The Massachusetts Bay Transportation Authority (MBTA or “the T” for short) has a problem. It has a $3 billion maintenance backlog, and must spend $470 million a year just to keep that backlog from growing. It has all kinds of wonderful plans to close that backlog, but those plans are all in the future. In the meantime, its latest budget proposal spares less than $100 million for maintenance.
So suppose someone offered the T a billion dollars. Heck, suppose someone offered it $2 billion. What percent of this money do you think the T would spend on maintenance?
Score a point if you guessed zero, for MBTA is currently spending $2 billion–half from the state and half from the feds–building a 4.3 mile extension of its Green light-rail line from Cambridge to Medford. When completed, this line is projected to increase the T’s total ridership by 7,000 “new” transit trips per day. Since the T currently carries about 1.4 million trips per weekday, the extension will increase ridership by 0.5 percent.
Of course, the extension wasn’t supposed to cost $2 billion or increase ridership by just 0.5 percent. Back in 2005, the major investment study for the project estimated it would cost just $390 million (see page 5-55). That’s in 2005 dollars, but adjusting for inflation to today’s dollars increases it to about $450 million, less than a quarter of the actual cost. Projected annual operating costs have also nearly quadrupled from $9.9 million in 2005 to $36.9 million today. The 2005 study also predicted the line would increase increase daily ridership by more than 14,000 trips, or 1 percent, not just 0.5 percent (same page).
The FTA no longer requires transit agencies to do major investment studies, but at the time such a study was supposed to fairly evaluate a wide range of alternatives. However, almost all the alternatives in this study were some form of rail or another. The one bus alternative was called “transportation system management” or TSM, and this is one case where the overestimated ridership and underestimated costs actually made it appear that rail would be more cost effective than bus. The document estimates that the proposed alternative (1A) would cost $9.59 per hour of travelers’ time saved, while the TSM alternative would cost $10.54.
Of course, quadruple the rail costs and cut riders in half and the actual cost per hour is much greater–at the FTA’s current specified discount rate of 2 percent, I estimate it is roughly $67. Until the Obama administration changed the rules, the FTA automatically rejected federal funding for any project costing more than $25 per hour of time saved.
So here we have another example of a transit agency and political system that can’t say “no” even when costs quadruple and benefits are cut in half. They can’t even say no when the existing system is falling apart. This is just one more example of why transportation should be funded out of user fees rather than tax dollars, as the managers of user-fee-funded systems have incentives to keep their infrastructure in good condition to keep from losing customers and not to build new projects they can’t afford.