top of page

Bail-In Out: What Happened to Bondholders in the UBS–Credit Suisse Merger?

By PATRICK HAGGERTY

Following the announcement over the weekend that Swiss regulators had forced through an emergency acquisition by UBS of its beleaguered compatriot Credit Suisse, questions have circulated among investors and other observers about a seemingly esoteric technical detail in the press release:


The extraordinary government support [provided by Swiss regulators] will trigger a complete write-down of the nominal value of all AT1 debt of Credit Suisse in the amount of around CHF 16 billion, and thus an increase in core capital.


For those who don’t obsess over the minutiae of bank regulation or occupy obscure corners of the bond market, this statement means that for investors who purchased certain bonds issued by Credit Suisse, their entire investment (more than $17 billion!) was unilaterally wiped out by Swiss authorities. This was surprising to most people, because UBS technically purchased Credit Suisse. Credit Suisse shareholders received more than $3 billion in compensation (in the form of UBS stock). Anyone who has studied basic accounting understands that the claims of bondholders rank higher in priority than the claims of stockholders. So what gives?


What are “bail-in” bonds and how do they differ from bailouts?


Banking regulators substantially revised global capital standards and resolution frameworks following the Great Financial Crisis. One of their policy goals was to prevent the need for future government bailouts to rescue or resolve a troubled “too big to fail” bank. To address this problem, regulators created the concept of contingent capital (or “bail-in”) debt. The idea is to require large systemically important banks to issue bonds that can either be wiped out or converted to common equity if the bank runs into severe trouble. If triggered, this action eliminates a liability of the bank, and thereby boosts the bank’s capital position (i.e., shareholders’ equity) without any capital injection from the government. Stated differently, the bail-in bond investors are forced to eat losses that otherwise may have threatened the claims of depositors. In the best case scenario, this might allow the bank to remain a going concern. In the worst case, it reduces the amount of loss suffered by depositors (or deposit insurance schemes) in the event the bank fails.


What are AT1 bonds and why were they wiped out by the Swiss regulators?


In Europe and other jurisdictions it’s common for banks to issue what are known as contingent convertible bonds (aka "CoCos"). These are a form of “bail-in” debt. Under the new capital rules, they qualify as a component of the bank’s regulatory capital if they meet certain criteria. The subset of CoCos at issue in the UBS - Credit Suisse merger were a more loss absorbing, and thus higher risk, variant designed to meet Swiss requirements for Additional Tier 1 capital (or "AT1” for short, hence the name).


The bonds were sold with clear disclosure that if the bank's financial condition deteriorates to the point that its solvency is threatened, the bondholder's claim as a creditor may either be extinguished or their rights converted into common equity. There were explicit triggers listed in the prospectuses for these bonds that could lead to their claims being wiped out if certain events occurred (see for example). In this case, the Swiss regulators cited a clause that allowed a full write-down of the bonds’ value if Credit Suisse received “extraordinary support” from the government, without which it would have become insolvent, bankrupt, or unable to carry on its business.


This was in effect a partial bail-in. It boosted the capital position of Credit Suisse and therefore improved the capital ratios of UBS following the merger, but was only triggered after the Swiss government pledged roughly $280 billion in government support to facilitate the deal. The bail-in did not prevent the bail-out.


Where does the money for a bail-in come from and why was there confusion?

Bail-in bonds, such as the AT1 bonds in the case of Credit Suisse, may only count as capital if they have been “paid in.” This means that the Bank has received the full proceeds of the bond sale. Therefore, the money for a “bail-in” comes from the investors who previously purchased the bonds. By extinguishing the bondholders’ claims, the bank retains full use of the money but sheds any obligation to pay it back.


The investors who purchased these bonds knew (or should have known) that they came with the risk of being written-down in the event of severe stress and they received higher yields as a result. However, it would not have been clear that the regulators would choose to permanently write-down the value of the bonds in a situation where common shareholders receive compensation as happened with Credit Suisse. That is a highly unusual move. It remains unclear why the Swiss regulators chose in the heat of the moment not to convert the AT1 bonds to equity.


Markets for AT1 and other bail-in bonds have predictably seen spreads widen dramatically following the Credit Suisse announcement. In response, banking authorities in jurisdictions where AT1 bonds have become popular quickly issued statements seeking to calm investor fears and paint the Credit Suisse deal as a unique occurrence (see for example the EU, Hong Kong, UK, and Canada).


The broader implications that the Swiss decision will have on banks’ ability to raise capital using similar instruments going forward remains to be seen, although there are technical differences between the various bond structures, and investors must pay close attention to the fine print.


Has the concept of bail-ins been incorporated into U.S. banking laws?

The Dodd-Frank Act required the Financial Stability Oversight Council (FSOC) to study the benefits and drawbacks of bail-in debt requirements (i.e., contingent capital requirements). The council issued a report in 2012 recommending that it "remain an area for continued private sector innovation" and encouraging the Fed and other regulators "to continue to study the advantages and disadvantages of including contingent capital and bail-in instruments in their regulatory capital frameworks."


Banking organizations may count certain preferred stock instruments with debt-like qualities as Additional Tier 1 capital under U.S. regulatory capital rules. These rules also allow banks to issue subordinated debt that qualifies as Tier 2 capital under certain conditions. However, neither of these instruments have the same “bail-in” features as CoCos found in Europe, Asia, and elsewhere.


In 2016, the federal banking agencies issued long-term debt (LTD) requirements in connection with a new rule requiring Global Systemically Important Banks (GSIBs) in the U.S. to hold a certain amount of "total loss absorbing capital" (or TLAC). The LTD portion of TLAC serves as a form of "bail-in" debt that can be converted into equity when a GSIB is facing a solvency crisis. However, these instruments differ from the CoCos found in Europe in that they do not contain contractual triggers that would cause or allow them to be written-down or recharacterized as equity prior to the bank being placed into receivership. Unlike CoCos in other jurisdictions, U.S. LTD instruments only serve as a source of “bail-in” capital after the banking organization has filed for bankruptcy or been placed into receivership. Therefore, as a practical and legal matter, extinguishing the bondholders’ claims would not allow the bank to remain a going concern.


One of the stated reasons why the U.S. has not adopted contingent capital requirements on par with what you see in other jurisdictions is that U.S. tax laws and accounting standards present unique challenges. Among other things, regulators have noted that a Swiss-style CoCo issued in the U.S. would likely not qualify as "debt" and thus the issuing bank's payment of interest would not be deductible as an interest expense. This would make issuance of such instruments significantly more costly than typical subordinated debt.


Did the 2018 tailoring rules impact the scope of “bail-in” requirements in the U.S.?


Not directly. The U.S. TLAC and LTD rules have always only applied to GSIBs. The tailoring rules adopted in 2018 did not impact the scope or application of those rules. However, in 2020, the banking agencies amended their capital rules to make it punitive for large banks subject to “advanced approaches” to hold LTD of other large banks. The purpose of the change was to reduce interconnectedness within the financial system and systemic risks. The 2018 tailoring rules increased the total asset threshold for applicability of the advanced approaches from $250 billion to $700 billion. So in a sense, the tailoring rules limited the scope of impact of the more recent changes. However, the tailoring rules were already in place when the 2020 rules were finalized.

Has a “bail-in” ever occurred in the U.S. and would it lead to the same confusion as the UBS-Credit Suisse deal?


There have been no “bail-ins” in the U.S. because the FDIC has thankfully not had to take any GSIB into receivership under its orderly liquidation authority. If a GSIB were to fail, and the claims of investors holding its LTD instruments were extinguished, this result should not be unexpected. What surprised investors in the Credit Suisse situation is that their claims were extinguished, while the more junior common stockholders' claims were not. That should not happen in the U.S. framework because the triggering of any writedowns of the LTD instruments would occur only as part of an orderly resolution process, which is clearly spelled out ex ante.


Would depositors lose money with a bail-in, were one to occur?


The main objective of bail-in debt is to absorb losses and thereby serve as a buffer to prevent those losses from reaching the bank's depositors. In any situation where debt is used to bail-in the bank, it would only serve to benefit depositors.


Did Silicon Valley Bank or Signature Bank have any bail-in debt?


No. Neither of the two banks that failed recently were GSIBs and thus they were not subject to the TLAC and LTD rules, or the FDIC’s orderly liquidation authority. Silicon Valley Bank’s parent holding company has filed for bankruptcy protection. The ultimate recovery of its unsecured creditors is already part of a brewing dispute with the FDIC.


Image by Credit Commerce from Pixabay

bottom of page