All entries for Monday 06 October 2014

USS employers shown to have falsified life expectancy figures

One of the chief reasons given by university employers for the changes to USS that they want is that we are all living longer in retirement, life expectancy has increased.

But my colleague Jane Hutton, professor of statistics at Warwick university, has shown that they have been overstating their case and posted false figures on their website.

The Employers Pensions Forum (EPF) Q&A webpage originally stated (Q9): "Current longevity patterns are significantly different to those when the scheme was set up in 1974. Then it was expected that a USS pensioner retiring at age 65 would live for 6 to 8 years in retirement so the cost of the scheme and the contribution rates were set on this basis. By 2014 the anticipated length of retirement is around 30 years, ..."

In other words, in 1974, 'our' life expectancy was half (49%) that of the general public, but by 2014 it had risen to 1.4 to 1.6 times greater! The Office of National Statistics has an increase of 1.3 to 2.3 years for each decade; the EPF has an increase of 5.8 years for each decade.

Jane wrote to the EPF pointing out that these figures are hardly credible since they would require us to believe not only that life expectancy was grossly underestimated when USS was set up in 1974 but also that rapidly increasing life expectancy had been ignored subsequently.

She did not receive so much as an acknowledgment of her email. But when a colleague looked at the same webpage he could not find the numbers because they were no longer there. Fortunately, Jane had printed the original version of the webpage as it had been (see Q9). The new page with the offending figures for life expectancy removed is here, but with no indication it has been edited and the same date as before.

Dennis Leech : 06 Oct 2014 17:38 |  Tags: Pensions; Uss |  Comments (1) |  Close comments |  Report a problem

Missing step in pension fund accounting: more information on the USS deficit needed

Not infrequently we hear that a pension scheme has a funding deficit, its assets failing to match its reported liabilities, while at the same time it is solvent in the sense that its income exceeds its outgoings. This seeming paradox demands an explanation yet the accountants and actuaries do not seem to think one necessary.

A salient example is that of the USS, whose liabilities were estimated last September to exceed its assets by some £13.1 billion. Yet last year its net surplus was £1.2 billion - income from contributions plus investments of £2.6 billion less payment of benefits of £1.4 billion, a sizeable figure which, while of course telling us nothing about the future pressures on the scheme, nevertheless suggests that there could well be considerable headroom before any solvency problems, if any, emerge. Or even that the deficit is a statistical artefact and not a useful guide to action.

The reported assets and liabilities figures for a large scheme like the USS, which covers a large a sector of the economy, are largely notional in that there are no sensibly conceivable circumstances in which they will ever need to be paid. They come out of the black box of finance theory. What matter are the real flows of income and expenditure every year and how these evolve over time.

There is a missing step in the accounting procedure for pension funds: what is needed is exactly how it is that the notional balance sheet deficit becomes manifest as an income/expenditure deficit and at what point in time that happens. Or if in fact it never happens. It should not be too difficult to provide that information.

(Letter submitted to the Financial Times)

Dennis Leech : 06 Oct 2014 11:59 |  Comments (0) |  Close comments |  Report a problem