Time to blow the lid on pensions secrecy

INDUSTRY VIEW

There was more than met the eye in British Gas's disclosure last week of Cedric Brown's handsome pounds 247,000 a year pension. As well as stirring up yet another row about fat cats, it also showed which way the wind was blowing over the Greenbury Committee's plans for disclosing the true costs of executive pensions.

After intensive lobbying by corporate Britain, they may be about to be watered down into a form that is rather more palatable to boards of directors than the original proposals.

The then director general of the CBI, Howard Davies, quietly pointed out when the Greenbury report was published in July that its innocent seeming recommendation that the full value of pensions should be disclosed was far tougher than people appreciated.

How right he was. By last autumn, by which time Mr Davies was safely away from the eye of the storm at the Bank of England, it was clear that the pay scandal would begin all over again if the recommendations of the committee - which was set up by the CBI - were taken at face value.

For example, actuaries calculated that a 55-year-old chief executive who receives a 50 per cent pay rise to pounds 500,000 a year could be shown in the annual report as benefiting by as much pounds 2.5m during the year he received the increase, on top of the annual pay package.

There was horror in many boardrooms as the implications sank in. The numbers to be disclosed would dwarf even the profits made on privatised utility directors' share options.

As well as magnifying the cost of genuinely earned increases, such as promotions, it would also put an end to an old Spanish custom in the boardroom of giving directors a pay rise to boost pension entitlements before departure. The cost of this disguised golden handshake has been invisible in annual reports.

Whether or not the motives for individual pay increases are respectable, the next annual report season would be a meaty one for the press and the opposition. It has not been lost on Conservative politicians, who know what a gift the fat-cat scandal was to Gordon Brown, the shadow chancellor, that this would be uncomfortably close to the election.

A second consequence of full pensions disclosure is that it would throw up such large rewards that boards may no longer be able to give out-of- the-ordinary pay increases at all, for fear of the spotlight to be put on them in the annual report.

Most people would regard this as no bad thing, of course, but it would perhaps cramp the generosity of many a remuneration committee.

So what was the significance of British Gas's decision to reveal Mr Brown's actual pension when he retires in April, when the company was not required to disclose it? It is simply that the CBI yesterday came down in favour of exactly that method for disclosing pension benefits.

Martin Broughton, chief executive of BAT and chairman of the CBI companies committee, rejected proposals to capitalise the increase in value of directors pensions each year, the method that would have led to some of the astronomical published figures mooted by actuaries.

Like any argument in which actuaries are involved, this one takes concentration to get the mind around. There are in fact five feasible methods for disclosing directors' pensions.

The present method used in annual reports is to show the cash contribution to the fund by the company, on the director's behalf. Everybody agrees this has become meaningless, because if the pension fund is in surplus there may be a contribution holiday and the cash cost is nil.

Another method that has virtually no support is called notional funding. This is based on calculations of the size of pension fund needed at retirement to pay the promised pension. But it is hard to follow and throws up nonsenses, including enormous negative remuneration for directors who leave early.

A method supported by a significant number of companies is to use SSAP24, the accounting standard already employed in company accounts to describe the pension fund as a whole.

This spreads pension costs out, avoiding huge annual leaps, but it would be difficult to interpret the figures for individual directors, and like the previous method it can also show negative values for early leavers.

In practice, then, two contenders are left. The one the CBI now favours is called accrued benefit. It certainly has the advantage of simplicity and ease of understanding.

As the table shows, what it does is give the extra annual pension at retirement which has been earned during the course of the year, as well as the total pension to be paid. This is a straightforward calculation based on salary and scheme rules.

Finally, there is the transfer value method which is favoured by senior members of the Institute and Faculty of Actuaries. This shows the increase in the transfer value of a director's pension during the year - in other words what he or she would be entitled to transfer as a capital sum into another pension fund, on leaving the company.

The method is in fact closely related to the previous one. What it does is take the extra pension earned during the year and, using actuarial assumptions, calculate the increase in the size of the fund needed to pay for it - hence the very large numbers it shows.

When actuaries were asked last autumn to come up with detailed proposals for implementing pensions disclosure, they proposed a transfer value method as closest to Greenbury's recommendation that companies should give the "value of a directors' pension entitlements earned during the year."

That was the point at which some big companies hit the roof. The actuaries were sent back to the drawing board with fleas in their ears and the result last month was a neutral consultation document explaining the pros and cons of all five methods. It was to this document the CBI was responding.

Those in favour of disclosing actual pension payments, the method chosen by the CBI, say it is simple and easy to understand, comparable between companies and capable of including the impact of discretionary payments. It certainly has the virtue of disclosing real money to be paid over.

For directors, the CBI method also has the virtue of making it harder for newspapers to write abusive headlines, since the total annual package will look much less dramatic.

Advocates of this method claim the alternative of disclosing the increase in transfer value is misleading and likely to be misunderstood, because it is based on actuarial calculations and represents an artificial sum of money the director never actually receives.

This misrepresents the case for disclosing transfer values made by the Institute and Faculty of Actuaries. They are just as much a matter of factual analysis as the pension payments the CBI favours. They are simply the cash payment that would be made to another pension fund if the director left. In other words, they are the director's accumulated pension wealth, so it makes a lot of sense to disclose on that basis.

As the two graphs show, a key reason for the enormous annual costs the transfer value method would give in some company annual reports is that many boards pay their directors' pensions on the basis of their latest year's salary.

If they were to switch to three-year rolling averages, the disclosed amount becomes much smaller each year and would alleviate the much feared publicity problem at a stroke.

It might also occur to some companies that the best way to eradicate the problem is not to pay ridiculous salary increases at all.