This week, financial markets were spooked by the write-down of Additional Tier 1 securities of Credit Suisse. Soon after the writedown, a discussion ensued regarding its legitimacy. One camp argues that investors in these securities should have read the instruction manual. See the pointy opinion piece in IFR by Prasad Gollakota, the former co-head of global capital at UBS. The other camp points out that the write-down was not really legit and that equity holders should have been wiped out first. See Jérôme Legras’s excellent deep dive on this matter in the Financial Times. To support their position, those in the second camp point out that the ECB and other supervisors published press releases informing financial markets that they would preserve the traditional capital structure hierarchy. The ECB would wipe out the holders of common equity before AT1 holders and holders of more senior securities.
I found the discussion on the legal base of the write-down fascinating because I spent some years at various working groups of the Basel Committee, the EBA, and the European Commission drafting the rules for these securities. This week was déjà vu all over again, because many of the discussions we held in 2009-2013 resurfaced; all the pros and cons. The difference between then and now is that we all aged a bit and that we learned from some notable glitches and failures.
I was particularly taken aback by the intense and passionate debate that transpired between the two opposing camps. Neither of them is really wrong. And yet, there is strong disagreement. How come?
This time is different
My explanation of the disagreement between the two camps is time. Shortly after the onset of the GFC, in 2008-2009, bank regulators felt betrayed by the inability of old-style CoCos to absorb losses. Losses were instead imposed on taxpayers, which did not end well. A strong backlash against these non-common equity securities followed.
The uptake of these securities had grown steadily since the nineties, with support from supervisors. For example, in 1998, the Basel Committee issued a press release that allowed banks to use up to 15% of hybrid capital to meet Tier 1 requirements. This was two years after the Federal Reserve decided to allow banks to use Trust Preferred Securities (TruPs) to meet Tier 1 capital requirements. The primary reason for issuing hybrid capital was that it was cheaper than common equity: payments on these securities, or coupons, were tax-deductible.
Old-style CoCos, or hybrids, had many debt-like features that prevented them from absorbing losses. For example, their coupons cumulated, and there were features such as dividend pushers that made them senior to equity. Unlike common equity, hybrids also matured, so investors expected these bonds to be repaid. Outside the US, hybrid securities would generally be redeemed after five years. In addition, various clauses, such as step-ups, incentivized banks to redeem the hybrids early, which created expectations among investors.
Post-GFC, bank regulators responded predictably to the failure of old-style hybrid capital to absorb losses. Hybrid Tier 1 capital could only return to the bank capital stack if it met way stricter requirements. EBA’s predecessor, CEBS, redefined hybrid capital as early as 2009 (see these guidelines). Out went the coupon cumulation, incentives to redeem, pushers, alternative coupon satisfaction mechanisms, and early retirement clauses. One year later, the Basel Committee published a new definition of capital: Basel III, which is now in force, features strict criteria for Additional Tier 1 capital, many of which are inspired by the failure of old-style hybrid securities.
Basel III compliant hybrid securities should be written down or converted into equity at a specified trigger point, in going concern. It is appropriate to write off these capital securities entirely when regulators determine that a bank’s viability is at stake.
The wilful juniorization of AT1 capital
As a result of initiatives such as Basel III, Additional Tier 1 CoCo capital has features that often render it junior to equity. One example is the option of cancelling coupons, another is the requirement that AT1 securities should not hinder recapitalization, and there is the write-down in going concern.
The juniorization of AT1 capital was all fine at back then. To please taxpayers, AT1 capital was designated as the whipping boy of the bank capital stack. No sympathy for the holders of these securities, who were deemed sophisticated, wealthy or institutional. In contrast, there was less hostility towards common equity holders. This makes sense because investing in common equity is popular. Bank shares are popular because they pay a steady stream of dividends. (An exception is the ban on dividend payments in 2020 due to the COVID-19 pandemic. However, this was likely a one-off event, and the ECB went on record to admit just that.)
So, why this change in sentiment? Wasn’t it agreed that AT1 should do a lot of the loss absorbing legwork, in some cases, first?
Something must have changed
My explanation relies on two events that likely changed supervisors’ attitudes. The first was the Deutsche Bank press release in early 2016, when the bank announced that it could pay its AT1 coupons. That press release sent markets in a tailspin, because investors realized the impact of a just published opinion of the EBA on Pillar 2 requirements. That opinion, endorsed by the ECB, would reveal that Deutsche Bank was sailing close to the wind and potentially about to breach its capital requirements, while at the same time, the bank was not in great shape. After a while and some concessions from the EBA, the markets calmed.
My takeaway from the Deutsche incident was that supervisors realized that AT1 was systemically important. Supervisors changed their positions on CoCos accordingly. Whatever happens to these wicked securities should not affect financial stability.
The second event was the failure of the Banco Popular in 2017, which prompted the ECB and SRB to use the European recovery and resolution framework. The use of the BRRD bail-in tool wiped out the investments of equity holders and investors in Tier 1 and Tier 2 capital securities. The ECB’s chief supervisor Danielle Nouy was confident about the success of the European recovery and resolution framework: “modestly… I would say that we have passed the test. … I fail to see even with the knowledge we have today what we could have done differently.” (FT.com June 20, 2017)
My takeaway from the failure of Popular: European bank supervision can rely on the recovery and resolution framework, the BRRD and everything associated with that directive.
So, in the context of a stronger focus on financial stability and the availability of a well-functioning recovery and resolution framework, it is not surprising that the ECB this week attempted to calm marks with that press release. Referring to the gold-plated European bank palliative care framework (BRRD), with all its tools and whistles, the European supervisor promised to stick to the capital structure hierarchy: “In particular, 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. This approach has been consistently applied in past cases and will continue to guide the actions of the SRB and ECB banking supervision in crisis interventions.”
Now, let us hope that the EBS press release does its work and calms markets.
On preserving that hierarchy …
The ECB press release promises to respect the hierarchy of capital structure. However, I feel a tad uneasy because the text of the press release is high-level, whereas a similar statement by the Monetary Authority of Singapore is more precise. The ECB press release also refers to common equity instruments, which under the relevant regulation is a subset of the items that define Common Equity Tier 1. Does the press release cover retained earnings and accumulated other comprehensive income as well? There is also a reference to troubled banks, which potentially signals a desire of the EBC to extend the reach of its powers.
However, more interesting is the ECB’s firmly communicated commitment to preserving the capital structure hierarchy of banks. In the presence of the well-functioning BRRD, I am not sure what to think. The tools are there, the ECB will use them. We know that. But the focus in the press release on resolution prompts the question if the ECB will respect the hierarchy only then. How does that dovetail with the rules on writing down a CoCo bond in going concern at the pre-specified trigger point? In the end, the Contingent Convertibility feature defines these securities; and Tier 1 is going concern capital.
In all, does the ECB’s commitment imply that the whipping boy features of AT1 are now on the chopping block? How will that contribute to the loss absorption of these bonds going forward?
Who knows. As with the question about the success of the French Revolution – regarding the success of the post-GFC regulatory initiatives: it is probably too early to tell.