National Academies Press: OpenBook

Effects of TOD on Housing, Parking, and Travel (2008)

Chapter: Section 1 - Literature Review

« Previous: Summary
Page 6
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 6
Page 7
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 7
Page 8
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 8
Page 9
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 9
Page 10
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 10
Page 11
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 11
Page 12
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 12
Page 13
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 13
Page 14
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 14
Page 15
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 15
Page 16
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 16
Page 17
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 17
Page 18
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 18
Page 19
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 19
Page 20
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 20
Page 21
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 21
Page 22
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 22
Page 23
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 23
Page 24
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 24
Page 25
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 25
Page 26
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 26
Page 27
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 27
Page 28
Suggested Citation:"Section 1 - Literature Review." National Academies of Sciences, Engineering, and Medicine. 2008. Effects of TOD on Housing, Parking, and Travel. Washington, DC: The National Academies Press. doi: 10.17226/14179.
×
Page 28

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.

6For the TCRP H-27A project, the panel identified a number of fundamental questions about transit ridership and TOD. For this literature review, the research team divided these questions into four general areas: 1) TOD travel characteris- tics; 2) transit system and land-use characteristics; 3) TOD ridership strategies; and 4) TOD resident/tenant characteristics. Findings related to these topic areas and specific questions follow. The existing research provides a largely complete story about transit ridership and TOD. There is significant and very detailed information about specific TOD projects in Portland, Oregon, Arlington County, Virginia (suburban Washington, D.C.), and the San Francisco Bay Area, where a significant amount of travel behavior data has been collected through resident surveys (and academic research). At the macro level, U.S. Census data also has been thoroughly analyzed to reveal differences between TOD households and other households in travel behavior and demographics. The findings are con- sistent with each other and consistent with economic and behavior studies that explain why people travel as they do. Many cities still lack detailed primary (survey) data. That said, it is reasonable to assume that transportation and eco- nomic forces that shape TOD residency and travel behavior in California, for instance, also would apply to other settings (e.g., Dallas). A lot more is known about the travel performance of TODs. Whereas the first generations of TOD focused primarily on advocacy and assisting early adopters, now there is increased measurement and understanding of TOD travel outcomes. Some key findings in this literature review include: • Between 1970 and 2000, transit ridership for work trips in- creased in TOD zones, whereas ridership declined markedly in the metro areas surrounding TODs. • TOD households are twice as likely to not own a car, and own roughly half as many cars as comparable households not living in TODs. • Among the factors that attract households to TOD, house- holds consistently place high value on neighborhood design, home prices and perceived value, and transit proximity. • Access to high quality transit is becoming increasingly important to firms trying to attract “creative class” workers in the knowledge economy. In addition to the literature on TOD, there are larger bodies of literature that address transit operations (to maximize rid- ership) and the travel impacts of development density, mixed uses, and urban design. This literature review does not describe all of those studies and focuses on research pertaining to TOD specifically. That said, some key findings from the general transit and land use literature are included, as they would not be expected to differ significantly for TODs. TOD Travel Characteristics 1. What are the travel characteristics (e.g., frequency of travel by different modes) of people who live or work in a TOD? 2. What was the travel pattern of the TOD resident prior to moving to the TOD? Key Conclusions • TOD commuters typically use transit two to five times more than other commuters in the region. TOD transit mode share can vary from 5% to nearly 50%. • Similar to findings for nonwork trips, transit share is two to five times higher, although mode shares are typically lower than commute trips (2% to 20%). • The primary reason for range is that transit use is heavily influenced by relative travel times with automobile and extensiveness of transit service, which can vary markedly across regions. As the transit network links to more job centers, educational opportunities, and cultural facilities, transit use increases. From this perspective, TOD type S E C T I O N 1 Literature Review

(e.g., suburban neighborhood versus suburban center) is less important than specific location within the region and the quality of connecting transit service. • The transit mode shares are statistically reliable, and for an existing rail system, one could reasonably infer the approx- imate transit mode share of a hypothetical new TOD by comparing it to similar TODs in the same system. • However, there is no rule of thumb or single mode share number that can be easily applied to a hypothetical new TOD along a new rail or bus system, due to widely varying local travel conditions and employment distributions. • A primary reason for higher TOD transit use is self selec- tion. Current transit users and those predisposed to use transit seek out TOD. • When work location is unchanged, often a significant per- cent (e.g., 50%) were transit users before moving to the TOD. • Among commuters with no previous transit access, transit use can increase (up to 50%). Findings The literature shows that those who live and work near transit stops patronize transit appreciably more than the typ- ical resident of a region. The most recent comprehensive study on the travel characteristics of TOD residents and workers is the 2003 study, Travel Characteristics of Transit-Oriented De- velopment in California (Lund, Cervero, Willson, 2004). In this study, ridership statistics were developed for those living at 26 residential sites near rail stations in California’s four largest metropolitan areas, as well as for a smaller sample of office workers, retail shoppers, hotel workers, and guests of projects near rail stations. Key findings about station-area residents include: • Commute mode share: From travel-diary responses, about one-quarter of the surveyed California TOD residents took transit to work. This was nearly five times higher than tran- sit’s commute-trip modal share by residents who lived in the surrounding community. This five-fold ridership bonus associated with transit-oriented living is similar to that found in a comprehensive survey of California TOD residents conducted in 1992 (Cervero, 1994). Patterns var- ied significantly across the state, with transit capture rates of nearly 50% for several Bay Area TODs, and less than 5% for some Southern California locales. About half of the working residents of all California TODs said they never take transit to work. • Frequency of travel: Across the 26 surveyed residential sites, 29% of tenants who responded to the survey indi- cated they commute by transit every workday, and another 7% reported they commute several times a week. In the case of the Pleasant Hill TOD, 49% of residents indicated they took Bay Area Rapid Transit (BART) to work every weekday. • Noncommuter mode share: Transit served, on average, 8% of nonwork trips made by surveyed station-area residents, again with considerable variation across TODs. At BART’s Pleasant Hill station for instance, transit served 15% of non- work trips compared to less than 2% for sampled projects in Long Beach and Los Angeles. The differential between transit’s modal splits for work versus nonwork trips high- lights the role that self-selection plays in shaping travel choices. Notably, people tend to move to TODs partly be- cause of the desire to rail-commute and express this pref- erence most visibly in their work-trip modal choice. • Trends: Transit’s modal share remained fairly stable over the 1993-2003 period for neighborhoods surrounding rail stations. However, since transit’s market share of trips gen- erally eroded over this 10-year period, it appears that TOD areas have weathered the secular trend toward declining transit ridership better than most settings. • Length of residency: There is some evidence that those who have lived the longest in California TODs tend to use tran- sit most often. Among those who lived in a TOD for more than a decade, the share taking transit for their “main trips” (both work and nonwork purposes) averaged 29% versus 17% among those who had lived in the TOD for less than five years. • Intervening factors: Consistent with other research on mode choice, many other factors played a critical role in influ- encing the modal choices of station-area residents. Policies that significantly affected modal choices included: free park- ing at the workplace, flex-time privileges, employer contri- butions to the cost of transit passes, and, to a less degree, land-use variables like density and street connectivity. Additional information about these intervening factors is included in subsequent sections of this literature review. Key findings about station-area office workers include: • Commute mode share: From the survey of those work- ing at 10 predominantly suburban office buildings near California rail stations, on average, around 12% traveled to work via rail transit. This is around five percentage points more than rail’s market share for TOD office workers who were surveyed in 1992 (Cervero, 1994). Modal splits varied markedly, however. For two of the 10 office projects, 25% or more of surveyed workers rail-commuted. These two projects are in downtown settings with comparatively high densities, good regional accessibility, mandatory parking charges, and within a block of the rail station. • Intervening factors: Besides proximity to rail transit, other factors that encouraged office workers to rail-commute included: availability of free parking at the workplace; 7

employer-provided transit passes; quality of the walking corridor from the rail station to the office building; and feeder bus frequency. Key findings about station-area hotel patrons and employ- ees and retail customers include: • Commute mode share: Of 111 workers surveyed at two hotels near rail transit in California, 41% traveled to work by rail transit. • Travel by hotel patrons: Transit was not used to access hotels near rail stations among the small sample of guests who were surveyed. More than half of the surveyed guests indi- cated that they used transit during their stay. • Travel by retail patrons: Of 1,259 retail patrons surveyed at three shopping facilities near rail stations in California, 13% had arrived by rail transit. Research from metropolitan Washington, D.C. also found higher transit market shares among station-area residents, attributable in part to the high levels of accessibility conferred by the Washington Metropolitan Area Transit Authority (WMATA) rail network (JHK and Associates, 1989). Over the past three decades, Arlington County has channeled new development into high-density, mixed-use projects around five closely spaced urban rail stations in the Rosslyn-Ballston Corridor, and employed a variety of techniques, including transportation demand management programs, to encourage residents to use transit. As a result, 47% of residents use modes of travel other than the automobile to get to work, and 73% arrive at rail stations on foot, providing a cost savings because neither the county nor WMATA have to provide long-term commuter parking; land parcels that were devoted to park- ing early on have all been developed. About 40,000 riders board daily at the five urban stations in the Rosslyn-Ballston Corridor. About 29,000 riders board at the four suburban stations farther out along the Orange Line; only 15% of these transit riders arrive at their stations on foot, while 58% arrive by car (Dittmar and Ohland, 2004). Dittmar and Ohland compiled 2000 Census Journey to Work data for selected TODs in three regions with high tran- sit ridership. These TODs were defined by using a half-mile radius buffer around selected transit stops. Table 1.1 shows high levels of both transit and walking at each of the stations, higher than the levels in the county as a whole. The Evanston and urban downtown stops had particularly high walking shares, indicating that many downtown residents both live and work downtown, and that transit supports this lifestyle. The walk shares in Arlington, however, were comparatively low, and the authors suggest this is due to the high number of regional jobs in the capital, and a historic neglect of the pedes- trian environment in Arlington (something that is currently being improved). Renne (2005) used similar census data to more thoroughly examine trends in travel behavior and vehicle ownership from 1970 to 2000 for households living in 103 TODs compared with averages for the 12 metropolitan regions where the TODs are located. TODs were defined by using a half-mile radius buffer around selected transit stops. While TODs may not have existed in these locations as far back as 1970 or 1980, today they are recognized as TODs and include a train station and dense housing at a minimum. Regions were classified into 8 Community Transit Share (%) Walk Share (%) Drove Alone Share (%) TOD Type Arlington County, VA 23 5 55 County Court House 37 43 Suburban Center Clarendon 34 6 6 7 8 47 Suburban Center Rosslyn 38 10 42 Suburban Center Ballston 38 42 Suburban Center San Francisco, CA 31 8 41 County Church/24th 34 38 Urban Neighborhood Embarcadero 24 44 19 Urban Neighborhood Cook County, IL 17 4 63 County LaSalle 25 37 25 Urban Downtown Chicago/Fullerton 44 8 36 Urban Neighborhood Chicago/Berwyn 38 5 42 Urban Neighborhood Evanston/Davis 19 24 42 Suburban Center Evanston/Dempster 22 14 49 Suburban Neighborhood Evanston/Main 55 22 7 Suburban Neighborhood Source: Dittmar and Ohland, 2004 Table 1.1. 2000 journey to work mode share for selected TODs.

three groups: older and redeveloping (e.g., Chicago, Illinois; New York/New Jersey), maturing heavy rail (e.g., Atlanta, Georgia; Miami, Florida; San Francisco, California; Washing- ton, D.C.), and growing regions with light rail (e.g., Portland, Oregon; San Diego, California; Los Angeles, California; Dallas, Texas; Denver, Colorado; and Salt Lake City, Utah). Renne’s results show that over the past 30 years, transit commuting has increased amongst TOD residents from 15.1% to 16.7%, while it has decreased across all regions from 19% to 7.1%. Despite the regions becoming increasingly auto-dependent for work trips, more than twice as many TOD residents used transit for commuting compared to the regional average (16.7% versus 7.1%) in 2000. Transit commuting was more than three times higher in maturing heavy rail regions, and more than twice as much in growing regions with light rail. (The data from New York/New Jersey produced unusual results, as transit ridership in suburban TODs, while robust, was outweighed by ridership in the rest of the MSA, which is very dense and metropolitan.) Table 1.2 shows detailed transit commute data from Renne’s study. From this data, Renne provides the following observations: • Maturing-heavy rail regions experienced the highest transit ridership growth and collectively have promoted TOD through development partnerships (e.g., joint de- velopment in Washington, D.C.) and supportive poli- cies. In comparison to Washington, D.C., Atlanta TODs have experienced declining transit mode share. Renne surmises this is because Washington TODs include more mixed uses and less parking, whereas Atlanta’s TODs include primarily office space surrounded by large parking lots. 9 Region Transit Share 1970 (%) Transit Share 1980 (%) Transit Share 1990 (%) Transit Share 2000 (%) % Change 1970- 2000 (%) Chicago TOD Average (n=8) 24.0 21.7 18.7 16.7 -30.0 Chicago MSA Average 22.1 16.6 13.7 11.5 -48.0 NY/NJ TOD Average (n=26) 15.7 13.1 13.6 16.4 4.0 NY/NJ MSA Average 35.5 26.7 25.4 24.9 -30.0 TOD Average 19.8 17.4 16.1 16.5 -17.0 MSA Average 28.8 21.6 19.5 18.2 -37.0 Atlanta TOD Average (n=4) 20.9 22.5 24.9 19.3 -8.0 Atlanta MSA Average 9.2 7.7 4.6 3.7 -60.0 Miami TOD Average (n=2) 0.5 2.7 5.4 6.5 1094.0 Miami MSA Average 7.1 5.0 4.4 3.9 -45.0 San Francisco TOD Average (n=18) 17.8 22.3 20.1 21.0 18.0 San Francicsco MSA Average 11.6 11.4 9.6 9.5 -18.0 Washington DC TOD Average (n=16) 19.0 27.4 32.5 30.0 58.0 Washington DC MSA Average 15.4 13.1 11.3 9.4 -39.0 TOD Average 14.6 18.8 20.7 19.2 32.0 MSA Average 10.8 9.3 7.5 6.6 -39.0 Portland TOD Average (n=5) 9.2 13.4 11.8 14.6 58.0 Portland MSA Average 5.5 7.6 5.0 5.7 3.0 San Diego TOD Average (n=6) 8.3 11.2 6.5 6.7 -19.0 San Diego MSA Average 3.7 3.4 3.5 3.4 -7.0 Los Angeles TOD Average (n=6) 6.2 11.5 10.2 8.4 37.0 Los Angeles MSA Average 4.2 5.2 4.7 4.7 11.0 Dallas TOD Average (n=6) 14.5 9.1 9.2 3.2 -78.0 Dallas MSA Average 5.2 3.5 2.3 1.8 -66.0 Denver TOD Average (n=2) 9.4 8.6 8.4 7.5 -20.0 Denver MSA Average 4.3 6.0 4.2 4.3 0.0 Salt Lake City TOD Average (n=4) 2.4 5.8 3.2 5.0 108.0 Salt Lake City MSA Average 2.2 5.0 3.1 3.0 36.0 TOD Average 9.9 8.2 7.6 -9.0 MSA Average 8.3 4.2 5.1 3.8 3.8 -9.0 Total TOD Average (n=103) 15.1 17.0 16.9 16.7 11.0 Total MSA Average (n=12) 19.0 14.1 12.0 7.1 -63.0 Source: Renne, 2005 Older and Redeveloping Regions Maturing - Heavy Rail Regions New Start - Light Rail Regions Table 1.2. Transit trends for journey to work trips for selected TODs.

• Portland also has experienced high growth in transit use, very likely due to aggressive policies to promote transit use and TOD. • Transit ridership growth also was realized in the TODs of Miami, San Francisco, Los Angeles, and Salt Lake City. • In San Diego, Dallas, and Denver, the rate of decline in transit use for TODs was greater than for the region, although transit use remains about twice as high. Since these TODs were not built until the late 1990s or after 2000, more time may be needed to fully evaluate the long term trend. Renne also compiled national work trip information for walk and bike trips as shown in Table 1.3. Key observations regarding these modes include: • TODs have about 3.5 times more walking and cycling than MSAs (11.2% in TODs versus 3.2% in regions). • Although walking and biking to work has declined nation- ally, the decline has been less pronounced in TODs. • The same cities that had the largest increases in transit ridership (Miami, San Francisco, Washington, D.C., and Portland) also had the lowest declines in walking and cycling to work. High-transit commute modal shares among station-area residents are significantly a product of self-selection: those with a lifestyle preference to ride transit consciously move to neighborhoods well-served by transit and act upon their preferences by riding frequently. A recent study by Cervero and Duncan (2002) used nested logit analysis to predict tran- sit ridership as a function of residential location choice in the San Francisco Bay Area. Around 40% of the rail commute choice was explained by residential location. Understanding how TOD residents and employees pre- viously traveled is important in sorting out the relative 10 Region Walk Share 1970 (%) Walk/Bike Share 1980 (%) Walk/Bike Share 1990 (%) Walk/Bike Share 2000 (%) % Change 1970- 2000 (%) Chicago TOD Average (n=8) 13.6 14.1 9.8 8.9 -34.0 Chicago MSA Average 9.6 7.9 5.7 3.4 -64.0 NY/NJ TOD Average (n=26) 16.9 14.3 8.6 8.2 -51.0 NY/NJ MSA Average 10.0 10.2 7.3 5.8 -42.0 TOD Average 15.2 14.2 9.2 8.6 -44.0 MSA Average 9.8 9.0 6.5 4.6 -53.0 Atlanta TOD Average (n=4) 13.1 16.1 7.9 7.4 -43.0 Atlanta MSA Average 4.4 3.2 3.1 1.4 -68.0 Miami TOD Average (n=2) 3.3 3.6 3.0 2.8 -15.0 Miami MSA Average 7.3 5.5 4.1 2.2 -70.0 San Francisco TOD Average (n=18) 19.8 19.1 14.9 16.1 -19.0 San Francicsco MSA Average 8.6 9.1 6.4 4.4 -49.0 Washington DC TOD Average (n=16) 17.3 18.3 14.9 14.2 -18.0 Washington DC MSA Average 8.4 7.0 5.4 3.2 -62.0 TOD Average 13.4 14.3 10.2 10.1 -24.0 MSA Average 7.2 6.2 4.8 2.8 -61.0 Portland TOD Average (n=5) 23.2 23.4 19.5 20.4 -12.0 Portland MSA Average 7.8 7.4 5.4 3.7 -52.0 San Diego TOD Average (n=6) 13.2 22.6 9.4 7.7 -42.0 San Diego MSA Average 9.5 9.1 6.1 4.0 -58.0 Los Angeles TOD Average (n=6) 15.2 13.5 10.7 9.5 -37.0 Los Angeles MSA Average 7.7 7.6 5.1 3.2 -58.0 Dallas TOD Average (n=6) 31.9 9.4 26.1 11.2 -65.0 Dallas MSA Average 5.8 3.4 3.2 1.6 -72.0 Denver TOD Average (n=2) 13.4 6.3 7.9 5.5 -59.0 Denver MSA Average 7.8 6.4 4.9 3.1 -60.0 Salt Lake City TOD Average (n=4) 12.9 8.0 6.9 7.1 -45.0 Salt Lake City MSA Average 6.5 5.7 4.5 2.3 -65.0 TOD Average 18.3 13.9 13.4 10.2 -44.0 MSA Average 7.5 6.6 4.8 3.0 -60.0 Total TOD Average (n=103) 17.4 15.8 12.3 11.2 -36.0 Total MSA Average (n=12) 7.8 6.9 5.1 3.2 -59.0 Source: Renne, 2005 Older and Redeveloping Regions Maturing - Heavy Rail Regions New Start - Light Rail Regions Table 1.3. Walk/bike trends for journey to work trips for selected TODs.

importance of self-selection. If most TOD residents patronized transit prior to their move, then net ridership benefits are somewhat reduced. Two California research projects throw some light on this question. The 1992 study of ridership of people living near California rail stops examined how they travel to work at their prior residence (Cervero, 1994). For those whose job location did not change, surveys showed that 56% of station-area residents rode transit to work at the pre- vious residence. Thus, TOD residency did not yield regional mobility benefits in the case of nearly half of the sample. However, impacts were not inconsequential. Among those who drove to work when they previously lived away from transit, 52% switched to transit commuting after moving within a half mile walking distance of a rail station. Similar findings have been observed in Portland, Oregon. At the Center Commons, an urban neighborhood TOD, about 56% of survey respondents currently use an alternate mode of transportation (i.e., transit, bike, walk, carpool) to get to work; about 46% use transit. Prior to moving into the TOD, about 44% used an alternate mode for work trips, and 31% used transit. In comparison, transit work-trip mode share for the city of Portland was 12.3% according to the 2000 Census. (Almost 75% of Center Commons respondents had an an- nual household income of $25,000 or less. About 78% of work trips on transit and 84% of nonwork trips on transit are by residents who make $25,000 or less per year.) For nonwork trips, 55% currently use an alternate mode of transportation, and 32% use transit. Previously, 42% used an alternate mode for nonwork trips, and 20% used transit (Switzer, 2002). At Orenco Station, a more affluent suburban neighborhood TOD, 18% of TOD commuters regularly use transit, 75% travel in single occupancy cars, and 2.7% carpool, bike, or walk (Podobnik, 2002). Sixty-nine percent of survey respondents indicated that they use transit more often than in their pre- vious neighborhood, and 25% use transit at about the same level. At The Merrick, an urban downtown TOD, 23% of resi- dents regularly commute to work or school by transit, 44% commute in a private vehicle, and 16% walk (Dill, 2005). Overall in Portland, 12% commute by transit, 76% by private vehicle, and 5% walk. The mode split for all trips at The Merrick is: 18% transit, 53% personal vehicle, and 29% walk. The Merrick residents also claim to drive less and use transit and walk more compared to where they used to live: • 45% claim to drive a lot less now; • 23% claim to drive a little less now (total of 68% drive less now); • 42% claim to use transit a lot more now; • 28% claim to use transit a little more now (total of 70% use transit more now); • 31% claim to walk a little more now; and • 16% claim to walk a lot more now (total of 47% claim to walk more now). The 2003 California survey of transit usage found a clear pattern of changes in travel behavior before and after mov- ing to a TOD. Among all residents surveyed, around 12% shifted from some form of automobile travel to transit for their main trip purposes; however, around 10% shifted from taking transit to auto travel after moving to a TOD, and 56% drove as much as when they lived away from a TOD. The change to car commuting was thought to reflect the trend toward suburban employment in automobile-oriented settings. The 2003 California study also provides longitudinal in- sights into ridership trends among TOD projects. Overall, no evidence was found that transit modal shares changed as TOD housing projects matured. In the case of several surveyed hous- ing projects near BART’s Hayward and Union City stations, the shares of commutes by transit were in the 26% to 28% range in 1992 and 2003. In a few TODs where transit’s com- mute market shares increased over time, results could reflect filtering effects: those who use rail transit may stay in place and maintain longer residences while those not using transit may be more likely to leave. In comparison to mode share, not much information about TOD trip generation rates has been captured. Because many TODs have grid-based street networks, there are more project access points than in conventional suburban proj- ects, which tends to increase the cost and complexity of trip generation studies (because more locations must be monitored). Lee (2004) reviewed and compiled TOD trip generation data from four locations, and this data is shown in Table 1.4. From the data, it is difficult to conclude how TOD trip rates compare to standard ITE trip rates, as the TOD rates generally fall between the two ITE apartment benchmarks. In Portland, Lapham (2001) found that the lower auto trip rates could only partially be explained by higher transit use; the TODs had transit mode shares of 16% in the morning peak period and 11% in the afternoon peak, compared to about 5% for the city. After including transit and pedestrian trips to analyze total trips, he still found the TOD trip rates to be lower than the ITE rates. Lapham notes that: • Few families were observed in the TODs, so smaller house- hold size may be a factor. • At suburban TODs, the AM peak period appeared to be earlier than the 7 AM to 9 AM recording period (i.e., TOD residents may travel at different times). • Some of the larger TODs may have had more internal trips that were not captured. 11

Table 1.5 shows trip rates for trips leaving The Merrick TOD in Portland, compiled by Dill (five bicycle trips were recorded and that mode is not shown). These numbers were recorded via travel diaries (not tube counters) and thus will be slightly lower than reality, as they do not include trips by visitors and The Merrick employees. However, these are likely to be a small number of trips. Assuming every resident who leaves The Merrick returns, the numbers can be doubled to approximate total trips to and from The Merrick. Thus, the daily trip generation rate is ap- proximately 5.4 total trips per apartment, and 2.8 auto trips. This is lower than the rate the MPO uses from the ITE Trip Generation book (about 6.6 total trips per apartment). Like Lapham, Dill speculates this is probably due to smaller house- hold sizes. The average number of people per apartment at The Merrick was 1.3, with 73% of the households having only one person. In contrast, in the 2001 National Household Travel Survey (NHTS), the average household size for people living in apartments was just over 1.9 persons per household, with 26% only having one person. In addition, about 40% have three or more people. Since the ITE rates are based on an average from trip counts taken at apartments all across the United States, it is likely that the average household size for the apartments measured by ITE is larger than at The Merrick. Given this likely difference in household size, the lower total trip rate seems reasonable, and highlights the fact that current ITE trip generation rates may differ significantly from actual TOD trip rates. Transit System and Land Use Influences 1. What levels of transit connectivity to desired origins and destinations are required to promote transit ridership at TODs? 2. What TOD land-use and design features (e.g., mixed land- use, traffic calming, bus bulbs, short blocks, street furniture) have had an effect on travel patterns, transit ridership, or the decision to locate in a TOD? Key Conclusions • Research shows that system extensiveness is positively correlated with transit ridership. • Extensive transit networks, worse traffic congestion (i.e., slow auto trip times), and higher parking costs work together to increase TOD transit ridership. • General consensus is that transit service headways of 10 minutes are ideal to support a transit lifestyle. • There is no single, definitive threshold for connectivity, and measures such as “track miles” and “number of transit sta- tions” on their own are not the best predictors of ridership. What matters is transit travel times relative to auto travel times. For example, an extensive but very slow transit system likely will attract few riders if highway congestion is not se- vere. Conversely, a single fast rail corridor adjacent to a highly congested auto corridor likely will attract high ridership. 12 AM Peak Hour PM Peak Hour Study Location Apartments (trips per dwelling unit) Office (trips per 1,000 sq. ft.) Apartments (trips per dwelling unit) Office (trips per 1,000 sq. ft.) Pleasant Hill BART 0.33 1.20 0.41 1.10 San Mateo 0.44 NA 0.49 0.92 Portland TODs 0.29 NA 0.38 NA Pleasanton Apartments 0.43 NA 0.47 NA ITE Apartments (use 220) 0.51 ITE Mid-Rise Apartments (use 223) 0.30 Source: Lee, 2004 Table 1.4. Selected TOD auto trip rates (total trips in and out). Per Week Per Day Per Week Per Day Total Trips 16.72 2.39 18.81 2.69 Private Vehicle 8.81 1.26 9.91 1.42 Walk 4.82 0.69 5.42 0.77 Bus 1.10 0.16 1.23 0.18 Light Rail 1.93 0.28 2.17 0.31 Transit (Bus + LRT) 3.03 0.43 3.41 0.49 Source: Dill, 2005 Trips From Merrick Per Person Trips From Merrick Per Apartment Unit Table 1.5. Trip rates by mode at The Merrick TOD.

• The systems that will generate the highest commute rider- ship will have a high percentage of regional jobs accessible by fast transit. • For work trips, proximity to rail stations is a stronger in- fluence on transit use than land use mix or quality of walk- ing environment. The most effective strategy to increase TOD ridership is to increase development densities in close proximity to transit. • Employment densities at trip ends have more influence on ridership than population densities at trip origins. It is crit- ical to locate jobs near transit in order to attract households to TODs. • Relative travel time (transit versus auto) is still more im- portant than any land use factor (density, diversity of uses, or design). • Mixed uses in TODs allow the transit service to be used for a variety of trip purposes throughout the day and week, but as a travel benefit, this is not a primary consideration for prospective TOD residents. Employment access is a pri- mary consideration. • Mixed uses (e.g., local restaurants) and urban design treat- ments (e.g., pedestrian pathways) are important for their amenity and design value in attracting residents and visitors/ customers. TOD residents highly value good neighborhood design in addition to transit access to work. Urban design and the local land use mix may influence which TOD prospective residents choose to live in. Good design also may make a TOD a more desirable location to travel to. Findings There is no absolute dividing line or tipping point for tran- sit connectivity that translates into high transit ridership. From a transit perspective, connectivity can relate to the number of origins and destinations that can be accessed, the speed of transit service, and/or the frequency of service con- necting origins and destinations. Mode choice studies of TOD residents and office workers typically show that transit travel times and their comparison to private car travel times is the strongest predictor of transit ridership. In other words, travel time differentials are a critical factor, and these differ- entials can vary greatly depending on local circumstances. Census research by Reconnecting America’s Center for Transit-Oriented Development (CTOD, 2004) provides a macro-level view of this dynamic. CTOD looked at 3,341 fixed guideway transit stations in 27 metropolitan regions. Transit zones were defined as the half-mile radius around the stations, and the 27 transit systems were categorized as small, medium, large, and extensive. Like Renne, CTOD found that commuters in transit zones were much more likely to use transit, and concludes that the size (i.e., extensiveness) and rel- ative speed of the rail transit system is a significant determinant of whether TOD households use cars or transit (Tables 1.6 and 1.7). That said, less is known about specific accessibility thresholds (e.g., number of accessible jobs, households) to support a given TOD. In TCRP Project H27, the research team noted that the highest recorded rail capture rates are in the Washington, D.C. area, and surmised this likely is related to the fact that Metrorail has the most extensive network of any recent- generation system in the country. Lund, Cervero, and Willson (2004) partly attribute higher transit mode shares for TOD residents in the Bay Area (e.g., Pleasant Hill, Alameda City) to a more extensive and mature rail system than other TOD places [e.g., Long Beach (LA), Mission Valley (San Diego)]. In that research, the authors found a significant relationship between transit ridership and an accessibility measure that divides jobs reachable by transit in 30 minutes by jobs reach- able by auto in 30 minutes. As one would expect, the more accessible a trip origin is to jobs by transit (relative to auto), the more likely the trip is to be made by transit. While regional travel models cannot predict the number of jobs or house- holds needed to support a particular TOD, they can predict reasonably well the ridership that will result from a TOD based on regional accessibility measures. Transit travel times have a strong bearing on relative acces- sibilities (by transit versus auto) and the decision to use transit. Cervero (2003) found that for non-transit users, auto travel was on average 42 minutes faster than transit (for all trip pur- poses), but for transit users, auto travel was only 23 minutes faster. This is consistent with many other studies that find Area Transit Zones Metro Area Chicago 25% 11% Washington DC 30% 9% Memphis 6% 2% Cleveland 13% 4% Denver 12% 5% Charlotte 4% 1% Los Angeles 16% 5% Source: CTOD, 2004 Transit System Size % Auto Commuters Small 72% Medium 77% Large 65% Extensive 49% Source: CTOD, 2004 Table 1.6. 2000 transit shares for work trips. Table 1.7. 2000 percent auto commuters by transit system size. 13

that slow transit travel times retard ridership growth. Riders also care a lot about service reliability. Riders have been shown to be more sensitive to unpredictable delay than predictable waiting times (Pratt, 2000, Chapter 9). TODs should be fo- cused toward transit facilities that offer clear travel speed and reliability advantages (e.g., rail lines or bus corridors with pri- ority design treatments). Numerous studies under the broader topic of transit oper- ations have been completed to understand how improved transit service (i.e., faster speeds, improved frequency, differ- ent configurations) affects transit ridership. These studies have typically been undertaken to increase transit ridership in general, although the findings are directly applicable to im- proving TOD-focused transit service and/or locating new TODs. These studies have not been exhaustively reviewed for this literature review. Rather, only some general findings are presented here. As would be expected, improved transit service levels makes transit more convenient to use and improves transit ridership. Services may be so frequent that riders don’t need schedules, and frequent service provides more flexibility regarding de- parture and arrival times. For TODs it is important to have good service levels all day. Because TODs typically have a di- verse range of land uses, they require good service frequency during both the peak and off-peak periods, to serve both work and nonwork trips. Table 1.8 gives a rough indication of rid- ership impacts due to different transit service changes, and shows that off-peak frequency improvements can improve ridership more than other strategies (the data indicate that a 10% improvement in off-peak service levels increases rider- ship by 7% on average). In Portland, for instance, TriMet has pursued a strategy of improving off-peak bus service in its most dense and mixed use (i.e., TOD-like) corridors to expand its nonwork trip mar- ket. From FY 99 to FY 03, TriMet improved service on 10 lines to “frequent service” (15 minutes or less all day, every day). On the improved lines, TriMet experienced a 9% increase in overall ridership, whereas ridership generally remained level for routes with only nominal increases in frequency. For the frequent lines, weekday ridership increased 8%, Saturday rid- ership increased 14%, and Sunday ridership increased 21%. Frequent bus service now accounts for 45% of weekly bus hours and 57% of weekly bus rides. A generally accepted service level threshold for TODs is headways of 15 minutes or less during most of the day (Dittmar and Ohland, 2004). It makes little sense to build TOD in places that receive only hourly bus service, as service is not frequent enough to make transit use convenient. Table 1.9 describes in more detail generally recommended transit ser- vice levels for different types of TODs. Other studies have focused more on the geographic aspects of transit service (e.g., system configuration) to see how rid- ership is impacted. Ewing (1995) and others have found that accessibility to regional activities has much more effect on household travel patterns than density or land use mix in the immediate area. Whereas accessibility to shopping or work- places alone is relatively less important, good access to shop- ping, services, schools, work, and other households has a strong influence on travel patterns. While Ewing’s research focused on vehicular hours of travel, the findings for TOD are clear. Even if TODs show a propensity to generate higher than average transit ridership, they should not be built in remote locations with reduced accessibility (by all modes) to a wide range of activities. Recent research on the relative performance of alternative transit configurations reveals that network orientation greatly affects the performance of rail and bus service. Based on data from the National Transit Database, Thompson and Matoff (2000) conclude that: • The best performing systems tend to be express bus-based systems oriented to strong central business districts (CBDs) in rapidly growing regions, and multi-destinational, coor- dinated bus/light rail systems in growing regions. In multi- destinational networks, a rail line is a feeder to suburban buses, just as buses are feeders to the rail line. Multidesti- national networks typically appear in two configurations: as a grid in high-density areas where frequent service on all routes can be supported and as a timed transfer network in 14 Factor Percent Change Peak Fare 0.20% Peak Frequency 0.20% Off-Peak Fare 0.58% Off-Peak Frequency 0.70% Out-of-Pocket Auto Costs 0.70% Table 1.8. Typical ridership response to one percent change in listed factor. Source: ECONorthwest, 1991. APTA, 1991. Note: Influencing factors are: preexisting service levels, geographic and demographic environment, and period of day or week. The response is greatest when prior service is less than three vehicles per hour, when upper and middle income groups are served, when a high number of short trips can be served, and the local economy is strong. In some suburban places, off-peak frequencies have achieved elasticities near 1% when the service expansion was comprehensive and carefully planned. (Pratt, 2000, Chapter 9)

lower-density places where frequent service on all routes can’t be justified. • Whereas express bus systems are more oriented to peak pe- riod commuters traveling to CBD’s, multi-destinational rail/bus networks are oriented to a broader mix of passengers and destinations. • In comparison, traditional CBD-oriented bus transit sys- tems in rapidly growing regions are in decline. In this case, individual routes, or collections of unrelated routes, cannot compete in a dispersed trip market as each route only serves origins and destinations on that single line. The implications for TOD are that ridership is likely to be maximized when TOD is located in express bus corridors linked to a healthy CBD, or located near rail corridors with robust connecting bus service. Land use variables that affect travel are frequently described as pertaining to density, diversity (i.e., mixed uses), and design - the 3 Ds. Cervero and Kockelman (1997) found that the elasticities between various measures of the 3 Ds and travel demand are generally in the 0.06 to 0.18 range, expressed in absolute terms. They conclude that the elasticities between the land use factors and travel demand are modest to moder- ate, and higher densities, diverse land uses, and pedestrian- friendly designs must co-exist if ridership benefits are to accrue. In its guidance for air quality conformance testing, FHWA notes that accessibility (i.e., the number of jobs accessible within a certain distance or time by mode) has a much stronger influence on travel than the 3 Ds, and unless density is above 7-10 dwelling units per acre, it is unlikely that the other Ds will have any effect, even in combination. (See www.fhwa.dot/ gov/environment/conformity/benefits/benefitsd/htm.) Density, or high shares of development within a 5-minute walk of a station, has generally been shown to be the strongest determinant of transit riding and walking among the land use variables. Cervero (2005) estimated the following density elas- ticities for transit ridership during the course of developing local ridership models for BART, Charlotte, North Carolina, and St. Louis, Missouri: • Charlotte Transitway TOD Scenarios: 0.192 (for persons per gross acre within a half mile of a station). • BART Extension: 0.233 (for population and employment within a half mile of a station). • St. Louis MetroLink South Extension: 0.145 (for dwelling units per gross acre within half mile of a station). While other studies have estimated much higher ridership impacts attributable to development density, these studies typically did not use control variables to hold the extraneous 15 Source: Dittmar and Ohland, 2004 TOD Type Urban Downtown Office Center Urban Entertainment Multiple Family Retail > 60 units per acre High Hub of regional system <10 minutes Urban Neighborhood Residential Retail Class B Commercial > 20 units per acre Medium access to downtown Sub regional hub 10 minutes peak 20 minutes off peak Suburban Center Office Center Urban Entertainment Multiple Family Retail > 50 units per acre High access to downtown Sub regional hub 10 minutes peak 20 minutes off peak Suburban Neighborhood Residential Neighborhood retail Local Office > 12 units per acre Medium access to suburban center Access to downtown 20 minutes peak 30 minutes off peak Neighborhood Residential Neighborhood retail > 7 units per acre Low 25-30 minutes Demand responsive Land Use Mix Minimum Housing Density Regional Connectivity Frequencies Table 1.9. TOD types with land use and transit characteristics.

factors of transit service levels, household demographics, and parking constant (e.g., prices). As a result, these factors may have influenced the results. The TCRP H-1 study, for instance, estimated a high population density elasticity of 0.59, but failed to include a measure of transit service levels. After accounting for transit service levels and other factors, Cervero re-estimated the density elasticity to be 0.192 (and the elasticity for the number of morning inbound trains was 0.59). Employment densities at destinations are more important than population densities at trip origins. Having an office or workplace near a transit stop is a strong motivator for many Americans to reside near transit and motivates people to buy into high transit-accessible neighborhoods. The end result is that having both ends of the trip within a convenient walk to and from a transit stop is key to high ridership levels. Several studies have shown that good job accessibility via transit is among the strongest predictors of whether station- area residents will take transit to work. The 1994 Cervero study of commute choice among TOD residents of Bay Area TODs found that having a workplace near a rail station strongly encouraged rail commuting. Commuting to a job in BART- served downtown San Francisco or Oakland, for example, increased the likelihood of taking transit by 35% to 60% among residents of suburban East Bay TODs. In another study of California TODs, Cervero (1994) found that four variables–employment density, employment proximity to transit, commute behavior at the worker’s previous job, and occupation–explained 92% of the mode split variation. Orig- inal research conducted by the team under TCRP H-27 for the Rosslyn-Ballston corridor of Arlington County, Virginia, showed that nodes of concentrated development along transit corridors translates into higher transit commute shares. In Arlington County, every 100,000 square feet of office and retail floorspace added from 1985 to 2002 increased average daily Metrorail boardings and alightings by nearly 50 daily board- ings and alightings. Research shows that proximity to rail stations is a stronger determinant of transit usage for work trips than land-use mix or quality of walking environment (Cervero, 1994). Concen- trating growth around rail stops often will yield high ridership dividends almost regardless of the urban design attributes of the immediate area. Still, all transit trips involve walking to some degree, thus the provision of safe, efficient, and comfortable-feeling walking corridors between transit stations and surrounding communities is an essential attribute of successful TODs. Mixed uses like housing, offices, retail shopping, and entertainment centers are important compo- nents of TOD since they produce all-day and all-week transit trips, thus exploiting available transit capacity. Studies show that the urban design features of TOD tend to have a modest influence (relative to physical proximity) on rid- ership patterns, and suggest the presence of an “indifference zone” for longer-distance work trips. That is, once work com- muters are within one-quarter mile of a rail station, factors like mixed land uses, traffic calming, pedestrian amenities, and even density seem to matter little. This is a consistent finding from studies on the ridership impacts of TOD, including the previously-cited research by Lund, Cervero, and Willson (2004). Availability, price, and convenience of parking strongly determine whether or not those working in TODs take transit. Lund, Cervero, and Willson found that the only neighborhood-design variable that explained commuting transit ridership among TOD residents was street connectiv- ity at the trip destination. Once controlling for the influences of factors like travel time and transit accessibility, no attributes of walking quality or land-use composition in the neighbor- hoods of TOD residents had a significant impact on transit mode choice. Some of the correlations with transit ridership found in that study are: • Pedestrian connectivity at trip destination: 0.37; • Sidewalks along shortest walk route: 0.16; • Street trees: 0.079; • Street lights: 0.178; and • Street furniture (benches, bus shelters): 0.137 The researchers also found that urban design variables exert a stronger influence for station area workers than for station area residents. Furthermore, within each TOD, some will value pedestrian treatments highly, while others will not be deterred by their absence if transit is nearby. Thus, resident attitudes matter considerably. That said, good urban design treatments probably make living at higher densities more attractive. Ewing and Cervero (2001) note that individual urban de- sign features seldom prove significant. Where an individual feature appears to be significant, as did striped crosswalks in one study, the causality almost is certainly confounded with other variables. In this case, painting a few stripes across the road is not likely to influence travel choices, and the number of crosswalks must have captured other unmeasured features of the built environment. Cervero (1994) concluded that for work trips, Within a quarter to a half mile radius of a station, features of the built environment (ignoring issues of safety and urban blight) matter little—as long as places are near a station, the physical characteristics of the immediate neighborhood are inconsequential. Another assessment underscores the importance of density and proximity to a station, however, more value was attached to the land-use composition of a TOD: “transit use depends primarily on local densities and secondarily on the degree of land use mixing” (Ewing and Cervero, 2001). For instance, using data on more than 15,000 households from the 1985 16

American Housing Survey, Cervero (1996) found the presence of retail shops within 300 feet of one’s residence increased the probability of transit commuting, on average, by 3%— ostensibly because transit users could pick up convenience items when heading home after work. Not all recent evidence diminishes the importance of urban design on the travel choices of TOD residents. The TCRP H-27 study found, for example, that the combination of high densities and small city block patterns significantly increased the share of station-area residents in the San Francisco Bay Area who took transit to work in 2000. In addition, auto- restraint measures, like traffic calming and car-free streets, likely have some marginal influence on ridership to the degree walking becomes safer, easier, and more enjoyable. The quality of walk and bus access to and from stations should also be considered. Although parking supplies and prices at the trip destination more strongly influence rider- ship among TOD residents than parking at the nearby rail station, the design and siting of station parking lots bears some influence on transit demand. Peripheral parking lots that do not sever pedestrian paths to nearby residential neigh- borhoods, for example, may induce transit usage, although this has not been tested empirically. Transit travel times, which tend to be short when transit enjoys high connectivity, are far stronger predictors of rail usage for TOD commuters than land-use, urban-design, and demand-management variables. Based on standardized model coefficients, the predictive power of transit travel-time vari- ables tends to be two- to three-times greater than land-use and policy-related variables, and based on modal travel time differences many travel models can predict transit ridership at TODs reasonably well. TOD land use features are more likely to affect travel behav- ior for shorter-distance, nonwork trips. To the degree that housing, offices, shops, restaurants and other activities are intermingled, people are less likely to drive and more likely to walk to nearby destinations. Similarly, while urban design is likely to only have a marginal impact on primary trips (e.g., whether and how to access work or a shopping center), it is more likely to affect secondary trips from an activity center, which can be made by car, transit, or on foot. Because of their pedestrian orientation and mix of land uses, TODs can significantly increase the number and per- cent of local trips made by walking and cycling in particular. Table 1.10 shows how the share of walk, bike, and transit trips for the Portland metropolitan region are higher in neighborhoods with TOD characteristics. Most notably, walk trips almost double when mixed uses are included in areas with good transit service. Using primary data from urban residents in the San Francisco-Oakland-San Jose MSA and San Diego County and negative binomial regressions, Chatman (2005) found that access by transit to nonwork activities increased by 22.6% for each 1,000 retail workers within a quarter mile of residences. This robust relationship was found for all of the nearly 1,000 residential households that were sampled. Adding a rail station yielded a significant further bump in ridership. For residences within a half mile of a light-rail station in San Jose or San Diego, the number of nonwork activities by transit rose an additional 6.5%. A far bigger bonus was found for high- performance regional rail services: for those living within a half mile of a BART heavy-rail or CalTrain commuter rail sta- tion, the number of nonworker activities via transit rose a re- sounding 284%. Besides retail density, pedestrian connectiv- ity increased transit’s mode share of nonwork trips. On the other hand, as walking quality increased, transit trips seemed to switch to travel by foot. Chatman’s work strongly suggests that the quality of the walking environment significantly influences travel choices for nonwork travel. Walk/bike travel to nonwork activi- ties was found to increase by 7.1% for every 1,000 retail work- ers within a half mile radius of sampled residences. These results show that the combination of intensifying retail activ- ities with good pedestrian facilities near regional rail stations can dramatically increase the use of transit for nonwork purposes. 17 Land Use Type % Auto % Walk % Transit % Bike % Other Daily VMT per Capita Good Transit & Mixed Use 58.1% 27.0% 11.5% 1.9% 1.5% 9.80 Good Transit Only 74.4% 15.2% 7.9% 1.4% 1.1% 13.28 Rest of Multnomah Co. 81.5% 9.7% 3.5% 1.6% 3.7% 17.34 Rest of Region 87.3% 6.1% 1.2% 0.8% 4.6% 21.79 Source: Metro 1994 Travel Behavior Survey Mode Share Table 1.10. Metro travel behavior survey results, all trip purposes (Portland, Oregon). VMT = vehicle miles traveled

Using 2000 data collected from more than 15,000 house- holds sampled in the San Francisco Bay Area, Gossen (2005) studied travel and sociodemographic attributes for seven distance/density categories based on households’ proximity to rails stations and ferry terminals. Regarding nonwork travel, Gossen found that transit made up these shares of nonwork trips for the following distance rings: 14.2% (up to 1/4 mile); 11.5% (1/4 to 1/2 mile); 6.1% (1/2 to 1 mile); 1.6% (> 1 mile - low- density suburbs). Gossen also found that VMT per capita in- creased with distance from rail/ferry stations in the following fashion: 19.9% (1/4 mile); 24.1% (1/4 to 1/2 mile); 29.4% (1/2 to 1 mile); 45.0% (> 1 mile - low-density suburbs). Evans and Stryker (2005) conducted research on Portland TODs to see if the presence of TOD design features is de- tectable using a travel demand model for nonwork trips. In other words, does designating a travel analysis zone (TAZ) as including TOD add explanatory power to a base travel model for non-work trips? In the Portland travel models, an urban design variable that captures the number of retail businesses, households, and street intersections within a half mile of each zone is cur- rently used to estimate nonwork trips. The variable is formu- lated so that places with a moderate mix of all three elements score higher than places with very high amounts of only one element. In a test model, the urban design variable was retained, and TAZs that contain built TOD projects were given an addi- tional code (the TODs were identified via a qualitative assess- ment by local TOD experts). Table 1.11 shows how inclusion of the TOD variable allows the model to more closely match observed mode share totals. Evans and Stryker’s results show that in centrally located and outlying TODs, walking’s share of nonwork trips is more than twice that for non-TOD areas, and that transit use is signifi- cantly higher in central TODs (7% compared to 1%) where local and connecting transit service is most robust. The results also show that the standard and urban form models capture most mode choice behavior for nonwork trips. Adding a TOD land use variable to account for the influences of unspecified factors (e.g., parking configuration, street lights) improves the model results only modestly and most noticeably for Central TOD transit use, which increased from 5% to 7%. [The urban form and TOD variables were not found to be correlated. The author also cautions against using TOD dummy variables in travel models, because 1) it is not good practice to overuse dummy variables, particularly ones that may measure a contin- uous attribute (e.g., degrees of TOD-ness) and 2) using a TOD variable requires an analyst to arbitrarily designate TODs in the base year and in future years, potentially introducing bias into the model.] 18 Area Source Walk Bike Transit Auto Actual 444 16% 50 2% 198 7% 2043 75% Standard Model 373 14% 53 2% 133 5% 2176 80% Urban Form Model 453 17% 56 2% 126 5% 2100 77% Ce nt ra l T O D TOD-Included Model 460 17% 50 2% 184 7% 2041 75% Actual 133 17% 11 1% 12 2% 626 80% Standard Model 101 13% 11 1% 14 2% 656 84% Urban Form Model 106 14% 12 1% 15 2% 649 83% O ut ly in g TO D TOD-Included Model 117 15% 11 1% 26 3% 628 80% Actual 1401 7% 217 1% 195 1% 19,388 91% Standard Model 1504 7% 214 1% 258 1% 19,225 91% Urban Form Model 1419 7% 210 1% 263 1% 19,308 91% No n- TO D TOD-Included Model 1401 7% 217 1% 195 1% 19,388 91% Actual 1978 8% 278 1% 405 2% 22,057 89% Standard Model 1978 8% 278 1% 405 2% 22,057 89% Urban Form Model 1978 8% 278 1% 405 2% 22,057 89% O ve ra ll TOD-Included Model 1978 8% 278 1% 405 2% 22,057 89% Source: Evans and Stryker, 2005. Table 1.11. Nonwork trip attractions by TOD types and travel mode (Portland, Oregon).

Mixed uses and urban design treatments can also reduce average trip distances. Evaluating shopping trips only, Handy (1993) analyzed the impacts of local accessibility on trip dis- tance and frequency, where accessibility reflected conven- ience to nearby supermarkets, drug stores, and dry cleaners nearby in small centers or stand-alone locations. In this case, accessibility was measured as a function of retail, service, and other non-industrial jobs in nearby zones (attractiveness) and off-peak travel times (impedance). The study concluded that high levels of local access are associated with shorter shopping distances, although no relationship was found for trip frequency. TOD Ridership Strategies 1. What motivates or impedes transit ridership in a TOD? 2. What strategies have been effective in increasing transit ridership at TODs? 3. What steps should transit agencies take in supporting TODs to maximize transit ridership? Key Conclusions • Factors that most influence transit ridership are station proximity, transit quality, and parking policies. • Fast, frequent, and comfortable transit service will increase ridership. • High parking charges and/or constrained parking supply also will increase ridership. • Free or low-cost parking is a major deterrent to transit ridership. • Successful strategies include: TOD transit pass programs, parking reductions, and car-sharing programs. • TOD transit programs will be similar to other transit pro- grams. Because by definition TOD residents and house- holds are the nearest to transit, TODs should be among the first locations that transit agencies implement specialized programs. • TOD (e.g., mixed uses, high densities, reduced parking) is still illegal around station areas in many cities and transit districts. • Steps transit agencies are taking to promote TOD include: reconsidering replacement parking requirements at park and rides, advocating for zoning changes with TOD entitle- ments, land assembly, joint development, and educational efforts (e.g., producing TOD guidebooks). Findings The travel fundamentals of TOD transit ridership are similar to general transit ridership. Among the variables amenable to policy change, transit service levels and prices are the strongest predictors of ridership in a TOD. Next in importance tends to be parking supplies and charges, followed by demand-management measures like employer provision of free transit passes. Least influential tends to be land-use and urban design factors. Mixed land use and high-quality urban design, however, can be important factors in drawing tenants to station areas in the first place, thus indirectly their role in shaping travel behavior in TODs can be substantial. While the factors listed above—transit service levels and park- ing management—strongly influence transit ridership, service enhancements and parking programs usually have not been introduced explicitly for the purposes of increasing ridership at TODs. In the transit planning literature, there is a large body of research on what strategies are the most effective in generat- ing increased transit ridership. The 1995 TCRP study, Transit Ridership Initiative, identified five main transit strategies to increase ridership: service adjustments; fare and pricing adaptations; market and information activities; planning orientation (community- and customer-based approaches); and, service coordination, consolidation, and market segmen- tation. It is reasonable to expect that this family of conven- tional transit ridership strategies also will be effective in gen- erating increased ridership at TODs (that study is not summarized here). Transit agencies interested in taking steps to maximize ridership at TODs would be well advised to start with these proven strategies. Among factors within the direct control of transit agencies, the provision of frequent, reliable, and comfortable transit services will induce ridership among TOD residents and workers more than anything else. Past ridership models reveal that the quality of transit services (in terms of speed and accessibility) are significant predictors of transit mode choice among station-area residents. To iden- tify the most effective transit service strategies, the key deter- minants of travel demand for a specific setting need to be known. One cannot easily generalize the findings from a few urban settings in California and the Washington, D.C. area to all parts of the country. That said, transit agencies also have shown considerable creativity in pursuing a variety of TOD-specific strategies to increase ridership at TODs. Transportation Demand Man- agement (TDM), initiating targeted pass programs, ad- dressing parking at a number of levels, car-sharing, modify- ing transit facility design, providing planning assistance, and developing TOD design guidelines are some strategies undertaken by transit agencies to maximize ridership in TODs. One of the best times to affect travel decisions and to en- courage transit use is when there is a change in home or job location. New TOD development offers a good opportunity to implement transit pass programs to attract individuals to use transit, and in general encourage others to change their transportation habits. A survey of commuters offered Eco Transit Passes through the Santa Clara Valley Transportation 19

Authority (VTA) found that after passes were given away the number of people driving a vehicle by themselves declined from 76% to 60%. It also found that transit’s mode share in- creased from 11% to 27%, while parking demand declined roughly 19% (Shoup, 1999). Portland’s TriMet initiated a TOD Pass Program in Sep- tember 1998 at four TODs in Westside suburbs in conjunc- tion with the startup of the Westside LRT project. Residents of these TODs were offered free transit passes. Among the key findings: in May 1999, 83% of Orenco Station respondents reported using transit, where only 30% of them used it prior to the Westside LRT opening. From September 1998 to May 1999, there was a 22% increase in the number of Orenco res- idents that use transit for commuting purposes. To estimate the collective impacts of increased parking charges and a new transit pass program, Bianco (2000) con- ducted a study of the Lloyd District, a TOD employment center near downtown Portland, immediately following the installation of the on-street parking meters. Programs imple- mented in the Lloyd District included: • The new on-street parking meters; • A new transit pass program (Passport); • Emergency Ride Home program; • Two new express bus routes to the Lloyd District; and • Transfer facility improvements. Survey respondents were asked to note how their commute behavior changed one year after these programs were started. For all workers, SOV mode shared declined 7%. For Passport eligible workers, SOV use went down 19%, transit use in- creased 12%, and carpools increased also. The mode share impacts were immediate and large. Twenty-five percent of re- spondents indicated that their primary reason for change was for lifestyle reasons, 22% noted the parking charges, and 19% because of Passport (other reasons included new transit avail- ability, change in car ownership, and other). Thirty-six per- cent of respondents listed Passport as their secondary reason for change. Transit agencies also have tailored car-sharing strategies for TODs. Research described later in this review shows that car ownership rates at TODs are significantly lower than average. At the same time, the need to use a car for some trips remains. Some TODs such as Buckman Heights in Portland have utilized car-sharing as a means to reduce the need for parking in the TOD while providing the option to drive if needed. Car-sharing allows individuals to have the benefits of auto use for personal trips without the hassles and cost of car ownership and reinforces transit-oriented lifestyles. Transit agencies have played an important role in advocating for and helping to set-up car sharing. Companies like Flexcar provide car sharing in communities such as Portland, Vancouver (WA), Seattle, Washington D.C., the San Francisco Bay area, Long Beach, and other Los Angeles areas (TriMet 1999). Together with density (i.e., proximity to transit) and good transit service, a major driver of TOD ridership is the provi- sion and management of parking. Market profiles of TOD residences (e.g., small households with few cars) suggest that parking-related strategies, like a relaxation of supply codes and the unbundling of parking and housing costs, could yield long-term ridership dividends. Thus, many transit agencies and local governments desire to affect the amount and price of parking provided. Numerous studies found that transit ridership increases when parking charges are implemented, and transit agencies and local governments try to affect these too. Mildner, Strath- man, and Bianco (1997) found that cities with interventionist parking policies, high parking prices and limited supply, frequent transit service, and a high probability that travelers pay to park, are most likely to have high transit mode shares. Shifting from free to cost-recovery parking (prices that reflect the full cost of providing parking facilities) typically reduces automobile commuting by 10% to 30%, particularly if im- plemented with improved travel options and other TDM strategies. (See http://www.vtpi.org/tdm/tdm26.htm. Some studies have focused on the impacts of reducing parking sup- plies, but parking supplies are generally limited where land use is intense and land costs are high. In these cases, it is common to see parking fees that correlate with land values, and the relationship between parking supply and transit demand is captive to the dominant role of parking pricing.) At The Merrick TOD, only 17% of workers commuted by private vehicle if required to pay for parking at school or work. In contrast, more than 70% of those with free parking used a private vehicle. The most recent California study found that the likelihood of transit commuting rose by nearly 70% if station-area residents enjoyed flex-time privileges and had to pay market rates for parking, compared to the scenario of no flex-time and free parking. The 1993 California study found the availability of abundant free parking to be the biggest de- terrent to transit riding among those living and working near transit (Dill, 2005). Restricted parking supplies at the workplace and employer financial assistance with transit costs also increased the odds of station-area workers opting for rail transit. Figure 1.1 from the Lund et al. study reveals the relationship. Based on the ex- periences of the typical California TOD office worker, the models showed with 25 feeder buses per day, a workplace with 50% more parking spaces than workers and no employer help with transit costs, just 9% of office workers near a California rail station likely will commute by transit. At the other extreme, for a worker leaving a station with 400 daily feeder buses and heading to a worksite where the employer provides transit-pass assistance and offers just one parking 20

space for every two workers, the likelihood the worker will commute by transit is 50%. For transit agencies involved in the development of agency owned land, the policies and procedures for encouraging TOD can have a major impact on the implementation of TOD and directly from that, TOD ridership. Park-and-ride lots often are viewed as land banking for TOD. Ohlone- Chynoweth Commons, located on the Guadalupe light-rail transit line in San Jose, is an example of transforming part of a park-and-ride into a medium density mixed-use TOD. The project’s housing, retail, and community facilities were developed on an under-used light-rail park-and-ride lot. For this project, Valley Transportation Authority (VTA) is- sued a request for proposal seeking a developer for the 7.3-acre site. The former 1,100-space park-and-ride now in- cludes: 240 park-and-ride spaces, 195 units of affordable housing, 4,400 square feet of retail, and a day care center (Parsons Brinckerhoff, 2002). One barrier to creating more TODs is that many transit agencies (WMATA in the Washington, D.C. region, the San Francisco Bay Area’s BART, MTA in Maryland and RTD in Denver, among others) have parking replacement policies that result in one-to-one replacement of park-and-ride spaces. The H-27 team estimated that replacement parking strictures affect at least one third of TOD settings. This has proven to be a major obstacle to TOD implementation on transit agency owned parking lots. With structured parking costs running on average between $10,000 and $15,000 a space ($23,000 to $25,000 a space with special features like a retail wrap), the cost of replacement parking can have a debilitating effect on the financial viability of a proposed TOD and the financial return to the transit agency. For a theoretical 5-acre resi- dential TOD project developed at 40 units per acre, the cost of replacement parking could add $30,000 to nearly $80,000 to the cost of each unit, making TOD infeasible in many places. Sometimes, transit parking has more to do with parking location than the amount of parking. There is a growing interest in designing transit parking to encourage TOD. Portland’s TriMet and DART in Dallas have moved parking at some stations away from the platform to accommodate TOD. Newly planned systems such as Phase II of the Gold Line in Los Angeles, Sound Transit in Seattle, and the Red Line in Baltimore are considering TOD early on in the location and design of stations. This balances the need for parking to generate ridership while preserving the opportunity to capture additional ridership from TODs within an interesting and attractive walk to the station. Transit agencies have served as an educator, advocate, and financial resource for local jurisdictions to advance the under- standing of TOD and facilitate the preparation and adoption of TOD plans and zoning. The presence of self-selection has clear implications for municipal land-use and zoning strategies. The desire of many households to live in a transit-accessible 21 0 0.1 0.2 0.3 0.4 0.5 0.6 Pr ob ab ili ty C ho se T ra ns it Employer help with transit costs No employer help with transit costs 0.5 parking space/worker 1 parking space/worker 1.5 parking spaces/worker 0.5 parking space/worker 1 parking space/worker 1.5 parking spaces/worker 0 25 50 75 100 125 150 175 200 225 250 275 300 325 350 375 400 Feeder Bus Frequency at Nearest Station (buses per day) Figure 1.1. Sensitivity of rail commuting to parking prices, availability of flextime work schedules, and travel time ratios via highway verses transit, based on modeling for predicting the likelihood of California station area residents commuting by rail transit in 2003 (Lund et al., 2004).

location argues for market-responsive planning and zoning. Introducing zoning and building codes consistent with lifestyle preferences of TOD residents means individuals can more easily sort themselves into transit-served settings and act upon their travel preferences. Preferential strategies, like Location Efficient Mortgages (LEM), also can make it easier for more households to sort themselves into highly transit-accessible neighborhoods. Transit agencies, in this regard, cover a seemingly ever expanding range of activities. MTA in Maryland has been investing $500,000 to $600,000 annually in TOD administra- tion and planning to create more livable places and increase ridership. MTA in Los Angeles, Sound Transit in Seattle, the RT in Sacramento, Triangle Transit in North Carolina, and TriMet in Portland are part of the growing list of transit agen- cies that have passed transit agency funds through to local governments to plan for TOD as part of developing new rail systems (Arrington, 2003). BART has active planning partnerships underway at a dozen different stations with the objective of building stronger part- nerships with local governments and to encourage ridership growth on its system. In an innovative twist on that theme, San Diego’s MTDB has a San Diego city planner assigned to work with MTDB’s planning staff as a liaison on TOD. NJ Transit, the nation’s largest state transportation system, pro- vides TOD assistance to cities through the Transit-Friendly Communities (TFC) program and the Transit Village Initia- tive. The TFC program, started in 1996, allocates roughly $100,000 per community to hire preselected consulting teams to get cities ready for serious transit village consideration. Charlotte Area Transit (CATS), together with the City of Charlotte/Mecklenburg County, has developed a 25-year regional transit/land-use plan, a joint development policy and station area plans to guide growth along centers and corridors. Metra, Chicago’s commuter rail operator, has developed strate- gies, principles and approaches to residential development in station areas targeted at communities and real estate profes- sionals. Finally, Parsons Brinckerhoff has identified nearly 100 transit agencies that have prepared TOD design guide- lines as part of a strategy to grow ridership and encourage the implementation of more TODs. TOD Resident/Tenant Characteristics 1. What are the demographic profiles of TOD residents and employers? 2. What motivates residents or employers to locate in TODs? Examples of motivators may include the quality of schools, access to jobs, housing affordability, presence of transit services, neighborhood services and amenities, and com- munity perception. Key Conclusions • The majority of TOD residents along new transit systems are childless singles or couples. • They are often younger working professionals, or older empty-nesters. There is a wide age spectrum. • They may have low, medium or high incomes; this is driven by the design and price of the specific TOD housing, and TOD developers will target/be able to predict their market. More higher incomes are being served as the United States continues to go through a robust construction phase of denser urban residential product. • TOD households typically own fewer cars because they have smaller households, and because they may forgo extra cars due to transit’s proximity. TOD households are almost twice as likely to not own any car, and own almost half the number of cars of other households. • The top three reasons households give for selecting a TOD are housing/neighborhood design, housing cost, and prox- imity to transit. Findings With an expanding inventory of built TODs to observe and learn from, there is a growing body of evidence about who is attracted to work, live, shop, and play in TODs. At the macro level, larger demographic trends washing over America with the aging of the baby boomers and the growth of the Gener- ation X’ers (24-34) are helping drive a growing demand for a more urban real estate product. New Urban News (January/ February 2003) cites the following factors as helping to drive the trend: a doubling of demand for homes within an easy walk of stores, and an increase in buyers who prefer dense, compact homes. New Urban News quotes Dowell Meyer’s research indicating that this market segment is expected to ac- count for 31% of 2000-2010 homeowner growth. In addition, the number of U.S. households with children is projected to decline. In 1990 they constituted 33.6% of households; by 2010 they will drop to 29.5% of households. These forces com- plement and reinforce the growing demand for TOD. Survey data and anecdotal case-study data offer strong in- sights into the demographic make-up of TOD residents. TODs often have large shares of childless couples, empty-nesters, Generation X’ers, and foreign immigrants (some of whom come from places with a heritage of transit-oriented living). Table 1.12 shows the demographic characteristics of TODs studied in the H-27 research. These data are consistent with other data showing that TODs attract smaller, typically child- less households. Other research about who lives in TODs reinforces these findings. A recent study of Transit Villages in New Jersey (Renne, 2003) reveals that they cater to a younger population 22

with more racial and ethnic diversity, more immigrants, more singles, and more lower-income households. AvalonBay, an apartment developer that has emphasized projects close to transit in high cost of entry markets, has learned that the prime market for its developments consists of Generation X’ers, singles, and couples with few children, as well as the over-65 market who want to sell the suburban home and move back to the city (AvalonBay, 2003). In Portland’s downtown Pearl District, where virtually all of the buildings are oriented towards transit, 6,400 units of new apartments and condominiums have been built in the past 10 years. According to school district demographers, only 25 school-age children live there, and less than 20 babies are ex- pected each year (Gragg, 2005). (In response, Portland recently adopted developer bonuses and potential tax abatements for family-size units and children’s play areas in new residential projects. In addition, the city will begin planning a neighbor- hood park for the northern end of the Pearl District with child play facilities.) Anecdotal reasons given for the lack of children include high housing costs (i.e., additional floor space for chil- dren is prohibitively expensive), a lack of outdoor play spaces and community center, and a lack of other children. At The Merrick TOD in Portland, the survey respondents were split evenly between men and women. In addition, the respondents: • Were primarily single-person households (73%); average household size was 1.3; • Ranged in age from 20 to 87 (median age is 33 years); • Have college degrees (68%) and work full time (75%); • Are childless; only one respondent indicated having a child under age 18; and • Have a wide range of household income levels, with 41% earning $50,000 or greater (Dill, 2005). The most recent California study of TOD found the fol- lowing attributes of 5,304 station-area residents residing in 26 housing projects near heavy-rail, light-rail, and commuter- rail stations (Lund, Cervero, and Willson, 2004): • Youth: The age structure of station-area residents was younger than that of the surrounding city; 62% of respon- dents were age 18 to 35. • Minorities: Because of a large affordable housing and re- development component, relatively higher shares of ethnic 23 Project Transit Mode TOD Type Demographic Snapshot The Pearl District, Portland, OR Streetcar Urban Downtown High income, retiring seniors, childless urban professionals, limited lower income units by developer agreement Mockingbird Station, Dallas, TX Light Rail Urban Neighborhood 30-45 year old professionals who can afford to own but prefer to rent The Cedars, Dallas, TX Light Rail Urban Neighborhood Lofts occupied by young professional couples and empty nesters Center Commons, Portland, OR Light Rail Urban Neighborhood Mixed income by design, 75% earn less than $25,000, seniors housing Village Green, Arlington Heights, IL Commuter Rail Suburban Center Empty nesters and childless professionals “Triangle TOD,” La Grange, IL Commuter Rail Suburban Center Over 50 empty nesters, under 30 professionals with no kids Market Square Townhomes Elmhurst, IL Commuter Rail Suburban Center Long term local residents seeking smaller easy to maintain properties in town (likely empty nesters) Addison Circle, Addison, TX Bus Suburban Center “Choice renters” singles, empty nesters, yuppies with no kids The Round, Beaverton, OR Light Rail Suburban Center Sales targeted to urban, “edgy” market (DINKS, retirees) Gaslight Common, South Orange, NJ Commuter Rail Suburban Neighborhood “Rail-based housing for childless households.” Just three school-age children live in the 200 apartments Table 1.12. Snapshot of TOD demographics from selected TCRP-H27 case studies.

minorities and non-whites were found among TOD hous- ing projects. • Office occupations: 70% of TOD employed-residents worked in office and professional occupations, which should be expected since California’s rail systems provide good and frequent radial connections to downtown white- collar districts. • Small households: TODs are more likely to have childless households; 83% of respondents lived in 1-2 person households. • New residents: TOD residents are newer to their current location than the typical resident of cities studied. The CTOD study, which looked at all built rail stations across the United States, also finds smaller households in sta- tion areas. Household size differences are more pronounced in areas with small and medium sized transit systems, com- pared to larger cities with more extensive transit systems, as shown in Table 1.13. In these latter cities (e.g., New York City), larger households are more inclined to live in smaller housing units more typically associated with TODs (attached condominiums, townhouses, apartments) due to land and housing constraints. CTOD also concluded that TOD trends towards smaller, childless households is likely to continue. Table 1.14 shows that nearly two-thirds of the total demand for housing near transit will be generated by single households and couples without children, a higher share than this group represents of the U.S. population as a whole. Households with children likely will account for only 20% of demand for housing in TODs. In addition, as shown in Table 1.15, CTOD projects that households headed by individuals age 65 or older will be dis- proportionately represented in TODs. In contrast, households in the 35 to 64 age range will be underrepresented, as these households are less likely to have a preference for TODs. Regarding the racial and immigrant status of TOD residents, Renne (2005) found the following: • Overall, in 2000 the percent of nonwhite and foreign born populations living in TODs was similar to the percent of nonwhite and foreign born residents within the larger region. • In San Francisco and Los Angeles, TODs have about 10% more nonwhites than their surrounding regions. In Miami, TODs have 18% fewer nonwhites than the MSA. • In Atlanta, San Francisco, Washington D.C., and Los Angeles, the percentage of foreign born was more than 10% higher in the TODs than the region. In Miami and Denver, the percentage of foreign born population is slightly higher in the region than the TODs. Generalizing about TOD income levels is more difficult than drawing conclusions about household size and lifestyle types. Apartment housing in older TODs often was built to serve lower income, transit dependent households, and some current TOD projects still are built to attract these house- holds. Examples of these projects are the Center Commons, 24 System Size Transit Zones Metro Area Transit Zones Metro Area Small 51% 27% 19% 40% Medium 38% 26% 31% 41% Large 38% 24% 34% 45% Extensive 34% 27% 36% 42% Source: CTOD, 2004 One Person Housholds Families of Three or More People Table 1.13. Household size by transit system size, 2000. Household Type % of Total 2025 Households Potential TOD Demand in 2025 Singles and Couples, No Children 55.5% 64.1% Other Households, No Children 12.6% 15.1% Married Couples with Children 21.8% 11.7% Single Parents, Other Households with Children 10.1% 9.1% Source: CTOD, 2004 Table 1.14. 2025 household types and projected TOD demand. Age Group % of Total 2025 Households Potential TOD Demand in 2025 15-34 22.0% 23.2% 35-64 50.4% 42.1% 65+ 27.5% 34.7% Source: CTOD, 2004 Table 1.15. 2025 age distribution of households.

Ohlone-Chynoweth Commons, and Fruitvale Transit Village, where public sector participation and funding were used to construct new, affordable TOD housing that the market would not provide otherwise. As policy makers have more consciously used TODs to shape development and increase transit ridership, the pool of prospective tenants has been expanded to include condo-living, higher-income groups that enjoy urban amenities (though they may live in suburban TODs). Thus, today’s TODs show a broad income range that reflects local land and construction costs, specialized developer niches, and local government policies (e.g., subsidies) to proactively build housing for targeted income levels. In the Portland region, for instance, downtown Pearl Dis- trict condominiums sell for more than $200 per square foot and are the most expensive housing units in the region. Orenco Station is an affluent suburban TOD where median monthly incomes range from $5,000 to $6,000. At Center Commons, however, about 75% of TOD residents’ annual incomes are less than $25,000 (this was a goal of the project). TOD income dis- parities like this exist throughout Portland and other regions. At the national level, CTOD found that the median in- comes of households in transit zones tend to be lower than those of households in larger metropolitan regions. For households with incomes between $10,000 and $60,000, the percent of households living in the region as a whole and in transit zones is similar. However, there are fewer households in transit zones than in the metro regions with incomes be- tween $60,000 and $100,000. In Houston, Tampa, and Pitts- burgh, transit zone median incomes are slightly higher than regional median incomes. Renne (2005) found higher than average TOD incomes in Chicago, Atlanta, Miami, Washington, D.C., and Dallas and suggests these cities are building more expensive and upscale TODs. Renne found that TOD zone incomes were substan- tially lower than regional averages in San Francisco and Los Angeles, and that these TODs also were the only regions to have both significantly more nonwhite and foreign born res- idents than the region. Research by Gossen (2005) suggests that in urban settings TOD residents generally have higher incomes than other households. The higher housing price premiums for TOD liv- ing could account for this. In the San Francisco Bay Area, Gossen found average incomes were higher within a quarter mile of rail stations than anywhere else in urban districts; only those living in suburban areas averaged higher incomes than TOD residents. The highest concentration of low-income households was within a half to one mile of rail stations. Regarding auto ownership, TOD residents tend to own fewer cars, and may be inclined to reduce car ownership upon moving into a TOD. Switzer (2002) found that at the Center Commons TOD, 30% of respondents owned fewer cars than they did at their previous residence, and that 37% of respon- dents did not own any car, as shown in Table 1.16. At The Merrick TOD, as shown in Table 1.17, only 8% of residents have no vehicle available, and 73% of households said moving to this place had no impact on the number of ve- hicles owned. Seventeen percent of households, however, said that they got rid of a vehicle because of the characteristics of the neighborhood. (Dill, 2005) In her recent study of Bay Area TODs in 2000, Gossen (2005) found that car ownership levels systematically fell with dis- tance from a station, consistent with other findings in the lit- erature. The average vehicles per person were: 0.5 (< 1/4 mile); 0.54 (1/4 to 1/2 mile); 0.61 (1/2 to 1 mile); 0.75 (> 1 mile - low- density suburbs). In fact, 70% of zero-vehicle households live within one mile of a Bay Area rail or ferry station. According to the 2000 Census, more than 12% of Arlington County households are without a vehicle, the highest rate in the region outside the District of Columbia. The pro- portion of carless households is even higher in Arlington County’s increasingly urban Metro corridors, approaching 20%. In several smaller communities along the Metro system across the Potomac River in Maryland, such as Takoma Park and Silver Spring (to cite two examples), there is also a high proportion of carless households: 16.2% in Takoma Park and 15.5% in Silver Spring. But in the surrounding suburbs, households without a car are a rarity. In Fairfax County, 4% are without cars. In Prince William only 3.5% are with- out cars. Arlington’s healthy proportion of households with- out cars is fueled in part by the number of singles who live in the county. According to the 2000 Census, 40% of house- holds are made up of singles (Dittmar and Ohland, 2004). Auto ownership for selected TODs is shown in Table 1.18. 25 Previously Currently Change No Car 21 36 42% One Car 60 54 -10% Two Cars 11 4 -64% Three Cars 3 -33% Five Cars 1 2 0 -100% Source: Switzer, 2002 % of Households No Car 8% One Car 75% Two Cars 14% Three Cars 3% Source: Dill, 2005 Table 1.16. Auto ownership at Center Commons TOD. Table 1.17. Auto ownership at The Merrick TOD.

In his analysis of 2000 census data, Renne (2005) found that: • TOD households own an average of 0.9 cars compared to 1.6 cars for comparable households not living in TODs. • TOD households are almost twice as likely to not own a car (18.5% versus 10.7%). • While about 66% of non-TOD households own 2 or more cars, only about 40% of TOD households own as many cars. • In TODs, about 63% of households own fewer than two cars, compared to 45% for other households. In the survey conducted for H-27, the reduction of park- ing requirements was cited as one of the most common in- centives offered by local governments to accomplish TOD. At the same time, respondents rated “allowing a reduction in parking” as only a marginally effective strategy to encourage TOD, since developers rarely use it. The policy relationship between parking supply and TOD ridership is clearly under- stood. However, a remaining challenge is to identify effective strategies to reduce parking in TODs that local governments and developers can actually embrace in the give-and-take of the real world. One of the factors that motivates residents to locate in TODs is referred to in research as self-selection. That is, those with a lifestyle predisposition for transit-oriented living con- scientiously sort themselves into apartments, townhomes, and single-family homes with an easy walk of a transit station. Being near transit and being able to regularly get around via trains and buses is important in residential location choice. High ridership rates in TODs are partly explained as a mani- festation of this lifestyle choice. In the Los Angeles Family and Neighborhood Survey ad- ministered by RAND (Sastry, et al. 2000), residents were asked an open-ended question about factors they weighed in choos- ing a neighborhood. Twenty-one percent cited transit access, more than highway access (11%). When asked: “For your personal commute to school or work, which transportation modes were important considerations in deciding where to live,” 14% cited only transit, 9% citied transit and walk/bike, and 9% cited some other combination involving transit— that is, around a third located with reference to transit com- muting. Auto access alone was cited by just 12%. In his 2005 doctoral dissertation based on a survey of resi- dents in the San Francisco Bay Area and San Diego County, Chatman found 74.4% of people living within half a mile of a sampled California rail station sought transit access when making a residential location choice. Furthermore, those seeking transit access to shops or services live an average of 1.8 miles closer to a rail stop. However, proximity to transit for nonwork activities is likely a minor factor in residential location choices. Ben-Akiva and Bowman (1998) simultane- ously modeled residential location choice and activity/travel schedules using a nested logit method, finding little relation- ship between nonwork accessibility and the choice of resi- dential neighborhood. Weighing the collective evidence, Chatman (2005, p. 150) concluded that “auto-oriented self selection does not appear to be particularly important in out- of-home nonwork activity participation, but transit self- selection does play a limited role.” The most recent California study (Lund, et al., 2004) found that proximity to transit was ranked third among factors in- fluencing households to move into TODs, behind the cost and quality of housing. The higher density housing found in 26 Community Cars/ Household TOD Type Arlington County, VA 1.4 County Court House 1.1 Suburban Center Clarendon 1.3 Suburban Center Rosslyn 1.1 Suburban Center Ballston 1.2 Suburban Center San Francisco, CA 1.1 County Church/24th 1.1 Urban Neighborhood Embarcadero 0.5 Urban Neighborhood Cook County, IL 1.4 County LaSalle 0.7 Urban Downtown Chicago/Fullerton 1.1 Urban Neighborhood Chicago/Berwyn 0.7 Urban Neighborhood Evanston/Davis 1 Suburban Center Evanston/Dempster 1.2 Suburban Neighborhood Evanston/Main 1.3 Suburban Neighborhood Source: Dittmar and Ohland, 2004 Table 1.18. 2000 auto ownership for selected TODs.

TODs tends to keep housing prices more affordable. While land prices are higher per square foot, this is more than offset by the smaller total area of dwelling units that are purchased or leased. The California survey found that proximity to tran- sit was most important among residents who had lived in the TOD the longest. This suggests those who self-select into rail- served neighborhoods tend to stay in place. The higher pre- mium they place on proximity to transit is reflected not only in survey responses but also ridership statistics. Because most TOD residents have no children, quality of schools was not a major factor in moving into TOD neighborhoods: fewer than one of 20 surveyed respondents identified this as a top three factor in influencing their residential location choice. In his survey of Center Commons residents, Switzer (2002) found that the most common reasons given for moving into the project were: new product/appealing design (20%), proximity to transit (17%), price (16%; the project includes a significant affordable housing component), and general location (15%). Other data, shown in Table 1.19, from Portland (Orenco Station) shows a similar pattern; while transit proximity can be an important factor in attracting TOD residents, the de- sign of the housing units and larger community may be more important. The Merrick TOD residents listed the following top 10 fac- tors they considered when selecting their current home: 1. High quality living unit; 2. Easy access to downtown; 3. Good public transit service; 4. Relatively new living unit; 5. Affordable living unit; 6. Close to where I worked; 7. Shopping areas within walking distance; 8. High level of upkeep in neighborhood; 9. Attractive appearance of neighborhood; and 10. Safe neighborhood for walking (Dill, 2005). These studies show that good transit access is a primary fac- tor in residential location decisions, consistent with studies that find high rates of self-selection among TOD residents. Other features that consistently rate as being important are the qual- ity of the housing and community design, and housing cost. In addition, suburban TOD residents often value local services and amenities (e.g., in mixed use buildings, or a TOD center), while households in more urban TODs value proximity to the full range of land uses and activities that cities offer. Not sur- prising, school quality does not even register among TOD households, as few TOD households have children. For projects incorporating affordable housing into a TOD, experience indicates that affordability often outweighs any transit considerations in making locational decisions. In mar- kets like Portland (e.g., Center Commons) and San Jose (e.g., Ohlone-Chynoweth) where there are shortages of new, well- designed affordable projects, affordability is a prime attractor to TOD (according to TOD project managers). The most important considerations for all retail develop- ments are location, market, and design; proximity to transit is not a prime consideration, and the market must be viable even in transit’s absence [Urban Land Institute (ULI), 2003]. Although a retail component may eventually become an excitement generator within a TOD, it cannot be the justifi- cation for the development. According to ULI, “Retail is the one land use that is least likely to succeed where it lacks strong support. Thus retail does not drive development around transit; it ‘follows rooftops’.” TOD plans should carefully consider the volumes that retail developers require, as the rules specifying the distance that customers will travel to any particular store are inflexible. High density offices and residences can be good sources of transit riders, but they do not always ensure retail demand, particularly if local retail demand already is being met. According to CTOD, which tracks national demand for TOD, firms and workers are increasingly exhibiting a pref- erence for 24-hour neighborhoods. In the past companies preferred suburban campus environments near freeways, and regions lured employers without regard to bigger picture development goals. Now other issues are coming into play, including the rise of the creative class and the increasing im- portance of technology and talent in a region’s economic de- velopment strategy. Because firms are chasing talent, which is choosing to locate in diverse, lively urban regions, firms now prefer these locations. According to a recent Jones Lang LaSalle survey (CTOD, 2005), access to transit is very impor- tant to 70% of new economy companies. 27 Feature Percent Design of Community 13.28% Greenspaces/Parks 12.24% Community Orientation 10.94% Town Center 10.42% Alley Parking/Garage Design 9.11% Design of Homes 8.33% Pedestrian Friendly 6.25% Close to Mass Transit 4.95% Small Lots/Yards 4.95% Quiet Community 3.13% Clubhouse/Pool 2.86% Safety 2.80% General Location 2.08% Close to Work 1.30% Other 7.55% Source: Podobnik, 2002 Table 1.19. Best aspects/things liked about Orenco Station.

In the Portland area, TOD is becoming increasingly inte- grated with the high tech sector. Orenco Station in Hillsboro is located very near to Intel (not part of the project), and a large share of Orenco residents are Intel employees that ben- efit from short commutes. Open Source Development Labs, a global consortium of leading technology companies dedi- cated to promoting the Linux operating system, located to The Round in Beaverton, in part to capitalize on rail access to downtown Portland and the airport. Just down the line in Hillsboro, Yahoo Inc. recently leased space right along the rail line, citing a mix of factors including: access to public transit, daycare options, affordable housing, and quality of life. ULI (2003) reiterates that if companies see transit as slow, unreliable, or not reaching enough of their workers, staff in charge of locations decisions will not pay attention to transit. When transit is viewed as a tool for recruiting scarce talent, however, companies will list good transit access as a criterion in site selection choices. ULI also notes that more companies indeed seem to be focusing on transit access for their em- ployees, even if management does not intend to use transit. Table 1.20 summarizes ULI’s perception of broad office loca- tion trends. From the perspective of the prospective TOD developer, the development process typically begins with an idea, either a site looking for a use, or a use looking for a site. A developer usually will initiate a TOD project based on experience with similar projects, but a TOD development also could be a nat- ural evolution for a developer with a background in urban or infill projects. Market analysis for such a project, as with all developments, will consider who will buy or rent in such a de- velopment at what costs. Land cost sets the broad parameters of the project, with an understanding of the development costs for such a project, and any special construction or as- sessment costs, such as participation with the transit agency in associated facilities. Once broad parameters of project costs have been estab- lished, often with some form of option to hold the property while further feasibility is examined, the developer initiates increasingly detailed studies of market, design, and finance. Market studies will examine not only potential clients, but also competitive projects, both supply and demand. The mar- ket analysis for a proposed multi-family residential project, for example, would compare rents for similar projects targeted to a similar clientele, including projects currently under con- struction. The analyst’s challenge is to estimate the market share of household growth that would select that project, as well as the likely absorption rate of the houses (how long it would take for the houses on the market to sell, be leased, or rented). The developer’s challenge is to make sure the esti- mated market share at the proposed price point is sufficient to ensure the project’s success. As indicated in H-27, there is growing experience with TOD projects in the development community (developers, market analysts, architects, transportation consultants, and lenders) and a growing base of information used to support the development process and understand the prospective clientele, residents, and businesses. In Washington, D.C., while there have not been statistically rigorous studies of the impact of transit access on property values, market studies es- tablish the premium for rental properties at about 7%. This means that for a project well served by transit, rents can be 7% higher than comparable properties not so well located. It also would mean, for example, that if a developer were able to offer the same rents, such a project would have enormous competitive appeal. The expanding portfolio of TOD projects is providing greater insight into TOD market advantages, as well as de- mographic and lifestyle characteristics of residents. These findings are useful not only in the product development phase, but also in marketing the product. There is growing awareness among developers that an important submarket of people are attracted to TOD projects, greater understanding of who the people are, and why they are attracted to TODs. 28 "Out" "In" Suburban/exurban campus locations Locations close to transit Corporate campuses Mixed-use developments Kiss and ride Live, work, play, and ride Location near CEO's home Location convenient for workers Free parking Free transit passes Driving to lunch Walking to lunch Errands on the way home Errands at lunchtime Commuting car Fuel-efficient station car Quality of the workplace Quality of life Source: ULI, 2003 Table 1.20. Workplace culture: what’s out and what’s in.

Next: Section 2 - Does TOD Housing Reduce Automobile Trips? »
Effects of TOD on Housing, Parking, and Travel Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB’s Transit Cooperative Research Program (TCRP) Report 128: Effects of TOD on Housing, Parking, and Travel explores the demographics of transit-oriented development (TOD) residents and employers, and their motives for locating in TODs. The report also examines the travel characteristics of residence before and after moving to a TOD and ways to increase transit ridership among these residents. In addition, the report reviews the potential effect of land-use and design features on travel patterns, transit ridership, and the decision to locate in a TOD.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!