I absolutely agree that long-term infrastructure planning is extremely important. We'll work together with them.
On the issue of risk transfer, I think a number of studies have said that.... Well, first of all, in terms of construction, most of the risk is at the upfront stage. Unless you're dealing with somebody who is going to have toll booths, there actually isn't a lot of risk in longer-term operations and maintenance. I think the Vining and Boardman study from UBC basically said that a lot of that risk, frankly, is not transferred.
At the end of the day—and I haven't seen any of the risk analyses that have done this—it's always the public sector that's responsible for providing that service. Pretty much all P3s in Canada are set up as special purpose vehicles, as I said, which means that the private sector can either go bankrupt or walk away. In Ottawa here, there are two small P3s that are profitable, but the big companies that own them—they're in arenas—walked away because they weren't making enough profit. Actually, one of them walked away; one of them demanded more money from the government.
The issue of capacity is really important. There was an experience in Nova Scotia with P3 schools. The contracts were about 2,000 pages long. The auditor there found that basically nobody in government had a handle on these. Some of the schools were owed hundreds of thousands of dollars. They didn't know about it. Nobody in government.... That's the provincial government.
Therefore the issue of having capacity to deal with these is really important, but as a number of people have said—and this gets back to another question—it's really only the big projects that should be viable in any sort of way for that, because of such high transaction costs and the need to have that ability.