Participants in the roundtable were:
Andrew Freestone, portfolio manager, Man GLG
Ola Littorin, head of long term funding, Nordea Bank
Alex Wall, head of capital, rating agencies and investor relations, Nationwide Building Society
Ervin Beke, credit analyst, banks, BlackRock
Isabel Rijpkema, head of long term funding and capital, Rabobank
Nicolas Bonis, portfolio manager, European financials credit and equity, Chenavari Investment Managers
Gildas Surry, portfolio manager, Axiom Alternative Investments
Matthieu Loriferne, executive vice president and credit analyst, Pimco
Wim Allegaert, general manager in finance, KBC Group
Charles-Antoine Dozin, head of capital structuring, Morgan Stanley
Khalid Krim, head of bank solutions, Morgan Stanley
Aarti Vasudeo, senior manager for capital issuance, ratings and debt investor relations, Lloyds Banking Group
Tyler Davies, moderator, GlobalCapital
: The market has had to weather a lot of volatility this year — it’s been nothing like it looked in 2017. How has that changedGlobalCapital thinking about the market?
Ervin Beke, BlackRock: It has actually become somewhat easier. As spreads have increased, the value has become somewhat better than it was back in January, when you had to decide whether or not to get involved in new deals at very tight spread levels. Overall, the way we have thought about this year is that it is going to be more about carry and not so much about spread compression, just because the market has become so tight.
The widening we have seen this year did not necessarily surprise us. The timing is not something we could have predicted, but the fact that spreads got wider has not changed our thinking about investment in today’s world.
We would want to look at relative value of debt against equity, trying to assess which part of the capital structure is more attractive and how risk is assessed in both markets. And then of course we look at macro trends, which to our mind have not really changed that much since the beginning of the year.
Matthieu Loriferne, Pimco: For me it is the liquidity in the market that has changed the most in recent weeks. After we saw how quickly the spread in Italian government bonds widened out and how difficult it was to find clearing levels, including in the short-end of the curve, it has made us want to reassess relative value across the capital structure — starting with the sovereign.
This not only has implications for buying in the secondary market, but also in primary. Execution of new issues will therefore be paramount. We were very selective throughout the year and we will continue to be.
Andrew Freestone, Man GLG: We saw a lot of off-the-run and relatively small sized paper being sold by banks in 2017 and early 2018. Given how market liquidity has changed, we believe that it is worth thinking about how these deals get priced going forward, who ends up holding the risk, and for how long those investors can hold the risk.
I do not think that these issues were out of people’s minds at the beginning of the year, but things are much tougher now in our view. We believe that the less usual names that have come to the market are going to have a lot more difficulty than the large and very liquid names. It will be interesting to see how this situation changes over the next few months and into next year.
Nicolas Bonis, Chenavari: The picture in the European banking sector is better than ever in terms of fundamental credit quality. We used to price to perfection in the past, but not anymore — so fundamentals are good but technicals are very difficult now. As Matthieu said, liquidity is a big part of that. I don’t see that situation changing in the short term. As investors we therefore remain very prudent and we are very lightly positioned now more than ever. We can see that issuers want to approach the market and be very proactive. That is the correct strategy for them. But for investors the correct approach is to be prudent. Liquidity is very poor.
: How can issuers square off a more cautious tone among investors with their desire to come to the primary marketGlobalCapital quickly?
Charles-Antoine Dozin, Morgan Stanley: Since we saw a resurgence in volatility in the second part of the semester, it’s been all about building optionality. It’s been about being able to seize the right windows and being quick to react. That means that issuers have to be prepared early to the extent that they can be. And they have to be able to make quick decisions to push the button and move forward when necessary.
Ola Littorin, Nordea: We sold our inaugural senior non-preferred transaction in June of this year. We updated our documentation after the first quarter to incorporate a contractual issuance capability for senior non-preferred.
That gave us the flexibility to control our timing for issuance without being dependent on the statutory introduction of the credit hierarchy directive in Sweden or in Finland.
We had also made internal preparations with the objective to create flexibility to act with very short notice should an appropriate issuance window appear. We had been monitoring the markets for some time, which had been volatile in offering very limited visibility. We spotted an opportunity to announce the transaction in the wake of the European Central Bank’s June 14 meeting when it announced a halving of its monthly asset purchases. The market responded well to the news allowing a positive backdrop to launch and bookbuild our deal in the following day.
Even though our agenda for senior non-preferred issuance is not massive over the next four years we deemed it prudent to commence issuance now in 2018.
Khalid Krim, Morgan Stanley: As well as the short term planning and the need to access a particular market, there is also the longer term capital and MREL planning that is happening behind it. Some banks are well advanced with that and some are just starting out, like Nordea.
If you think about the AT1 market, it took quite some time for all of the issuers to go out and raise capital and get to the point where they are now as they think about refinancing and replacing the first wave of deals. On the MREL and TLAC side, it is more about building up the stock and articulating that to the market.
Wim Allegaert, KBC: I would agree with that. From an issuer’s perspective the most important thing is not necessarily the current volatility or poor liquidity. It’s crucial to have a good idea of where the market is, but your multi-year planning is actually step one. It starts with your budget.
Aarti Vasudeo, Lloyds: That’s right. We’ve been quite active in the market this year in the terms of senior funding. But as we have already heard, the higher levels of volatility have made banks actually step back and question their entire strategy for funding and capital planning over the next couple of years.
This has made it difficult as you have to think about the build-up of MREL over the next few years. When the regulation comes into place, we have to be clear about what buffers one really needs to have in place. And then there is also the fact that AT1s will come up for calls.
Banks will have to make the decision whether to refinance them or not, depending on where the market is and where their capital stack is at that point in time.
Isabel Rijpkema, Rabobank: We always try to anticipate and be ahead of the curve with regards to capital planning. In anticipation of MREL — and in absence of the option to issue holdco or non-preferred senior to meet MREL requirements, we have been issuing quite a significant amount of tier two since 2012. Although tier two is obviously more costly than non-preferred senior, I am confident that we can benefit from the strong starting point in terms of capital buffers at the time of introducing non-preferred senior. With non-preferred senior we can continue working towards our MREL requirement, while optimising the capital structure
We also try to look quite far ahead at what might be coming at us that could impact market conditions and hence issuance windows. This is especially true for products like AT1, because you have to plan well in advance to be able to use the right windows and not come under any pressure to issue. The slightly higher buffer of AT1 than the 1.5% required that we are targeting helps in giving you a little extra leeway in which to plan your transactions.
Alex Wall, Nationwide: Don’t underestimate just how long the planning cycle is and just how far in advance issuers are thinking about the timing of transactions — notwithstanding what might be happening on the business side of the balance sheet.
It was in 2015 that Nationwide started talking about what we could do about 2019 as a potentially difficult year for AT1 refinancing. We knew that would be the year in which the first significant wave of issuance of the product came due, and we knew that there was going to be a lot of sterling refinancing that year.
That was the genesis for the thinking around issuance of more core capital deferred shares. It was also the genesis for our strategy of issuing a significant volume of tier two in advance of being able to issue senior non-preferred. All of this was three years in advance of the hypothetical refinancing event.
: KBC has already refinanced some of its AT1 for 2019. Was this similar thinking?
Allegaert, KBC: It is about being prepared to go early if there is an opportunity. We have also been quite vocal that we want to keep the 1.5% buffer of AT1 filled, so we didn’t want to face a problem when looking to refinance at the beginning of 2019. We did not go as far as Rabobank and look to build up an extra buffer of AT1 above this, because we are quite comfortable with protecting our other debt investors with a comfortable layer of core equity.
: It is obviously not just issuers that have to think in advance. If we go to the AT1 market specifically, people are obviously going to have to start thinking about the first wave of AT1s coming up for call. How are investors looking at the probability of deals being refinanced?
Gildas Surry, Axiom: Issuers have communicated extensively about the economic incentives behind calling or not calling, so the probability comes down to a transparent and objective rationale that is quite easy to anticipate.
Those with high reset spread levels are very likely to be called. You can also look at the potential impact on CET1 of calling foreign currency issues that are accounted for as equity. There are UK bank AT1s and an older instrument from Rabobank that could present issues in this regard, but in most of the cases it is fairly predictable.
It’s a bit trickier to approach what we call legacy instruments — deals that have been issued before 2016 at the very latest. Some have been called and some have not. That’s why there is a lot of value and opportunity in this sector.
The UK is an interesting example. When you look at a bank’s ring-fenced entity now, they are better capitalised and better protected. It is not difficult to expect the cost of their funding instruments to be in the single digits, which can give you some clues about which legacy instruments will be taken out.
But if an instrument is not called, it is no longer a strongly negative market event. The important thing is that there remains an active and transparent dialogue between investors and issuers about their funding strategy, and ultimately call policy.
Freestone, Man GLG: Even at the time when the earliest AT1s were issued, it was pretty clear to us that the environment had changed. There is a whole set of different incentives now and everything is designed so that you could expect the product to behave a bit more like the US preference share market.
Nevertheless, it must be on everyone’s mind that at some point an AT1 is not going to be called. I’m sure that some people will be surprised at different points in time, and clearly you’ve got an awful lot of complexity built into a number of these instruments — so if you do see market movements, prices could move quite considerably.
But I would be surprised if we were to see one of the early issuances left outstanding, just because of where spreads were at the time.
Beke, BlackRock: I would add that the AT1 market seems to be pricing in these probabilities reasonably efficiently. New deals are being priced more as perpetuals, and old deals to their first call dates, with a higher reset spread, so if we see a purely economic decision of not exercising an AT1 call, I don’t think the market would be surprised. But if we see security with a high back-end not being called, that’s something that could prompt us to re-evaluate how we were thinking about AT1 calls.
Bonis, Chenavari: I would agree that it’s pretty easy to know which ones will be called, so we do have a clear distinction in the market now, between the ones that will be called at first call date and the ones that should be priced to perpetuity.
I don’t think investors have been ready for that, because when we look at the repricing of some bonds now on the market it feels like they could go lower. Most investors used to price AT1 to the first call date. Maybe not the specialists, but other investors used to price to first call date.
From my perspective, if some issuers aren’t able to refinance the high reset bonds I would be surprised and my perception is that there should be a market impact. But it’s a very low probability.
Dozin, Morgan Stanley: The flip side is that at the very top of the market earlier this year we had a couple of instances where issuers had to make some moderate concessions when it came to reset spreads. There was less of a probability that if conditions were to deteriorate they would be able to or willing to call the instruments.
When AT1 valuations are very high and the cost is very effective from the point of view of the issuer, there is obviously more of a risk that they will decide to sit on the capital for longer and maybe not necessarily exercise their calls at the first call date.
If we were to go back to the valuation levels that we had earlier this year, then I wouldn’t be surprised if in some situations investors demanded a little bit more of a spread to compensate for that.
Krim, Morgan Stanley: There won’t be too much discussion about the wave of AT1 refinancing that is coming, because most of them have high back end and reset values. It is probably more a topic of discussion for the wave that is being issued now and that will come for call in five or seven years, depending on the call date.
Would the first non-call of an AT1 be a problem for the market, though? It is going to be very specific. If it is linked to an understandable problem with one institution then I’m sure people can get over it. If it is something that implies a repricing of the AT1 market then people may well feel differently.
Loriferne, Pimco: A lot will depend also on the environment. If it comes in a strong market environment like we had last year or in January, it would be easier to digest than if it comes in the more volatile market environment that we have had in the middle of this year.
If we look at what happened last year with a major UK bank, they were able to explain a call decision on some of their legacy instruments in a very transparent manner. It was well received by the market and it avoided a significant negative repricing of their bonds, unlike some of their peers a couple of years before.
So it really matters how issuers communicate their intentions. That helps a lot in smoothing the price action and ultimately it helps with their continuous access to the market at a decent funding cost. What the market hates the most is a surprise, particularly one difficult to explain.
Freestone, Man GLG: To your first point there, Matthieu, if we see a rapid and substantial spread widening resulting in the non-call of an instrument, then that could be a big concern — especially in the context of the banks saying that they plan their capital issuance years in advance. What would have gone wrong in such a short timeframe?
If there is just a broad widening over a long period of time, it is a slightly different set of circumstances where you might not have had a chance to refinance inside that spread.
: Would you then take into consideration what banks are saying about their call policy when valuing an instrument? Would you rather see a predictable issuer or an issuer that makes decision on an economical basis?
Freestone, Man GLG: Can you make a decision on something other than an economic basis? That’s the real underlying question.
Rijpkema, Rabobank: One of the key drivers behind our decision making is having continuous access to the market and a consistent storyline. That could also mean that we could call an instrument potentially in five years, six or seven years, even if it’s economically irrational — because it should pay off for consistency purposes and all your instruments in the longer term.
Freestone, Man GLG: Yes, but I’m slightly concerned about the regulatory environment and the pressure that you could face when making those call decisions one way or the other. Irrespective of what you might say and how you might think of long term value of your relationship with investors, there might be shorter term pressures — over three, four or five years — that force you to make slightly differently choices.
At the moment we are in an environment where we haven’t seen any non-calls in new-style instruments, so we don’t have any examples on which to base our decisions. It is not that I don’t believe what issuers are saying, it is just there are other factors involved.
Rijpkema, Rabobank: If you meet all your capital requirements, in my view there shouldn’t be any ground for the regulator to not allow to an AT1. I have the feeling that regulators also value the long term perspective for issuers and market access.
Beke, BlackRock: The problem with reputation and call policy is that we have not seen the proof in this new environment yet. We agree that calls must be based on economic decisions — that is why it has been written into the CRR/CRD.
We are, however, not involved in the discussions that issuers have with the regulator. We do not know how they think about it, and that’s why we don’t know how they would approach an argument where an issuer would say that it is going to call an instrument even though it is going to replace it more expensively.
That is why I will want to see first the proof that it is a viable strategy, before I give too much credit to the idea that the regulator can accept uneconomic calls if the issuer has enough capital. It is being on the safe side for an investor to treat calls on more of an economic basis.
Vasudeo, Lloyds: I do think that it is more institution specific. If the focus is on P&L then the call decision will be on an economic basis. Banks would not want to pay extra for a new issuance and call the old issuance.
Given that there are very limited windows to issue an AT1 and given that banks are required to obtain regulatory permission and everything else in place first, banks would rather pre-fund the AT1 as much as possible than do something very close to the first call date. If the markets widen completely then there is a risk of a non-call scenario. The cost of carry is also to be considered when timing the AT1 issuance.
Littorin, Nordea: The market is moving into uncharted territory when it comes to call behaviour of new-style instruments. A number of factors will be relevant here including prevailing market conditions, back-end spreads for the ISINs coming up for a call, regulatory considerations in the approval process and, last but not least, the issuing bank in question.
We may see different behaviours from different issuers with different drivers. As an issuer we cannot comment on future calls. However, I expect investor focus will be high on this topic.
: As we have been saying, the long-term trajectory of the AT1 market will be important. Coupon levels hit new lows last year, but how do people anticipate AT1 pricing changing?
Dozin, Morgan Stanley: We got the sense that a lot of issuers were accelerating their plans in January and February to be able to take advantage of the perfect market conditions that we had at the time. Things have been getting worse and worse ever since then, so everyone has wanted to get into the market as soon as they can.
Littorin, Nordea: Yes, the market has woken up to the fact that we have had a long period of very benign issuance conditions supported by a hunt for return for investors. But we saw the turning point in market sentiment earlier this year resulting a flurry of AT1 transactions in the primary market despite non-perfect receptivity of some deals. That illustrates that there has been a rethinking of funding plans among issuers and an acceleration in bringing more strategic trades to the market such as AT1s in anticipation of higher spreads in the future.
Krim, Morgan Stanley: The interest rate environment is also at play here. People may disagree as to whether we should look at the coupon or the spread when valuing the product, but as rates are going up, it will clearly affect the way in which both investors and issuers assess pricing.
Allegaert, KBC: Precisely. At the beginning of the year we got to a point where we felt that you would have to factor in some extension risk for AT1s regardless of the interest rate scenario at the time, but we also got to a point where some AT1s had yields that were actually lower than share dividend yields. So investors were taking the downside on their notional without taking any of the upside.
: How has the way in which investors value AT1s changed?
Surry, Axiom The number of uncertainties has increased. That’s true whether you look at political developments or economic prospects.
But it’s also true if you look at the regulatory environment, because the relationship between issuers and investors can be sometimes distorted by the regulator. Over a year ago we had the resolution of Banco Popular, when the regulator missed an opportunity to make the AT1 format work. The regulator was also slow to take sides when Aviva moved the market with an announcement about its UK preference shares.
Clarity regarding how regulators will treat these products would be always be welcome. When you compare AT1s with equity, you might wonder why the product trades with accrued interest for example. If you are a US investor you might also wonder why it is often taxed as a fixed income instrument, when US preference shares are subject to a different tax treatment.
We would like regulators to take more ownership of the product and align its characteristics with what has been done elsewhere.
Wall, Nationwide: But the reality is that regulators across Europe have missed the opportunity with examples like Popular to demonstrate how a CoCo should work properly, which is frustrating from an issuer’s perspective.
The initial wave of refinancing is coming due and people are already replacing bonds with new deals that will be out there for at least five, seven or 10 more years.
I would hope that over time we get a greater variance in how investors look at the instrument idiosyncratically. They will look at whether it’s a Nordea or a Rabo or a Nationwide AT1 and they will price it according to the credit of the institution. The concept of the AT1 market will be less to the fore, so we won’t have an event like we did in February 2016 when a single bank’s activities dragged the whole market wide.
Vasudeo, Lloyds: Yes, when the regulator introduced AT1, they were at a very nascent stage of developing their regulatory framework. They then bolted buffers and buffers on top of the trigger levels, so the 7% and 5.125% figures have become absolutely irrelevant.
Bonis, Chenavari: Indeed, it absolutely clear that the 5.125% trigger level is totally inefficient. Issuers pay a high cost for that. We can see now that AT1 does not really work as a going concern instrument — it’s more a gone concern instrument now. Normally we would try and change the format of this kind of instrument by maybe putting a higher trigger on issuances to make it clear that there is a risk and they can absorb losses if a bank faces trouble.
Wall, Nationwide: But the difficulty is that if the regulator came out now and said that bonds would have to have a 9% or 10% trigger to qualify for all of the benefits originally intended for AT1, then it would potentially put an incentive on issuers to redeem the existing class of instruments. So they’ve missed their window.
Surry, Axiom: You can still have a grandfathering period.
Andrew Freestone, Man GLG: That’s a legislative choice though. That’s not a regulatory choice.
Loriferne, Pimco: The various iterations of the CRR have shown that there is always a reform or a fine-tuning of capital requirements and new definitions of what does and doesn’t count as a capital instrument. The legacy tier one asset class is a perfect example of that.
So it is conceivable to think that a couple of years from now we may have a new set of requirements and a new set of going concern instruments. That’s totally plausible. The fact that they may create an incentive to call — so be it. It’s also in the interest of the regulators to replace inefficient instruments with instruments that will work better.
Krim, Morgan Stanley: I have mixed feelings here. How can we explain that banks are issuing something that may have regulatory benefit but cannot or should never be used? That is the issue with the AT1 market. There is still a stigma about flipping a piece of debt into equity, about not paying a coupon and not calling an instrument.
So we can start to think here in Europe about creating a going concern tool that gives banks the potential to have true contingent capital that would be included as part of early intervention measures and recovery capital. As long as its holders accept that on day one they are a debtholder but on day two they may be turned into a shareholder and must be comfortable owning the bank’s shares through a liability continuum, then why not issue these kinds of securities.
It may not be for everyone, but at least we could think about having something available as going concern capital for stress tests or for recovery planning.
: What do some of the issuers think of selling a product that genuinely offers going concern capital?
Rijpkema, Rabobank: We would all benefit from having an instrument that actually works as going concern capital. Practically speaking, there is quite limited value in having AT1 on your balance sheet at the moment other than for meeting requirements. It is not expected to properly function as going concern capital and although the layer protects more senior instruments like tier two and non-preferred senior, the layer is relatively small for the majority of banks. If we take a step back, what we all intend to do is make the financial system safer and avoid big shocks if something goes wrong. It is always good to design an instrument that can function for a good cause.
: So is it worth an experiment?
Rijpkema, Rabobank: Preferably, you do at least want to have some sort of regulatory benefit first.
Loriferne, Pimco: A regulatory experiment is always a bad idea. Issuers should not try to depart too much from regulatory requirements.
Hybrids have proven to have value in times of significant stress. We all remember that the first CoCo conversion was actually a major US bank switching their entire stack of preferred securities into equity back in the second quarter of 2009. So at times of real stress and when significant capital is required, CoCos have absorbed losses in the past and proven their usefulness.
We can debate going concern and gone concern, but when it matters you will have an extra 1.5% or more to absorb losses and share the recapitalisation burden. That has to have a value for the issuer as well.
Rijpkema, Rabobank: In 2010 Rabobank experimented with senior contingent notes which in stress serve as an extra buffer that will be written down at a critical moment. The structure as such has never received any regulatory recognition unfortunately.
Wall, Nationwide: What Khalid has described in terms of having an instrument that converts to equity and gives the owners a stake in the running and functioning of a new entity on a going concern has to be preferable for all parties. Unfortunately, the one time we have seen a going concern instrument used in anger, it wasn’t used that way. That’s why it was a missed opportunity.
Allegaert, KBC: The 1.5% AT1 bucket does not actually provide a lot of protection for everything behind it given the relative thickness of this layer. If you look at most banks in our peer group, they run with about 14% in core equity. So you get to a situation where you can say that equity levels are sufficient or more than sufficient for a lot of economic scenarios, but then you have this whole stack of much smaller layers if you run into trouble. So the distinction between AT1, tier two and even bail-inable senior debt is going to be more blurred.
: If all bank debt instruments are gone concern, should there really be much of a distinction between each of the different tiers?
Rijpkema, Rabobank: Clearly it depends on the size of each of the layers, but there is still quite a big difference with regards to risk profile driven by the layer size and the point at which they can be written down from a regulatory perspective. Non-preferred senior is still to be introduced in the Netherlands, but when we issue it will be protected by over €50bn of capital buffers with comfortable layers of each part of the capital stack — based on a bail-in buffer of 26.8% at the end of 2017.
Freestone, Man GLG: The head of at least one resolution authority has said in the past that subordinated instruments have zero recovery in almost all resolution cases; that means that value would break in the non-preferred senior class. If you have a non-preferred senior tranche that is large enough, then they might become the new owners of the bank.
Today, we believe that we are not at the point with MREL that any credible resolution authority would be able to do that. Popular was one example case in which this did not happen — losses stopped at the subordinated bond classes.
Your loss given default has to drive your valuation difference in these cases.
Loriferne, Pimco: Losses in a going concern scenario always look smaller than the losses actually incurred when the bank is a gone concern. The losses you are expecting tend to be completely amplified once you switch from going to gone concern, because one quickly realises the magnitude of losses after, for example, marking down portfolios to more appropriate clearing levels and over a shorter time period. So it is not inconceivable that a large bank, levered 15 times, can incur losses of €20bn, €30bn or even €50bn in a resolution. This has to be reflected in the price of non-preferred senior debt.
Krim, Morgan Stanley: But the point of having an MREL requirement is also to have a buffer. I do not expect the entire non-preferred senior bucket to be touched.
There is also an interesting question around where tier two sits in the capital stack. We know that it can be used in a resolution, but also outside of a resolution. So in discussions about total capital, tier two is probably more of a buffer to the next layer than something that is safer than AT1 and riskier than non-preferred senior.
Wall, Nationwide: The way we always look at it is by using stress testing as a starting point. It provides us with the building blocks to calculate how much of a buffer of going and gone concern capital we will need, and how useful these instruments could be through a five year stress period.
That is when you do see a clear difference between tier two and non-preferred senior. If you have a one or two year period of extreme market distress and dislocation, you can withstand having to absorb losses as a going concern by having an MREL buffer on hand and callable tier two instruments. The optionality around the call of tier two during that stress period has real benefits in the depths of a stress.
Rijpkema, Rabobank: I fully agree that tier two will have value in a capital structure. I see the benefit in making sure you protect your senior non-preferred creditors and hence more senior creditors with sufficient buffers.
From a regulatory standpoint there will probably remain quite a distinction between capital instruments and gone concern instruments, which they will only want to use for resolution.
Surry, Axiom: We are thinking here in terms of a Basel II concept of going and gone concern. Basel III is more about non-viability and bail-in.
If you ask the regulator whether a non-preferred senior is a gone concern instrument, they would rather refer to the point of non-viability — or PONV. These are new and untested concepts that give investors a lot to analyse — and ultimately value to capture in their portfolios.
But one may remain sceptical about the greater good contributed to the sector. Complexity is the enemy of efficiency. We prefer to spend time with bank management trying to push them into improving their business models rather than fine-tuning their capital stacks.
And it is quite ironic that senior non-preferred is trending in pricing towards where tier two was before the crisis. Is non-preferred senior therefore the new tier two? Tier two may have value in stress tests, but what if value were given to non-preferred senior in stress tests for instance? Ultimately there should be convergence between the two asset classes and that is what we will be trying to identify in the long run.
Littorin, Nordea: There is a logic for a differentiation between the various capital classes. But it’s important to reflect not only on the respective sizing of tier two and AT1 but also on CET1. As a Nordic bank our CET1 ratio is higher than most European banks resulting in AT1 and tier two levels closer to minimum requirements. With a comparatively larger CET1 ratio the loss probability for AT1 and tier two is reduced impacting absolute and relative pricing. This total capital stack will also act as a cushion for non-preferred debt, which is to be used in resolution.
I think a reasoning around valuation parameters of capital instruments need to be referencing the issuer as different banks may have different requirements on capital, have different ratings and are in different stages of the build-up of capital including bail-inable debt. This creates challenges for instrument valuations but is something the market will become more accustomed to.
Loriferne, Pimco: The regulator wants to be sure that there are buffers above buffers above buffers, so that when losses are incurred it is not too damaging for the rest of the institution. It is not inconceivable as we build up these buffers that we realise in a few years’ time that the Swiss solution — of having a significant CET1 requirement supplemented by a large stack of AT1 — offers a credible alternative to the current set-up in other jurisdictions. The distinction between tier one and tier two is likely to diminish once a bank meets all of its regulatory requirements. We can already see that happening, given that tier two issuance volumes have fallen significantly prior to the introduction non-preferred senior. It feels like issuers were just waiting for clarity on the regulatory side.
Tier two is helpful when building the capital stack, but when the transition is complete, the value of the instrument diminishes. That’s probably a 2022 conversation.
Krim, Morgan Stanley: The market is not marking any difference between issuers that have capital stacks with 4% or 5% of tier two, versus those that have 2%. If people are not getting the benefit from the investor community or from the rating agencies, then they will say there is no uplift for tier two and they will issue what is cheapest to deliver for MREL, which is non-preferred senior.
Rijpkema, Rabobank: The fundamentals may not be reflected in the current market environment, but if we have worse conditions then I would hope to see that valuations are not only driven by technical factors. I would hope that analysis around the probability of default and around loss given default would be driven by the make-up of a bank’s capital stack, as well as its business model. There should therefore be a role to play for tier two to protect non-preferred senior debt.
Allegaert, KBC: But it does come at a cost. Your interest expense is there if you build up more AT1 or tier two, and let’s not forget about the build-up in the core equity capital stack either. That has been quite spectacular. The regulator does not come to you to tell you to build up larger buffers of AT1 or tier two in the first place. They say that you should focus on your core equity first and foremost. And that’s reflected in their target setting.
Dozin, Morgan Stanley: From that perspective, the capital requirements naturally force you to operate with de minimis AT1 and tier two, because they are the limits to which you can substitute CET1 for non-equity instruments.
If there was a split of the Pillar 2 requirements between those different buckets, you might be able to increase the role of these AT1 and tier two instruments — as in the UK and Sweden. But in the long term I would agree that when issuers have reached the stage where the loss given default of the senior non-preferred stack is where it needs to be and they have eased into their steady state requirements, the probability of default becomes a little less important.
Total capital requirement levels will remain important for the next few years, but the concept will probably be a little diminished when issuers comply with all of their targets. At that stage we might see issuers reverting back to the minimum levels of 1.5% of AT1 and 2% of tier two and not more.
: What about the other end of the liability stack? Will banks continue issuing senior preferred when the non-preferred senior layer has been built up?
Vasudeo, Lloyds: Preferred or opco senior debt still provides some funding benefit, and there is also a price element to it whereby issuers can execute at lower costs. So banks will continue to rely on preferred senior, but it will purely be from a funding perspective.
For UK banks like Lloyds there is also the question of what happens when ring-fencing comes in. The new and different entities will have to fund themselves, and they will require a mechanism for doing that.
Wall, Nationwide: Yes, and from a mix and diversity of funding sources. We are all left in a fairly homogenous space in that AT1 and tier two are clearly long dated product, as so are non-preferred senior bonds. Most of us, even Rabo, have now moved to the wonderful world of covered bonds, which again are long dated products. The obvious thing to say then is that preferred senior is a very flexible and useful short dated funding tool.
Rijpkema, Rabobank: I fully agree. Clearly your ability to issue preferred senior will depend on how much of your total funding requirements have been eaten up by capital and non-preferred senior. But the product will continue to play a role in our funding structure
Dozin, Morgan Stanley: The elephant in the room is that a certain portion of your senior preferred debt may be able to count as MREL, if you want it to be eligible.
This might not apply to institutions with holding company structures or those banks that intend to operate with full subordination, but I suspect that preferred senior will have a role to play in the MREL picture for smaller banks. To the extent that this is possible, I would not be surprised if a number of banks did not move too far away from their current funding mix and into the non-preferred senior asset class.
Rijpkema, Rabobank: I have a hard time understanding these discussions; why design an instrument specifically for the purpose of MREL and to protect senior preferred as a pure funding tool and then allow banks to use preferred senior again under conditions. I cannot see how that would work together with the concept of no creditor worse off either.
Loriferne, Pimco: Allowing 2.5% of preferred senior for the build-up of MREL would have to be a transitional measure. If it became permanent it would be a mistake. At the point of resolution we would then once again enter into the question of where the regulator should stop.
Preferred senior is simply a very useful funding tool that allows you to access the market without having to rely on central banks and without encumbering your assets too much.
In my opinion there is potentially a structural weakness in the Canadian system compared to the others, in the sense that all senior debt will have to be bail-inable. This is because it removes the extra funding flexibility that a senior unsecured funding tool can provide.
Littorin, Nordea: I completely agree that preferred senior debt is vital as a funding instrument for liquidity, which becomes even more important in times of stress. We will continue to be an active issuer of senior preferred parallel to senior non-preferred.
We have stated that we aim to build our bail-in-able requirement with subordinated instruments like senior non-preferred as it would be logical in the context of “no creditor worse off”.
Allegaert, KBC: Senior preferred is actually tier four. That is the way you should look at it.
But its value is for funding purposes. Strictly speaking we do not need it, because we have a low loan to deposit ratio. We have to issue everything else for regulation — tier one, tier two and tier three. It increases your interest expense, but you have to issue it. And so senior preferred is different because you can either issue it for funding purposes or you can just stay on the sidelines if you have no need for it.
: Regulators have stressed that they can bail anything in, so I am interested to hear that preferred senior would be inappropriate as MREL. Do people consider that there is a remote chance of preferred senior ever being bailed in?
Wall, Nationwide: The regulator beats us up as issuers on a regular basis about barriers to resolution. The inclusion of preferred senior within an MREL bucket is a barrier to resolution. Over time this will go.
Beke, BlackRock: We saw this last year as well, when we had two EU resolution cases in which senior instruments were left untouched because it would have been too difficult to bail them in in a timely manner. There was a Spanish bank with a capital deficit and even though the regulators could have reached for the senior debt as well as the subordinated instruments, they decided against it.
Bonis, Chenavari: We are talking here about banks with good market access, but there are numerous banks across Europe that have no market access — especially for the new layer of non-preferred senior debt. I am trying to understand how they can comply with MREL regulation. That is a real concern for me.
The other part of this is that I prefer the holding company/operating company approach to MREL, but unfortunately we cannot apply this structure across Europe because of the mutualist set-up that is still in place for banks in the region.
When you put a bank with a holdco approach into resolution, it is much simpler to have all of the necessary resources at the holdco and the equity at the opco. It makes it easier to keep the bank open for the next stage. This helps with keeping all of the layers open to investment. When you compare Europe with other jurisdictions like the US and Australia, you see that banks in those places still give the ability for all investors to gain exposure all across the capital structure — even retail.
Vasudeo, Lloyds: Having a holdco/opco structure has made the UK structure simple. We haven’t really had to deal with the issues that other banks have had to face, with non-preferred senior coming in. At Lloyds, we are also ahead of some of our peers in terms of building our MREL layer.
But there has been a big push from the regulator to clean up the entire capital stack. Historically we have legacy instruments that were issued of our operating companies, like old style tier twos that could potentially be an impediment to bail-in. The push from the regulator is to see that move from the opco to holdco so that all the MREL and capital sits at the holdco, which would make bail-in and resolution easier.
: What role can green bonds play in helping banks meet their funding or bail-inable capital requirements?
Littorin, Nordea: We sold our first green bond last year which was in senior preferred format. For us the value is monetary, in the form of attractive issuance terms, but also strategic, in the form of addressing sustainability targets for us as a bank but also for our customers. The volume of green loans on our balance sheet that are eligible for green bond issuance is increasing. We will continue to issue green bonds in senior preferred format. Whether the green bond concept will expand into capital or senior non-preferred in a larger scale is a bit too early to say.
Allegaert, KBC: We did our inaugural green bond this year as a holdco senior bond. The deal had been in the making for quite some time and it actually underpins our position and progress in the overall sustainability of the corporate strategy. A bond deal was the next step in bringing this topic in front of a wider audience of investors. As Ola said, there is definitely more to follow.
Krim, Morgan Stanley: It is clear that there is a bigger focus on sustainability from the industry and in Europe we have initiative at policy level from the European Commission that will keep this topic live in the next 12 to 18 months and on top of the agenda of banks and the financial services industry. Banks need to think about how they will be able to support green finance, and green or sustainable bonds can help banks to convey that as part of their ESG strategy.
Surry, Axiom: If you look at the universe of green bonds, it is about €40bn equivalent now. That is sizeable but it is probably not sufficient for us to initiate a dedicated green strategy or mandate yet. We welcome and support the trend and we try to incorporate questions about sustainability into our conversations with bank management, because it is something that we expect them to address as much as cost reduction or business digitalisation.
Beke, BlackRock: Yes, it is certainly something that is gaining more traction from the investor base. Institutional funds especially are becoming more interested in ESG factors and they want this to be reflected in their investments as well. We are going to see more and more issuers trying to meet that demand with appropriate issuers. This is certainly not going to be a small sector in the future.
Loriferne, Pimco: It is extremely important that investors engage with banks about these very important issues. While we obviously value a more focused approach towards ESG issues, we would always balance that with the price at which those instruments come at. And at the end of the day the credit risk we take will be the same in a green bond as in a normal bond.
: What else will we be sitting here talking about in 12 months’ time?
Bonis, Chenavari: The Banking Union is a key topic. It could be very difficult to implement the changes now, and this is not necessarily the priority of all politicians across Europe. But progress towards a stronger Banking Union could be a game changer for continental banks.
Dozin, Morgan Stanley: It seems as though a lot of work has been done across the bloc to reduce the overall risks in the banking system, especially in the periphery. CRD 4 and the BRRD have been implemented and there has been a lot of pressure on non-performing loan positions.
What is missing is the element of risk sharing. That is going to be the topic of conversation in the coming months. We will see if there is more willingness from countries like Germany to accept further mutualisation of risk once we have looked at the final terms of the risk reduction package. s