Re 2a) – surely a traditionally procured construction contract that was tendered on a fixed-cost basis (i.e. the contractor commits to doing the work at a set price and has to absorb any cost overruns himself) would achieve the same thing?

If I recall correctly, Gordon Brown’s London Underground PPP left the taxpayer with the cost overrun bill. By way of contrast.

It’s true PFI is an excellent accounting fudge, but perhaps another important reason PFI is so entrenched is precisely because it is so vastly inefficient from a taxpayer perspective. It’s so massively profitable for so many people (consultants and lawyers as well as the actual contractors) that it’s become a powerful political lobby in itself.

Excellent: incentives matter, even to politicians. Especially perhaps.

Being able to announce the building of something, attend the opening of something, on your watch but the bills come in to hte people who turn up 15 years after you’ve retired? Of course this will be popular with politicians.

“surely a traditionally procured construction contract that was tendered on a fixed-cost basis (i.e. the contractor commits to doing the work at a set price and has to absorb any cost overruns himself) would achieve the same thing?”

This would, if the politicians (or the civil service, each is as bad as the other) would write a tight contract in the first place. What actually happens is that they try to leave a lot of wriggle room (all politics is like that) plus then interfere and try to change the contract during the work.

And contractors just loooove that. A contract change sir? Why, our pleasure, that’ll be another 20% on the total bill!

If the political process were able to write contracts as tight as, then stick to them as, Tesco does on its store building then schools n’ospitals would be built as cheaply as Tesco stores are.

But they can’t do that so they ain’t.

“PFI isn’t going anywhere. This report changes nothing.”

Did anyone think it would?

Much of this report concentrates on the down side in the public sector. The upside is the big money made by the contractors – contributors to party funds.

Maybe nobody’s noticed, but the bank bailouts had a big downside in the public sector, and a huge upside in the City.

Anyone notice a pattern?

Government is funded by private corporations and largely populated by individuals seeking non-executive directorship sinecures – which makes PFIs a win-win deal.

@3. Tim Worstall: “What actually happens is that they try to leave a lot of wriggle room (all politics is like that) plus then interfere and try to change the contract during the work.”

That’s feature creep during the planning and building stages. It’s why so many IT projects (private sector, not just government) fail. Ask your mates at The Register, Tim. The problem is not unique to PFI.

Assuming that government purchasers could stick to the original plan, PFI would still cause financial problems. When Tesco authorise construction of a shop, they wait until it is built and then they run it. With PFI, a government service orders a building and building support services on a complex contract. The assumption is that the required building support services will be similar ten years after the contract is agreed. Because nothing changes in ten years: surgery doesn’t change, libraries and record keeping work the same way, essential utilities such as air conditioning, water supply, power source/requirements are immutable.

“That’s feature creep during the planning and building stages. It’s why so many IT projects (private sector, not just government) fail. Ask your mates at The Register, Tim. The problem is not unique to PFI.”

Indeed. In fact, spot on. It’s just that government (at any level) is worse at this than private sector.

Pater worked a lot of his career on the one and only public sector engineering project that came in under budget and under time. Polaris (much of the rest of his career was on Trident, the replacement). Leave aside the nuclear annihlation bit and I do recall asking him why it had actually worked.

“Interfaces” was the answer, with a chuckle from him and the several of his muckers who heard my 18 year old self’s question.

You can do whatever you like within your area. But you cannot change an interface: you cannot change the infor that you’re passing onto hte next stage, cannot demand different such from hte previous stage. Sure, with missiles and subs, this goes further, it’s not just information. It’s also power flows, water flows etc.

But the basic insight would have prevented, for example, the NHS IT disaster. As Yorkshire Ranter has been banging on about for years.

“Here is the interface for sharing information”.

How you collect it, how you manipulate it, how you process it, we just don’t give a shit. But this is absolutely the way you will present it.

Would have been much simpler, no?

@6. Tim Worstall: ““Interfaces” was the answer, with a chuckle from him and the several of his muckers who heard my 18 year old self’s question.”

That sounds like one of the points that I raise in annual appraisals that flags me up as a shit stirrer. To which interface should I dedicate my time? The sexy tech or the old one that works? Should I plumb something in and disregard the consequences?

William Bushnell Stout: “Simplicate and add more lightness.” (Later attributed to Colin Chapman.)

If I recall correctly, Gordon Brown’s London Underground PPP left the taxpayer with the cost overrun bill.

Kinda. The taxpayer ended up paying for the improvements that are currently happening, but only because a) the PFI company went bust b) London Underground is too important to just say “well, it failed, screw you all”. So while the taxpayer *could* have refused to pay to fix the problems, that would have cost the London economy far more than the cost of stepping in.

OTOH if a PFI new-build hospital contractor goes bust in the early stages of construction, the government can completely legitimately say “not our problem” and create new capacity elsewhere.

This is another important point on PFIs – they only work for projects that aren’t too-big-to-fail. If the government can plausibly tell the PFI company “no, we’re not bailing you out for your cock-up”, things are much more likely to work out in the public interest than if letting the project tank is unthinkable.

“the one and only public sector engineering project that came in under budget and under time.”

There is a bypass near me that was built ahead of schedule and under budget. I’ve worked on numerous small scale projects that have achieved the same. (Of course one could say that the timescale and budgets were too generous in the first place, and thus easily achievable.) Its just you only hear about the cock ups.

Why the worst infrastructure gets built and what we can do about it (pdf)

7 PROVEN FACTS THAT THE WORLD IS HEADED FOR AN ECONOMIC APOCALYPSE IN 2012

“Indeed. In fact, spot on. It’s just that government (at any level) is worse at this than private sector.”

I suspect that the cause is simply that they are a big organisation controlled by people terrified of having their careers terminated if the slightest shortcoming or perceived shortcoming should become known to the public/politicians. We get disastrous levels of caution, expensive obsession with needless procedure in order to pretend every single decision has been technocratic and scientific (and therefore no-one is responsible), and a tendency to stick tenaciously to doomed plans (to avoid admitting error) for far longer than any impartial outside observer ever would.

I see no reason there should be special magic that goes on in an office due to it being run by a profit-making entity. I suspect private companies merely tend to be efficient because they concentrate on simple profitable activities, an option not open to people trying to run public services.

We’ll see how much better the various local authorities planning to contract out their entire operations commercially do at this sort of task. The reality is, I think, that they’ll be not any better at all. Actually, no, I think it’ll be worse because they’ll probably have screaming conflicts of interest, due to the small number of big players in any given part of the “outsourcing” industry.

““Here is the interface for sharing information”. … Would have been much simpler, no?”

Not really, because much important information for public projects (outside the narrow sphere of engineering) is probably not usefully expressed in numerical form. Primarily, people need to talk to each other.

Besides, I think I can guess how senior local authority figures would respond to the idea that “interfacing” was essential to running a credible project: they would immediately fear being held responsible for failing to do “interface” properly, and would go massively overboard: they would hire “high-level interfacing consultants” with extensive private sector experience, who would most likely suggest inserting an “interfacing protocol development stage” into each project (no matter how small), and suggest the production of project-specific “interfacing protocol framework documents”. These would initially be conceived of as being “a side of A4″ but in practice would run to hundreds of pages because of the outputs demanded. There would, of course, be an extensive “interface scoping consultation process” taking some months, carried out by the consultants in question. For each project, they would obviously recommend contracting out the strategic interfacing protocol process (SIPP, no doubt) in ill-defined “tranches” or “packages” to various private contractors who know sod-all about the actual project, and spend most of their time in dispute with the holders of the other “tranches” about who should do the difficult bit.

OK, so I’m exaggerating for effect, but you get the picture. Pointless bureaucracy is very much an innovative private sector-led strategic partnership these days.

As I am finance guy, it’s worth adding that many of the PFI contracts were drawn up with “heroic” (ie wrong) assumptions about how low the interest rates available to the private sector would be. Indeed, Private Eye have been banging on about the alleged manipulation of this figure regarding the tube deal for years now.

It is arguable that the tube deal – put together by Brown – was the worst decision taken by that government. (Though competition is fierce for that accolade.)

“I suspect private companies merely tend to be efficient because they concentrate on simple profitable activities, an option not open to people trying to run public services.”

An example of this is Swansea City Council, who last year briefly tried to outsource their entire child protection functions. It was dropped fairly quickly as no private operator would touch such a thing.

It is arguable that the tube deal – put together by Brown – was the worst decision taken by that government. (Though competition is fierce for that accolade.)

Only by insane people, though. The Tube deal cost a few billion quid – a lot, but not much as a % of GDP – and led to the Tube getting much better (objectively, even if it doesn’t always feel that way).

It’s not hard to come up with a list of decisions taken by the government that did far more harm. Hell, it’s not hard to come up with a decision taken by the current government that did far more harm, even though it’s only been in office for a year and a half…

“Being able to announce the building of something, attend the opening of something, on your watch but the bills come in to hte people who turn up 15 years after you’ve retired? ”

It also locks future governments into spending commitments, and gives the private sector a financial interest in government spending. Most of us are familiar with the concept of the military industrial complex in the US, and the argument that this explains the massive over-spending on the military. Well PFI in schools and hospitals is one way of achieving something similar in health and education, and thus explains its popularity with New Labour.

Does anyone believe that the PFI stuff really, really, will end up back in the public sector? I don’t.

16: this is the one point that works massively in PFI’s favour that is often missed: it does mean that governments like the current shower *can’t* cut back on key services.

‘Governments will always love PFI, oppositions will always hate it’

Any evidence the opposition hates it? Quite a u-turn if so.

PFI is the worst of both worlds: driven by greed and free from democratic accountability but backed up by the force of the State.

Its New Labour corporatism in its purest form.

Shatterface: it was invented by the Tories, FFS. It’s the Thatcher consensus in its purest form – and yes, NuLab’s economic policy was the Thatcher consensus plus a bit of redistribution.

Its not just Labour and Tory politicians who have had this conversion once in office.

The SNP derided PFI and PPP so much, they tried to bring in a new form of financing using their Scottish Futures Trust.

The SFT doesnt actually bring in any new money, but instead, for one of the first projects to be built, merely allows the local authority to pay a third party contractor (who the SFT will help them find!) an annual “fee” for the use of the building which the contractor will design, build and maintain for 30 years, after whcih point it reverts to the authority.

If it looks like PFI, acts like PFI, then it probably IS PFI.

@20. john b: “Shatterface: it was invented by the Tories, FFS. It’s the Thatcher consensus in its purest form…”

Further to that, PFI is an expansion of the creative accountancy tricks that local authorities deployed in the 1980s. PFI permits the establishment of projects that are not measured on the books as capital, and because we have fooled the book keepers we are supposed to fool ourselves that we have taken a wise course of action.

But the creative accountants of the 1980s understood that they were buying time and that their projects were financially inefficient (ie directly borrowing money would have been cheaper). PFI supporters have yet to acknowledge long term cost and have tossed out arguments that PFI mitigates the expense of managerialism. In which case, sort out managerialism.

So the Conservatives are to blame twice. Firstly, for creating an environment where public accountants signed off bizarre local government capital schemes in the 1980s. Secondly for kicking off PFI at an national level.

All the same, John Prescott and those who permitted him to promote PFI as a low cost capital funding mechanism have earned ritual humiliation.