Government Operations Committee on Oct. 4th, 2012

[Peter Braid, Conservative, Kitchener—Waterloo, ON]

That’s great.

My final question is about financing. Both in your presentation and in responses to questions, you've spoken about the fact that risk is transferred to the private sector. Risk obviously has a cost. The private sector will build in cost for risk.

In addition, appropriately so, the private sector needs to make money. Given those conditions, how are these arrangements good for the taxpayer?

[John McBride]

That's a great question. That is one of the most fundamental points of a P3, and that's why it doesn't work in all circumstances. The question is, who is best able to manage that risk?

If the government were to take that risk, would it cost the government more than if the private sector were to take the risk? If the government did a procurement in its traditional way, what would the probability of a cost overrun be?

We'd look at the empirical evidence. We do workshops with experts about the probability and the likelihood—we do it on a whole-risk register, a whole set of risks. So there's quite a systematic evaluation of the risks, probability, and results. We do things called Monte Carlo simulations to figure out the evaluation of those kinds of risks, what those risks are worth, and what they will likely cost the government. Then the question is whether the private sector is better able to manage those risks. They will charge you for them, but will they charge you less than if you did them on our circumstances?

That's why, if it's low-risk, standard, for example, you're building or replacing sidewalks—but if you're building the Sea-to-Sky Highway from Vancouver to Whistler, up that coast…. If any of you have driven on it, think of the engineering and the issues of that Sea-to-Sky Highway. There are significant engineering complexities in that kind of project, or a major hospital, but they can manage those kinds of risks better. They will charge you, and that's the evaluation. If they can't, you don't do a P3, and if they can, you do. That's really a technocratic evaluation.

On average, our prior investments, our estimates...well, you have to see what it is in reality—they are about 8% better value for the projects we invest in. Because we invest only 25% in the projects, on a $1.5 billion project, on an average of 8%—I'll do my math—that's $120 million or $130 million of better value for the taxpayer, plus we're getting the infrastructure.

[The Chair, NDP, Pat Martin]

Thanks, Peter.

John McCallum, go ahead.

[John McCallum, Liberal, Markham—Unionville, ON]

Out of this $1.2 billion, are you able to tell us what is spent in 2009–10, 2010–11, 2011–12, and perhaps projections for 2012–13?

[John McBride]

I can certainly tell you what is committed, yes.

[John McCallum, Liberal, Markham—Unionville, ON]

Okay. I think that would be useful.

You've stressed the fact that sometimes a P3 is good and sometimes a P3 isn't good. Could you tell us, with concrete examples or criteria, the types of situations where it's not good? I can think of traditional areas where P3s are often used, such as for highways, bridges, and hospitals, and maybe even prisons—although that would be controversial—airports.... Is it a question of scale and complexity?

[John McBride]

Those would be the two most important factors, but there are others. Size—people have different views, and it's a bit different depending on the sector. You've got to think of size, both capital costs and operating, but less than $50 million of capital cost. Why is that? That's both a question of the transaction costs...so there's the same sort of bid cost to these kinds of things, whether it's a $200 million project or a $50 million project.

To attract the interest of the private sector and really get good competition and to really justify the incremental transaction costs—which is part of the weighing of these kinds of things—it has to be of a certain type of scale.

The second one is risk. That's complexity, but what is the risk to the private sector? Partly it's complexity and partly it’s correlated with scale. Large projects tend to be riskier, but risk is also a question of experience. If it's a project you don't do very often—for example, you're a regional health authority and you build a hospital once in a generation—it is probably better to engage somebody who has built 50 hospitals rather than learning on your own. So it's your own expertise in being able to deliver that.

[John McCallum, Liberal, Markham—Unionville, ON]

Generally, you're saying the greater the risk, the more likely you want a P3.

[John McBride]

Correct.

The third one would be your ability to transfer that risk to the private sector. Sometimes there's a lot of risk but you can't transfer it to the private sector. What would be inefficient things to transfer to the private sector? Often permitting risk is a difficult risk to transfer because it's with the public sector. There can be risk of external change. That's why I mentioned the IT example. The private sector will do anything for you, but they will price that risk. Is it a price of risk that you think is actually better, or is it a risk that you should take yourself? It's really a question of how to partition risk.

[John McCallum, Liberal, Markham—Unionville, ON]

The social housing, that's a pretty difficult one, isn't it?

[John McBride]

Yes.

[John McCallum, Liberal, Markham—Unionville, ON]

Why is it difficult?

[John McBride]

You're taking existing buildings that are 100 years old and they have to rehabilitate them. How do you deal with the existing buildings? It's easier for greenfield, but people are looking at how you do it for rehabilitation. When it's a greenfield, all you're really taking is geotechnical risk. But when they're 100-year-old buildings and you start to take down the walls, how is the private sector going to price a risk it can't know before it actually signs on the contractual line? How do you deal with that, and how do you give them the best information on the existing state of it? That was possible only because there had been some done in a traditional way. They got experience on those types of buildings in the east side of Vancouver, so they were willing to do it. When you're taking pre-existing risks, it's much more difficult to legally and contractually split them.

[John McCallum, Liberal, Markham—Unionville, ON]

Thank you.

[The Chair, NDP, Pat Martin]

You're finished, John?

[John McCallum, Liberal, Markham—Unionville, ON]

Yes.

[The Chair, NDP, Pat Martin]

Thank you.

Next, Ron Cannan.

[Ron Cannan, Conservative, Kelowna—Lake Country, BC]

Thank you, Mr. Chair.

And thanks, Mr. McBride.

Coming from British Columbia, I'm well aware of the P3, of Mr. Blain and Partnerships BC. We have two wonderful post-secondaries, Okanagan College and UBC Okanagan, which is a $100-million-plus investment. One great example besides the Sea-to-Sky is the William R. Bennett Bridge, which is a $144-million, five-lane bridge. It's been an icon for Partnerships BC. If you want to come and take a tour, you'll also come across some award-winning wineries at either end.

[Voices]

Oh, oh!

[Ron Cannan, Conservative, Kelowna—Lake Country, BC]

Just make sure you don't drink and drive.

That one was a design, build, finance, and operate. Maybe you can elaborate to the committee, share a little bit about the different models. There's BOOT and all these other acronyms, so maybe you can enlighten us a little bit on that.

[John McBride]

B.C. was a leader in the country in the P3 model. There are lots of great examples in Quebec. I talked about how there are various steps in doing an infrastructure project. There's the design, there's the build, there's the operate, maintain, and finance. So there are various models of P3. P3 is really an umbrella of models. It's a question of what's the most efficient procurement mechanism for a particular project. Is it the traditional way? Is it a design/build? Is it a design/build/finance? Is it a design/build/finance/maintain? You can package different things together depending on the outcome.

I'll give you an example. We'll stay in B.C. The Evergreen Line in Vancouver has been procured as a design/build/finance. Why not include operate and maintain? It's the extension of an existing line, and the integration of the operating and maintenance would not be effective. In water and waste water, operation and maintenance is critical because it's integral to the success of the project. It's difficult to partition risk for design and build.

So it actually requires a thorough analysis of the various models. P3 is a grouping of various models. It's really a technocratic question: what produces the best outcome for the public sector in the range of available models?

[Ron Cannan, Conservative, Kelowna—Lake Country, BC]

Thank you.

You mentioned earlier in your comments that it's a very complex contractual deal. I discussed this with my legislative assistant, who's been on the Hill for a number of years. When they reviewed this P3 model before, they talked about the different templates or guidelines and how every deal is different. From your experience, is it then the more precise the contract, the less likely there will be any problems down the road, and the better the result?

[John McBride]

Absolutely. At the end of the day, these are contractual arrangements, and your ability to negotiate and have clear and transparent contractual terms makes them easier to manage.

It's one of those things...why Canada is a global leader. It has established Ontario-B.C. standard contractual documentation, which lowers transaction costs for the legal review of them every time; you can start with basic templates and contractual documentation. Obviously, performance specifications and things like that you need for the project, but you don't have to review every single clause again. They have been amended and updated based on experience, as you get better contractual relationships.

If you went back and talked to Larry, he'd say, “If I knew what I know now, I would have provided this or I would have provided that.” We've continually learned, and updated the contractual templates. I've been very fortunate to be able to benefit from Larry's work and from David Livingston in Ontario, who have been leaders in Canada on this, and to take the best of their best practices and try to apply them at the federal level.

[Ron Cannan, Conservative, Kelowna—Lake Country, BC]

It's an evolving learning process, I agree. Thanks.

In regard to the dollar value, I thought at one time they were talking about $50 million. How did you settle at $100 million?

[John McBride]

At the federal level? At the end of the day, that was the government's decision.

In B.C., it's at $50 million. Alberta is at $100 million. I think to start with...and you'd have to ask the Treasury Board and the Department of Finance why, at the end of the day, they settled on $100 million, because it was their decision.

It's pretty clear that at over $100 million, you should be at least thinking about whether or not a P3 is viable. To get started, it was the 80-20 rule. I think you're going to get 80% of what you should. You will also note that it's mandatory over $100 million and encouraged for under $100 million. So it's not that people aren't thinking about it for under $100 million. For it to be mandatory, people wanted to have less focus on the ones that are likely going to be the most beneficial as P3s.