Bank capital: Banks brace for TLAC terms, clarity still needed
Tier-3 debt faces contractual hurdles; FSB accedes on structure disparities.
A series of leaked documents throughout the summer have provided limited clarity on the Financial Stability Board’s final thoughts on total loss-absorbing capacity (TLAC), which are set to be revealed imminently.
“We could have agreement on TLAC within the next week or so,” says one official close to the negotiations. “There has been huge progress over the last month.”
Not all countries are
at the same place in
terms of what you can
do in resolution
Publicly, however, following its plenary meeting on September 25, the FSB has merely said that “good progress was made in agreeing key remaining policies to help end too-big-to-fail”.
It is sometimes forgotten that while TLAC applies to roughly 60 G-Sibs, the minimum requirement for own funds and eligible liabilities (MREL) in Europe applies to around 6,000 banks and is due for implementation in January 2016. Unsurprisingly, resolution strategy has been top of the agenda for most European banks in recent months.
Prospects of a TLAC announcement coincided with many European banks rushing to fund at the end of September, eager to get ahead of any developments. In the last week of the month, UBS issued $4.3 billion in its first holding company issuance, BNP Paribas issued $1 billion tier-2 notes and Société Générale issued its first AT1 trade.
“People want to get on with this,” says Chris Bates, partner at Clifford Chance. “There is no lack of desire by issuers to get to the endgame. They want to say to the market – we are done, we are resolvable. But uncertainty could run on for two more years after the proposal.”
And that could still be a long way away. “This is the end of the beginning in achieving bail-in,” warns one EC official. “We might have the rules but we don’t know what their impact will be – the jury is still out.”
Banks are undeniably under pressure to be seen to be on top of what is an extremely fluid situation. “There has been constructive alertness,” according to Gerald Podobnik, global head of capital solutions at Deutsche Bank. “Whenever there is a big new concept there is pressure to immediately disclose the estimated effect of the framework.” Podobnik, speaking at the Brussels meeting, adds: “Large banks are under substantial pressure to comply as soon as possible.”
That can be tough to do when the implications of bail-in are still so unclear. At the core of any bank’s resolution strategy is loss-absorbing capital. The requirement for this has been evident since 2010 but even now banks are still struggling with the mechanics of exactly how bail-in can be achieved. It is a legal minefield.
There are effectively three routes to subordination for banks needing to issue bail-inable debt. The first is structural subordination, whereby debt is issued through a bank holding company and is subordinated to debt held at the operating company level. The second is statutory subordination, where debt is subordinated to excluded liabilities by statute.
The third is contractual subordination, whereby senior subordinated, or tier-3, debt is issued that is bail-inable but ranks above pre-existing tier-2 debt in the waterfall.
Of the three routes, contractual subordination is by far the trickiest to pursue. The path of least resistance is the one that Germany and Italy have taken. In Germany, draft legislation has been introduced to subordinate bank bonds in insolvency and in Italy the legislature has proposed to make deposits that are not covered by deposit protection super senior. The two opposing routes essentially achieve the same aim. This makes life easy for the banks, but it is not clear the extent to which other European jurisdictions might follow their lead.
Wait and see
Many might take a wait-and-see approach. “If countries decide to do things on their own, other countries will look at this and wonder what the impact will be,” says one observer. “There are always first movers. People are watching to see whether this will be an advantage or disadvantage.”
Issuing debt out of a holding company is an easy solution for banks in certain jurisdictions, such as the UK or US, but is a very tough challenge for those elsewhere.
“To move towards a holding company model in a jurisdiction that has never had one is a complex and difficult undertaking. This is not an easy thing to do. It is not an easy route,” says Podobnik. Swiss banks Credit Suisse and UBS have now both issued at holdco level this year.
This leaves contractual subordination, which sounds simple but could be very difficult.
“For bail in to work, it has to be workable in practice,” says one official. “You need contractual consent and the consent of the home regulator to bail in. This raises issues we are aware of and it is important to find a way out if we can’t find contractual agreement.”
That might well turn out to be the case for some banks. Ever since bail-in was proposed in 2010 banks have realized the need for an extra cushion of loss absorbency below senior debt and many have issued large volumes of tier-2 bonds as a result.
“Bail-in has changed the way we look at tier-2 completely,” says Podobnik. “Investors want to see an additional cushion of loss absorbing capital below senior debt.” But what tier-2 investors probably don’t want is for a new layer of loss-absorbing debt to be inserted between them and senior bondholders.
“The contractual subordination route is a problem for many banks with legacy tier 2 issues,” says Tom Grant, partner at Allen & Overy in London. “In order to achieve tier 3 we need the existing tier 2 to be capable of being subordinated to a new layer of debt sitting immediately below senior debt. Most of the time tier 2 ranks directly below senior so there is no room to squeeze in any new tier 3.” The Spanish regulator has already introduced legislation to subordinate tier-2 creditors in the statutory hierarchy.
Beside and below
Grant tells Euromoney: “We haven’t seen anything that purports to be tier 3 yet. There is still a question over how each European authority will implement the MREL requirement for each bank.”
“Investors have become very, very interested in what is below them in the capital structure and what is beside them,” says Bates. “The reaction of first takers of tier-3 could be that they demand a significant premium.”
When Credit Suisse issued its first euro-denominated holdco bond in April, the premium investors received over its opco debt was around 50bp. However, recent market volatility has seen holdco debt from issuers such as Barclays, RBS and the Swiss banks widen in secondary.
How to price tier-3 debt is just the latest conundrum that FIG investors have been presented with as a result of the implementation of the Bank Recovery and Resolution Directive.
How to price the additional risk of bail-in when the practical process itself is still so unresolved is fiendishly complicated. Existing tier-2 terms may restrict the issue of tier-3 and the very issuance of tier-3 could violate pari passu clauses in a bank’s existing debt. Banks therefore have the choice of issuing tier 3 debt that ranks pari passu with tier 2 even though it is contractually different or to undertake a liability management exercise to amend the terms of its tier 2 bonds and exchange them.
Resolution is not the new insolvency. It is just an option
Axel Kunde, Single Resolution Board
“Being a debt investor in banks has become a lot more complicated,” says Bates. “And being a debt investor in Europe is more complicated than being a debt investor in the US.”
This is because managing capital for multi-bank entities is much more complicated and there are far more such banks in Europe than there are in the US. One of the main discussion points following the leak of the draft TLAC term sheet on August 24 was the addition within the section on minimum TLAC of language aimed at smoothing any disparity between TLAC quantums required by banks with single point of entity (SPE) or multiple point of entry (MPE) structures in resolution. The FSB seems prepared to allow MPE banks such as HSBC and Santander to now count any excess internal TLAC held at subsidiary level to their TLAC requirement at parent level. Essentially, the G-Sib’s home and host regulators will determine where the TLAC deduction will take place. The purpose of the new drafting is to try to iron out any differential between TLAC quantum required that is due to banks’ differing resolution strategies: SPE or MPE.
“Not all countries are at the same place in terms of what you can do in resolution. There can therefore be a material difference between the effectiveness of internal and external TLAC,” says Etay Katz, partner at Allen & Overy. “MPE banks have been lobbying to get recognition for excess internal capital to count towards their external requirement. They have argued that the holding company should be allowed to take account of this in the total TLAC calculation. It is quite extreme for a banking group to be entirely SPE or MPE, most are somewhere in between. But the impact of TLAC is meant to be neutral regardless of whether you are SPE or MPE.”
In a speech on July 17, Andrew Gracie, executive director of resolution at the Bank of England, made the same point. “Many SPE firms will have around the periphery subsidiaries that could be resolved separately or allowed to enter insolvency without adverse consequences for the rest of the firm. And many MPE banks are, in practice, collections of SPE strategies.”
What is sometimes lost in the discussion of how and what to bail in is the fact that just because an instrument is not TLAC or MREL-eligible, does not mean that it will not be bailed in. And it also doesn’t mean that investors won’t be hit.
“Even if liabilities are not bail-inable, that doesn’t mean that investors won’t suffer losses,” says Katz. “Structured notes, for example, will obviously suffer substantial losses if a bank is in trouble.”
“Both MREL and TLAC are minimum requirements,” Axel Kunde, head of resolution planning at the Single Resolution Board, observed at the Brussels seminar. “It doesn’t mean that bail-in will be limited to MREL eligible liabilities.”
He also points out that even with the BRRD in place it does not mean that banks will always be resolved. “Resolution isn’t for everyone,” he says. “For many of these 6000 banks if there is a problem it might be acceptable to liquidate them. Resolution is not the new insolvency. It is just an option if insolvency doesn’t deliver the required outcome.”