Written evidence from Mr Mark Davies; Mr Masroor Ahmad LDI0021
Our experience in LDI started in 2006 when we built an independent LDI manager. We grew this business to c. 150 UK Pension scheme clients and in February 2022 this was sold to a leading fund manager. We stayed on to transition the business until September 2022.. As such our evidence is based on our general experience of talking to Trustees about LDI over that time as well as our specific experience of LDI mechanics. We have kept out responses short but we would be more than happy to field any follow up questions.
The most important point in answering this question is the definition of “impact”.
The task of a DB pension scheme, arguably, is not to maximise assets, more it aims to secure pensions of its members. As such we believe “impact” should be measured by relative affordability of securing pensions. The only objective way of doing this is by considering the cost of securing those liabilities – the most secure being an insurance contract. The assets available to a scheme to secure liabilities are the assets in the scheme plus the contribution commitment from the sponsor. Clearly in this construct more assets will always be better however the assessment of any pension scheme success should never be based on whether the assets could have been bigger as there will always be a strategy that could have performed better if the job was to maximise assets.
We would suggest impact is therefore measured as the difference (deficit) between the cost of securing the liabilities and the assets of the scheme. If this is zero then the scheme can guarantee (as near as possible) pensions. The smaller this number is then the more affordable it is for the scheme sponsor. As an aside we prefer the deficit approach to funding level (the ratio of assets to insurance cost) as it is much more aligned with sponsor affordability which is, after all, the ultimate fallback for schemes.
For most of 2022 the cost of insuring pensions will have fallen. All else being equal, from a pure LDI perspective, we would expect that this would have improved the funding position (reducing the deficit) for most schemes. Those with LDI would still have likely seen an improvement as most pension schemes will have a hedge ratio of less than the insurance cost of the liabilities. As a result, whilst assets of those with LDI will have fallen they will not have fallen by as much as the liabilities.
There are numerous trackers out there that monitor this sort of position and shows this picture. The XPS one estimates assets and compares this to liabilities on a Gilts + 0.5% discount rate:
https://www.xpsgroup.com/what-we-do/technology-and-trackers/xps-dbuk-tracker/
Clearly the performance of the other assets of the scheme will also have an impact, but that will be scheme specific and unless they were sold as a function of the LDI situation then we do not comment here.
In our view the key element of understanding from September and October is to what extent did LDI make Trustees of schemes do things that they wouldn’t have chosen to do and to what extent did that impact the deficit.
The best way of considering this is to think about the characteristics of a scheme that would have fared worst: It would be one that had LDI that was removed/reduced when interest rates peaked and in the background perhaps they were selling assets that they wouldn’t have normally chosen to sell. Our expectation is that where this is likely to be most acute is with schemes with pooled LDI funds and/or a high proportion of illiquid assets. Whether this action was a negative (or positive) will only be known if the position is either reversed or at some point in the future dependent on relative movement of assts and gilt yields.
We summarise above the impact on DB schemes. We have no information on any impacts of individual pension scheme members only that we would not expect an impact given the funding position of a scheme does not change benefits (except in a sponsor insolvency and we have heard nothing of a sponsor insolvency caused by this).
Clearly DC investors holding gilts will have seen a fall in assets value. However annuity prices will have also fallen.
We do not think that the regulator should regulate the use of LDI explicitly just as it does not regulate other types of investment. To regulate the amount of LDI a scheme has is to restrict a schemes ability to manage risks. Regulating LDI products we would argue is not the role of the Pensions Regulator. Finally, some may suggest regulating over the use of “leverage”. We believe that doing this will be both challenging to define as well as difficult to implement. This may have some potential unintended consequence – for example restricting the ability for pension schemes to use currency hedging.
If anything, we think the best approach would be for the regulator to provide guidance on the types of scenarios and stresses that pension schemes should consider. The challenge with this approach is that schemes may design strategies only around these specific scenarios and therefore could lead to some negative outcomes particularly if the scenarios are not well thought through.
We have spent over 15 years educating Trustees on LDI. The risk of LDI – that rates might rise and LDI go down in value, in our experience has always been flagged to Trustees. We have also seen discussions around what would happen if interest rates rose significantly –the idea of selling assets to support LDI if rates were higher (particularly if funding was improved) was common in our experience.
We only managed segregated mandates (primarily for small and medium sized pension schemes) and not pooled funds so we do not know to what extent Trustees didn’t understand the mechanisms of pooled funds in more extreme scenarios.
LDI is a hedge of a risk – that is how Trustees in our experience have seen it. On the basis that hedging a risk makes sense then LDI is the only tool that allows that to happen. LDI using derivatives (as with currency hedging) is the only way to hedge that risk whilst still retaining the ability to invest in growth assets (again this is analogous to currency hedging).
As such the question is about management of leverage in LDI. Clearly one thing that will be looked at is the allocation to LDI for a given amount of hedging as well as liquidity of assets outside the LDI portfolio but this is less to do with LDI itself.
On the LDI side, we have always believed that pooled LDI funds have some fundamental challenges due to the isolation of the leverage in a limited recourse structure – hence our only offering was segregated mandates. Pooled funds will clearly need to change (and some have already done so by reducing inherent leverage and adding more prudence to top up mechanisms). This is highlighted in Sarah Breeden from the Bank of England’s speech.
Segregated mandates have more flexibility so will need very little changing in themselves aside from the consideration of general liquidity in the other scheme assets as discussed above
We believe that the experience will have highlighted governance challenges to Trustees. In may cases there may not have been a problem, in some it may have been strained but manageable and in others they arrangements may not have worked. In all cases we believe that most Trustees will already be looking to make changes to governance approaches as a result. Therefore, we believe that governance rules may just reinforce changes already being made.
Whilst many commentators cite regulations (pension and/or accounting) as the driver for using LDI we believe that the motivation for LDI would exist even without these. As discussed in the first question the aim of a pension scheme is to secure liabilities. The most secure way to do this is with an insurance contract and the pricing of that contains interest rate and inflation risk. Some Trustees and employers may still want to hedge these risks even if regulations did not highlight the risks to them.
November 2022