Aside from the up-front capital costs, rail transit comes with many hidden costs that rail advocates either downplay or ignore entirely. The first of these is interest on debt. Transit agencies rarely if ever have to borrow money to buy buses, so pay no interest. But, even with the federal government paying half the costs, most transit agencies have to borrow money to pay much of the other half, and then repay that money out of the same sales, property, or other taxes that fund their operations. This means that money that could otherwise go to improving transit operations instead must be spent on interest.
While transit planners and advocates may mention interest in planning documents, they almost never include it when accounting for the cost of transit projects. For example, proponents of a recent light-rail proposal for Pinellas County, Florida, would state that the light-rail project cost $1.6 billion. But this didn’t include the nearly $700 million in interest that local taxpayers would have to pay on bonded debt to build the rail line.
Another hidden cost is inflation. Most preliminary estimates of rail projects are based on constant dollars for a recent year for which inflation is known. For example, the projected $1.6 billion cost of the Pinellas rail project mentioned above was in 2011 dollars. But by the time the project had reached the voters in 2014, inflation had already added about 5 percent to the cost. By the time the project would have been completed, inflation would have added about $650 million to the cost. Of course, inflation affects everything, but public officials often express shock when they learn that a project that was supposed to cost $500 million actually turned out to cost $600 million due to inflation.
A cost that is even less frequently mentioned by transit planners is maintenance. A well-patronized rail transit line may appear to cost less to operate per rider than buses, but that’s partly because transit agencies count maintenance as a capital cost rather than an operating cost. While buses share streets (and the costs of maintaining those streets) with cars and trucks, most rail transit lines are dedicated exclusively to transit and so maintenance costs must be covered by transit agencies.
When transit systems are new, maintenance costs are low, but as they age those costs rise. After about 30 years, transit agencies must replace much of the tracks, power facilities, and other infrastructure, and that cost is often as great as the original construction cost. Thus, about the time the bonds to build the lines are paid off, agencies must find or borrow similar amounts of money to keep the rail lines going.
A 2010 study by the Federal Transit Administration found that transit agencies with older rail systems had close to a $60 billion maintenance backlog due to transit agencies putting money into building new transit lines rather than maintaining existing ones. The backlog has only grown since then because transit agencies weren’t even spending enough on maintenance to keep their rail lines in their current state of poor repair.
Although the 2010 report did not break out the backlog by transit agency, rail systems in New York, Chicago, Washington, and Boston each have unfunded maintenance needs of roughly $10 billion. Philadelphia and San Francisco also have significant backlogs, while newer rail lines in cities such as San Diego and Portland are just beginning to suffer from deferred maintenance due to lack of funds.
Cost overruns, interest and finance charges, and maintenance and capital replacement costs all significantly add to the cost of rail transit. In order to pay these costs, transit agencies almost always end up cannibalizing their bus systems. Since buses carry far more transit riders than rail in all but a handful of urban areas (namely Atlanta, Boston, New York, San Francisco, and Washington), cutting the bus system to build rail often means an overall reduction in transit service and ridership.
For example, in 1979 Atlanta opened the first 17-mile line of its heavy-rail system. At first the system appeared to be successful, with rail ridership quickly growing to nearly 60 million trips per year in 1985 while bus ridership also grew from about 80 million trips in 1979 to 98 million trips in 1985. As Atlanta expanded the rail system to 26 miles in 1985 and eventually to 52 miles, however, bus ridership declined to 60 million trips per year in 2012, while rail ridership peaked at 84 million trips in 2000 (and 91 million trips during the year Atlanta hosted the 1998 Olympics) and then fell to 70 million trips in 2012. From 1985 to 2012, for every new rail trip, Atlanta transit has lost two bus trips. Since the population of the Atlanta urban area grew by about 150 percent during this time period, the decline in overall transit ridership translates to a huge decline in per capita ridership.
Part of the reason for the decline in ridership was a decline in bus service, which went from 25.2 million revenue miles of service in 1985 to 22.8 million in 2012. Part of the reason was that, as Atlanta built rail, buses that once took people straight to major job centers were rerouted to be rail feeder lines, forcing people to transfer from bus to rail to bus and thereby making transit less desirable. Finally, rail transit proved ineffective at countering the decentralization of both jobs and housing that have made auto travel far more attractive than transit commuting.