Reflections on the 2022 LDI volatility

Joseph Evans

Joseph Evans

Senior Vice President, Investment Consulting
(Monday, Oct, 03, 2022)
|   5 mins

This article was written shortly after the peak of the LDI volatility, on 3rd October 2022.

In the below video I recorded back in May, I was describing the steps pension schemes could take to efficiently manage LDI collateral. In reality, I’d spent my entire career in a low and declining interest rate environment, so I had no real concept of what a collateral squeeze could look like.

The moves in gilt yields last week were remarkable – larger than anything predicted in even the most extreme risk model scenarios.

Risk models typically use a 1-year period, but these moves happened over a few days – the speed of the moves is what has caused the headaches across the DB pensions industry over the last week.

Let’s put these moves into perspective using the two charts below.


Almost 20 years of gilt yield compression unwound in 9 months.


(Source of charts: Bank of England)

How large were the 1-day moves from last week in historical context?

The largest 1-day move in 30-year real gilt yields on record before last week was 0.31%. The move on Monday was 0.68%, and the move on Tuesday was 0.76%. So, we had daily moves more than double the previous largest 1-day move on historical record …in consecutive days.

Around 10:52am on Wednesday, 30-year real yields were up another c.0.55%. Fast forward to the end of that day, we had seen the record for a 1-day move broken for the third day in a row – but this time the other way around! The £65bn gilt-buying promise the BoE announced at 11am had sharply driven 30-year real yields down 1.08% that day and given pension schemes some room to breathe.

Thursday and Friday increased that breathing room further, with 30-year real yields decreasing by another 0.81% across the two days. Now that’s some volatility!

There’s no doubt I’ll remember this week for a very long time. But before this becomes just a distant memory, I wanted to share some of my immediate reflections:

  1. The power of teamwork (cheesy, I know) 🧀 The rapid collaboration between Redington colleagues, trustees, in-house teams and LDI managers has been the key to guiding our clients through in one piece. The traditional advisory structure is sometimes criticised for slow implementation – however, the pace of decision making from our clients has been brilliant.
  2. The value of ‘live’ tech tools 💻 Investors have had a need to see the latest information possible – I am so grateful to our tech team for producing a tool that presents an intuitive picture to clients of their latest position. Being able to see the first-order impact (i.e. move in their funding position) is obviously very helpful, but also having answers to the second-order questions at the tip of our fingers – e.g., how much collateral would we have left if rates increased another 1%? What does our asset allocation look like now? What is our latest expected return and risk? What is the attribution of our funding level move? – has allowed our clients to make better-informed decisions quicker.
  3. ‘Risk management should be put in place in the good times to have most effect in the bad times’ 🔐 This is one of the key Redington investment principles and it’s never been more relevant than last week. Decisions made by clients years ago to set up LDI collateral rebalancing frameworks and a pre-agreed ‘collateral waterfall’ of assets has significantly reduced risk over the last week.

And here are some of my takeaways moving forward:

  1. There’s a mountain of work left to do in the immediate term 🌄 The BoE intervention is temporary – they say their gilts purchases are only up to 14th October. The intervention might have stabilised markets for now, but no one knows whether this will remain. Schemes will need to replenish LDI collateral pools to healthy levels before 14th Oct, where possible – this means more asset transfers to arrange.
  2. Trustees should reassess their journey plan 🚅 Most DB pension schemes in the UK will have a meaningful level of liability hedging, which would have protected their funding level from most of the impact from the overall negative move in real yields last week (assuming they were not ‘stopped out’ of their hedges). Trustees should work with their advisers to check if their current journey plan is still appropriate.
  3. Trustees should ask – how liquid are our  ‘liquid’ funds? 💧 Now more than ever, schemes need to be confident they can rebalance from their liquid funds quickly. Funds labelled as daily/weekly/monthly liquidity, but with long redemption notice periods and settlement periods, might no longer be viewed by trustees as a liquid fund.
  4. Barbell’ strategies might become more common 🏋🏻 Risk models across the industry will need to be recalibrated. The historic moves we have seen might mean the ‘new norm’ for minimum levels of capital to hold in LDI portfolios to back hedges will increase. This implies schemes would have less capital in return-seeking assets. If schemes maintain a similar return target, this implies the return-seeking assets in the portfolio will need to have a higher expected return.
  5. Potential issues for CDI strategies? 🧩 One of the core tenets of Cashflow Driven Investment (“CDI”) strategies is to hold bonds to maturity – if implemented well, it can be an intuitive and effective solution for pension schemes. I would guess the last week has tested the flexibility of these strategies and some schemes may have had to sell bonds earlier than they were expecting. If so, this may have had a knock-on effect on the cashflow matching design.

It has been a stressful but memorable experience to be in the thick of it. But after several financial ‘crises’ in the last couple of years, I’d be happy to see a calmer period in markets for a while! 😅

If you have any views and/or challenges to add to my thoughts above and experiences from this week, please get in touch:

Unless indicated, these are the views of the author and may differ from those of the firm.





Pin It on Pinterest