Ezra Klein advances the Whiny Siblings model of transportation infrastructure funding:

In answer to this, some conservatives suggest Amtrak should be able to stand on its own two feet. But is that fair? Last year, Congress appropriated $40 billion in highway funding — that goes to maintenance and new road construction. And a similar sum is appropriated every year. Amtrak received around a billion. In 2002, the nearest year for which I can find cross-statistics, Congress appropriated $14 billion for the FAA and air travel, $32 billion for highways, and $521 million for Amtrak. Between 1971 and 2001, Air and Highways received 62 times the funding Amtrak did. That the train system isn’t as plush should not, under those circumstances, come as a surprise.

The idea, I suppose, being that if highways and air traffic control get two cookies and an 11:00 curfew, Amtrak should too. But this doesn’t actually make a whole lot of sense. Leave aside that air and roads each have vastly more annual passengers than rail. Government funds highways for the very good reason that it’s infeasible to charge users directly on all roads. (For the moment, anyway: Systems like EZ-Pass provide an obvious model for how it could be done.) Drivers are still paying indirectly, of course: The same year that Congress appropriated $32 billion for highways, the government took in 24.5 billion in federal gasoline taxes. But every Amtrak passenger already pays directly for every train ticket, so there’s no parallel argument for preferring a tax-and-subsidy arrangement to direct user fees.

Also, while Ezra suggests rail is trapped in some kind of vicious cycle of falling ridership and higher ticket prices, Amtrak did apparently have a record number of passengers this past year. Sure, I wish a subsidy could cut into that $120 ticket so I could ride to New York in a comfortable three hours rather than a cramped 4-and-a-half on the Chinatown bus. But unless they can add substantial capacity (and I don’t know, maybe they can), a lot of that subsidy is just a benefit to the relatively more affluent people riding the train anyway.

Maybe the subsidy argument is that even if drivers internalize some of the costs and externalities of driving through tolls and taxes, it’s politically impossible to make these high enough to fully capture those costs, so as a second-best, we should underwrite rail travel in order that choice between transit methods not be skewed. I suppose that’s reasonable in principle. But that’s not an argument for giving rail as much as it needs to be viable; it’s an argument for underwriting rail if it turns out to be viable when the cost displacement is neutral across transit methods, and supposing we’re not further just promoting inefficiently high levels of travel. I don’t know how the numbers work out, but just pointing to the large amounts spent on other modes of transportation doesn’t really tell us much in itself.