The ghost of Archegos returns to haunt Simm

A recent letter from the Bank of England’s Prudential Regulation Authority (PRA) setting out the conclusions of its review of the standard initial margin model (Simm) reminded some in the industry of an old game from Sesame Street.

The UK regulator cites three events at the top of its review of Simm for non-cleared derivatives: the Covid-19 pandemic; the Russia-Ukraine crisis; and the default of Archegos Capital Management.

As market participants are quick to point out, one of these things is not like the others.

“Why are they citing Archegos?” asks one industry source. “Maybe there is something there, but most of us don’t understand it.”

Archegos wasn’t margined using Simm, which was developed by the International Swaps and Derivatives Association in response to the non-cleared margin rules. The now-defunct family office wasn’t even subject to those rules at the time of its default.

Some bat off the regulator’s mention of Archegos as a red herring. The PRA itself makes clear its conclusions about Simm – that more work needs to be done to account for non-modelled risk factors – were based on an analysis of the model’s performance during the Covid-19 market stress.

Others, though, believe the family office’s dramatic failure – and the $10 billion losses it inflicted on global banks – lit the touchpaper for the latest round of sabre-rattling on margin.

After all, most of Archegos’s equity swap positions, which totalled $120 billion notional at its peak, were booked into the UK subsidiaries of its dealers. If there are more Archegoses in the market, they will be swept up in the sixth and final implementation phase of the initial margin regime, beginning on September 1.

Few – if any – models would have adequately margined a client that went to such lengths to hide its exposures

Dealers say the regulator has stepped up its scrutiny of how margin models handle hedge fund exposures in recent months. “They’ve basically been asking: if Archegos happened under Simm, would it have been covered?” says a quant analyst at a European bank.

The answer he gives is an emphatic “no”. But that’s no reason to target Simm, he adds. Few – if any – models would have adequately margined a client that went to such lengths to hide its exposures. The PRA’s recommendations to improve Simm governance – enhanced backtesting with actual profit-and-loss data – would have made little difference with Archegos.

More worryingly, the regulator’s own standard margin schedule for non-cleared derivatives – billed as the most conservative of all approaches – may have fared even worse. The so-called regulatory grid assigns a 15% margin charge to equity exposures. Risk weights under Simm for large-cap developed market tech stocks are calibrated at 27%. Dealers say that may need to be increased to at least 40% to cover the Archegos losses.

PRA’s pot shots threaten Simm’s ubiquity

It is possible that the upshot of all this is that regulators universally agree to increase margin charges under grid. Such a move may even be welcomed by some dealers, who privately concede that the standard schedule is insufficient for some hedge fund counterparties.

Others believe the family office’s blow-up should be kept out of the margin debate altogether.

“Archegos was not to do with initial margin models,” says the quant analyst at the European bank. “Whenever banks experience a big loss, regulators have a tendency to assume it must be because they didn’t collect enough margin. But with Archegos, the real question is why banks allowed a client to place such large, risky bets in the first place.”

The answer is unlikely to be found in an analysis of margin models.