Writing up a cool paper on the positive externalities of mass transit, Paul Krugman describes transit investment as a kind of useful second-best policy: "the right answer is to get the incentives right, and charge large fees for driving in congestion."
That's correct, but I also think it's important for transit fans to see that congestion fees are a kind of force multiplier for transit. After all, in some big American cities the peak congestion charge would have to get quite hefty at some times of the day. Some folks will respond to that by paying the fee, some by time-shifting their driving to a less-crowded hour, and some by riding transit. A bus, after all, is a great mechanism for spreading the cost of road access across a large number of people. And while with highways the quality of the service provided declines with the number of users (traffic jams), with well-designed transit it goes the other way. The more people who want to travel on a particular transit route, the more financially viable it is to provide high-frequency service. And high-frequency service is the key to real-world transit useability.