As the global banking giant Credit Suisse Group AG struggled to survive, UBS Group AG agreed on Sunday, March 19, 2023, to take over it in an all-share transaction brokered by the Swiss government. The takeover surprised investors by wiping out 16 billion francs of AT1 CoCo bonds while paying 3 billion francs to equity holders.
Among the investors in Credit Suisse AT1 Bonds were Saudi National Bank, Qatar Investment Authority, Pimco, Invesco, BlackRock and several major hedge funds. All these investors are sophisticated and expected to understand the risks of AT1 bonds. However, the risk of wiping out was not reflected in the price until the wipeout was announced. When investors were struggling to understand whether AT1 bonds are debt or equity, the Swiss regulator told investors to read the prospectus. Credit Suisse treated AT1 bonds as debt and equity simultaneously. It used the bonds as debt for cheaper funding, and it used AT1 bonds as equity to meet regulatory capital requirements. Was Credit Suisse using CoCo bonds to arbitrage between bond market and banking regulation?
The Financial Stability Oversight Council (FSOC) reported to Congress that U.S. regulators would leave CoCo bonds as a private sector innovation instead of using them as regulatory capital.
The wipeout of Credit Suisse AT1 bonds rattled the global AT1 market. The price of AT1 bonds collapsed the takeover announcement. It recovered partially a couple of days later as investors started realizing that not all AT1 bonds have the 100% write-down feature. In fact, most AT1 bonds either convert to certain shares of equity or write down partially when triggered. The global AT1 market spans across Europe, North America, and Asia. The countries with most banks issuing CoCo bonds are China, India, UK, and Switzerland. None of the U.S. banks have issued any CoCo bonds. In 2012, the Financial Stability Oversight Council (FSOC) reported to Congress that U.S. regulators would leave CoCo bonds as a private sector innovation instead of using them as regulatory capital. One can question whether the main purpose of CoCo bonds is to arbitrage between regulation, tax law, and bond markets. Any economic theory that explains why banks issue CoCo bonds should explain why U.S. banks do not issue CoCo bonds.
The main selling point by the proponents of CoCo bonds is that CoCo bonds help financial stability and protect taxpayers. However, the takeover of Credit Suisse by UBS put Swiss taxpayers at risk again 15 years after they had bailed out UBS. As part of the takeover deal, the Swiss National Bank provided a 100-billion-franc liquidity line to UBS, and the Swiss Department of Finance provided a 9-billion-franc guarantee for potential losses in the assets of Credit Suisse. Evidently, the 16-billion-franc CoCo bonds did not insulate Swiss taxpayers from bank risks. A poll conducted a few days after the takeover announcement shows three quarters of Swiss people are unhappy about the takeover brokered by their government. The convincing logic in the selling point of CoCo bonds must be flawed somewhere. It is important to understand where the flaws are.
With market trigger, CoCo bonds and bank stock can have either multiple equilibrium prices or no equilibrium price at all.
The late recapitalization of Credit Suisse points to a problem of CET1 ratio. It is an accounting measure based on book value and does not signal the status of a bank. In the month before Bear Stearns was taken over by JPMorgan Chase on public assistance, the estimated CET1 ratio of Bear Stearns was 13.5%. In the month before Lehman Brothers bankrupted, its CET1 ratio was 10.1%. Right before Silicon Valley Bank was closed by FDIC, its CET1 ratio was 15.4%. If the CET1 ratio trigger is not put into use, banks are left to regulators to decide when to recapitalize. As we have always seen, regulators often act so late that the bank is no longer a viable going concern. The ineffectiveness of accounting triggers was anticipated by many economists. It is also widely recognized that accounting ratios are subject to management manipulation. Exactly for these reasons, some economists advocate market trigger: a trigger on the market value of equity. They argue that market trigger is timely, objective, and hard to manipulate. These arguments sound reasonable, but CoCo bonds with market triggers have a fundamental issue as first pointed by Sundaresan and Wang. With market trigger, CoCo bonds and bank stock can have either multiple equilibrium prices or no equilibrium price at all. In either case, information about bank value will be lost, causing large pricing uncertainty, which may be open to market manipulation. These concerns are confirmed by economic experiments. Since then, economists have tried to develop alternative trigger mechanisms, but so far none is practical. Additionally, CoCo bonds with write-downs clearly subsidize equity holders when triggered. The 100% write-down of Credit Suisse AT1 bonds is clearly a wealth transfer from AT1 bond holders to equity holders. Research is needed to understand whether AT1 bonds exert incentives for managers to run banks down.
A near-trigger experience by Deutsche Bank in 2016 further suggests that a trigger event is likely to be viewed as a negative signal about the bank. When the Deutsche Bank CET1 ratio was near the triggers of some its CoCo bonds in 2016, the bank CDS spread jumped to 248 bps, and the bank stock price lost more than 40% in two months. Deutsche Bank tried to avert the trigger by offering to buy back 3 billion euro-denominated debt and 2 US billion dollar-denominated debt. The bank’s CEO John Cryan later lamented his experience with CoCo bonds: “They are not great instruments. We shouldn’t really need them going forward. ... when you want them to be debt they are equity and when you want them to be equity they are debt.”
Some basic math for the takeover of Credit Suisse shows that the wipeout of AT1 bonds helped the Swiss government to maintain financial stability
Perhaps a positive feature of AT1 bonds is making finding a buyer possible, thus avoiding nationalization or liquidation. Some basic math for the takeover of Credit Suisse shows that the wipeout of AT1 bonds helped the Swiss government to maintain financial stability by finding a buyer to take over Credit Suisse. Nationalization or liquidation of Credit Suisse may throw the world into a much deeper banking crisis with uncertain damage to Swiss taxpayers. In this sense, AT1 bonds are better than regular bonds for stabilizing the banking industry.
Even though AT1 bonds helped the takeover of Credit Suisse, it is still questionable whether CoCo bonds are better capital than common equity. Suppose Credit Suisse had issued 16-billion-franc common equity, instead of 16-billion-franc AT1 bonds. Then, there would be no need for regulator to decide when to wipe out the complicated bonds and to surprise investors, causing large volatility and uncertainty in the global bond market. The market would wipe out the value of common equity in response to information about the bank. For the takeover, the per share value of Credit Suisse equity would have been lower, but the total equity value would have been the same. Credit Suisse would not have been able to arbitrage between banking regulation and bond market. Would this have been a better scenario? More research is needed to understand which security, CoCo bond or common equity, makes financial markets and institutions more stable.
By Zhenyu Wang, Edward E. Edwards Professor of Finance in Kelley School of Business, Indiana University. He was formally a vice president and the head of Financial Intermediation Function in the Federal Reserve Bank of New York.
To read the unabridged version of this article, "CoCo Bonds: Are They Debt or Equity? Do They Help Financial Stability? — Lessons from Credit Suisse NT1 Bonds", click here.
To watch Prof. Zhenyu's lecture entitled “CoCo Bonds: Are They Debt or Equity? Do They Help Financial Stability?” for The Institute for Corporate Governance Public Lecture Series, click here.
If you would like to read further articles from the 'Banking Crisis 2023' Blog edition, click here.
The ECGI does not, consistent with its constitutional purpose, have a view or opinion. If you wish to respond to this article, you can submit a blog article or 'letter to the editor' by clicking here.