Who really bears the risk for P3s?

Canada is now the second biggest market for public private partnerships (P3s) in the world, as a recent Conference Board report showed (on page 30, see my initial critique here).

P3s are big business: Canadian governments closed deals on a reported $7 billion in P3 contracts in both 2010 and 2011.  This was the highest in the world in 2010 and second only to France in 2011 (see also note 1 below).  We’re well above the U.S., Australia and even the U.K., which had been a world leader before their PFI program imploded.

The amount of money from the Canadian public going into P3s is only going to increase, with more P3s in the pipeline and federal and some provincial politicians aggressively pushing P3s on their own and other levels of government, including municipalities and First Nations.

While Canada may be one of the leaders in the market for P3s, we’re far from a leader when it comes to transparency, assessment and accounting for P3s.   P3s are already a murky business when it comes to financial transparency—and we’re close to the bottom of that pool.  The value for money assessments used to justify P3s in Canada are simply not credible for a number of reasons, including for a specific reason outlined below.

Early P3s in Canada were often used by governments mainly to shift their debt off-book, but then auditors started catching up to this and they have subsequently justified for a number of other reasons, all largely proven wrong.

All the Canadian P3s I’ve seen in the past decade or so have been justified on the basis that they transfer large amounts of risk to the private sector.   This is also what U of T professor Matti Siemiatycki and Naeem Farooqi found when they analyzed all the P3s conducted by Infrastructure Ontario for an article published in the prestigious Journal of the American Planning Association last year.  Every single P3 project was justified on the basis of Value for Money assessments that claimed P3s transferred large amounts of risk from the public to the private sector.

The average amount of risk calculated for these projects was almost half (49%) the base project costs.   For some projects, the value of the “risk” calculated amounted to over 80% of the base project cost, averaged over $100 million for each of the 28 projects and over $3 billion in total.  That’s a lot of money, no matter how you count it.   Just to be clear: not one of these P3s would be justified on the basis of the central Value for Money assessments without this assumption that large amounts of risk were transferred to the private sector.  And Ontario isn’t unique: P3s in other provinces are also justified on the basis they transfer large amounts of risk to the private sector. (see note #2 below).

But how is this risk calculated?   They don’t say.   The value for money risk assessment templates Infrastructure Ontario provides are frankly embarrassing from a public policy perspective, especially for decisions that have involved billions of dollars of the public’s money.  There’s no evidence provided for any of the numbers proposed in their risk matrix—and other provinces are no better.  The value for money assessments release for each P3 project are superficial window dressing that provide none of the details necessary for an independent assessment.    And in the instances where auditors have reviewed the actual finances of P3s, they’ve generally always found that the project would have cost less if it were publicly financed and not run as a P3.

The way risk is calculated for specific P3s may be more sophisticated and complicated but there’s very little transparency: they hold risk workshops where people apparently come up with numbers adding up to tens and hundreds of millions of dollars, but nothing is revealed about the specifics.  When I’ve directly asked officials for these calculations, they claim it’s “proprietary” or confidential information—and all the P3 documentation that’s made public leave out the these details and other details on how much the P3s will actually really cost by year.  (see note #3 below)

But no matter how complicated—and secret—these key calculations of risk transfer for P3s are, none I’ve seen acknowledge a crucial fact: the real risk the private sector assumes through a P3 is limited by the net amount of unsecured money they have put into the project.  This amount is represented by the equity they provide and any net cash they have committed, less funds received. The initial equity share of the cost of P3s is usually no more than 10-15% and sometimes as low as 8% or less.   Since P3s are invariably set up as “special purpose vehicles” (SPVs), the big companies behind them can simply walk away if they aren’t making enough profit or if things go wrong, thanks to limited liability laws for corporations.  The maximum they lose is any equity and any net cash they’ve put in, less what they’ve been paid.   And a number of P3 companies have abandoned their projects, or used the threat of doing so to get more money out of the government.

Government always bears the ultimate risk because it’s ultimately responsible for delivering the service.  This is a fact that also seems to be ignored in these P3 risk assessments.  The government can then end up being responsible for paying off the bond-holders, whose money is secured through the asset and project agreements.  As we saw with the Ontario gas plant scandal that led to former Premier McGuinty’s resignation the cost of paying off the bondholders (in that case, hedge funds based in the US and Cayman Islands) can amount to many times the actual cost of the project.  And, as that example shows, P3s often magnify the risk for the public sector, instead of reducing and transferring it.

There may be some fancy modelling and a lot of creative accounting involved in the risk calculations used to justify P3s, but they still can’t escape the fact the private sector’s risk is limited by the net amount of money and equity they have in the project.  This amount (typically no more than 10-15% of the initial cost) is invariably considerably less than the value of risk that is assumed to be transferred.   For the Infrastructure Ontario projects I looked at, this ran from a minimum of 16% up to over 50% for a number of projects.   And since the equity share in P3s is usually paid a return of at least 10% a year, the net stake the private sector has in P3 projects after these returns diminishes quickly.  This doesn’t mean calculations of risk transfer less than the net private equity at stake are credible: that can’t be determined unless the details are made fully public–and they’ve aren’t doing that.

Governments may never reveal the details of these P3 risk assessments (because it is supposedly “proprietary” information or simply to cover up the indefensible degree they engage in creative accounting to justify P3s), but any calculations of risk transfer that exceed the net cash and equity the private sector puts up should have no credibility, nor are they justified by the real arbiter when it comes to P3s, which is the private money involved.

There are numerous other problems involved in the way governments and private sector firms assess and account for P3s in Canada and elsewhere, and it will take considerably more to outline them.


A few additional notes:

  1. These calculations of the financial close value of P3s understate their real liabilities for the public, because P3 deals commit governments to pay billions each year for decades to P3 operators.  The UK government revealed last year that their outstanding PFI liabilities amounted to over £300 billion.  That’s equivalent to over C$500 billion or almost $20,000 per UK household.  Canada is likely not far behind the UK in building up a “P3 Debt Bomb” of the same magnitude, but our governments refuse to reveal the extent of it.
  2. BC also double counts risk transferred by using higher discount rates, which is another issue.  It should also be noted that few P3s in Canada have to deal with demand risk, or revenue risk related to demand conditions, such as through tolls.  Most now have their annual revenues guaranteed by the public entity for decades, so these revenues streams resemble a government bond, but at a much higher rate.
  3. There’s big money to be made operating P3s—and there’s also big money to be made on the consulting side for the accounting, finance and legal firms that prepare the P3 business cases, risk assessments, P3 contracts and bids.  These P3 “transaction costs” amounted to an average of over $8 million per project in Ontario for the government.  And, as emblematic of the lack of real risk transfer involved with P3s, even the companies that lose bids for P3s are paid expensive “honoraria” by governments involved.   Children at the junior elementary level may all receive participation medals for some activities (something my children now mock), but I don’t know of that many adult competitions where even the losers receive a generous monetary award–except in the coddled P3 industry.

One comment

  • Lesley Durham-McPhee

    Calculation of risk becomes emotional, especially for governments. There’s a perception that the private sector is less emotional, I doubt that’s true. They certainly have learned how to exploit the emotions of politicians and ultimately the public.

Leave a Reply

Your email address will not be published. Required fields are marked *