macrocephalopod

yes, but I still think the simplified standard approach is a useful starting point to get an idea of the transaction-level costs.

This is one area where the modelled capital requirement is quite a lot lower than the standardized, because if you aren't capable of getting a model approved, you arguably shouldn't be doing this kind of business at all in any size. CFDs just treated as exposure to the underlying

so basically the capital requirement is whatever the notional is, and if you hedge via buying the stock, it's zero? no counterparty surcharge?

I have tried to avoid counterparty credit risk my whole life because it is hellish complicated but I suspect there should be some capital requirement for it here, basically because what is happening today is an utterly predictable risk.

especially funny because the lads at GS and DB appear to have made out without any material losses versus 2 yards at Nomura and something material at CS/UBS.

this might be only tangentially related to the question but fwiw this was helpful in understanding the mechanics, which lead me to believe the only thing out of the norm was the amount of leverage Hwang used, + insufficient/wrong hedgehttps://twitter.com/macrocephalopod/status/1376431761169788933?s=20 …

Thanks. The wrong hedge would - hopefully - have shown up in their own risk measures. But wouldn’t for the PBs: they’re perfectly hedged on each side.

which is why it's a little confounding that $NMR is on the hook for up to $2B and $CS for something material as well