By Graeme Muir FFA, Partner and Head of Public Sector at Barnett Waddingham
Anyone else remember those days when LGPS Funds had one valuation every five years? No? Me neither. These days we seem to have one every five minutes – annual accounting valuations, Section 13 valuations, cost management valuations, cessation valuations and of course, the one that really matters, the triennial funding valuations. These are now set to become every four years – so quinquennial became triennial and is now turning quadrennial – another one for your spell checkers!
So where to start?
One of Lord Hutton’s key recommendations following his review of public service pensions was that retaining defined benefit schemes in the public sector should come with the condition that there is a “cost cap” mechanism to control the cost of future pension promises.
Costs of defined benefit pension arrangements are uncertain and so there should be a ceiling on how much tax payers are exposed to in meeting these costs. This seemed like a good idea at the time but like many things in life is easier to say than actually do.
The "how long" and the "how much"
The cost of a pension promise depends on many things but mainly it’s size and duration; i.e. how much will the annual pension be and how long will it be paid for? After many long hours of discussion about how to build such a cost cap mechanism, HM Treasury decided that it was mainly the “how long” aspect of the cost that should be included. So if the “how long” turned out to be longer than anticipated, we should reduce the “how much” or increase the member contributions to reflect that they will get more pension payments.
Whilst the original Hutton concept was a cost “cap”, it was also argued that there should be a floor. That is, if the “how long” turned out to be shorter than expected, we should increase the “how much” or reduce member contributions. Of course there was no chance this would ever happen….
“Mortality rates have improved and longevity has increased since Adam and Eve were around, even allowing for the odd plague or war along the way. The models used to predict future longevity are fairly complex these days and allow for some level of future improvements, based to differing extents on both short-term and long-term historical trends.”
Such models were used in the cost cap mechanisms which were being built at around the same time as the rate of improvement in mortality showed some signs of possibly slowing down to below long-term trends. Maybe this was a blip? We now know that it’s probably a bit more than a blip and it’s accepted that the trend line has flattened a little. So at the first time of asking, the mechanism designed to cap pension costs has in fact produced numbers that indicate the “how long” might not be as long as we thought, triggering a review of the “how much”.
This was a consistent outcome across all the public service schemes, albeit the LGPS being ahead of the game as always had its own mechanism. The LGPS Scheme Advisory Board then proposed some relatively straightforward changes to the Scheme – lower levels of member contributions for those on low pay, setting a minimum level of lump sum death benefit and effectively improving ill-health retirement benefits.
The McCloud case judgement
Then, just as we were about to get these changes in place, along comes the judgement on the McCloud case. All those quite sensible transitional changes to public service schemes when moving from final salary to career average are now deemed, or likely to be deemed, to have been unlawful, directly on age discrimination grounds and indirectly on the grounds of sex or race. The Government is hoping to be able to appeal this judgement but, on the basis that this is likely to take quite a while, the cost management process is on hold. The promise has been made that whatever needs to be done will be backdated to 1 April 2019. It’s a situation that is less than ideal.
Of course 2019 is LGPS funding valuation year – probably the last triennial one – with new employer contribution rates required from 1 April 2020. So, apart from the usual actuarial assumptions (inflation, investment returns, mortality etc) we may also have to make some assumptions about the outcome of the McCloud appeal and the cost management process. On the basis that nothing will have been sorted out on McCloud and cost management before we need to start reporting preliminary valuation results, the Scheme Advisory Board is to issue guidance to Funds and actuaries on what to do. Do we just ignore for now and revisit employer contributions once sorted? Or do we make some assumptions about the outcome and reflect this in employer contribution rates, to be resolved at the next valuation?
And talking of which, when will the next round of valuations be?
This is widely expected to be 2024 and likely to be LGPS Funds both north and south of the border. So that means five years after the 2019 valuation in England and Wales for LGPS Funds to fall into the same four year cycle as other public sector schemes. That is because the funded LGPS is just like these other schemes, apart from having assets. Employers are all paying contribution rates specific to their circumstances, which are also dependent on how the assets perform.
Of course the whole purpose of valuations is to check how the funding plan is doing. As it’s highly unlikely to be completely on track, what we need to do to get it back on track. Arguably the longer you leave it before looking at it, the more likely things could have gone seriously off track. This may not be too much of a problem for statutory or tax raising bodies but could be an issue for employers who may only be in the LGPS for a short time, where they may need to get back on track very quickly. So our hope is that we will be able to carry out interim valuations for such employers, or any groups of employers, where Funds determine they can have interim valuations.
Some interesting times ahead……
However, we do lots of valuations these days – the biggest number being accounting valuations each year for most employers in the LGPS. Due to reporting deadlines we need to use roll forward and estimation techniques to prepare the numbers in time. The “estimation error” in each roll forward, however, gets picked up at the triennial funding valuations when we get full data and we then rebase the accounting numbers for the next roll forward. It’s a fact though that accounting bodies and auditors are getting more excited about the “error” in the accounting numbers and can set some quite challenging tolerance levels. So if we are moving from three to four yearly valuations then we will have four years of estimation to sort out rather than three – a 33% increase. We suspect this might lead to more challenge from auditors so perhaps interim valuations will become the norm rather than the exception, albeit some will only be to get more “precise” accounting numbers.
|
