Do PPPs have a future?

The below article, which has been cross-posted from The Conversation, examines the future of toll road Public-Private Partnerships (PPPs) in Australia. The author is Dr Stephen King, who is an Economics Professor at Monash University.

BrisConections has been placed into administration only seven months after opening the Brisbane Airport Link toll road/tunnel. It has not had sufficient users to make the project viable. So what does this mean for future public-private partnerships (PPPs)?

In the short term, it will mean very little. The citizens of Brisbane have a great tunnel that (from my experience) cuts significant time off a trip to the airport. The investors have done their dough. And there may be various lawsuits about who misled whom.

However, this is the fourth in a series of PPP toll road failures, including Sydney’s Lane Cove and Cross City tunnels, and Brisbane’s Clem7. If PPPs are to have a future, we need better ways to handle the project risk.

The risk associated with large infrastructure projects can be significant. For toll roads, the viability of a project depends on projections of future traffic flows. But these flows may be highly variable, depending on a range of choices by the government and car users.

Under a traditional PPP contract, much of this risk is directly borne by the private investors. However, this risk will be reflected in the contract that underpins the PPP. So the risk will be indirectly borne by car users and taxpayers.

Most obviously, the greater the risk, the higher the tolls that will be demanded by the private participants in the PPP. So car users bear the risk of the project through high toll charges. This can undermine the social benefits of the toll road. Instead of taking traffic off congested suburban roads, high tolls may mean too few cars use the toll road.

More subtly, car users may bear the risk through limits placed on the government. The PPP contract may restrict future government transport policies that would alter traffic flows – even if these policies were in the publics’ best interest. If the government wants to implement these policies in the future then it will need to renegotiate the PPP contract. This can be a messy and costly process, meaning that desirable transport policies are left in the ‘too hard’ basket.

Taxpayers may also bear the risk of a PPP through guarantees on revenue or via ‘take or pay’ contracts that guarantee a flow of government funds.

In the extreme, taxpayers bear the risk through the potential for a government bail out. If the government decides that a PPP can’t be allowed to fail for political reasons, taxpayer funds may be used to protect private investors.

If car users and taxpayers are going to bear the risk of a PPP toll road, what is the point of using private funding? The government can fund a PPP and directly bear the project risk, even if it is built and operated by the private sector. And government funding is currently significantly cheaper than private funding. Indeed, as Michael Pascoe notes in the Age:

“Australian governments can borrow more cheaply than the private sector to invest in infrastructure. The federal government in particular can borrow extremely cheaply on very long terms”.

Unfortunately, this option for improving the nation’s infrastructure appears to be off the agenda. The current ‘budget surplus’ fetish means that long-term government borrowing and investment in public projects is ruled out on the grounds of short-term political pragmatism.

So, if we want on-going investment in public infrastructure, we need better PPPs that handle the risk in clever ways. One alternative, being investigated by Melbourne University’s Centre for Market Design, is to provide the private investors with a fixed return in current dollar terms.

Rather than specifying a length of time for the toll operator to charge road users, this approach allows the private operator to charge tolls until it receives a certain amount of money. This shares the risk between the private operator and the car users. If traffic volumes are high, the private operator will get their return quickly and the road will move back into government hands sooner than expected. If traffic volumes are low, the private operator will have a longer time to get their return.

Such an approach to a PPP will not save private investors when traffic flows are so poorly predicted that they can never get their money back. But it does protect them from short-term fluctuations.

The approach also improves flexibility over future government policies. To the degree that government policies change traffic flows, the private operator is protected. Changed traffic flows automatically change the length of time for the payback to the private investors. The PPP contract will only need to be renegotiated if the changes in traffic flows are so significant that the private operator cannot ever receive the full return on their investment.

The failure of BrisConnections does not spell the end of PPPs. If we want infrastructure investment and government budget surpluses, then PPPs are a must. But it does spell the end of naïve PPPs and it signals the need for research in order to design better PPPs. That is, unless we can find some more private sector bunnies who are happy to lose their money building roads for the rest of us.


Leith van Onselen
Latest posts by Leith van Onselen (see all)


  1. The fundamental weakness with most PPP projects is that they are not viable on the basis of user charges alone.

    As a consequence the ‘projected’ usage figures are tricked up to get the projects happening and there is usual some clause that provides that once the private investors are wiped out the taxpayer picks up the tab.

    Who benefits in all this?

    The shiny suits clipping fees in setting up the project and the construction companies.

    If a project is worthwhile and has economic benefits (adds value to the community) it should proceed with the costs of doing so recovered from:

    1. The users

    2. Land owners

    But also more importantly the people living in the region that benefits.

    In some circumstances this will be land along the route or services by the project and in other cases of state significance all land in the state.

    See where I am going?

    Yep – recover a large chunk of the cost from those who benefit from taxing the land that benefits.

    A higher rate impost for 20-30 years should do the trick.

    Fund the project using government borrowing and outsource the building and project management to those with the competency and lowest charge.

    • Re: “A higher rate impost for 20-30 years should do the trick”:

      That gets us into arguments about what locations benefit from the project and how much of the change in land values is due to the project, and how much to other causes. But there’s no need for any of that. If the government responsible for a particular type of infrastructure claws back, through the tax system, a certain fraction of EVERY uplift in land value, then infrastructure projects of that type whose cost/benefit ratios do not exceed that fraction will automatically pay for themselves through the ensuing uplifts in land values. The government sets the tax rules, and the market decides who pays how much for each project.

      • Excellent point!

        That would avoid the predictable debate about who actually benefited from a project.

        If their land value goes up – there is the answer.

        When the land values goes up – the tax receipts increase and the cost of the project is recovered.

        Needless to say dud projects with no real value still need to be avoided as they will incur a cost but not produce the tax receipts from increased land values.

        • Agree completely, Pfh007 and Gavin. This is one of the many things in favour of land taxes, period.

          Even the value of good schools capitalising into property values, gets tax revenue captured.

          The transfers of wealth to property owners that occur via spatially-based spending of taxpayers money, is one of the biggest rackets in economies and possibly the least understood.

          “…..Needless to say dud projects with no real value still need to be avoided as they will incur a cost but not produce the tax receipts from increased land values…..”

          Commuter rail projects are the worst of the lot – there are vested interests who stand to capture almost all of the “benefit” while paying little towards the total cost, which gets lumbered onto the nation’s taxpayers and motorists. Benefit-cost ratio might be appalling, like $1 benefit to every $4 cost; but if the people gaining the $1 are only paying in 10 cents (with the other $3.90 picked up by the nation’s income tax payers and petrol tax payers), you can bet they will be backing the pro-rail activists to the hilt via the back pocket.

          Road spending has far better benefit cost ratios and does not concentrate the benefit. TOTAL “economic rent” generated is higher, but far more dispersed so that the myriad owners of individual sites have little incentive for pork barrel shenanigans. Sure the developers of a new fringe suburb will “benefit” from having a road provided compared to if there was none, but the economy as a whole suffers drastically if this small-minded thinking is used as an excuse NOT to do these systemic rent-thinning urban expansions.

        • One argument for being a specific about the land that gets charged for a project is that when the tax is levied on a property the land owner can be provided with a list of the projects funded by the tax.

          It could be divided into three parts


          That way the people out in the bush can see they are not being hit for the costs of building a subway in the eastern suburbs but they are contributing to the highway, dam, power grid that provides benefits to the entire state.

          The might be less opposition to a land tax if people can clearly see the connection between the tax and the amenity of their land and thus its value.

          Plus the constant reminder of dud projects on the tax bill will cause everyone to take new proposals seriously.

          If the NBN involved a levy of $X on each land tax bill for the next 40 years people might take more seriously the idea of extending fast broadband only to those houses they currently don’t have access to it or new houses and paying a lower NBN charge on their land tax.

  2. Excellent logic, Pfh007.

    A further tidbit; to your rhetorical question regarding cui bono?

    Response: the forecasting consultancies responsible for the projections of ‘user numbers’, e.g. the consistent ‘over guessers’ Arup, who get paid upfront, with absolutely no clawback provisions.

    They are analogous to the credit rating agencies, who rated as ‘AAA’ all manner of faecal matter in the credit boom of the last decade. NB some of these, e.g. S&P, are being sued for their egregious behaviour.

    Rhetorical Question #2: Why hasn’t Arup, by the hapless infrastructure investors?

    • I know someone in specialist transport consultancy who is really, really strong on this subject.

      Besides consultants who get paid regardless, there is also the people who earn fees selling the bonds.

    • Ultimately they just produce the modelling tools with a massive disclaimer. The financing consortium teak the models – just change the projected gap between cars a smidge and … 😉 whadayaknow.

    • Don’t forget the pollies who pick up cosy gigs after they retire as ‘special consultants’ to banks etc who structure these deals and make millions in fees.

  3. Umm… don’t we all bear the risk for large projects (and for all investment generally)?

    All those thousands of workers and machines could have been building something else society values. But they didn’t. So we all lose.

    All the other details are simply rent-seeking games between egos with more money than sense.

    • Yes – that is a very good point.

      Allowing public investment decisions to be limited to those that can packaged by shiny suite into PPP ‘vehicles’ means that:

      1. Other more important public investments that could generate much greater value are not considered.

      2. The fact that we all pay for mal-investments is concealed behind the smoke and mirrors of private-public partnerships.

      • I am actually somewhat pleased when local taxpayers end up getting a road partly paid for by suckered bondholders. But the investment fraternity surely should wise up after a while.

        I say let the PPP’s rip meanwhile, as long as it is bondholders who are going to be suckered, not the taxpayer.

        • And therein lies the problem. There’s now talk of government-guaranteed PPPs.

          Of the models identified, the partial capital contribution model represents minimal intervention.


          The next level of intervention would see
          government consider forms of risk sharing and guarantees for specific projects. The more material variations to the standard
          PPP finance structure relate to government directly investing equity or debt and would need to be carefully evaluated against clear criteria.

          • That was a bit messed up. Quote is as follows:

            “Of the models identified, the partial capital contribution model represents minimal intervention.

            The next level of intervention would see government consider forms of risk sharing and guarantees for specific projects.

            The more material variations to the standard PPP finance structure relate to government directly investing equity or debt and would need to be carefully evaluated against clear criteria.”

  4. WRT to PPP’s the first question in my mind is always “cui bono”.

    When the whole deal is done it is always easy to identify who the mark was, the real trick in gambles like this is identifying the mark before the game starts or worst case before the losses start to pile up. As they say in poker if you have not identified the mark within the first 3 rounds then assume it’s you and cut your losses right then and there.

    IMHO PPP’s are silly financial structures for everyone except the developer of the structure.

    • Wasn’t this Macquarie’s whole business model for quite a while? Create a new investment structure to be the mark, spin the investment to the market and cream off the fees.

      • Definitely a silver doughnut model but I was thinking more along the lines of failed products from the mini-M’s (babcock etc). If I tried to do this sort of deal on George st. the police would arrest me within a day, yet the same shenanigans involving $100M with a PPP and I get politicians shaking my hand, with full media releases. go figure.

    • Austin, Texas, actually has a complex system whereby individual sites are assessed a special “road use” tax as part of their local property tax bill.

      It is complex but it seems to be working. The success is probably because when spread over everybody, it does not have to be a particularly high tax. It is not really worth anyone’s while to dispute their assessment, which is generally fair.

  5. PPPs are crap. However, the author should take a look into the process by which the demand forecasts were modelled before the projects were commissioned. Choice models by experts clearly predicted these schemes had insufficient willingness to pay on the part of potential users. Try speaking to the academics who were eliminated in the RFT process because their demand forecasts were “too low” for the powers that be (i.e. correct).

  6. Do they have a future? Hell yeah – romancing fees and cash out of disinterested fund managers and gormless politicians is like shooting fish in a barrell.

    Big big future – muddies at the wharf anyone?

  7. It’s not entirely clear to me why this is a problem, or at least why the PPP model is the problem.

    If the problem is that the tollroad was a bad bit of infrastructure that should never have been built at any price, isn’t that just a normal government cock-up which doesn’t really have anything to with the PPP model?

    If the problem is that the tollroad shouldn’t have been built because demand was insufficient, surely it’s not expected that doing it through a PPP prevents this from ever happening. That seems horribly naive. Isn’t the point that a commercial analysis mitigates the risk, and that selling the risk is part of making that process meaningful and real.

    If the problem is that not as many people use the tollroad as expected, well I’m not sure what the problem with that is. The reality of societal infrastructure is that some bits of it get more use than others, but it operates in aggregate and has to exist in aggregate. Accurate usage forecasts might be important for the financiers, but I’m not sure why it’s that important for society.

    If the problem is that investors lost money, isn’t that the whole point? The investors bid, if they get it right, they make money, if they get it wrong they lose money, and if they all submit low enough bids the project gets canned.

    In this case it seems the model has worked perfectly. We have a nice new tollroad, which people are using, and the government didn’t get burnt

    • Investors that lose money are generally taxpayers and those stupid enough to have their pension in a big fund. If you could see how much the consortium pulls in fees and exec salaries along the way you might not feel so balanced.

      The modelling on the lane cove tunnel was an insider joke before a shovel was lifted.

  8. Roads, bridges, dams, tunnels ie essential infrastructure is the domain of government. They are intergenerational in character and governments can raise the funding via the bond markets and service the debt via recurrent income and direct revenue from the investment. The notion that somehow a single generation should pay for a 30 -50 year lifespan investment is just plain ridiculous. The Snowy scheme would never have happened if it left up to a PPP, nor the Sydney harbour bridge etc. This is an entirely appropriate type of investment that governments should be involved in and to use long term debt instruments to fund them.

    • Hear hear. And we are long overdue for a significant amount of spending on infrastructure. Time for our governments to get busy.

      • Yep. But it does not sit easily with the prevailing ideology that governments should be involved as little as possible in the running of their states and nations. Much better in their veiw to hand over responsability for running basic civil services to for-profit entities while still paying for it with public money.

  9. All these project bankruptcies look all too frequent to be ‘mistakes’ IMHO.

    $700 million airport link collapsed just six months after its opening, The The Cross City Tunnel collapsed just 16 months after opening. The Lane Cove Tunnel was sold for 57c in the dollar to Transurban barely a year after it opened.

    The Lane Cove tunnel is an example of corporate ineptitude at work. The route it replaced, Epping hwy, had a steady 55,000 cars per day using it year after year. For some illogical reason traffic was projected to almost double to 100,000 within 2 years of the tunnel opening. When this never happened the project went bankrupt despite it’s concession lasting until 2037.

    I seriously can’t see how any right-headed investor would sign off on those kinds of projections…. it’s a bit too suspicious to continuously be just ‘bad judgement’

    With entities like the Abu Dhabi’s sovereign wealth fund and Canada Pension funds owning significant shares of the revenue streams from these projects PPPs seem to be a great vehicle to export revenue from local infrastructure to overseas owners.

    • I seriously can’t see how any right-headed investor would sign off on those kinds of projections…. it’s a bit too suspicious to continuously be just ‘bad judgement’

      because those who sign off, it’s not their money. They are Australian-quality execs who gouge in terms of salary and fees, then ride off in the sunset leaving the bond holder and equity holder carrying the can.

      We know this is the probable outcome, agency theory says so.

    • This phenomenon is all part of a near total failure in mainstream economics and transport planning, to understand urban land economics, economic land rent, and the way the transport system interacts with and affects these things.

      I agree completely with pessimists who say that most roads, even those with a good benefit-cost ratio from the point of view of the local economy, will not pay for themselves via tolls on drivers. There is simply too much “benefit” that is not captured by the drivers themselves, whenever there are “free” alternative routes that experience reduced congestion. Much congestion on urban highways is caused by motorists who use it for only a segment of its length. These people are absolute “naturals” to use the existing secondary roads when the new highway itself is tolled.

      Central to this, is the failure on the part of the greatest number of people, to understand the dispersion of urban employment and other reasons for trips.

      But NOTE, a new bridge or a tunnel that dramatically shortens a trip, is a different story altogether. The eventual driver demand for these is consistently UNDER-forecasted. These things really DO “induce traffic”. The writings of Cesare Marchetti are interesting on this.

      • “But NOTE, a new bridge or a tunnel that dramatically shortens a trip, is a different story altogether.”

        I think the two City Link tunnels in Melbourne are good examples of this. Now you can (usually) drive under the centre of Melbourne in 5-10 minutes when it used to take up to an hour.

  10. Thanks for this excellent analysis – where on earth did they get such optimistic traffic projections? I’ve heard that there’s also been issues with Eastlink (here in Melbourne) and that their projections were way off.

    My concern is that the East-West tollway mooted to connect the Eastern Fwy with Citylink (and/or the Western Ring Road) here in Victoria is going to be a financial black hole and that our state government will underwrite their profits/projections in order to get extra private money. As you rightly point out, major infrastructure projects can be really cheap as the Commonwealth has a significant borrowing capacity but there’s an unwillingness to use this credit.

    Is there any way major projects can be completed without these complicated financial vehicles?

  11. I refuse to use toll roads. Every time I purchase petrol the government takes a chunk of money which is supposed to be used for funding roads. Why should I pay twice to use the same road? Why should profits some of these toll roads generate go into private hands?

    • Why should profits some of these toll roads generate go into private hands?

      Rent seekers love monopolies.

  12. Can you let me know what is the typical expected repayment period pitched to investors for these projects ? It would seem to me that these long term investments in infrastructure assets that should last many decades. Therefore providing steady income for many years to come. Perhaps Super funds would be better co-investors with governments looking for longer term returns ?
    Perhaps the Future Fund should be an investor of such projects as well, thus maintaining local ownership of all such strategic national assets.

  13. “The approach also improves flexibility over future government policies. To the degree that government policies change traffic flows, the private operator is protected. Changed traffic flows automatically change the length of time for the payback to the private investors.”

    Yawn. That’s how the Dartford Second Crossing was financed back in the mid-1980s. It raises the obvious question: “Why not do a BFT and re-finance with government bonds on completion?”

    If you really want to know about PPPs, look at the following:

    (Note: the discussion is continued over several comments)

    and briefly on the [former] very successful “semi-private” system used in France from the mid-1970s to 2005:


    • Some very to the point links there, Stephen. With your background in financing private roads, you are clearly well aware of all the drawbacks.

      Thank you!

  14. Whether PPP has a future or not, I am in favor of the “users pay” principle for new roads. Perhaps the costs can be partially covered by increasing the petrol tax and car registration fees.

    After all, why a taxpayer who does not own a car should subsidize the construction costs of a road?

    • The problem with that suggestion is that current taxes on petrol and registration fees already pay for road construction and maintenance about five times over.

    • Well that is the problem with user pays, is that it doesn’t capture every beneficiary.

      Those who have cleaner air because traffic is smoother and/or diverted, they may not pay for this improvement.

      How do you capture the value they received?

      why a taxpayer who does not own a car should subsidize the construction costs of a road?

      Reminds me of lorax drinking coffee in paris…

      I can guarantee you that person who doesn’t drive benefits from delivery trucks carrying goods that they purchase, or rides a bus, or a numebr of things.

      Infrastructure benefits everyone, be it direct use, indirect use, or enterprise leveraging off it to provide wealth and jobs.

      ‘user’s pays is an attempt to make only those in curring direct use pay, and that is absurd… but used because it is most easily quanitifiable.

      Once you start trying to extract payment from indreicts beneficiaries, it then is only qualitative… and in some apsects, politics is about assigning a price to those qualitative aspects.

    • The user should pay but other beneficiaries who don’t actually use the road but benefit from its existence ( increased land values due to faster travel times etc ) will also contribute via their rates / land tax.

      • People who want to “not pay for roads because they don’t own a car” should go and live somewhere where no-one owns cars and everyone travels by some other means.

        It is impossible to have a modern economy without roads and cars. Roads and cars are not a heinous rent-seeking scheme perpetrated by “big oil” or car makers; they are an essential and necessary feature of getting an economy from the third world to the first. It is never going to happen any other way.

        It was automobility that rendered Marx’s theories irrelevant. Without automobility, the rentier class would have captured all the gains of increased productivity and income. Economic land rent would rise as fast as incomes did.

        Certainly the land-rentier class does capture a far greater share of income growth in the heavily “planned” cities of the UK, than in the most automobile-based US cities. Growth containment urban planning is a handy substitute for lack of automobility, when it comes to the land rentier class capturing the gains of rising productivity.

    • Because they benefit in other ways. Everything from the truck that delivers groceries to the store through garbage collection to the bus you take to work.

      The idea of modern society without roads is laughable.

    • But somebody has to foot the bill. And it is not fair if car owners do not bear the majority of the share.

      Besides, these “indirect users” will also pay because the owners of the buses, trucks, etc., will surely pass on the costs to the users of their services.

      Of course, if the petrol tax and the registration fees already pay for road construction and maintenance about five times over, then I have nothing to complain about!

  15. If I may, I would like to say ‘perfectly analyzed and concluded by the following referenced lucid hyper-intelligentsia.

    Pph. Pph007, aj, Rumplestaskin, China Bob, Alex Hayworth, then a wee bit for Rusty Penny.
    I now ask the how and why of the smug snouts and Scorpions that designed, implemented and profited by way of this non-honest form of financial ledgerdemain, in that have not been called to account, for they are the actual source of the glut-greed disease spreading within and throughout Australia’s major cities?

  16. Those particular PPP’s were great results for the people and bad for the investors and banks whose projections were so wrong.

    It’s the same as Allco and Babcock they went well during constant growth but the era ended and those who made the last investments tended to get burnt.

    Those who invest now in similar projects (assuming they bid enough to get a deal approved by government) will be ultra conservative and make a decent return and possibly off the back of minimum flow per annum take or pay agreements – ie transfer more risk to government.

    In the end a lot of this stuff is just off balance sheet financing, sometimes arbitraging depreciations deductions for the private investor not available to state governments.

    Remember leveraged leasing and the tax rulings they had to introduce? And how Lveraged leasing was banned for state government authorities like transport. Westrail was in the middle of a leverage lease tender at the time the Commonwealth government closed the gate.