A suffocating man craves the foulest stench.


Such is the pressure on trustees and sponsors of our defined benefit schemes to appear solvent, that they welcome the transfer of their potential pensioners to “cash” as a suffocating man breathes fetid gas.

It is within the capacity of trustees to reduce cash equivalent transfer values (CETVs) where they consider there is insuffecient money in the scheme to meet the full payment, but very few trustees trigger the “insuffeciency report” that allows this to happen.

The result is that schemes pay CETVs in full that are likely to do long-term damage to the scheme , though the act gives short-term respite. To understand this extraordinary state of affairs, we need to delve beneath the covers and understand how schemes can use different discount rates to value the same liability. In dong this, we can begin to understand the inanity (in not the insanity) of applying a mark to market valuation system to long term liabilities.

I write as a non-actuary, you may read as an actuary – if I mistake, forgive.

Here’s how it works

Actuaries give a long-term estimate of the likely returns on each class of assets. This is based on history and adjusted to the particular circumstances of today. Although there are times when bonds outperform equities, the long-term trend is for the reverse to happen. This is what is called the equity risk premium.

Schemes that invest 100% in bonds get none of the equity risk premium. Transfers that are offered from these schemes are very hight, that is because the “discount rate” that is the rate at which the long term cost of paying a pension is reduced by investment returns is very low. At the moment, the discount rate using gilt returns is lower than the projected rate of inflation. That is what is making some transfer values extraordinarily high.

Schemes that invest primarily in equities have much higher discount rates. This is because the assumptions of long-term returns on equities give for a much higher anticipated return. This means that the future liability of paying a pension is much lower. Schemes which invest in equities give much lower transfer values.

An ordinary person would question why any trustee would deliberately invest over the longer term, in bonds, which are expected to give lower long-term returns. The expert will reply that though bonds give less return, they also are less volatile. Since volatility impacts the corporate balance sheet, companies would rather take lower long-term returns than higher volatility. They will also point out that if a company goes bus, the holders of bonds get paid before the holders of equity, bonds are a safter bet.

You can already see how conflicted a trustee can be. It’s in the long-term intersts of the scheme to be invested in equities but in the short-term interest of the company to have stable funding. In the short term, the scheme is safer in bonds than equities, in the longer term, being in bonds suffocates the scheme which will require artificial respiration (though massive cash injections from the sponsor).

In the CETV – these conflicts crystallise.

Why transfers are such a problem is that they crystallise a long-term liability into an immediate payment.

The transfer value is calculated to the particular asset mix of the scheme. As we have seen, this depends on the bond/equity mix.

However the valuation of the scheme, for the purposes of the company’s account and reports, is based not on the bond/equity mix but on a decision determined by the accounting standards prevailing under IAS19. These demand that the discount rate, whatever the asset mix, is consistent and consistent with the bond discount rate. This makes liabilities very expensive and is why most schemes are showing an accounting deficit. Were schemes able to report using the method used to calculate transfer values (known as best estimates) then in aggregate, schemes would be reporting a surplus of assets over liabilities.

Looking at the chart below, the blue line shows how our top 6000 funded DB schemes would look if they were paying transfer values (best estimates) and the purple line shows how they look for accounting purposes (PPF solvency).fabi-graph

Why trustees will breath foul air (rather than suffocate)

Every time a trustee pays out a transfer value, he/she is paying out less than the accounted for liability (unless the discount rate for accounting and transfers is the same=100% bonds). So the technical solvency (for accounting purposes) is improved by paying transfer values.

But it breaks some trustee’s heart; for trustees want to pay scheme pensions, it’s what they were put on the planet to do (well that’s a bit of an exaggeration but you get my drift). More importanly, trustees have sleepless nights worrying what happens to the transfer value after they’ve given it up.

Trustees will let their babies go because the alternative is even more horrible. They will breathe toxicity rather than suffocate, even if the toxicity means they suffer later. For the money paid out today, will be needed in years to come, and where schemes are seriously underfunded (on any basis) and where the scheme is not reducing transfer values (for whatever reason) then the payment of CETVs today will be to the long-term detriment of the scheme.

Why don’t trustees reduce transfer values (when they could)?

This is a question that the Pensions Regulator ought to be asking right now.

Talking to my colleagues, I think there are two answers.

  1. Trustees are naturally inclined to have confidence in the recovery plans they have negotiated with their sponsors. An insuffeciency report is a vote of no confidence in the recovery plan, and by extension in the capacity of the sponsor to meet deficit payments. Not only is such a plan an insult to the sponsor, it’s an admission of defeat for the trustees.
  2. By reducing CETVs offered, trustees are reducing CETVs taken up.This is simple maths and has nothing to do with discount rates, if you reduce the CETV by 20% you increase the critical yield needed from the investment of the transfer by 20% making the taking of the transfer much less likely.

So trustees will pay transfer values they cannot afford rather than not pay transfer values at all, that’s because of the pressure from mark to market accounting from the IAS19 standard.

And what of the future?

If we see inflation rise, we can expect to see interest rates rise. If Mark Carney decides to stop issuing bonds to fund QE, then the yield on bonds will rise and if he does this as interest rates increase, the yield on bonds will rise quite sharply. This will ease the pain of the value of the bonds falling (the great virtue of a fully matched portfolio). But it will mean that transfer values will fall too, for the bond component will drive up best estimate discount rates. Only schemes with a high equity investment will see stable transfer values.

Ironically, what economic theory tells us, is that all this will happen because the economy is growing again. Equities do well when there is economic growth. But if schemes have sold off their family silver to pay large numbers of transfer values, they may not have the equities to properly benefit from any upturn.

At this point, the long-term toxicity of breathing foul air (paying full CETVs) will be fealt.

We are seeing (with the Royal Mail and perhaps we will see with USS) the impact of moving a scheme fully into bonds. There is no scope to pay future accruals at the Royal Mail (other than to call on the employer to fund the scheme at unaffordable levels).

The reason schemes like the Royal Mail adopt high bond exposures is to reduce volatility for the employer and to please the members. These are precisely the reasons trustees give for paying high transfers (members get “freedoms”) employers get away with paying out on the blue line (above) which improves their purple line (above).

But the future of accruals at the Royal Mail currently looks bleak, and the capactity of many schemes denuded of cash from over-payment of CETVs also looks bleak.

The problem (as a former Government Actuary pointed out this week) is that we haven’t worked out whether we love our DB schemes or loathe them. If we loathe them, then lets buy them out or consign them to the PPF (pre-packing our businesses into administration). If we love our DB schemes, let’s take care of them, let’s use insuffeciency reports where they are needed, let’s live with some balance sheet volatility and let’s get our scheme into equities while we still can.

And for heaven’s sake – let’s stop talking about CETVs as de-risking. What happens to transferred money is fraught with risk!


About henry tapper

Founder of the Pension PlayPen,, partner of Stella, father of Olly . I am the Pension Plowman
This entry was posted in advice gap, pensions and tagged , , , , , . Bookmark the permalink.

6 Responses to A suffocating man craves the foulest stench.

  1. George Kirrin says:

    You own up to not being an actuary, Henry, and I know you’re not an accountant either.

    Accounting standards in fact refer to using best estimates too, as this extract from IAS 19 (the international accounting standard for pensions accounting) shows:

    “The overall actuarial assumptions used must be unbiased and mutually compatible, and represent the best estimate of the variables determining the ultimate post-employment benefit cost. [IAS 19(2011).75-76]”

    The accountants go off message, however, when they then presume mark-to-market offers the best estimate value for pension schemes and sponsoring businesses which are (mostly) going concerns. Market prices are marginal prices, not reliable or “decision useful” (which is another of the accounting clichés) valuation measures for the whole business or the whole pension scheme portfolio of investments.

  2. Con Keating says:

    There is a fundamental problem at the heart of the CETV issue. These transfers are in essence options granted for free to the members of schemes – an option on the (quasi-market) level of the discount rate – the cost of this option is borne by the scheme, and very expensive – of the order of thirty percent of scheme liabilities. Interestingly the only other common option of similar term in finance is that of equity – which suggests that the hedge instrument could be equity.
    There is a further problem which this CETV exercise makes evident – the super senior position of svheme members under this view. This was not the intent of any regulation. It arises from the misconception that schemes must be self-sufficient. The recent ILC paper put it, incorrectly as: “The current TPR remit for pension scheme trustees is to ensure members receive full benefits.” The Trustee duty is in fact to “secure benefits”, a very different matter. The ILC mantra is a common but incorrect and very dangerous meme.
    There may be apparent savings, but these are cosmetic. The exercise of these transfer rights raises the cost of provision and weakens the position of remaining members. The only defence for trustees would be that this additional cost will be borne by the sponsor.

  3. Con Keating says:

    George’s comment is helpful. The CETV values are not accounting values. If someone transfers from a scheme based upon a higher discount rate than that used in the IAS 19 accounting, it will throw up a surplus. However, if that transfer entails an asset movement above the level of funding current, this will crystalise a deficit. Say the IAS value of the liability is 100, the actuarial best estimate is 80, and the level of funding 65. Then originally the accounts show a deficit of 35. This 35 deficit is realised as 25, throwing up a surplus of ten.

  4. henry tapper says:

    I don’t disagree at all, though I think that the amount of discretion given to actuaries in calculating the IAS19 values isn’t that great – it always seems to revert to discussion around the corporate bond yield – but that’s me talking as neither an accountant or an actuary! The variations in best estimate calculations on CETVs is a matter of consternation to the ordinary member but (I hope) is explained here.

    The day to day rise and fall of CETVs is also a matter of consternation and the reasons for that are also contained within the blog. It is easier for people to engage with their transfer value’s progress than the progress of scheme funding – but the two are linked.

    All this is of academic interest unless a transfer value calculation is asked for, but with CETV calcs. up 30% on the year, it looks like this is and will remain a hot topic.

  5. Pauline Armitage says:

    What would happen in the (admittedly hypothetical) situation of everyone in the scheme asking for their CETV?

Leave a Reply