3 ms·

Yes, they wiped out these CoCo (Contingent Convertable) bonds, which are floated to provide money to provide a "capital cushion" in the case of stress. As bonds

by kurthr 4y ago

Yes, they wiped out these CoCo (Contingent Convertable) bonds, which are floated to provide money to provide a "capital cushion" in the case of stress. As bonds they're definitely weird, because (in this case) they are effectively subordinate to equity. This wipeout may make it difficult to issue any more AT1s, which removes another barrier created in 2008 that was supposed to prevent public bailouts. I am a bit surprised they didn't just convert an payout at the fairly cheap stock price, which it sounds like other regulators are saying they will do? https://www.investing.com/news/stock-market-news/explainerwhy-markets-are-in-uproar-over-a-risky-bank-bond-known-as-at1-3034847 https://www.investing.com/news/stock-market-news/explainerwh...

thechao 4y ago
The article actually discusses AT1's at length, and pre-addresses all of your arguments. I'm not sure I find Levine's argument 100% compelling — I mean, we're dealing with a sentiment based sale, not calculating engines — but I don't think he's as close to wrong as you seem to imply.
- kurthr 4y ago
  I don't have a strong opinion, and simply posted one of the many articles discussing it today. It's definitely water cooler talk, and I wouldn't say it's discussion is really 'at length' or base any significant legal or monetary decisions on it. edit: If you have Bloomberg Intelligence there's a nice graphic on Additional Tier 1 bonds with Triggers, Loss Absorption, and and Write Up by Issurer. It was known that both UBS and CS were non-recoverable.
seanhunter 4y ago
The thing that's interesting about cocos is that research has established that cocos don't really do what they were originally designed for and instead incentivise risk-taking[1] by banks and especially so if the conversion terms protect existing shareholders too much from dilution. [1] https://www.bankofengland.co.uk/-/media/boe/files/working-paper/2021/risk-taking-and-uncertainty-do-contingent-convertible-bonds-increase-the-risk-appetite-of-banks.pdf?la=en&hash=FBBFE9D7F4D99329D9EB48D59D9D3019E3D5B4DC https://www.bankofengland.co.uk/-/media/boe/files/working-pa...
- chii 4y ago
  So why does anyone buy these "bonds"? it seems like their pay out isn't high enough to really be worth the potential risk. Or is it a matter of investors not reading the fine print (which i find to be not likely)?
  - seanhunter 4y ago
    Although you may not find it likely, that is sort of the case. Bond prospectuses are very long (100s of pages sometimes) and complicated with tons and tons of boilerplate language so most people focus on exactly 1 page of that (the "waterfall" describing payout and loss priorities).[1] Things like under what circumstances an instrument will convert are sometimes very complicated and people like Elliot Capital have basically made a career out of actually doing the homework and figuring this out better than everyone else. You can bet all sides have teams of securities lawyers pouring over the conditions of conversion of every single coco bond out there and these most of all. Part of the calculus for AT1/Coco bonds seems to have been that investors thought the bonds would basically never convert (and were certainly not pricing in the likelihood of being wiped out in this way) so they were getting paid for nothing. That is why the coco bond market has been so spooked by this. Some investors have been saying the cbanks have made a mistake because noone will ever buy AT1 bonds again, but cbanks seem to have gone off the whole AT1 idea and would prefer just real normal tier 1 capital instead so probably wouldn't mind that outcome too much. [1] Real life example: I was involved in the structuring of a bond and after a team of us working on it for 6 months and having poured over the prospectus through multiple drafts, after it had been in investors hands for 6 months we disagreed with lawyers about a very basic point (whether there were physical bond certificates or not) and ended up being wrong (there were - stuck in a custodian vault somewhere).
    - chii 4y ago
      > investors thought the bonds would basically never convert .. so they were getting paid for nothing sounds like the calculus for the CDS (credit default swaps) for AAA mortgage bonds...before the GFC!

by kurthr 4y ago

thechao 4y ago

The article actually discusses AT1's at length, and pre-addresses all of your arguments. I'm not sure I find Levine's argument 100% compelling — I mean, we're dealing with a sentiment based sale, not calculating engines — but I don't think he's as close to wrong as you seem to imply.

kurthr 4y ago
I don't have a strong opinion, and simply posted one of the many articles discussing it today. It's definitely water cooler talk, and I wouldn't say it's discussion is really 'at length' or base any significant legal or monetary decisions on it. edit: If you have Bloomberg Intelligence there's a nice graphic on Additional Tier 1 bonds with Triggers, Loss Absorption, and and Write Up by Issurer. It was known that both UBS and CS were non-recoverable.

seanhunter 4y ago

The thing that's interesting about cocos is that research has established that cocos don't really do what they were originally designed for and instead incentivise risk-taking[1] by banks and especially so if the conversion terms protect existing shareholders too much from dilution. [1] https://www.bankofengland.co.uk/-/media/boe/files/working-paper/2021/risk-taking-and-uncertainty-do-contingent-convertible-bonds-increase-the-risk-appetite-of-banks.pdf?la=en&hash=FBBFE9D7F4D99329D9EB48D59D9D3019E3D5B4DC https://www.bankofengland.co.uk/-/media/boe/files/working-pa...

chii 4y ago
So why does anyone buy these "bonds"? it seems like their pay out isn't high enough to really be worth the potential risk. Or is it a matter of investors not reading the fine print (which i find to be not likely)?
- seanhunter 4y ago
  Although you may not find it likely, that is sort of the case. Bond prospectuses are very long (100s of pages sometimes) and complicated with tons and tons of boilerplate language so most people focus on exactly 1 page of that (the "waterfall" describing payout and loss priorities).[1] Things like under what circumstances an instrument will convert are sometimes very complicated and people like Elliot Capital have basically made a career out of actually doing the homework and figuring this out better than everyone else. You can bet all sides have teams of securities lawyers pouring over the conditions of conversion of every single coco bond out there and these most of all. Part of the calculus for AT1/Coco bonds seems to have been that investors thought the bonds would basically never convert (and were certainly not pricing in the likelihood of being wiped out in this way) so they were getting paid for nothing. That is why the coco bond market has been so spooked by this. Some investors have been saying the cbanks have made a mistake because noone will ever buy AT1 bonds again, but cbanks seem to have gone off the whole AT1 idea and would prefer just real normal tier 1 capital instead so probably wouldn't mind that outcome too much. [1] Real life example: I was involved in the structuring of a bond and after a team of us working on it for 6 months and having poured over the prospectus through multiple drafts, after it had been in investors hands for 6 months we disagreed with lawyers about a very basic point (whether there were physical bond certificates or not) and ended up being wrong (there were - stuck in a custodian vault somewhere).
  - chii 4y ago
    > investors thought the bonds would basically never convert .. so they were getting paid for nothing sounds like the calculus for the CDS (credit default swaps) for AAA mortgage bonds...before the GFC!