An Analysis of the Pensions Regulator’s
Review of impact on DB landscape following (the) LDI Episode
A “short” blog from Iain Clacher and Con Keating
Nausicaa Delfas, TPR CEO
Iain Clacher and Con Keating
As TPR’s Report is 58 pages long, we are going to analyse the report and its findings in a series of short blogs.
This report was produced by TPR in response to a recommendation from the Work and Pensions Select Committee. It is interesting that in the covering note to the report there appears to be some migration of TPR’s ambition:
“TPR’s key priorities as a regulator are to protect savers’ money by making sure trustees and employers comply with their duties; enhance the system through effective market oversight, influencing better practice; and to support innovation in savers’ interests.”
The cover note highlights the following findings in the Report:
“Our modelling shows aggregate funding levels of DB schemes improved over 2022 on a variety of measures:
- Funding levels improved for 87% of schemes on the statutory ‘technical provisions’ (TP) basis.
- Only 5% of DB schemes experienced both a deterioration in their funding level and either an increase in their existing funding deficit, or a movement from surplus to deficit, over 2022.
- By the end of 2022 our analysis suggests that broadly 80% of schemes were in surplus on a TP basis and on a buyout basis.
- About 4 in 10 schemes are estimated to be fully funded as at the end of December 2022 compared to less than 10% at the end of December 2021.”
NB – please refer to TPR’s statement at the foot of this blog referring to these bullet points. there is a restatement
It is deeply disconcerting to find at the level of the executive summary, such an obvious inconsistency as highlighted in bold text above.
If 80% of schemes are in surplus on both TP and Buy-out bases (estimated to be 15% – 20% above TP, then only about 5% of schemes would be in deficit of a TP basis alone. This is completely inconsistent with 40% (4 in 10) being fully funded. These figures are also inconsistent with the figures reported by the PPF on an s179 basis. In December 2021, PPF reported 2,156 schemes (41.3%) in deficit and 3063 in surplus. In December 2022, the PPF reported 686 (13.4%) in deficit and 4,445 in surplus.
What is even more concerning, is that TPR quite simply do not know the full extent of the impact of LDI.
“We have in place a data strategy to provide us with a more robust picture of the resilience of both pooled and segregated LDI funds, the liquidity levels of DB schemes and the governance and operations of DB schemes enabling them to respond quickly to a similar situation in the future. This is achieved by having regular flows of data from the five UK fund managers that between them hold almost 90% of the DB market in leveraged pension investment assets, and by including additional questions regarding LDI strategies into our annual DB scheme return starting from 2024. These questions will seek to understand the level of leverage used by schemes, the degree and deliverability of liquidity in the overall asset portfolio, and the trustee governance around collateral transactions. This will also allow us to distinguish between schemes which used pooled funds, and which have segregated funds.”
Almost 18 months after the LDI crisis, and two years since LDI started unwinding as a strategy[1] there is a “data strategy” and some additional questions that will be included in annual DB scheme returns from 2024. Given the severity of the LDI crisis (it has been subject to parliamentary investigations in both the Lords and the Commons) to not know with any greater confidence than this is highly questionable. This is not to say that it is an easy job, but what has been presented does not feel like this has been given the urgency and imperative that was needed.
This lack of urgency is also highlighted by the following statement:
“Over the last 18 months we have also doubled the number of investment consultants. We are bolstering this with the recruitment of a senior economist and seeking access to senior market participants to better gather market intelligence and understand potential wider market risk which could impact pension investments.”
As with almost everything that comes from TPR, no such letter would be complete without reference to the DB Funding Code and Regulations:
“Furthermore, the anticipated introduction of the funding and investment (FIS) regulations in early 2024, will result in TPR receiving richer scheme valuation data from trustees. This will be provided irrespective of whether a scheme is in deficit or surplus.”
We presume this is a reference to The Occupational Pension Schemes (Funding and Investment Strategy and Amendment) Regulations 2024. The impact assessment for which was largely the work of TPR and is the subject of a separate brief review.[2]
Our conclusion in that review was:
“We do not believe that this impact assessment is fit for purpose, and that these Regulations will bring substantial further costs onto employer sponsors, costs which far exceed any benefits they may bring to pensioners and scheme members.”
As differences in data are a recurrent issue in our commentaries, we shall begin with TPR’s discussion of these In the Report, the section headed: “Differences from TPR estimates to those published elsewhere”.
This begins with:
“137. TPR is the only organisation which collects data from all registered DB schemes subject to scheme funding regulations. Summarised scheme valuation data is submitted to us by schemes in deficit on a technical provisions basis at least every three years, according to legislative requirements to submit triennial valuations. Annual data is also collected in DB scheme returns, which provides information such as membership and the asset splits at the date of the latest trustee report and accounts.” and “138. Our data is shared with the PPF, and they use it for their modelling and publications such as the annual PPF Purple Book.”
The Purple Book describes TPR’s data in its most recent edition, 2023:
Only 245 (4.8%) of these valuations are relevant for discussion and analysis of the LDI crisis which may be considered to begin with the increase in gilt yields from December 2021. At most, just 15 of these schemes may reflect the full effects of the turmoil. Indeed, the evidence is that there were still some LDI-related costs being experienced as late as June 2023.
The Report continues with:
“139. The Office for National Statistics (ONS) is the main other organisation that collects DB scheme data on a frequent basis. It surveys c.10% of the universe on a quarterly basis, including all DB schemes over 10,000 members and a selection of smaller schemes (c. 6% – 7%). As the ONS does not have full coverage with smaller schemes, it extrapolates their results in order to make an estimate for the whole of the DB market.”
And further,
“141. We do not know how many schemes are surveyed where these adjustments are made as we have not been able to obtain the ONS data for individual schemes to enable us to undertake such comparisons.
It is therefore difficult to know with certainty where the differences lie, though we suspect in part it is with the extrapolation of some of the smaller scheme data and part of it is through actual experience being different to that assumed. Examples of where experience may differ to the assumptions we have adopted include actual investment returns, the application of when trustees are buying and selling assets, or if trustees make strategic changes to their asset portfolio.”
The ONS publishes its sampling of schemes. We have discussed these issues previously here and in A Commentary on the oral evidence of the Pensions Regulator to Work and Pensions Committee entitled “bad data leads to poor regulation”.
To quote from this review of TPR testimony:
“The sample size for ONS is 614 schemes (around 12% of schemes), but it captures some 70% – 75% of scheme assets. To attribute a difference of 14% in total values to differences in sampling techniques as applied to 25% – 30% of the overall universe values seems highly unlikely. Until December 2021, the values being reported by TPR, the PPF and the ONS differed only by small amounts and could fairly be attributed to sampling errors. The large difference we now observe has grown steadily since then. Between the ONS and PPF, the amount was £196 billion at March 2023 and with TPR £180 billion. The difference between PPF and TPR figures can be attributed to the valuation of insurance policies. (Incidentally, we believe that valuation to be somewhat low.)”
It is perfectly sensible to apply survey techniques to data collection and ONS is without equal in this regard, and the Financial Survey of Pension Schemes has the kitemark of being National Statistics, which is a separate benchmark within the ONS for the veracity of the data.
This section of the report continues with:
“143. We would also highlight that TPR and the ONS collect data for different purposes. TPR has a statutory duty to collect data, some of which we can change and have discretion over, and some of which is set out in legislation. Our universe data is collected from all schemes to enable risk-based regulation in an appropriate way. Where we have concerns and engage with an individual scheme we may request additional information at that time in order to understand the issues in greater detail and understand the trustees’ decision-making processes. The ONS surveys are more frequent for the purpose of their statistical publications, however they do not engage with schemes for other purposes.”
[Emphasis added]
The purpose for which data is collected is irrelevant; the value of the assets held by schemes should be a matter of objective fact. While it is true that the ONS does not engage with schemes other than for data collection purposes, they do engage with schemes extensively in that.
This section ends with:
“145. We want to be as joined up as possible with other regulators and public bodies. We would like to have access to the underlying data used by the ONS in its publications, but it has not been able to provide this to us so far – for legitimate reasons around what data it is able to share with third parties. We continue to meet regularly and discuss this with the ONS.”
The ONS currently publishes data principally on the assets held by schemes but also includes data on financial liabilities such as repo and derivatives but does not cover pension liabilities to scheme members.
It has a project under way to collect and publish these scheme liability values commencing with the year 2022. We understand this is currently undergoing quality control and assurance testing. It has apparently been the cause of the delays in publication of the June and September editions of the existing financial survey of pensions schemes, which are expected shortly.
We find it interesting that TPR has a desire to have access to the data that the ONS holds, as this would suggest that perhaps that the ONS data is richer and has more up-to-date information on what actually happened. Moreover, given that TPR will only be asking about LDI from 2024, the ONS data may well be the only good source we have for some time to come.
TPR STATEMENT ON A FORMATTING ERROR IN ITS REPORT
TPR has written to me with the following statement
In light of this blog from Con Keating and Iain Clacher, we would like to point out that there was a formatting error in the covering letter to the Work and Pensions Committee accompanying TPR’s analysis of the impact of the LDI episode on DB funding. This inadvertently changed the sense of the penultimate bullet point. We have notified the committee of this point.
The two bullet points in question should read:
- By the end of 2022 our analysis suggests that broadly 80% of schemes were in surplus on a TP basis.
- On a buyout basis, about 4 in 10 schemes are estimated to be fully funded as at the end of December 2022 compared to less than 10% at the end of December 2021.
I have contacted Con and Iain to explain this error.
[1] In July 2022, the FT highlighted that further collateral calls were coming for schemes, See: UK pension schemes confronted by growing liquidity strains, Financial Times, 1st July 2022. (free link)
[2] https://henrytapper.com/2024/02/10/keating-and-clachers-review-of-the-dwps-proposed-funding-regulations/