˚A Crown Guarantee for USS?

Mike Otsuka wrote to me recently about the possibility of the government ending the dispute over University Superannuation Scheme pensions with a crown guarantee. The government could simply take on all the risk associated with USS. Its balance sheet would be absorbed into the Whole of Government Accounts. More on that in a bit.

Something like this was suggested by the Liberal Democrat leader Vince Cable. So far nobody from the government has, as far as I know, seriously discussed it.

For some details on how crown guarantees for pension funds work, have a look at pp.42–4 of this document, published by the National Audit Office.

Politically, the advantage is, potentially, an immediate end to the dispute. USS could accept its own original valuation (the September valuation), according to which there is no immediate future funding problem and the capped Defined Benefits pensions that employees are demanding could be preserved in something close to their current form, at least in the meantime. The Pensions Regulator could relax a bit. This would end the strike, satisfy USS, and (assuming some degree of yet-undemonstrated rationality) be acceptable to the more risk-averse members of Universities UK, since the risk they were so worried about would get swallowed up onto HM Treasury’s balance sheet.

Here I want (1) to say a bit about how this would work, (2) to deal with some possible objections, (3) to say why the government should accept it, and (4) give some more general opinions about pension funds.

The NAO report linked above notes that:

there are several examples where central government has provided financial support to a range of funded public service pension schemes. In the four cases identified, government financial support totalled £1.4 billion (p.42).

A guarantee for USS would potentially need more than that. But notice that the allocated funding is only a guarantee. If the fund performs normally the guarantee wouldn’t have to be called. But it would permit USS to keep the September valuation without the Pensions Regulator breathing down its neck. Deficit recovery could be satisfied with normal contributions and no violation of ‘Test 1’. The relation to the Pensions Regulator would be equivalent to that of the Local Government Pensions Scheme:

The Pensions Regulator assesses the appropriateness of assumptions and the length of the recovery period for private sector pension schemes and some public sector funded schemes but not the LGPS. Its role has expanded to include assessing how well the LGPS funds meet the governance and administration legal requirements and the standard to which they are being governed. It uses this information to build risk profiles for individual public service schemes. However, the Regulator’s assessment does not cover deficits and recovery periods specifically. (p.37)

In other words, a crown guarantee would give the scheme a bit of breathing space to resolve the dispute in a way that could satisfy the union.

In reality — behind the various self-imposed financial constraints — the result of a crown guarantee would be to absorb USS’s balance sheet into the Whole of Government Accounts. Its deficit would then appear as ‘taxpayer’s equity’ (here’s a sample from the most recent published WGA — 2015–16):


The units there are billions of pounds. So it’s not a huge difference to add USS’s deficit, on any valuation.

But more important is the nature of the liabilities. The ‘liabilities to be funded by future revenues’ are what politicians and the media refer to as The Deficit. Now different people think different things about the acceptable size of The Deficit. But two important points are:

(1) HM Treasury can have no cash-flow problems, ever. The Bank of England clears its payments before the Debt Management Office engages in whatever funding operations are required to satisfy various self-imposed financial restrictions. The DMO sells various government assets, particularly Gilts, to cover liabilities. If there is demand for Gilts at the prices required to keep yields in line with the Bank of England’s target interest rate, then the liabilities are adequately funded. Which brings me to:

(2) Pension funds like USS buy lots of Treasury Gilts. USS currently buys a lower proportion than other funds, relative to equities and property, etc., but it aims to shift this over time. And the sheer volume is enormous. Thus when HM Treasury guarantees a pension fund, the pension fund uses the financial support to purchase the Gilts that fund the financial support! If this sounds like bootstrapping to you, it’s because it is.

In fact, in accounting terms, there isn’t much going on with a crown guarantee for a pension fund besides some consolidation and a portfolio adjustment. Pension funds protect against risk by buying Gilts. This is, as I argued in a previous post, effectively pushing risk onto HM Treasury, which absorbs it without trouble given its imperviousness to cash-flow risk (and the infinite time-horizon for its ‘budget constraint’ in most models). If HM Treasury guarantees the pension fund, the fund can buy fewer Gilts to cover the same risk, while HM Treasury effectively takes on contingent liabilities to the fund (which fall due only in cases of poor performance). Bonds turn into a sort of reverse-equity.

This is the answer to people who complain that a government guarantee would shift risk from employers and scheme members onto taxpayers. See below.

I’ve heard various objections to the idea of a crown guarantee for USS.

Moral Hazard: Some people say there’s a problem of moral hazard if the government guarantees any fund, including a pension fund. Why wouldn’t the fund managers just pick the highest-risk, highest-return investments, knowing that the bill for any losses would be picked up by HM Treasury?

The complaint, in my view, ignores the regulatory structure still in place around guaranteed funded pension schemes. They’re still assessed by the Pensions Regulator. They’re still heavily controlled on what they can invest in.

In one sense ‘moral hazard’ pervades our entire financial system. Why doesn’t the Bank of England’s Sterling Framework (which takes a wide range of assets as collateral in exchange for temporary liquidity) create a moral hazard problem? The answer is that it does, but it’s controlled by a regulatory framework. Financial units don’t get government guarantees for free.

The key is this: the government guarantee is meant to protect against cash-flow risk, not against portfolio risk. The fund is required by regulations to stay solvent. It’s just that unnecessary disputes concerning cash-flow risk are gone. USS, as I’ve argued before, doesn’t seem to me to face any cash-flow risk de facto. A government guarantee would just make what is true de jure match what is true de facto.

Anyway, there are currently several funded pension schemes under crown guarantees. They don’t seem to have crumbled under a problem of moral hazard:

https://www.nao.org.uk/wp-content/uploads/2016/06/Evaluating-the-government-balance-sheet-pensions.pdf — p.42

Nationalisation: Some people complain that a government guarantee is a slippery slope to the full nationalisation of the scheme. The scheme would then be at the mercy of politicians, and we know they’re not always on our side. I don’t know what to say about this except that I would recommend gritting the slope. The schemes listed above don’t seem to be on their way to nationalisation; if USS came under a crown guarantee, it wouldn’t have to be nationalised.

Burden on the taxpayer: Some complain that a crown guarantee would shift the burden of risk off employers and scheme members and onto the taxpayer. But I don’t think there’s much transfer of risk involved. Again, you’re really just changing bonds to equity. If anything is changing here, it’s the classification of HM Treasury’s liabilities to the fund, not where the burden of risk falls.

Additionally, I can’t resist noting, if the fund performs badly in tough economic times, HM Treasury can refinance its liabilities rather than increasing taxes on anyone. This is called a fiscal expansion, and it’s exactly what Treasury should do in hard economic times. If anything, a crown guarantee would equip Treasury with a new macroeconomic auto-stabilizer.

If I give one to you…: Here the objection is one of fairness. If the government guarantees this scheme, why not every DB scheme in the country? Well, I’m not opposed to that idea. But in terms of pragmatic politics, this is the DB scheme over which there is currently an industrial dispute. There are already some schemes covered by crown guarantees. We’re just talking about adding another.

You can argue, of course, that this dispute started in part because guarantees are given to schemes like LGPS. Why, ask USS members, can’t we be given the same deal as the members of LGPS? Perhaps a crown guarantee, resulting in the same deal, would start off a domino effect. Industrial disputes would break out over every DB pension scheme.

Well if that happens, it happens. The point is, we’re in a halfway house right now, with schemes like LGPS guaranteed and others not. If the nation is forced to reconsider its entire pensions system in search of consistency, I can’t see that as a bad thing. See section 4 below.

Here I can be brief. To the hard-headed Conservative politician, I would suggest that the opportunity cost of a government guarantee is negative. It requires only temporary financial support in extreme cases of downside risk. That’s a much lower cost than the cost of ongoing industrial disputes over pensions. The cost of this dispute has already been off the scale. A government guarantee seems a cheap way to get out of it and avoid future disputes.

To the Labour politician, I would promote the scheme as a move towards more equitable risk-sharing. When the media complain about risks being shifted, or costs being imposed, onto the ‘taxpayer’, they make it sound as if we’re in some flat-tax world. But, from a Labour point of view, the nice thing about public funding is that you can draw it from the people most able to afford it. People who can’t work still need income, including the retired. We live in a social democracy in which this is publicly provided. And we live in a progressive-tax society, in which the burden can be placed on those who can bear it most easily.

If I’m really honest, I don’t think pension schemes should exist. When everyone tries to save for retirement, you get a Keynesian paradox of thrift effect. One person’s saving reduces another person’s income, so that the overall attempt by everyone to save just reduces the total income from which saving could be drawn. Saving requires investment: if I am to earn more than I spend, someone else must spend more than she earns.

We’ve built a system designed to maximise paradox of thrift effects. Saving is heavily tax-incentivised and facilitated through large pension funds. Investment is extremely difficult: banks don’t lend readily for capital development; they lend for asset-acquisition, which is a really a way of borrowing money in order to save it. Policy doesn’t adjust to a high-saving world, and economists retain their belief in the power of monetary policy to undo paradox of thrift effects with a zeal that Christ himself would envy.

The problem with trying to guarantee a certain amount of saving for everyone through a pension scheme is that you never know there’ll be enough private investment to sustain the saving. Worse still, investment decisions are based on volumes of sales: current consumption. So when pensioners find themselves with insufficient income, the result is less investment to supply the savings for the next round of pensioners.

The way to solve this is to break the nexus between pensions and saving. Just set up a guaranteed earnings-related state pension, up to some reasonable cap. In high-investment periods, incomes would be high, proportional tax revenues would be high, and the state-run scheme would be fully funded or even in surplus. In low-investment periods, the tax revenues would be lower, the scheme would move into deficit, and in this way it would provide a needed fiscal stimulus. Pensioners would spend their income, making up for whatever shortfall in sales led private investment to collapse.

That’s where I’d want to get to in the long run. Just to lay my cards on the table. But as far as ending this dispute goes, I don’t see much wrong with the idea of a crown guarantee for USS.

Lecturer in Philosophy, University of St. Andrews — personal website: https://axdouglas.com/