LDI

Today’s post looks at the LDI kerfuffle from a couple of weeks ago.

What is LDI

LDI stands for liability-driven investment.

• It’s a risk-management strategy used by DB pension funds to help ensure that they can meet their future liabilities.

UK Pension funds apparently hold £2.5 trn in assets, of which 72% are in bonds.

• This breaks down 25% government bonds, 28% corporate bonds and 47% index-linked bonds.

LDI uses derivatives (such as interest-rate swaps) to minimise the impact on pension portfolios from changes in inflation and in interest rates.

• The big players in LDI are BlackRock, Schroders and Legal & General, along with specialists like Cardano and Insight.

LDI is reasonably popular – according to the Investment Association, the amount of liabilities that it hedges rose from £400 bn in 2011 to £1.6 trn in 2021.

• Which would seem to be slightly more than 60% of UK pension fund pots, and the vast majority of bond holdings.

---

On the Principles and Interest blog, Toby Nangle from Columbia Threadneedle gave a more detailed explanation:

Imagine I’m a scheme with £100 of assets and £100 of liabilities. I put £20 into ‘matching assets’ – say, long gilts. I put the remaining £80 into low-volatility growth assets that might beat cash like short-dated bonds, private credit, diversified growth funds and equities.

This creates a mismatch between assets and liabilities (falling yields act on all of the liabilities through discounting, but only on the 20% of matched assets).

To hedge against this risk that bond yields move, I enter into an interest rate swap, receiving fixed for 20yrs and paying floating on £80.

Now Toby is matched, but his swap counterparty needs security against him not being good for paying the floating rate over 20 years.

Luckily I have £20 in long-dated gilts! This is more than enough, and more than I am ever likely to need. Most of this will count as ‘excess collateral’. If yields rise, this eats into my ‘excess collateral’. I might need to sell down some of my growth assets to replenish it.

Schemes are protected against sharp market moves by over-collateralisation calculated through stress tests.

• But you aren’t protected against moves larger than your stress tests imagined.

What’s wrong with LDI?

In MoneyWeek, Max King pointed out a couple of problems:

Without knowing how long the members would live, matching assets to liabilities was an actuarial estimate, not a certainty.

More importantly, there weren’t enough long-dated gilts to match the liabilities, so it was just assumed that the proceeds from maturing gilts could be reinvested at the same yield. If yields fell, as they did until 2021, a gap between assets and liabilities would open up and the scheme could become underfunded.

To make LDI work for underfunded schemes, you need derivatives (swaps) rather than physical gilts.

Even with the required collateral, much less capital needed to be invested, thereby freeing capital to be invested in other assets in an attempt to catch up on the shortfall.

This required leverage which was fine as long as the swaps generated profits (because gilt yields fell).

The switch into LDI is likely to have driven gilt yields to unsustainable lows (0.68% for ten years at the end of 2021) and the selling of UK equities may have depressed valuations.

But when yields rise, there are problems.

What is the impact of rising bond yields?

Rising bond yields are not bad for pension funds – they both reduce pension scheme liabilities and assets.

• The Pension Protection Fund (PPF) calculates that a 30 bps rise in gilt yields reduces scheme assets by 2.7% but reduces liabilities by 5.6%.

But they cause problems for LDI.

• The collateral used in LDI contracts is typically bonds, so the rise in yields makes this collateral worth less.

if the value falls enough, the pension fund will receive a margin call to top up the collateral, which will normally happen from cash and sales of other liquid assets.

• Often they will have days to make this payment.

But the moves last month were so large and so fast that funds didn’t have enough cash on hand.

• So they needed to sell another liquid asset, which was largely long-term index-linked gilts.

Unfortunately, this forms a vicious circle (also known as a “doom loop”) – selling gilts reduces prices and increases yields further.

• Which in turn means that more collateral is needed.

And the realisation that there is a doom loop leads to speculation about the size of the problem and to fears of contagion (a la 2008).

What happened?

After the mini-budget of 23rd September, UK bond yields rose by 30 basis points (their prices fell).

• The fall continued overnight when Asian markets were open and were expected to continue on Monday 26th.

Yields rose another 50 bps and there were fears that LDI fund managers would have to sell lots of long-term gilts into an illiquid market.

• £50 bn of daily sales were expected, in contrast to average daily volumes of £12 bn.

On the 27th, yields rose by another 67 bps.

• This meant that LDI funds might have negative value, and be forced to liquidate (meaning lots of sales of gilts).

What did the BoE do?

On the 28th, the BoE announced that it would start to buy gilts, which led to a 100 bps fall in yields.

• The initial series of purchases would run to October 14th and total £65 bn (a maximum of £5bn per day for 13 days).

The planned gilt sales scheduled for early October have been delayed until November.

Note that this is not a £65 bn bailout of pension funds, as the mainstream media has reported.

• Less than £5bn will be used on most days, and bonds are being purchased.

The BoE might even make money in the medium term.

• Alternatively, yields might continue to rise and the scheme could be extended. (( Indeed, it has already been extended to include index-linked bonds ))

It is also worth noting that DC pensions would not receive the same government underwriting in times of stress,

What would have happened without intervention?

The Bank of England told the Treasury Select Committee that without intervention, some of the investments in LDI funds would have become worthless.

Deputy Governor Jon Cunliffe wrote:

Had the bank not intervened on September 28, a large number of pooled LDI funds would have been left with negative net asset value and would have faced shortfalls in the collateral posted to banking counterparties.

Liquidity conditions were very poor, and market intelligence calls identified the first concerns from LDI fund managers about the implications of market  developments, should they persist.

In some LDI funds, the speed and scale of the moves in yield and consequent decline in net asset value far outpaced the ability of the DB pension fund investors to provide new capital in the time available.

While it might not be reasonable to expect market participants to insure against all extreme market outcomes, it is important that lessons are learned and  appropriate levels of resilience ensured.

What should pension funds do now?

According to consultant John Ralfe, they should buy more bonds:

Schemes must use this (apocalyptical) opportunity to lock in their improved funding position by selling risk assets and buying matching bonds.

John was once in charge of the Boots pension schemes and moved it entirely out of equities into bonds. (( Since he did this in 2001, he also avoided losses on stocks in the bear market of the time ))

• This was the start of LDI, though not the flavour involving derivatives.

Jeremey Warner wants them to own fewer bonds:

Over-investment in bonds that finance government spending – rather than equities that drive business expansion – is ultimately a curse on growth and productivity-enhancing investment.

In practice, most DB schemes (apart from public sector ones) are now closed and are moving into run-off (as members retire).

• This is likely to mean that they become net sellers of gilts, and it’s not clear who might replace pension funds as buyers.

The Pensions Regulator wants better risk management:

We again call on trustees of DB schemes and their advisers to continue to review the resilience and liquidity of their investments, risk management and funding arrangements, and plan accordingly to protect the interests of scheme members.

Will the BoE intervention work?

Max King sees problems:

The liquidity provided by the Bank of England increases the money supply and will filter through into higher inflation and hence higher gilt yields.

And there will be continuing losses from the LDI schemes since gilt yields aren’t heading back to 1%. None of the options looks great:

They can admit that the LDI scheme has failed and go cap in hand to the government for a bailout. They can seek to reduce the benefits to members for which they are contractually liable.

They can seek to pass the scheme on to the already overstretched Pension Protection Fund. Or they can seek to make up the shortfall by switching the fund to invest in supposedly higher-risk, higher-return assets such as equities.

There is also the nightmare scenario (which doesn’t seem to have happened) of a fund being whipsawed out of its LDI hedge just before the BoE action crashed yields by 100 bps.

To be continued …

I expect to update this article next week, or perhaps write a follow-up.

• Until next time.