Thanks for the invitation to speak with you.
Infrastructure seems to be having a real moment here in Canada as well as abroad. It seems to be on the lips of media and political leaders, as well as getting policy and public attention. There's a lot of interest in this.
When it comes to infrastructure for government, government really has two main decision points. One is which projects to build, and the second point is how they should be procured. I want to touch upon both of these topics briefly within my 10 minutes.
With respect to which projects to build, we have a vast need for infrastructure right across this country. We're going to be spending billions and billions of dollars over the next decade. I think the key—and maybe this doesn't even bear saying, but I'll say it anyway—is that we need to be picking the right projects.
There's this sense that building any infrastructure is valuable, but really, to get the top social value and the top community value from these projects, we have to have mechanisms to pick the right projects. Investing in projects that don't deliver, that aren't the highest priorities, can lead to wasted public money. The project has to be maintained over decades. Also, there's a potential for loss of public confidence in our leaders to actually solve problems.
So we first pick the right projects, and second, focus not only on new projects but also on operations and maintenance. Once you build these things, they're in our communities for decades to come. They need to be maintained. That costs money. I've heard some estimates that the cost of construction is just 20% of the total life-cycle cost of a project. There could be as much as 80% future costs just in terms of operating and maintaining these. These are big dollars. We have to have the money available to actually keep these projects running and in good order. Once you let them wear down, the costs really spiral to keep them up and running.
Let me shift now to talk about how we procure projects. In this vein, I want to speak specifically about public-private partnerships, which has been the majority of my research.
Public-private partnerships make up generally a small fraction of all infrastructure in the country. They're generally best for very large projects. Maybe 10% to 15% of all infrastructure investment goes through public-private partnerships, but these are also the biggest, highest-profile, costliest projects in our country. Generally the numbers are $50 million to $100 million and up. These are big projects, and there's also a lot of attention on these projects. This topic, then, has become one of great interest in terms of how we can deliver these big projects effectively.
Public-private partnerships have been used right across the country. There have been about 200 of them, maybe a little bit more, that are either done or are in the procurement process.
P3s means a lot of different things, but basically it contains three components. One is the parts of the process you are bundling and providing to the private sector. The private sector can take on anything from design, build, finance, operate, and maintain. That's one aspect, and the various components can vary. As the private sector takes on more functions, it also takes on more responsibility for delivering the project.
The second component of a public-private partnership is risk, and which risks in particular the private sector is going to take on. The big risks are construction risks, potential for cost overruns and delays; availability risk, that the facility is going to work as expected once it's opened; and then finally, demand risk, that the revenues are going to be there as was predicted. These are the types of risks, and they can vary on public-private partnerships.
The final point is the repayment mechanism. With what mechanism are they being repaid? There are only really two. They get repaid either through user fees and direct tolls, or they get repaid through availability payments, and that's direct payments from government. In our country, most projects are the availability payment type of deals, where the government pays them back entirely over the course of the project. What that means, importantly, is that public-private partnerships are not new money.
One of the motivations for using public-private partnerships historically has been to bring in new money. The public sector can tap private money to pay for infrastructure. It turns out that, the way we're doing these projects now—mostly as availability payment-type deals—these are not new money.
Then what are the other reasons you might do this? Really, this is about value for money. Can we leverage the private sector to deliver better projects than government would deliver on its own? That can be things such as innovation being brought forward, or life-cycle maintenance, that you have money actually ring-fenced in the contract to maintain the projects. That's another aspect that's very important. But there's also risk transfer, that the risks of major cost overruns are transferred from government to the private sector. That's another area where public-private partnerships are seen as a real opportunity.
In terms of how these projects have performed, the record on public-private partnerships here in Canada, I would divide public-private partnerships between the first projects that took place up until probably the end of the nineties, early 2000s, and this most recent generation over the last decade or so. In the most recent generation the projects have tended to be built on time and on budget, which is positive, so there's cost certainty. Once the projects are operational we haven't seen any major failures, any major contract renegotiations, or bankruptcies. That's been a concern about public-private partnerships and we haven't seen that.
If I were to try to identify why we're having some level of success with P3s, or public-private partnerships, so far, we tend not to transfer demand risk, which is very hard for the private sector to control. That's better maintained by government. We seek to strategically use private sector finance, so government has recognized that they should use private finance to transfer risk but not as a way to raise new money. Government is ultimately, for most projects, going to pay back almost the entire shot by themselves, so this is not new money.
We tended not to transfer operations to the private sector. These are mostly for design, build, finance, and some of the maintenance of the hard asset. We're not transferring the operations of this service, so this has maintained government flexibility, which I think has been a positive and has meant that our deals are not quite as inflexible or quite as prone to controversy and tensions as in other countries. Finally, we're counting these projects on the books, so this is on-balance sheet investment. This is not an accounting mirage.
While we're seeing successes in public-private partnerships here in Canada, I think there are some points and some outstanding issues we need to consider.
Public-private partnerships are not a cheap way to deliver infrastructure. In fact, up front they're quite expensive. They have higher construction costs. They have higher transaction costs. These are for the lawyers, the accountants, and the advisers to structure these deals. They also have much higher project finance costs. The private sector borrows money at much more expensive rates than what government can borrow at, so they have considerably higher costs. The Office of the Auditor General in Ontario did a study. It found that doing P3s was $8 billion more expensive than if government had delivered the projects directly and effectively managed the risk. That's the key point, to effectively manage risk, then they could have saved $8 billion. That's a real open question, but there is potential for savings there.
On the construction side, studies in Europe have found that public-private partnerships cost up to 25% more in terms of their upfront capital costs, so you're paying a premium. It's like buying an insurance policy against future risks. You're really paying a premium. The problem is that we don't have good data on whether that premium is actually value or not. There's not the evidence on what risks have happened on past projects. There's not the detailed studies of that. I think that's really problematic because, while we can say that public-private partnerships are delivered on time and on budget for the most part, we don't know what that's compared to. How much are we paying for that insurance premium and how much value is that giving? Could government actually deliver that project more effectively and manage the risk, instead of trying to transfer it, because risk transfer comes at a high cost?
There are a few other issues to raise. One is loss of policy flexibility. When you have these long-term contracts, it can pose real problems for government, who needs to make changes to how the facility is used or the rates that are being charged. There are all sorts of other issues. We can lose flexibility. That's caused tension on international projects. We're fairly early on in our experience with public-private partnerships, and so far so good, but we'll have to see down the road how that issue of flexibility comes up.
Another issue I want to talk about quickly is “the only game in town”. Public-private partnerships are one option, but we have to be very careful that we're not setting up structures that make this the only option that's available for, especially, municipalities to access senior-level government funding. This poses the potential issue that we're not using public-private partnerships because they deliver value but really just because we can access money. That can lead to real problems in terms of the incentives and projects being used that are not necessarily the best value. I think it's very important, then, that when we have funding models for delivering money to municipalities, especially, but also provinces, that these are not tied to a specific model. Public-private partnerships are one option for delivering infrastructure, but they need to be used in the ideal setting. We shouldn't be choosing in advance so that governments can access money. That can really lead to potentials of not carrying out accurate studies on the incentives and why we're using public-private partnerships.
A final point is around innovation and design, because we've heard a lot about how these projects are structured and that they drive innovation. The questions are what types of innovation, and innovation for whom. The types of innovation we tend to find are those around construction means and methods, innovations around finding ways to shrink the building, to make them so that they still provide the service but in smaller sizes and lower costs. These are really cost-saving innovations.
When it comes to architecture design and those types of issues for public-private partnerships, the observation is that they tended to be fairly average buildings and not necessarily great architecture or great design. That's not necessarily true in all cases but is a general observation. They haven't won a lot of major architecture awards. They're not necessarily the signature buildings in your community.
Not every building has to be a signature building, necessarily, but these facilities are public infrastructure that's going to be in our communities for decades to come. We need to make sure that the quality of these buildings in terms of architecture and design is there and is at the highest level possible. That's another area—a flag to raise—that we should pay attention to.
To wrap up very quickly, I have a few recommendations. I think we need to be carrying out studies of risk. We need to understand what the value of that risk transfer is. We're paying high premiums up front to transfer risk to the private sector, especially for construction. We don't know if we're getting value for that. Cost certainty is important, but not necessarily at all costs. Government might be able to manage risk rather than just transfer it and thus save money for taxpayers and citizens.
We need to develop a bureaucracy that has the skills to analyze and take part in these projects. These are complicated projects. As part of that, I think we need to be focusing on infrastructure broadly and not necessarily just on public-private partnerships. To that point, I would would say with respect to PPP Canada that I think the organization should be rebranded and made broader. It should be made “Infrastructure Delivery Canada”.
We should be focusing on effectively delivering all infrastructure projects, not necessarily just public-private partnerships, and on having funding mechanisms that focus on effective procurement, not necessarily just on incentivizing public-private partnerships. There are all sorts of innovative types of procurement models that might deliver value for Canadians. We need an agency that pushes effective and innovative procurement, not necessarily just public-private partnerships.
The final point I want to make is that we need to be leveraging information and becoming analytical organizations. Infrastructure and public-private partnerships have a lot of data that come out of them. We should be using that data to systematically evaluate how our projects are performing and to come up with new mechanisms and new tools to make sure the projects we build are performing and that in the future we come up with the mechanisms that are the most effective to deliver projects successfully.
Thank you. I'll leave it there.