The Ahistorical Vision of “Bus Before Rail”

There is a widespread assumption that urban transit should develop incrementally, progressing from bus to light rail to heavy rail, with the success of each mode justifying the viability of the next.
This model of development is ahistorical, if not revisionist.  We know that historically profitable rail transit thrived long before buses arrived on the scene.  Moreover, there are essentially zero documented cases of high ridership bus routes evolving* into rail lines; as Eva Wood et al write, “no known conversions from BRT to LRT have occurred” (2006).
We know that rail-first development can result in a sustainable urban fabric that supports high transit ridership (e.g Tama Garden City).
Urban fabric resulting from bus-first development has a mixed record (Curitiba may be the only success story?)
The smart thing would be to leave it there; the systems involved are so complex that self-deceptive rationalization is far more likely than true understanding.  Wiser to accept that “trains first” is the historically proven model than to talk ourselves in circles with theories.
That said, here is some theory:
The key difference between trains and buses occurs at the stations; rail stations improve walkability, while bus stops generally do not.
Within one block of a rail station is a high value “direct access” (直接) zone with high foot traffic; within one block of a bus stop is a high noise, high pollution “diesel” zone with comparatively low foot traffic.
The natural result is that the articulation/clustering of walkable density around train stations is much higher than around bus stops.
Rail stations provide sufficiently concentrated foot traffic to support pedestrian oriented commercial life, thereby creating a convenient environment for car-free living, with broad appeal.  This is the virtuous circle by which rail transit generates its own ridership.
The high capital costs of rail greatly increase the likelihood of joint transit+property development.  The high cost of grade separated stations promotes greater stations spacing, vastly increasing their chances of achieving a critical mass of foot traffic at any one station.  Moreover, the cost and quality of rail justify smaller lot sizes (i.e. fine-grained development), another key ingredient for walkability.
The greatest weakness of bus routes might be their relatively low cost; it is too easy to trade frequent service for wide coverage, and it is too easy to add stops; spreading out boardings reduces the ability of any one stop to support pedestrian oriented commerce.  The result is that bus systems usually do not promote articulation/concentration of commerce and the built environment.
*In fact, there are several BRT systems for which conversion to rail is actively being planned (e.g. Bogota).  However, the rapidity with which these lines achieved high ridership indicates that they may as well have started out as rail; the conditions justifying rail development evidently existed before the implementation of BRT, not primarily as a result of BRT+Property joint development.  Curitiba may be the lone exception.
Curitiba’s implementation of BRT has been extremely successful, and holds many (mostly unheeded) land-use and zoning lessons for North American transit planning.  It is the absolute poster-child for BRT.  However, it is not an example of bus to rail evolution; despite carrying over a million passengers as early as 1990, the long awaited dream of rail conversion has yet to materialize.
The Ottawa BRT is relatively successful within the North American context, but its success is largely dependent on an urban fabric originally built on rail transit; “in 1929, the Ottawa Electric Railway carried more than 30 million riders on 90.5 kilometres of track.
Related Reading
Transforming Cities with Transit (Suzuki, Cervero, Iuchi)

5 thoughts on “The Ahistorical Vision of “Bus Before Rail”

  1. There is a bus route in Utrecht, the Netherlands, that is currently in the works of being replaced by a LRT system. It’s a line to the university which is now done with double-articulated busses, which have reached the capacity limit. The new trams can be combined up to 72m long if i’m not mistaking, and can be driven with a lowest interval of 5m. It’s called the Uitholijn, and it has a website in Dutch

    Liked by 1 person

  2. Cortwatsjer has provided the following links discussing a bus->LRT evolution in Aarhus. Google provides a very readable machine translation from the original Danish (I can’t vouch for the accuracy):

    This should be a good case study, as Aarhus has been growing, and will continue to grow:

    “Accelerating growth since the early 2000s brought the inner urban area to roughly 260,000 inhabitants by 2014. The rapid growth is expected to continue until at least 2030 when Aarhus municipality has set an ambitious target for 375,000 inhabitants.” ~Wikipedia

    Aarhus also has a fairly unique bike-share system:

    “Aarhus has a free bike sharing system, Aarhus Bycykler (Aarhus City Bikes). The bicycles are available from 1 April to 30 October at 57 stands throughout the city and can be obtained by placing a DKK 20 coin in the release slot, like caddies in a supermarket. The coin can be retrieved when the bike is returned. Bicycles can also be hired from many shops.”


  3. Yes, but:

    “..even in cities with extensive rail networks, the bus still takes a huge share of public transport.

    “According to the latest numbers from Transport for London, their city’s iconic Underground (402 kilometers long) served 1.2 billion passenger trips from 2013 to 2014. But London’s buses served double that number with 2.4 billion passenger trips. The total passenger kilometers serviced by buses also dwarfed the subway rail service 6-to-1. Buses served 491 million kilometers while rail only served 76.2 million kilometers. (Passenger kilometers=total distance traveled by all the passengers who rode the service.)

    “Similarly, Hong Kong has more than 210 kilometers of heavy and light rail and yet 55% of Hong Kong residents take the bus. Only a quarter use rail. Taipei has over 120 kms of rail that carry 14% of the mode share, while Taipei buses carry 18%. In Singapore, the mode split is 25% for bus, 19% for rail. In Guangzhou, it’s 35% for bus, 14% for rail.

    “Even in cities where the rail service edges out the bus, its still a very close split. Some 12% of New Yorkers take rail and 10% take the bus. In Seoul, rail service takes 35% of the riders while buses carry 28%. The one outlier seems to be Tokyo, where rail carries 48% of riders while buses only carry 3%. But the world’s largest megacity has such an incredibly fine grain of mixed uses–something we should aspire to follow–that 23% walk to their destinations, while 14% take a bike.*

    “Building a new, extensive rail system may take us a decade or more. Reforming the bus system can be done much, much more quickly. We can build extensive and high quality bus rapid transit (BRT) and bus feeder services in less than two years. We can reform the bus value chain in one year. It can be done. Our sister megacities have shown the way.

    “Enrique Peñalosa opened the first phase of Bogota’s Transmilenio BRT system within three years of the inception of the project. Construction took less than a year.”

    *Data from Passenger Transport Mode Shares in World Cities, JOURNEY, Nov. 2011.


Leave a Reply to Christian Cancel reply

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

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

Facebook photo

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

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.