In brief, O'Toole argues that light rail is a bad deal, particularly in light of an alternative that can share infrastructure with cars and trucks (double-decker buses). O'Toole also notes that the projections made by rail supporters about ridership and the like are usually...let's say optimistic.
We get light rail because Baptist-and-bootlegger coalitions are enabled by ignorant voters (the link is to an EconTalk podcast in which Bruce Yandle explains the concept). The Baptists are environmentalists who see rail as a green alternative to automobiles and civic boosters who want economic development (which rail won't cause) or for their town to be a "world-class city" ("whatever that is," O'Toole asks). The bootleggers are the contractors and construction workers as well as land owners in downtown areas who will see demand for their services increase. This is further made possible by voters who don't grasp one of the first lessons of economics: resources are scarce, and they have alternative uses. Again, we're back to Frederic Bastiat's "What is Seen, and What is Not Seen": it's easy to see the shiny new rail line and say "that made us richer!" It's harder to see what we sacrificed in order to get that shiny new rail line.
Political incentives in transit are especially perverse. You get to write press releases and get your picture taken at a ribbon-cutting ceremony and have your name put on a plaque at the new train station that was built under your leadership. You don't get much recognition for making sure potholes get filled. Construction is overfunded, and maintenance is underfunded. Discussing maintenance problems in Washington, Chicago, and Boston, O'Toole writes:
Such maintenance shortfalls are almost guaranteed in a transport system that is not funded entirely out of user fees. Politicians love to support grandiose capital projects, especially if they can get some other level of government to fund them. That allows the politicians to bask in glory when the projects open for business. But they routinely underfund maintenance, as there is little political benefit in replacing a worn-out rail, brake shoe, or electrical signal, while accidents, delays, and other problems can always be blamed on someone else.
For almost all American cities, it's a mistake to try to green the planet and relieve congestion by building more rail. O'Toole's conclusion:
In the end, building new rail transit lines, at least in the Americas, is almost always a mistake. Putting the same amount of money to use in relieving congestion for everyone by undertaking such projects as coordinating traffic signals and building high-occupancy toll lanes adjacent to crowded highways would produce far greater benefits. Alternatively, providing the same transit capacity with buses instead of trains would cost far less.
ATSRTWT. It's a useful corrective to mistaken thinking about urban planning.