This is Con Keating’s and Iain Clacher’s submission to the Work and Pensions Committe’s call for evidence on “Defined benefit pensions with Liability Driven Investments”. Written in March of this year, it represents an important perspective on the subject that is given insufficient airing in conventional media .
Written evidence from Professor Iain Clacher & Dr Con Keating
We have previously submitted written evidence to the Committee as well as having given oral evidence at the first hearing of the Committee’s investigation into LDI. We welcome the opportunity to provide additional evidence to the hearing on several aspects that we did not cover in our initial submission or testimony, and that have since emerged in the intervening period.
Taxation and the cost to the Exchequer
UK DB pension schemes had some £1.8 trillion in assets at the beginning of 2022. This had fallen to £1.4 trillion by PPF estimates at year-end, a decline of £400 billion, but more by our estimates, which show assets declining by some £500 billion.
Occupational pension schemes are privileged with respect to taxation and so these assets had benefitted from tax concessions.
Contributions made to pension schemes by employer sponsors are tax deductible to the employer. This applies to both normal contributions made in respect of new awards and also to special contributions, for example those made to repair scheme deficits. In the latter case, the tax deduction, if the amount exceeds £500k, is spread over four years. Pension funds also benefit from tax exemptions from income and capital gains taxes.
In recent times, the value of the tax benefits has commonly been estimated to be in the range of 15% to 20% of the asset values.
Perhaps the simplest way to describe the loss to the taxpayer is to draw attention to the fact that some 15% – 20% of the asset value lost, arises from these tax concessions – by our estimates that would be £75 billion to £100 billion.
In normal circumstances, there is an offset to these tax concessions, which is that pensions in payment are taxable as income. Except for a lump sum of up to 25% of the commuted value of the pension, which is tax exempt, and may be taken in whole or in part by pensioners, pension income payments are subject to individual income tax.
This income tax goes far in offsetting the tax concessions enjoyed by schemes – indeed the combined situation has often been described as one of tax deferment.
Despite recent movements in interest rates, although the present value of pension liabilities has declined with rising rates, the pensions ultimately payable are unchanged. This means that the losses of the past year will need to be recouped in order to pay the promised pensions. This may come from either further employer sponsor contributions or from investment returns – it really does not matter which, both will be tax privileged.
The central point we are making is that this is a tranche of some £500 billion of the funds used to pay pensions that will ultimately have benefitted twice from tax concessions, as the shortfall in assets has to be made good. However, the tax receipts from pensions paid will be unchanged. This means that the second of these sets of tax benefits to the contributions and/or returns on investments has no offsetting income tax receipts as there is only one pension paid to someone in retirement, and this will cost the Exchequer markedly.
New Funding Regulations and DB Funding Code of Practice
The proposed Funding Regulations and DB Funding Code of Practice compound the tax problem highlighted above. The point of “significant maturity”, as set out in the proposed regulations and code, however measured, means that beyond this point, schemes must hold asset portfolios which exhibit “low dependency” on the sponsor employer, which could also be described as a self-sufficient asset portfolio. This objective accelerates and increases the funding requirement on sponsors of DB schemes, and with that, the cost to the Exchequer in both amount and timing.
The aim of funding to self-sufficiency levels also renders the Pension Protection Fund redundant, which itself raises issues. The shortened horizon of the proposed point of “significant maturity”, also means that more of the cost of funding to the required level will need to be borne by sponsor employers, since there is less time for investments to make good the shortfall in assets.
Higher discount rates, which have been lauded as improving the funding ratios of DB schemes, carry with them shorter times to the date of significant maturity, and compresses the funding requirement into the near-term, and with that the tax cost. For a typical scheme[i], defining the point of significant maturity, as the scheme having a 12-year modified duration, as is currently proposed, will shorten the time horizon from 14 years at a 4% discount rate, to immediately if the discount rate were to be 7%.
The present value cost of funding to self-sufficiency currently lies between 15% and 23% above the prudent statutory funding level.
There are many other aspects of the proposed Funding Regulations and Code which concern us. We will mention just one more here; the treatment of open schemes. They may be few in number, but they are among the largest schemes in the UK. There is again a failure by the DWP and The Pensions Regulator to recognise the role of the sponsor(s) and that this constitutes an asset to the scheme. In the case of open schemes, the present value of future contributions may well be its largest asset.
The market for de-risking
A frequent analysis to feature in industry commentary and the pensions trade press since the LDI induced gilt crisis, is that because of the “improved” funding of DB schemes, schemes are now in much better shape and closer to their “endgame”, which typically means a transaction with an insurer. This can take the form of a “buy-in” (a bulk annuity purchase) or a “buy-out”, which is the passing on of the pension scheme assets and liabilities (for a premium) to an insurer.
The latest statistics below, taken from the Purple Book 2022[ii] show that between the 2nd half of 2013 and the 1st half of 2022, there were only £55.8 billion of buy-out transactions and £129.4 billion of buy-in transactions.
The most recent statistics from the PPF7800 Index show the present value of PPF level pension liabilities to be £1076.2 billion[iii]; this would currently be £1362 billion at buy-out level. Even with the pensions industry predicting a big year for risk transfers of around £40 billion[iv] to £60 billion[v], what is very clear, given the size of the risk transfer market relative to DB pension liabilities, is that most DB schemes will have to pay pensions in full, on time, as they fall due, for the foreseeable future.
While there are both capital and human resource constraints on the ability of insurers to increase their activity, it is the skilled and experienced human resource which is likely to prove to be the binding constraint. With that in mind, we would expect insurers to be able to exploit their position and increase their profit margins. It is interesting that a number of stock market analysts have already drawn attention to this possibility.
While risk transfers may make sense from the perspective of a DB scheme or from a corporate sponsor, they do not result in investment in the real economy. They are structured financial transactions as opposed to productive investments.
There is a functional equivalence between insurance buy-in and buy-out, and LDI, when LDI is complemented by a longevity swap. As such, there really should be no economic motivation for a scheme which has adopted full (but unleveraged) LDI and an appropriate longevity swap to entertain the idea of buying in or buying out scheme liabilities.
Longevity swaps, as a form of risk transfer, are used to manage the risk that members may live longer than the standard actuarial mortality models used by pension schemes predict. However, with mortality increasing[vi] and longevity having been on a downwards trend[vii], such risk transfers are hedging a risk that is very much a tail risk, and one that is highly unlikely to experience a rapid spike in improvement. However, for those schemes that have undertaken longevity swaps, cash is being drained away from their funds in the interim, and schemes are facing substantial capital losses on these contracts.
While we have only looked at a small number of longevity swaps, all of which are offside i.e., in the red and losing money, we would be very surprised if there are many, if any, longevity swaps that are currently profitable to schemes. The capital losses experienced in some of the cases we have looked at are also very large, with one extending to 35% of the notional value of the swap.
This again means that there is less money at the disposal of trustees to pay pensions, as there is a leakage of current cash flow from the scheme today, to manage a risk that is remote in time. In one case we have seen, the swap would only ‘come good’ if the life of the average scheme member proves to be more than 90 years.
While recognising that the call for evidence is concerned with LDI, we feel that we should state that we are most concerned by the timetable now being imposed by the DWP and The Pensions Regulator on the new Funding Regulations and Code.
The similarities between the proposed funding requirements and LDI far outweigh the differences. The incomplete nature of the information made available on the final text of the Regulations and the absence of any meaningful impact assessments is a major concern. As well as this, it appears from the text of the proposed Code that The Pensions Regulator has learned very little as to the economic consequences of the past decade and more of their regulatory emphasis on funding over covenant, nor have they learned anything from the resultant crisis in LDI.
The timetable for implementation of the Funding Regulations and Code being imposed will also mean that any changes and remedies suggested by the parliamentary committees investigating the LDI crisis are most unlikely to be actioned before the new regulations and code are enacted. At this point in time, we do not even have reliable estimates of the overall cost of the crisis. However, we are being called upon now, to make judgements on complex matters, for which neither the DWP nor The Pensions Regulator have provided any cost estimates for the risks to members, despite risks to members being the primary motivation of the proposed Regulations and Code. Moreover, there is no analysis of the cost of these regulations to the taxpayer.
We would also urge the Committee to read and consider the recommendations contained in the letter from Lord Hollick to the Economic Secretary to the Treasury, Andrew Griffith MP, and the Pensions Minister, Laura Trott MP, on LDI.
[i] The liabilities of this scheme extend over the next 70 years. It has an average life of 20.5 years and 90% of its projected liabilities fall due within 39 years.