Towers Watson’s latest study of the assumptions used to calculate the pension numbers in the accounts of FTSE 350 companies with 31 December year-ends has found that:
• Greater accuracy is working in companies’ favour: For a typical defined benefit (DB) scheme, a seemingly technical decision over which bond yields to use in calculations could change the liabilities disclosed to investors by around ten per cent. Companies have been able to disclose smaller pension liabilities in their accounts by more accurately reflecting how much time will elapse before pensions must be paid out.
• New accounting rules will reduce recorded profits: When 2012 accounts are restated to reflect new accounting rules that were introduced this year, the pre-tax profits of companies with DB schemes will fall by an average of around 3 per cent – or around £5 billion for the FTSE 350 as a whole.
• Life expectancy assumptions are stable: On average, companies assumed that male scheme members aged 65 in 2012 would live to the age of 87.5 and females to 89.4. Life expectancy assumptions are very similar to last year’s.
• Deficits have grown since accounts were prepared: Aggregate FTSE 350 pension deficits were similar at the end of 2012 as at the end of 2011 but have since increased, from £56 billion to £70 billion by early June.
Pensions on the balance sheet – more accurate calculations produce smaller liabilities
Under the IAS19 accounting standard, expected future pension payments should be converted into a single liability number using the yields on high quality corporate bonds of a duration that is consistent with the company’s pension commitments.
Yields on bonds in the iBoxx AA >15-year corporate bond index fell by 0.6 per cent between 31 December 2011 and 31 December 2012 (from 4.7 per cent to 4.1 per cent). Although some companies calculate their pension liabilities using this index, the average discount rate disclosed in company accounts was higher to begin with and fell less sharply (from 4.8 per cent to 4.4 per cent). Higher yields make liabilities look smaller.
Neil Crombie, senior consultant at Towers Watson, said: “The explanation for this is that many companies are not just looking at an index; they are going for more precision by weighting bond yields according to when their pension payments fall due.
“Until the past couple of years, the approach used would not have made much difference, but going for greater accuracy currently works in a company’s favour. This is because very long-dated bonds now pay higher rates of interest than bonds that are merely quite long-dated. Even with an index of long-dated bonds, the average duration will be shorter than a typical company’s pension liabilities.
“The discount rate is the most important assumption that a company must make when preparing pension numbers for its accounts and seemingly technical decisions can make a big difference to the balance sheet. While many do this already, all companies will in future have to highlight to investors how sensitive the results are to the assumptions chosen.”
Profits warning on accounting standard change
For accounting years beginning on or after 1 January 2013, revisions to the IAS19 accounting standard mean that the credit on pension fund assets must be the same as the discount rate. Previously, if the company’s expected return from their pension scheme’s assets was higher than the discount rate, this would lead to higher profits being recorded. Amongst companies who disclosed an average return on assets in their 2012 accounts, this exceeded the discount rate by an average of 0.8 per cent.
Neil Crombie said: “2012 profits will have to be restated next year so that investors can compare results on a like-for-like basis. Across the whole FTSE350, 0.8 per cent of pension scheme assets would be around £5 billion or around three per cent of pre-tax profits for companies with DB schemes.
“We don’t think that the ability to record bigger profits by investing in riskier assets was acting as a significant brake on pension scheme asset de-risking. Most companies were already more interested in trying to reduce the volatility of the pension numbers on their balance sheets.”
Life expectancy assumptions stable
On average, FTSE350 companies with 31 December reporting dates assumed that a male scheme member aged 65 in 2012 would live for 87.5 years, compared with the average figure of 87.3 years disclosed in 2011 accounts. Females are on average assumed to live for 89.4 years, compared with 89.3 in 2011. Current 45-year-olds were on average assumed to live for 1.8 years longer than current 65 year-olds.
Neil Crombie said: “At first sight it looks as though companies have revised up their life expectancies slightly but this is not really the case. They were already assuming that someone aged 65 in 2012 would live longer than someone aged 65 in 2011. Taking this into account, the assumptions have hardly changed.
“There is a difference of about six years between the highest and lowest life expectancy assumptions. This is partly because schemes are looking in detail at what the characteristics of their members mean for life expectancy assumptions. Another reason is that only some companies will have stripped out the prudent margins they use for other purposes when disclosing their best estimates of members’ life expectancy, though the proportion of companies that do this is increasing.”
What has happened since the year-end?
On 31 December, when snapshots of pension schemes’ assets and liabilities were taken, markets factored in some likelihood of the way RPI inflation is calculated being changed. This would have produced lower inflation-linked pension increases. On 10 January, the National Statistician announced that this would not happen after all.
Neil Crombie said: “Unpicking the change-that-never-happened means that companies now have to budget for bigger pension increases. Wrapping this together with other market changes, we estimate that aggregate deficits in all FTSE 350 pension schemes will have grown from £56 billion to £70 billion since the end of 2012.”
|
