Tony Blair was a classy politician when it came to the level of political talent he seemed capable of. How sad that like his political counterparts in Australian State Labor governments he and his Chancellor Gordon Brown established the kind of spiv financing that saw Greece go to the wall as standard operating procedure.
By John Kay, Published: February 15 2011 23:25 | Last updated: February 15 2011 23:25
Last week, the UK’s parliamentary public accounts committee published a damning critique of plans to widen the M25, London’s orbital motorway, under the private finance initiative . The proposal to outsource Britain’s air and sea rescue service under a similar arrangement collapsed amid recrimination. When finally unravelled, the complex story of the PFI may illuminate the murky – and to taxpayers costly – relationship between government, business and finance in the era of New Labour.
The PFI represented an attempt to adapt the methods of project finance – developed for activities such as North Sea oil exploration – to public sector infrastructure projects. Project finance was necessary because the industries concerned were better at managing projects than raising money. The government, however, was bad at managing projects but very good at raising money. Public sector projects have routinely been plagued by delays and cost overruns. But the government can borrow on the finest terms: it has all the resources of future taxpayers.
In the climate of the past two decades, it was perhaps inevitable that the bankers would dominate the engineers. Funding structures became more elaborate and complex. The iconic refurbishment of the Treasury’s own premises was undertaken under a PFI, which raised senior debt at 163 basis points above the government’s own cost of funds.
We do not have adequate information on the capital and operating cost of PFI projects, relative to those that would have been incurred by direct public procurement. The assertion that PFI projects came in under budget more often than directly procured activities is valid, and often made, but means little. Under PFI, cost escalation is built into the process at an earlier stage. The public accounts committee correctly judged that bids to maintain the M25, at far less cost than allowed for by the Highways Agency, demonstrated not shrewd purchasing but that the agency had little grasp of what it was doing. Infrastructure projects in Britain almost invariably come in towards the top in international comparisons of costs and there is no sign of general improvement.
There is no doubt, however, that the funding component of PFI projects has been expensive. The rate on the Treasury refurbishment illustrates the point. But the effect, and intention, was to take substantial volumes of government borrowing off balance sheet. Before pointing a finger at Enron or Greece, take a walk down Whitehall.
PFI contracts are inflexible. Schools and hospitals are locked into agreements that may last 30 years or more. This follows from insistence on bundling finance, capital procurement and facilities management into a single contract. Any change in requirements during that period must be negotiated with a monopoly provider who generally has little incentive to be economical or accommodating, as George Osborne, the chancellor of the exchequer, discovered when faced with a quotation for a Treasury Christmas tree so exorbitant that he went out to buy one himself.
Perhaps most seriously, PFI has not improved the public sector’s capacity for project management. Rather than learning to manage construction companies and facilities providers better, public agencies became expert at negotiating elaborate contracts and innovative debt finance. That is not a skill government needs. From the start, PFI conflated the desirable aim of exploiting private sector management skills in project supervision with the undesirable aim of obscuring public finances with complex funding structures. It created an industry of advisers with a vested interest in its own expansion.
The present coalition has agreed to end off-balance sheet funding and to observe international financial reporting standards. It should renegotiate or buy back many existing PFIs and adopt simpler structures to outsource much government activity and develop project skills while minimising the role of bankers and financial modellers.
Nicholas,
I’m still not convinced that PFIs (or PPPs here) are such a bad thing. It seems to me that, at least from this article, the main bad thing about them is this: “The iconic refurbishment of the Treasury’s own premises was undertaken under a PFI, which raised senior debt at 163 basis points above the government’s own cost of funds”, and the rest just shows that the British government, like many Australian ones, is essentially incompetent at actually negotiating decent contracts and also corrupt enough to try and use PPPs to sweep debts under the carpet, like Australia also. However, these two things really have nothing to do with PPPs being bad or good in general. The situation is probably worse in Australia, because the choice is often between some bit of infrastructure we need or nothing at all, not between something that is government funded or a PPP (especially at the state level).
I can also see some upside to this:”PFI contracts are inflexible”, especially for roads (schools and hospitals are another matter). This being that charges on roads (e.g., the M4 in Sydney) are going to be far harder to get rid of if scrapping them comes under popular pressure (cf. having a better a road system) than they would be than if the roads were purely public.
Conrad, I agree with what you’ve said. But somehow it reminds me of the National Rifle Association’s slogan that only people kill people, not guns. I’ve never argued that the category of PPP is bad. It’s just that it’s usually deployed in bad ways and for the wrong reasons (ie to hide debt). Pretty obviously the basic architecture for PPPs makes sense – which is to go through a structured process of deciding who is best to bear and manage which risks and then arrange the project accordingly.
I am just too slow to get the link with Greece.
Nic is right – its not that PPPs are always bad in principle, but that the incentives the players in them face usually make them bad in practice.
The dead giveaway is the need to shroud things in “Business-in-Confidence”, and to make the financing so complex that outsiders can’t understand it. As soon as you see players pursuing non-transparent means you should assume prima facie that the deal is not a good one for third parties – ie in this case the punters.
DD, without wishing to detract from your point – particularly your agreement with me ;) – keeping things commercial in confidence may well be conducive to getting a better price (as firms don’t want low prices they’ve agreed to bandied about). That’s not a (very strong) argument against releasing the information because probity should generally trump a few pennies, but it shows that the dilemmas are real, not just the result of deliberate obfuscation.
Patrick, Greece’s government got into trouble by shunting things off its government’s balance sheet – just like we have done. Just not quite so egregiously.
I’ve had a bit of experience with these. They are almost always a bad bargain – because there is no sensible way to allocate risk that cannot be gamed. Long complex contracts drag in lawyers, simple short ones get interpreted in favour of whoever holds the whip. This is usually the private party, who has the technical expertise and all the time in the world. Why assume governments are uniquely bad at complex projects? Notice Boeing’s latest troubles? Or Siemens? Or the 00s of millions NAB (and every other major bank) wrote off on IT? Notice that when State Rail was run by engineers it mostly worked? Private is better at publicity, and avoiding exposure. Public cannot write off debt (which is the usual fate of anything large and innovative in private hands – a la most railways in the C19, the Channel Tunnel, Global Crossing….). Straight purchasing by well-staffed government has a long track record of success – why throw it away?
Contracts can be as simple or as complex as you need them to be. If you desire flexible options and detailed stipulation of outcomes, then the contract is going to end up complex… but at least the desired outcome will actually be documented somewhere (hopefully in view of the voters, so they can later compare the intended outcome with the actual outcome). If “innovative debt finance” includes research into costs vs benefits and figuring out who will pay for a particular service… I’d argue that would be exactly the skill that government needs.
The other skill they need is the ability to not get heartset on a particular project. Widening the M25 sounds like a nice thing to have, but if after doing the research, the finance seems a bit expensive and the investors seem a bit reluctant — hey just maybe those guys know something, huh? Sometimes it’s better just to walk away before the trouble starts.
Ahh yes, the secrecy factor… but that’s really up to the voters to demand and end to the whole unaccountable, inscrutable, under-the-table dealings that go on when people are not sufficiently vigilant. The concept of spending public money in secret is incredibly undemocratic. Government officials caught misleading the public need to be treated harshly.
Which happens in pretty much every sphere of government influence.
Exactly right.
I disagree, if everyone tendering for a project knows that ALL tenders (winning and losing) will be published they will stick closely to their catalog prices and those prices are regular market rates. You can be sure that if they are bidding significantly lower than market then they have figured out some other way to claw the profit back.
Why is it a “write off” if no one else can achieve the same deliverable at a lower cost? Do you want those ATM’s to work every time, or is near enough good enough?
Hey Nic, here’s the same argument from me, 5 years ago, at JQ.
Got the jump on you for once. ;)
Goddamn link doesn’t work.
http://johnquiggin.com/index.php/archives/2006/03/10/weekend-reflections-11/
Hmm, on Greece, haven’t you confused a symptom with the disease?
One could equally usefully say that if the Greek government had been even capable of executing PPPs Greece might not be a basket case.
The best strategy for government would be to set up PPPs and then buy them out at a firesale price when the private entity goes bust. A number of NSW transport projects (airport rail line, cross-city tunnel etc) could have been purchased cheaply this way.
Well bugger me if that isn’t the most concise critique of PPPs ever written down.
Here’s another example of the secrecy trick — create a wholly government owned PTY/LTD company that pretends to be “private” and thus avoid FOI or other scrutiny. All the benefits of a private company without worrying about shareholders asking questions, all the powers of government without the voters asking anything either. How good does it get?
http://www.theaustralian.com.au/national-affairs/julia-gillard-backs-foi-exemption-for-taxpayer-funded-nbn/story-fn59niix-1225990173260
And still asking for an itemised list of what the Building Education Revolution actually built and how much each one cost… unlikely it will ever become public knowledge, but gotta plug on and keep asking.
I think that the standard FOI regime is to exempt such bodies ‘in relation to documents in respect of its commercial activities‘.
Treating any part of costing or planning for the NBN as a commercial activity is a bit rich to my mind. The definition, after all, requires there to be a degree of competition:
(emphasis added).
In Secretary, Department of Employment, Workplace Relations & Small Business v The Staff Development & Training Centre Pty Ltd [2001] FCA 382, Drummond J commented that:
In that case Drummond J went on to specifically reject as an argument
OTOH, in Australian Postal Corporation v Johnston [2007] FCA 386, Greenwood J interpreted commercial activities quite broadly, with reference to activities carried out with a purpose, perhaps amongst others, of competition with the private sector.
But I would struggle to see how anyone could imagine that the NBN is in competition with the private sector or could reasonably be expected to be given that the whole point is to construct a monopoly provider.
So I would hope that Drummond J’s reading would apply and the mere fact that the NBN Co will compete with everyone else for tendering etc would be irrelevant to the costings and business planning.
Great comment Patrick (we seem to be forming a mutual admiration society here). From memory Telstra was exempt from Ombudsman, FOI etc even back in the days when it was still a publicly owned monopoly. Of course it remains to all intents and purposes a monopoly for landline installation and repair. The competition is only at the retail end. The extent to which Telstra has long been effectively impervious to any meaningful form of accountability, even before it was privatised, is underlined by this extract from my lecture notes about the public/private divide:
BTW I think there would be some vaguely plausible basis for exempting some classes of NBN documents relating to the tendering process and other activities involving negotiations. It wouldn’t really be in a position to get a good deal on anything if businesses negotiating with it could compel production of material revealing NBN’s negotiating ambit etc while the negotiation was on foot. However, even there, there’s no plausible basis for such an exemption to be held still to apply once the negotiation has been completed. This conundrum has been a recurring problem with FOI ever since it was introduced, and not only with GBEs potentially competing with the private sector. The exemption has historically been used by government agencies generally to avoid disclosure of just about anything relating to tendering and procurement. I’ll have to check out that Drummond J decision and how it’s subsequently been treated, because it looks like a decision which would have significantly restricted this practice if widely applied. A quick check of the AustLII noteup function shows that it wasn’t appealed and has not been considered or applied since at federal level. However AustLII’s noteup function doesn’t reliably pick up every instance where a case has been considered or applied.
I agree with Ken and Patrick that FOI should apply in this case, but am also relieved that the the CSO’s imposed on Telstra (and others) can’t be used to force Telstra to wire up Twiggy Forest’s holiday bungalow in the Bungle Bungles (if he had one).
haha indeed Austlii does suck for research!! Drummond J was actually reversed at [2001] FCA 1375
Albeit, apparently affirming the specific chain of reasoning relevant to the present question:
I would need to read that a few more times to get comfortable that I have understood it!!
The Appeal seems to have been considered only by the AAT: [2007] AATA 1569 and [2004] AATA 1364.
Maybe the issue just doesn’t come up a lot for some reason obvious to an administrative lawyer but which escapes me ?
To clarify, I too had relied on austlii when I wrote my first comment.
Pattrick
It appears that the Drummond J decision related to s43(1)(a) and (b) of the FOI Act. Note that the Act was recently substantially amended and overhauled, so the following relates to the Act as it previously existed. The exemptions in s43(1)(a) and (b) are:
However, there is also s43(1)(c) which I think is the main one government agencies rely on in resisting applications for disclosure of material relating to procurement etc:
Incidentally, the corresponding provision in the overhauled FOI Act is s47 and s47G. All they seem to have done in essence is split s43 into 2 separate sections, one providing complete exemption and the other qualified. I suspect it would still allow agencies to resist disclosure of documents relating to procurement from the private sector:
S 47G is the most relevant for tender/procurement purposes:
One thing I haven’t checked is whether “organisation or undertaking” for the purpose of this section can include the organisation or undertaking of a government department or agency, but I suspect the answer is that it could.
Well it looks like Gilla has conceded that the existing FOI laws wouldn’t protect NBN as initially discussed – i.e. that what they were originally planning was not at all the ordinary operation of FOI: