is there any evidence or calculations that show roads to be the least subsidized form of transportation?

Well, I think that's a pretty easy question to answer without vast compilations of data. Overall, the total dollars that have gone to subsidize roads is certainly many times greater than the total for other forms of transportation, while on a percentage of total dollars spent, other forms probably far exceed the percent of road subsidy per total spending.

This doesn't mean we shouldn't use more market mechanisms, merely that we shouldn't be surprised if the results in a more market based world were at least as car-centric as those in the current one.

I wouldn't be surprised either. But, I also wouldn't be surprised if the transaction costs associated with private roads in a free society encouraged innovations that would have made the car obsolete. But, all other variables held constant, I think it's safe to bet that development patterns would be more compact and diverse without the heavy subsidies to transportation since the progressive era.

Please don't mistake me for someone who is advocating one form of transportation over another. This is where I think O'Toole and I diverge. He says, “let's not criticize roads because transit it more subsidized.” I say, “I don't care what type of transportation it is, it's all heavily subsidized.” So, my overall objective isn't to advocate one form of transportation over another, but to show that present land use patterns are a reflection of the vast subsidies to transportation, in general.

First, you raise some interesting negative externalities of highways that are not included in cost discussions, but there are also positive externalities (e.g. more commerce). Is there any strong reason to believe either of these effects dominates?

I tried to be careful not to push the externalities angle to hard, but if I wrote it over I think I would downplay the externalities even further since the case can be pretty clearly made with explicit and implicit costs. It's too hard to quantify externalities and I find the concept of “costs to society” highly subjective. So, you could be right – you could be wrong.

Second, your use of opportunity costs with respect to tax revenue is one that as best is not unique to transportation, but rather is an argument that the economic cost of taxes always exceeds the nominal amount of such taxes– even apart from any efficiency losses with respect to the activity/item being taxed. It is probably correct, but it is hard to measure the effect of this loss because the subsidized project might generate value in excess of the nominal cost. This is unlikely to be true overall given all of the ways in which government resource allocation is inferior to private resource allocation, but I think your post overstates the negative effect rhetorically by not mentioning the value that projects might have in excess of their costs.

My main point here (perhaps the wording could be improved) is that a private road owner/operator would face the additional burden of taxes on revenue, land, and financing. Whereas the public sector can ignore this. So, to claim something “pays for itself” that benefits from reliefs that private businesses cannot take advantage of, significantly distorts the truth. Either we should examine all private businesses on a pre-tax basis, or we should apply a comparative burden when analyzing public endeavors to see if they “pay for themselves.”

Given that this seems like a hard area in which to make calculations, I guess I am more inclined than you to cut Randal O'Toole some slack.

Slack for what? OK, let's completely ignore externalities. I think that it's quite clear that roads don't even pay for themselves when only looking at explicit costs. Throw in implicit costs, which is necessary and proper when asking “does x pay for itself?”, and roads don't even come close to paying for themselves except maybe a few cases. If they paid for themselves, we would have a vastly competitive private road market – and we don't.

With respect to a couple of specific points:

First, you raise some interesting negative externalities of highways that are not included in cost discussions, but there are also positive externalities (e.g. more commerce). Is there any strong reason to believe either of these effects dominates?

Second, your use of opportunity costs with respect to tax revenue is one that as best is not unique to transportation, but rather is an argument that the economic cost of taxes always exceeds the nominal amount of such taxes– even apart from any efficiency losses with respect to the activity/item being taxed. It is probably correct, but it is hard to measure the effect of this loss because the subsidzied project might generate value in excess of the nominal cost. This is unlikely to be true overall given all of the ways in which government resource allocation is inferior to private resource allocation, but I think your post overstates the negative effect rhetorically by not mentioning the value that projects might have in excess of their costs.

Given that this seems like a hard area in which to make calculations, I guess I am nore inclined than you to cut Randal O'Toole some slack.

I’d also like to point out that, private highways may make payment slightly more complex. However, what if I told you optimal-flow tolling (that maximizes profit by charging more for busier times) will almost completely eliminate congestion? The only time you will ever have congestion is if there is an accident. Since the private operator looses revenue for every second the highway is not at peak flow, it has greater incentive to keep the road safer, clean accidents quickly, and keep construction disturbances to a minimal.

Congestion will be so rare, you may forget what the word means. Wouldn’t that make your life less complex overall?