The “Private Finance Initiative” (PFI) policy was a curate’s egg – sometimes it worked, but in many (most) cases it was probably mistakenly conceived and implemented.
But what irritates me most about PFI is not the mistakes that were made around it, but the complete (wilful?) ignorance of many of its critics in understanding what most PFI deals were.
They are frequently critiqued as PFI project X – e.g. build a hospital – will cost 10 zillion times what the cost of the hospital through direct state spending.
The misunderstanding, or “wilful blindness”, is quite simple – the PFI contract was not to “build a hospital” but to “build a hospital, do all the maintenance on it and provide all sorts of building and back-office services for (usually) its lifetime.”
The main bit of the “massive costs” of PFI therefore come from maintenance and service contracts for the life-time of the building.
Were these often over-priced – yes. Were they often badly designed – yes. Were the interest charges on the original capital build over-priced – often.
But what many comparisons of PFI and non-PFI costs do is ignore the fact that maintenance and services would have had to be paid for anyway. They compare apples and oranges by ignoring this. Certainly this is the case for many of the ‘politically motivated’ attacks on PFI.
There were serious errors in setting up many PFI contracts that led to over-priced build, maintenance and services elements. But they are massively exaggerated if you pretend these were ‘build only’ contracts.
The biggest problem with PFI, in my opinion, was the lack of flexibility they built in. Public service needs are changing rapidly and being tied into 30-year contracts down to micro-detail levels is insane.
But if you are really interested in understanding the negatives, and positives, of PFIs rather than just scoring political points then a good place to start is the many National Audit Office reports on the subject. They take a more balanced view. See this for example, but there are lots more.
——–
And a “PS”
One interesting little positive ‘unintended consequence’ of PFI ought to be mentioned. I had an accidental discussion with a couple of commercial building industry execs (on the same train). They said PFI had changed the way they approached private sector commercial builds.
The standard model was always build it and hand it over and forget it. What PFI had made them think about was the ‘life-cycle’ of a building, including how it was maintained. This changed thinking about original design – if you were going to have to maintain a building it paid to design it so it was low-maintance, or at least lower-cost maintenance.
They told me they were now offering this as part of their commercial packages for new builds – slightly more expensive initial build but cheaper life-cycle maintenance.
In public sector PFI deals they’d been doing this to maximise their own profits, but in the private sector the same approach gave them a “USP” over rivals.
This is of course just an anecdote and I have done no research to see if its true. But it certainly sounds plausible.
STATECRAFT - improving government 
I don’t believe in most cases it’s willful misrepresentation. Most politicians are bad at understanding the issues they’re all too keen to pontificate on! Last night’s Question Time or any Question Time for that matter is an example!
I am also irritated by those who claim PFI more expensive on basis of interest rates charged by private financiers versus govt borrowing costs. See here for example.
http://crookedtimber.org/2015/11/19/private-infrastructure-finance-and-secular-stagnation/
(although John Quiggan does not himself make that argument and I am sure he is familiar with the argument I am about to make …)
however, if I am honest my thinking here is based on a half-remembered lecture by Paul Grout many years ago and I am not at all sure I have got it entirely right (although I am pretty sure superficial comparisons of financing costs are misleading).
Sorry if this is an abuse of your comments section, but here’s my tentative argument for why interest rates comparisons are misleading. I would be grateful if anybody would point out where I am going wrong:
Interest rates charged by private investors putting money into PPPs are higher because there is some risk of them not getting their money back. If the investment was risk free, then the interest rates should be closer to a risk free rate like the interest rates on government bonds. Under conventional public financing, these risks don’t go away, and the comparison of costs between public and private alternatives needs to account for that. Comparing interest rates charged does not do that.
Suppose we are talking about a hospital that is expected to cost £100m to build and then under a PPP an annual charge will be made of the supply of hospital services. The annual charge has to cover financing costs, so if the returns demanded by those putting up the £100m are high, that will feed through to the annual charge. Suppose also that once the hospital is built, there are some running costs that also need to be covered by the annual charge. These are expected to be £2m annually. Let’s say that once contracts are signed, there is a risk that running costs will turn out to be higher than £2m, and that’s where the risk that investors will not get repaid comes from. The risk could also come from building costs being higher than anticipated, but I don’t see that would change the argument, and I think this way is easier.
Let’s say the risk of things going wrong are such that private investors charge 8% interest rate, and that implies a contracted annual charge for hospital services from the PPP of £10m.
If the public sector can borrow at 2%, that means that keeping all else constant (but supposing for sake of illustration the public sector signs a PPP with itself) the annual charge for hospital services if the £100m was financed at 2% could be £4m a year, which goes to show that private finance is a terrible idea, right?
Well, no. Because the risk of things going wrong and running costs being higher than anticipated hasn’t gone away, and when that happens under the public sector PPP then the public sector has to bear those costs. When things go wrong under the private PPP, the cost of hospital services doesn’t change but project has to bear those higher costs meaning that the private investors lose money. That difference isn’t visible by comparing the interest rates charged in the two cases.
I haven’t worked this through, but I am guessing that if you assume that the rates charged by private investors merely reflect risks (and do not assume they make excess returns on top of that) then the true costs of hospital services under the two alternatives are identical, once you factor in the expected extra costs the public sector would bear in the things going wrong scenario (whatever the numbers would be to make the above all make sense).
Now of course there’s lots of things you can think of to add to this story that would modify the conclusion and make the private route worse. If you were so inclined, you could add stories about incentives and efficiency to make the private route better. Nonetheless, I guess the idea is that the main result (you cannot just compare financing rates because you need to also thing about where risks and the associated costs end up) would hold however you amended the story.