2016: New headwinds hit
The year began inauspiciously, with regulatory pronouncements casting a pall of fear and uncertainty over an AT1 market already spooked by macroeconomic news. We asked leading market participants how they expect key narratives behind the volatility to unfold and what strategies they are adopting in light of the latest developments.
Participants:
Wim Allegaert, general manager, finance, KBC Group
Frédéric Baudouin, head of regulations, liquidity and solvency, financial management department, Crédit Agricole SA
Aravind Chandrasekaran, senior analyst, Camares
Antoine Cornut, CIO, Camares
Doncho Donchev, capital solutions, debt capital markets, Crédit Agricole CIB
Nadine Fedon, global head of funding, Crédit Agricole SA
Stéphane Herndl and Dung Anh Pham, banks analysts, credit research, Amundi Asset Management
Vincent Hoarau, head of FIG syndicate, Crédit Agricole CIB
Olivia Perney Guillot, senior director, financial institutions, Fitch Ratings
Isabel Rijpkema, capital management, Rabobank
Tim Skeet, board member, International Capital Market Association, and chairman of the ICMA Asset Management & Investors Council bail-in working group
Neil Day, managing editor, Bank+Insurance Hybrid Capital
Note: Participants kindly responded to questions in the second half of February as market and regulatory positions were developing — please see the news section for updates on some of the key developments.
AT1 AND THE SUBORDINATED DEBT MARKET
Neil Day, Bank+Insurance Hybrid Capital: The correction of the AT1 market intensified at the end of January. What have been the main drivers of the turmoil?
Amundi: In our view, a number of factors were at play. These included a flow of regulatory publications — the EBA’s opinion on the SREP, the SSM’s subsequent publication — that indicated the bar had been raised further in limiting banks’ ability to pay coupons on AT1s, but also left room for varying interpretations as to the treatment of Pillar 2 capital requirements. There was also disappointment on fourth quarter earnings from the equity investor community, which further exacerbated fears that some banks may be unable to make AT1 coupon payments.
Until last year, the AT1 market had been supported by the search for yield of some investors and the highly accommodative central bank measures. The tensions observed in recent weeks are the first shock to affect this not yet matured market and some investors are likely trying to adapt.
Vincent Hoarau, Crédit Agricole CIB: From a pure market perspective, AT1 indeed in February suffered its worst performance. Cash price moves were exacerbated when negative headlines surrounding Deutsche Bank hit screens. The price action has been phenomenal, with losses of four full points in a single trading session in some particular names. There were some forced sellers in the market. And in such circumstances investors who have taken significant mark-to-market hits are obliged to sell while there is nobody on the other side of the trade. In the meantime, in financials’ equity, people were selling on hard facts — i.e. global growth and oil concerns, negative interest rates, etc — that affect bank earnings. This cocktail encouraged the downward spiral in financial institution-related securities, whether that be equities, senior, Tier 2 or AT1. To cut a long story short, the collapse in AT1 was strongly correlated with equities and amplified by the negative convexity. Meanwhile, nobody felt ready to do some bottom-fishing given the too concentrated investor base.
But the sector should rebound. From a pure fundamental perspective it looks extremely cheap versus high yield, while — from a capital standpoint — banks have never been so robust.
Antoine Cornut, Camares (pictured below): On AT1 the main issue has been the complexity of the product, especially after the EBA announcement. I’m not sure that everyone has really understood what the criteria are for the coupon to be paid or not.
The consensus at the start of 2016 was that the AT1 market was an asset class to own, especially after the performance of last year. It contributed to the bad positioning of investors whereby effectively everybody is long and nobody is short, meaning that only small selling needs create some very large movements in the AT1 market.
At the same time, I note that the AT1 market has outperformed the equity market. Is the equity market getting tainted because of the AT1? Probably. I don’t think equity investors fully understood the way AT1 works either, the way that coupons and dividends interact.
So I think all this amounts to a lack of transparency and understanding, which is creating a lot of confusion and even more repricing.
Since the AT1 market started there has been a lack of distinction between issuers. It was very well behaved, but I don’t think the individual credits were actually taken into account. As soon as people have concerns regarding the capacity of one bank to pay a coupon, we’ll see much more the differentiation between credits. Hence the headlines around a few names — Deutsche Bank, UniCredit or Popular. In that respect, it is good news for the AT1 market to the credit consequences of what you buy. I think that’s a positive development for this asset class.
Doncho Donchev, CACIB: As an investor you are at the coalface of the market, and I fully agree on the point about idiosyncratic risks, which must reflect individual credit quality — with this being more pronounced in AT1, because it is the highest beta debt product in the capital structure of banks. And you are certainly right about some market participants perhaps not having such a detailed understanding of the product, which is complex.
Indeed, we have seen an unfortunate confluence of events, with the other point being pronounced regulatory uncertainty, apropos the EBA announcement. The whole Pillar 2 process has been quite unclear and full of contradictions for investors and hence troubling for the asset class, and has certainly also been a contributing factor to the sell-off. How do you view that?
Cornut, Camares: This is going to make it very complicated for new people to actually look at the AT1 market, and I would argue that the existing buyer base is pretty much full on the product, or close to being full. I think raising new money for the product is going to remain challenging.
Tim Skeet, ICMA: There has been movement across all the classes of bank paper, especially the AT1. This may prove to be a temporary correction, and markets may well stabilise again. Nevertheless, the volatility has once more forced the debate on where spreads should be for investors in bank paper given the heightened levels of risk under bail-in
The events of late 2015, added to current market noise, suggest that there is an unpredictable element of risk that has spooked the market. There appears to be a lingering suspicion that investors are being asked to take the risk of a bank not fulfilling its capital requirements in a variety of ways — the immediate focus perhaps being on potential non-payment risk on AT1 instruments.
It is unclear how exposed investors might be to this risk. Issuers can certainly claim they are rock solid, but ultimately this is the regulator’s call. The public debate between Governor Carney in the UK and John Vickers, author of the Independent Banking Commission Report, for example, suggests that there may not yet be a real consensus on capital levels. This potentially exposes investors to the risk of moving goalposts.
Many of the terms imbedded in the original AT1 instruments appear superfluous and outdated in today’s market. Reset levels and triggers look irrelevant given market moves and capital requirement levels.
It is likely that the European market will be fragmented, with differing perceptions of risk depending on the perceived robustness of individual banks and local systems. Outstanding levels of legacy NPLs will have to be examined.
Day, BIHC: In the current circumstances, what are the main parameters/metrics you use in the valuation of the AT1 instrument?
Amundi: We consider the risk of coupon skip to be the main risk, given our view on fundamentals and as it would materialize ahead of a conversion/write-down of the principal, barring any exceptional case. The main parameters we consider are therefore the firm’s earnings generation capacity and volatility (e.g. stability of earnings, litigation risks), the headroom to a breach of the Combined Buffer Requirement (CBR) but also the capacity to service AT1 coupons in case of restrictions on Maximum Distributable Amounts (a function of the amount of coupons and the firm’s dividend policy).
For now, the market’s interpretation is based on the banks’ disclosures. But the European regulatory bodies have yet to confirm their position on a number of critical aspects, such as the treatment of AT1/Tier 2 instrument shortfalls and whether these do reduce the headroom to a breach of CBR.
Cornut, Camares: What I find amazing is the way that the sell-off in AT1 is actually lower than the sell-off in Lower Tier 2 securities when adjusted by beta — if you look at Deutsche Bank, for example — its AT1 securities have traded lower on par with the sell-off in Lower Tier 2, so the price differential between the Lower Tier 2 and the AT1 basically is not that big. Are you meant to be buying AT1 securities today at 75 cents or Lower Tier 2 securities at 84 cents? Lower Tier 2 where there is no uncertainty regarding payments or final maturities, an asset class that is the core holding of many asset managers and insurance companies? I don’t think valuing the recent sell-off in AT1 is as trivial as some people would want to portray it.
Day, BIHC: What importance do you give to the reset spread levels after the recent significant market correction?
Amundi: The market is seemingly pricing a growing risk of AT1 extension beyond their first call dates. This repricing is arguably driven by the more challenging macroeconomic environment and the resulting lower earnings expectations, as well as more demanding solvency requirements. Should this repricing become the norm, this would maintain yields higher in the longer-term for AT1s, as investors would seek compensation for extension risk. And the higher the incremental cost of issuance, the greater the likelihood that issuers think twice before calling an AT1.
Dung Anh Pham, Amundi
Hoarau, CACIB: AT1s have a perpetual structure with a call option, and the extension risk is more pronounced when the instrument trades well below par with a low and “off market” reset spread at the call date. This implies a “negative convexity”, with a greater probability of extension risk. This was another source of the downward spiral in the asset class, because for many securities the first call dates are just a few years away. Investors trade on the assumption that those bonds won’t be called. So basically, low reset spreads, i.e. negative convexity, leads to further weakness in valuations in a distressed situation. So yes, reset spreads are a very hot topic in AT1 valuations, and we will have a two tier market in that regard as soon as the AT1 sector reopens in primary.
Day, BIHC: What could drive a rebound of the segment? What would make you open to buying new AT1 issuance?
Cornut, Camares: I just want to understand what I am buying. What are the rules for the payment to be made or not? Some issuers don’t even publish the amount of MDA and ADI that they have.
Ultimately there is a risk-reward conversation to be had. For some issuers who are seeing their AT1 bonds trading 10 points lower purely because of the sell-off, I would be happy to buy now. If you look at some of the UK names or some of the Swiss names, for example, I find the conversation to be a little easier. I’m not sure that they would be interested in issuing — I don’t think they want to pay a 10% coupon yet — but some secondary levels are attractive. It’s purely a question of risk-reward at the end of the day.
For some of the names where there are still some concerns regarding the amount of disclosure or what you are truly buying, making an investment decision is far less easy.
Amundi: We would welcome greater clarity from regulators on the treatment of Pillar 2 for European banks, as discussed earlier. Furthermore, a smooth phasing-in of the recently announced requirements would also likely provide some relief on the AT1 segment, albeit potentially at the expense of the rapid capital build-up that is being pushed by regulators.
We remain constructive on the AT1 segment, but we remain selective on the fundamentals and with regards to relative value.
Hoarau, CACIB (pictured below): Clarity and the softening of regulatory constraints will certainly help, but this is not on the agenda of regulators. Interest rates and yields across the board remain extremely low. So we should reach a bottom simply because the correction in the security is overdone in terms of relative value. Elsewhere, any improvement in the global market backdrop and market sentiment will re-boost the segment, which is highly driven by the equity market. Central bank messages and actions, and less negative economic data points could be a short term catalyst of a rebound and retracement of January losses.
SREP, PILLAR 2 & MDA
Day, BIHC: What is the long term impact of the current concerns surrounding SREP and potential coupon restrictions?
Wim Allegaert, KBC (pictured below): The uncertainty over coupon risk hampers a correct pricing of AT1 risk while the uncertainty regarding required capital levels going forward makes capital planning a complex task. These uncertainties also complicate matters on portfolio fit for these products for the money management firms. The recent EBA clarification removed the uncertainty whether MDA is only applicable under Pillar 1 or also under Pillar 2, and although coupon risk — strictly speaking — is higher since MDA also applies under Pillar 2 now, the coupon risk actually has always been there, since AT1 coupons are anyhow fully discretionary, and investors have always clearly understood this item and taken this into account. The recent turmoil is hence not linked to this in my view.
Amundi: As the AT1 market develops and banking solvency requirements continue to be raised further, investors are increasingly factoring in the risk of a first-time AT1 coupon skip. This risk was exemplified by the recent movements in the AT1 markets on concerns that a coupon may be skipped due to insufficient Available Distributable Items. Over time, this may push some investors to become more selective, all the more so as regulatory publications remain subject to a number of interpretations and the macro environment remains fragile.
Isabel Rijpkema, Rabobank: The disclosure of SREP requirements reduces uncertainty for investors and creates more transparency, which is positive. The recent market movements made it clear that investors are concerned about the risk of coupons not being paid. Taking into account the fundamentals of the bank you are investing in will be an important assessment for investing in this type of product.
Skeet, ICMA: Investors will go on asking for transparency, predictability and fairness in how they will be called upon to take the pain of a restructuring or capital rebuild. As noted earlier, there is still a lack of clarity over the precise levels of capital tripwires being imposed on banks and how they are measured. The EBA’s Adam Farkas rightly asks that investors are told what they are and how they are met. Investors have to be able to measure and price their risk.
Any suggestion of moving goalposts, capriciousness or political impact on regulatory decisions for headlines rather than hard economic reasons will worry investors further. There is a need to a level playing field.
Investors will continue to ask for the ECB to be open and accountable. The ICMA investor bail-in working group was set up to provide a platform for investors to work with regulators and engage in a constructive dialogue to ensure that regulatory policy fairly takes into account investor needs. Ultimately investors and regulators will want to work together to ensure risk is fairly and correctly priced under the new rules.
Naturally regulators will have to accept that this may result in some financial institutions becoming shut out of certain types of funding from time to time. This is part of market discipline, but regulators must also take into account that harsh handling might inadvertently lead to loss of confidence in an otherwise sound institution, potentially leading to a death spiral.
Market speculation and rumour cannot be regulated away. But the more investors are confident that they know the risk and are assured that regulators will be consistent, the hope should be that markets will stabilise more rapidly after scares and irrational fears.
Consistency between different regulatory measures and points of intervention is crucial. Any suspicion that regulators may not be predictable, consistent or transparent would have a hard impact on the market.
Day, BIHC: What is the impact of the EBA classification on the interaction between Pillar 2 and MDA restrictions on AT1 coupons?
Rijpkema, Rabobank (pictured below): As already assumed by many European market participants, the fact that the EBA has clarified that Pillar 2 requirements “sit” below the combined buffer requirement increases transparency, which is a positive. On the other hand, there is not a real level playing field in Europe with regards to Pillar 2.
Allegaert, KBC: It is interesting that EBA recognizes that a cancellation of AT1 coupons does not support access to the financial markets. This is very important as market access is key as one of the measures to remedy the solvency situation. It remains to be seen under which conditions AT1 coupons indeed are actually cancelled.
Amundi: It has had a nuanced impact. On the one hand, it provides more information (and sometimes some reassurance) for those issuers that published their SREP requirement. But on the other hand, the EBA’s opinion remains subject to a number of interpretations. Equally importantly, when it comes to Pillar 2 and MDA restrictions, supervisory authorities failed to follow a unique approach, with the SSM stating it would follow the EBA’s opinion (for now…) and some national supervisors saying they would not abide by it. This uneven playing field makes it more difficult to compare AT1s across jurisdictions, in our view.
Cornut, Camares: It is unfortunately not very clear. When you look at BNP’s earnings presentation, for instance, they noted that they were not including Pillar 2 in their calculations because apparently the ECB told them not to. This is not what the EBA implied in early January. So are you meant to trust what the ECB tells some banks privately — don’t worry about it — or are you meant to be following what is a public document from the EBA? This complete lack of transparency makes investing in this asset class a difficult task. Thankfully not all AT1 have been issued under the ECB/EBA framework.
Donchev, CACIB: It’s definitely not an easy question, and we will see how the clarifications turn out. Clarity and transparency about the application of the rules is key in this context and regulators should be in a position to deliver on these key items. A level playing field is also a key concern — how can Pillar 2 be binding for MDA in the SSM from the outset, whilst in Sweden, a member of the EU, it can only become eventually binding upon a “formalisation” of the Pillar 2 requirement? What about deficits on AT1/Tier 2 under Pillar 1?
Day, BIHC: What disclosure requirements are you anticipating from the ECB in relation to Pillar 2/SREP?
Amundi: Following the publication of the SREP requirement for a number of banks within the Eurozone, we would expect others to follow, absent any legal prohibition to do so.
Notwithstanding this, supervisors have to balance the much-needed public disclosure in this respect with maintaining room for manoeuvre for dealing with stresses, as evidenced recently. Hence, we would not expect the SSM to push banks to publish more than what has been disclosed thus far. Instead, we would expect the SSM to communicate more on how they would deal with SREP requirements and breaches, practically speaking.
Allegaert, KBC: KBC already disclosed SREP before the EBA paper regarding Pillar 2, given our assessment that this is stock price-sensitive information that we have to communicate.
Rijpkema, Rabobank: Rabobank disclosed SREP requirements in a press release dated 18 January 2016.
Day, BIHC: Has the recent publication of Pillar 2 requirements affected your perception of the sector? Does the SREP requirement inform ratings?
Olivia Perney Guillot, Fitch (pictured below): For securities with going-concern loss-absorption characteristics, it is not so much loss severity, as the probability of non-performance relative to hitting the point of non-viability that is the biggest rating variable. Activation of a going-concern loss-absorption feature does not mean a bank has failed, but is treated as “non-performance” at a security-level rating.
We believe fully discretionary coupons to be the most easily activated form of loss absorption. Although in many instances it is unclear at exactly what point a bank will be “required” by its regulator to omit coupons, this is likely to arise (and may even automatically kick in) when a bank is within a capital buffer zone.
Bank hybrid securities with fully discretionary coupons are likely to exhibit the widest degree of notching in a bank’s liability structure, irrespective of any other features.
For issuers with investment-grade Viability Ratings, Fitch’s base case will be to notch down by three notches for incremental non-performance risk. However, more notches may be assigned where Fitch has particular concerns, for example over a bank’s flexibility in avoiding falling into a capital buffer zone whether due to internal (e.g, capital management policies) or external (e.g, regulatory/SREP/Pillar 2) influences.
Day, BIHC: How should one go about deciding on the appropriate size of the management buffer?
Aravind Chandrasekaran, Camares: We don’t even know what the target is pre-the management buffer to have a view on what the management buffer should be for a generic issuer. It’s jumping the gun, right? We don’t even understand how Pillar 2 works, because there’s no clarity, we don’t know whether there’ll be relevant grandfathering… Unless you have a view on what the current surplus or deficit is, it’s very difficult to work out what the excess you need on top of that is.
Rijpkema, Rabobank: Rabobank capital strategy focuses on a minimum CET1 ratio of 14%, subject to regulatory requirements. On a fully phased-in basis, this provides a management buffer of at least 1.5%, which should give sufficient comfort, especially given that Rabobank has always been profitable and showed a stable earnings/profit pattern. Current SREP (9.5%) plus phased-in Systemic Risk Buffer (0.75%) requirement adds up to 10.25% with a CET1 ratio of 13.5% (FY 2015), so this provides a comfortable buffer over and above the minimum requirement.
Allegaert, KBC: The appropriate size of the management buffer should take into account some or all of the following elements: the impact of adverse economic conditions; the capital needed to absorb regulatory uncertainty (e.g. Basel IV); and non-organic and organic growth expectations versus expected profit. This is a dynamic exercise, taking into account timings of events and different scenarios (fly, float, sink scenarios). In the end, it remains a risk appetite decision.
Amundi: Sizing the management buffer is likely to be based on several aspects. To start with, one should measure existing capital and earnings generation capacity against the level of capital required. This management buffer needs to factor in the complexity of the group’s structure and its risk profile. It should also be compared to that of its peers.
Frédéric Baudouin, Crédit Agricole: Regarding Crédit Agricole, at group level our management buffer above our Pillar 2 is quite high, more than 350bp. At Crédit Agricole SA level this buffer is smaller, but we have committed to reach a buffer of 150bp by the end of the year, which we think is enough to manage the prudential constraints on Crédit Agricole SA in an environment where Crédit Agricole benefits from the support of the regional network. Please note as well that Crédit Agricole group must respect its Pillar 2 constraints from 1 January, but the Pillar 2 constraint for Crédit Agricole SA will apply only from 30 June this year.
Chandrasekaran, Camares (pictured below): It’s quite complicated this whole concept with the two entities that you have. From the outside it’s very difficult to work out exactly where the binding constraints are in terms of each entity, and how much flexibility you have to address a deficit in one place versus a surplus in another place. You therefore have to rely on some degree of management or issuer flexibility in being able to ensure capital fungibility between the two levels.
Nadine Fedon, Crédit Agricole: We have been communicating quite regularly on that topic when disclosing our quarterly results — for both equity and bond holders — so I think the communication is perfectly clear about the binding constraints for the two entities. In our press release issued last December, we disclosed that Pillar 2 at Crédit Agricole group level includes the G-SIB buffer which is phased in resulting in a 9.75% requirement, and at Crédit Agricole SA level we do not have a G-SIB buffer, and that’s why we have a 9.5% requirement. So it’s perfectly clear in our press release.
Donchev, CACIB: When we talk about the management buffer, it’s about the buffer above SREP — the Pillar 2 requirements or any binding constraints on AT1 coupons, which is probably one of the main criteria investors look at.
This should depend on the credit profile of the institution and what risk it carries. Technically, a bank that is bigger, more systemic, more complex, more diversified across jurisdictions and so forth, and where there are difficulties with capital flows across it — especially banks that also have large investment banking operations — should have a higher buffer. For banks that are smaller and have a more clearly defined model, the buffer could be smaller, from a certain perspective, because they are clear and easier to understand for investors. But this comes with a flipside that could be a negative point, which is that if you are less diversified, then obviously you have less diversification benefits. So it’s difficult to say where the buffer should be. This is an issuer-specific discussion.
But what we have clearly witnessed since the SREP decisions of 2015 started to become public is that anyone who has pure buffer of less than 100bp in their reported ratio versus the fully-loaded SREP target, including any buffers that can apply on top, has suffered disproportionately in terms of AT1 price with investors taking fright.
Chandrasekaran, Camares: My point is that if you say you are 200bp higher but you still need to issue 2% of Lower Tier 2 and some AT1, that buffer means nothing, because we don’t know what the constraint on coupon payments is.
Donchev, CACIB: Yes, absolutely. Are deficits in either AT1 or Tier 2 additive on top of the SREP requirement plus fully-loaded buffers? This is a question that only the ECB can answer — or the ECB, the EBA, I’m not sure who. But clearly we need clarification on this, absolutely. And communication on this point has been lacking. There is no clear communication from issuers, probably because they also cannot make such communications or have judged that they cannot. All they say is that, well, actions speak louder than words: they’ve paid the coupon and they show you the SREP requirement versus CET1, and they ignore AT1 or Tier 2 deficits for the time being. Unfortunately, yes, I agree with you on that point.
Day, BIHC: Are AT1/Tier 2 deficits additive on top of the SREP
Amundi: Based on the disclosure of a number of banks for their fourth quarter results, this seems not to be the case. But absent any official clarification, we ought to err on the side of caution.
Allegaert, KBC: Reading CRR/CRD IV and the EBA opinion 16/12/2015, the answer is clearly positive. But this does not yet show up in the SREP letter from the ECB, which is limited to CET1 requirements.
Rijpkema, Rabobank: The uncertainty around this point was not helpful for the AT1 market, but it is good to see that for 2016 at least it will not impact the MDA intervention point. For Rabobank this is not an issue as Rabobank has fulfilled its AT1 and Tier 2 requirements with 16.4% Tier 1 capital and a Total Capital Ratio of 23.2%.
TLAC vs MREL: SIZING THE CHALLENGE
Day, BIHC: How do you expect MREL and TLAC to be implemented in the EU?
Allegaert, KBC: The recent Information Note for the European Commission created some uncertainty, on top of the existing one given the different national initiatives. Clearly more time will be needed to come to a common platform for the Member States. Ultimately, this uncertainty may slow down the issuance of MREL and TLAC-eligible instruments by EU banks.
Rijpkema, Rabobank: MREL requirements have not been defined yet, but are expected to be compatible with the TLAC requirements for G-SIBs. MREL requirements could in a conservative scenario end up higher than TLAC requirements. Rabobank is not directly impacted by the TLAC requirements as we are not a G-SIB, but we expect TLAC requirements to be included in the MREL assessment.
Amundi: Unlike the Basel capital requirements, TLAC and MREL introduce bail-in tools for a wider variety of debt classes and therefore need to interact with local insolvency laws. This explains why thus far we have seen diverging proposals put forth by European Member States, in order to transpose BRRD in a way that would make domestic banks TLAC-compliant. In addition, the proposals made by each Member State clearly appear to be tailor-made to fit best with their banking systems’ structures and funding mix.
Despite recent discussions held at the European Commission, the prospects of a convergence appear minimal, for now.
Market participants meanwhile expect a convergence of MREL and TLAC rather than MREL surpassing TLAC.
Skeet, ICMA: Whatever is promulgated needs to work with every country’s insolvency laws. Better cooperation and communication between all the regulators bodies and governmental agencies is imperative to avoid regulatory clash, overlap and inconsistencies. It may not be possible for TLAC to find a single common definition, but the basis has to be fair, respect the waterfall of creditor rights and be transparent.
If the level is to be set high, regulators must take full account of refinancing risk and investors’ appetite and line availability. They should not be complacent about this.
What investors expect is as clear as possible definition of where the triggers and tripwires will be set for each bank by whatever measures the regulators apply. Transparency on this is the key to confidence.
Day, BIHC: Are we heading towards MREL of 25%-30% RWA required capital?
Donchev, CACIB: I think we have clarity on where TLAC will be, so 16% in 2019 rising to 18% of RWAs in 2022, with the applicable buffers on top, so you have the capital conservation buffer on top of that, so 2.5%, and the systemic buffer, and potentially any other buffers which may apply, so we are talking… 16% plus 2.5%, takes us to 18.5%, call the systemic buffer somewhere around 1%, so between 19.5% and let’s say around 22%, 24% depending on the bank.
Obviously, the sizing of MREL is under discussion, notably the now publicly acknowledged now difference in opinions between the EBA on one side and the Commission on the other side, on the RTS for MREL. MREL has a certain specificity such that it doubles up Pillar 1, plus Pillar 2 requirements, plus certain buffers… there’s upward, there’s downward adjustments, but as said, there is no clarity about how it may work and how high the capital requirements may be. It could be something that goes more in the region of 25%-30%, so clearly there is also regulatory clarity needed on that point.
Amundi: MREL of 25%-30% of RWA would imply 10.8% of total liabilities in own funds and eligible liabilities while the market still expects a minimum of 8%. Although the MREL level will be set on a case by case basis, the gap seems to be too large.
Allegaert, KBC: You can build up a metric that comes to those 25%-30% numbers. We anticipate, though, that European Resolution Authorities will not aim for these ratios in the years to come and focus first on phasing in 8% MREL (as a percentage of liabilities) by 2019. MREL in the 25%-30% of RWA range would not necessarily accommodate a recovery of the European economy which the ECB aims to achieve through its expansive monetary policy. Market participants are very preoccupied with the whole MREL-TLAC debate.
Rijpkema, Rabobank: Anticipating TLAC/MREL, Rabobank targets a total capital ratio with a minimum of 25% of RWA. MREL requirements will be determined on a bank by bank basis but we see the 25%-30% as a potential range where requirements could end up.
Baudouin, CASA (pictured below): I think it is important to remember that Crédit Agricole’s MREL ratio was already 8% excluding eligible senior debt at 30 September, which means it would be possible for the group to have access to the Single Resolution Fund in case of resolution. I agree there is still uncertainty about regulation and SRB decisions regarding the implementation of TLAC and MREL in the European legislative framework. The Commission is likely to make a legislative proposal on TLAC/MREL implementation in 2016. We strongly hope that the EU legislative framework (BRRD) will be amended consistently with the TLAC international standard adopted last November by the G20 and that the TLAC/MREL duplication will not remain. This overlap between two ratios that have the same goal confuses investors and needs to be clarified as quickly as possible. The market disturbances caused by the recent Commission proposal on Tier 2 have shown the need for a clear and quick proposal of TLAC implementation.
BRRD & NATIONAL LAWS/EU COMMISSION PROPOSAL
(See section on France at end of article for more)
Day, BIHC: There is a proliferation of national solutions regarding the role of senior unsecured within MREL/TLAC. What does it mean for the integrity of the European bank senior unsecured debt market? A return to pre-euro days? A complexity premium?
Perney Guillot, Fitch: EU countries are adopting diverging paths to establish workable foundations for bank resolution, but coordination is important because national laws and the characteristics of each country’s banking system are being married with EU-wide regulation and Banking Union.
The EU authorities are encouraging countries to explore an EU-wide solution to remove obstacles that could hinder an effective bail-in in the event of bank failure.
Recent proposals for legislative changes in Germany, France and Italy provide clues about the likely paths for these countries. Germany is ensuring TLAC-eligibility of existing senior bank debt by subordinating it to all other senior liabilities from 2017. Italy is considering full deposit preference whereby senior bank debt still be TLAC-eligible if equally ranking excluded liabilities are small. France is proposing to create a new class of senior bank debt, “non-preferred”, which will be subordinated to existing senior debt and bailed in more quickly in the event of resolution under BRRD. These options contrast with the UK, which is requiring structural subordination through holding company structures for the country’s large banks, as is the case in Switzerland and the US.
Other EU countries have provided little detail about their preferences. All proposals will have to consider both existing national insolvency laws and the specific features of a country’s banking system. We believe resolution authorities will maintain a high degree of flexibility when dealing with each resolution case for deciding which liabilities will be bailed-in or not transferred to a bridge bank, subject to the overriding “no creditor worse off than in liquidation” principle. Clear and credible agreed group-wide resolution plans would smooth the process and avoid the additional risk of unilateral intervention by host regulators.
The BRRD came into force on 1 January 2015, but some EU countries are still determining what legislative changes, if any, need to be made, while others are already amending national insolvency laws and changing bank group structures.
Skeet, ICMA: As noted earlier, a proliferation and differing standards of subordination language might be inevitable. The hope is that investors will face consistent and predictable measures.
Complexity is already posing enough problems. Some investors question the need to increase complexity through the introduction of new instruments whose status may not be clear versus other instruments and whose risk may be difficult to price.
Amundi: The proliferation of national solutions has definitely rendered the senior unsecured debt market more difficult to assess. Unlike pre-TLAC, investors now need to understand the implication each national law (i.e. of each transposition of the BRRD into national insolvency law) has on their position in resolution or liquidation and their loss severity.
One also needs to understand the interaction of various insolvency laws and each resolution strategy. Indeed, for now one can doubt whether the Single Resolution Board would be able to impose the bail-in of senior unsecured debt without risking breaching the BRRD’s “no creditor worse off” principle, if the resolution entity has debt written in both German and French insolvency laws, for instance.
Notwithstanding these fundamental questions, we expect the market to adopt some simple standards for pricing each solution, similarly to what has been observed for the senior unsecured debt issued by UK and Swiss holding companies, with a 40bp-70bp premium across the board, irrespective of the issuers’ fundamentals and of the overall capital structures.
Rijpkema, Rabobank: Rabobank’s capital strategy is focused on protecting senior unsecured for the unlikely event of bail-in and Rabobank has therefore built up a comfortable capital stack and plans to target a Total Capital Ratio in the range of 25%-30%. With regards to national solutions, Rabobank has published a position paper on the “German” solution in which it does not support such a proposal as Rabobank is in favour of building of a stack that protects senior unsecured debt.
Rabobank supports the “French” solution as it gives all institutions the flexibility to match their capital stack to their business model and at the same time make sure that no creditor is worse off. A European solution is preferred to create a level playing field.
Chandrasekaran, Camares: There was the European Commission document saying that you basically need to satisfy all the requirements with Lower Tier 2, and I think the EBA dissented to it. So, even from the perspective of how much Lower Tier 2 an issuer needs to have, there is actually no clarity, because ultimately the European Commission is responsible for putting into legislation the TLAC and MREL regulations, and if their proposal is that it all needs to be Lower Tier 2… It’s highly unlikely to happen, but there is a large amount of uncertainty about how much Lower Tier 2 banks need.
Donchev, CACIB: We have seen three different routes: the German route, where they take the senior unsecured, existing and future, and from 1 January 2017 it ranks below all the other problematic liabilities like derivatives, corporate deposits and so forth; there is the French route, which will effectively split senior unsecured, with the old stock of senior not changing in the ranking and remaining pari passu, and new senior unpreferred instruments that are in essence economically like Tier 3 but still called senior; and the Spanish with the Tier 3 solution. So we have this proliferation of national solutions, and then you have the European Commission coming out saying actually we would like to have Tier 2 as the liability of choice for eligible liabilities and ignoring what has been happening in the individual Member States.
Again, I can only agree with what others have been saying, that we need more regulatory clarity and we need to see in which direction this journey will go. Unfortunately, now as an investor you don’t know — in terms of pure demand and supply dynamics — how much Tier 2 will be needed on a systemic level versus senior unsecured, and when you invest in senior unsecured you may not know what you are investing in.
Perney Guillot, Fitch: Although several EU countries are adopting diverging paths to establish workable foundations for bank resolution, it is also possible that European authorities might propose a harmonised solution at the euro area or even EU level as an alternative to individual national approaches. This could even go as far as revisiting the MREL and/or depositor preference aspects of BRRD, in order to harmonise TLAC/MREL bail-in buffer approaches for all member states.
Day, BIHC: In the absence of law, are we seeing individual issuers draft their own strategy?
Allegaert, KBC: The intention of having a clear hierarchy for creditors is obviously a good and necessary one. The multitude of different systems doesn’t necessarily add to clarity and puts a challenge on portfolio fit, grandfathering risk and spread pressure.
Nevertheless, looking through this, KBC has already drafted and communicated its MREL strategy, in which the HoldCo KBC Group is positioned as issuing entity to satisfy MREL. Previously, KBC also moved up the AT1 and T2 issuance from KBC Bank to KBC Group. KBC Group will issue senior debt insofar as it is needed to reach its MREL target. The target has not yet been decided by the Resolution Authority, but it is assumed to be at least 8%, in line with BRRD. Also, our existing 17% total capital target remains. KBC Group will downstream the proceeds in subordinated format to KBC Bank. What adds to our benefit is the fact that, compared with other financials, KBC Group is a diversified holding company with investments in KBC Bank, KBC Insurance and KBC Asset Management. This diversification mitigates the structural subordination for investors in senior debt of KBC Group.
Rijpkema, Rabobank: In the absence of specific law in The Netherlands the Rabobank capital strategy targets high capital ratios to protect senior bondholders.
Amundi: In many Member States, some banks have long made public their strategy for complying with MREL. This is notably the case for banks with a high structural reliance on senior debt funding, which seek to meet an 8% MREL ratio with more subordinated instruments. These issuers may reassess their current strategy as their local insolvency laws are changed, though.
MARKET IMPACT & PRICING OF NATIONAL LAWS
Day, BIHC: What are the implications of the Novo Banco/BES senior unsecured debt transfer?
Amundi: The transfer of the Novo Banco senior debts to BES was not expected and came as a further evidence that regulators have broad powers to resolve a bank and can be unpredictable. This has harmed the relative trust of the market vis-à-vis regulators and has also partly driven the recent sell-off in the banking sector.
Stéphane Herndl, Amundi
Perney Guillot, Fitch: The Bank of Portugal’s decision to transfer almost Eu2bn of senior bonds out of the Novo Banco bridge bank and back to Banco Espírito Santo (BES), which is to be liquidated, shows bridge bank investors can face legacy retransfer risks until resolution is concluded. In our opinion, the handling of Novo Banco’s resolution process raises questions about how future bank resolutions might be dealt with in the EU, particularly because the timeframes could prove quite lengthy. The Novo Banco experience also means that any future ratings we might assign to bridge banks might be constrained to reflect legacy retransfer risks to investors until that risk is sufficiently remote.
When announcing resolution proceedings, the Bank of Portugal said that it could transfer assets and liabilities between Novo Banco and BES at any time during the resolution process, and this has now taken place. The re-transfer of the five senior bonds announced in December 2015 aimed at addressing Novo Banco’s solvency shortfall. We do not rate Novo Banco. But if we did, we would probably have viewed the retransfer of the bonds into BES as a restricted default of Novo Banco.
Cornut, Camares: Draghi discussed the case of Novo Banco in the European Parliament and made the point that BRRD can only apply to certain bonds in a case where it actually does create some kind of systemic risk for the financial system. So does he mean that some bonds that are retail-held — which can create some kind of systemic risk, being a run on the bank — should be removed from the BRRD? Does he mean that the Novo Banco example where they excluded some of the retail-held bonds will actually happen going forward? He really placed a strong emphasis on the fact that we will not do it for some bonds, some liabilities if it actually creates some kind of financial instability. So haircutting bonds held by a large asset manager like a BlackRock, a Pimco is fine because they can bear the losses; but doing it for some smaller investors who have their life savings in these bonds is out of the question? How are we meant to interpret these comments?
Fedon, Crédit Agricole: I think it is unfortunate that the Novo Banco case has given the impression that pari passu securities might be treated differently. But we think that this case cannot be taken as a general one for the Banking Union, and as you know the ECB has stated publicly that it was not involved in the decisions around the senior debt transfer from Novo Banco to BES.
Cornut, Camares: Unfortunately for us it doesn’t mean that the ECB did agree or disagree on the way Novo Banco was recapitalised in December.
Fedon, Crédit Agricole: Well, it’s not a general case.
Day, BIHC: What is the future for retail Tier 2, if there is one?
Perney Guillot, Fitch: The bail-in of retail junior bondholders in the liquidation of Banca Romagna Cooperativa (BRC) highlighted the increased likelihood that losses will be sustained by bank creditors now that effective resolution tools are available in the EU. When the Italian authorities began BRC’s liquidation in July, they decided to bail in its equity and junior debt, all of which was held by retail depositors. This first-time bail-in of retail creditors in Italy shows how bank resolution procedures have changed under BRRD. The initial plan was to use funds from Italy’s Deposit Guarantee Insurance Fund to make up the shortfall between BRC’s assets and liabilities. But the European Commission ruled that this would constitute state aid. Bail-in followed, while retail insured depositors, protected under BRRD, suffered no losses. Despite the bail-in of junior debt, no loss was ultimately suffered by the retail bondholders as the Italian mutual sector’s Institutional Guarantee Fund decided to reimburse them in full to preserve the reputation of the sector. But junior creditor losses are far more likely under BRRD and we believe this will reduce retail investors’ appetite for this type of bank debt instrument. Also, bailing in subordinated debt can be done outside (i.e. before) formal resolution action on a bank.
Amundi: In some countries, notably Italy, retail Tier 2 funding is significant and imposing a ban on retail Tier 2 placement would be detrimental to the system. However, as the resolution of some Italian banks has recently shown, regulators need to make sure that the instruments are only sold to retail customers who clearly understand the risks and are capable of absorbing potential losses. Regulators will therefore need to ensure the placement of regulatory capital within retail networks does not create any material risk of mis-selling, notably by carefully monitoring the retail investor base, ensuring fair pricing and strong legal documentation.
THE SHADOW OF BASEL IV & IFRS 9
Day, BIHC: Mario Draghi also told the European Parliament: “No Basel IV” — what do you make of this? What impact might RWA inflation otherwise have on absolute required levels of capital? And IFRS 9?
Cornut, Camares: I think it is difficult for Mario Draghi to say publicly there will be “no Basel IV” when people are talking about Basel IV. He cannot say in private conversations to dismiss the EBA recommendation and that the ECB will take a more pragmatic and lenient view, and then say publicly that he will follow EBA recommendations (which no-one really understands). He is trying to please everybody. People are concerned that there will be more capital requirements for European banks — which is what the equity market is pricing in — so he’s trying to create a circuit-breaker for the recent sell-off in banks’ equity on the basis that no more capital is needed — which would be positive. But it goes against the work done by the Basel Committee, so maybe he means no Basel IV for the next two years, which is probably true, and then Basel IV when the market can actually cope with it. But he cannot please everybody. Clearly there will be Basel IV one day.
Baudouin, CASA: Our view is that you have a lot of work underway on the revised standardised approach in the Basel capital framework for credit markets and counterparty risks. All that work is likely to be implemented by 2019, and indeed may have an important impact on the RWA calculation. We are participating in different consultations, and what we can see is that you have a lot of new different rules — like the application of input and output floors based on the standardised approach to be applied on the internal approaches; the fundamental review of the trading book, with revised boundary between the trading book and banking book; also revised internal models approach for market risk; and there are also changes in the revised standardised framework for credit risk in the second consultation document — for example, there is a reintroduction of the use of external ratings for exposures to banks and corporates, and also the use of LTV ratios in order to risk weight real estate loans. Therefore our opinion is that these changes to the Basel III framework are quite important, so we agree that we are heading towards a Basel IV framework rather than a modified Basel III framework.
Donchev, CACIB: Looking at numbers from Pillar 3 disclosures — which are actually quite helpful since banks started providing them, with LGD and PD ratios, risk weights, etc — for certain banks there is no impact, but for other banks, depending on jurisdiction there could be a 20%, 30%, 50% and sometimes as much as 100% estimated increase in risk weighted assets — so a potential doubling of RWAs. Based on this estimated quantitative impact there is clearly a mismatch with saying that there will be no Basel IV.
Perhaps the authorities are trying to be reassuring by saying there is not going to be Basel IV. My personal reading of this is that perhaps the final impact will be somewhat smoothed over, and that perhaps there will be kind of a transitional period that will be appropriate. But if we are talking about an overall increase in RWAs, let us assume, of above 10%, then you cannot say that is neutral. So while public regulatory statements might be getting gradually less hawkish over time, I therefore think that yes, we can clearly talk about Basel IV.
Allegaert, KBC: At this point, it is challenging to quantify the impact, given the uncertainty there still is around issues like floor levels, aggregation issues, etc. However, we understand European supervisors (e.g. the Bank of England, ECB) have indicated that the objective of such new requirements is not to significantly raise the capital in the system but to install a level playing field. This could be achieved through the calibration/floors and/or lower Pillar 2 add-ons. As such, it seems that Basel IV should mainly be a concern for banks that are outliers today.
Amundi: The potential rise in RWA will definitely impact capital. Nevertheless, the regulators now seem more concerned by the macroeconomic environment to which banks’ lending capacity is critical.
IFRS 9 is a major overhaul of the accounting standards that is due to be implemented in 2018 (effectively “tomorrow”). Yet its day-one impact on banks’ regulatory capital and the volatility it will induce on banks’ earnings are far from clear. IFRS 9 provisioning will be based on internal models. Paradoxically, this comes at a time when regulators are calling for a lower reliance on internal models for banks’ regulatory capital. The degree of oversight banking supervisors will have also remains a key unknown: unlike for Internal Ratings-Based capital requirements, IFRS 9 models will be validated by external auditors and not supervisors.
Skeet, ICMA (pictured below): Investors have raised the issue that capital triggers/tripwires might be activated by movements in the risk asset weightings and recalibration of a bank’s capital base through regulatory intervention, leaving investors potentially exposed. This might be another example of moving the goalposts in the name of macroeconomic policy or prudential supervision that subsequently leads to losses for investors. Predictability is therefore a key ask of investors and credit analysts — who nevertheless accept that there are unpredictable market and operational risks that regulators will legitimately need to address.
THE FRENCH SOLUTION: NON-PREFERRED SENIOR
Day, BIHC: Are you satisfied with the French solution that has been come up with to address the MREL/TLAC issue?
Baudouin, CASA: Yes. The French solution is really similar to what we had advocated for during the negotiations with other banks, other French G-SIBs, and the French Treasury. So we are very satisfied with the French solution. It presents many advantages over solutions that have been decided in other European countries, such as Germany or Italy.
First of all, there is no retroactivity for investors, as the stock of existing senior unsecured instruments will be preferred, which means that the ranking of existing senior unsecured debt will remain unchanged. Furthermore, it really clarifies the eligibility of debt for TLAC, because liabilities excluded from TLAC, like short term instruments and structured liabilities, cannot be issued in this new category of senior junior debt, which means that this new category of debt qualifies as TLAC with no ambiguity, and while it is less risky than subordinated debt, it is certain that it will qualify as TLAC. And finally, the French solution gives flexibility for French issuers, since they will have the option to issue preferred senior unsecured debt or non-preferred senior unsecured debt after the French law has come into force. So given those three big advantages of the French solution, we are really satisfied.
Perney Guillot, Fitch: We think the French proposal will provide banks with flexibility in their funding plans, assuming investor demand: non-preferred senior securities could help French G-SIBs comply with TLAC requirements, whereas more cost effective preferred senior securities would be issued to fund the banks’ structural deficit in deposits.
BNP Paribas is likely to be an active issuer of the new non-preferred senior debt because its current total capital adequacy ratio is far lower than its anticipated TLAC requirements. Other French G-SIBs either have smaller or no TLAC shortfalls. But they might issue some non-preferred senior debt, depending on pricing and market appetite.
Fedon, CASA (pictured below): The French solution is in our view superior to the German solution because the existing stock of debt is preferred. It is also clearer for investors in terms of the hierarchy of creditors when compared with the Italian solution. The Spanish proposal is not optimal because firstly, there are legacy ranking clauses in Tier 2 debt that prevent the issuing of Tier 3, and secondly given the EU momentum towards eligible liabilities being senior unsecured debt rather than subordinated.
We think that the majority of the Member States that have not adopted any solution for the moment could also follow the French route, in particular the Dutch central bank could be interested as well as Belgium, even if certain Benelux issuers benefit from legacy HoldCo/OpCo company structures and could use those. And even Spain could follow the French route because of the aforementioned complexities.
Chandrasekaran, Camares: Perhaps the design is advantageous for the reasons you mentioned, but do you think the market is going to let you sell non-preferred senior somewhere close to exiting senior or where Lower Tier 2 is?
Fedon, CASA: Regarding pricing, we think that the pricing should be set looking rather at the existing HoldCo-OpCo differential than at Tier 2 pricing. In fact, it should be a good proxy for pricing the new French non-preferred debt, and we think that the thickness of the tranche will not be the main driver of the pricing.
Chandrasekaran, Camares: Why? There is no element of structural subordination. It is entirely a function of the thickness of the tranche.
Fedon, CASA: Compare, for example, two issuers: the first one is completing its TLAC obligation with 90% of Tier 2 and 10% of senior non-preferred, and the second one is completing the same TLAC obligation with 50% of Tier 2 and 50% of senior non-preferred. For the first one, there is very little non-preferred debt protected by a lot of Tier 2, resulting in case of resolution in a low probability of failure, but a higher loss given failure, and therefore the pricing of the first issue of senior non-preferred should not be worse than the pricing of the tenth issue. And furthermore, if you have just one issue of senior non-preferred it will have a scarcity value. So the difference of HoldCo/OpCo should probably be a better proxy.
And you also have to bear in mind that the rating of the instrument will be involved in the relative valuation of the new instrument, as ratings are meaningful in the pricing rationale, just as they are for HoldCo-OpCo spreads.
Cornut, Camares: We will see. We think it is going to depend on the rating agencies, what type of rating you have, and hence what kind of real money guys are involved. Depending on their constraints, can they actually allocate to senior preferred or Lower Tier 2? I think that will be the main driver. And then how much the market is going to be excited about this new unpreferred asset class.
Day, BIHC: Do you have a sense from the rating agencies whether the new senior unpreferred will have the same rating as Lower Tier 2, or the current senior unsecured?
Donchev, CACIB: The rating agencies have publicly provided only initial views on the proposed French legislation, but not profound analyses. I think the good news perhaps is that for French banks this instrument will be largely rated investment grade. Obviously it will be between classic senior unsecured and Tier 2, and if you look at all the French banks these ratings are all investment grade – even if the differential varies widely according to the rating agencies’ different methodologies. If you have a particularly large differential in terms of notches, it might be close to Tier 2 and then over time increasing towards the preferred senior unsecured rating. But obviously it should not be the same rating as classic senior unsecured, this new non-preferred senior, because that does not appear to make sense (although, here, again, this will depend on each rating agency’s rating methodology).
We are talking here about debt which is supposed to be explicitly bail-in-able after Tier 2, but bail-in-able, so there are no systemic considerations to factor in. What all the rating agencies have in common is that to the extent that whatever state support they may still factor into the ratings of classic senior unsecured debt or old-style preferred senior unsecured debt, this will not apply to the new debt instrument, so it will be notched or it will be rated based on the standalone credit rating of the issuer, whatever this is called for each rating agency.
Perney Guillot, Fitch: The new French non-preferred senior debt class would effectively become the reference debt class used by Fitch for setting a bank’s Long-Term Issuer Default Rating (IDR). This is because our IDRs rate to the third-party, private sector senior debt category with the highest risk, and the default risk on the new class of non-preferred senior debt would be higher than the risk associated with preferred senior debt and other senior liabilities like large, wholesale deposits.
In theory, if a bank issued very large and stable volumes of non-preferred senior securities and built up a considerable buffer, the default risk of the preferred senior debt would reduce and it could be notched up from the non-preferred debt rating. But we are some way off this, as the French option still requires debate and we think legislation is unlikely to be passed before end-June 2016.
Day, BIHC: To what extent will the fact this solution is now available change your plans?
Fedon, CASA: With the new liability class, we get complete flexibility to fulfil our TLAC obligations without increasing the cost of our regular funding, and in fact we might effectively issue less Tier 2 in the future and replace it partially by senior non-preferred issues. But it also remains to be seen what, if any, rules emerge at the EU level with respect to eligible liabilities for MREL and TLAC purposes.