Written evidence from Richard Britton (LDI0005)
Defined Benefit pensions with liability driven investments - Call for evidence
This written evidence is submitted by Richard Britton in a personal capacity.
Date: 9 November 2022
Introduction
I make this submission from the perspective of a person with 40 years’ experience in financial markets and their regulation. This includes 12 years as a broker in international bonds and US government securities with the US investment bank, Merrill Lynch; 15 years at the two predecessor bodies to the current FCA where I had several roles in supervision of the wholesale bond and equity markets; and 12 years as an independent consultant on International financial market regulation, working primarily for the IMF and World Bank. I am retired and I am a beneficiary of a DB pension scheme. With this background, I hope I can make a useful contribution to the inquiry.
I was profoundly disturbed by the events following the mini-budget, media reports on the potential losses, and possibly worse, suffered by DB schemes, and the role of LDI as the cause. I was aware that my pension scheme, along with many others, is running at a deficit and has done for many years, as disclosed by the annual report I receive from the trustees, and that therefore there is a risk, over the medium to long term, that my pension might need to be adjusted downwards, or no longer be increased in line with inflation. Although the trustees are always reassuring in that regard. I was not aware that substantial proportions of the assets of DB pensions schemes (including mine) have been managed under LDI and this methodology exposes the schemes to significant risk of loss over the short term (that is actual loss and not merely paper loss).
By which I mean that if the trustees had simply invested that portion of the schemes required to be invested in long dated gilts, to match their long term liabilities (contractual obligations to current and future pensioners), the drop in gilt prices in reaction to the mini budget would merely have resulted in a drop in the current value of the gilt portfolios. There was never any suggestion that the government would not make interest payments as they arise or pay back the principal at maturity. Furthermore, as the interest rate on long-term gilts rose, on a discounted cash flow basis, the current level of assets required to meet the schemes long term liabilities actually declined.
One might add that in these circumstances, the reaction of the gilt market would have been far less severe and would probably not have required the substantial (or indeed any) intervention by the Bank of England to stabilise the market. That intervention appears to have been essential because, for the last 20 years, DB schemes have increasingly, through the adoption of LDI strategies, exposed themselves to the risk of having to crystallise losses, not as a result of the slow build-up of deficits which might render them unable to meet their ongoing long-term obligations to pensioners, but as a result of severe, but short-term, market disruptions, as happened in the last week of September. The first, most pensioners fully understand. The second will have come as a disturbing surprise to almost all of us. It seems to have come as a surprise to The Pension Regulator and perhaps to the Bank of England also.
If the Committee’s experience mirrors mine as a regulator, its enquiry will be viewed as potentially threatening the status quo. The Committee will receive submissions, often highly technical in nature, from investment management firms with substantial commercial interests in maintaining the status quo. These will far outweigh (in volume if not in analytical quality) submissions by others providing critical analysis. These submissions will, as is to be expected, identify and emphasise what they view as the positive elements LDI brings to dealing with the issues around meeting DB funds’ long term liabilities. They will almost certainly minimise the short term risks LDI has created for the funds and which recent events have exposed.
In the first part of my submission, I will deal with the Committee’s questions on DB scheme governance, trustees understanding of the risks involved and whether LDI is still fit for purpose. In the second part I will discuss the competence of The Pensions Regulator (TPR) and suggest possible structural changes. I have made some amendments having regard to the thoughtful speech given by Sarah Breeden, Executive Director, Financial Stability, Strategy and Risk of the Bank of England on November 7.
DB governance, trustees’ understanding of risk in LDI and whether LDI is still fit for purpose.
While seeking to educate myself in the use of LDI by pension funds I came across several documents which I can only describe as marketing brochures produced by some of the biggest LDI managers. Generally, these emphasise the benefits and minimise the risks. This is not unusual in such material. However, given that the target readership is presumably DB scheme trustees, many of whom (particularly perhaps those with responsibilities for small DB funds), will have very limited understanding of the technicalities of LDI and its reliance on the use of derivatives and leverage, it is essential that risk as well as reward is fully and clearly explained. In my view this is not the case in the documents I have seen.
I wish to draw the Committee’s attention to the content of one such document in which, in my view, that required level of transparency is not met. A one page appendix (page 11 of 13), purports to explain how swaps and repurchase (repo) agreements work and touches briefly on risk, though not on the risks to which, realistically, LDI managed funds are primarily exposed. As the Committee is aware, these products were at the core of the recent disruption. I do not claim to be an expert in the technicalities of pension fund management in general or LDI in particular. However, I do understand bond markets and the ways they are financed. The appendix describes products with which I am familiar but in a very high level way and which, while not untruthful, obscures the short term risks (actual and not paper) to which LDI exposes DB funds.
Two short paragraphs from the appendix, and my comments thereto, will illustrate my concerns. I have numbered the sentences for ease of reference. The first reads as follows.
The first two sentences are factually correct, although they omit recognition of the legal complexity of these contracts if the rights of both parties are to be secure.
The third is misleading. If the value (market price) of the bond declines, not only does the value of the repo to the fund decline but the value of the gilts put up by the fund as collateral also declines. This requires the lending institution to demand additional collateral, by way of a ‘margin call’ and for the fund to meet this in cash or additional gilts, usually on a same day or next day basis. This is what happened in September, although some commenters have noted that it was already happening, though to a manageable degree, to some funds over the summer as interest rates rose.
The fourth implies that using repo is a win/win position for the pension fund. It is not. As with sentence three, it is misleading. The fund has exchanged long term risk (which can be mitigated over time) for significant short term risk which, if it crystallises, as it did as a result of the mini-budget, must be dealt with immediately. The fund has entered into a secured borrowing arrangement, usually with a bank, in which it has put up its gilts as security for what is, in economic terms, a cash loan for an agreed period of time. As with any secured loan, think of a home mortgage, the funds obtained are less than the market value of the collateral (gilts) at the time the transaction is entered into. This initially protects the interests of the lending bank. Far from “offsetting the impact of changes in interest rates”, if rates rise, ie the value of the gilts decline, the bank will continue to protect its interests by demanding more collateral as I have described re the third sentence. This transaction is beneficial to the pension fund only if rates decline ie - gilt prices rise. In which case the fund may be able to take back some of the gilts it has put up as collateral. I would suggest that, given that until recently, rates have been at historic lows, it would have been a brave, or reckless, fund manager who continued to bet on rates not rising at some point in the near term.
I found sentence five in this paragraph even more alarming. Although not a specific recommendation as an investment strategy, the reference to “many pension schemes” strongly implies it is a suitable investment management technique. Simply put, it suggests that having borrowed cash from a bank by putting up its gilts as collateral, it should use the cash to buy more gilts which it can in turn repo and buy more gilts, and it can repeat this process until the money it generates on each repo is used up. This presumably goes a long way to explaining why, the report commissioned by TPR, DB Pension Scheme Leverage and Liquidity Survey in December 2019, found that some funds are leveraged up to six times. This is getting into the realms of leverage expected in high risk hedge funds or companies acquired by private equity firms using very high levels of debt to equity and not, I would suggest, in supposedly conservatively managed DB funds.
An alternative way of looking at what is going on, as the fifth sentence makes clear, is that highly leveraged LDI managed funds have been creating additional assets in the portfolio. This is similar to the way that the monetary multiplier in bank deposit taking and lending operates to increase the amount of money in the economy. But banks are heavily regulated in this activity by the Bank of England and the PRA, with substantial mandated reserves to meet unexpected losses. No such regulation covers pension funds.
The second short paragraph purports to explain risk in swaps and repo. But it only deals with counterparty risk in undertaking these transactions. That is the risk that the lender will default at the maturity of the contract, and how that can be mitigated. While this risk is real, it is unlikely to result in crystallised losses for a fund if, as the paragraph recommends, the risk is mitigated by trading with a range of counterparties (such as leading British and international banks with strong credit ratings and balance sheets - my addition).
A trustee without sufficient detailed knowledge of the operation of swaps and repo might take from this that this is the only risk to which a fund is exposed if it is managed using LDI methodologies. This is not the case. And for the reasons I have set out above, it is the risk least likely to arise.
There is no reference to, or description of, market risk and in particular interest rate risk, that is the risk that short term interest changes, and the associated changes in gilt prices, can very rapidly expose a fund to significant losses, including, in extremis, losing significant proportions of the gilts it purchased to meet the funds’ long term liabilities. This will happen, if a fund fails to meet a margin call and the lender exercise its legal right to take control of, and dispose of, the gilts the fund have put up as collateral. To find any warning as to the risks in the use of derivatives and leverage in LDI one must look to the final page. It is found buried among the regular boilerplate that is standard at the end of all such documents. It takes up all of six lines.
Conclusions on LDI and its adoption by trustees of DB funds
First, in my view, LDI invested DB funds have been transformed from long term, conservatively managed assets pools, acting to preserve the interests of many thousands of ordinary people and the firms which employ or employed them into highly leveraged, short term focussed, speculative vehicles. As such LDI is not ‘fit for purpose’ if indeed it ever was.
Second, the marketing document I have analysed fails to properly disclose the substantial, and short-term nature of the risk inherent in LDI that a fund will have assumed. It also actively encourages increasing risk by assuming ever higher levels of leverage. This is probably not an isolated example.
Third, when I was a regulator, and if I then had the power to approve such marketing documents, I can, without hesitation, state that I would have required substantial changes before granting approval.
Has The Pensions Regulator taken the right approach to regulating the use of LDI and has it had the right monitoring arrangements?
I note that in her November 7 speech, Ms Breeden declared that the Bank intends to take action to deal with ‘poorly managed leverage in the non-bank financial system’. Which raises the question as to how long this threat to DB funds, the gilt market and financial stability has been going on. Which brings me to the second part of my submission; the role of the regulator.
The recent history of TPR does not engender confidence. Your 2018 joint Committees’ report on its handling of the Carillion collapse described TPR as ‘feeble’. Despite its Board accepting MPs criticisms and committing to a cultural change, have matters really improved? Furthermore, has its technical and analytical expertise increased to keep pace with changes in the industry and the markets? As far as I can see, the report commissioned by TPR, DB Pension Scheme Leverage and Liquidity Survey in December 2019, referred to above, raised no alarm bells.
Conclusions on the role of The Pensions Regulator
In my view the Committee’s questions should both be answered in the negative. I will unpick that statement as follows.
Whatever changes (if any) to the institutional structure of regulation of pension funds will be made, I would suggest that consideration be given to giving the regulator a stress testing power, at least for the largest DB funds, as the Bank has for banks. According to the DWP’s recent consultation paper, out of a 2021 total of 5,220 DB schemes, there are 186 with more than 10,000 members and 160 with 5 – 10k members. Though many smaller funds are managed using LDI on a pooled funds basis, the operation of these pools raises additional risks as the Bank has highlighted. Ms Breeden has also referred to a need for stress testing in her speech.
November 2022