Transit Supportive Parking Policies and Programs (2016)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

8 to use the same catchment area for providing parking and the built environment, regardless of station access. The finding also conflicts with the work of Cervero (2006), who showed that the greatest ridership payoff comes from intensifying station-area housing. It also is at odds with Willson and Menotti (2007), who show that the success, in terms of transit ridership of trad- ing parking for development is highly variable, depending on proximate factors such as the local real estate market. However, another study of BART park-and-ride programs found that to generate the same ridership as a surface parking lot, housing development must be four to five stories, assuming the lot and housing are built on a 10,000-square-foot site, or 110–150 units per acre (Wilbur Smith Associates 2011). This analysis assumes that 25% to 35% of residential trips would involve transit, and that one parking space per unit is provided. Although not spe- cific to station areas, supporting research shows that less park- ing provision on the residential side could result in even more transit ridership (Weinberger 2012). Although Duncan’s analysis outlines the system ridership advantages of park-and-ride, he notes several disadvantages including that parking lots can be dangerous and unpleasant and deter nonauto transport modes, station area land optimal for TOD is instead used for parking, and emissions are higher for shorter trips—typical of those taken to access a park-and- ride facility (Duncan 2010). Park-and-Ride Location Multiple studies discussed the ideal locations for park-and- ride lots within a regional context and suggested that they belong on the outskirts of the region as a way to intercept driv- ers. An autocentric approach in which the authors recommend placing park-and-ride lots “ahead” of the congested portion of a road or highway to give drivers the choice between using a congested road or parking and taking transit is provided in TCRP 153: Guidelines for Providing Access to Public Transportation Stations (Coffel et al. 2012). The downside is possible upstream congestion on the transit system that pre- vents people from accessing the system closer to the central business districts (CBDs) that are typically most effectively served by transit. In a 2013 study of parking at GO Transit (a regional public transit service for greater Toronto and Ham- ilton) commuter rail stations in Ontario, a similar approach on a local scale was recommended, placing park-and-rides at stations on the edges of towns or activity centers. They suggest There is an array of research on parking at or near transit sta- tions. However, it is not clear from the literature which policies support and promote the use of transit. A finding consistent with the Synthesis survey reveals that two agencies may have the opposite policy to achieve the same end; for example, some will price parking in order to boost ridership and others may provide free parking to achieve the same end. The most significant debate is over the question of whether to develop the areas surrounding the station, with one author arguing that development densities would have to be unacceptably high in most jurisdictions in order to surpass the transit rider- ship associated with park-and-ride (Duncan 2010) and others providing the counter argument that the best way to increase ridership is by developing such station areas (Willson and Menotti 2007). This review examines reports and documents on the topic that were published between 1991 and 2014. Sources were found using online databases such as TRB’s Transportation Research Information Documentation (TRID), APTA, Springer Link, Science Direct, the Research Division of the Federal Bank of St. Louis, and Google Scholar. Following the major literature trends, the review of published works is organized around these themes: 1. Park-and-ride facilities 2. Transit-oriented development (TOD) 3. Nontraditional parking and station access. PARK-AND-RIDE FACILITIES The choice to provide parking at transit stations has “con- sequences for the viability of a rail project in terms of cost, ridership, political support land use impacts, and broad sus- tainability goals” (Duncan and Christensen 2013). Despite this importance, there is a paucity of literature on park-and-ride lots in the United States and little in the way of studies that have quantified the relationship between the provision of park- ing lots and transit ridership. A study of Bay Area Rapid Transit (BART) park-and-ride facilities indicated that more than one housing unit or job must be placed adjacent to the station for every parking space that is removed—at densities that most municipalities find either politically or practically unable to accommodate. “[T]hese results suggest that the provision of station parking usually represents the most practical way to maximize ridership” (Duncan 2010). This assumption appears chapter three LITERATURE REVIEW

9 focusing on nonauto access at stations within towns, such as walking, biking, and feeder bus service. In all situations, they suggest providing a “strong evidence base” to often-skeptical stakeholders on the benefits of providing less parking and offering alternatives to driving (Engel-Yan et al. 2014). Policies for designing park-and-ride lots emphasize that a high-quality waiting environment where commuters feel safe and comfortable helps to promote the transit service. Shirgaokar and Deakin (2005) found that commuters in the San Francisco Bay Area preferred park-and-rides within a short walk of the transit station with amenities for waiting passengers, includ- ing restrooms, snack machines, and water fountains. Safety was also an issue, with commuters raising concerns about the frequency of theft and vandalism at park-and-ride lots and the desire for greater lighting and visibility (Shirgaokar and Deakin 2005). When, Where, and How Much? An overarching concern is determining whether to provide parking; weighing the tradeoffs between building park-and- ride facilities, allowing and promoting development, or pro- viding access by other modes. The goal of parking policies in the station area is typically to maximize ridership and station access. Dunphy et al. (2003) noted that “As Goldilocks might say, parking around transit must be ‘Not too much, not too little, but just right.’” The authors suggest four key principles for good parking policy: 1. Move it away from the platform to conserve the most valuable real estate, 2. Share the parking (a station in San Diego is cited for sharing its parking with a multiplex theater), 3. Deck parking and charge patrons to provide financing for the facilities, and 4. Wrap it to improve attractiveness and provide space for mixed uses. Too little parking can also be blamed as a deterrent to transit ridership. In Los Angeles, the MTA estimates that the Red Line loses as many as 1,500 riders a day because of the North Holly- wood station parking lot filling up by 7:30 a.m. (Nelson 2014). It is unclear whether this is a problem of too little parking (a very subjective idea) or perhaps an opportunity for better access management. The San Francisco Bay Area examples indicate that increased parking user fees, for the same supply, are corre- lated with system growth. This relationship could imply that an access management strategy that favors more walking access for riders who are closer to the station leaves parking spaces available for those who must travel farther to reach the station. Pricing Pricing parking spaces provides a means of influencing the use of parking facilities, influencing access mode choice, and recovering some of the construction and operating costs of the parking. All of the studies on park-and-ride policy acknowledged the possibility of charging for parking at such facilities, but did not agree on whether to charge or how much to charge. Shirgaokar and Deakin’s survey of park- and-ride users revealed that many customers were willing to pay $1 or $2 to park in a safe, secured lot, citing con- cerns about theft, vandalism, and visibility, or up to $4 for reserved parking (Shirgaokar and Deakin 2005). Shaheen et al. (2005) noted that the guaranteed availability of park- ing at a transit station, even if it costs extra, gives drivers an incentive to use transit. The complementary analysis is that it also encourages driving to transit when other access options may be available. Transit Parking 101 (APTA 2014) gives special consideration to local demographics and envi- ronmental justice, noting that there is “ongoing concern” that the implementation of parking charges can have dis- parate impacts on transit customers who can least afford them. In the San Francisco Bay Area, changes in parking fees require that BART conduct an environmental justice analysis to determine if there would be adverse impacts on low-income and minority populations. The literature notes that charging for parking may also reduce both demand for parking and ridership (although in the case of BART, pricing has not). In cases where parking is more than 90% utilized, increases in parking pricing has shown no effect on occupancy rates although it is possible that price increases will stimulate a shift in users; when park- ing is less than 90% utilized, there is a 0.33 elasticity (Coffel et al. 2012), meaning that demand is reduced. This suggests that where parking is oversubscribed; that is, there is greater than 90% occupancy, there is a considerable amount of con- sumer surplus. As a general guide, to make park-and-ride an attractive option parking fees in combination with a round trip transit fare should be less than all-day parking costs in the CBD (Coffel et al. 2012). A case example of the Regional Transportation District in Denver demonstrates that parking pricing provides rev- enue, but also reduces the number of long-term parkers and shifts demand to facilities with unused capacity. The issue of parking pricing beyond the confines of the immediate sta- tion area—such as on-street parking—is considered impor- tant to “internalize” the artificially low cost of driving and encourages mode shift to public transit. However, one sur- vey showed that most transit officials oppose or are neutral regarding parking pricing, because of a fear of losing riders (Yoh et al. 2013). Smart parking technology allows commuters to reserve spaces and pay for parking; technology facilitates the intro- duction of parking pricing policies that may significantly reduce car travel and increase transit ridership. Shaheen et al. (2005) speculates that paying for parking at BART sta- tions may be more palatable “if they feel they are getting an

10 advance benefit” and may “pay a premium for the luxury of knowing that they won’t have to circle for parking once they reach their destination.” Parking Impact on Transit Ridership Overall, the literature that documents the empirical relation- ship between parking capacity and demand for transit services is limited, although understanding the relative influence of the factors that affect transit ridership “is central to public policy debates over transportation system investments and the pricing and deployment of transit services” (Taylor and Fink 2003). In their review of the literature on the factors influencing rider- ship, the authors noted that the “relative importance of these various factors . . . is not well understood.” Merriman (1998) studied the impact of increasing park- ing capacity at commuter rail stations in the Chicago area where parking was limited. Depending on the time period, definition of constraint, and other variables, between 0.6 and 2.2 additional boardings are associated with each additional parking space. Although there is some empirical evidence that increasing parking capacity slightly reduces boardings at adjacent stations, the net overall impact is positive. Kuby et al. (2004) addressed ongoing questions regard- ing how many riders light rail transit can attract, given its resurgence in popularity and that many of today’s cities are low-density and autocentric. The presence of park-and-ride lots was studied as one of a number of factors influencing boardings. Regression analysis determined that land use and accessibility were the most significant factors in determining ridership; parking was less significant, along with other fac- tors such as the percentage of renters within walking distance, employment, and population. Cervero (2006) noted that TOD has been shown to produce an appreciable “ridership bonus” in California. Reasons for this include life-style preferences for transit-oriented living and employer-based policies that reduce free parking and auto subsidies. As noted previously, a study of nearly 1,000 resi- dents living in 26 housing projects within ½ mile of California urban rail stations showed that the greatest ridership payoff comes from intensifying station-area housing, not parking or station-area design. Although parking provisions “might influ- ence the attractiveness of station-area housing among tenants . . . [parking] appear[s] to exert minimal influence on whether station-area residents opt for transit or not” (Cervero 2006). Willson and Menotti (2007) provide some insights into the quantitative relationship between parking and ridership in a paper that addresses the tradeoff between providing commuter parking and encouraging TOD. The model explores the effect of converting one acre of surface parking to a TOD, with no replacement parking at locations in the BART system. A sur- face parking lot of one acre provides approximately 124 spaces, generating 136 daily boardings, according to the model, and the authors run several scenarios with various development densities, parking supply, and pricing alternatives. The paper concludes that “replacement parking and TOD depend on local real estate and transportation conditions” and “leaving transit agencies’ land resources in surface parking involves substantial opportunity costs” (Willson and Menotti 2007). TRANSIT-ORIENTED DEVELOPMENT POLICIES There is a significant amount of literature concerning questions of parking at TOD sites. Cervero (2006) asserts that TOD can create a balanced demand for trips throughout the day when it offers a mix of uses, such as housing, offices, and retail. This reduces the pressure on both the transit network and on parking. It also creates opportunities for shared parking on evenings and weekends. Thus, policies that make it easier to provide shared parking make TOD easier to build by reducing the capi- tal and operating costs of overbuilding parking (Cervero 2006). The literature recommends that the redevelopment of park- and-ride facilities for new construction not include one-for- one parking replacement; a finding borne out in the case example of the Lindbergh Station TOD in Atlanta. Additional parking adds both hidden costs (such as land) and direct costs (maintenance and operations) to housing and other uses in the station area, while also undermining the trip-reducing benefits of TOD (Willson 2005; Cervero et al. 2009). How- ever, previous assumptions that development would generate fewer transit fares (and thus lower farebox recovery) than a park-and-ride lot has not discouraged efforts to provide one- for-one space replacement (Willson and Menotti 2007). In “Commuter Parking Versus Transit-Oriented Development,” the authors did a fiscal study of TOD and park-and-ride lots in different settings. They discovered that transit agencies are likely to find a net fiscal benefit in projects with medium- to high-density development and partial or no parking replace- ment, provided that they were built in urban or urbanizing locations. In suburban locations, it was found that TOD often had no or negative impact on transit ridership compared with park-and-ride lots (Willson and Menotti 2007). In “Vehicle Trip Reduction Impacts of Transit-Oriented Housing” and related works, Cervero and Arrington (2008) reviewed several residential projects in TODs in California, Maryland, New Jersey, Oregon, and Pennsylvania, and found that the number of vehicle trips generated by each unit was dramatically lower than what was anticipated by the Institute of Transportation Engineers; in some cases 60% lower. They noted an inverse relationship between residential density and auto trip generation. Both Cervero and Arrington (2008) and Jacobson and Forsyth (2008) noted several variables in parking use, including proximity to the transit station, urban design of the station area, and the ability to make other trips in the surrounding area without a car such as to shopping, schools, or other amenities.

11 The land use around transit stations can also have an impact on parking demand and use. In “Office Development, Rail Transit, Community Choices,” Cervero (2006) studied ten office buildings in seven locations in California (Los Angeles, Orange County, Sacramento, San Diego, Alameda County, Contra Costa County, and Santa Clara County) to examine their parking use. The study found that workers are much more likely to commute by rail if their workplace is closer to a rail station than if their home is near a rail station. Office build- ings with high transit use were in dense, mixed-use neigh- borhoods where parking was priced at market rates, whereas office buildings with free and plentiful parking (greater than one space per worker) had lower transit ridership. This sug- gests that one parking policy that could support transit use is to reduce parking requirements (or make them more flexible) for offices near transit, which might provide incentives for office development and increase ridership. Design Standards and Policies Parking has a direct relationship to the quality of urban design of transit station areas, because parking lots and garages at street level can have a deleterious effect on the walking environment, making it less safe or comfortable for walking. One study reviewed the impacts of parking on urban design in TOD. Look- ing at seven TODs in Missouri, Oregon, and Virginia research- ers concluded that parking can undermine the benefits of TOD, and thus transit ridership, by reducing the density of station areas and increasing the distance from the station to housing, offices, shops, or other uses. However, it also noted that on-street park- ing can calm traffic on busy streets, making walking more pleas- ant, which might reinforce transit use (Jacobson and Forsyth 2008). This is supported by Shirgaokar and Deakin’s 2005 study of commuters in the San Francisco Bay Area, which showed a preference for park-and-ride lots within a short walk of the transit station and with amenities including restrooms, snack machines, and water fountains. Commuters also expressed a desire for increased lighting and visibility. Zoning Policy Several studies examined the use and availability of parking in residential and office projects in TODs. Multiple studies (Willson 2005; Cervero 2006; Willson and Menotti 2007; Cervero and Arrington 2008; Cervero et al. 2009) found that many TOD residential projects had overbuilt parking, possibly because of local zoning codes that mandated more parking, banks hesitant to lend to projects that had less parking, and developers concerned about market viability. In addition, Manville et al. (2013) found that minimum parking requirements reduce housing and population densi- ties while increasing vehicle density. This makes TOD less effective, as there are fewer residents living within the transit station area and those that do have easy access to cars, which encourages them to drive. That in turn encourages more auto- oriented uses that make driving more convenient and walking or bicycling less so, further reducing transit use. Multiple studies alluded to the importance of shared parking in reducing parking supply at TODs, particularly those with multiple uses that may be required to provide their own park- ing. In “Parking Policy for Transit-Oriented Development: Lessons for Cities, Transit Agencies, and Developers,” Willson (2005) emphasized the need for both developers and transit agencies to collaborate with municipal governments and lenders to find creative ways of reducing the parking in resi- dential projects, noting that the parking is often overbuilt. He also recommends unbundling parking costs from housing costs, which allows residents to choose whether or not to pur- chase parking and provides an incentive to ride transit instead. NONTRADITIONAL PARKING AND STATION ACCESS Carsharing Research shows that providing carsharing vehicles at or near transit stations can reduce the demand for parking in tran- sit areas while supporting a car-free or car-light lifestyle. At transit stations, carsharing programs can be managed either by transit agencies or by the municipality, with spaces provided within a park-and-ride facility or on nearby streets. One study, “Carsharing Parking Policy” (Shaheen et al. 2010), provides a rundown of carsharing policies in North America and around the world. In most countries that have carsharing, vehicles can be kept in on-street or off-street spaces, and parking for car- sharing vehicles is either provided for free or at a reduced rate. In the United States, the United Kingdom, and Australia car- sharing vehicles also have access to dedicated parking zones. Shaheen et al. (2010) also studied carsharing policies in 17 local governments, one state government, and eight pub- lic transit agencies in North America. Not surprisingly, the local governments had dedicated carsharing space on-street, whereas the public transit agencies set aside carsharing park- ing within park-and-rides. A handful of jurisdictions (Arling- ton, Virginia; Portland, Oregon; Washington, D.C.) and transit agencies [BART, Chicago Transit Authority (CTA), Washing- ton Metropolitan Area Transit Authority (WMATA)] had a cap on the number of available carsharing spaces, which might have the unintended consequence of discouraging carsharing and thus transit use (agency profiles indicate that caps are often lifted if demand warrants additional carsharing supply). However, nearly all of the municipalities and some of the transit agencies [Metropolitan Atlanta Rapid Transit Author- ity (MARTA), New Jersey Transit, Translink, and WMATA] allowed carsharing vehicles to park for free, thus creating an incentive over private car use. An intercept survey of carshare users in the San Francisco Bay Area found disagreement on whether carshare parking should be located in curbside spaces, an issue for transit

12 stations that lack park-and-ride facilities. More neighbors supported reserving some on-street parking spaces for car- sharing vehicles than for people visiting the area for work or errands (Shaheen et al. 2010). Catchment Areas and Station Access There is general agreement in the literature that providing park- and-ride facilities extends the radius of a station’s catchment area by several miles, which can have an exponential effect on the potential ridership served. According to the 2000 U.S. Cen- sus, there were roughly 100,000 Bay Area households within a ten-minute walk (one-half mile) of a BART station, whereas there were nearly 1 million households within a ten-minute drive (3.5 miles) (Duncan 2010). The idea of a larger market should be considered with the question of market penetration; thus, an “expected value” for the catchment and penetration potential could be considered as a way to consider the issue more robustly. This has significant policy implications on a number of levels; a larger catchment area means that a greater number of taxpayers who subsidize a transit agency also have access to it, which is politically beneficial even if actual transit usage is low. Duncan (2010) states this succinctly: “A large service area population with a low individual rate of transit usage may produce more riders than a small service area popu- lation with a high individual rate of usage.” Studies found that bike parking is possibly the most cost- effective way to support transit ridership, as bicycling can extend the catchment area of a transit station, similar to park- and-ride lots and fixed-route feeder bus service, but in a way that requires less land than park-and-rides and is cheaper to implement than feeder buses. Other benefits cited include pro- viding greater mobility to customers at the beginning and end of a transit trip; a benefit not available to drivers using park- and-ride lots. In a study of bicycling to BART stations, Cervero et al. (2013) found several characteristics of transit stations that had high rates of bicycling access. One was making bike park- ing available and free while charging for car parking, which creates an incentive to bike to public transit. Another was the availability of secure bike parking, such as bike stations, sturdy bike racks, and electronic lockers, which discourage theft and vandalism and reduces capacity issues with bikes on trains. As part of an overall strategy to integrate bicycles and transit, bike parking is seen as good for marketing and com- munity relations. Some transit agencies also worked with local municipalities to include bike parking in transit facil- ity construction. Few agencies surveyed collect data about bicycle parking (Schneider 2005). AREAS FOR FURTHER INVESTIGATION It can be concluded, as does Steiner et al. (2010), that “over- all, there is a shortage of literature that is able to quantify the effects of parking policies on transit performance.” Hence, the following several related areas of useful inquiry are pro- posed that are largely absent from the literature: 1. Policies relating to land-banking parking areas (or set asides) and how this relates to strategies to avoid over- building parking supply and reducing environmental impacts. 2. When to provide parking and when not to. This appears to be an especially pertinent topic given the relatively high cost of land within transit agency service areas, the high costs of providing parking, and the increasing demands for transit agencies to control costs and help meet air quality targets. 3. Related to this and also absent from the literature is a comprehensive and detailed analysis of policies toward supply and demand and the quantitative relationship between providing parking and levels of transit ridership. 4. Shared and leased parking arrangements. 5. Station typologies and their catchment areas with regard to the role of parking and parking replacement. There is a lack of compiled data on individual sta- tion areas, their placement on the transit route or line, catchment areas and land use context, and how these variables can be used to manage parking supply and regulations. The investigators did not find any research that connected station typologies, parking policy, and transit ridership. 6. Exploration of the relationship of transit fares, cost of parking, and transit ridership. 7. Parking distribution in the transit system and how this impacts transit capacity. 8. Relationship of parking price change to parking sup- ply and demand at a station level. 9. Evaluation of parking performance in relation to tran- sit performance. Some of these topics and others may be discussed and fur- ther investigated in TCRP Project H-52: Decision-Making Toolbox to Plan and Manage Park-and-Ride Facilities for Public Transportation (expected to be completed in August 2016). SUMMARY The literature review summarizes the findings and conclusions of transit agency parking policies and programs. The exist- ing literature has been categorized into three primary themes: park-and-ride facilities, TOD, and nontraditional parking and station access. The existing research is inconsistent in terms identifying parking policies that support and promote the use of transit. The literature shows, for example, that agencies may have the opposite policy to achieve the same end; some will price parking to boost ridership and others may provide free parking to boost ridership. The literature also presents differing findings on transit ridership impact when compar- ing development around a station with providing parking. One author argues that development densities would have to be unacceptably high in most jurisdictions to surpass the transit ridership associated with park-and-ride, whereas others provide the counter argument that the most effective way to increase ridership is by developing station areas.