Iain Clacher and Con Keating
Iain Clacher & Con Keating
The Office for National Statistics has just published its half-yearly Financial Survey of Pension Schemes for the two quarters ended December 2023 and March 2024. There are some massive and important discrepancies between the figures produced by TPR and PPF and the ONS survey results.
Table 1, below, shows the total assets held by private sector DB schemes for the period December 2021 to March 2024.
Our own work simulating the ONS survey method, based on a sample of around 1,000 schemes, suggests strongly that the ONS figures are the more accurate.
Table 1: Value of assets held by private sector DB schemes – December 2021 to March 2024
We have not seen any value for scheme assets published by the Pensions Regulator (TPR) for March 2024, though Neil Bull has cryptically referred to schemes having £1.3 trillion in assets. At year-end 2023, TPR reports schemes as holding £327 billion more that the ONS survey reports. PPF reports £237 billion more than the ONS in December 2023 and £255 billion in March 2024.
The discrepancies have been growing significantly over the period. They are highly statistically and financially significant. For example, the PPF s179 funding ratio of 146.5% reported in March 2024 would be just 120.5%. If overall scheme funding reported using TPR.s values was 100%, this would fall to 78.5% using the ONS values.
As PPF’s figures are based on returns to TPR, it is perhaps not surprising that a similar trend is present in their data. However, it is reasonable to expect PPF values to lie above TPR’s due to the presence of buy-in insurance policies held by schemes – these policies pay the full benefits of scheme members, not the PPF’s reduced benefits, and would represent a source of profit to them not available to the scheme. However, in December 2023, PPF reports £90 billion fewer assets than TPR.
The ONS survey reports insurance policies held by schemes. It is unclear what, if any proportion of these are buy-out policies. The ONS survey reports policies held by schemes as £145 billion in December 2021, unchanged at £145 billion in March 2024. These policies represent an increase from 8% to 12.3% in overall scheme assets. If we assume that the policies have a duration of ten years, using gilt yields as the discount rate (which rose from 0.98% to 3.89% over the period), this unchanged value would represent an increase in nominal coverage of £52.5 billion.
The most worrying aspect of the discrepancies is the Impact Assessment for the new low-dependency funding requirement was based on TPR figures. It seems that the new DB funding regime is likely to prove extremely costly to schemes, sponsors and the taxpayer. At the very least, some explanations are due.
A Work and Pensions Select Committee report in March 2024 recommended that the PPF and TPR should work with the ONS to obtain more accurate results and publish these.
There seems to have been no public response to this recommendation.
TPR representatives are said to read these blogs, so why don’t they respond to Clacher and Keating? What are they trying to hide?
TPR hid behind this (to me) feeble explanation in oral evidence to the WPC in November 2023:
“First of all, they are used for different purposes. The ONS uses a sample of schemes, around 10% to 15%. It tends to focus on the larger schemes, and it will look at the funding levels and the level of assets over recent periods. It will then use that data to effectively scale for the smaller pension schemes. It does not hold the data on the smaller pension schemes. We hold the data on all the pension schemes. That is why our numbers differ.”
PPF was more open than TPR and admitted this much to the WPC in writing in December 2023:
“We are clear in our 7800 index publications that we don’t hold enough data to capture the structural changes to asset allocations, nor to capture changes to in any leveraged LDI portfolios (these factors have been particularly pronounced since March 2022) and as a result, the impact on assets will often be less accurate than the ONS survey.”
The underlying cause for concern with TPR comes from its lack of accountability.
That leads to a mindset that it doesn’t even think it needs to offer a coherent explanation for the data issues. Everything they say so obviously suffers from spin, lacking actual substance. They’re seem at least to be aware of the old saying about being “better to be thought a fool, than open your mouth and remove all doubt…”.
And that lack of clear accountability is what created the gaping disconnect from the devastating impact it’s derisking (disinvestment) policies have had / are having on the UK economy, and for TPR’s refusal to acknowledge any connection or culpability, again choosing spin.
Pingback: Doubt cast on the impact of pension DB funding code. | AgeWage: Making your money work as hard as you do
A comment on yesterday’s blog quoted a TPR response to questioning on the data discrepancy
““First of all, they are used for different purposes. The ONS uses a sample of schemes, around 10% to 15%. It tends to focus on the larger schemes, and it will look at the funding levels and the level of assets over recent periods. It will then use that data to effectively scale for the smaller pension schemes. It does not hold the data on the smaller pension schemes. We hold the data on all the pension schemes. That is why our numbers differ.”
This does not hold water. The value of assets being measured by TPR and ONS is the same, purpose does not enter consideration. It is true that buy-in policies should be worth a little more to the PPF, since they pay full benefits rather the reduced benefits paid by PPF but with policies held by schemes at just £145 billion, that difference might be around £20 billion, an order pf magnitude lower than is seen.
The ONS sample of 614 schemes may be 10%-15% by number of schemes but it is around 70% – 75% of assets held by value. The discrepancy is 29.7% of the ONS number. TPR would actually need to be saying that small schemes have almost twice as many assets to explain the £327 billion discrepancy.
TPR does not hold current data on schemes. The Purple Pook shows just how old their data is:
“Figure 2.4 | Distribution of assets and s179 liabilities in The Purple Book 2023 dataset by date of s179 valuation. Over two-thirds of assets and liabilities at 31 March 2023 have been rolled forward from s179 valuations dated between April 2020 and March 2022.
For The Purple Book we roll forward the assets and liabilities from the values featured in the latest s179 valuation, as submitted in the annual scheme returns. The following table shows the distribution of assets and liabilities (as at 31 March 2023) across the effective dates of the s179 valuations from which we have rolled them forward.
s179 valuation dated between: Number of schemes Assets (£bn) S179 Liabilities
01/10/2022 and 31/03/2023 15 6.8 5.6
01/04/2022 and 30/09/2022 230 36.5 28.0
01/04/2021 and 31/03/2022 1,466 435.6 332.3
01/04/2020 and 31/03/2021 1,630 529.7 379.9
31/03/2005 and 31/03/2020 1,710 395.9 299.6
Total 5,051 1,404.4 1,045.5
Source: PPF”