Is Infrastructure a Public Good? No, Sort Of, and What Role for the Public and Private Sectors

by Randall Bartlett

Economists like to talk a lot about public goods. But depending on which side of the political fence you fall on, what is considered a public good may differ. This is because many goods and services have some or none of the characteristics that define public goods but may offer significant additional benefits beyond their private return (known as positive externalities). Infrastructure falls into this category, along with education and health care. Hence why the debate as to the role of the public sector in these areas continues unabated and differs so much across countries.

This debate has resurfaced recently as the federal government seeks to create a new institution for funding infrastructure investment––the Canada Infrastructure Bank (CIB). In Canada, it is governments at all levels that tend to hold the primary role in the planning, building (i.e. as sponsor), managing and financing of infrastructure. Other than in the construction phase, the private sector is a relatively niche player in the lifecycle of infrastructure assets. The CIB aims to be a catalyst for including the private sector in the other aspects of the lifecycle. The untested presumption is that service levels, externalities and competition will remain constant (at least) while risk will be transferred from the public to the private sector. However, this has rarely been the case with public-private partnerships (PPPs)––the oft-cited private-sector solution to public-sector infrastructure folly. Indeed, with the private sector owning and managing infrastructure assets, the cost of capital is likely go up.

With this in mind, the questions of ‘Why the CIB?’ ‘Why now?’ and ‘Why the rush?’ remain unanswered. Without being able to clearly and credibly address these questions, it is not possible to discern whether the CIB will be able to deliver value for money to taxpayers. And it is ultimately value for money which should matter to policymakers and taxpayers alike.

What is a public good?

As economists traditionally define it, a pure public good has two characteristics: it is both non-excludable and non-rival. Non-excludable is exactly as it suggests––it is not possible to exclude people from using a good or service even if they don’t pay (known as freeriding). Non-rival simply means that one person’s use is not diminished when other people use it. The classic example of a pure public good is the defense of national borders. If you pay for the defense of a country’s borders and I don’t, we benefit equally if we both live in that country. It is therefore non-excludable. Moreover, it also non-rival, in that my being defended does not diminish the level of defense you receive. Hence, the defense of national borders is a pure public good.

Unfortunately, examples of pure public goods are few and far between. Instead, we are generally faced with goods and services that have only one of the necessary characteristics. A lecture hall is a good example, as it easily excludable (one can close and lock the doors) but adding one additional student is unlikely to reduce the quality of the experience for other students, and so it is non-rival. Health care is similarly conflicted, and infrastructure can be as well.

However, some goods and services that are often considered to be public goods have neither of the attributes that characterize public goods. Instead, they are incorrectly referred to as public goods because they have positive externalities. Externalities are the (often unintended) consequences of an activity that effect other parties but are not reflected in the price. Negative externalities are generally the most discussed, with common examples being things like pollution or carbon emissions that compound climate change. Positive externalities are the benefits to society from a private activity that aren’t priced into the good or service itself. Again, public education comes to mind. On average, better-educated people tend to not only have higher incomes (a private benefit), but also better health, are less likely to commit crimes, have more successful marriages, etc. They also tend to raise children that have these same attributes. Therefore, there are benefits to society beyond just the private return to education for the individual. Health care can be similarly considered, although the benefits are often less immediately obvious. Hence the ongoing debate south of the border.

Is infrastructure a public good?

Infrastructure certainly does not fall into the category of a pure public good. A toll road is clearly excludable. And, if the toll is set below the price that would be determined in the market, congestion could still occur, meaning it could also be rival (although one would expect the owner of road to raise the toll if this were to be the case). This is, of course, often an issue on public roads where there is no immediate cost of use. All roads or parts of roads could also be privatized, meaning tolls could be ubiquitous for use. This is more generally the case for airports, ports, railways, international and some interprovincial bridges, etc., in the form of user fees.

However, the private ownership of infrastructure can also lead to problems. For instance, ownership of the single route between two places creates a natural monopoly, as the cost of building alternative routes is likely to be high. Take the section of road directly in front and adjacent to your house, for example. To get access to your house, you need to use that section of road, as there is no other way to access your house (except maybe by air but at a significant expense). As such, you would be willing to pay a great deal for the use of that section of road. Armed with that knowledge, the owner of the road can charge a toll that is well above where it would be in a competitive market, thereby making a monopoly profit. The same is likely to be true for the owners of each section of road. And not only would the tolls be high, but the inefficiency of endless tolls on all roads would be steep, with all of these additional costs being passed on to consumers in one way or another.

As a result, most roads are owned and maintained by the public, not only in Canada but in the majority advanced economies. This is particularly true as it pertains to main thoroughfares, as it is generally recognized that ready access to highway infrastructure has enormous positive externalities for economies. And these extend to infrastructure asset types beyond just roads and highways. The April 2015 Australian Infrastructure Audit succinctly captures the positive externalities of infrastructure: “Productive, sustainable infrastructure is essential if we are to drive economic growth, increase employment and enhance the quality of life of all Australians.” Indeed, the greater the positive externality – the benefit to the broader public beyond just the private benefit – the greater should be the effort to avoid monopoly ownership and pricing.

What role for privately-owned infrastructure and Canada Infrastructure Bank?

Some have stated that private sector involvement in infrastructure is a good thing, often citing public-private partnerships (PPPs) as the primary example. In theory, this is absolutely the case. While private sector financing costs for infrastructure are generally higher than for the public sector, the benefits of the increased likelihood of projects being completed on time and on budget, as well as the risk transfer from the public to private sector, are thought to outweigh this additional cost. However, in practice, PPPs have frequently fallen short of this ideal in Canada. This is because risk is generally not transferred from the public to private sector, meaning the pricing of the asset is out of step with the actual amount of risk that is assumed by the private owner. As a result, taxpayers often do not get the value for money they had anticipated. For this reason, the jury remains out on whether PPPs are effective in lowering the overall cost of infrastructure to taxpayers in the long run. That said, many people remain optimistic that the success of PPPs is only a matter of negotiating the right deals on the right projects, which may ultimately prove to be the case.

This brings us to the proposed Canada Infrastructure Bank. As was discussed in a recent IFSD blogpost by Azfar Ali Khan and me, the value-for-money proposition posed by the CIB remains a mystery as investments are not a guarantee of realizing outcomes or positive externalities. If it is meant to help fill an infrastructure gap, certainly we should know how big that gap is and in which regions and types of infrastructure before throwing money at a problem that we aren’t sure even exists. If, instead, the CIB is meant to fill a gap in the financing of infrastructure, surely this is not an issue. As Azfar and I pointed out in our recent note, with the lowest net debt-to-GDP ratio in the G7, Canada has room to run when it comes to borrowing (increasing liabilities) to invest in infrastructure (a non-financial asset). This would raise the level of the general government net debt, but would like still leave us in a better spot than Germany, our closest net debt-to GDP neighbour in the G7. Further, the fact that the cost of borrowing to finance investment is lower for the federal government than the private sector is beyond dispute. That’s not to say that an infrastructure bank of this nature wouldn’t be effective in helping to finance investment in more cash-strapped jurisdictions, such as municipalities, U.S. states, or emerging markets. It’s just that this case cannot be made in Canada.

Other benefits from private sector involvement in infrastructure investment should also be considered. Could it be the expertise that exist in the private but not public sector? Maybe, but most of this infrastructure supported by the CIB is going to be new or ‘greenfield’, an area where Canadian institutional investors have limited experience when compared to their foreign contemporaries. Is it possible that bringing institutional investors on board will increase the likelihood that projects will be completed on time and on budget? Maybe, but we can’t be certain, particularly as these are ‘greenfield’ projects where the federal government will continue to have ‘skin in the game’, meaning a lot of the benefits traditionally associated with private-sector involvement may not materialize. Is it the transfer of risk from the public to the private sector that is the primary benefit? Well, if PPPs provide a useful comparison in the Canadian context, then the answer is probably no, as the public will very likely continue to be exposed to much of the assets’ demand and revenue risk.

This brings us to the key question surrounding the establishment of the CIB: If much of the risk is likely to continue to be borne by the public sector and private-sector financing costs are higher, why would we allow the private sector to have majority ownership of assets that typically allow for monopoly or near-monopoly pricing? Does this suggest that taxpayers will ultimately be paying less for these assets, from beginning to end, if the private sector is involved? Maybe, if these deals are structured properly, but Canada’s history with PPPs suggests otherwise.


To conclude, infrastructure does not satisfy the definition of a pure public good, nor does it necessarily have any of the attributes of a public good. Instead, infrastructure creates benefits to society and the economy (positive externalities) beyond just the private returns that come from its ownership, much in the same manner as education or health care. And given the monopoly-like characteristics of infrastructure, which can lead to pricing which is in excess of what would be reached in a competitive market, there is arguably a role for government to play in participating in the market for, and ownership of, infrastructure.

In theory, there are benefits to private-sector involvement in the designing, building, financing, operating, and maintaining of infrastructure for public use. However, in practice, these benefits have generally fallen short of expectations. Hence, the jury remains out on whether PPPs are actually effective in reaching their stated goals and delivering value for money to taxpayers.

Regarding the Canada Infrastructure Bank, the case for its existence is even less clear than that of PPPs. Is it meant to help increase infrastructure spending in Canada? Sure, but to fill a gap we don’t know with certainty exists. Is it meant to support a financing need? Well, no. Is it meant to increase the overall benefits, or positive externalities, that come from infrastructure investment? Absolutely. But then the question becomes: How will the CIB do this by increasing private ownership of infrastructure assets characterized by near-monopoly pricing power? This question remains unanswered. Does this rule out a role for the CIB? No, but it certainly begs the questions: Why the CIB? Why now? And, why the rush?