Reason #9 to Stop Subsidizing Transit
It Doesn’t Relieve Congestion

Congestion relief is one of the most used–and often most persuasive–arguments in favor of increased transit subsidies. Transit carries more than half of New York City workers to their jobs, and as such it prevents that city from being more congested than it already is. However, at least since 1970, almost nowhere in the United States has a subsidized expansion of transit service led to a reduction in overall congestion.

Transit’s Share of Travel in 1970 and 2015

Urbanized Area19702015
Atlanta10.4%3.8%
Baltimore16.9%8.2%
Boston18.2%15.5%
Buffalo12.3%4.4%
Charlotte9.1%3.0%
Chicago24.4%13.7%
Dallas-Fort Worth-Arlington5.7%1.9%
Denver4.8%4.5%
Houston6.0%2.6%
Los Angeles4.8%5.6%
Miami-Ft. Lauderdale-W. Palm Beach6.2%4.0%
Minneapolis-Saint Paul9.5%6.1%
New York39.0%34.6%
Norfolk-Virginia Beach9.0%2.0%
Phoenix-Mesa1.3%2.5%
Portland7.0%8.5%
Sacramento2.7%2.8%
Salt Lake City2.3%4.2%
San Diego4.8%3.9%
San Francisco-Oakland16.0%21.2%
San Jose2.4%4.5%
Seattle6.6%10.7%
St. Louis9.2%3.3%
Washington17.6%17.6%

In most urban areas, subsidies to transit began in earnest in 1970, plus or minus five years. As shown in the table above, since then transit usage in most of these areas has declined despite the subsidies. The only major exceptions are Phoenix, Salt Lake City, San Francisco-Oakland, San Jose, and Los Angeles. Portland and Seattle also saw an increase in transit’s share, though for what it’s worth, all of the increase in Portland and most of Seattle’s increase took place in the 1970s.

Despite the increased transit shares in a few urban areas, congestion has grown dramatically in those same areas as job growth has meant that auto commuting has also increased. In all of these urban areas, congestion is several times worse today than it was in 1970.

While it might be argued that, were it not for transit subsidies, congestion would be even worse than it is, in most cases the reverse is true. The average transit base carries only 10 passengers at a time over the course of a day, and in many urban areas the average is much smaller. Yet that bus has the footprint of three or more cars, especially since it frequently pulls off and on the streets at transit stops. Light rail also adds to congestion, and often commuter rail does as well. Only heavy rail never makes congestion worse than it already is.

From the table above, it seems that San Francisco-Oakland is the most likely candidate for transit actually reducing congestion. Yet, despite the increase in transit’s share of commuting, per capita transit ridership has dramatically declined, at least in the last 35 years. While the San Francisco BART system is often credited with relieving congestion on the San Francisco-Oakland Bay Bridge, elsewhere in the urban area congestion has gotten far worse.

“Increasing transit utilization does not lead to a reduction in traffic congestion,” say Thomas Rubin and Fatma Mansour in this study of the relationship between transit and congestion in 74 urban areas. Outside of New York and a few major downtown areas–Boston, Chicago, Philadelphia, San Francisco, and Washington–transit simply does not carry enough people to have a discernible influence on congestion. Even if transit ridership doubled in most corridors–something that has probably never happened since 1970–no one driving in the corridor would even notice.

In short, transit subsidies do not have a major influence on congestion. Anyone who supports transit subsidies on the grounds that transit relieves congestion is either in New York, deluded, or misrepresenting the facts.

Share

3 thoughts on “Reason #9 to Stop Subsidizing Transit
It Doesn’t Relieve Congestion

  1. Dave Brough

    “Even if transit ridership doubled in most corridors–something that has probably never happened since 1970–no one driving in the corridor would even notice.”
    You only notice it in places where they ‘repurpose’ lanes that used to carry hundreds of cars to a service that now runs empty double-decker buses every twenty minutes (Sundays and holidays excepted). http://mitchellsfrontpage.com/wp-content/uploads/2015/01/49572_1332673774_bus-lane-surfacing1.jpg

  2. rojwilco

    It would be interesting to see per capita ridership, transit share of travel, and congestion all charted on the same graph for 1970 to now. This brings up a question for me, however: how do you quantify congestion?

Leave a Reply