Unfit to govern

What can governments do better than the private sector?
The answer isn't quite "nothing". There's one thing states are better at - borrowing money. Because governments can bleed taxpayers dry, they'll almost always be able to borrow at lower interest rates than the private sector.
However, the story of the Norfolk and Norwich PFI hospital (full report here) reminds us that New Labour has relinquished even this function. It's an old story, but a dirty one.
The gist is that in 1998, the NHS trust gave a contract to a private consortium, Octagon, to build and  run a hospital. In 2003, Octagon refinanced the debt it incurred in building the hospital, thus raising the rate of return to shareholders from 19% to 60%, whilst exposing the NHS trust to greater financial risks.
The dead trees say the scandal is the "windfall for shareholders" in the 60% return.
This is only part of it. The scandal is also the original cost of finance -  of 19% at a time when 20-year gilt yields were under 6%. Octagon says such a high cost was necessary because this was a risky deal.
Maybe it was - to them. But they shouldn't have borne the financing risk. The government should have, because the risk to it was lower.
So why was the risk shifted to the private sector? It's an accounting fiddle. If the government borrows directly to build a hospital, headline government borrowing rises. Private financing just hides the debt, even though it costs more. In real terms, the taxpayer loses.
There are two messages here.
1. New Labour is not "pro-market". In a proper free market, risks are transferred to those who can most easily bear them; for example, entrepreneurs transfer some risks to shareholders, who are better able to spread risks. However, under the PFI, risks are transferred away from these best placed to bear them. The government shifts a risk onto the private sector, thus compelling the tax-payer to pay an unnecessary risk premium in exchange for little reward.
2. New Labour has no conception of the proper functions of government. Just as it has abandoned the core duty of the state - to protect its citizens - so it has abandoned something else the state should do best, raising money.
If the chief executive of any private business showed  no awareness of his business's core competence, and preferred accounting trickery to real risk management, he'd be booted out of a job.
So why is New Labour still in office?
There's more: Please remember the distinction between private management and private financing. There might be a case for introducing the former into the public sector, but there's not case for the latter. For more on the PFI, try these links.

Comments

[angry economist (Glenn)]

In my time appraising government expenditure plans in latter years we looked at discount rates of 6%.

Seems bonkers that taxpayers are paying so much for borrowing against a guaranteed income stream for PFI financiers.

[Luis Enrique]

they've since changed things so that the state receives 50% of the gains from refinancing haven't they?

Is it really a guaranteed income stream? Guaranteed revenues aren't guaranteed profits and some PFIs will end up being big loss makers won't they? But how does this affect financing - the lenders want higher rates because of a higher risk of default? I'd be interested to know what the PFI companies quote as their cost of capital and how it's worked out.

[Tim Worstall]

I remember when this was all being first thought through (mid 90s? well, OK, I don’t actually remember when it was) and having long arguments with a couple of PFI enthusiasts. I just couldn’t see what was in it for the State. Yes, private management, bearing risk, might make things more efficiently. Yes, accounting trickery. But why PFI? Why not simply fixed price contracts with private companies?
The only conclusion I could come up with then was that the accounting trickery was sufficiently juicy to make it worthwhile for politicians. And I didn’t (nor do) think that sufficient in toto.

[Jack]

"Enron for governments" according to the FT.

As I understand it Gordon Brown's enthusiasm for PFI is mostly because it allowed him to spend a lot more than he could borrowing the normal way. It's partly wishful thinking but also partly a workaround of a shortcoming of national accounts.

I think it works like this. Suppose a hospital can save enough by building its own laundry that the investment required would have paid for itself in three years with great certainty. If the hospital borrowed the capital without PFI it counts as national debt and gets nixed by the Treasury. Contract out to a PFI scheme and it looks like savings immediately.

I don't think PFI is the solution but the problem is real enough. What is the right way to do it and who does it that way? The US Social Security scheme has much better accounts than the UK equivalent. Do things like this also work better over there?

[dearieme]

"New Labour has no conception of the proper ...". Lack of propriety is there greatest vice. They seem to have no virtues at all.

[dearieme]

Arnold Kling says
"Public-private partnerships are problematic, in my view. Power corrupts, absolute power corrupts absolutely, and private-public partnerships absolutely corrupt the private sector." And that from a country where the package doesn't include a seat in the Lords. Though, come to think of it, it might include a seat in an embassy somewhere nice.

[Alex]

Right from the beginning, the PFI accounting was incredibly weird. The public sector comparators, for example, Treasuryspeak for the baseline that the public sector was supposed to achieve on its alonio were often set below the current performance it was actually achieving.

Then there was the farce of "Reputational Externality" - when, despite the handwavey "risk transfer" figures and the deliberately terrible public sector comparator, the deals still didn't add up, they came up with the idea of including a notional cost to the public of the government being seen to borrow more, even if the total cost was going to be greater!

There being no reasonable methodology for this, conveniently, it could be pretty much any figure needed to close the deal.

On the left, we've been saying this since they were invented back in 1996..