Tier 1 capital: Difference between revisions

From The Jolly Contrarian
Jump to navigation Jump to search
No edit summary
No edit summary
Line 3: Line 3:
Of a regulated [[financial institution]], the capital level below everything else that gives comfort to the bank’s creditors — in particular, its depositors — that their debts will be met and deposit withdrawals honoured.   
Of a regulated [[financial institution]], the capital level below everything else that gives comfort to the bank’s creditors — in particular, its depositors — that their debts will be met and deposit withdrawals honoured.   


If you are a regulated financial institution (a bank) — but ''only'' if you are one of those— you must “hold” a certain percentage of tier 1 capital, though a certain type of financial analysts get annoyed if you say “hold”, for the pedantic reason that tier 1 capital is a really just what is left of your assets after you deduct your liabilities, and isn’t something you “hold”, as such.
If you are a regulated financial institution (a bank) — but ''only'' if you are one of those — you must “hold” a certain percentage of tier 1 capital.


Less pedantic types feel that since you have to monitor it every day, and do something, like issuing more tier 1 capital securities, if it isn’t there, this isn’t really a distinction worth getting het up about.
=== Pedantry alert ===
There is a certain type of financial analyst who get annoyed if you say banks “hold” capital, for the pedantic reason that capital is a really just what is left of your assets after you deduct your liabilities, and isn’t something you “hold”, as such. It is a difference between two other things, rather than a thing in itself.


What are tier 1 capital securities, then?
Less pedantic types feel that since you have to monitor that difference every day, and do something, like issuing more tier 1 capital securities, if it isn’t there, this isn’t really a distinction worth getting het up about.
 
Put actually that pedantry can be important, as we will see. Now: what are “tier 1 capital securities,then?


==Tier 1 common equity==
==Tier 1 common equity==
The classic tier 1 capital is the institution’s ordinary share capital. This is known, by the same people who know coronavirus as “COVID-19”, as “[[tier 1 common equity]]”, or “[[CET1]]”.   
The classic tier 1 capital is the institution’s ordinary share capital. This is known, by the same people who know coronavirus as “[[COVID-19]]”, as “[[tier 1 common equity]]”, or “[[CET1]]”.   


Until 2008, that is all there really was  Then the [[global financial crisis]] happened, and the world’s various bank regulator committees, councils and forums got together, promulgated largely coordinated set of [[bank resolution and recovery regime]]s, in the process savagely increasing tier one capital requirements with which banks had to comply.
Until 2008, that is all there really was  Then the [[global financial crisis]] happened, and the world’s various bank regulator committees, councils and forums got together, promulgated a largely coordinated set of [[bank resolution and recovery regime]]s, in the process savagely increasing tier one capital requirements with which banks had to comply.


==[[Alternative tier 1 capital]]==
==[[Alternative tier 1 capital]]==
When banks complained — equity capital is quite the drag on performance — the committees conceded there could be a layer of tuer 1 which wasn’t ''actually'' common equity, but could be made to ''behave'' like it if a bank’s chips ever got really down.
When banks complained about this — equity capital is quite the drag on performance — the committees conceded there could be a layer of tier 1 which wasn’t ''actually'' common equity, but could be made to ''behave'' like it, should a bank’s chips ever got really down.
 
On the good days, this layer could behave a lot like debt: fixed [[Coupon|coupons]], redeems — ''if'' it redeems — at par. Most likely, everyone thought the prospect of lots of banks’ chips getting really down again, all at once, was pretty low, so this was a largely academic issue — but it is March 2023 and here we all are.
 
Anyway, this new layer of “quasi common equity” came to be known as [[alternative tier 1 capital|“alternative” tier 1 capital]], or “[[AT1]]” which, when said aloud, sounds like “[[eighty-one]]”.  


We suspect everyone thought that a large number of banks’ chips would not simultaneously get down again, so this was a largely academic issue — but it is March 2023 and here we all are. Again.
AT1 capital takes the form of [[Interest|interest-bearing]] [[subordinated]] debt which the bank may, but need not, call after a few years, As such, from an investor’s perspective, it is ''perpetual'' in the same way ordinary shares are. But banks can decide to repay it, at par, if they like. This they will only do if times are good and it is cheap to issue more AT1 capital.  


Anyway, this new layer of quasi common equity came to be known as [[alternative tier 1 capital|“alternative” tier 1 capita]], or “[[AT1]]” which, when said out loud, sounds like “[[eighty-one]]”.  
Why does it count as tier 1 capital then? Because it contains an embedded, “contingent,” bomb.


AT1 capital takes the form of [[subordinated]] debt which the issuer may, but need not, call after a few years, As such, from an investor’s perspective, it is as theoretically ''perpetual'' as ordinary shares are.  
In certain disaster scenarios it is ''convertible'' into ordinary shares, at which point it ''becomes'' [[CET1]], or may even be written off altogether. To zero. A [[Donut|duck]]. Bupkis.


In certain disaster scenarios it is also ''convertible'' into ordinary shares, at which point it ''becomes'' [[CET1]], or may even be written off altogether. Conversions and write-downs are  “contingent” on defined events, like capital thresholds being breached — ''[[der Teufel mag im Detail stecken]]'' to the max — so AT1s are also called “[[contingent convertible securities]]” or  “[[co-cos]]”.  
Conversions and write-downs are  “contingent” on defined events — ''[[der Teufel mag im Detail stecken]]'' to the max — things like capital thresholds being breached, or regulators concluding the bank is not viable otherwise — so AT1s are also called “[[contingent convertible securities]]” or  “[[co-cos]]”.  


It became clear in March 2023 when [[Credit Suisse]] finally gave up the ghost, that many in the market, including AT1 investors, didn’t fabulously understand how they worked.  
It became clear in March 2023 when [[Credit Suisse]] finally gave up the ghost, that many in the market, including AT1 investors, didn’t fabulously understand how they worked.  


===[[Debit Suisse]] and the irate noteholders: co-co go loco===
==[[Debit Suisse]] and the irate noteholders: co-co go loco==
Famously, in that panicked spring weekend in 2023 when it slipped into history<ref>We have a sense [[Credit Suisse]]’s history is not done just yet but that, like Disaster Area frontman Hotblack Desiato, it is merely spending a year dead for tax (and, er regulatory capital) purposes. It may well be back, at least as a high-street banking brand in Switzerland.</ref> the “trinity” of Swiss regulators put a gun to UBS’s head, forced it to absorb [[Lucky]]’s equity (and all the baubles, jewels and hellish instruments of madness and torture embedded in it) — and, by ordinance, directing [[Credit Suisse|Lucky]] to write down its AT1s — called “Perpetual Tier 1 Contingent Write-Down Capital Notes,” and names are sort of important here — to zero.
Famously, in that panicked spring weekend in 2023 when it slipped into history<ref>We have a sense [[Credit Suisse]]’s history is not done just yet but that, like Disaster Area frontman Hotblack Desiato, it is merely spending a year dead for tax (and, er regulatory capital) purposes. It may well be back, at least as a high-street banking brand in Switzerland.</ref> the “trinity” of Swiss regulators put a gun to UBS’s head, forced it to absorb [[Lucky]]’s equity (and all the baubles, jewels and hellish instruments of madness and torture embedded in it) — and, by ordinance, directed [[Credit Suisse|Lucky]] to write down its AT1s — which were “Perpetual Tier 1 Contingent ''Write-Down'' Capital Notes,” and names are important here — to zero.


This — and there isn’t really a delicate way to put this, readers, so let’s just come out with it — ''pissed the AT1 noteholders the hell off''.
This — and there isn’t really a delicate way to put this, readers, so let’s just come out with it — ''pissed the AT1 noteholders the hell off''.
Line 36: Line 43:
{{Quote|If a Contingency Event, or prior to a Statutory Loss Absorption Date, a Viability Event occurs, the full principal amount of the Notes will be mandatorily and permanently written down. '''''The Notes are not convertible into shares of the Issuer''''' upon the occurrence of a Contingency Event or a Viability Event or at the option of the Holders at any time. <ref>Emphasis added. Full documents [https://www.credit-suisse.com/about-us/en/investor-relations/financial-regulatory-disclosures/regulatory-disclosures/capital-instruments.html here].</ref>}}
{{Quote|If a Contingency Event, or prior to a Statutory Loss Absorption Date, a Viability Event occurs, the full principal amount of the Notes will be mandatorily and permanently written down. '''''The Notes are not convertible into shares of the Issuer''''' upon the occurrence of a Contingency Event or a Viability Event or at the option of the Holders at any time. <ref>Emphasis added. Full documents [https://www.credit-suisse.com/about-us/en/investor-relations/financial-regulatory-disclosures/regulatory-disclosures/capital-instruments.html here].</ref>}}


They were fortified in their dudgeon by other central bankers (BOE, ECB, the Fed) unhelpfully announcing, for the record, that that is not how ''they'' would expect to treat [[AT1]]<nowiki/>s (you can just imagine FINMA honchos going “yeah, thanks Pal,” when a central banker from ''Greece'' — yes, yes, ''that'' Greece — went on record as saying “well needless to say we’d never do anything like that. We Greeks are civilised, not like the Swiss!”<ref>This is a paraphrase, and an exaggeration for effect, I freely admit.</ref>) and now ambulance chasing [[Litigation lawyer|litigator]]<nowiki/>s are whipping up even more foment, indelicately trawling [[LinkedIn]] to raise a pitchfork mob of aggrieved investors to go and sue — well, it isn’t clear ''who'' they would sue, or for what, since this was done by legislation — and even the normally mild-mannered financial analyst commentariat has been periodically erupting into virtual fist-fights about what the AT1s do or do not say.
But, docs schmocks.
 
Wounded AT1 holders were fortified in their dudgeon by other central bankers (BOE, ECB, the Fed) unhelpfully chipping in, saying, for the record, that that is not how ''they'' would expect to treat [[AT1]]<nowiki/>s. You can just imagine FINMA honchos going, “yeah, thanks, Pal,” when a central banker from ''Greece'' — yes, yes, ''that'' Greece — remarked “well, needless to say we’d never do anything like that. We Greeks are civilised, not like the Swiss!”<ref>This is a paraphrase, and an exaggeration for effect, I freely admit. [[https://www.cnbc.com/video/2023/03/20/awe-are-close-to-the-end-of-the-tightening-cyclea-bank-of-greece-governor-says.html?&qsearchterm=stournaras Full interview here.]</ref>  
 
Now ambulance chasing [[Litigation lawyer|litigator]]<nowiki/>s are whipping up even more foment, indelicately trawling [[LinkedIn]] to raise a pitchfork mob of aggrieved investors to go and sue — well, it isn’t clear ''who'' they would sue, or for what, since this was done by legislation — and even the normally mild-mannered financial analyst commentariat has been periodically erupting into virtual fist-fights about what the [[AT1]]<nowiki/>s do or do not say and whether a trigger event did or did not happen.
 
Meanwhile, from the investors, there is lots of jilted lover energy: “how could I ever trust a central banker again?” sort of thing, and lots of “who knew Switzerland was a banana republic?” vibes, too.
 
Now the JC ''likes'' Switzerland, so he is staying right out of that debate: There are plenty of thought pieces from those more learned and temperate than the JC about that. But Switzerland is ''not'' a banana republic, and it is known for its banking acumen. This, we think, will be borne out over time.  


Meanwhile, from the investors, lots of jilted lover energy: “How could I ever trust a central banker again?” sort of thing, and lots of “who knew Switzerland was a banana republic?” vibes, too.
=== On creditors ranking behind equity-holders, feelings and so on. ===
But the conceptual question this all throws up, in the abstract, is an interesting one: should creditors, however subordinated, ever rank ''behind'' common shareholders?


Now the JC ''likes'' Switzerland, so he is staying right out of that debate: There are plenty of thought pieces from those more learned and temperate than the JC about that.
Surely not?


=== But still ===
Everyone knows AT1s can get converted into equity, at which point they rank equally ''with'' shareholders, and even written off — but there seemed to be the expectation that a write-off would only happen if common shareholders are getting written off too. 
But the conceptual question this all throws up, in the abstract, is an interesting one: should creditors, however subordinated, ever rank ''behind'' common shareholders? Surely not?


Everyone knew AT1s could get converted into equity, at which point they rank equally ''with'' shareholders, and even written off — but there seemed to be the expectation that a write-off would only happen if common shareholders are getting written off too.  
First, a little spoiler: ''effectively'' ranking behind shareholders and ''actually'' ranking behind shareholders may ''feel'' similar — especially if you have just been written down to zero while the shareholders live to see another day — but they are different things. When an is written down AT1 down to zero, its creditors ''actually'' rank ''ahead'' of shareholders. It is just that their claim is zero.


First, a little spoiler: ''effectively'' ranking behind shareholders and ''actually'' ranking behind shareholders feel similar — especially if you have just been written down to zero while the shareholders live to see another day — but they are quite different things.
Another spoiler: this cannot have come as a surprise. Issuers ''must'' have contemplated writing AT1s down while shareholders survived: otherwise, why even ''have'' write-down Notes? A write-down contingent on total shareholder annihilation is no different from a normal conversion to equity: you get what the shareholders get: zero. If that is all you wanted, you would just issue normal contingent convertible bonds.  ''But these AT1s were not convertible''. There were “Perpetual Tier-1 Contingent ''Write-Down'' Capital Notes”. Again, that name. Important.


Two spoilers, in fact: issuers ''must'' have contemplated writing AT1s down while shareholders survived: otherwise, why even ''have'' a write-down option? A write-down contingent on total shareholder annihilation is no different from a normal conversion to equity: you get what the shareholders get: zero. That kind of write-down option would be meaningless.  
The whole point of writing down AT1s is to deliver a capital buffer and stave off an insolvency ''so the bank can carry on''.


The whole point of a write down to zero is to deliver a capital buffer and stave off an insolvency ''so the corporation can carry on''. If it succeeds, the shareholders will live to see another day.
=== Pedantry redux ===
This is where that pedantry we mentioned at the top is important: “capital” is not a thing: it is a ''difference between things''. If the AT1s are vaporised, it follows that the tier 1 capital — now comprising only common equity — is worth 17bn more. If the Write-Down succeeds, the shareholders will live to see another day.


So the JC thinks those central banks who are on record as saying “we’d never write off AT1s before shareholders” are flat out ''wrong''.
So the JC thinks those central banks who are on record as saying “we’d never write off AT1s before shareholders” are flat out ''wrong''.
Line 61: Line 77:
Therefore, the AT1 investors do not ''actually'' rank behind shareholders. They ''can’t''. They either ''become'' shareholders, or they are ''goneski''. If they get converted into shares they ''may'' get some recovery, but only once all the company’s other creditors have been repaid in full. A written down AT1 ''has'' been paid in full. The liability was just zero.  
Therefore, the AT1 investors do not ''actually'' rank behind shareholders. They ''can’t''. They either ''become'' shareholders, or they are ''goneski''. If they get converted into shares they ''may'' get some recovery, but only once all the company’s other creditors have been repaid in full. A written down AT1 ''has'' been paid in full. The liability was just zero.  


But AT1 investors whose notes are written off  still feel as if they are ''effectively'' ranking behind shareholders. This is because they get nothing and shareholders get something.  
=== Did the AT1s ''really'' do worse than common equity? ===
But AT1 investors whose notes are written off  still feel as if they are ''effectively'' ranking behind shareholders. This is their [[lived experience]]: they get nothing and shareholders get something.  


But is that really true?  
But is that really true?  


The panel above illustrates our best guess of the cumulative shareholder return — the diminishing cash dividends paid plus ongoing share price, which was in of course in insistent decline over 5 years — and cumulative AT1 return, which is a a much fatter, fixed coupon plus its market redemption value , which we have just made up, but on the premise that until things get truly dire, it will stay somewhere near par. It was trading at 25% on the last trading day before it was vapourised.
The panel above illustrates our best guess of the cumulative shareholder return — the diminishing cash dividends paid plus ongoing share price, which was in of course in insistent decline over 5 years — compared with the cumulative AT1 return, which are those much fatter, fixed coupons plus the AT1s’ market value, which we just made up, on the premise that until things got truly dire, it would have been somewhere near par. they were vaporised.


The obvious thing about this is that, for a buy and hold, long term investor, it's return has been better than the common equity, ''including after it was nixed''. The combined coupons since issue are easily more than the final acquisition price. In all circumstances ''except a thermonuclear meltdown'' the Cocos were a vastly better investment.
The thing to notice here is that, for a buy-and-hold, long-term investor, the AT1s’ return, ''event after they were nixed'', was ''miles'' better than the common equity. The accumulated coupons since issue totalled about a third of its issue price. That is far more than the final acquisition price for the common equity.  


“Ah yes, you counter, but tell that the the distressed investors who bought the AT1s on Saturday.
“Ah yes, you counter, but just try telling that the the distressed investors who bought their AT1s last Saturday.


And we should not feel undue sympathy for distressed — ahem vultures — looking to buy a 7% fixed instrument for cents on the dollar when the issuer is in the midst of a well telegraphed existential meltdown? ''We should not''. Even if the ones who ''did'' read the prospectus.
But should we feel sympathy for distressed — ahem, ''vultures'' who buy a 7.25% capital instrument for 20 cents on the dollar during a well-telegraphed existential meltdown?


One invests in common equity to take advantage of rapidly changing market conditions. An equity price is a capricious, will o’ the wisp sort of thing that flits about impishly, by driven by the unpredictable humours of the market. One can, and many do do make a living trading short-term movements.
''We should not''. Even those who ''did'' read the prospectus.  


In ordinary times, interest-bearing notes — when AT1s — are much less volatile  than common equity. Their market value will not fluctuate much day by day. Their main attraction is their coupon yield. You only benefit from that over time. AT1s reward long-term investment. In ordinary times their return is a linear function of ''how long '' you are prepared to hold them, and therefore how long you fund the bank’s tier 1 capital.
=== On the difference between equity and debt investment ===
One invests in shares to take advantage of rapidly changing market conditions. A share’s price is a capricious, will o’ the wisp sort of thing: it flits about, impishly, by driven by the unpredictable humours of the market, whether they are fundamental, structural, geopolitical, or just the product of the delusional madness of crowds.  


It is different in a bank distress scenario. Here AT1s are unusually vulnerable: this is the very contingency they are designed to protect ''the bank''  against. As the banks capital ratio approaches criticality, their performance more and more to resembles the equity.  
One can, and many do, make a living trading short-term movements in shares.


''Convertible'' AT1s which, in the worst case, will turn into common equity, will converge on the common equity exactly.  
In ordinary times, debt instruments — even AT1s — are much less volatile. They do not hop about much, day-to-day. They are sensitive to interest rate environment, and ultimate solvency of their issuer, but their value does not jump around. Their main attraction is their scheduled interest income. Investors only benefit from that over time. AT1s reward long-term investment. In ordinary times, their return is a linear function of ''how long ''you have held them, and therefore how for long you provide the bank with tier 1 capital. Day-trading undistressed bonds is not much of a laugh.


''Write-Down'' AT1s will become even ''more'' volatile than common equity.  They are, effectively, binary options: either they are triggered, in which case they are worth zero, or they are not, in which case they recover, will eventually be called at 100, and in the meantime will continue to pay fat slugs of interest.
It is different in a distressed scenario. Here AT1s are unusually vulnerable: this is the very contingency they are designed to protect ''the bank'' — not the investor; the bank — against. As the bank’s capital ratio approaches the trigger threshold, their performance more and more to resembles the equity: ''Convertible'' AT1s, in their worst case, ''become'' common equity. Their value will converge on the common equity exactly. ''Write-Down'' AT1s will become even ''more'' volatile than common equity.  They are, effectively, binary options: either they are triggered, in which case they are worth zero, or they are not, in which case they recover, will eventually be called at 100 and, in the meantime, will continue to pay fat slugs of interest.  


Indeed, this is exactly what we saw.  
Indeed, this is exactly what we saw.  


Should we feel bad that speculators looking for a quick buck, who held the notes for a couple of days, got hosed? No. This is exactly the bet they were taking.
So, should we feel bad for opportunistic speculators, who picked up the AT1s for a song, but called the market wrong and got hosed? ''No''. This is exactly the bet they were taking.


Should we feel bad for buy and hold investors who held from issue and were written down to zero? No. They did much better than common equity holders over that period.
Should we feel bad for loyal investors who bought at issue, held till the death and were written down to zero? Again, ''no''. They did much better than common equity holders over the life of their investment. It is sad that they got written down, but that is exactly the option they sold when they bought the Notes.  


The tier one capital layer is there to protect depositors and vouchsafe the stability of the wider financial system, whose collected interests are best served by the bank remaining a going concern. That they happen to share that interest with the banks ordinary shareholders is beside the point. The bonds reward long-term investors — those who read the terms and clocked that “Perpetual Tier 1 Contingent Write-Down Capital Notes” meant these were notes that could be written down in a time of capital stress most likely had a bank to sell last week.  
=== Financial stability wins ===
The tier one capital layer is there to protect depositors and ensure the stability of the wider financial system, by helping banks to remain a going concern even in times of great stress. That the bank’s ordinary shareholders happen to share that interest is beside the point. AT1s are meant to reward, and did reward, long-term investors. Those who read understood that “Perpetual Tier 1 Contingent Write-Down Capital Notes” meant their notes that could be written down in a time of capital stress most likely had a bank to sell last week.  


It sounds like there were plenty of buyers.
It sounds like there were plenty of buyers.

Revision as of 10:57, 24 March 2023

Regulatory Capital Anatomy™
The JC’s untutored thoughts on how bank capital works.
CS tier 1 chart.png
Lucky’s CET1 and AT1 compared since issue, yesterday.
From our machine overlords
Here is what, NiGEL, our cheeky little GPT3 chatbot had to say when asked to explain:
Tier 1 capital is the core capital of a bank. It is the most “reliable” form of a bank’s capital — reliability being in the eye of the beholder — and is composed of equity capital, disclosed reserves, and certain non-redeemable subordinated securities called “AT1s”, which can be converted to common equity or written off if the bank’s capital ratio falls through a trigger.

Tier 1 capital is is used to absorb losses without the bank being required to cease operations. When opportunistic AT1 investors find this out, they get mad.

Under Basel III, a bank’s tier 1 and tier 2 assets must be at least 10.5% of its risk-weighted assets, up from 8% under Basel II.


Disclaimer: NiGEL’s a neural network, he drinks a lot, and he spends too much time on the internet, so if you listen to anything he has to say you only have yourself to blame.

Come to think of it, that is also true of the JC in general.

Comments? Questions? Suggestions? Requests? Insults? We’d love to 📧 hear from you.
Sign up for our newsletter.

Premium version of this page

Tier 1 capital
/tɪə wʌn ˈkæpɪtl/ (n.)

Of a regulated financial institution, the capital level below everything else that gives comfort to the bank’s creditors — in particular, its depositors — that their debts will be met and deposit withdrawals honoured.

If you are a regulated financial institution (a bank) — but only if you are one of those — you must “hold” a certain percentage of tier 1 capital.

Pedantry alert

There is a certain type of financial analyst who get annoyed if you say banks “hold” capital, for the pedantic reason that capital is a really just what is left of your assets after you deduct your liabilities, and isn’t something you “hold”, as such. It is a difference between two other things, rather than a thing in itself.

Less pedantic types feel that since you have to monitor that difference every day, and do something, like issuing more tier 1 capital securities, if it isn’t there, this isn’t really a distinction worth getting het up about.

Put actually that pedantry can be important, as we will see. Now: what are “tier 1 capital securities,” then?

Tier 1 common equity

The classic tier 1 capital is the institution’s ordinary share capital. This is known, by the same people who know coronavirus as “COVID-19”, as “tier 1 common equity”, or “CET1”.

Until 2008, that is all there really was Then the global financial crisis happened, and the world’s various bank regulator committees, councils and forums got together, promulgated a largely coordinated set of bank resolution and recovery regimes, in the process savagely increasing tier one capital requirements with which banks had to comply.

Alternative tier 1 capital

When banks complained about this — equity capital is quite the drag on performance — the committees conceded there could be a layer of tier 1 which wasn’t actually common equity, but could be made to behave like it, should a bank’s chips ever got really down.

On the good days, this layer could behave a lot like debt: fixed coupons, redeems — if it redeems — at par. Most likely, everyone thought the prospect of lots of banks’ chips getting really down again, all at once, was pretty low, so this was a largely academic issue — but it is March 2023 and here we all are.

Anyway, this new layer of “quasi common equity” came to be known as “alternative” tier 1 capital, or “AT1” which, when said aloud, sounds like “eighty-one”.

AT1 capital takes the form of interest-bearing subordinated debt which the bank may, but need not, call after a few years, As such, from an investor’s perspective, it is perpetual in the same way ordinary shares are. But banks can decide to repay it, at par, if they like. This they will only do if times are good and it is cheap to issue more AT1 capital.

Why does it count as tier 1 capital then? Because it contains an embedded, “contingent,” bomb.

In certain disaster scenarios it is convertible into ordinary shares, at which point it becomes CET1, or may even be written off altogether. To zero. A duck. Bupkis.

Conversions and write-downs are “contingent” on defined events — der Teufel mag im Detail stecken to the max — things like capital thresholds being breached, or regulators concluding the bank is not viable otherwise — so AT1s are also called “contingent convertible securities” or “co-cos”.

It became clear in March 2023 when Credit Suisse finally gave up the ghost, that many in the market, including AT1 investors, didn’t fabulously understand how they worked.

Debit Suisse and the irate noteholders: co-co go loco

Famously, in that panicked spring weekend in 2023 when it slipped into history[1] the “trinity” of Swiss regulators put a gun to UBS’s head, forced it to absorb Lucky’s equity (and all the baubles, jewels and hellish instruments of madness and torture embedded in it) — and, by ordinance, directed Lucky to write down its AT1s — which were “Perpetual Tier 1 Contingent Write-Down Capital Notes,” and names are important here — to zero.

This — and there isn’t really a delicate way to put this, readers, so let’s just come out with it — pissed the AT1 noteholders the hell off.

Their indignance was largely driven by foundational conceptions of what subordinated debt securities are meant to be — that is, senior to equity — rather than even a cursory glance at the terms or, goddammit, even the title of their Notes. This, from the termsheet, gives a clue:

If a Contingency Event, or prior to a Statutory Loss Absorption Date, a Viability Event occurs, the full principal amount of the Notes will be mandatorily and permanently written down. The Notes are not convertible into shares of the Issuer upon the occurrence of a Contingency Event or a Viability Event or at the option of the Holders at any time. [2]

But, docs schmocks.

Wounded AT1 holders were fortified in their dudgeon by other central bankers (BOE, ECB, the Fed) unhelpfully chipping in, saying, for the record, that that is not how they would expect to treat AT1s. You can just imagine FINMA honchos going, “yeah, thanks, Pal,” when a central banker from Greece — yes, yes, that Greece — remarked “well, needless to say we’d never do anything like that. We Greeks are civilised, not like the Swiss!”[3]

Now ambulance chasing litigators are whipping up even more foment, indelicately trawling LinkedIn to raise a pitchfork mob of aggrieved investors to go and sue — well, it isn’t clear who they would sue, or for what, since this was done by legislation — and even the normally mild-mannered financial analyst commentariat has been periodically erupting into virtual fist-fights about what the AT1s do or do not say and whether a trigger event did or did not happen.

Meanwhile, from the investors, there is lots of jilted lover energy: “how could I ever trust a central banker again?” sort of thing, and lots of “who knew Switzerland was a banana republic?” vibes, too.

Now the JC likes Switzerland, so he is staying right out of that debate: There are plenty of thought pieces from those more learned and temperate than the JC about that. But Switzerland is not a banana republic, and it is known for its banking acumen. This, we think, will be borne out over time.

On creditors ranking behind equity-holders, feelings and so on.

But the conceptual question this all throws up, in the abstract, is an interesting one: should creditors, however subordinated, ever rank behind common shareholders?

Surely not?

Everyone knows AT1s can get converted into equity, at which point they rank equally with shareholders, and even written off — but there seemed to be the expectation that a write-off would only happen if common shareholders are getting written off too.

First, a little spoiler: effectively ranking behind shareholders and actually ranking behind shareholders may feel similar — especially if you have just been written down to zero while the shareholders live to see another day — but they are different things. When an is written down AT1 down to zero, its creditors actually rank ahead of shareholders. It is just that their claim is zero.

Another spoiler: this cannot have come as a surprise. Issuers must have contemplated writing AT1s down while shareholders survived: otherwise, why even have write-down Notes? A write-down contingent on total shareholder annihilation is no different from a normal conversion to equity: you get what the shareholders get: zero. If that is all you wanted, you would just issue normal contingent convertible bonds. But these AT1s were not convertible. There were “Perpetual Tier-1 Contingent Write-Down Capital Notes”. Again, that name. Important.

The whole point of writing down AT1s is to deliver a capital buffer and stave off an insolvency so the bank can carry on.

Pedantry redux

This is where that pedantry we mentioned at the top is important: “capital” is not a thing: it is a difference between things. If the AT1s are vaporised, it follows that the tier 1 capital — now comprising only common equity — is worth 17bn more. If the Write-Down succeeds, the shareholders will live to see another day.

So the JC thinks those central banks who are on record as saying “we’d never write off AT1s before shareholders” are flat out wrong.

A corporation’s shareholders take all the profit and all the losses of the undertaking. You can only work out what those profit and losses are once every other claim on the enterprise has been settled. Those other claims have the feature of being debtor claims. Debtor claims all have defined payoffs; equity claims are, “whatever’s left”.

So, when resolving a company that has gone bust, you must deal with AT1 creditors before you finally settle up with shareholders. You can do this two ways: you can convert the AT1s into shares or, if its terms permit, you can just write them off altogether. Either way, by the time you deal with shareholders, no AT1s are left. Only shareholders remain.

Therefore, the AT1 investors do not actually rank behind shareholders. They can’t. They either become shareholders, or they are goneski. If they get converted into shares they may get some recovery, but only once all the company’s other creditors have been repaid in full. A written down AT1 has been paid in full. The liability was just zero.

Did the AT1s really do worse than common equity?

But AT1 investors whose notes are written off still feel as if they are effectively ranking behind shareholders. This is their lived experience: they get nothing and shareholders get something.

But is that really true?

The panel above illustrates our best guess of the cumulative shareholder return — the diminishing cash dividends paid plus ongoing share price, which was in of course in insistent decline over 5 years — compared with the cumulative AT1 return, which are those much fatter, fixed coupons plus the AT1s’ market value, which we just made up, on the premise that until things got truly dire, it would have been somewhere near par. they were vaporised.

The thing to notice here is that, for a buy-and-hold, long-term investor, the AT1s’ return, event after they were nixed, was miles better than the common equity. The accumulated coupons since issue totalled about a third of its issue price. That is far more than the final acquisition price for the common equity.

“Ah yes, you counter, but just try telling that the the distressed investors who bought their AT1s last Saturday.”

But should we feel sympathy for distressed — ahem, vultures — who buy a 7.25% capital instrument for 20 cents on the dollar during a well-telegraphed existential meltdown?

We should not. Even those who did read the prospectus.

On the difference between equity and debt investment

One invests in shares to take advantage of rapidly changing market conditions. A share’s price is a capricious, will o’ the wisp sort of thing: it flits about, impishly, by driven by the unpredictable humours of the market, whether they are fundamental, structural, geopolitical, or just the product of the delusional madness of crowds.

One can, and many do, make a living trading short-term movements in shares.

In ordinary times, debt instruments — even AT1s — are much less volatile. They do not hop about much, day-to-day. They are sensitive to interest rate environment, and ultimate solvency of their issuer, but their value does not jump around. Their main attraction is their scheduled interest income. Investors only benefit from that over time. AT1s reward long-term investment. In ordinary times, their return is a linear function of how long you have held them, and therefore how for long you provide the bank with tier 1 capital. Day-trading undistressed bonds is not much of a laugh.

It is different in a distressed scenario. Here AT1s are unusually vulnerable: this is the very contingency they are designed to protect the bank — not the investor; the bank — against. As the bank’s capital ratio approaches the trigger threshold, their performance more and more to resembles the equity: Convertible AT1s, in their worst case, become common equity. Their value will converge on the common equity exactly. Write-Down AT1s will become even more volatile than common equity. They are, effectively, binary options: either they are triggered, in which case they are worth zero, or they are not, in which case they recover, will eventually be called at 100 and, in the meantime, will continue to pay fat slugs of interest.

Indeed, this is exactly what we saw.

So, should we feel bad for opportunistic speculators, who picked up the AT1s for a song, but called the market wrong and got hosed? No. This is exactly the bet they were taking.

Should we feel bad for loyal investors who bought at issue, held till the death and were written down to zero? Again, no. They did much better than common equity holders over the life of their investment. It is sad that they got written down, but that is exactly the option they sold when they bought the Notes.

Financial stability wins

The tier one capital layer is there to protect depositors and ensure the stability of the wider financial system, by helping banks to remain a going concern even in times of great stress. That the bank’s ordinary shareholders happen to share that interest is beside the point. AT1s are meant to reward, and did reward, long-term investors. Those who read understood that “Perpetual Tier 1 Contingent Write-Down Capital Notes” meant their notes that could be written down in a time of capital stress most likely had a bank to sell last week.

It sounds like there were plenty of buyers.

See also

References

  1. We have a sense Credit Suisse’s history is not done just yet but that, like Disaster Area frontman Hotblack Desiato, it is merely spending a year dead for tax (and, er regulatory capital) purposes. It may well be back, at least as a high-street banking brand in Switzerland.
  2. Emphasis added. Full documents here.
  3. This is a paraphrase, and an exaggeration for effect, I freely admit. [Full interview here.