An anonymous person writes. I am bound like Ixion to a wheel and must publish.
Hot on the heels of Clacher and Keating’s excoriation of the skin and bones I have gleaned from the interweb, I get by carrier pigeon the following answers to the very same questions. The post is anonymous, the pigeon divulged nothing, but was heard to mutter , under interrogation, don’t shoot me, I’m only the messenger. NB
I do not understand how an open ended investment scheme needs new members to make payments to current members. I am thinking that the discussion on open schemes needs some reality checks.
New members aren’t needed to make payments to current pensioners. But having a flow of new money being paid into a scheme by active members and the employer, to fund the accrual of benefits, provides more flexibility on investment strategy. There will be less of a (or no) requirement to sell assets to meet benefit payments. This allows a longer term investment horizon to be adopted. The reduced focus on liquidity requirements should allow schemes to generate higher returns over the long term, reducing the ultimate cost of delivering the benefits.
Why should pensioners and deferreds have less security in an open scheme than a closed one?
The DWP, TPR and the Pensions Minister have frequently stated that members of all DB schemes should have the same level of security, irrespective of the status of their pension scheme. This is a great soundbite. It is a lot harder to analyse what it actually means.
TPR’s consultation seems to consider that increasing member security means taking less investment risk. But this leads to a reduction in expected return, which will increase reliance on the employer covenant, as the employer is likely to need to pay more money into the scheme over a longer period. I have yet to see any analysis which demonstrates that taking this approach improves member security.
In my opinion one way of considering member security is to analyse the likelihood of members being paid their benefits in full. (A more rigorous approach would be to also include allowance for the proportion of benefits which the members would receive if they do not receive their full benefits; in most cases this would be expected to be the vast majority of the benefits they have been promised, either through the PPF or because the scheme will have enough assets to purchase annuities which exceed PPF compensation, taking into account any recovery from the employer in an insolvency process.)
LCP have carried out analysis which demonstrates that reducing investment risk can significantly reduce the likelihood of paying all pensions. They consider an example of a well-funded scheme with a strong sponsoring employer, where adopting an investment strategy targeting 1% pa above gilts gives a 95% probability of paying member benefits in full. This compares to a probability of 88% with an investment strategy targeting 0.5% pa above gilts. So this apparent “risk reduction” more than doubles the risk that members do not get paid their pensions in full.1
The reason that targeting higher returns leads to better outcomes for members is because the current strength of the employer covenant means that the scheme can afford to take more investment risk in the short to medium term. Adopting an investment strategy with a lower expected return reduces the short term volatility of the funding position (I am deliberately avoiding calling this “reducing short term investment risk”, as it is not reducing risk to the members), but replaces this with a longer term reliance on the employer covenant. As there is less visibility on covenant in the longer term, this leads to worse outcomes for members. It is very rarely possible to “reduce risk” in a pure sense; usually risk can only be reduced in one area if more risk is taken elsewhere.
Adopting an appropriate level of investment risk in the short term is also in line with the commentary from TPR in the Code of Practice consultation regarding covenant visibility, as it should allow less risk to be taken over the longer term, when there is less covenant visibility.
There is a further material limitation with a policy focused on maximising security for open schemes. This objective places great importance on the £1 of pension which was accrued yesterday, and has no regard whatsoever to the £1 of pension which will be accrued tomorrow. Trustees will usually be advised that their core objective is to protect the security of members’ accrued benefits, and that future benefit provision is solely a concern of the employer. (Although personally I have never agreed with the implicit suggestion that the inverse is true, i.e. that employers only focus on future benefit provision, and historic pension promises are merely legacy issues for which they will always seek to contribute the smallest amount possible. In my experience this was rarely true from the corporate perspective.)
There is a clear conflict here with the perspective of members. They will care just as much about the pension benefits they will earn in their future career as those they have earned in the past. When they reach retirement and receive their monthly pension, they will care not a jot about how much of it was earned before and after an arbitrary date. Further, they are likely to take an interest in the sustainable growth of their employer, given that this will influence their career prospects and future pay prospects. It is therefore crucial to strike a balance which takes into account the affordability of ongoing accrual.
Depending on how “security” is defined, it could be argued that even in the same open scheme the deferred and current pensioners will have more security than actives. In the event of employer insolvency, members over retirement age would not be at risk of a 10% haircut to their benefits if their scheme entered the PPF, unlike those below retirement age. Further, in an insolvency the actives would be far more heavily affected by the risk of job losses and the loss of future pension provision. Finally, in light of the PPF compensation cap, members with smaller pensions are less exposed to pension cuts than those with larger pension
What happens when an open scheme closes and it is more poorly funded than a similar closed scheme at that point?
There is huge variation of characteristics in the DB universe. It is therefore always hard to imagine a genuine like-for-like comparison.
But let us consider the hypothetical scenario in which we have two mirror schemes, with a material active population. One of them is expected to remain open to new entrants for the foreseeable future and the other was closed to new entrants yesterday.
The closed scheme would be very immature. It has a large active population still accruing benefits. It is likely that for many years to come the contributions payable will still go a long way to meeting current benefit payment requirements, meaning no step change in the investment strategy is required. The closed scheme should still be able to take a relatively long term approach when setting its investment strategy.
The two schemes will have the same duration at the moment, as duration is usually calculated based on only accrued liabilities. As the actives would still be expected to accrue benefits for some time (perhaps 40 years, on the basis that the youngest member of a scheme which was until recently open to new entrants might be in their early 20s), the duration of the scheme will fall very gradually over time. This scheme might therefore still be 20 years away from “significant maturity”.
Returning to the original question, if an open scheme is closed it is likely to have a longer duration than a comparable scheme which was closed some years previously. Its point of “significant maturity” will therefore be further away, giving it more time to achieve a Long Term Objective (LTO).
TPR’s proposed new Code of Practice envisages open schemes needing to assume they mature in future, with no allowance for the flow of new entrants reducing (or reversing) the maturation process. I would suggest that a reasonable alternative would be to allow open schemes to set their funding and investment strategy based on the trustees’ expectation of the future flow of new entrants, but require all schemes to carry out contingency planning regarding the actions they would take if their scheme were to close to new entrants (and even potentially future accrual). This would include consideration of how they would achieve their LTO (funding and investment strategy) and their expectation of how long they would have to get there (i.e. when they would be expected to become “significantly mature”, if they were to close to new entrants tomorrow).
On this basis all open schemes could set an LTO now, as outlined in the consultation document, which they would target achieving when they are “significantly mature”. But in determining that time period they should be able to take the current status of their scheme into account. In many cases this would mean that there is currently no defined point at which they expect to become “significantly mature”, as the calculation suggests that they will remain immature in perpetuity, due to their open status.
Trustees could be required to make an assumption about the flow of new entrants as part of their valuation process. This assumption would be based on historic practices and actuarial advice, as well as the employer’s views of the future. In this way it would be very similar to the approach already adopted to setting a salary inflation assumption.
The alternative suggestion in the Code of Practice would be that schemes should set their funding and investment strategy on the basis that no new members will join the scheme. The position will then be recalibrated at the next valuation, at which point it is likely that the scheme will not have matured as expected, due to the new entrants. But this appears very inefficient, particularly as de-risking of the investment strategy is likely to need to be unwound when it transpires that the scheme has not matured as expected.
I have heard suggestion that defining all schemes as either “open” or “closed”, with different funding regimes applies, introduces a risk of “gaming”. Theoretically an unscrupulous employer could open their scheme and admit an incredibly slow rate of new entrants to reduce their funding requirements. But by requiring schemes to assess when they expect to be “significantly mature”, based on a realistic assumption about the flow of new entrants, and applying the same funding regime to all schemes, this binary separation between open and closed schemes can be avoided. Indeed, some open schemes might be expected to mature more quickly than some closed schemes. For example, an open scheme which has a relatively low flow of new entrants might mature more quickly than a scheme which has a large active population (e.g. due to a bulk transfer) but is closed to new entrants.
Currently accruing DB benefits get indexation up to 2.5% and a higher cap than originally intended. Plus once over NPA there is no 90% or cap. So PPF provides high protection.
The PPF does provide a relatively high level of protection. But schemes pay a levy to the PPF which is based on the PPF’s assessment of the likelihood that the PPF will need to provide compensation to these members. So all else being equal, an open scheme which is immature will pay a higher PPF levy than a more mature closed scheme.
Time for reality bites on funding level. There are good arguments for why accruing pensions for younger people should be matched by riskier assets. Let them have them in a CDC format not relying on the PPF.
This seems to contradict stated Government policy relating to the protection of DB schemes.
Yes, yes, yes!
This enthusiastic endorsement suggests you are not the author of the piece Derek!
I wish I was!
In a piece which is all good, let’s highlight and reinforce two major points:
“… one way of considering member security is to analyse the likelihood of members being paid their benefits in full.”
I agree. This is an excellent criterion for evaluating a proposed course of action: does it make it more or less likely that benefits can be paid in full?
The key task is not to make investment returns more certain, but to make benefit payments more certain. Making investment returns more certain is not a proxy for making benefit payments more certain. Achieving the former does not imply the latter is also improved.
A high but uncertain investment return can be better support for benefit payments than a more certain return: more certain to be a low return, perhaps too low to pay the benefits in full. Being certain we can’t pay the benefits in full is not the kind of certainty we are after.
“… allow open schemes to set their funding and investment strategy based on the trustees’ expectation of the future flow of new entrants, but require all schemes to carry out contingency planning regarding the actions they would take if their scheme were to close to new entrants (and even potentially future accrual). This would include consideration of how they would achieve their LTO (funding and investment strategy) and their expectation of how long they would have to get there (i.e. when they would be expected to become “significantly mature”, if they were to close to new entrants tomorrow).”
An excellent suggestion. This is exactly what I envisage doing for an open scheme client whose valuation is starting up now. Such a contingency plan would be useful to share with the sponsoring employer, so they are aware of the consequences of closure.
Pingback: TPR goes blog to blog with its critics | AgeWage: Making your money work as hard as you do