los Angeles, Orange County, and Santa Barbara County.
Any analysis of the entitlement process based on individual project case histories must begin with two caveats. The first is that every project really is different in some critical way. The second is that sample sizes of 24 and 22, respectively, are not always large enough upon which to base definitive conclusions. This is particularly true when those samples are then subdivided into smaller regional sub-samples.
Comparing review performance by region, average approval periods were shortest for Bay Area/Northern California projects (5.7 months), and longest for southern California projects (14.3 months). The average review time for Central Valley projects was 11.3 months. Particularly in southern California, larger projects were reviewed faster on average than smaller ones (the unit-weighted average review time for southern California projects was 10.4 months). In the Central Valley, by contrast, larger projects took slightly longer to review than smaller ones.
Average review times in both southern California and the Central Valley were adversely affected by a few controversial projects. Excluding the most controversial of the southern California projects—Sungate in Santa Barbara County—the average southern California review time was only 8.6 months. Excluding Creekside Greens in El Dorado County and Shepherd Ranch in Fresno—two complicated and controversial projects—the average review time for the other six Central Valley projects was only 5.2 months.
The average Bay Area/Northern California project was subject to 2.8 discretionary reviews; larger projects were subject to slightly fewer reviews than smaller projects. Only two of the six Bay Area/ Northern California projects were reviewed by bodies other than a planning commission or city council. An exception, Evelyn's Ranch in Santa Rosa, was subject to three separate outside reviews. The average Bay Area/ Northern California project was subject to more than seven continuations. Creekside Meadows in Fairfield was subject to 13 separate continuations, boosting its total approval time to a region-high 10 months. There was no consistent relationship between the number of project reviews, the number of continuances, and total project review time. Indeed, Evelyn's Ranch in Santa Rosa, which was subject to four separate reviews, was approved in less than five months. None of the Bay Area project decisions were appealed.
The average Central Valley project was subject to 3.8 reviews. As an outlier, the aforementioned and controversial Shepherd Ranch project in Fresno was subject to seven different discretionary reviews. Seven of the eight Central Valley projects were subject to one or two separate reviews by bodies other than the local planning commission, city council, or board of supervisors. The average Central Valley project was subject to 6.3 continuations. Shepherd's Ranch in Fresno was subject to a region-high 15 continuations. Only one Central Valley project, Coleman Ranch in Sacramento, was the subject of an appeal.
The average southern California project was subject to 3.3 reviews. As in the Bay Area, larger projects were subject to fewer reviews on average than smaller ones (The region's smallest project, Sungate in Santa Barbara County, was subject to the most reviews—seven.). Most southern California projects were subject to additional discretionary review by an agency other than the local planning commission, city council, or board of supervisors. Outside review agencies included the Coastal Commission (Deauville in Carlsbad), local citizen's planning councils (Springfield in Orange County, Del Webb in Riverside County, and Carmel Valley Neighborhood 10 in San Diego), and various inter-agency review committees (Champion Collection in Chino Hills, Northridge Premier in Los Angeles, Sungate in Santa Barbara County, Sunset Hills in Santa Clarita, and Silverthorne in Simi Valley).
Most southern California projects were subject to between two and five continuances. Sungate in Santa Barbara County, and Silverthorne in Simi Valley were subject to the most continuances, as well as to the most reviews, both of which added substantially to their total approval times, as well as raised the various regional averages. Only one southern California project, Sungate Hills in Santa Barbara County, was the subject of formal appeals.
Yields: Somewhat surprisingly, the approvals process did not result in any form of consistent downzoning or density reduction. Among the six Bay Area/Northern California projects, the total number of constructed units actually exceeded the number of approved units, albeit by a tiny amount. 9* Among the eight Central Valley projects, the number of constructed units fell short of the number of approved units in only one case, Fairway Oaks in Bakersfield; in another case, Silverado Oaks in Roseville, the number of approved units exceeded the number of requested units. Among the southern California projects, the number of constructed units fell short of the number of approved units in only two cases: Del Webb in Riverside County (by 3 percent), and Sungate Hills in Santa Barbara County (by 8 percent).
Multi-family projects are typically more controversial and complicated to develop than single-family homes. This is particularly true for affordable rental projects. At the same time, compared to single-family subdivisions, apartment projects tend to be smaller, consume less land, and be located in areas earmarked for multi-family development by existing planning documents.
As a group, the eight Bay Area/Northern California projects have more units and are denser than the four Central Valley and 10 southern California projects. The average number of units in the Bay Area/Northern California projects is 198, versus 153 units in the Southern California projects and 108 units in the Central Valley projects. The average density of the Bay Area projects, 25.6 units per acre, is 50 percent higher than the average density of the southern California projects, and nearly double the density of the Central Valley projects. The higher densities of the Bay Area/Northern California projects reflect the region's much higher land costs and traditional acceptance of higher-density living.
The eight Bay Area/Northern California projects range in size from 30 to 442 units, and in density from 12.2 units per acre to more than 251 units per acre. The 10 southern California projects have almost the same span of sizes (28 to 344 units) and densities (13.3 to 164.7 units per acre). The four Central Valley projects are somewhat more homogeneous: they range in size from 48 to 196 units and in density from 9.1 to 17.6 units per acre.
Compared by affordability level, the 14 market rate projects have more units, occupy larger sites, and are slightly denser than the eight affordable projects. The "average" market-rate project includes 195 units on 10.2 acres, at an average density of 19.2 units per acre. The "average" affordable project, by contrast, includes 102 units on 5.5 acres, at an average density of 18.4 units. As multi-family densities go, neither level is particularly high. Excluding the two highest-density market projects, the market-rate Bay Towers in San Francisco and the affordable Firenze Apartments in Santa Monica, causes the average density of the market-rate and affordable projects to fall to exactly the same level—17.9 units per acre. The smaller size of the affordable projects reflects a number of economic and regulatory realities. In some cases, the fewer the number of affordable units, the smaller the required (or implicit) subsidy, so the easier it is to raise scarce subsidy and financing dollars. Smaller affordable projects are also much easier and less expensive to manage. And since affordable projects often draw undue neighborhood opposition, smaller projects tend to be easier for their sponsors to get through the regulatory process.
A third of the multi-family case study projects were processed as part of a specific or area plan. Given their smaller site sizes and infill locations, this is actually a larger percentage than might otherwise be expected. By contrast, two-thirds of the single-family projects were processed as part of a specific plan or PUD.
Only four projects required some form of review by a state or federal agency. Because its site was partially in a designated wetland, the Willow Park project in Redding was required to obtain a permit from the Army Corps of Engineers. In Carlsbad, the Villas project was subject to review by the California Coastal Commission. Neither of these reviews added significantly to the time or complexity of the overall review process. 10* Not so the Gabilan Hills Townhomes project in Salinas, which was put on hold for several months until receiving its tax credit allocation from the (state) Tax Credit Allocation Commission. 11* The 1010 S. Van Ness project in San Francisco, which was partially subsidized under the federal HOME program, was delayed by community opposition and the involvement of the U.S. Department of Housing and Urban Development. 12*
CEQA reviews were not required of most of the projects, primarily for reasons of location: most were located in areas covered by prior environmental review documents. A full environmental impact report (EIR) was required of only one project, Positano Apartments in Santa Barbara County. Eight projects met CEQA requirements with negative declarations; another six were processed under mitigated negative declarations. Six projects, all in Southern California, were processed under a specific plan EIR, or were "tiered" off a previously certified area plan EIR.
For affordable housing projects, the interaction between the development approvals process and the granting of federal or state subsidy dollars can also prove troublesome and confusing. As required under the Low Income Housing Tax Credit program, affordable housing projects are supposed to have all their land use approvals in place as a condition of subsidy or tax credit approval. Because non-profits typically face difficulties gaining site control, and because development reviews and subsidy reviews often follow different timetables, this is often not the case. It is not unusual, for example, for a non-profit sponsor to begin the development approvals process and simultaneously submit an application for tax credit or federal subsidies. Depending on the size and complexity of the projectùand in the tax credit case, the degree of competitionùa subsidy may be awarded before the final development approval is granted. More commonly, the process of obtaining subsidy dollars or finalizing tax credit syndication extends far beyond the entitlements process.
The average multi-family project was subject to 2.3 separate reviews, one of which was typically a non-legislative review. Non-legislative reviews included design reviews (in Redwood City, Santa Barbara County, and Santa Monica), reviews by state and local agencies (in Redding, Salinas, San Francisco, and Carlsbad), and reviews by administrative and neighborhood advisory groups (San Jose, Sacramento, Chula Vista, Los Angeles, and Santa Barbara County). Non-legislative reviews were far more common in southern California than in the Bay Area or the Central Valley. Affordable projects were subjected to slightly more total and non-legislative reviews than market-rate projects.
The average project was subject to four continuances. Continuances were far more prevalent among Bay Area/Northern California projects than among Central Valley or southern California projects. The average number of continuances per Bay Area project was 6.5, regardless of project size. The average number of continuances per southern California project was only three, while the average for Central Valley projects was two continuances. Affordable projects were slightly less likely to be subject to a continuance than market rate projects. Two projects, 1010 S. Van Ness in San Francisco, and Camino Real Homes in San Diego, were approved on appeal of a prior turndown by the same legislative body. Both were affordable projects. While most of the case study projects were processed in five months or less, a few—1010 S. Van Ness and Bay Towers in San Francisco, Vista Angelina Apartments in Los Angeles, and Camino Real Homes in San Diego—took a year or more to review. Three of the four are affordable projects. Vista Angelina in Los Angeles, an affordable project, was delayed by the granting of financing approvals from the Los Angeles Housing Department. Camino Real in San Diego, also an affordable project, was delayed by a HUD grant approval. (Bay Towers in San Francisco was delayed by design changes and the need to accommodate an adjacent historic building.) 1010 S. Van Ness, a small (30 units), affordable project partially designed for people with AIDs, was delayed by relentless neighborhood opposition.
As with the single-family projects, required review times and continuances varied with the complexity of the approvals process:
Project density and size also affected review times. The average review time for the eight projects with a density of 20 units per acre or more was 9.1 months. The average review time for the 14 less dense projects was only 5.4 months. The higher-density projects also required an average of 5.6 continuances, compared to only 3.2 continuances for the lower-density projects. Project size was of slightly lesser importance. The ten projects with fewer than 100 units were approved in an average of six months and required 3.3 continuances. The 12 projects with more than 100 units were approved in an average of 7.3 months and required 4.8 continuances. As with the single-family projects, readers should be careful not to draw too many conclusions from comparisons based on small differences and small sample sizes.
Yields: Surprisingly, perhaps, the approvals process did not result in any consistent loss of units. The review process resulted in a unit reduction in only one Bay Area/Northern California project (The Hamptons, in Cupertino). In southern California, the number of units were reduced in three projects: Bolero in Chula Vista, the Villas at El Camino in Carlsbad, and Vista Angelina Apartments in Los Angeles. Both the Hamptons and the Villas at El Camino are very large projects. Three projects, Bay Towers in San Francisco, Brookfield Loop in Bakersfield, and Positano Apartments in Santa Barbara County, were ultimately constructed with a few more units than initially approved or requested.
All the case study projects fall under the provisions of the Permit Streamlining Act (PSA, pursuant to Government Code Sections 65950 and 65957) and CEQA Guidelines (Sections 15107, 15108, and 15109), which stipulate maximum permissible review times. Negative declarations are to be adopted within 180 days from when a project application is accepted as complete; a 30-day extension may be granted due to applicant delays. Projects for which a negative or mitigated declaration has been adopted must be approved within 60 days, with allowances for a 90-day extension with the applicant's consent. Thus, allowing for all possible extensions, projects for which a negative declaration has been adopted must be approved in not more than 360 days (12 months).
Similar requirements apply to EIRs. By State law, a Final EIR must be certified within one year from when a project application is accepted as complete, allowing for a 90-day extension by mutual consent, or a suspension of the review clock due to unreasonable applicant delays. Projects are to be approved within 180 days of final EIR certification, allowing for a single 90-day extension with the applicant's consent. Thus, allowing for all possible extensions, the combined EIR/ project approval process should take no more than 755 days (25.2 months).
Based on the times outlined in Exhibits 30 and 31, and Appendices L and M, these standards appear to have been exceeded in three of the single-family case projects (Creekside Greens in El Dorado County, Sungate in Santa Barbara County, and Shepherd's Ranch in Fresno), and two of the multi-family case projects (Bay Towers in San Francisco, and Camino Real Homes in San Diego), even allowing for authorized time extensions. 13* Assuming extensions were not forthcoming, project review times for two additional single-family projects (Northridge Premier in Los Angeles and Silverthorne in Simi Valley) and one additional multi-family project (1010 S. Van Ness in San Francisco) would also have exceeded State PSA/CEQA permitting-time standards.
Because the researchers did not request access to all project review documents and correspondence—as it was not the focus of the research—we are unable to determine whether the PSA/CEQA standards were actually exceeded. It is clear, based just on this limited sample, is that permit times still vary widely throughout California, and that the spirit of the PSA/CEQA standards, if not the letter, is not always observed. What is also clear is that State laws requiring local permitting processes to be streamlined cannot be effective without adequate monitoring and enforcement.
The 46 case studies assembled for this analysis provide the most comprehensive and far-ranging look at the residential entitlement system ever undertaken in California. Taken together, the case study profiles suggest that while the housing entitlements process in California may be convoluted and frustrating (particularly to project sponsors), it does not, as a rule, result in interminable approval delays or in ad hoc unit and density reductions—at least in communities which are generally favorable to development. Most of the case study projects were processed with only two or three separate reviews in an average of six to eleven months, depending on the project type and location. Despite their higher densities, the multi-family case study projects were generally processed faster and with fewer review rounds than their single-family counterparts (Due to the difficulties of obtaining financing, affordable multi-family projects took longer to approve on average than market rate projects.). Perhaps most surprisingly, in a couple of cases, the number of units approved was generally equal to or greater than the number of units requested.
This rather benign picture of the residential entitlements processing California is subject to five important caveats. First, the cities and counties from which the case study projects were solicited are not a representative cross-section of the entire spectrum of residential permit processing in California. As noted previously, we selected jurisdictions based jointly on prior building permit activity and on projected housing demand. As a result, the sample generally excludes jurisdictions where there is significant need, but which were unwilling or unable to accommodate recent production. In other words, compared to all California cities and counties, the sample is more biased toward growth-accommodating jurisdictions.
Nor are the 46 case studies necessarily representative of the full range of residential projects. They were not selected at random, but by local planners as reasonably representative of the development approvals process in their jurisdictions. Case selection was not subject to review by the project developer, the research team, or HCD. Moreover, the period for which they were selected, 1994-97, was one of moderate development activity, so the types of project approval delays typically associated with rapid growth and frenetic development were less likely to have occurred. We make these observations not to suggest that local planners sought to "stack the deck," but rather to note that we imposed no controls on the case studies to insure that they be uniformly representative of the true population of housing projects under review.
As a third caveat, we did not examine whether or how the entitlements process added to the cost of developing housing. It is entirely possible that the "price" of obtaining rapid and non-controversial development reviews is for the developer to undertake additional and costly design changes, impact mitigations, and infrastructure and service commitments. Again, we make this observation not to criticize, but simply to point out that we do not know.
One thing we do know is that two-thirds of the single-family case studies were processed as part of a pre-approved specific, community, or area plan. This suggests that for many of the reviewed projects, the most onerous, time-consuming, and controversial part of the development approvals process had already been completed. Prior to the recession of 1991-93, developers throughout the state (but especially in southern California) had stockpiled hundreds of thousands of full or partial development approvals in the form of development agreements, specific plans (including tiered EIRs), zone changes, and vested tentative maps. When, after three years of recession, development activity started picking up in 1994, many developers—or more likely, their successors—were able to makes use of these stockpiled approvals. As that stockpile is now mostly drawn down, more and more developers will have to entitle entirely new sites, adding to the time, difficulty, and cost of obtaining future development approvals.
This last discussion points to the most serious flaw in our analysis: while we have a relatively complete picture of what the approvals process looks like after the first formal permit application has been filed, we know little of what transpired prior to that point. We know little, for example, of prior annexation battles, or of the difficulties of gaining prior specific plan approvals, or of pushing through a substantial general plan amendment, or of undertaking a master, program, or specific plan EIR. We know little of how many housing units (if any) were lost or conceded in the earliest (and largely undocumented) part of the development approvals process. Some of the difficulties inherent in entitling land for housing were indicated in our limited follow-up interviews with builders, but in many cases the earliest—and perhaps most critical—phase of the land delivery system remains inadequately understood.
California Government Code, Sections 65950, 65957
California Resources Agency, Governor's Office of Planning and Research. 1995. CEQA Guidelines. Sections 15107-15109. Sacramento.
Construction Industry Research Board. Residential permit statistics. 1995, 1996, 1997.
Daniel Curtin, Jr. 1997. California Land Use & Planning Law. Point Arena: Solano Press Books.
William Fulton. 1991. Guide to California Planning. Point Arena: Solano Press Books.
John Landis, Robert Olshansky, Rolf Pendall, and William Huang. 1995. Fixing CEQA: Options and Opportunities for Reforming the California Environmental Quality Act. Berkeley: California Policy Seminar.
Ned Levine, Madelyn Glickfeld, and William Fulton. Home Rule: Local Growth... Regional Consequences. (forthcoming.)
Rolf Pendall. 1998. Environment and Planning B.
Copyright ¨ 2000. All rights reserved.