Does Bus Rapid Transit Reduce Driving?

17

1

Source: photo by Lucas Conwell

Introduction

Bus rapid transit (BRT) systems have received ample attention in recent years for their ability to offer travel time reductions and mobility benefits comparable to those of rail transit for less than one third of the cost.  One major argument for investing in BRT lines, which incorporate subway-like features such as separate rights of way, long articulated buses, and stations with pre-boarding fare payment (Gilbert 2008; Hidalgo and Gutierrez 2013), holds that they overcome the stigma often associated with bus use and therefore attract new transit riders who formerly drove, thereby reducing congestion and greenhouse gas emissions.  This study tests the ability of BRT upgrades to convince drivers to switch to transit by conducting an intercept survey of riders on the VIVA Yellow BRT line in Newmarket, Ontario, which opened about 5 months before the study.

Literature

Previous studies in urban planning come to a variety of conclusions as to the ability of BRT to cause mode shift.  One set of papers uses econometric mode choice models, which estimate a regression to explain the probability of choosing each mode of transportation using demographic, built environment, and modal characteristics, to evaluate the impact of transit upgrades.  This branch of research finds, among other conclusions, that riders of the new TransMilenio BRT in Bogota, Colombia perceive time spent riding the new BRT as less costly, and thus will stomach longer waiting and walking times in order to use the system, even if additional transfers might actually make the trip longer than on the old bus system (Lleras 2003).  Additionally, TransMilenio succeeded in reducing car ownership among lowincome households in dense neighborhoods with high levels of mixed-use development (Combs and Rodriguez 2014).

Similarly, a study using panel survey data and discrete mode choice models finds that commuters whose bus travel time was reduced by at least 10 minutes by the opening of the Fastway BRT outside London were significantly more likely to take the bus (Chatterjee 2010).  Simulations as part of a 2011 study of the existing VIVA system based on a discrete choice model in fact found that the effect of transit improvements on commute mode share outweighs the effect of congestion by up to a factor of nine (Forsey et al. 2013).

However, this conclusion is far from unanimous in the literature.  Discrete choice modeling using panel household travel survey data from Chennai, India shows that the accessibility of transit positively influences the probability of transit use only for workers who relocated into the catchment area of transit stops (Srinivasan and Bhargavi 2007).  According to Cao and Schoner (2014), existing residents of the new Hiawatha light rail corridor in Minneapolis, MN increased their transit use less than commonly assumed; furthermore, commuters who relocated into the vicinity of stations rarely take light rail, possibly because of the lack of pedestrian infrastructure around stations.

Finally, a panel survey of Santiago residents both before and after the implementation of the new Transantiago bus and metro system in Santiago, Chile, found that most new low-income metro riders are former bus riders; in addition, the percentage of high income residents commuting using cars and other “private modes” actually rose after the system’s opening (Yáñez, Mansilla, and Dios Ortúzar 2010).   Thus, while many evaluations of transit improvements do find appreciable impacts on travel behavior, these effects clearly merit further investigation.

VIVA Yellow

Figure 1: Newmarket, Ontario

2

Source: “Toronto Cleaning Services – Service Area.”

The VIVA Yellow line is the second stage of the VivaNext project, which will eventually be extended into an extensive bus rapid transit network in the York Region in suburban Toronto.  The VIVA Yellow line runs east-west along Davis Drive, a main suburban arterial in Newmarket, an outer ring suburb a full 30 miles north of downtown Toronto.

The line begins at the Newmarket Bus Terminal at the western end of the municipality, which offers bus connections to other local VIVA and York Region Transit lines as well as GO Transit lines to Toronto and other regional destinations.  From Yonge Street (just east of the Newmarket Terminal) to Roxborough

Road (just east of the Southlake stop), 1.6 miles to the east, VIVA Yellow runs in a dedicated “Rapidway.”  The line then runs another 1.43 miles in mixed traffic to a park and ride lot at Highway 404, a major regional transportation artery (“Davis Drive/Newmarket”; “Presentation Boards”).

Figure 2: Viva Yellow bus rapid transit map

3

Source: “Maps by municipality: Newmarket-Aurora.”

The Rapidway, which opened along with the rebranded line on Nov. 29, 2015 (Novakovic 2015) incorporates dedicated red-painted bus lanes, pedestrian and streetscape improvements, as well as elaborate stations.  These three “Vivastations” feature futuristic glass canopies, heated waiting areas, real-time arrival information screens, and fare machines to allow pre-payment (“Presentation Boards”).  Additionally buses receive transit signal priority at intersections (“Viva BRT service opens in Newmarket, Ont.”).

Figure 3: Southlake Vivastation

4

Source: photo by Lucas Conwell

Figure 4: Off-board fare pre-payment at Southlake

5

Source: photo by Lucas Conwell

Even the more simple curbside stops outside of the Rapidway at Huron Heights, Leslie, and Hwy 404 feature modern glass shelters with fare machines and real-time information screens.

Figure 5. Leslie curbside stop

6

Source: photo by Lucas Conwell

VIVA Yellow uses brand new vehicles shorter than the articulated buses used on other VIVA routes; these vehicles are fully adequate to cover the present levels of ridership.

Figure 6: VIVA Yellow bus

7

Source: photo by Lucas Conwell

VIVA Yellow is highly unusual among BRT systems in that it runs through a low density, suburban, historically car-dependent area (Novakovic 2015); thus, finding that VIVA Yellow has done little to coax drivers out of their cars does not necessarily discount the effectiveness of BRT in more urban settings.  On the other hand, its location makes the project a fascinating test of the impact of high quality transit on suburban travel.

Methodology

Originally, our two person survey team planned to follow a random sampling plan of VIVA Yellow stops, surveying passengers at a randomly selected station for one hour at a time from 2-6pm on Friday April 1st and 1-4pm on Saturday, April 2nd.  Due to a lack of transit agency permission, we were unable to conduct surveys on board buses, which could have produced a higher response rate.  Passenger numbers on Friday and Saturday were unexpectedly low, so low that we made the decision to abandon the random sampling plan in favor of focusing on stops that seemed to have the most passenger activity.  Many of the stops selected in our random sampling plan were virtually devoid of any boardings or alightings, so following the plan would have depressed sample size so far as to make the results almost meaningless.

Initially, we planned to survey both boarding and alighting passengers, but alighting passengers proved extremely averse to our requests, so we instead focused on surveying passengers waiting to board the bus.  Passenger numbers, even at the most frequented stops, were low enough that we did not have to use an “every n-th passenger” sampling method; instead, we asked every passenger waiting at the stop whether he or she would be willing to fill out a survey to help improve bus service.  Even among these passengers – who would be waiting at the bus stop regardless of whether they chose to participate – response rates varied widely from stop to stop, from around 25% to 75%.

Not surprisingly, ridership was higher on Friday, and the distribution of the 29 surveys collected reflects this disparity.

8

The Leslie stop not only had reasonably high numbers of passengers, but these passengers also proved especially willing to participate, perhaps due to a type of snowball effect.  Main, Southlake, and Newmarket also provided sizable numbers of survey respondents; the response rate at Newmarket was lower than its ridership numbers would suggest because the waiting bus seemed to encourage passengers to hurry on board even when overhead signs indicated that the bus would not leave for several minutes.

9

Results

Demographics

The educational distribution of passengers reveals that our sample does not only consist of low-income riders; a full 38% possess either a Bachelor’s or graduate/professional degree.

10

Correspondingly, our sample contains a surprisingly high number of car owners, given the usual demographic profile of suburban transit riders in North America.  In fact, almost half own a car and so are, at least to some extent, riding VIVA Yellow by choice.

11

More than three-fourths of respondents were born in Canada; of those who selected “other,” only one reported the specific country.

12

The sample is also relatively young, with a median age of 27.  Two respondents declined to report their age.

13

Trip characteristics

Respondents were also asked to report the primary purpose of their trip; while we intended for them to select only one response, many selected multiple.  Thus, we report the shares of trips for which the respondent selected each purpose.  Around half of trips were made to commute to or from work, and almost as many involved errands or shopping.  Just under a third were social or recreational trips, and a fourth were to or from school.

14

Almost two thirds of respondents walked to the stop where they boarded VIVA Yellow; 31% transferred from another bus to VIVA Yellow, while just under 10% were dropped off at the stop by car.  Fewer respondents – just under half – walked from the stop where they left the VIVA Yellow bus to their final destination; virtually all others planned to transfer from VIVA Yellow to another bus, either VIVA, York Region Transit, or GO Transit.

15

Impact of BRT on mode shift

Despite the ambitious investments made in the VIVA Yellow BRT, the overwhelming majority of riders most frequently used the bus to make their current trip prior to November 2015, when VIVA Yellow opened.  Once again, we intended for survey respondents to choose only one mode they used “most frequently,” but a minority selected multiple.  We report the exact responses below.

Beyond the 83% who selected “bus” when asked which mode they used most frequently to make this trip prior to Nov. 2015, 7% selected only “walk,” and 3% reported driving alone or with family (as distinct from carpooling).[1]  One respondent reported taking the bus and driving alone prior to Nov. 2015, and one further respondent selected walking, biking, and driving alone.  Perhaps these respondents intended to indicate that they sometimes used each mode selected, or they actually drove to the bus stop, parked, and then took the bus.

16

In any case, 90% of respondents reported walking or taking the bus before the opening of VIVA Yellow to make the same trip they were currently making.  Thus, only 10% of surveyed VIVA Yellow riders might have driven instead of taking transit before the BRT improvements, and most of these so-called drivers seem to also have used other modes at least some of the time.  Based on this admittedly small sample, the VIVA Yellow BRT upgrades seem to have played only a minimal role in shifting drivers onto transit.

Conclusion

Our survey of 29 VIVA Yellow riders suggests that fewer than 10% of riders would have driven before the completion of BRT upgrades; these results suggest that BRT, at least in the suburban context of Newmarket, has little potential to alter drivers’ commuting behavior, take cars off the road, or reduce transportation emissions.

Certainly, this simple survey did not account for other factors, such as rising incomes, which might have counteracted the effect of BRT, but it seems unlikely that any such confounding factors would have changed enough in the span of five months to alter aggregate commuting patterns.  More importantly, a sample of 29 riders is certainly much smaller than optimal.  However, the share of former bus riders is so high that it seems unlikely that a larger sample would include a substantial number of former drivers.

Accepting that VIVA Yellow did not have a significant impact on area residents’ mode choices does not necessarily mean that BRT in more urban contexts could not do better.  Indeed, VIVA Yellow runs along a suburban arterial where the car-oriented layout of businesses and streetscapes as well as the low residential density create a hostile environment for transit.  Building a similarly high quality line in Toronto or one of its more densely populated inner ring suburbs could yield vastly different results.  VIVA planners clearly recognize that transit crucially depends on conducive land uses, as plans to encourage urban growth and density along Davis Drive reflect (“Presentation Boards”).  In any case, improving and speeding the trips of those who choose or have no choice but to take the bus is a worthy goal in and of itself.

References

Survey instrument

18

[1] Three respondents selected “bus” and “walk;” we include these with the bus-only responses, as we assume that they meant that they walked to the bus stop and then took the bus.  Thus, bus would still be the primary mode.


My name is Lucas Conwell, and I’m a senior at Tufts University majoring in Quantitative Economics and minoring in Urban Studies.  My focus is on transportation economics and planning, and I’ve had the opportunity pursue my passion for sustainable transportation through internships at German Railways (DB), the Massachusetts Smart Growth Alliance, and most recently, the Volpe National Transportation Systems Center in Boston.

Tags: , , ,

10 Comments on “Does Bus Rapid Transit Reduce Driving?”

  1. Max TC April 28, 2016 at 1:02 pm #

    Great article and study! I’m interested in what your thoughts are on transit route length and termination points in a suburban context. Based on the experiences I’ve had navigating Connecticut’s current transit system, it seems to me that BRT (or even BRT “Lite” – essentially frequent bus service with a few benefits) has tremendous potential in suburban locations along low to moderate density corridors. The more successful examples I have seen, however, are longer (anywhere between 8-22 miles in length) and connect at least a few major centers of population and employment.

    Like

  2. Lucas Conwell April 30, 2016 at 10:57 pm #

    I think that the lack of major destinations along the VIVA Yellow route definitely contribute to its low ridership and inability to, at least so far, draw former drivers – there’s a mall at the Newmarket Terminal end that isn’t connected to the Terminal at all, the other end is a park and ride, and in between is pretty much your typical suburban arterial. Even if there were more major employers or destinations along the route, it seems to me that the lack of residential density along it and other VIVA routes still doom it to low ridership, since the riders accessing the major destinations would still have to come from somewhere. Jarrett Walker also seems to be pretty pessimistic about suburban service (http://humantransit.org/2012/05/does-suburban-local-service-get-cars-off-the-road.html), but at the same time, I’ve heard that Ottawa has used BRT very successfully in its suburbs to funnel people into downtown.

    Like

  3. urbphile August 10, 2016 at 8:34 pm #

    http://www1.nyc.gov/site/planning/about/department.page

    City planning lies at the nexus of many disciplines. It blends theoretical insights from the social sciences and natural sciences with urban, planning, and design concepts. Through this merger of ideas, it engages communities to bring about healthier, more inclusive, more vibrant places. To fulfill its mission, DCP draws on a variety of geographic, functional and technical talent to:

    Work with neighborhoods and government agencies to develop sound ground-up frameworks for growth that align strategic planning priorities with individual community needs

    Capitalize on the City’s competitive advantages, including its built environment, natural setting, diversity and standing as a robust center of commerce and culture, and advance regulatory changes when appropriate

    Encourage high-quality, sustainable development that respects and enhances its context

    Promote neighborhood economic development

    Maintain the Zoning Resolution text and maps that provide the policy and regulatory basis for land use and development, as mandated by the State and local laws

    Advise government agencies and the public by providing policy analysis, technical assistance and data on housing, zoning, urban design, community facilities, transportation, demography, waterfront/public/open space data to inform strategic and capital planning decisions

    Maintain an online Community Portal so that the public has easy access at all times to land use, housing and population data, as well as other planning resources

    Serve as technical support to the City Planning Commission in its annual review of approximately 450 land use applications and ensure that processes, reviews and approvals are efficient, consistent, and user-friendly

    Like

  4. Lux August 31, 2016 at 5:27 pm #

    cement v asphalt
    https://answers.yahoo.com/question/index?qid=20070621062540AAs6MvT

    And at bus stops, sometimes they’re cement since they are heavy. And sometimes there are bulbs and ADA ramps, hopefully. solar panels, etc.

    http://www.worldstandards.eu/cars/driving-on-the-left/

    Like

  5. Henry Hungry September 3, 2016 at 1:48 pm #

    all depends on the type of road engineering
    good, big signage, flexible, good communication…

    Like

  6. Bart October 26, 2016 at 6:16 pm #

    Frequency and speed matter the most. Reliability too.

    Like

    • Bart October 26, 2016 at 7:32 pm #

      also it is also just circuits and electricity. not too hard

      Like

  7. Rapido December 7, 2016 at 11:38 am #

    Like

  8. SAS! January 5, 2017 at 5:14 pm #

    look at the driving here — new cable-stayed bridges everywhere, LGA construction…
    better engineering, machines, materials, economical for that length/volume/weight


    versus this bus lane in harlem

    right next to 125th metro north


    Like

  9. Abraxas January 29, 2017 at 10:13 pm #

    How many riders take this system?
    Clearly, the more investment, usually the more ridership…

    For instance:

    The Metropolitan Transportation Authority has announced that Long Island Rail Road and Metro-North Railroad are seeing record ridership numbers, with LIRR carrying 89.3 million customers in 2016, a 1.9% increase over last year and the highest ridership since 1949. Metro-North Railroad carried approximately 86.5 million in 2016, the highest ridership in Metro-North’s history.
    The LIRR’s growth continues recent trends in which the LIRR has registered a 1.97% average growth per year over the past 5 years. The railroad’s ridership has grown 10.2% over five years, from 81.0 million in 2011. Metro-North’s ridership for 2016 surpasses the previous record of 86.3 million, set last year. Metro-North’s total ridership has more than doubled since the railroad was founded in 1983.
    The ridership figures come at a time when Governor Andrew M. Cuomo has proposed a major capacity increase to the LIRR by expanding the Main Line from two tracks to three between Floral Park and Hicksville and as the LIRR is building a second track from Farmingdale to Ronkonkoma. Governor Cuomo has also announced the re-envisioning of Penn Station, which is expected to host some Metro-North New Haven Line service, via four new stations in the Bronx to be built in the coming years.
    “The ridership figures underscore the importance of the LIRR and Metro-North capacity expansion projects that are underway or proposed,” said MTA Chairman and CEO Thomas F. Prendergast.
    Changing Economic Patterns Point to Further Growth for LIRR
    Underlying economic and demographic trends portend ridership growth continuing into the future, as a generation now entering the workforce shows a greater reliance on the railroad than older generations. A detailed demographic and travel analysis of LIRR customers shows an increasing reliance on the LIRR on the part of younger generations, and the beginnings of a reverse-travel market segment that the MTA expects would be expanded if Governor Cuomo’s proposed Main Line Expansion project is built as expected.
    The study showed that millennials, defined as those born between 1981 and 1997, have lower levels of access to automobiles than older New Yorkers, and are more likely to reach their local station by walking, bus, or being dropped off by others.
    “Our data reinforces what we’ve seen elsewhere that millennials are more likely to opt for the railroad as matter of choice, and to embrace a lifestyle built around downtown activities and living than previous generations,” said William Wheeler, MTA Director of Planning. “We know that habits that are developed early in one’s adult life tend to stick with them through their entire working lives. So the trend bodes well as a long-term positive for LIRR ridership.”
    The survey of LIRR customers found that for weekday travel via LIRR, 65% of trips were made to Manhattan for work, 14% were for westbound work travel elsewhere, 9% were for non-work travel to Manhattan, and 11% were for eastbound travel for work or non-work.
    “These results will be very valuable to the railroad as we make decisions regarding service planning, capital program expenditures and marketing in the years ahead,” said LIRR President Patrick Nowakowski. “There is an intrinsic demand for reverse-peak travel to the Island that today is very difficult for the LIRR to accommodate as a two-track railroad. This data shows that if and when the Main Line is expanded to a third track, our reverse-commute service would fill an immediate unmet need.”
    Ridership Records on Metro-North’s Harlem, Hudson and New Haven Lines
    All three of Metro-North’s East of Hudson Lines surpassed records. The Harlem Line and the Hudson Line beat last year’s record by over 125,000 each, with 27.7 on the Harlem Line and 16.6 million rides on the Hudson Line. The New Haven Line, Metro-North’s busiest, had another exceptional year, with 40.5 million annual rides, surpassing last’s year’s record by approximately 20,000.
    East of Hudson ridership numbers are strong for both customers commuting to and from work and non-commuters. Annual commutation ridership is 0.6% above 2015. Non-commutation ridership for 2016 remains consistent with 2015’s increase of 2.3%
    West of Hudson annual ridership, which was negatively impacted by September’s Hoboken Terminal train accident, dipped to 1.7 million, down 61,368 from last year.
    More customers took advantage of Metro-North’s connecting services in 2016. Combined ridership on the Railroad’s three connecting services – the Hudson Rail Link, Haverstraw-Ossining Ferry and the Newburgh-Beacon Ferry – grew by about 577,000, up 3.8% from 2015. Ridership increased by 10.8% on the Haverstraw-Ossining Ferry, by 4.3% on the Newburgh-Beacon Ferry, and by 1.5% on the Hudson Rail Link.
    “We’ve worked diligently to improve service for our customers by providing more frequent train service and enhancing service reliability, and we’ve accomplished these goals while maintaining the highest safety standards,” said Metro-North President Joseph Giulietti. “We’ve delivered technological advancements that make service even more convenient, including eTix and the expanding availability of real-time information. We’re pleased and grateful that customers are responding to our efforts. But this record isn’t an end point for Metro-North, and we’ll continue to strive to improve service for our customers.”

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: