Yes, but the thing is the numbers of fares that would have to be paid to cover the cost. I start with an analysis where things cover their cost. Then subsidy, you might say, is a way of addressing shortfalls in that. The primary shortfall is that you don't have enough passengers to do that. I mean, you can envisage a high-speed service between Calgary and Edmonton that carries 20,000 people a day for all trips in all directions. My estimate is that this would cover its cost.
Maybe realistically you're never going to get above 10,000 a day, which means, very roughly, that you have a 50% subsidy of the thing. But the big factor here, I think, is the oil price. It's not a question of social engineering and how you get the people out of their cars. Above a certain oil price, they will get out of their cars if the train is available.
The trick is to try to estimate that. My guesstimate—and I really wouldn't call it more than that—is something in the order of a $1.60 a litre. It's where I arrived at and what I presented at this Red Deer conference. It may well be wrong, but I don't think it's hopelessly wrong. It may very well be $2.30 or something like that.
It's the numbers. You have to figure out the numbers. I don't have the numbers for Toronto-Montreal and Toronto-Ottawa. I'm inclined to think that they might be a little bit better than the Calgary-Edmonton ones, but I don't have them.