Mini Budget Crisis – What’s the damage? Part 2
9th December, 2022
To reference my previous blog discussing the LDI crisis and schemes which are now in a weaker position than before, let us recap the characteristics of schemes most likely to be worse of:
- Smaller than average
- High hedge ratio
- Lower funding level
- Operating off an “investment platform”
- Invested in a pooled LDI fund
- Allocated non-trivially to illiquid assets
- Inexperienced consultants (i.e. do not understand the collateral implications of derivatives)
- Inexperienced trustees (i.e. do not understand the collateral implications of derivatives)
By definition, you might be able to protect against 40 year interest rates by buying a gilt. But if you have to fund that gilt, you will be subject to margin calls. And those margin calls are for “immediate” settlement, a contingent “T+1 Risk” (“T+X” is my shorthand for how quickly you need the money or the money gets to you, where X equals the number of business days for the cash to move). Collateral calls are always contingent because on trading day 1 there is no collateral effect, and any “negative” effect (ie a draw on my liquidity) will not become apparent until (i) day 2 and (ii) only if rates move up.
So the scheme risks set out above matter because (taking them in order):
- Smaller than average tends to have lower quality of advice and resources to solve the problem of a DB scheme. A DB scheme is one of the most devilish asset-liability management (ALM) issues to solve for. It is far more complicated than any bank’s balance sheet. Banks’ balance sheets are run by bankers and fully trained risk professionals and scrutinised on a daily basis by the regulator. A DB scheme is, in general, run by people who have never run a risk-on balance sheet, who cannot afford proper advice and are regulated by at least an arms’ length.
- High hedge ratio means, all things being equal, more derivatives (aka “leverage”). These instruments are more exposed to T+1 Risk.
- Lower funding level means, all thing being equal, more derivatives. Only fully funded schemes which can afford to invest 100% in gilts can have a 100% LDI hedge ratio and no derivatives. A 90% funded scheme, which wants a 100% hedging ratio will have to have some derivatives with their embedded T+1 Risk, since if it spent all its assets on gilts, it would never fill the deficit. So within the portfolio mix, it might look like 60% allocated to LDI which would include 40% of derivative cover and 30% allocated to risk seeking assets.
- Operating off an investment platform adds “layer risk”: communications. To “see” what is going on within the LDI portfolio, a trustee has to talk to the platform provider and the platform provider has to talk to the LDI fund manager. And the response has to come back up the chain. And there has to be someone at each level who can action any inevitable instructions immediately because any actions have to settle the next day (T+1) and therefore, cash needs to be available the SAME day. So there is an inefficient communications mechanism which will simply fail if pushed too hard.
- Investing in a pooled fund versus a segregated account is, definitionally, another (part) layer of complication. The assets in the fund have their own liquidity and the fund has (likely) a different level of liquidity. The manager, in the LDI case, has a liquidity framework to adhere to but it is unlikely that the rules relating to fundholder failure (e.g. injecting liquidity on a T+1 basis) have ever been thoroughly practiced and it will not have been part of the sales pitch.So in the result of failure to get more cash into the fund by a fundholder, this will quite possibly been the first time the fundholder has been aware of the second order liquidity risk they have been running. And it is quite likely the fund manager is ill-prepared, as the MBC move in yields was outside any reasonable risk scenario generator (as the FCA recently acknowledged). Which is likely to have led some fund managers to operate outside of the fund’s documentation. This is of interest to those who were “knocked out” and to those who were “knocked in”: were these actions consistent with the discretions of the fund manager?
- For those schemes that were managing the crisis well so far, the next hurdle was to generate the liquidity to pay into the T+1 gaping maw of the LDI portfolio. It turns out that many fund managers and consultants, when describing investments as “liquid” failed to point out that this is a relative term. “Liquid” strategies trading on a more than a next day settlement basis are illiquid in the face of a T+1 collateral call. I suspect the Government’s efforts to persuade pension schemes to invest in illiquid infrastructure assets is doomed to failure since even “liquid corporate bond” funds were illiquid in this context and, if truth be told, corporate bonds are not liquid and never have been. Liquidity is only there when you don’t need it.
So, if I now needed to deliver £1m tomorrow (T+1), but had no money coming in until T+X (X being any number higher than 1), where did you turn?
Pension funds do not have lines of credit with banks (and I suspect there are legal hurdles to jump if you want to have one and use it in this liquidity scenario).
Pension funds do have sponsors, but the funds most likely to be most in trouble may have sponsors who are not able to lend. Further, a sponsor would reasonably not want to fall foul of any sums advanced being “surplus trapped”. And there is the board to agree (and not just UK boards). Solving this, even if the spirit is willing, is a major corporate event where time is not available. Tough stuff.
Schedules of contribution were, I believe, the softest option (but only assuming the Sponsor had the liquidity to provide an accelerated schedule and anything but the next payment on the schedule would need to be subject to increasing scrutiny and start to look, from a sponsor risk perspective, more like a loan rather than an inevitable payment).
Meanwhile, your fund manager may have decided to de-lever you (ie knock you out). And you may not know it yet. More layers lead to worse communications. Knocking you out is, if you had not already been told when you bought the product, the last line of defence against a liquidity driven collateral call. And that decision may not be taken with your knowledge.
- Inexperienced advisors might not have built into the schemes’ risk framework actions to respond to an eventuality of the MBC. Liquidity cascades are a standard feature of any ALM risk framework. Too often, high standard deviation scenarios (i.e. events unlikely to happen) are dismissed on the basis that “they will never happen”. What the response should be is “what are we going to do if it does?” This area, in my opinion, is where the system will have been seen to have most significantly failed (I could be wrong…..): the system did not plan sufficiently for such an eventuality or even recognise upon what circumstances this eventuality would arise and explore what the options were. As ever, those with plans did better than those without.
- And finally, if at the end of this, the trustee board is lay or has little external risk management or even “other scheme” experience, the board is vulnerable to operating outside its comfort zone. A trustee board is not a governance organ that is designed for quick responses unless it has already thought through such eventualities thoroughly. The trust board system “does not do financial crises” and the lay board is significantly more exposed than a board with the expertise included.
My main takeaway is that, if you are managing a sophisticated ALM balance sheet (as all defined benefit pension schemes are), you need to wear a proper risk management hat. And, as a trustee board, at least one member of that board should be so attired. You can of course have someone else to wear it, higher up the food chain, but the higher up it goes, it may well end up on (or inside) a leopard!
We begin to look forward in our next blog from John Wilson who asks – What next?