Should valuation assumptions be a function of the investment strategy? What does the regulator say about the gilts-plus approach? Paul McGlone from Aon Hewitt, David Weeks from the Association of Member Nominated Trustees, Jonathan Reynolds from Capital Cranfield, Leslie Scrine from the M&G Group Pension Scheme, Andrew Cheeseman from Pan Trustees and Andrew Young from the Pensions Regulator discuss scheme valuation methods.

Andrew Young: I always like to say, ‘Are we talking about gilts plus a fixed amount or gilts plus something that is variable in some way?’

Simply assuming gilts plus a fixed amount for each actuarial valuation is fundamentally wrong. The process is wrong. The answer may happen to be right. Trustees should however be able to give evidence as to why they consider that their gilts plus something fixed is appropriate this time around.

In practice the valuation assumptions become a definition of a model portfolio against which you are going to be benchmarked

Leslie Scrine, independent trustee

Leslie Scrine: On funding, the Pensions Regulator’s code basically says trustees should start from the yield on gilts in framing their valuation assumptions, and there are then about 20 paragraphs amplifying that, including one saying that if you are invested in equities you might wish to allow something for the expected returns over gilts but you must be very prudent about it. That was what pushed us into gilts plus.

Andrew Cheeseman: I sign off about 20 valuations a year as a chairman of trustees, and I know the first thing that the regulator is going to say to me is, ‘Why did you move from using a gilts plus half a per cent and make the assumptions less prudent?’ Because of this our actuaries will normally be steering us to keeping the same assumptions, which I do not agree with.

It seems to me that our assets should be reflecting integrated risk management. So we set our stall out with our assets, and then we look at our assets and our cash flow to work back to what our assumptions should be. Not set the assumptions and then end up by saying, ‘Oh well, because we have set all the assumptions for the liabilities, we are going to have high volatility if we do not match our liability modelling, so let us all put our money into gilts’.

David Weeks: We had a canter round this ground at our last quarterly conference of the Association of Member Nominated Trustees.

Their view was the valuation should reflect the circumstances of the individual scheme more. Are we looking at a defined benefit scheme that is moving pretty close to buyout, or one with people coming in at the bottom?

The one that perplexed the most was the question of how you are basing it on discount rates. Lay trustees find it quite difficult to grasp how a discount rate is set and how a small change in an assumption can make a big impact on the liability.

There needs to be greater awareness among trustees of what the purpose of the valuation is and what you are seeking to achieve with the information; it is a guide, not an immutable rule.

Jonathan Reynolds: My feeling is that it is all about where the drive for the alternative valuation method comes from. If it is purely there to make the technical provisions more palatable, I would question whether that is a valid way of doing it. Ultimately, what we are looking for is the security of member benefits, and what I am actually interested in is, ‘Does this recovery plan get me to where I want to be?’

A lot of that depends on the future of the scheme; if I have got a plan to buyout then gilts plus across the curve, gradually reducing the plus amount over time, makes sense.

If I have a long scheme – dare I say even an open scheme – then it is fundamentally different; maybe one of the stochastic-type valuation bases might work well there.

For the majority of schemes, I feel the current valuation method serves the purpose well, so long as you know exactly what you are trying to achieve and how you are going to get to it.

Scrine: In practice, the valuation assumptions the actuary chooses become a definition of a model portfolio against which you are going to be benchmarked.

There is a bit of a chicken and egg situation here; the investment strategy should be decided by the trustees independent of the valuation

Andrew Young, the Pensions Regulator

So if the actuary is valuing on gilts plus 1 per cent, you are going to try to devise a strategy that gives a high confidence of getting gilts plus 1 per cent without too much volatility.

If the actuaries, as they used to, value off a growth dividend model using the yield on the FTSE All-Share, you are quite likely to devise an investment strategy that aims to outperform the FTSE All-Share.

Young: There is a bit of a chicken and egg situation here; the investment strategy should be decided by the trustees independently of the valuation. The valuation is a tool that comes later to manage the scheme.

Paul McGlone: An important distinction, and we are seeing it more and more, is whether the assumption you are making in your discount rates is meant to be a prudent estimate of the expected return you are going to get from your portfolio, or whether it is completely unrelated to that.

On the one hand you might say, ‘We are discounting our liabilities at gilts plus 2.5 per cent pre-retirement because we think that is a sensible expected return on our portfolio,’ but I have schemes that set their discount rate as gilts plus half a per cent because that defines the end position they want to get to. It bears no relation at all to the investment strategy they want to run to get there; it is purely about saying, ‘I want to be within a certain distance of buyouts,’ perhaps.

By setting gilts plus a margin as the place you want to get to, you can drive investment behaviour you didn’t intend. Your actuary may say, ‘If you do not like volatility then you should do something about hedging it’, and the investment consultant says, ‘There is a good way of hedging it and it is called LDI and gilts,’ and you back yourself into an investment strategy which may not be what you intended.

Young: All closed schemes should have a long-term target, and most of them are quite likely to have a relatively low-risk target. The issue is how you get there, what returns you want to achieve and how prudent you want to be over that period as well.

I do not see very many people saying we are going to have a valuation basis of gilts plus 50 basis points throughout, but investing in equities for the first 15 years.

McGlone: I see quite a few who say, ‘Our long-term target that we all recognise is, let us say, gilts plus a half, but we do have a strategy that does something quite different in the meantime with the aim of getting us there.’

In fact, the regulator is encouraging us to use that on schemes that have weak sponsors, saying, ‘Your technical provisions should be your endpoint you want to get to and, if you have an asset strategy to help you get there, put it in your recovery plan’.

Young: The question then is, ‘What is the assumption of the recovery plan?’ Whether you have it in the recovery plan or the technical provisions, it comes to the same thing in practice.