I have a lot of respect for Komanoff, and the idea isn't stupid by any means. There are probably many situations where it makes sense, such as on buses in Midtown, as suggested by the RPA. However, I've come to the conclusion that free transit is generally not a good idea.
Kheel and Komanoff acknowledge that transit fares can function as congestion pricing to discourage overuse of the system, and because of this their latest proposal includes subway fares during rush hour. Matt Yglesias expands on this idea in a recent blog post.
I've realized, though, that in this issue (as in many issues regarding support for transit), people come to the table with different goals. I can think of a few reasons for supporting free transit, and depending on which one(s) motivate you, you may or may not be interested in alternative methods or discouraged by negative consequences. Free transit has the potential to:
- Encourage choice of efficient/less-polluting/community-building modes
- Encourage travel (and therefore commuting, shopping, socializing)
- Relieve the burden of poverty
- Reduce the cost of collecting fares
The fare-collection issue is a big one, but the cost of fare collection can be reduced by using proof-of-payment systems. The burden of poverty can be reduced by tax credits, need-based subsidies or other systems (in Curitiba, Jaime Lerner paid people in bus tokens for recyclables brought to collection stations). An effective transportation system encourages travel all by itself; in order to be effective it doesn't have to be free, just affordable.
That leaves us with the first reason: encouraging use of transit because it's more efficient, less polluting and builds better communities (and yes, I know that last one is very loaded). Of course it's not transit itself, but encouraging transit use relative to car use. And to do that you don't need to have free transit, only transit that's a better value than driving.
Free transit is not guaranteed to be a better value than driving. In many places, for a significant segment of the population, transit is practically free anyway. I mean that in the literal sense of "practically": it costs so little relative to some people's incomes that it might as well be free. But it's not a better value because in these places it has non-monetary costs: it requires more waiting and offers less flexible schedules, and often doesn't even go near where a potential user wants to go. So the freeness doesn't make it worthwhile. It may even lead people to believe that transit is worthless.
One major objection that I have to free transit is that it can discourage the development of for-profit transit. How can you make a profit if you don't charge for anything you sell? Or if you're competing with the same thing offered for free by the government? It can be done (I've paid good money to for-profit water suppliers when there was free government water available right nearby), but it's another challenge in a field fraught with challenges.
I'm also a proponent of farebox recovery. When a public transit operator can reduce the subsidies necessary to run a route without driving people away from transit and into cars, then I think they should.
So that's why I think the Kheel Plan still goes too far, and why free-transit absolutism is misguided. Free public transit? In certain places at certain times, yes. Everywhere, all the time, no.