The Federal Transit Administration has published its New Starts recommendations for 2017. The recommendations include profiles of more than 60 different transit projects, including bus-rapid transit (BR), streetcars (SR), light rail (LR), commuter rail (CR), and heavy rail (HR). Only 60 projects are shown in the table below as a few, such as the San Antonio streetcar, have been cancelled or at least are “on hold.”
|San Fran.||Van Ness||BR||2.0||163||81|
|San Jose||El Camino||BR||17.4||188||11|
|Chicago||Red & Purple||HRC||5.6||957||171|
|Dallas||Red & Blue||LRC||48.1||119||2|
For the projects in this year’s report, the average cost of new streetcar lines is $45 million per mile; light rail is $163 million per mile; heavy rail is $347 million per mile; and commuter rail is $38 million per mile. Of course, cost overruns are likely for a majority of these projects, so the final costs are likely to average 20 to 50 percent more. Even without the overruns, these costs are outlandish, as the nation’s first light-rail lines cost only around $10 million to $25 million per mile (after adjusting for inflation), and streetcars were supposed to be even less expensive than that.
|City||Project||Mode||Daily Riders||Cost/trip||Prius cost|
Ridership projections are only available for about 29 projects that are well advanced in the planning process. Based on this ridership data, most rail projects will cost well over $10 per rider while most bus projects will cost less than $5 per rider. Of course, projections are usually high, so the actual costs will be more.
The Federal Transit Administration is no longer interested in the cost-effectiveness of projects, so it rarely bothers to publish how many projected riders will be “new” transit riders. In most cases, at least half of all riders would have taken transit even without the project, so the cost of getting someone out of their car and onto transit would be more than double the cost per rider shown here. In the case of the L.A. Regional Connector and San Francisco Central Subway light-rail projects, almost all of the riders would have taken transit anyway, so the cost per new rider will be many times higher than the cost per average rider.
I like to calculate how often taxpayers could give a new Prius to every daily round-trip transit rider for these projects. For example, the Orlando South commuter-rail project is so expensive and will carry so few riders that we could give every daily round-trip rider a new Prius every 18 months. The extension of the existing North commuter-rail line is even worse: we could give every rider a new Prius every nine months and still save money. Of course, this statistic is most meaningful for commuter-rail lines and other lines used mainly by commuters as opposed to shoppers, event patrons, and so on.
Given that cost-effectiveness is no longer relevant to the FTA, the agency seems to be using the “Field of Dreams” criteria for handing out money: if you apply for it, they will fund it. To any reasonable person, the idea that taxpayers should spend nearly $70 million, plus more than $6 million in annual operating costs, so that 300 people could commute on the North extension of Orlando’s commuter train, for example, seems absurd. Yet the FTA proudly includes this project in its list of recommendations. I’m sure the fact that the line serves the district of one of top-ranking members of the House Transportation and Infrastructure Committee has nothing to do with it.