OTI is a nonprofit corporation dedicated to providing transportation information.
Webber is professor emeritus at the University of California, Berkeley and Director
of the University of California Transportation Center.
The fundamental idea that underlay the BART plan held that spatially dispersed patterns
of employment, housing and gathering places ought to be converted into concentric
ones--more like those of Eastern European cities. How to do that? By installing the
kind of transport systems that had earlier shaped those older metropolitan areas.
The other major objective was to eliminate automobile congestion -- an outcome that
would, of course, reinforce urban centering. The key lay in converting the auto-dominant
transport system to a mixed system in which rail-transit would carry much of the
peak-period traffic. By selectively placing rail stations at the mid-points of potential
subcenters, these places would become points of high accessibility, thus attractive
to developers of high-density housing, office buildings and retail complexes. In
turn, as subcenters became densely settled, more and more people would leave their
cars at home and ride trains to work within concentrated employment centers.
Now, something over twenty years after he decision to build BART and ten years after
operations began, a great deal of office space has been built in downtown San Francisco
and downtown Oakland, but neither the suburban centering effects nor the traffic-congestion
effects have been realized. BART may have contributed significantly to CBD growth,
but it has not yet as expected restructured the suburbs. To be sure, BART had a slow
start, owing in part to equipment failures that hampered reliability and hence discouraged
some patronage. The system is still not operating to designed capacity. Until it
is fully fixed patronage will continue to be depressed below potential levels. Nevertheless,
it is now apparent that it is unlikely ever to attract as many passengers as its
designers had hoped for. Fewer people than expected are switching from automobiles
to trains.
As one result, traffic congestion within BART's district is about where it was before.
That's in part because these motorists who did switch to BART left vacant highway
space that was then occupied by others, including others making trips they would
not otherwise have made. It's in part also because BART has drawn many of its riders
from buses, thus displacing low-cost transit service with high-cost service, while
not significantly affecting traffic congestion. And, of course, Bay Area populations
have risen over the years, thus expanding highway demand. By now, BART is carrying
somewhere around 3 percent of vehicular trips within its 3-county district.
Suburban centering has been slow in coming. A few new buildings now stand next to
BART's suburban stations, but the expected concentration has just not happened. Suburban
construction has moved apace, but station sites have not become the magnets they
were expected to be.
Studies of BART's impacts, conducted at the University of California and at the Metropolitan
Transportation Commission, suggest three main explanations for these unexpected outcomes.
1. The Bay Area, like all the other Western metropolitan areas, contains a dense
overlay of streets, roads and freeways, supplemented by extremely high auto owner
ship rates. These regions enjoy virtually complete accessibility. Every place is
directly connected by road to every other place. Despite frequent traffic tie-ups,
most people enjoy exceptional mobility. New rail access at suburban stations added
new accessibility as planned. But the increments proved to be insignificant in proportion
to overall accessibility. As a result very few developers were enticed into seeking
sites adjacent to rail stations. They sought good pieces of land accessible to the
road network instead. It looks as though it is now simply too late to use limited-access
rail transit as an instrument for inducing urban centering.
2. Because BART is laid out essentially as a mainline railroad, rather than as a
network of lines, very few people find either their origins or their destinations
adjacent to stations. For most people, a train ride requires a supplemental trip-leg,
either by foot, car, or bus at either the origin end, the destination end, or both.
Recent studies reveal that travelers find these supplemental trip-legs to be most
onerous, clocking the time costs at two to three times the rate assigned to travel
along the main leg of a trip. Neglect of that factor alone would account for much
of the exaggerated forecasting of rail-transit patronage we have recently seen.
3. Recent research also reveals that people choose automobiles not because they have
a love affair, or that cars confer social status, or that they go fast. People choose
cars because they offer door-to-door, no-wait, no-transfer service at tolerable dollar
costs. Car time spent starting and ending a trip is brief, hence that total trip
time is typically shorter by car than by transit. With a car waiting in one's own
garage, access time is zero; the access leg of the transit trip is eliminated. By
using a single vehicle from origin to destination, the high fractional costs of transfers
are reduced to zero. Where public transit runs adjacent to both origin and destination,
a transit trip can approximate the total travel time of an auto trip. Transit systems
do then effectively compete with cars.
Unfortunately, the geometry of a mainline railroad does not match the geometry of
Western metropolitan settlement patterns, and so the automobile trip is fastest overall.
Therein lies the failure of the rail rapid-transit idea for the Western metropolis.
It is based on the assumption that commuters can be enticed out of their cars and
into modern trains that go fast. But it is total travel time that matters, not speed.
The way to entice people out of cars is with transit that can compete with the automobile
on its own terms--with transit that approximates door-to-door, no-wait, no-transfer
service. That calls for transit technology that is more like an automobile than like
a suburban railroad train. The bus threads through the local neighborhood, then runs
non-stop into the city center, offers one successful model. (AC-Transit buses continue
to serve large numbers of San Francisco-bound East Bay commuters on routes that parallel
BART's tracks. Its advantage: it can collect morning passengers from their local
street comers, then head directly into the City on freeway and bridge.)
The ideal transit system to serve the Western metropolis has not yet been invented,
The closest we've come so far is a shared taxi or jitney--an automobile used in public-transit
mode.
The effective transit system that does evolve will surely be more like a shared taxi
than a train-- adaptable to low-density, dispersed settlement patterns; capable of
providing random-access -- from anywhere to anywhere; approximated door-to-door,
no-wait, no transfer service; thus providing short trip-times and lower dollar-costs
than automobiles allow.
Despite the confidence we all had in the initial idea of a Bay Area rapid-rail system
and in its urbanization and transportation effects, we were wrong. It has not worked
as we expected.
Of all Western metropolises, the Bay Area's urban pattern comes closest to matching
the linear geometry of a suburban rail line. Since it didn't work here, we can be
confident it will not work in a Los Angeles or a Houston, where the urbanized areas
occupy wide plains.
