Jordan:
The correct lesson to learn from this is that the car manufacturers should make small, light, safe cars (cf HyperCar from RMI[*]) which will both a) benefit the drivers and the environment (safer and cleaner); and b) remove the irrational decisions made by stupid pressures that lead people to buy hummers.
[WS:] Granted, but how on earth are you going to remove the stupid pressure to buy hummers, if not by making driving them prohibitively expensive? By whipping or jailing the owners?
Another point - your argument seems to change the subject a bit - I did not argue just for more efficient cars, but for more efficient means of transportation, including public transit. With that in mind, you should have said "the correct lesson is to build better transit alternatives, including small, light, safe cars, as well as efficient public transit."
More importantly, however, you seem to believe that people are somehow pressured to buy hummers instead of making rational decisions to do so. I challenge you to identify that compulsion by buy hummers. Until you do that, I keep claiming that people buy hummers and "environmentally incorrect" means of transportation in general because they find it less costly than alternatives. If one wants to change that, one needs to alter that cost/benefit structure by making the status more costly, and the alternatives less so.
You also seem to believe that simply making a viable alternative available will make people switch to that alternative. That reasoning seems fallacious because it does not take into account the transaction cost of change itself. Changing the established habits is inherently costly, and that cost should be added to the cost of the alternative. To illustrate that point: if the cost of the status quo is X, and the cost of an alternative is X-1, that itself is insufficient reason to change, because there is a cost (or a perceived cost) of the change itself (Y). Therefore, the cost of the alternative is X-1+Y (cost of change) which often renders that alternative more costly than the status quo.
It is very difficult to change the value of Y (cost of change) by policy means, because that value is determined subjectively by the people who are to make that change. One of the key reasons is the so-called endowment effect http://en.wikipedia.org/wiki/Endowment_effect. Therefore, to affect the consumer choice by policy means, the only viable alternatives are a significant reduction of the cost of the alternative or increasing the cost of the status quo. In case of public transit, the cost reduction of the alternative is naturally limited, since transit fares are already low (e.g. day pass in NYC is $7, which is a drop in a bucket for most people). Therefore, the only effective policy alternative is to increase the cost of the status quo (driving), which does not face a similar limitation (indeed, sky is the limit here.)
That also answers Dwayne's comments alleging that I want to "punish" people. I do not know where you got the idea of "punishment," Dwayne, since I definitely did not use that concept. I believe that people are fundamentally rational, albeit that rationality is always bounded and framed/discrete as opposed to being fully informed and continuous as the rat-choice theorists seem to assume. That means that they make their decision based on some form of cost/benefit calculus: they avoid choices that seem more costly to them and opt for those that appear less so.
Therefore, to change the choices people make you need to alter the cost/benefit balance. It has nothing to do with punishment. In fact, it avoids punitive measures to change behavior (e.g. coercion, fines, jail, whips, chains, etc. which would make Brian happy :))
Wojtek