If gas prices don't go up, then high-speed rail, the way it's presently conceived, is not going to break even. It's going to need a subsidy. What I was talking about was trying to explore--and I think this should be done much more rigorously than I've had the opportunity to do--what would be the break-even price, and is that a relevant price? If you were to do your sums and find that it was $20 per litre for gasoline that was the break-even price, then you wouldn't worry about that argument. But you can do the sums and find out that it's actually not $20 per litre; it's something actually quite close to where we are now, or not so much where we are now, but where we were last year. That makes it very interesting.
So you're not doing any societal engineering. What you're doing is saying, look, we have a reasonable chance that the oil price is going to go up that high, and thus the gas price will go up that high, and therefore a goodly number of people, if the rail is there, will use the rail.