Estimates of pension scheme liabilities facing the UK’s largest companies vary by around 20% depending on the assumed life expectancy of members, according to new research by Mercer Human Resource Consulting.
The consultants analysed 28 FTSE 100 companies – representing 49% of the total market capitalisation of the FTSE 100 – of which more than half provided information on the mortality rates used last year. Converting the information to allow meaningful comparisons revealed that most assumed life expectancies for a 65 year-old of 18 to 22 years. The lower the life expectancy assumption used, the lower the value of a scheme’s liabilities will be.
Mercer says that voluntary disclosure in the annual reports of some companies means that for the first time this year the stringency of the mortality assumptions used by these companies for their employees and pensioners can be assessed. This data is important in evaluating the prudence of pension deficit calculations under International Financial Reporting Standards, says Mercer.
“Given the social and geographic differences in memberships, one would expect pension schemes to use a range of mortality assumptions,” says Tim Keogh, Worldwide Partner at Mercer. “But those schemes that use particularly low assumptions may come under increasing pressure to justify their positions.”
In the sample of companies analysed, the distribution of assumptions used was roughly even between 18 and 22 years, with a single outlier company that used an assumption of 16 years. It is widely accepted that there are significant variations in life expectancy between different occupational groups and in different parts of the UK, so some variation is to be expected, says Mercer.
“When assessing the impact of a pension scheme on a company’s value, a challenge for equity analysts is to make sense of life expectancy assumptions without any information on a scheme’s membership,” explains Keogh. “In some cases this will be obvious, but many schemes have groups of members with very different characteristics to the current workforce, and without this information misleading conclusions can be drawn.”
The top-end life expectancy figure of 22 years is consistent with the “PMA92 medium cohort” mortality projection, based on data extrapolated from those receiving annuities from insurance companies, says Keogh. Most purchasers of these annuities had white collar occupations.
The study was conducted as part of Mercer’s regular review of pension deficits and trends in the FTSE 350 companies and covered the period from 31 December 2004 to 31 March 2006. The story, says Mercer, was one of pain and gain. The consultancy says that in many respects 2005 was remarkably like 2004 – assets performed well in absolute terms, but the value of liabilities also grew.
At 31 December 2005 deficits were well above 31 December 2004 levels, although bond yield increases thereafter meant that the position at 31 March 2006 (gross deficit 68bn) was slightly improved from that at 31 December 2004 (gross deficit 75bn).
The study provided further evidence of a gradual rather than dramatic change in investment strategy, mainly away from equities (around 4% of equity holdings appear to have been switched to other asset classes during 2005).
On average, companies contributed around 60% more to their schemes than the cost of covering new benefits, says Mercer. But this statistic disguises an underlying pattern where a few companies are making very significant contributions on a sporadic basis, while the rest are contributing little more than enough to cover the cost of new benefits. “Even based on current contribution rates, deficits will not be eliminated for some considerable time unless there is a dramatic rise in equity markets relative to bonds or bond yields continue to rise,” says Keogh.
While the trend to close defined benefit schemes to new entrants has slowed in recent years, a number of companies are now going a step further and closing their schemes to future accrual – Mercer estimates up to 5% of FTSE350 companies have now done so.