Written evidence from John Ralfe Consulting Ltd. (LDI0059)

The Rt Hon Sir Stephen Timms MP

Chair Work and Pensions Committee

House of Commons,

Palace of Westminster,

London SW1A 0AA                                        12th December 2022                                                      

Inquiry into DB pensions and “Liability Driven Investment”

Dear Sir Stephen

Following my earlier written[1] and oral evidence[2], I would like to make some further comments.

1. The Committee has heard a lot about LDI versus “Leveraged LDI”, and Members should be clear about the crucial difference between the two.

LDI is simply matching pension assets and liabilities. This matching or hedging reduces risk for scheme members, the sponsoring company, the Pension Protection Fund, and the financial system as a whole.

Meanwhile, “Leveraged LDI” is identical in economic terms to a pension scheme borrowing, and then buying assets which do not match liabilities – ie quoted equities, private equity, hedge funds, property etc. This is not hedging, but a bet that the value of the “risky” assets will go up more than the value of the borrowings.

Unlike LDI, which reduces risk, “Leveraged LDI” is speculation, and increases risk for scheme members, the sponsoring company, the Pension Protection Fund, and the financial system as a whole.

With the scale of “Leveraged LDI” now being disclosed, it seems many pension schemes, and their sponsors, have become badly-run hedge funds in the last few years.

Furthermore, I question how many trustee boards using “Leveraged LDI” really understand the risks they are taking in these opaque, complex and expensive arrangements.

2. Some evidence the Committee heard – especially from investment consultants who I described as “the villains of the piece” in my evidence - has downplayed September’s meltdown as just a problem in the  “plumbing”, and once this is sorted out, “Leveraged LDI” life can continue as before.

We know borrowing by pension schemes is banned by law. But “Leveraged LDI” allows pension schemes to drive a coach-and-horses through this legal ban, in a way which is hidden from scheme members, shareholders, credit rating agencies and regulators.

This hidden borrowing, and speculation, led to September’s meltdown, and a taxpayer guarantee through the Bank of England – “moral hazard” in action. My view is that “Leveraged LDI” should be prohibited.

2. My original conclusions and recommendations still stand.

“The Committee should be absolutely clear that the real underlying problem is not “LDI” – which is hedging - but “Leveraged LDI” – which is speculating.

My “interim” recommendations to get-to-grips with “leveraged LDI” are:

a. The Pensions Regulator, the Financial Conduct Authority and the Bank of England should produce better, and more coordinated information about the current extent of leverage in pension schemes. The Purple Book [3] on DB schemes  produced by TPR and the PPF is silent on “Leveraged LDI”.

b. The International Accounting Standards Board and the UK Financial Reporting Council should require employers to disclose more information in their published accounts about the current extent of leverage in their individual pension schemes.

c. Current regulations from 2005 prohibit borrowing by pension schemes with more than 100 members.[4]  These regulations should be strengthened and clarified to prohibit pension schemes taking on leverage, however cleverly disguised, as well as explicit “borrowing”.

3. My two recent Financial Times articles also stand:

“Investigation needed to hold those behind UK pension crisis to account” [5]

“BT’s enduring pension problem” [6]

Please feel free to ask any questions.

Yours sincerely

John Ralfe

Additional comments:

1 The trigger for the liquidity squeeze was “Leveraged LDI”, not LDI

"LDI" is simply matching DB pension assets and liabilities - selling equities, and buying long dated matching bonds, exactly what Boots pioneered 20 years ago, to hedge obligations to pay pensions.

Most asset and liability matching can be done through selling equities and other “risky” assets, and buying bonds. Interest rate swaps can also be used to improve matching, especially inflation matching, again as Boots pioneered 20 years ago.

Swaps are “marked-to-market” with collateral or margin calls being paid or received by the pension scheme as interest rates move up or down. The collateral can be paid by the company sponsor directly, as with the Boots swaps, 20 years ago.

Suppose a pension scheme had a £100m “vanilla swap” in place this September. As gilt rates rose sharply after the “mini” Budget the scheme would have been required to post some collateral, but only for the change in value of £100m.

But a scheme with a £300m three-times  “leveraged swap” would have had to post three-times as much collateral, aggravating its liquidity squeeze. This is just a matter of arithmetic - the more swaps outstanding, the higher the required collateral.

The same principles apply to other “Leveraged LDI” mechanisms, such as “gilt repos”.

Smaller schemes tend to use “pooled” leveraged gilt funds for “Leveraged LDI”.

Suppose a scheme held £20m in a 2055 gilt fund (as well as other gilt funds). As gilt rates rose sharply in September the value of the gilt fund would have fallen, but the scheme would not have been required to come up with any cash.

But suppose a pension scheme held £20m in a 2055 Leveraged gilt fund, with a leverage of three-times.

As gilt rates rose sharply in September the fall in value of the Leveraged gilt fund would be three-times that of the vanilla gilt fund. To maintain the agreed leverage limits the scheme would have had to put in cash equal to three-times the fall in the value of the underlying gilt.

2 What was the Bank of England trying to avoid in buying gilts?

It is not clear why the Bank of England felt it needed to intervene, who it was trying to protect, and what it believes would have happened without its intervention. Were we really on the edge of a financial meltdown, with obvious and immediate public policy consequences, which justified the “moral hazard” of underpinning gilt prices?

Without Bank of England intervention either pension schemes would have sold assets, or their company sponsors would have posted collateral directly on their behalf, or extended a line of credit.

We don’t know how many of the 5,000 companies with DB pension schemes had “Leveraged LDI”, but I believe that most – especially the larger companies - would have just posted collateral or extended a credit line.

Some companies, including J Sainsbury [7] and DS Smith [8] have disclosed they did just this.

There is also some ambiguity about whether Lloyds Banking Group at least discussed providing a credit line, although it is unclear what, if anything, actually happened (see point 6).

Some smaller, or weaker companies would not have been able to post collateral or extend a credit line. These are likely to have been a small percentage of total liabilities and members.

The worst that would have happened is that this could have triggered administration of these smaller and weaker sponsoring companies.

If so, pension scheme members would have gone into PPF and company assets would have been sold. No one would wish this, but it is a million miles away from a financial apocalypse.

3 Evidence from Barnett Waddingham

a. In my oral evidence I said that the extent of “Leveraged LDI” is “hidden”, and not disclosed to pension scheme members,  shareholders, credit rating agencies and regulators in company accounts,  and I used BT as an example.

In the December 7th evidence session Sir Desmond Swayne asked the panel to respond to my claim that “Leveraged LDI” is “hidden”.

Mr Rod Goodyer, of Barnett Waddingham, said, “I don’t believe leverage is being hidden. It is in the public domain”.

Using my example of BT, he said BT’s Statement of Investment Principles was clear about the nature and extent of “Leveraged LDI”, and that all scheme SIPs were publicly available, because they could be “Googled”.

The BT Pension Scheme Statement of Investment Principles certainly can be found through Googling. [9] However, BT’s 8-page SIP contains absolutely nothing about the extent and nature of BT’s “Leveraged LDI”

I have also downloaded the SIPs of a number of other pension schemes, chosen at random:

J Sainsbury[10], Northern Foods[11], University of Manchester Superannuation Scheme[12], Lloyds Bank Pension Scheme No2 [13] and D.S. Smith [14]

None of these SIPs give any information about the extent and nature of their “Leveraged LDI”. Clearly this is a small sample, but it would be helpful if Mr Goodyer could provide examples of SIPs which do contain information on “Leveraged LDI”.

Pension scheme accounts – not usually publicly available – may provide some information, but it is still down to guesswork, even for experts.

Even if it is possible to make some educated guesses by looking at the SIP and pension scheme accounts this is wholly inadequate. The information on “Leveraged LDI” should be crystal clear in each company’s published report and accounts, submitted to Companies House.

I challenge Mr Goodyear to open a copy of any company’s published accounts, and point to the explanation and disclosures on “Leveraged LDI” in the pension notes.

I stand by my evidence that the scale of “Leveraged LDI” is hidden, and not in the public domain.

Because there are no accounting disclosures required for “Leveraged LDI” in IAS19 [15] or FRS102 [16] I have recommended that the IASB and FRC should require companies to be explicit about the extent and nature of Leveraged LDI in its pension schemes.

At the macro level the information on “Leveraged LDI” produced by the Pensions Regulator and other regulators is also sketchy.

b. Barnett Waddingham’s written evidence said that:

“DB pension schemes are forced, by their very construction, to mismatch assets and liabilities.  There is therefore a tension between managing asset/liability mismatch (i.e. buying gilts to match liabilities) and generating the returns required in the funding plan to meet benefits in full.  This is the reason why leveraged LDI was invented” [17]

This claim is very misleading.

DB pension schemes are not “forced” to mismatch assets and liabilities, they can simply choose to hold long dated bonds, and interest rate swaps, to match their pension liabilities.

The claim that somehow “Leveraged LDI” “was invented” as the only way to allow benefits to be paid is like saying the only way I can pay off my debts is to bet money at the casino.

Pension deficits must be paid by sponsoring companies, so it is up to them to get out their cheque books and increase contributions. And if they want to take bets on their own balance sheet, that is entirely between them and their shareholders, but they should not bet with pension scheme money.

c. Barnett Waddingham also said that:

“ LDI has had very real benefits to both members and sponsors of DB schemes by protecting them against increasing funding deficits over the last decade or so.  LDI has been a stabilising force for the UK economy”. [18]

It may be correct to say that pension scheme “Leveraged LDI” bets have paid-off “over the last decade or so”, ie risky assets have outperformed matching bonds, so schemes’ funding position is better than it would have been otherwise.

But this is no more than saying “pension schemes have taken bets and the bets have (so far) paid off”.

“Leveraged LDI” increases risk for the overall financial system, so is a “Destabilising – not stabilising force- force for the UK economy”.

And the extent of “Leveraged LDI” is hidden from pension scheme members,  shareholders, credit rating agencies and regulators, because of poor accounting and poor corporate governance.

The history of financial markets teaches us that  “hidden leverage” always eventually ends-in-tears.

5 Evidence from BT Pension Scheme

BT Pension Scheme said that: [19]

“at the Scheme’s last triennial valuation in 2020, the Scheme’s funding deficit was £8.0bn. We estimate that in the absence of the LDI hedging programme, the deficit would have been £7.6bn higher (i.e. £15bn or more) that would have required BT to pay significant additional contributions to repair the deficit. For context, at the date of the triennial valuation, June 2020, the market capitalisation of BT was c £11bn.”

BTPS’s statement could be re-written as: “We estimate that in the absence of speculating through borrowing to buy equities, private equity, hedge funds, and property, the deficit would have been £7.6bn higher”.

If BT, the company sponsor, wants to plug the huge pension deficit not by cash contributions, but by speculating, it should do so on balance sheet, transparently to shareholders and creditors, and not bet with pension money.

Bad disclosures and bad corporate governance allow this betting in BT’s Pension Scheme, couldn’t happen on BT’s company balance sheet.

Although 60 per cent of BTPS’s assets were in matching bonds, cash and “secure income”, 40 per cent of assets at June 2022 - about £19bn - were in  “equity-like assets” — public equities, private equity, property, hedge funds, infrastructure and “non-core credit”.

The £19bn in “equity-like assets” is much larger than BT’s £12bn market capitalisation. BT is on the hook for all pension deficits, so in economic terms, holding £19bn of “equity-like assets” in its pension scheme is identical to BT borrowing £19bn long term, and then buying those assets directly.

BT’s market cap is therefore sensitive to movements in the value of “equity-like” pension assets — as a matter of arithmetic, a 10 per cent fall in their value hits BT’s market cap by about 10 per cent after tax.

About £15bn of these assets are unquoted, and BT’s auditors list “the valuation of unquoted plan assets” as their first “key audit matter”.

I would also point-out that BTPS has a “Crown Guarantee” so the “Leveraged LDI” bets are, ultimately underwritten by unsuspecting taxpayers.

And, for the sake of accuracy, BT’s market capitalisation at June 2020 seems to have been £13.75bn not £11bn, as stated by BTPS [20]

6 Evidence from Mr Henry Tapper on Lloyds Banking Group

Mr Tapper's written evidence [21] to the Committee - which he has now partly withdrawn - has created real confusion about the impact of September's financial meltdown on Lloyds Banking Group three pension schemes.

If he still stands by his other comments it would be helpful if he provided his evidence.

Mr Tapper's comments have helped in bringing attention to the LBG pension schemes, which are very large in relation to the size of LBG itself.

The 2021 accounts show pension liabilities of around £50bn, more than LBG's market cap, and a £6bn deficit, which the bank is paying off over time. There is no suggestion that LBG has a problem with pensions, but it would be helpful if the trustees and LBG made a formal statement explaining exactly what happened.

7 Evidence from Professor Iain Clacher on the “missing £500bn”

For the sake of completeness, I include below my earlier letter to the Committee.

Dear Sir Stephen

Work and Pensions Committee Inquiry into DB pensions/ LDI

Thank you for the opportunity to give evidence to the Committee, which I hope you found useful.

During the session Professor Clacher said that £500bn of pension fund assets had been “lost” as a result of the LDI debacle.

The Financial Times reported this on its front page, with the headline. “Gilt crisis was major factor in £500bn hit to UK pension funds, MPs told”[22]

The FT said: “Giving evidence to the Commons’ work and pensions select committee, Iain Clacher, a professor at Leeds University Business School, said based on his calculations “roughly £500bn is probably missing somewhere”. “And this isn’t a paper loss,” he added. “This is a real loss because pension funds were selling assets to meet the collateral calls.””

I am writing now to draw your attention to the inconsistency between Professor Clacher’s £500bn figure and the monthly “official” figures produced by the Pension Protection Fund.

1. The PPF’s figures show the drop in pension scheme assets from August 2022 - the month end before the “mini”-Budget – to September- the month after -  was just £115bn.

Even if we ascribe all of this £115bn to the “mini”-Budget, this is a fraction - less than a quarter - of Professor Clacher’s £500bn figure.

2 This £115bn is the drop in market values of pension scheme assets – a ”paper loss”, not the “real loss”, as Professor Clacher suggests. The “real loss”, based on actual asset sales by schemes, will be significantly less.

Professor Clacher’s explanation of the £500bn figure may be in his written evidence, but it would be helpful for him to provide a short explanation.

Fall in asset values attributable to the “mini”-Budget

The PPPF publishes monthly estimates of the value of assets held by all private sector schemes -  https://www.ppf.co.uk/ppf-7800-index 

Recent figures are :

At August 31st – the month-end before the “mini”-Budget - pension scheme assets were £1,565bn.

By September 30th – the month after the “mini”-Budget - assets had fallen by 7 per cent - £115bn - to £1,450bn.

By October 30th, asset values had risen slightly, following the Bank of England’s bond buying intervention.

The PPF says the impact of small changes in equity values during September was neutral, so it is reasonable to assume the fall in gilt and bond asset values due to the “mini”- Budget was the full £115bn.