This is a very interesting & important question so let me elaborate a bit. Quick thread. https://twitter.com/EuroYield/status/1336907537787068416

When Basel 2 was introduced with internal ratings based on PD, it faced criticism of procyclicality. The PD & RWA methodology mitigated a bit, but the main problem remained: when the economy goes south, PDs go up, capital goes down, and banks can’t lend.

Fast forward & the GFC happened. By far, procyclicality was NOT the biggest problem (VaR on negative basis trades was) but it still was one. So, as part of the big reform package, the concept of capital buffers was introduced.

The idea is that there is some capital that you can use during bad times – i.e. your capital ratios can go down in recessions. Country-wide it’s the countercyclical buffer, on a bank by bank basis, it’s the capital conservation buffer.

Sounds like a great plan, but the Eurozone crisis messed it up. Why? Because everyone realized that capital doesn’t tell all the story. The EZ crisis triggered bank failures (like our good friend Dexia) with capital at 17% CET1 ! So what happened next is two things:

European supervisors kept asking for more capital (ratios have stabilized a long time ago in the US, but NOT in the EU, they keep going up) and they did not hesitate to introduce BRRD & make banks fail even if they have high capital.

The result is a total lack of trust on both sides: when supervisors say that they will allow banks to operate with lower capital, banks don’t believe them. They think Popular, they think dividend limitation, etc. The recent ban didn’t help either.

So banks just don’t want to reduce their capital ratios, simply because everything that supervisors have done over the last ten years tells them it’s a BAD idea - if they do it, they'll struggle to get them up again

(if you use 8% of CET1 on a loan on which the margin is 1%, it takes 8 years to get your capital back... ratios don't go up that fast!)

But guess what: supervisors don’t trust the banks either! They don’t really believe the IFRS models, they don’t really believe the AIRB models, they don’t really believe the capital planning, etc.

And you can't really blame them! There have been so many f***-ups, from total fraud at BES to nonsense and endless provisioning at MPS.... and hundreds of horror stories

But at the end of the day the result is simple: the buffer framework doesn’t work.

Sure, it can be tweaked, improved. Replacing a big chunk of the CCB by a big chunk of CCyB would help. Making the MDA rule non automatic but subject to supervisory discussion would help a LOT.

But ultimately, this is about trust: banks need to trust their supervisors when they say that a rule will be consistently applied. Supervisors need to trust the banks that their capital planning (in the broad sense) are consistent.

And this will take time - the time required to heal from two massive crisis (GFC + Eurozone) and hoping we don't get a third with Covid.