The Lombard Review हिन्दी

Why Credit Suisse bondholders lost everything before shareholders

Looking north to the Loop from the rail yards west of the Field Museum, Chicago, 1978.
Looking north to the Loop from the rail yards west of the Field Museum, Chicago, 1978. Photo: Downtowngal/Wikimedia Commons · CC BY-SA 4.0

Credit Suisse's Additional Tier 1 (AT1) bonds, worth CHF 16 billion ($17.3 billion), were written down to zero on Sunday, 19 March, as the Swiss authorities pushed the bank into an emergency marriage with UBS. The Swiss Financial Market Supervisory Authority (FINMA) made the call. Shareholders, who normally lose everything first in an insolvency, walked away with roughly $3.25 billion in UBS stock. With one regulatory decree the usual order of the capital structure was turned upside down, and the $275 billion global market for contingent convertible bank capital was thrown into turmoil.

Modern insolvency law rests on absolute priority. Shareholders take the first loss and junior debt takes the next, while senior creditors are protected until all the junior capital is gone.

A slow collapse, then a fast one

Credit Suisse had been weakening for two years. In 2021 it lost billions on the collapse of Archegos Capital, a family office, and had to freeze supply-chain finance funds linked to Greensill Capital. A restructuring announced last October did not stop clients leaving, and the bank disclosed outflows of about CHF 110 billion in the final quarter of 2022. This month it delayed its annual report after questions from American regulators and then admitted material weaknesses in its financial controls.

The end came within a week. On 15 March its largest shareholder, Saudi National Bank, ruled out putting in more capital, and the shares fell to a record low. The Swiss National Bank offered support and Credit Suisse said it would borrow up to CHF 50 billion from it, but deposits kept leaving. By the weekend the authorities had arranged the takeover, using emergency law to bypass a shareholder vote.

The choice of a merger was itself a departure. Since 2008 Switzerland has built a resolution regime for its big banks, under which a failing lender would be recapitalised by converting its bail-in bonds into equity while its Swiss operations carried on. The authorities judged that route too risky in a week of global banking stress and opted for a sale to UBS instead. Once the resolution plan was set aside, the treatment of each class of investor depended on emergency decisions rather than on a pre-agreed script.

What the prospectus allowed

FINMA leaned on obscure wording in Credit Suisse's Tier 1 prospectuses, which allowed a complete write-down if the government stepped in with extraordinary support (a "Viability Event"). Since the Swiss state had provided a CHF 9 billion guarantee against losses and a large emergency liquidity line, FINMA argued that the trigger had been met. The decision may well stand up under Swiss emergency law. For the wider asset class it was a commercial disaster.

AT1 bonds were invented after 2008 for exactly this kind of moment. They are perpetual securities whose coupons a bank can skip without defaulting, and they convert into shares or are written down if the bank's core capital ratio falls below a set trigger or if the regulator declares it no longer viable. The design was meant to make bondholders, rather than taxpayers, absorb losses in a failing bank. Nobody disputed that AT1 holders could lose everything. What surprised the market was losing everything while shareholders kept something.

The BlueScope steelworks at Port Kembla, Australia.
The BlueScope steelworks at Port Kembla, Australia. Photo: Marek Ślusarczyk/Wikimedia Commons · CC BY 3.0

Pricing a bond that can vanish

The quantitative problem for investors is how to price that risk now. A credit spread compensates for two things, the chance that a loss occurs and how much is lost when it does. Before Sunday, analysts modelled an AT1 loss as a function of the bank's capital ratio and assumed recoveries would at least match those of shareholders. FINMA has shown that in a state-supported rescue the loss on an AT1 can be total even when the equity is worth something.

That changes the maths across the market. If loss given a rescue is close to 100 per cent regardless of the equity, the probability of state support becomes the variable that matters, and state support is most likely precisely at the systemically important banks that issue the most AT1 debt. Models that treated these bonds as a senior slice of equity risk now need a separate term for political discretion, which is hard to estimate from past data.

Every AT1 now looks riskier

Investors buy subordinated bank debt on the firm assumption that shareholders will be wiped out before bondholders lose a penny. By putting bondholders below shareholders, the Swiss authorities added a political risk premium to every AT1 bond issued by a European bank.

Regulators in Frankfurt and London hurried out statements saying that in the European Union and the UK, shareholders will always take losses before AT1 holders. Past European resolutions support that promise. When Banco Popular failed in 2017, Spain's resolution authority wiped out its shareholders, wrote down its AT1 and Tier 2 debt and sold the bank to Santander for a single euro, in the order the rules intended. The Dutch state's nationalisation of SNS Reaal in 2013, which expropriated subordinated bondholders, also left shareholders with nothing. The Swiss decision stands out because it inverted that sequence.

Who wins and who loses

UBS is the plain winner, buying a rival for a fraction of its book value with a state guarantee against part of the losses, and the write-down of AT1 debt strengthens the combined bank's capital at no cost to it. Credit Suisse shareholders lost most of their investment but kept something. The AT1 holders, a mix of institutional funds and wealthy private clients, received nothing, and legal challenges to the write-down are widely expected.

The wider losers are banks that rely on AT1 debt to meet capital requirements. Their instruments fell sharply in price on Monday before recovering part of the drop after the European statements. The first test will be whether any large bank chooses to redeem its AT1 bonds at the next call date, since calling them makes sense only if new bonds can be sold on acceptable terms. Skipped calls, or a long pause in new issuance, would show that the market has not believed the regulators' reassurance.

FINMA spared Credit Suisse a formal liquidation by putting Swiss convenience ahead of the creditor hierarchy. The price will be paid by other banks, which are likely to face much higher costs to raise AT1 capital for several years.

Write to Arnav Goel at contact@thelombardreview.com

More From The Lombard Review