Thoughts from the WPC

I have re-watched the Work and Pensions Committee LDI evidence sessions, made notes on what was said and my comments. This took a lot longer than I thought but was well worth it as it helped me recognise some key themes from the 2 hours or so of evidence.

I have consolidated this thinking into the things I agreed with, the things I would challenge and also the things I think were missed. I have tried to group these into the general themes that ran through the meeting

The look and feel of the meeting was like a Trustee meeting

First of all though what struck me was the feel of the meeting. As someone who has been speaking to Trustees on LDI for over 15 years the WPC meeting actually had a look and feel of a Trustee meeting – with the committee as the Trustees and the panels as the advisor: The committee searching for answers and the best strategy going forwards and the technical experts trying to stay concise and to the point without sacrificing accuracy. The style of the questions being asked, some misunderstandings and the way they were answered really did seem familiar. Would love to hear whether other people felt the same?

Why pension schemes used LDI

I agree with both panels that regulations were one of the drivers and motivations for LDI.

Importantly though I do not think it they are the?only?motivation. What I mean is that I disagree with the implication though that if there were no regulations then there would be no LDI. This was very briefly touched upon in the second panel when sponsor covenant was mentioned and the failure of it resulting in benefit cuts - but not enough in my opinion. As I have posted before if you agree that the aim of a pension fund is to secure benefits then one way to do this is with an insurance contract and/or portfolio of bonds (the latter being the strong view of John Ralfe in the first panel). If this is your aim then hedging the risk of this ultimate strategy will be a motivation whether the regulations tell you it should be or not.

I have a strong disagreement with the statement made in the first panel about there being an “arbitrage” in using leverage because short rates (what you pay in a derivative) were low and long rates (what you receive) were higher. This is technically incorrect in two areas. Firstly the implication of using the word “arbitrage” is that you are certain to make a profit. This is not the case – the swaps that were described with low short rates and low long rates have zero value when traded and if there was an arbitrage this wouldn’t be the case. Secondly the reason the swap has zero value is because the long rate is the cumulative expectation of all of the short rates (including the current one) – the swap may involve cashflows flowing one way at certain points but these will be offset by cashflows flowing the other way at others making the whole thing zero value. Another point related to this was that this effect will be included in the value of the liabilities as they will be valued using the same market rates.

Leverage – is it legal?

Both panels agreed that reducing leverage should be (and has been) done. Clearly the first panel thought the leverage reduction should be greater than the second but importantly panel 1 (or at least part of panel 1) seemed to claim that leveraged LDI was illegal. This was picked up in?The Times?so I think it is important to address it.

There were two legs to this as far as I could tell: The first was that leverage (specifically in the form of Repo) is borrowing and therefore not allowed. The second was something to do with the transposition of European regulations into UK ones the essence of which I think was that you are not allowed to use derivatives to hedge liability risks.

I am not a lawyer so cannot comment on the specific legalities but in my view these two points don’t make sense.

On the first point yes there may be some ambiguity with Repo but you can achieve LDI with swaps and not use repo. Given swaps are derivatives that manage risk then it appears to me that it is permitted by the regulations. A strong point made in the second panel was that no scheme lawyers, in their experience, have ever raised the legality of LDI. I echo this in my experience.

On the second point I have no idea about EU investment regulations but what I do know is that Dutch pension funds do LDI using derivatives (e.g. as mentioned in this story:?https://www.dnb.nl/en/statistical-news/snb-2022/dutch-pension-funds-sell-record-amount-in-assets/). As a result even if the European regulation wasn’t transposed as mentioned then the fact that European institutions are doing LDI seems to suggest those regulations permit it. Again I may be misunderstanding and am no lawyer.

Leverage – was it hidden?

It was claimed in the first panel that the level of leverage was a “surprise” and was “hidden”. This was picked up in some of the pensions press: (https://www.pensionsage.com/pa/Hidden-leverage-blamed-for-DB-liquidity-issues.php?and?https://www.professionalpensions.com/news/4060665/hidden-leverage-blame-ldi-crisis-wpc-told)

To start with I agree that hidden leverage is bad.

I will park the inconsistency in the first panel of the leverage being both hidden and yet the consequences predictable and forewarned – in my view from a Trustee perspective it was certainly not hidden from a regulators perspective it is less clear.

In looking at LDI, in my experience, the following topics are generally covered: Why hedge liabilities, how much risk to hedge, how leverage allows you to hedge but retain growth, the downsides if rates go up, the mechanisms if rates go up. As such I would argue most Trustees will have been aware of the mechanisms and behaviours that happened (perhaps not prepared for the speed of them) so from their perspective it was certainly not hidden.

The regulators perspective is an interesting one. TPR undoubtedly does not collect as much information as it could do – with this I agree. On the other hand since around 2014 under European regulation (the European Markets and Infrastructure Regulation “EMIR”) all derivative positions have to be reported to regulators. I also believe there is equivalent reporting regulation for Repo as well. As a result the data on the size of derivatives positions was out there. The fact that regulators began to consider stress tests etc says to me that perhaps it wasn’t as hidden as the headlines suggested.

When does LDI work and when does it not?

This was a thread throughout the session – it was generally discussed that schemes and sponsors were happy having LDI when rates were falling and not happy when they were rising. I was actually left wondering is that the right way around?

LDI is a hedge and what you are doing with a hedge is managing the risk that you are wrong – it may have been “obvious” that rates may only rise which in my experience has generally been the view of Trustees but there is a risk they don’t. LDI hedges against that risk. Something not touched on in enough detail was the hedge ratio – i.e how much of this risk should be hedged. For those who are certain rates will rise (or there are no liabilities to hedge) then this ratio will be zero. For those who have no idea/don’t want to worry about interest rates this number is nearer 100%. In practice I think most schemes were somewhere between the two. If you look at the PPF Purple book it provides some funding sensitivities which implies the hedge ratio across the industry was 50% ish.

Working through this – if interest rates rise then Trustees are happy – their view is right and rates rose, the Liabilities fell and the LDI portfolio didn’t fall by as much as the liabilities – funding improves.?Yes?they may have regretted not hedging 0% but that is investment for you – very easy to do in hindsight. If interest rates fell on the other hand – i.e. the view of rates rising was wrong, then the funding position gets worse but not by as much as without a hedge – so yes you are pleased you hedged but you are worse off.

This distinction for me is critically important to understand because a lot of the portrayal of LDI in the press and in this committee was that Trustees were speculating that interest rates were going to stay low. In my experience this is just plain wrong. In my experience Trustees hoped rates would rise but wanted to temper that view with LDI because they couldn’t afford to be wrong.

Are schemes better off?

The follow on from the above point is around whether schemes are better or worse off. On one hand as described above they are better off as liabilities have fallen (caveat to this is covered in my small schemes point). On the other hand the assets are lower by ￡300bn-500bn as mentioned in the first panel.

On the latter point I think if you look at the PPF 7800 index (https://www.ppf.co.uk/ppf-7800-index) asset falls of this magnitude look right. Clearly a scheme with more assets is better than a scheme with less assets. However I think the absolute size of the assets as a measure of success has two key challenges to it.

The first is that the job of a pension scheme is to secure benefits – this can either be with bonds or with an insurance policy. If they don’t have enough money then this deficit falls on the sponsor and/or investment returns. LCP estimate that this deficit has shrunk considerably meaning that schemes are closer to being able to afford to secure pensions than they were a year ago.

The second is that only thinking of asset value as a measure of success is always doomed to fail – there will always be a strategy that will have resulted in more assets than the scheme currently has investing all assets in Tesla or some other single stock will inevitably have given more assets still but that isn’t to say it would have been the right thing to have done.

The nub of this point is that the schemes saw an improvement over time but some schemes will have done better than others but that isn’t to say that the strategy with more assets was the right one.

I do agree with the first panel through that those who didn’t do so well will have been those whose hedge was cut at the peak of rates and then put back on when rates were lower. In some cases I expect this would have been a choice but the more uncomfortable cases will be where this was done by the LDI manager unilaterally – I would expect this to be the case for pooled LDI funds as was described.

Investment consultants and investment advice

In my experience the evidence of the second panel is far closer to reality in terms of experience and knowledge of LDI.

As an aside the “test” about number of margin calls in the first panel for me was a bit false – 4 calls per day is for cleared derivatives and not all derivatives are cleared so I do not think that “test” was as useful as it came across.

The one challenge I would make is around pooled funds. For all other asset classes pooled funds are a sensible way to access an asset class. I have always said that this is not the case for derivatives as there are consequences of putting unfunded derivatives in a pooled fund. As such I think there is certainly a risk (in some cases not all) that it was presumed that pooled funds were more straightforward than they turned out to be.

Small schemes

There was a statement in the second panel that they thought most LDI was done by large schemes. I do not think that this is correct. Henry Tapper posted on LinkedIn after the meeting with some rough numbers and I would be inclined to agree more with that analysis.

This is important because small schemes are likely to be in pooled funds which is where the acute problems were.

Pooled LDI and product design

Those who follow me will have seen I have always questioned the efficiency of pooled LDI. The events of September highlighted the issues of isolating leverage within a pooled fund. I would write a whole piece on it but in essence as Con Keating said, a pooled fund cannot have a negative NAV and therefore the motivations of a pooled fund in times of stress are self-preservation. In the crisis there are two things that could have blown the funds up – a negative NAV and/or the counterparty banks closing out their trades (i.e. the pooled fund defaulting). As such the pooled fund, quite rationally, will exhibit behaviour to prevent this. In more normal market conditions this will manifest itself in the pooled fund asking for more money, this may move into asking for money more quickly and then finally it may involve unilateral cutting of exposure if the fear of default is too high.

One clarification I would make is that when schemes in pooled funds were asked for “collateral” this will have been as a result of the manager deciding what to ask for, how much and by when. It is unlikely to be a pass through straight from the counterparty bank.

A related point in the second panel was the worry about competitive pressures driving leverage up. In pooled funds I doubt that this will happen. The reason being that everyone is more alive to the risks of pooled LDI – including credit teams at banks. Arguably the primary worry of a pooled fund is not being closed out by the bank which means that what level of leverage is acceptable to the bank is probably the driving factor – product design is unlikely to be in the gift of the manager. If this is right then pooled LDI is likely to become more homogenous.

To be clear whilst segregated mandates will have been surprised by the size and speed of the movements, they do not have the same pressures about self-preservation and therefore have more tools to deal with a crisis – general commentary seems to indicate that this was the case in practice.

All in all it was an enjoyable session to watch and by and large what was said was reasonably sensible. My hope is that the right lessons are learned.