Michael Bennon, Managing Director at the Global Projects Center at Stanford University, discusses infrastructure Public-Private Partnerships in an era of compounding deferred maintenance deficits.
Gridium: Hello everyone, and welcome to this conversation with Michael Bennon, Managing Director at the Global Project Center at Stanford University.
Michael served as a Captain in the U.S. Army and the U.S. Army Corps of Engineers for five years leading engineer units, managing projects, and planning for infrastructure development in the United States, Iraq, Afghanistan, and Thailand. His research interests span public sector finance, infrastructure and real estate investment, and project organization design.
My name is Millen, and I’m with Gridium. Buildings use our software to fine-tune operations.
Michael and I will be discussing his research into infrastructure of public-private partnerships–also known as PPPs–and what it means to consider new projects in an era of compounding deferred maintenance deficits.
Firstly, thank you for your service Michael, and also, I’m really excited to be speaking with you today.
Michael: Well, thank you, and it’s great to be on the show.
Gridium: Before we explore your work at Stanford, and some ideas around PPPs–public-private partnerships–I’d like to ask about your boots-on-the-ground experience working on new infrastructure projects, domestically and abroad.
Michael: Yea, well I think for a lot of folks in the infrastructure/development industry and engineering, I think building the new projects is kind of what got them passionate about the industry, for sure, even though a lot of what we have to do in public infrastructure and buildings is about asset management and maintenance.
A lot of what we were working on with new projects… my first deployment to Iraq, part of my job was learning in a crater repair engineering unit, and so for that project we kind of worked primarily in central Iraq, but these units are interesting because essentially when a roadside bomb goes off on kind of a major road essentially part of our job would be–usually late in the evenings–we would go to the site and there would, of course, a big hole in the concrete and we would go and kind of try and repair the road in the evening.
So, it was some pretty rough repair work, but we would kind of spend the evenings doing that work, and I think it kind of… it’s certainly an interesting maintenance situation.
Gridium: Yes.
Michael: Most road maintenance programs don’t have to deal with regular bombings.
Gridium: Right.
Michael: But (Laughs), it really did kind of motivate my unit because a lot of the roads that we would go out and work on, if we weren’t kind of repairing those day in and day out, they would be…
Gridium: Critical.
Michael: …completely bombed out and there were a lot of towns and U.S. Army bases that really depended on those roads. That’s how they got their food and their fuel and kept the economy running.
So, it kind of really hammered home the need for regular maintenance.
Gridium: Indeed. I can only imagine.
Larry Summers, the 71st secretary of the Treasury and the former director of the National Economic Council talks about deferred maintenance as a debt burden and that it compounds at 7% a year.
How should we think about the balance between new infrastructure projects versus repair and maintenance of existing assets?
Michael: Yes, I think that’s a common question in the public infrastructure industry, it’s kind of–for a developed economy, like the United States–how much do we want to weight new infrastructure development versus kind of maintaining our existing infrastructure stock?
And I think, one of the things Larry has certainly pointed out is that in a lot of cases, there’s an even clearer economic argument for maintenance spending for infrastructure, in terms of a return on investment even more so than there is on a lot of new capital projects, which still can obviously benefit the economy in a lot of different ways.
But the costs and benefits to those projects are usually based more on kind of microeconomics or the specific benefits of the projects. So… you know I think, in terms of a balance between maintenance spending and new capital projects, I guess one of my biggest general recommendations is simply that we do a fairly poor job in the United States of incorporating maintenance costs, or a plan for asset management and maintenance into initial project planning.
So, when we plan new infrastructure projects or large public capital investments, in a lot of cases, a lot of that analysis is really focused on capital costs and the costs to build the asset, and not enough focus is on the long-term costs of maintaining that public infrastructure which, of course, if you look in aggregated, a much bigger spending commitment that we’re making when we build a new asset.
Gridium: And when we talk about building a new asset, that’s a good example of where public-private partnerships come into play. So, if we can back up a little bit, what is a PPP, and why are they important to the next wave of development in our cities, and in our nation, and of course, globally as well?
Michael: Yeah. So, the term public-private partnership is used pretty widely as a procurement option for developing new infrastructure.
In a lot of cases, there is a lot of ambiguity about how that term is used and so, it can cause a little bit of confusion in terms of what that procurement model looks like. So, the definition that I like to use for a public-private partnership is a procurement for an infrastructure asset in which, eventually, one contract or one concession is used to cover the full lifecycle of the project.
And so, to break that down more specifically, there’s five kind of key functions that go into any major infrastructure project: it has to be designed, it has to be built or constructed, it has to be financed–so we have to raise the capital to build it–and then it has to be operated and maintained.
And so, for kind of the vast majority of infrastructure that is built through traditional procurement, you know, the government sponsor for the project will either design it with their engineering staff or they might hire an engineering firm to design it.
Then, they’ll kind of… once they have a design and a cost estimate, they’ll have contractors bid for the construction of the asset. So, those are private contractors that are actually building our infrastructure.
They will typically finance it on the state or city balance sheet, or there might be a federal grant.
And then they’ll operate and maintain it either through outside contractors or there might be a public agency that performs a lot of the routine maintenance. So, all of those various tasks are kind of broken up into separate contracts and managed by the kind of public sponsor that is ultimately responsible for the project.
Under a PPP, there’s essentially one concession contract for a private concessionaire or consortium of companies to do all of those tasks. They are going to design it, build it, they are going to finance it at the project level–so it’s financed as a project, not necessarily on the balance sheet of the public sponsor–and then they are going to operate and maintain it in the long term. Typically, the procurement model is used… so PPPs, the main reason they are used is to transfer risk away from the taxpayer and to that private concessionaire.
In terms of the risk transfer, there is kind of three buckets of risk that can occur on a very large, complex project.
It can run way over budget and take too long-and those are correlated. So, it can take much more capital to build the asset. It can go over budget in maintenance, and that can lead to a shorter useful life, deferred maintenance or just increased costs. And then the other risk is for certain projects that are funded by user fees, like a toll road for instance… there can be demand risks. So, not enough drivers to adequately pay off the cost of the road.
And so, that P3 procurement model is really used to transfer some of those risks to the private concessionaire. And because of that you see this procurement model used primarily for very large, complex projects. Not really the smaller budget projects that are the majority of most infrastructure projects.
Gridium: I see, and as you were walking through those risks that are hedged and transferred as part of a PPP, it had me thinking. Is there a maintenance risk associated to a PPP project?
Michael: Definitely, and that’s one of the core risks that can be transferred through the procurement model. Where PPPs for infrastructure or public buildings, other public facilities, where they overlap with maintenance risk is in a few different areas.
So, one is the… there is an opportunity, because it’s a full lifecycle procurement, for that concessionaire to kind of align incentives across the project’s lifecycle because that company is not only building the asset, they have to maintain it for the long run. So, the procurement model kind of opens up opportunities for innovation where a contractor might spend more money on the asset itself even though in a typical procurement, you wouldn’t really see that, right?
The contractor is trying to build it as cheaply as possible and still meet the government’s technical specifications. While in a PPP, you could have a situation in which a contractor would be willing to over-invest because the asset would be cheaper to maintain in the long run. One of the other areas where it overlaps is around deferred maintenance. So, PPPs have performance-based contracts, which basically states that instead of the government coming up with the technical requirements to maintain the asset and then hiring a contractor to implement that plan, the concessionaire is required to maintain those assets at a certain condition, and if they don’t, then the government can actually dock payments for them over the lifecycle.
And typically, what you see with these PPP is they’ll be a longer-term concession from maybe around 20 years to maintain the assets all the way to 99 years of maintenance.
And they’ll be required to have a government inspection to make sure the concessionaire is meeting maintenance and the government’s not on the hook if it doesn’t pass inspection. That’s a risk that the concessionaire takes on. They also have pretty stringent hand-back requirements, so it’ll still be in a good condition when it reaches the end of that maintenance.
Gridium: Fascinating to think about a 99-year long maintenance contract. I mean, it’s a significant amount of work that the parties are agreeing to.
Michael: And a significant amount of risk that can be transferred to the concessionaire. Now at the same time, with a term that long, there could be technological changes that reduce the costs of maintaining an asset, just as kind of one example. The risk can kind of run both ways, right?
Gridium: Yeah.
Michael: But typically, what you’ll see from these contracts, that a 99-year concession contract is probably an extreme example. Typically, you’ll see these contracts range in the kind of 20 to 35-year range.
Gridium: Speaking of risks running both ways, I know that your research has explored some recent examples where the demand levels for public-private partnership funded toll road didn’t meet expectations.
And so my question is around the expectations: is this a trend where PPP projects aren’t meeting expectations? And if so, why is that happening?
Michael: Yeah, well I wouldn’t say it’s a trend because there’s been a lot of studies of this phenomenon; but, in a lot of cases infrastructure assets don’t necessarily meet their projected demands. And part of that is simply a function of the fact that it’s really hard to tell how many people are going to want to drive on a toll road that doesn’t exist yet.
Gridium: (Laughs) Yeah.
Michael: So, usually there’s pretty extensive studies that go into these projects but, in a lot of cases, they end up overestimating demand to some extent. What we’re interested in is several of these public-private partnerships for toll road in which the government transferred revenue risk to the private concessionaire.
So, the private company was on the hook if not enough cars used the road. And then, essentially what happened is several of these roads were financed in the mid-2000s, and when the recession hit, it turned out that people in general, drive a little bit less on toll roads when the economy is in a downturn.
And so, there’s some interesting cases that are kind of coming to fruition, just over the last few years, of a lot of toll roads both developed using traditional financing and public-private partnerships have seen these kind of traffic downturns. Now that’s subsided, but during the recession there was some kind of reduced demand for a lot of toll roads in the United States, so it’s an interesting risk to look at for a case study.
Gridium: Indeed. And it wasn’t until I started to dig into your research, Michael, that I noticed the link between my introductory finance classes at university and public-private partnerships. So, curious and interested to know your thoughts on the link between discounted cashflow analysis, and Black-Sholes optionality models, and what they have to do with infrastructure projects.
Michael: Yeah, it’s actually a pretty, kind of heavy modeling industry, I’d say, in terms of infrastructure investment generally.
I talked a little bit about some of these demand models for new toll roads, for instance.
But even more kind of the less-risky projects have pretty complex financial models and the reason for that, in part, is because these are very long-term concessions. And so typically, when there’s a need for increased analysis, both on the part of the government, and then, of course, on the part of the private contractors that are bidding for the project.
And part of that is just due to the nature of the fact that it isn’t a short-term construction contract. This is a potentially a 20-year maintenance agreement. There’s a greater need for scrutiny on the project-both from the company that could be financing it and actually investing capital into the project. And then also on the public side, there’s an increased need for transparency. And if a government’s going to enter into a longer-term contract-and these large infrastructure projects are usually very complex–there’s a greater need for kind of an analysis to look at whether this is creating value for the taxpayer, right?
And so, you do have a lot of analysis that goes into these projects at the outset when this procurement model’s used, and that can actually be a good thing for incorporating risk assessment through the discounted cash flow process and risk assessment models to make sure that the government’s getting a good deal by entering into one of these contracts. So, there’s a lot more analysis that goes into it.
Gridium: And as a citizen, I’m curious what your thoughts are on the best practices our audience should think about when they consider revenue risk, and concessions made to develop and reason an overall project selection.
Michael: Yea, so… this is kind of a relatively newer area of study in terms of kind of what makes a good public-private partnership structure and a good PPP assessment process. But if I could probably pick out kind of a higher-level rule of thumb, it’s that if you’re looking at it from the perspective of a public sponsor, a government that’s considering its options for meeting a need, it’s really about risk transfer. And so, some of the higher-level questions around that:
Is the capital cost of this project greater than $100 million, which is a pretty large infrastructure project? If it’s below that, maybe the amount of risk that the government can transfer to a private partner is simply going to be lower. Has the public agency implemented a project like this before? Is this similar to other projects they’ve done in the past, or is this going to be completely unique, and thus, kind of entail more risk?
And then, some of the other factors, it’s kind of similar to what I eluded to early, but: are there opportunities for innovation over the lifecycle of the project based on the technologies or just how the project’s structured?
And then I would say another factor that governments should consider is really, is there a need for flexibility? ‘Cause that’s another factor, as I mentioned, with a PPP procurement, it’s a long-term contract. And so, in certain situations, for certain types of infrastructure projects, the public sponsor might need additional flexibility. And so, maybe a longer-term contract simply couldn’t work in that case.
So, those are a few of the factors, but I would say the biggest one is really the amount of risk that can be transferred through the procurement.
Gridium: It’s no wonder to me that Stanford has a research initiative dedicated to this subject. Can you tell us a little bit about the Global Project Centre, and also its P3 FLIPS Program?
Michael: Sure. Our center at Stanford, so we do kind of research on innovation within infrastructure development and investment, kind of at large. So, it’s an interdisciplinary program, meaning we can kind of pull faculty from multiple different parts at the university and if infrastructure development is one that really looks across the different silos at a university, like the School of Engineering and the Business School and Public Policy.
And so, we’ve kind of–for the most part–done research on kind of innovative models for infrastructure development, and that’s part of the reason why we’ve more work on these public-private partnerships, because they’re a relatively new model for infrastructure procurement, at least in the United States. Now, they’ve been used kind of globally for some time now.
And then we also do research on kind of “infrastructure allocation.” So, what you’ve seen over the last decade is institutional investors, like a public pension fund, have created allocations to invest some of their capital into infrastructure alongside the other parts of their portfolio, like stocks and bonds. And so, we’ve done some research on the performance of those investments as well, and the risks and the term and characteristic of the asset class.
Gridium: Yeah, it’s really fascinating stuff. And it couldn’t be more applicable in today’s market and in our society today.
For example, New York City’s subway and train system is experiencing, well for the lack of a better word, a maintenance crisis, to the point where the major is now seeking $500 million in annual revenues to pay for the fixes.
How do you think New York City got into this mess?
Michael: Well, I think there are probably multiple contributing factors that go into the one situation like the one that New York City is experiencing and that a lot of public transit agencies in the United States are experiencing.
You know, one component of this is that it’s quite old, right? So, this is a system that’s been around for a while and you’d expect there to be kind of incremental maintenance expenditures over time as it starts to reach the end of its useful life.
Gridium: Good point. (Laughs)
Michael: But I would say there’s probably something more there in a system like the MTA, in that there’s probably a considerable amount of deferred maintenance.
So, you know, I think what a lot of the studies that have been done-and most of the studies that have been done on this topic have been kind of in the road sector–but it’s pretty generally applicable that if maintenance is deferred initially, that those costs compound over time.
And it’s very hard to measure exactly how badly those costs are compounding until you get in a situation like the New York MTA is currently in where they are having kind of a maintenance crisis. And so, what you see happening in a lot of cases in the United States is that maintenance is deferred over time for public infrastructure and then that drastically shortens the useful life of a lot of our public assets which can lead to some of these problems. And that probably is a factor or is part of a cause for these crises like you see in New York.
Gridium: Michael, what do you think it will take for the country to start working down our ballooning deferred maintenance deficits? I think also, you’ve done some work around the accounting standards for deferred maintenance?
Michael: Yes, we… this is, it’s actually a study we’re currently working on in and around deferred maintenance and reporting. It’s pretty specific to the United States.
So, part of this is specifically looking at deferred maintenance in public buildings and that component of the study is really looking at that interest rate that I mentioned early, right? How much does deferred maintenance in various asset classes compound over time? And the other component of that study that we’re really interested in is related to infrastructure maintenance accounting standards for U.S. states and cities.
Currently, in the United States, a lot of our public infrastructure is accounted for using the straight-line method. And when I say accounted for, I mean on the city’s financial statements, they kind of just assume that infrastructure has a certain useful life and that it’s depreciating at a certain rate.
And kind of, the second method that some cities and states use in the United States–but I would say it’s fairly rare in general–is where the government will actually try to come up with a technical estimate of the costs that it would take to bring that system back to kind of a fully-functioning status, right? So, they’ll do an engineering estimate of the cost that it would take to kind of rebuild that asset to make it functional again, and they’ll list that amount (cost) as the total amount of deferred maintenance on their balance sheet.
And so, we’re interested in this topic and for kind of that model of accounting to hopefully propagate a bit in the United States and I think that would obviously have several benefits. One, of course, is simply that taxpayers should be able to understand exactly how much the debt burden of their deferred maintenance is for their public assets, and it’s hard for them to get a picture of that currently.
Gridium: While it’s hard to get a picture of that currently, there certainly doesn’t seem to be a lack of interest, at least on infrastructure, in our nation today. So, it must mean that you and your team are super busy. Is there, besides this accounting standards work, anything else that you and your team are particularly excited about over the next year or so?
Michael: Yeah, no. We’ve been doing some research that kind of relates to a few of the policy proposals that have come out of the new administration. I think the jury’s still out a bit in terms of what’ll actually be implemented in terms of U.S. Federal Infrastructure Policy.
One of the programs that we’re currently doing some research on is what’s generally referred to as an Asset Recycling Program, and the current administration has kind of floated this idea generally, providing some incentive for states to enter into such a program. And so, probably the most notable Asset Recycling Program that’s actually been implemented is by the Australian Federal government, and then in particular, in New South Wales.
What the state did, in this case, is they created a new agency that came up with a 20-year infrastructure strategy for the state. And part of the reason they did that is because they wanted this kind of central agency to kind of look across all of their needs; water, transportation, energy.
They came up with almost like a prioritization based on economic studies of kind of the top projects for the state and they rank-ordered all of them. And then, what the state did, is they procured concessions on some of their existing infrastructure. So, they would kind of procure a long-term contract to maintain existing economic infrastructure.
The example that is commonly cited in the U.S. is the Indiana toll road: the state of Indiana sold a concession on a toll road. And then this agency in Australian then uses that capital that comes from those concessions and reinvests it in that prioritized list of needs. So, they reinvest the capital in new projects. So, that’s one kind of interesting policy that we are doing more research on, just to look at the kind of costs and benefits of a program like that.
And then, I’d say the other topic that we’re interested in from a federal policy perspective, is to look at federal spending and the incentives that it creates for the state and local political leaders in developing new infrastructure and then maintaining existing infrastructure. I think what’s interesting about the United States that’s really unique is that our federal government spends a lot more on infrastructure relative to state local governments than what you might see in another large federal democracy in a developed economy like Australia or Canada.
So, our federal government spends more, and our federal government really focuses a lot of that spending on new projects. So, a lot of their spending goes to new infrastructure projects as opposed to maintenance.
And it’s declined a lot, but it’s still a pretty big component. So, back in the 1950s and the 1960s, federal spending was significantly higher than it is today; but even today, it’s still a pretty large component of total infrastructure spending. And that’s kind of interesting because there’s always benefits to federal spending on infrastructure, and it’s justified generally because there’s kind of economic spillovers from these projects, that, even though most of the benefits go to the city or state in which the project’s located, there’s benefits generally.
And what’s interesting though is that most projects–the organizations actually responsible for getting projects done and planned, and maintaining those infrastructure projects–that’s state and local government.
When you have kind of one level of government that is investing a lot in new capital costs and kind of another level of government that’s responsible for putting projects together, but then also kind of on the hook for long-term maintenance, that can create bifurcated incentives in infrastructure investment generally. And I think even though most infrastructure funding is raised by state and local governments, federal policy has a huge impact on the incentives of kind of state and local political leaders and managers of an infrastructure agency. And so I think those incentives are interesting, specifically in the United States.
Gridium: Thanks again Michael. I want to say that we really appreciate you taking the time to chat with us. This has been fascinating, and it’s been great to dig deeper into public-private partnerships and the role that maintenance plays in our infrastructure.
So, thank you.
Michael: Yeah, thanks so much for having me.
