The Curious Incident of Credit Suisse AT1

Why a policy mistake should not spell doom for the asset class

Following a week of market turmoil and a crisis of confidence, the Swiss Financial Market Supervisory Authority FINMA approved the takeover of Credit Suisse (CS) by UBS on March 19^th. As part of the deal, FINMA decided to write down in full CS’s AT1 securities, while shareholders were allowed to retain some recovery value.

AT1s are a type of fixed income security introduced after the Global Financial Crisis (GFC) for the purpose of bolstering banks’ capital levels, if needed, without having to resort to taxpayer money. The idea behind the creation of AT1s was to introduce a ‘safety buffer’ between shareholders – who bear the first loss in case of bank resolution or restructuring – and uninsured depositors, whom the resolution authority may want to preserve as much as possible to avoid the risk of a crisis of confidence and a bank run.

FINMA’s decision to fully write down CS’ AT1s without fully writing down shareholders has surprised market participants and triggered a negative reaction. We believe this was a policy mistake that may have longer-term consequences for the credibility of the Swiss banking resolution framework, as it introduces significant uncertainty as to the actual hierarchy of Swiss banks’ creditors resolution and restructuring. The value of Minimum Requirement for own funds and Eligible Liabilities (MREL) and Total Loss-Absorbing Capacity (TLAC) of Swiss banks may start being regarded differently than in other jurisdictions, and we may start seeing equity conversion clauses becoming a more prominent feature of AT1s issued by Swiss banks, who will need to reassure AT1 investors that they will not be wiped out ahead of shareholders.

At the same time, we believe this policy mistake is linked to a specificity of the Swiss banking resolutions framework and would hardly be replicable under the EU Banking Recovery and Resolution Directive (BRRD) or the UK bank resolution framework. This conviction is further reinforced by the publication of a joint statement by the EU Single Resolution Board (SRB) and EU Banking Authority (EBA) clarifying that under EU law, equity instruments are the first ones to absorb losses, and only after their full use would AT1s be required to be written down. A similar statement was put out by the Bank of England.

Protecting senior creditors in line with the hierarchy of claims is key to preserving financial stability, in case banks need to be restructured or resolved. This approach has been consistently applied in the US and Europe in past cases – and most recently in the handling of Silicon Valley Bank in the US, and of Silicon Valley Bank’s UK subsidiary. We see no reason to doubt that it will continue to be applied by resolution authorities in these jurisdictions, and in Europe we think this event might even prompt a re-opening of the long-dormant policy discussion on completion of the Banking Union.

What happened to the CS AT1s?

AT1s can typically be written off in two situations.

The first case is that of a capital breach: if the Common Equity Tier 1 (CET1) ratio of a bank falls below the regulatory minimum (7% or 5.125% depending on jurisdictions) of its Risk Weighted Assets (RWA), the convertible instruments are written off or converted into equity. This would happen for any value of CET1 below the regulatory minimum, even if CET1 were to remain above 0% – which implies that in the case of a capital breach, AT1s could be written off even with shareholders keeping residual value.

This was not the case with Credit Suisse. CS’s last reported CET1 ratio stood at 14% and in a press-release issued on Wednesday 15^th the Swiss authorities confirmed that the bank was in compliance with the liquidity and solvency requirements, which would be much higher than the above-mentioned thresholds for AT1 loss absorption. A second statement issued on the day of restructuring (March 19^th) the Swiss Authorities explained they needed to take action because due  to a crisis of confidence “there was a risk of the bank becoming illiquid, even if it remained solvent”, reason why “it was necessary for the authorities to take action”.

While illiquidity can obviously turn into insolvency if dragged on for too long, insolvency does not appear to have been the direct trigger of the AT1 conversion in the case of Credit Suisse.

The second case where AT1s can be written off or converted is when public support is given to the bank in the context of restructuring or resolution. This follows directly from the post-GFC doctrine that banks’ creditors should bear the costs of restructuring and resolution before public money is deployed. It implies that some degree of bail-in of creditors is a pre-requisite for public support.

Yet, the expectation in the case of public support is that bail-in would still follow the “pecking order” implicit in the hierarchy of claims – i.e. shareholders would be wiped out first, followed by bondholders, up to uninsured deposits. In the case of CS, this did not happen. Despite being technically senior to equity, AT1s were fully wiped out while shareholders were allowed a recovery value of CHF 3 billion.

So, why were CS AT1s written down?

What happened appears to be linked to a specificity of the Swiss Banking Act when it comes to the power FINMA may exercise during restructuring proceedings. These include the power to convert into equity partially or fully and/or write-down the obligations of the bank, including AT1s.

What is special about FINMA, however, is that they seem to retain more flexibility than most global resolution authorities when it comes to how they can do the above. When deciding to write off liabilities in restructuring proceedings, FINMA may not be required to follow any order of priority, which means debt could be cancelled in whole or in part prior to the cancellation of any or all of the equity capital.

The key word here is may, as it suggests that FINMA has no obligation to proceed in the way it did in the Credit Suisse case but retains the flexibility to do so. And so it did, on Sunday. According to Swiss Finance Minister Karin Keller-Sutter (see minute 13:14 of the press conference here), all decisions made between March 16^th and 19^th by the Swiss Federal Council were based on Notrecht, i.e. emergency law. This can be thought of as an executive order – which is regulated under Article 184 (3) and Article 185 (3) of the Swiss Federal Constitution. Emergency law is temporary and must be replaced by ordinary law – which means the Federal Council will have to submit a new bill to parliament within six months.

In our view, the decision to exercise this discretion was a significant policy mistake on the side of FINMA. It introduces a distortion in the hierarchy of claims in the Swiss financial system, and it raises questions as to how senior debtholders really are in Swiss banks. As a result, the value of other debts securities issued by Swiss banks to meet MREL and TLAC requirements may now be regarded differently than in other jurisdictions and Swiss banks may have more difficulty and may need to pay more to raise funds to meet those requirements.

In Europe, Swiss banks’ AT1 are virtually the only ones to have permanent write-down language, while most other large banks in the EU and UK are 50-50% split between equity conversion and temporary write-down. From now on, we may expect equity conversion to become a more prominent feature of AT1s issued by Swiss banks – who will certainly need to reassure AT1 investors that they will not be wiped out ahead of shareholders.

Can this happen in the EU?

We do not believe so – and the recent statement by EU regulators clearly states that common equity instruments are the first ones to absorb losses, and only after their full use would Additional Tier 1 be required to be written down.

In the EU, extraordinary public financial support to a bank would typically trigger resolution (and thus a bail-in of 8% of liabilities), unless the public support is of a “precautionary” kind – i.e. required to “remedy a serious disturbance in the economy of a Member State and preserve financial stability”. A State guarantee to back liquidity facilities provided by central banks, a State guarantee of newly issued liabilities, and even an injection of own funds would typically fall under this category – meaning that a situation similar to the Credit Suisse case could be dealt with in an orderly fashion within the BRRD framework without national authorities resorting to domestic emergency powers.

In case bail-in is triggered, then Article 48 of the EU Bank Recovery and Resolution Directive (BRRD) is extremely clear as to what the pecking order is, when writing down liabilities: CET1 would be hit first, followed by AT1s, Tier 2 (T2) instruments, any subordinated debt that is not AT1 or T2 in accordance with the hierarchy of claims in normal insolvency proceedings, and only if all the above is not enough to restore capital to its regulatory minimum, uninsured deposits.

Whenever bail-in is applied, BRRD explicitly states that creditors can be written off “if and only if” the category of debt that is immediately junior to them has been written off first. Article 48(5) of BRRD explicitly states that “when deciding on whether liabilities are to be written down or converted into equity, resolution authorities shall not convert one class of liabilities, while a class of liabilities that is subordinated to that class remains substantially unconverted into equity or not written down”.

There is some flexibility in the EU framework for resolution authorities to exempt certain liabilities from bail-in. But this flexibility is worded in much narrower terms than what appears to be the case in the Swiss framework. Article 44(3) of BRRD lays out the “exceptional circumstances” in which this may happen, but while opening to the exemption of deposits, none of those exemptions seem to suggest that debt could be fully bailed in before equity is wiped out. More specifically, EU resolution authorities could exempt liabilities from the regular cascade of bail-in if:

1. It is not possible to bail-in that liability within a reasonable time;
2. the exclusion is strictly necessary and is proportionate to achieve the continuity of critical functions and core business lines in a manner that maintains the ability of the institution under resolution to continue key operations, services and transactions;
3. the exclusion is strictly necessary and proportionate to avoid giving rise to widespread contagion, in particular as regards eligible deposits held by natural persons and micro, small and medium-sized enterprises, which would severely disrupt the functioning of financial markets, including of financial market infrastructures, in a manner that could cause a serious disturbance to the economy of a Member State or of the Union; or
4. the application of the bail-in tool to those liabilities would cause a destruction in value such that the losses borne by other creditors would be higher than if those liabilities were excluded from bail-in.

So, while a solution like the one implemented by US authorities for Silicon Valley Bank – with full wipe-out of shareholders and creditors but full protection of depositors – would be entirely possible under BRRD based on the above, a solution like the one implemented by FINMA for CS – with full wipe-out of AT1 holders without full wipe-out of equity holders absent a capital breach– would not appear to be possible.

Disclaimer: These podcasts should not be copied, distributed, published or reproduced, in whole or in part. These recordings are for informational purposes and have been prepared by Algebris (UK) Limited (“Algebris”). Each podcast is not intended to be relied upon as a forecast, research or investment advice, and are not a recommendation, offer or solicitation to buy or sell any securities or to adopt any investment strategy. Opinions expressed are as of the date of recording and are subject to change. The views and opinions of any guest participants on these podcasts are not necessarily those of Algebris and its affiliates.
The information and opinions contained in these recordings are derived from proprietary and non-proprietary sources deemed by to be reliable by Algebris and are not guaranteed as to accuracy or completeness. They may contain ’forward looking’ information that is not purely historical in nature. There is no guarantee that these will come to pass. Reliance upon information on this site and/or the recordings is at the sole discretion of the reader and/or listener. Past performance is not indicative of current or future results. Algebris is not providing any financial, economic, legal, accounting, or tax advice or recommendations in these podcasts.