Design for confidence: the write-down of AT1 bonds in the last days of Credit Suisse
The reputation of Swiss regulator, Finma is on the line as the holders of Credit Suisse’s AT1 bonds take legal action, seeking compensation after securities they owned were written down, as part of the bank’s forced marriage to UBS.
Public image ramifications aside, the outcome could also bring into question the effectiveness of post global financial crisis (“GFC”) banking regulation.
But do the bondholders have a point? To better understand their position, we need to turn the clock back to 2008. After the GFC, regulators imposed a set of new rules to avoid another international crisis. These new requirements forced banks to hold large buffers of liquid assets against the risk that they might suffer in a bank run. The amount of capital that banks were required to hold was increased and the definition of ‘capital’ was narrowed. The regulators backed this up by providing an alternative to the chaos caused by previous disorderly failures of large banks. This became known as a “resolution framework”.
The orthodoxy amongst regulators was that this new policy worked. They believed that a bank that met the higher capital and liquidity requirements would inevitably gain the market’s confidence.
This belief informed the design of capital securities, like Credit Suisse’s AT1 bonds – considered capital under the new rules. The reasoning was that capital ratios were key to survival. If a bank’s capital ratio fell dramatically – which would happen if it lost a lot of money – the AT1 bonds would be written down, improving the ratio again. Regulators believed that this would restore confidence. The bonds could also be written down if “customary measures” to restore the bank’s capital ratio were “inadequate or infeasible”.
Credit Suisse’s AT1 bonds were built with this failure mode in mind. If there were losses, it would cause capital ratios to falland there would be attempts to raise new capital. If these worked, great: if not the bonds would convert and save the day.
In the case of Credit Suisse, however, the bank remained well-capitalized throughout. Despite its healthy capital ratio, there was still a loss of confidence which triggered depositors and other counterparties to run anyway.
When this happened, it left the Swiss authorities with a problem. The AT1 bond write-off had not been automatically triggered because Credit Suisse’s capital ratio was far above the required point and therefore, there was no need to raise new equity. By the usual doctrine that central banks should lend freely to a solvent bank against sound collateral if no one else would, the answer was clear: keep Credit Suisse afloat.
The authorities decided not to do that. Instead, they brokered a forced sale of the bank to UBS. Given that only one buyer was considered, and that the sale was hurried, UBS also took over its biggest rival on the cheap while it was decided that the AT1 holders should share the pain. Therefore, an emergency law was passed which allowed the bonds to be written down.
With all this in mind, the bondholders have a strong case for redress. Credit Suisse did not need capital: it needed liquidity. Rather than providing that liquidity, the authorities chose to sell the bank, unilaterally changing the terms of the AT1 bonds and forcing bondholders to absorb some of the associated cost.
It is reasonable to ask why post-GFC liquidity regulation did not help here. The answer again lies in confidence. Large banks are required to keep large cash buffers at all times. They must calculate their so-called “liquidity coverage ratio: every day, which is the ratio of qualifying liquid assets to the bank’s estimated need. If the ratio starts to fall, investors and ratings agencies become nervous. Therefore, banks are reluctant to use their liquidity buffers as it could create or exacerbate a loss of confidence.
This phenomenon was already evident during the Covid-19 pandemic. Despite large cash outflows, many banks were reluctant to draw on their liquidity buffers. Both the Bank of England and the European Banking Authority issued papers encouraging buffer use, but to little avail: European banks’ average liquidity coverage ratios was 146 per cent during Covid. Similarly, Credit Suisse had a ratio of over 150 per cent on the Tuesday before it was sold. Post-crisis liquidity regulation, it turned out, was not enough to provide bank stability once a run on the bank had begun.
The resolution framework did not fare well in the face of events either. The framework was not used to avert a collapse of Credit Suisse which the Swiss finance minister, Karin Keller-Sutter,said “would have triggered an international financial crisis”.
Despite this, regulators are not examining the major pillars of post-GFC policy making. In July 2023, Michael Barr, the United States Fed Vice Chair for Bank Supervision, said that he thought the existing approach to capital requirements was sound and advocated for a raise in required capital. It’s clear that Mr. Barr has not learned from Credit Suisse’s demise.
The whole saga casts serious doubt on the actions taken in two Swiss cities: Basel, where the post-GFC banking regulations were designed, and Bern, where the authorities decided to write down the Credit Suisse AT1 even though the grounds for doing so were, at best, tenuous. When the bondholders get their day in court, we should hope that the whole story is revealed. At play is not just the confiscation of bondholders’ property, important though that is, but also the assumptions underpinning the foundation of post-GFC bank regulation.