What alternative valuation methods can schemes use?

Paul McGlone: You can do a stochastic valuation where you do not have any assumptions about liabilities and discount rates; you model the assets and you say, ‘How likely is it in 25 years that we will have enough money to pay the residual lump of people given we know we can probably buy an annuity at something like gilts at that time?’

If a different methodology became commonplace and was found useful and appropriate, I am sure the Pensions Regulator would not stand in its way

Andrew Young, the Pensions Regulator

That has some validity to it, but at some point you still need to convert it into a liability for reporting purposes, which feels like an unhelpful step.

Andrew Young: If people move towards that kind of approach to arrive at their recovery plans, we do need to think about the sensible way to report it.

If a different methodology became commonplace and was found useful and appropriate, I am sure the Pensions Regulator would not stand in its way.

Jonathan Reynolds: Do you actually have schemes that work on that basis, Paul?

McGlone: Not many. The majority of schemes probably still express their discount rate as gilts plus a margin. They may construct it in other ways, but then express it in that way for convenience. The number who do a full stochastic valuation is limited.

Reynolds: What is seen as an acceptable probability?

McGlone: If it is too high the model will tell you to invest everything in equities, and if it is too low it tells you to invest everything in gilts. Commonly it is somewhere in the region of 60-70 per cent.

Reynolds: As I was coming here today I thought, how would this play with members if I was saying to them, ‘We are basing our investment strategy, our valuation, on a 75 per cent chance of being able to pay you your full pension’ - would people feel comfortable with that?

McGlone: Let me answer a simpler question: what probability of success does gilts plus 2 per cent for a growth portfolio have over a 10-year period? It is about 70 per cent. That is the level of certainty you have embedded into a typical strongish sponsor assumption. It is not high.

David Weeks: To put it another way, you could say you have a three in 10 chance of not having these benefits.

McGlone: Well, you have a three in 10 chance of having to ask the sponsor for more money.

Scrine: And you have a three in 10 chance of exposing members to the risk of not being paid.

Weeks: The member may then turn around and say, ‘At the moment we have two options; either keep the current level of benefits, or we have to go into the hands of the Pension Protection Fund, and there is no halfway in the middle.’

One of the issues is whether there should be some sort of halfway house, whereby there is some way of abating the level of benefits and so strengthen the opportunities for coming out with a better valuation.

Reynolds: Ultimately, it all comes down to how much the employer is paying in. 

Andrew Cheeseman: If you work out your cash flow and say, ‘My current portfolio has a 70 per cent chance of being self-sufficient’ or whatever it happens to be, you still have the opportunity to be flexible with the 30 per cent, and that is where the employer comes into it on how he is able to pace that.

McGlone: If for 29 per cent out of the 30 per cent the employer can stump up enough, you are fine.

Cheeseman: That is right; as you say, it is that bit in between. This is where we come back to integrated risk management.

McGlone: And valuation methods do not pay benefits, in the end.

Young: They are just a tool for arriving at decisions. That is all they should ever be, but I can see that they then become targets for investment returns because that is the assumption that is there. That is a dangerous move.

Scrine: For us, it was not just the ‘gilts plus’ valuation that made us start investing in gilts. Part of it was that revaluation promises to preserve leavers’ benefits and the fall in inflation meant there was no longer a great deal of discretionary benefit.

I am interested in [Andrew Young’s] point that gilts plus fixed and gilts plus a variable margin are different. I can see how that might work for us.

Young: You start from the assets you hold and work out your expected return on each type of asset.

You should allow for the fact you may be derisking in future. Allowing for that, what is the expected return on your portfolio? That might be 3 per cent, say, and you could express that with reference to a gilt return, which might be gilts plus one; that plus one is derived from having done all the work on the return on the assets. It is really just a way of expressing it.

There would be a case for expressing your assumptions describing the process that you go through, rather than the number you come up with

Paul McGlone, Aon Hewitt

McGlone: The way we see it working in practice most commonly is, a trustee board will say, ‘We would like an assumption that has a 70 per cent chance of happening.’ We might say, ‘Gilt yields are 2 per cent and our best estimate return on an equity portfolio is 7 per cent, and if you want an assumption that has a 70 per cent chance of happening, that is 5 per cent, so that is gilts plus 3 per cent.’ So we then set the assumption as that for the next three years.

At the next valuation, we might say, ‘Nowadays, you can only assume gilts plus 2.5 per cent if you want a 70 per cent chance of success, so let us re-express the valuation this time around.’

You are then left with that disconnect and, from a hedging perspective, that is quite interesting; you have hedged, perhaps, 100 per cent of gilt yield movements and then I come in at the next valuation and change the plus, which means I have just wiped out a big piece of the hedge you have put on. That may be wiping out in a positive or a negative way, depending on which way the market has moved.

Scrine: But it is interesting. Could you deal with the practicalities, so that every quarter the ‘plus’ moves a little bit rather than there being a huge change every three years?

Pensions Expert: Would that be a case for more frequent valuations?

McGlone: I think it would be a case for expressing your assumptions in your statements of funding principles in a more detailed way that describes the process you go through, rather than the number you come up with.