Does zoning help or hinder transit-oriented (re)development?

Abstract

Chinese

Despite its reputation as a car-oriented city, the Los Angeles metropolitan area has made substantial investments in developing rail transit since 1990. Most new stations were added to an already dense built environment, with auto-oriented zoning and established land use patterns. In this paper we ask whether redevelopment is occurring around Los Angeles rail stations, and whether zoning and related policies are facilitating or constraining transit-oriented development. We conduct case studies of five stations, documenting zoning near stations, as well as the amount and type of new development after stations opened. Results illustrate that incompatible zoning and related land use policies may constrain growth near stations, but TOD-friendly zoning alone is not sufficient to spur development.

Keywords

housing markets land use planning local government public transportation urban spatial structure zoning

Introduction

It is difficult to imagine New York City without the subway, London without the Tube or Paris without the Métro. In cities that built intra-city passenger rail systems more than 100 years ago, the current built environments have grown up around transit infrastructure. Successive waves of growth have added height and density near stations, creating clusters of jobs and housing. Patterns of high-density mixed use development are still emerging around rail systems built in the 1960s and 1970s, in cities such as Washington DC, the San Francisco Bay Area, Mexico City and Seoul. As cities around the world grapple with increasing traffic congestion, local governments are developing new rail systems and expanding existing ones, from Beijing to Brasilia, Manhattan to Chennai. These large ongoing public investments raise a number of questions. When rail infrastructure is added to an already dense built environment, will new development patterns emerge? What form might redevelopment near stations take, and when will it become apparent? Does zoning facilitate or impede transit-oriented development around new stations?

In this paper, we investigate these questions by documenting zoning and new development around five Metro rail stations that opened in central Los Angeles between 1993 and 2003. Despite its reputation as a car-oriented city, the Los Angeles metropolitan area has for decades been investing substantially in rail transit. In the 1960s, when the city was losing population and jobs to rapid suburbanisation, regional leaders believed that a subway along the Wilshire Corridor would revitalise the central city and so established a new transit authority. After three unsuccessful attempts, in 1980 voters approved a proposal for a countywide rail network, along with a one-half cent sales tax to fund development (Fulton, 1996). Additional sales tax measures were passed in 1990 and 2008. Together these taxes generate approximately US$2.3 billion annually, of which about half is available for rail capital construction.¹ To date, LA Metro has invested US$9 billion (in nominal terms) in rail infrastructure, building more than 75 stations along six rail lines, and has by far the largest transit construction programme in the USA (Nelson and Weikel, 2016). City and county governments (and many voters) see rail transit as the key to reducing congestion, restructuring urban form, creating a livable city, and attracting economic development.

However, several factors may hinder ridership and the potential for economic growth near stations. Although the Los Angeles–Orange County MSA has a dense residential population, employment is distinctly polycentric and dispersed throughout the MSA, with a relatively weak CBD (Giuliano et al., 2015). Development patterns are highway-oriented; the largest employment clusters are located along major freeway corridors. In most instances, the LA Metro rail stations were built in already dense neighbourhoods with auto-oriented zoning and established land-use patterns. The station locations were selected more for political expediency than economic efficiency, which may diminish the potential for ridership and nearby development (Elkind, 2014; Taylor et al., 2009). Despite the region’s long-standing and substantial investment in rail infrastructure, LA Metro currently has fewer daily boardings than 30 years ago, when the system was only buses (Nelson and Weikel, 2016). Anecdotal observation suggests that relatively few stations have experienced physical and economic growth after station opening.

Urban economics models predict that building new rail stations should increase the accessibility of surrounding neighbourhoods, leading to higher land values and attracting higher density development (Alonso, 1964; Mills, 1967; Muth, 1969). But redevelopment in dense urban areas is generally more costly than ‘greenfields’ development, requiring demolition of existing structures, possible environmental remediation, and land assembly from multiple owners (Wheaton, 1978). Thus redevelopment near new stations will only occur if land values around the station have increased enough to support substantially higher density development than the existing structures.

Additionally, zoning may constrain higher-density redevelopment near stations through a variety of mechanisms, including bans on TOD-compatible land uses, density limitations, or procedural rules that add to ‘soft’ development costs (Glaeser et al., 2006). Of particular concern in California is the statewide California Environmental Quality Act (CEQA), which requires public review and comment as part of an environmental impact review for nearly all development projects (Olshansky, 1996). Neighbours and public leaders may use CEQA and other zoning rules to impede unwanted redevelopment. Conversely, public officials and nearby residents can facilitate change near stations by increasing the allowable density, granting density bonuses, fast-tracking proposed development, or soliciting development through public–private partnerships (Cervero, 2004). In short, whether redevelopment occurs near new stations depends on the interaction between property values, pre-existing land uses, and zoning.

This paper’s primary contribution is to investigate the mechanisms by which zoning influences redevelopment around rail stations, using a rigorous comparative case study design and careful measurement of zoning and development outcomes. Our study augments the prior literature on transit-oriented development in several ways. First, although numerous quantitative studies have examined the effects of transit investments on economic outcomes, controlling for economic conditions and demographics, the regression models do not include controls for zoning or existing land uses.² Second, in contrast with most qualitative case studies, we draw clear contrasts between successful and unsuccessful TOD sites. For instance, studies by Belzer and Autler (2002), Cervero (2004), Ellis (2005), and Greenberg (2004) take a ‘best practices’ approach, evaluating the effectiveness of zoning, planning and fiscal policies used at successful TOD locations. Comparing zoning across stations with and without TOD generates insights into how zoning may constrain development. Third, our research design focuses on five cases drawn from a single city, better allowing us to isolate the role of site-specific zoning from citywide economic conditions and institutional factors.

This paper complements our prior quantitative analysis that estimates longitudinal changes in employment and housing outcomes for 28 LA Metro stations over a 20 year period (Schuetz et al., 2015). Comparing station areas to matched control neighbourhoods, that study finds no evidence that station openings led to immediate changes in employment or housing markets, but a few stations saw employment growth within five to ten years after opening. This qualitative analysis seeks to better understand and contextualise the quantitative results, namely, why post-station growth was relatively rare and slow to materialise. The contrasting empirical methods of the two papers have different relative strengths. The quantitative analysis controls for observable demographic and economic conditions near stations, in order to isolate the impact of station opening, and estimates the average impact across a large number of stations. The case studies offer rich detail on zoning, land use, and idiosyncratic neighbourhood characteristics that influence redevelopment outcomes. This paper – the only case study matched to a quantitative analysis of the same city – provides unique insights into the successes and challenges of TOD.

The remainder of the paper discusses our empirical approach and presents findings. We document existing land use patterns, zoning, and changes in land use and the built environment since the stations opened. Two station areas – Hollywood/Vine and Pershing Square – have seen growth in housing and commercial uses. These areas have TOD-compatible zoning, strong real estate markets, and benefited from targeted local government efforts to encourage development. The other three stations – Vermont/Santa Monica, Highland Park and Civic Center – have seen few changes in land use or buildings. Two areas have incompatible or ambiguous zoning, and two areas have had relatively soft real estate markets. These results suggest that zoning may constrain redevelopment, but that TOD-compatible zoning alone will not induce development. Although the analysis focuses on Los Angeles, the findings are relevant for other auto-oriented cities in the USA, Canada and Australia that are trying to alter travel behaviour and land-use patterns through rail transit investments.

Data sources and empirical approach

Our research design examines cross-sectional variation and changes over time in zoning, land use, and physical environment, around five LA Metro rail stations that opened between 1993 and 2003.

Selection of study neighbourhoods and stations

To ensure variation in outcomes, we identified two station areas that have had substantial post-station redevelopment as well as three stations with little or no change. Based on prior quantitative analysis of the LA Metro system, the stations were selected to be broadly representative of stations throughout Los Angeles (Schuetz et al., 2015).³ Comparing outcomes for several cases in the same city allows us to focus on highly localised variations in zoning and land use (King et al., 1994). Although studying five stations somewhat limits the depth of each case, it allows us to move beyond site-specific idiosyncrasies and draw more generalised inferences. Pairwise comparisons between cases that share similar attributes (i.e. same immediate neighbourhood) further control for potential confounding factors.

The five stations are located in three neighbourhoods (Figure 1). Pershing Square and Civic Center are in Downtown Los Angeles, which is the region’s largest employment centre but historically had little residential population. Hollywood/Vine and Vermont/Santa Monica both lie within the Hollywood neighbourhood in Central Los Angeles, a relatively high density mixed-use corridor, with commercial uses concentrated along major streets and medium-to-high density housing along side streets. The third neighbourhood, Highland Park, is a lower-density, mostly residential area in northeast Los Angeles.

Figure 1.

Rail stations and employment centres in Los Angeles County.

We document pre-existing land-use patterns, baseline zoning, and supplemental neighbourhood plans, as well as targeted government efforts to promote development near stations. We assess the compatibility of each station’s zoning with the principles of TOD: namely, the feasibility of developing medium-to-high density residential and commercial structures. Case studies are presented in descending order of TOD-compatibility and extent of local government promotion of development. We also provide context on broader neighbourhood demographic and economic factors that may influence nearby development. While our results suggest that zoning and prior land uses are associated with the likelihood of TOD, they do not establish a causal relationship, because there may be unobservable neighbourhood characteristics correlated with both redevelopment and zoning.

Data sources

Information on zoning and land-use plans were obtained from the Los Angeles Department of City Planning website. The city’s online Zone Information and Map Access System (ZIMAS) was used to determine the current zoning classification for all parcels within one-quarter mile of the station.⁴ We follow the literature in defining TOD-compatible zoning as a regime that allows both residential and commercial uses (such as retail, household services, hotels and offices) at medium-to-high densities, as set by floor-to-area ratio (FAR) or dwelling units per acre (Belzer and Autler, 2002). We also assess zoning complexity, based on the number of different land-use plans that regulate development in each neighbourhood. Land-use plans and adoption dates were obtained from the Los Angeles City Planning website.

To track redevelopment and land-use changes near stations, we document the existing structure type, land use, year structures were built, and number of residential units (if relevant). Property-level data on housing sales volume, sales prices, and new housing construction from 1988 to 2012 were obtained from DataQuick. Transactions data have limited coverage of multifamily rental buildings and conversions from non-residential buildings, therefore we also rely on parcel-level data from the Los Angeles County Tax Assessor’s website and the online building permits database of the Los Angeles Department of Building and Safety.

The location and opening dates of rail transit stations were assembled from the LA Metro website and supplemental documentation. Station addresses were geocoded and matched to census geographies. Population characteristics were assembled from tract-level data from the 2005–2009 American Community Survey. Because stations often abut multiple census tracts, we calculate weighted average characteristics for each station, based on the land share in each adjacent tract. Demographics around stations are therefore approximate.

Station case study findings

For each station, we present information on three key questions: are zoning and land uses near the station compatible with TOD?; has zoning changed since the station opened?; and has any redevelopment occurred near stations after opening? We also provide general context for each neighbourhood’s current physical and economic environment.

Hollywood/Vine Station: Compatible zoning, government as development facilitator

The Hollywood neighbourhood is a densely developed urban area that includes tourist-oriented retail and entertainment, several residential neighbourhoods, and office and industrial zones. The Hollywood/Vine Metro station sits at the intersection of two major commercial thoroughfares; the Hollywood Walk of Fame borders the station entrance on Hollywood Boulevard (Figure 2). Many of the surrounding hotels, apartment buildings and theatres date from the 1910s through 1930s and are architecturally notable. The demographically and economically diverse population mirrors the city’s overall composition (Table 1).

Figure 2.

Hollywood/Vine station.

Table 1.

Station area characteristics and development outcomes.

	Hollywood/Vine	Pershing Square	Vermont/SantaMonica	Highland Park	Civic Center
TOD-compatible zoning	Yes	Yes	Yes	Uncertain	No
Real estate market strength	Strong	Strong	Weak	Increasing	Strong
Post-station redevelopment	Yes	Yes	No	No	No
General
Geographic area	Hollywood	Downtown LA	East Hollywood	Arroyo Seco	Downtown LA
Year station opened	1999	1993	1999	2003	1993
Line(s) served	Red	Purple, Red	Red	Gold	Purple, Red
Avg. daily boardings (2013)	6175	10,503	5860	2718	6107
Zoning and land-use plans
Plan name (year)	Hollywood Community Plan (1988)	Central City Community plan (2003); Adaptive Reuse Ordinance (1999)	Hollywood Community Plan (1988); Vermont/Western TOD Specific Plan (2001)	Northeast Los Angeles Community Plan (1999); Highland Park Historic Preservation Overlay Zon (1994); Avenue 57 TOD Neighborhood Plan (2002)	Central City Community plan (2003); Bunker Hill Specific Plan(2013); Adaptive Reuse Ordinance(1999)
Neighbourhood characteristics
Population per acre	245	77	189	86	117
Median household income	36,983	20,822	32,529	45,013	18,411
Black (%)	6	14	4	3	11
Hispanic (%)	32	21	67	76	18
1–4 family housing (%)	10	1	15	59	2

Notes: Station opening date and ridership obtained from the MTA. Information on zoning and other plans assembled from the Los Angeles City Planning Department. Development outcomes from DataQuick housing transactions, Los Angeles County Tax Assessor, and City of Los Angeles Department of Buildings. Neighbourhood characteristics from 2005–2009 American Community Survey.

Zoning near the station allows high-density residential and commercial uses (Table 1). Some older buildings exceed currently allowable densities, meaning that if they were demolished, replacement buildings would be smaller than current structures. Development in the area is regulated by the Hollywood Community Plan, which was adopted in 1988, during the Metro planning process but before completion of the Red Line. The Plan calls for the preparation of new station area master plans ‘if higher intensity development is to be encouraged’. A new Hollywood Community Plan Update was adopted in 2012, which would have increased allowable density and encouraged additional development along commercial corridors such as Hollywood Boulevard (Zahniser, 2013). However, the new plan was highly controversial among residents, who challenged it in court. It was rescinded in 2014 after a judge ruled that the Environmental Impact Report contained errors. Until 2012, the station area also fell under the statewide Community Redevelopment Agency (CRA) boundaries, which granted additional tax revenues for development (Blount et al., 2014).⁵

Local government exerted targeted and sustained efforts to facilitate development of a mixed-use TOD project above the Hollywood/Vine station. Through a 2001 Joint Development Agreement between the MTA and private developers, the MTA used eminent domain to acquire several small lots, demolished the existing structures (mostly small retailers) and merged the lots into a single large parcel (City of Los Angeles, 2006). The redeveloped site, completed in 2009, consists of a W hotel, condominiums, rental apartments (of which about 20% are required to be affordable to low- and moderate-income households), and approximately 60,000 square feet of ground floor commercial use (Figure 3).⁶ Monthly rents for the market-rate apartments in the building – ranging from US$2000 for a studio to over US$7000 for a two-bedroom – are likely out of reach for many of the neighbourhood’s existing residents, given the median household income of US$31,000.⁷ The redeveloped site represents both a shift in land uses and an increase in housing and commercial square footage.

Figure 3.

Pershing Square station.

The broader real estate market around Hollywood/Vine has also strengthened since the Great Recession. Residential sales near the station have increased substantially since about 2006, with roughly 40 sales per year and a median price of US$660,000 per unit. Retail activity near the station has increased, with nearly 40 new restaurants opened since 2006.

Pershing Square station: New Adaptive Reuse Ordinance encouraged housing

The Pershing Square station is located in the Historic Core sub-neighbourhood of Downtown Los Angeles, an economically and demographically diverse area that has undergone substantial population change in the past 10 years. The station’s name derives from the adjacent public plaza, which houses various activities, from a weekly farmers’ market to free meals for Downtown LA’s sizeable homeless population (Figure 3). The station is surrounded by high-rise office buildings with ground-floor retail and restaurants, high-rise residential buildings, and surface parking lots. Approximately two-thirds of buildings date from the 1910s and 1920s and have distinctive architectural features.

Nearly all land near Pershing Square is zoned for high-density commercial and residential activity (Table 1). However, many existing structures exceed the current allowable densities, some by as much as 30%. The Historic Core neighbourhood has seen substantial change since the city’s adoption of the Adaptive Reuse Ordinance (ARO) in 1999. The ordinance encouraged conversion of vacant or underutilised commercial buildings into housing through an expedited approval process and waivers of certain zoning and building code provisions.⁸

No new structures have been built within several blocks of the Pershing Square station, but at least six nearby high-rise buildings underwent residential conversions since 2005, including two conversions on adjacent or facing blocks.⁹ These buildings contain roughly 500 housing units (both condominiums and rental apartments). No residential sales occurred near Pershing Square station from 1988 through 2004, rising to about 50 sales per year from 2005 to 2008, and more than 100 sales per year from 2009 to 2012. Median condo prices range from US$250,000 to US$500,000.

With the increased quantity and quality of housing stock in the neighbourhood, the residential population has more than doubled since 2000. More affluent households have moved into the newly converted apartments and condominiums. Nearby retail is also shifting from stores and services oriented to working-class Latinos, towards trendy, upscale stores and restaurants (Immediato, 2015; Wotapka, 2013).

Vermont/Santa Monica station: Compatible zoning, weak demand

The Vermont/Santa Monica station is also located in the Hollywood neighbourhood, about 2.5 miles southeast of Hollywood/Vine, at the intersection of two active commercial streets (Figure 4). Low-rise commercial buildings surround the station, including a shopping centre with parking lot, fast-food restaurants, and a gas station. Nearly all structures are fewer than five storeys and date from the 1950s through 1970s. The station is one block north of Los Angeles City College, a public community college, which is the area’s largest employer. While both Vermont Avenue and Santa Monica Boulevard are primarily commercial, the side streets are mostly residential, with modest one- and two-storey single-family houses and small apartment buildings dating from the 1920s through 1950s. The median household income and population density are similar to that of the Hollywood/Vine station area, and nearly two-thirds of the area’s population was Hispanic (Table 1).

Figure 4.

Vermont/Santa Monica station.

Zoning near Vermont/Santa Monica reflects the area’s mixture of uses: parcels along the two main streets are zoned medium-density commercial while the side streets are zoned for medium-density multifamily housing (Table 1). Land use is governed by the Hollywood Community Plan and the Vermont/Western TOD Specific Plan, adopted in 2001 to encourage additional development near the station. The TOD Specific Plan allows a 50% increase in commercial densities along the main intersection, while preserving existing densities along the residential side streets. The Specific Plan enumerates various other goals: encouraging pedestrian-oriented streetscape improvements, ensuring ‘a well-planned combination of commercial and residential uses with adequate open space’, as well as the ‘creation of a lively civic atmosphere’ and less specifically defined aesthetic goals.

Despite the adoption of the TOD Specific Plan, no new residential or commercial buildings have been developed near the station since it opened. The existing land-use patterns have not altered much, nor have the streetscape improvements envisioned in the specific plan emerged. Unlike the Hollywood/Vine and Pershing Square stations, housing transactions suggest that the real estate market near Vermont/Santa Monica has not been strong enough to support new, higher density development since the station opened. From the mid-1990s through the housing collapse in 2007, sales of small apartment buildings were modest but consistent, with average prices around US$200,000 per unit. Beginning in the Great Recession, sales volume declined to fewer than ten sales per year, with prices around US$150,000 per unit. In the absence of rising property values, TOD-compatible zoning and land-use plans have not generated redevelopment.

Highland Park station: Ambiguous zoning, emerging demand

Highland Park is an older neighbourhood about six miles northeast of Downtown Los Angeles. The above-ground light rail station is located one block northwest of the neighbourhood’s central commercial strip along North Figueroa Street (Figure 5). The strip is composed mostly of one- and two-storey buildings dating from the 1900s to 1920s, currently occupied by a mixture of retail and services, as well as an historic 1924 movie theatre. Three-quarters of residents are Hispanic, and many of the businesses cater to Latino customers with services such as specialty foods, payment remittances, and immigration-oriented legal services. The side streets off Figueroa are mostly residential, with one- and two-storey single-family and small multifamily buildings dating from the early 20th century, and a notable concentration of Craftsman buildings that are designated for historic preservation. Like the Historic Core in Downtown Los Angeles, in the past few years, the neighbourhood has drawn attention – and controversy – from local and national media as an example of gentrification, with a growing population of younger, more affluent college-educated non-Hispanic whites (Clark, 2015).

Figure 5.

Highland Park station.

Zoning in Highland Park is complex, governed by three sets of plans: the Northeast Los Angeles Community Plan (adopted 1999), the Highland Park Historic Preservation Overlay Zone (adopted 1994), and the Avenue 57 Transit Oriented Development Neighborhood Plan (adopted 2002). All three of these were in effect before the Gold Line began operations, although the Community Plan and TOD Neighborhood Plan were adopted during the rail system’s development. Figueroa Street is zoned for high density commercial use, while the majority of the surrounding neighbourhood is zoned for medium-density housing (Table 1). Some existing commercial and residential structures are at or above currently allowable densities. The three plans not only have different rules governing allowable uses, dimensions, procedures, and other specific components, but goals stated in the plans reveal some contradictory expectations. The Community Plan largely frames development as a challenge to be controlled, rather than a desired outcome. It repeatedly mentions the need to separate commercial and residential uses – more typical of older, single-use zoning rather than New Urbanist-inspired mixed-use TOD – and calls for more parking to serve commercial areas. According to the Community Plan, the TOD-specific plan ‘is being prepared to regulate … and guide development so that the mistakes of the recent past are not repeated’ (Community Plan, p. I-5). Throughout the document, the destruction of historic buildings and loss of neighbourhood character incurred by new development are referred to as ‘mistakes of the past’. These motivations are also frequently referenced in the HPOZ, which establishes procedural requirements to review any proposed demolition, alteration, or redevelopment of existing structures.

The language of the TOD Neighborhood Plan reflects an attempt to balance multiple economic, physical, aesthetic and social goals. The plan outlines some incentives – reduced parking requirements, increased FAR and streamlined approval process – designed to induce reuse of existing structures and introduce residential elements along Figueroa Street. The plan calls for live-work spaces for professional and creative occupations, and designates an ‘Artwalk’ area that allows artistic production activities in residential areas. Although some density bonuses are offered in exchange for improving pedestrian amenities, the plan still stresses the need to maintain the ‘historic character’ of the neighbourhood, including ‘limiting the massing of parcels to maintain an appropriate scale of development’. It is unclear whether the type of land assembly and redevelopment which occurred at Hollywood/Vine would be allowed in Highland Park. The TOD Neighborhood Plan also outlines an ambitious set of social goals:

maintain a diverse community, where people of many different ages, incomes, family formation types, and cultural perspectives will live, work and shop in harmony … support and expand the traditional local population of working writers and artists.

No new development has occurred adjacent to the Highland Park station or along the Figueroa Street commercial strip since the station opened, and minimal development has occurred in the adjacent residential neighbourhood. Media reports on gentrification describe increased renovation and sales activity among older buildings in the neighbourhood since 2012, resulting in higher rents or sales prices, but it does not appear that these activities have yet increased building supply. The overall land-use mix has not changed, although tenant turnover along Figueroa Street shows signs of commercial gentrification; upscale restaurants replacing fast food, for instance. Building permits suggest most commercial property renovations have been relatively modest, such as replacing HVAC systems or façade improvements. The housing market around Highland Park is fairly active, with approximately 50 to 100 sales per year during the 2000s, including single-family and small multifamily buildings. Single-family house prices in recent years have averaged US$300,000, while multifamily buildings sold for around US$150,000 per unit. Residential sales volume and prices decreased sharply in the Great Recession and had not recovered by 2012. The strengthening real estate market suggests increasing demand for residential and commercial space near Highland Park. Whether new development occurs over the next few years likely depends on which of the many land use plans will dominate, and on political support from residents and public officials.

Civic Center station: Zoning limits TOD-compatible land uses

The Civic Center station is located in the Civic Center sub-neighbourhood of Downtown Los Angeles, two blocks west of City Hall and about 0.7 miles north of Pershing Square (Figure 6). The Central City Community Plan describes the neighbourhood as the ‘governmental, financial and the industrial hub of Los Angeles’. Land uses near the station area are predominately city and county government buildings, a public park, and a large performing arts complex, including the Frank Gehry-designed Disney Hall and newly opened Broad Art Museum. There is limited housing and thus little residential population immediately adjacent to the station, but the station is within walking distance of the growing Historic Core neighbourhood.¹⁰ The Civic Center station area has by far the highest employment density of the study stations; two-thirds of jobs are in public or institutional sectors.

Figure 6.

Civic Center station.

Reflecting existing land uses, more than half the land near the Civic Center station is zoned exclusively for Public Facilities, with allowable uses limited to government buildings, fire and police stations, libraries, and similar public uses (Figure 6). On the south side of the station, parcels are zoned for high-density commercial and multifamily residential (Table 1). These parcels are occupied by large office and apartment buildings that were built prior to the station opening.¹¹ The Central City Community Plan, adopted in 2003, encourages more residential growth, supporting locally serving businesses, inducing more pedestrian-oriented development, and taking advantage of unused office space. The plan calls for projects to ‘maximize the development opportunities of the future rail transit system while minimizing adverse impacts’.

No residential or commercial buildings have been developed around the Civic Center station within the last 25 years, nor has the land-use mixture altered. The government office buildings were mostly built during the 1960s and 1970s, although new buildings for the Police Department and California Transportation Department were completed in the past six years. The most notable change in the physical environment near the station was the 2012 conversion of a surface parking lot into Grand Avenue Park, which extends several blocks east and west of the station (Figure 6). The City of Los Angeles developed the park, which has landscaped open space and hosts public programmes such as concerts and holiday celebrations. The strong real estate market around Pershing Square suggests that there might be demand for residential or commercial development near Civic Center, but the current zoning and public land ownership are likely barriers to private-sector development near the station.

Discussion and policy implications

Los Angeles is among many large cities worldwide that are currently investing substantial resources to build intra-city rail systems. Local policymakers hope that, in addition to improving transportation flows, their investments will spur physical redevelopment around stations. Our results show considerable variation in redevelopment and land-use change near five LA Metro stations. The neighbourhoods in which these stations were built – central business districts, mixed-use urban corridors and older residential neighbourhoods – have parallels in other car-oriented cities that are seeking to transform the built environment through rail transit, particularly those across the USA, Canada and Australia. Although the institutional context varies across cities and countries, our research suggests some generalisable results of opening stations in densely built neighbourhoods. To answer the questions posed in the introduction, the probability, form and timing of transit-oriented redevelopment depends on compatible zoning, strong real estate markets, active engagement by local government agencies, and (in most cases) political support from neighbourhood residents.

Will redevelopment occur around new rail stations in dense urban areas?

New rail stations in densely built neighbourhoods may experience noticeable amounts of redevelopment, but only where land values are high and it is possible to redevelop at substantially higher density than current buildings. Three station areas – Hollywood/Vine, Pershing Square and Civic Center – have had strong real estate markets for a sustained period after stations opened. At two of them, targeted public efforts removed prior obstacles to redevelopment: using eminent domain to overcome the barrier of fragmented land ownership (Hollywood/Vine) and allowing conversion of underutilised existing structures (Pershing Square). Underlying land values and real estate markets are highly localised, so incentives for redevelopment can vary considerably even within small geographic distances. Hollywood/Vine and Vermont/Santa Monica are located roughly 2.5 miles apart on the same rail line and within the same Community Planning area, yet the Vermont/Santa Monica area has substantially weaker demand. No development has occurred near the station – nor have existing house prices risen – despite the adoption of TOD-compatible zoning. This contrasts with some predictions in the literature, which assert that TOD will emerge naturally as long as zoning does not constrain it (Bernick and Cervero, 1997; Levine, 2006; Suzuki et al., 2013).

While research in other cities suggests that ridership often predicts successful TOD, it is not clear that improved accessibility explains redevelopment in our cases. Both Hollywood/Vine and Pershing Square stations have relatively high ridership, but Civic Center and Vermont/Santa Monica have roughly comparable ridership (Table 1) yet have experienced no redevelopment. Highland Park’s strengthening real estate market has not increased ridership at the station. Public transit, especially rail, plays only a minor role in Los Angeles, compared with other cities: rail transit accounts for only 1% of commuters in LA, far lower than cities such as Boston, Chicago, or San Francisco (19%, 13% and 8%, respectively; ACS, 2010–2014). Under such conditions, stations may not significantly influence land values.

What will station-oriented redevelopment look like, and how long will it take?

Compared with TOD around stations built in greenfields areas, redevelopment near urban stations will be more variable in appearance and may take many years to emerge. The form and timing of redevelopment reflects land values and zoning, as well as other public-sector actions – or lack of actions – around stations. Land assembly through eminent domain at Hollywood/Vine enabled an exemplary TOD project on the station parcel. However, the city’s actions were politically controversial and challenged in court, creating a lengthy delay. Redevelopment in Pershing Square took a notably different – and less controversial – form, due to the prior built environment and passage of the Adaptive Reuse Ordinance: all new housing was built through conversion of older commercial buildings. Redevelopment timing coincided with adoption of the ARO, more than seven years after the station opened. Highland Park offers an interesting case for continued observation: real estate markets near the station began strengthening about 10 years after the station opened, and any future redevelopment is likely to be shaped by the neighbourhood’s historic preservation policy.

How does zoning facilitate or hinder redevelopment?

Zoning can impede station-oriented redevelopment if it limits TOD-compatible uses, restricts height or density too much, or creates excess uncertainty in the development process. Whether zoning constitutes a binding constraint on development depends on the written rules, implementation by public officials, and political pressure from constituents. The contrasts between Civic Center and Pershing Square offer the strongest evidence of zoning constraints: the two stations are less than one-quarter mile apart, share a common real estate market, and have similar ridership levels, but differ widely in the TOD-compatibility of their zoning. The cases also illustrate the complexity and range of zoning tools around stations. Highland Park is the best example of multi-layered zoning and potential incompatibility across plans, but zoning around all five stations present non-trivial challenges for development. Complex rules and procedures may deter some potential developers, while the uncertainty and length of the process will increase development costs and result in higher prices or rents. Notably, near some stations with ostensibly TOD-friendly zoning, allowable densities for new buildings are lower than existing structures.

Our five case studies include widely varying examples of local government efforts to change zoning near stations, reflecting the City of Los Angeles’ piecemeal approach to encouraging TOD. By contrast, the neighbouring city of Pasadena adopted new zoning that increased allowable densities near all three Gold Line stations in downtown Pasadena at the time that Gold Line service began. In most cities, the public agency responsible for planning, financing, and developing transit infrastructure does not directly control land-use decisions. At most, regional transit agencies can offer incentives for local governments to alter zoning.

Beyond written land-use policies, political support from neighbourhood residents and/or within-city governing bodies matters. Los Angeles has a relatively strong city council and weak mayoral system, so support from relevant council members is critical to adopting compatible zoning or obtaining a variance. Moreover, the city’s Neighborhood Councils can mobilise residents around both zoning changes and specific development projects. Facilitated by CEQA, opposition groups may bring legal action that delay or end development and zoning changes, as in the Hollywood Community Plan Update. The Highland Park case illustrates how ambivalent community sentiment may produce plans with conflicting goals and elements.

This research also highlights some important measurement issues. In the sample neighbourhoods, both new development and zoning are extremely difficult to measure accurately. The complexity of zoning has long challenged empirical researchers; even basic questions of what uses are permitted and at what density may be difficult to ascertain. Most quantitative analyses of real estate markets use sales prices from observed transactions or new building permits. These outcomes are less relevant in dense urban areas, where the existing building stock contains commercial buildings that transact infrequently and the housing stock is largely renter-occupied. It is useful to consider subtler changes to the building stock – such as interior renovations and façade improvements – that may indicate rising property values in the short run but are not easy to track with existing data sources. Data limitations such as these enhance the utility of small-scale qualitative research in complement to larger quantitative studies.

The findings point towards several potential future research areas. First, comparing the timing and form of TOD across different economic, political and institutional contexts, including in internationally, would be useful. Second, research could examine the interactions between various policy interventions and underlying real estate market conditions (both local and macroeconomic). Third, how can planners to encourage local governments to adopt TOD-compatible zoning when they expand transit systems? Equally important, how can public officials build support for higher-density development among neighbourhood residents? Investigating these questions would enhance policymakers’ ability to maximise the return to public investments in rail transit.

Footnotes

Acknowledgements

The authors are grateful for thoughtful comments from Liz Falletta, Leslie McGranahan, Nick Saponara, and participants at the Federal Reserve System Regional Analysis Conference.

Funding

This research was supported by a grant from the John Randolph and Dora Haynes Foundation. The analysis and conclusions set forth are solely the responsibility of the authors and do not indicate concurrence by the Board of Governors or other staff in the Federal Reserve System.

Notes

References

Alonso

(1964) Location and Land Use. Cambridge, MA: Harvard University Press.

Baum-Snow

Kahn

(2005) Effects of urban rail transit expansions: Evidence from sixteen cities, 1970–2000. Brookings-Wharton Papers on Urban Affairs : 147–206.

Belzer

Autler

(2002) Transit-Oriented Development: Moving from Rhetoric to Reality. Washington, DC: Brookings Institution.

Bernick

Cervero

(1997) Transit Villages in the 21st Century. New York: McGraw-Hill.

Blount

Nakano

et al . (2014) Redevelopment Agencies in California: Benefits, Excesses and Closure. U.S. Department of Housing and Urban Development, Economic Market Analysis Working Paper EMAD-2014–01.

Boarnet

Crane

(1997) LA story: A reality check for transit-based housing. Journal of the American Planning Association 63(2): 189–204.

Bowes

Ihlanfeldt

(2001) Identifying the impacts of rail transit stations on residential property values. Journal of Urban Economics 50(1): 1–25.

Cervero

(2004) Transit Oriented Development in the United States: Experiences, Challenges and Prospects. Washington, DC: Transit Cooperative Research Program Report 102.

Chatman

Tulach

Kim

(2012) Evaluating the economic impacts of light rail by measuring home appreciation: A first look at New Jersey’s River Line. Urban Studies 49(3): 467–487.

10.

City of Los Angeles (2006) Memorandum from Richard Benbow, Acting Chief Executive Officer, Community Redevelopment Agency of the City of Los Angeles to Agency Commissioners.

11.

Clark

(2015) Highland Park Swap Mall: A Year Later. Marketplace podcast. Available at: www.marketplace.org/2015/08/20/wealth-poverty/york-fig/highland-park-swap-mall-year-later.

12.

Duncan

(2011) The impact of transit-oriented development on housing prices in San Diego, CA. Urban Studies 48(1): 101–127.

13.

Elkind

(2014) Railtown: The Fight for the Los Angeles Metro Rail and the Future of the City. Berkeley, CA: University of California Press.

14.

Ellis

(2005) Zoning for transit-oriented development in Minneapolis: Challenges and opportunities. Master’s Thesis, Hubert Humphrey Institute of Public Affairs, University of Minnesota.

15.

Fulton

(1996) The Reluctant Metropolis. Point Arena, CA: Solano Press Books.

16.

Giuliano

Agarwal

(2010) Public transit as a metropolitan growth and development strategy. Brookings Institution working paper. Available at: https://www.brookings.edu/wp-content/uploads/2016/07/urbanandregionalpolicyanditseffectsvolume3_chapter.pdf.

17.

Giuliano

Hou

Kang

et al . (2015) Accessibility, Location and Employment Center Growth. Final Report, METRANS Project 11–06. Los Angeles: METRANS Transportation Center, University of Southern California.

18.

Glaeser

Schuetz

Ward

(2006) Regulation and the Rise of Housing Price in Greater Boston. Cambridge, MA: Pioneer Institute for Public Policy Research and Rappaport Institute for Greater Boston.

19.

Greenberg

(2004) Regulations shape reality: Zoning for transit oriented development. In: Dittmar

Ohland

(eds) The New Transit Town: Best Practices in Transit-Oriented Development. Washington, DC: Island Press, pp. 57–82.

20.

Hess

Almeida

(2007) Impact of proximity to light rail rapid transit on station-area property values in Buffalo, New York. Urban Studies 44(5): 1041–1068.

21.

Immediato

(2015) Downtown LA is a retail field of dreams. Los Angeles Magazine. Available at: http://www.lamag.com/theclutch/downtown-l-a-is-a-retail-field-of-dreams/.

22.

King

Keohane

Verba

(1994) Designing Social Inquiry. Princeton, NJ: Princeton University Press.

23.

Kolko

(2011) Making the most of transit: Density, employment growth, and ridership around new stations. Public Policy Institute of California working paper. Available at: http://www.ppic.org/content/pubs/report/R_211JKR.pdf.

24.

Levine

(2006) Zoned Out: Regulations, Markets, and Choices in Transportation and Metropolitan Land Use. Washington, DC: Resources for the Future Press.

25.

Mills

(1967) An aggregative model of resource allocation in a metropolitan area. American Economic Review 57(2): 197–210.

26.

Muth

(1969) Cities and Housing. Chicago, IL: University of Chicago Press.

27.

Nelson

Weikel

(2016) Transit ridership declining despite massive investment, more rail lines. Los Angeles Times, 27 January.

28.

Olshansky

(1996) The California Environmental Quality Act and local planning. Journal of the American Planning Association 62(3): 313–330.

29.

Schuetz

(2015) Do rail stations encourage neighbourhood retail activity? Urban Studies 52(14): 2699–2723.

30.

Schuetz

Giuliano

Shin

(2015) Is Los Angeles Becoming Transit Oriented? Finance and Economics Discussion Papers 2016–004. Board of Governors of the Federal Reserve System. Available at: https://www.federalreserve.gov/econresdata/feds/2016/files/2016004pap.pdf.

31.

Suzuki

Cervero

Iuchi

(2013) Transforming Cities with Transit: Transit and Land Use Integration for Sustainable Urban Development. Washington, DC: The World Bank.

32.

Taylor

Kim

Gahbauer

(2009) The Thin Red Line: A case study of political influence on transportation planning practice. Journal of Planning Education and Research 29(2): 173–193.

33.

Wheaton

(1978) Urban spatial development with durable but replaceable capital. Journal of Urban Economics 12(1): 53–67.

34.

Wotapka

(2013) In downtown L.A., a housing revival. Wall Street Journal. Available at: https://www.wsj.com/articles/in-downtown-la-a-housing-revival-1385508551.

35.

Zahniser

(2013) Judge deals major blow to Hollywood growth plan. Los Angeles Times. Available at: http://www.latimes.com/local/lanow/la-me-ln-judge-hollywood-growth-plan-20131211-story.html.