[FabHK]

Matt Levine in his excellent Money Stuff newsletter had a neat summary of what happened there [0].

Basically:

1. A pension fund has to pay a $100 pension in 30 years (=liability) [1]. So, it buys a 30 year bond today (=asset) that pays $100 in 30 years. The asset and the liability are matched, and it is flat rates: long one bond (which it bought), and "short one bond", namely the pension it has to pay. +1 -1 = 0

2. However, rates have been quite low over the last two decades, making bonds expensive. So, the pension fond decides to try something new: mix bonds with stocks. They have higher expected returns, so it can get away with buying a bit less upfront, say half a bond and some shares. There will be a shortfall, but if the shares outperform the bond, as expected, it will sort itself out in 30 years.

3. Now the pension fund is long half a bond (which it bought), and "short one bond", namely the pension it has to pay, so in total not flat, but short half a bond: +0.5 -1 = -0.5. This means that the fund is unhappy when bonds rise in price, ie yields fall.

4. But even though rates were low two decades ago, they have fallen even lower after the GFC etc. [2] So, funds were unhappy, as they appeared to get worse shortfalls. So, they got the great idea of hedging, ie synthetically going long bonds: happy when bond prices go up, yields fall.

5. But recently, yields have shot up, bond prices fell. That's good, per se, for the pension funds - they're naturally short bonds, after all. Except that now they've hedged - and their hedges went massively against them.

6. Bonds are so cheap now, with high rates, the pension funds should be buying them. But instead they might be forced to sell to cover their margin calls, driving bond prices down even further, and ... we're in the spiral you mentioned.

[0] https://www.bloomberg.com/opinion/articles/2022-09-29/uk-pen...

[1] They actually have to pay a pension more or less every year over the next N years, but one can simplify this particular story by just picking one year.

[2] see e.g. https://fred.stlouisfed.org/series/DGS10 (click on "MAX" to see the full history)

[benj111]

I understood it (after reading Matt Levine) that it was more an accounting requirement.

Ie the books need to balance today, even though they're always going to balance in 30 years.

You seem to be suggesting that they took an unnecessary risk.

Is this my misunderstanding or yours or is it a combination?

[FabHK]

The books don't need to balance now: there can be a shortfall. That, however, is more of an accounting artefact (the books are not balanced now, but it is going to be ok in 30 years if the stocks outperform as expected).

The problem is that as rates fall, the shortfall increases. That's still just an accounting artefact, but it looks bad. To mitigate the bad look, the funds entered real hedges, which are now (with rates moving the opposite way) blowing up.

So, my understanding is that the funds took a real and unnecessary risk to mitigate what was basically just "bad optics".

[uup]

But did they really enter into these hedges after the GFC? It seems like rates were already near-zero so it wouldn’t have made much sense to hedge against lowering rates.

[benj111]

Id read the Matt Levine article https://www.bloomberg.com/opinion/articles/2022-09-29/uk-pen...

The hedge is because they moved to stocks because of low yields on bonds.