The effects of the recent storms have highlighted the importance of long range weather forecasts to give us time to prepare for the anticipated problems ahead. The evidence suggests we are in for a lengthy warm spell, which means sea levels are going to be rising. Although we do not know for sure how events will actually unfold, no-one is seriously suggesting that carbon emissions are cooling the planet. So scientifically-derived models projecting possible future outcomes are useful in deciding what to do in the here and now. We might not like what we see, but at least we do not have to wait until the last polar bear sinks from view before we realise that something is amiss. We have the opportunity to avoid the disaster, or at least mitigate its effects.

How to avoid pension scheme deficits
So why do so many pension schemes have huge deficits? Were there no similar projections available ten or more years ago that could have warned trustees of the effects of overheating in the financial world? Sadly the answer is yes, but there was relatively little serious interest in gloomy forecasts when surpluses were large and everyone was on a contribution holiday. Asset liability modelling ("ALM") is the unsnappy name for a very valuable tool. What it offers is not just a projection of how the assets grow but also how the liabilities of a scheme might be expected to develop too. It is designed for Defined Benefit (DB) schemes, and whilst these are losing favour (at least with private sector employers) there are an awful lot of them still about, and they are likely to remain of very substantial size for a very long time. So it would be nice to know what the next valuation might bring, or what the position might be at the end of a deficit repayment period.

The chart shows an example of the projected funding level for a DB scheme at the next valuation for a range of investment strategies. Each bar represents a different strategy.The height of a bar shows the range of projected outcomes.

Reliability of the projections
There is also the level of certainty to consider: how reliable are the projections? The calculations can answer this question too. Reasonable estimates can be made of the likelihood of different outcomes. Given the variability of asset prices in the past, probabilities can be assigned to asset values in the future. This gives an indication of whether a projected median funding level is a very good indication of the likely outcome or whether there would be a wide spread of likely outcomes.

Who might find an ALM study useful?
Trustees of DB pension schemes, whether these are closed to future accrual of benefit or not, should find such an analysis invaluable in reviewing investment strategy. Many already do. The assumptions used to project future funding levels can be the same as the funding valuation. Employers, too, are increasingly interested because an FRS17 or IAS19 basis can be used. This will allow an employer to see the projections relevant to a company’s accounts. There are also proposals from the UK Accounting Standards Board described as “horrendous” or reasonable, depending on whom you ask. The ALM could take these into account, too, so that projections of the effects of these proposals could be assessed well before discussions were concluded.

What does an ALM do?
An ALM study looks at a wide range of different investment scenarios, representing the current mix of assets and a range of alternative mixes. These could be used to explore the bounds of possible asset choice and more practical, realistic alternatives. The details and results relating to the most recent actuarial valuation are also captured. Then the sophisticated software does its work: rolling both the assets and pension benefits forward in time. The main items of interest are the funding level and the size of the scheme deficit or surplus. These are the main items that are output from the analysis.

When is the best time to conduct an ALM study?
The best time to conduct an ALM study is immediately after an actuarial valuation. The reason is that this is the point when the liability data are most up-to-date. However, this does not preclude studies being undertaken at other times: the liabilities would then have to be rolled forward appropriately.

Summary
We may not like projections of the effects of global warming any more than we would have liked projections of pension scheme deficits. But good indications of possible futures allow action to be taken, whilst there's plenty of time left.