LONDON: Actuarial science may be one of the more modern forms of wizardry.
By altering the assumptions about inflation, investment returns, bond yields and salary growth, actuaries can make pension scheme deficits mushroom or shrink at a whim. Since those forecasts require peering into the future, and no one has a crystal ball to say what will happen with certainty, a wide variety of forecasts is not only predictable but, arguably, legitimate.
Indeed, that fact was one of the key reasons why the UK's accounting profession introduced the rule known as FRS 17.
It was intended not only to make pension obligations more transparent to investors but also to enable them to compare one company's accounts against another.
A new study by Pension Adviser Review, a provider of consulting services to trustees and companies, shows that some firms of actuaries are particularly inclined to recommend assumptions that help to flatter accounts.
Even within firms, different clients may have widely varying assumptions built into their accounting records, suggesting that some actuaries are prepared to make recommendations that clients want to hear
The study's conclusions were bolstered by another recent report issued by actuarial consultants Lane Clark & Peacock -- a firm whose own range of assumptions are generally more conservative than most -- which looked at the FTSE 100 companies.
Even for companies with the same year-end, the report found a wide range of assumptions about such things as inflation and discount rates that ought not to vary significantly from company to company.
Keith Faulkner, managing director at PAR, said analysts looking at company accounts should perhaps inquire about which actuarial practice helped to prepare them. "If you use actuarial firm A, B or C, that might affect the answer to the size of pension liabilities," Mr Faulkner said.
The report from PAR was the first to look at what has long been a troubling issue in the pensions world but which is considered so sensitive that few will make any public remarks.
However, it has long been common currency that some firms of actuaries, and some partners within those firms, will recommend assumptions that allow companies to pay less into their pension schemes. Other firms and individuals have been associated with the reverse.
Indeed, current and former partners at Mercer Human Resource Consulting -- which is shown in the PAR study to recommend liberal assumptions -- were the actuaries whose recommendations of conservative assumptions led the Boots pension scheme to move its assets into bonds.
Watson Wyatt's median discount rate, at 5.40 per cent, was in line with that of many of its competitors, although the difference between the largest and smallest rates used for any of its clients was 0.95 percentage points. The smallest range, 0.20 percentage points, could be found at Hymans Robertson and Lane Clark & Peacock.
Mercer had the widest "gap" between discount rate and salary rises, which PAR says is more meaningful in terms of the effect on liabilities than just looking at the discount rate or the rate of salary increases in isolation.
Mercer's gap was 3.01 percentage points while Watson Wyatt's gap was 2.85 percentage points. Those two firms advise 110 of the 188 companies in the survey.
Hewitt Resources, the UK's third-largest actuarial firm, advises a further 39 clients. Its "gap" was a far more conservative 1.9 percentage point while LCP was the most conservative at 1.4 percentage points.
The UK company with the single largest gap between these two numbers in its accounts to the end of 2004-05 was J Sainsbury, a Watson Wyatt client, which had a spread of 2.75 percentage points.
He added that the variations found within actuarial firms on such things as the discount rate was very surprising because accounting rules clearly state that the rate must be that of AA corporate bonds of appropriate duration.
That variation also shows up clearly in the LCP report on pensions. Companies whose accounts were dated at March 2004, for instance, had discount rates ranging from 5.20 per cent (Gallaher) or 5.25 per cent (Reuters, William Hill) to much less conservative assumptions of 5.4 per cent at ITV and Royal Bank of Scotland and 5.45 per cent at HBOS.