The reason why public transportation requires so much of a subsidy in the US is because we undervalue the cost of all transportation and have heavily subsidized highways, automobiles and low-density single family development for the last 6-7 decades.

Automobile dependency in the US is not simply a matter of personal transportation preferences; but is the result of policy decisions and public infrastructure spending which have heavily favored the automobile. Auto ownership is essentially mandated through land-use policies which require separation of uses, low densities and high parking ratios – all of which conspire to make traveling by any other mode unreasonable.

In order for public transportation to be profitable in the US, a city must have relative high density, have a transit mode share north of 30%, and the primary transit system would have to be grade separated and automated.

Vancouver’s public transportation system (including the elevated and automated Skytrain) has operated in the black for most of the last 12 years.

Achieving a transit mode share greater than 30% in a US city (outside of Manhattan) is likely not possible without a significant scaling up of carsharing service. Carsharing has gone from having about 2,000 member in the US in 2000 to nearly 500,000 members in 2010 – and it continues to expand quickly.

US cities should enact policies which enable and support carsharing service and plan their public transportation systems around grade-separated, automated rail with aggressive transit-oriented development planning.

In addition, everyone seems to overlook toll roads – a great proof that highways can and do pay for themselves. True, at about 8% of total US auto traffic, they are far from setting the tone for surface transportation.

you mean they won’t go spend hours in traffic and hours waiting to board an airborne cattle car?

when we say “mass transit” – do we include air travel?

Once you have this it becomes easier to raise transit fares. Although it would also help if we had good enough transit that people were willing to pay extra in the first place. We certainly have efficient enough transit. Looking at Chicago as an example, our CTA and Metra fares are so far below London it’s almost comical and yet the farebox recovery ratio is actually respectable. If Metra could even charge half London commuter fares, it wouldn’t even need an operating subsidy. Metra BNSF is already close to break-even as it is.

To the small extent that those countries practice market-side redistribution, those reduced fares do it with greater precision and less damage than the walmart transit pricing that American liberals routinely advocate. I am left with the impression that middle class people (straphangers campaign, etc.) just don’t want to pay more for their own rides, which is understandable but ethically nil. If you want to make progress, focus on the coming once in a half-century opportunity to fix the tax code that Reagan killed and Bush buried, and secondly on transit discounts for those below the poverty line. But please don’t keep transit crippled by demanding that base fares stay low. Those superficially progressive politics cheapen transit, reduce it to its current state of filth, and ultimately undermine progressive aims.

“Doc, rich people in New York seem to have no problem taking the subway.”

I guess you mean some rich people. Those that need to get downtown quickly from the upper east side would be crazy not to take the relatively well run 4/5/6. It’s a rather different experience riding transit in from further out places, largely because the one-zone system makes those routes extreme money losers for the authority. (The socialists in France seem to have forgotten to march in the streets against Paris’s extensive ‘regressive’ fare zones.) Not that outer Queens and Brooklyn would be for it, but zones would eventually make their commutes faster and decidedly less degrading, closer to the (much more expensive) Metro North service. Unfortunately they still think they can drive everywhere, so we have those types defunding transit on the tax side and flimsy progressives defunding it on the fare side. The result is America’s globally embarrassing public transportation.

The last paragraph of the quote goes to a crucial point. The subsidies of the last decades enabled relatively cheap car travel and therefore subsequent land use and settlement decisions that have locked us into a situation where it is that much more difficult (decreased density, different settlement patterns) for transit to make money. Private streetcar companies used to be profitable before mass motorization. New roads and highways came along, and a new transportation mode with its unique advantages and downsides took over.

But once you get to a certain density you need to build transit no matter what, since with transit (at least with rails) you get the requisite speed and capacity (but not a lot of flexibility). With cars you get speed and flexibility but not a lot of capacity which in turn doesn’t help with speed. The investment in transit can at first seemingly be avoided when your cities spread out like Houston. But that of course will bite you later on.

So how to compare costs/subsidies per passenger mile between travel modes? What counts and what not?

Direct subsidies to car travel are the construction / maintenance / policing of roads which are paid for by tax money. Other indirect subsidies include the cost of parking where every parking spot (land value + construction) on average reaches $10,000 or more.

There are positive externalities such as making our industrial economy as we know it possible; flexibility and speed i.e. huge gains of productivity; ‘driving force’ in some technological progress.

There are negative externalities such as air quality, 4-5 million accidents, 40,000 deaths every year and resulting health care costs, congestion i.e. loss of productivity, ever higher costs and crazier stunts in foreign affairs to secure access to oil, contribution to negative trade balance, dependence, a lot less disposable income for families etc.

It’s a complicated picture and I haven’t heard of someone who has put all the factors for every mode together. Maybe it’s a moot point because, in total: We need everything – so we subsidize everything (car travel / transit / rail / air transport / ships).
With shrinking energy sources the balance of subsidies will probably shift towards the modes with more capacity and efficiency. But don’t count on it. Denial is one of the strongest human forces.

Anyway, please excuse my musings.

At any rate, while I agree with what Bibi has quoted, I think there is a case to be made that transit users deserve even more subsidy than auto users. And it’s precisely because of those societal benefits.

It all comes down to this: If you want to maximize passenger flows (not vehicle flows), presumably because you’re running up against congestion, then you have to give people reason to choose the more efficient, non-single-occupancy option. Decreasing the congestion and allowing more potential employees to get to work is a societal benefit.

The relevant measure in determining whether transit or roads are more heavily subsidized is subsidy _per passenger-mile_. The figures you cite–total dollars–ignore the fact that roads carry many more people to many more destinations–so of course we should expect that they get more money! Suppose the government were to offer citizens a $100 discount off the purchase of a unicycles and $.50 of the purchase of bicycles. Chances are, because many more people are interested in bicycles than unicycles, the government would end up spending far more money in total helping people by bicycles than they would helping people buy unicycles. But if we were trying to figure out which product “was more heavily subsidized” in the sense of “for which product did the government’s subsidy cause the levels of consumption change more dramatically from what they would have been, without the subsidy”–the answer would clearly be unicycles. The relevant statistic is the subsidy per good consumed. And in the case of mobility, the good is passenger-miles.

Also, dr_doodle is totally correct in pointing out that the figures cited also compare only total social cost of cars to operating cost of buses.