Secrets of Success: Have the actuaries got the sums wrong?

Is it possible that virtually all pension fund providers, not just Equitable Life, are understating their ability to meet future claims? This issue has exercised the pensions and investment industry for some time. The question has implications for all of us, not just for the big corporate schemes that most worry regulators and the stock market. It also strikes at the heart of how we should be making our investment plans.

The most recent pundit to start stirring this pot is the City economist Andrew Smithers. He says the way most consultants and actuaries calculate pension liabilities is way off beam. In his view, there is a bias in the calculations which seriously overstates the returns likely to be achieved by pension funds (or indeed by any of us).

He is not alone in this line of thought. Actuaries work out how much money companies and to a lesser extent individuals should be putting aside to generate the pension that employees expect. Stage one is to work out what the future liabilities are - in other words, what pension has to be provided in the future. Stage two is to calculate the amount that has to be put aside in order to meet that liability. This in turn depends on three assumptions: how much money is put aside each year to be invested, how long there is to run before the invested fund has to be realised, and what rate of return you assume can be made in that period.

The tide has turned so sharply against most defined final-salary company pension schemes partly because of their lack of flexibility. Companies have woken up to the fact that committing yourself to fund a pre-defined level of pension at a fixed date, the employees' retirement date, could be a hugely expensive burden. A final salary scheme is a benefit whose value is guaranteed in advance.

As such it is not all that different from the guarantees that have provided so much grief for Equitable Life and other with-profits pension providers. The Equitable case revolved around the management's belief that they were entitled to ask policyholders with guaranteed annuity rates to (in effect) pay for them by accepting a lower terminal bonus, should investment returns be insufficient to provide a big enough pension pot to meet the claims.

When the House of Lords said they could not do that, it virtually finished the society off. Guaranteed policyholders got a benefit that nobody had provided properly for, and which could only be paid for by other members of the society. In the same way, companies that are unable to wriggle out of the obligations they have guaranteed to their employees in the past could also find themselves struggling. The difference is that companies which are still in business do in theory at least have the chance to earn enough from future trading to make up the shortfall.

But the assumptions you make in forecasting the pieces of the pensions jigsaw that are not fixed in advance are just as important in working out whether the right amount is being put aside. Retirement ages can be extended, and almost certainly will be, but future investment returns are another matter. Forecasting them can never be an exact science, and there is clearly room for discretion in making assumptions.

But who is kidding whom if the assumptions are too optimistic? The point that Mr Smithers makes is that actuaries and pensions consultants are in danger of overestimating the potential for future investment returns by a large margin.

The rate of return from the stock market is central to this issue. For many years most pension funds have allocated 50 to 70 per cent of their assets to the stock market, in the belief that this is justified by their higher expected returns.

It certainly worked like a dream in the period up to the market peak in 1999, but is that the right policy? Only, says Mr Smithers, if you assume that stock market returns follow the pattern of the most recent past, not their long-term course. As the chart shows, the returns from shares in the past 30 years have consistently exceeded the long-run average of 6.5 per cent in real terms. Yet many companies and consultants work on the assumption that the long-run rate of return is still the basis on which to calculate their future liabilities.

If you believe that in the very long run, the average return must revert to the long-run average, it follows that there is a greater risk that your pension fund will come up short in the next 10 to 20 years. Most probably future returns will be lower than the long-run average: 4.5 per cent per annum for shares seems a more reasonable assumption. In that case, you will either need to put more aside to increase the likelihood of obtaining the pension you once expected, or change your investment strategy.

It is legitimate for individuals to adopt a riskier investment approach in order to make their retirement potentially happier. What is not sensible, says Mr Smithers, is for companies to assume in advance that the increased risk of owning so many shares will pay off. It doesn't help that sticking to current return assumptions makes company balance sheets look better and encourages today's company bosses to leave the problem to their successors.

At least the Chancellor seems to see the problem. His pension simplification rules introduce a degree of greater flexibility for those still working. The gulf between those with guaranteed pensions and those without can only grow wider in the next few years if the actuaries have got their sums wrong. You won't err by being realistic about what is coming.

jd@intelligent-investor.co.uk