2024 Outlook Roundtable: Everything Can Happen
Financial institutions investor roundtable with the EBA: macro & structural drivers for 2024 issuance.
The calendar may be turning from 2023 to 2024, but many of the questions stay the same. When do rates start falling? How might CRE hit banks? What future AT1? To discuss these and other critical matters, on 13 December, Bank+Insurance Hybrid Capital and Crédit Agricole CIB invited investors and the European Banking Authority to explore the key macroeconomic and structural considerations for financial institutions issuance in the coming 12 months.
Participants:
Filippo Alloatti, head of financials (credit), Federated Hermes Michael Benyaya, co-head of DCM solutions and advisory, Crédit Agricole CIB Jenna Collins, portfolio manager, Brevan Howard Vincent Hoarau, head of FIG syndicate, Crédit Agricole CIB Raoul Leonard, senior analyst, Sona Asset Management Delphine Reymondon, head of liquidity, loss absorbency and capital unit, European Banking Authority (EBA) Michael Roper, portfolio manager, PGIM Jordan Skornik, senior portfolio manager, UK, Amundi Neil Day, managing editor, Bank+Insurance Hybrid Capital, and moderatorPlease click here to download the roundtable in pdf format.
Neil Day, Bank+Insurance Hybrid Capital: Inflation and monetary policy have dictated market direction in 2023. Will 2024 be more of the same?
Vincent Hoarau, Crédit Agricole CIB: Looking back a year, in December 2022 markets were expecting the first rate cut to be in late summer 2023. Clearly the reality was completely different. Only a few weeks ago, in October, the market suffered an ongoing increase in the term premium on US Treasuries. So to answer your question, everything can happen and we are in for another wild ride when it comes to monetary policy in the next 12 months.
Central banks have certainly made tremendous progress in cooling inflation — this is what has fuelled the recent phenomenal rally in credit, in rates and in equity. But we are not there yet. Yesterday’s US CPI signalled a bumpy road ahead regarding when the inflation target will be reached. Based on the latest GDP figures and job data, the US economy continues to cruise at a very high altitude. In Europe, the situation is completely different — economic datapoints are weak, particularly in Germany. Indeed, there is a narrative whereby the ECB could be in a position to cut rates before the Fed. But we have to bear in mind that, in contrast to the Fed, the ECB has a single mandate, which is price stability, and we are by no means at the 2% inflation target yet — the next print might be back around the context of 3%, taking into account some base effects that are perhaps not at the forefront of everyone’s minds.
When it comes to the US, following the latest CPI and NFP figures in recent days, if we have a similar set of reports again, we will potentially see markets significantly repricing the rate curve in Q1. That’s one of the main risks for markets in the early part of 2024. Our economists do not believe the Fed will be ready to cut rates before we see a series of negative non-farm payrolls — clearly we are not there yet and the earliest this could potentially happen, according to our colleagues, will be summer 2024. So we have a much more hawkish view with regard to central bank action in 2024 versus what the market is currently pricing, which is a first wave of rate cuts as early as March or April. This sounds a bit like science fiction, particularly if we look back at what happened over the past 12 to 15 months. To me, the extremes would be anything from zero to as many as six or seven rate cuts — again, everything is possible. If we try to sequence 2024, H1 is potentially going to be a repeat of H2 2023. H2 2024 is very likely to be more critical, in the sense that the rate cut cycle will very probably have started, together with a negative equity-bond correlation that can obviously be tricky for markets. Quantitative tightening will be well underway — unless we have a serious credit event or market shock in the first half.
In any case, bondholders are much better positioned going into January 2024 than was the case in January 2023: we have reached peak rates. The worst that can happen is a long plateau, with rates staying at relatively high levels for quite some time, but monetary tightening is definitely behind us.
So I’m quite convinced that when it comes to the primary market in the early days of 2024, people will be chasing highly rated and quality longer-dated paper to avoid ending up chasing such bonds in the secondary market in Q1. We are therefore quite constructive. The carry is great and it’s very likely going to be a year of good capital gains — however, it will definitely be another chaotic year, because rate volatility will remain intact.
To complete the picture, maybe just a few words on what central banks still have at their disposal in the toolbox to decrease excess liquidity. We all know that they, and particularly Christine Lagarde, feel committed to doing this, and it will be important to carefully monitor when the ECB will stop the reinvestment of redemptions under PEPP — the last active asset purchase programme — as well as the evolution of the ECB and Fed balance sheets. Nonetheless, reducing excess liquidity remains a great challenge for the ECB.
Raoul Leonard, Sona AM: In this context I would note that 2024 sees an election cycle where 60% of the democracies of the world are going to the polls — including most importantly the US, of course. It’s going to be a politically-charged environment, and if the US is going into an election with the economy strong and fiscal spending at super-high levels, and everyone is trying to buy votes and to avoid recession, that raises the question whether or not the Fed will go ahead and cut into that. That makes me a little more bullish on the market, although it could mean rates staying higher for longer.
Filippo Alloatti, Federated Hermes: Indeed, it seems that the bond market is finally believing we have peak rates — having been wrong five or six times over the last 18 months. And I agree, there are forces that speak in favour of disinflation. When we consider what could derail this process — and noting Raoul’s point on the politically-charged year — it is a potentially unexpected increase in demand, either in the US, the UK, or even mainland Europe, for whatever reason — governments spending their way into the next election, which is typically the case, or because people realise that the so-called cost of living crisis is not as harsh as they were expecting it to be, for a number of reasons, because mortgage rates are coming down, because house prices haven’t collapsed as was forecast, and maybe because parts of the population abandon the sort of pessimism that took hold over the last 18 months. And then there is the price of oil and gas. Last year, potentially big increases were anticipated, but this year oil prices dropped. This could, however, put pressure on OPEC to do something more radical in terms of supply — the market has always focused more on demand, anticipating a bigger or smaller slowdown/recession, but we should not forget about the supply side.
Jordan Skornik, Amundi: We could also get a positive demand shock, because real wages are back in positive territory. Consumption has been very resilient over the last year thanks to savings. And in the UK, for example, we have just seen mortgages increasing again. I think that the last bit of inflation is going to be probably the hardest to tackle. So while the market is pricing in quite a lot of cuts, it feels like we are going a bit too far, a bit too fast. We can feel there is still a lot of tension on this front — every data point gives some food for the doves or for the hawks — so I suppose this volatility will still be apparent for some time yet.
Jenna Collins, Brevan Howard: I think issuers are worried about the second half, and that’s why we’ll probably see a good amount of issuance in the first half. We may make it to this no landing type scenario, but if we do get a recession, it’s probably going to be more apparent in the second half of the year. But given that everyone’s expecting something to happen in the second half, my guess would be that it will actually happen in March, like this year, or in early 2025.
Hoarau, Crédit Agricole CIB: Some issuers give us the sense that they are already looking into 2025 in terms of funding redemptions, so we may end up in a situation where many are trying to finish full-year 2024 funding plans no later than the summer before starting pre-funding 2025. I believe refinancing risk is a potentially serious topic for the year ahead — but not the early part, where we are very well positioned for the first half of 2024.
Day, BIHC: How do all these dynamics affect your positioning?
Collins, Brevan Howard: Going back to basics, why do we buy credit? Because it offers a pick-up to government bonds. Well, government bonds have come down in yield, so for now, and probably for the first half of the year, the implication is that you’re supposed to buy credit, you’re supposed to go down the capital structure — in good names — and we will probably have a continuation of the compression theme. To put some numbers on it, if Treasuries are at 4%, low 4%,and Santander AT1 are 9% or high 8s, I think you’re supposed to buy the Santander.
Regarding the economy, there is some evidence of a slowdown, but not much, and not in the US. So there’s this push and pull, as Jordan mentioned. And there’s a somewhat controversial view: what if we saw the mini recession in 2022, with demand then increasing again — because of the support of low petrol prices — but then maybe there’s a bigger recession is coming? We’ll see, but for now rates are kind of leading the way, so take duration, moving down the capital structure in banks with good profitability; you don’t need credit protection right now, because we’re not seeing evidence of a downturn yet, but I don’t want to be caught in low quality, low profitability banks.
Skornik, Amundi: But most of us here are credit investors; if you are just a duration guy, would you still be long duration from here? I’m quite happy to buy the all-in yield in the credits, but if I took just duration, I would be a bit cautious — tactically, at least for the next two or three months, maybe I would reduce duration a bit. But given I’m long credit, I’m quite happy to hold the duration — with a bit of protection; as maybe we don’t need protection, but you never know.
Alloatti, Federated Hermes: Just because it has gone so fast over the last month and a half?
Skornik, Amundi: We are pricing in so much for next year. Even as a relatively dovish house, what we are expecting is what is priced in by the market now.
Collins, Brevan Howard: I totally hear you. But this is a very consensus view, that we’ve come too far.
Actually, even though US Treasury yields came down from 5% in October, if we go back to, say, June, July, we were sub-4%. So I’m not saying we can go a lot further, but we can definitely go sub-4% in the near term on the 10 year. We’ll see what happens after that, but post-SVB we thought we were in a worse state, wondering what the repercussions would be, and now we feel a lot better.
Michael Roper, PGIM: If you look at an index level where, say, the dollar market is from an IG perspective, there does come a point where it feels like credit spreads are quite tight. Everybody’s got on this soft-landing narrative, but there are definitely tail risks out there. Perhaps people will start thinking in terms of all-in yield, and the percentage of that which is actually the credit spread, and ask whether they are being appropriately compensated for those risks — even though it feels like the duration point is still something people want to get on board with.
Skornik, Amundi: Buy subordinated versus the rest?
Roper, PGIM: I think that’s where the AT1 world is. In terms of spread, there’s still a conversation to be had, but as you go up the cap structure or into something offering more security, the all-in yield argument is a bit less obvious.
Skornik, Amundi: On the sub strategies, we are remaining very long down the capital structure, but on the main global corporate strategies, we are reducing here and there.
Alloatti, Federated Hermes: Taking not the 100 year view, perhaps, but a 16 year view: since the global financial crisis, the idea for the average European high yield manager has been to buy duration, no matter what. It probably worked well until March 2021. Rates volatility may have subsided, but it’s still there. In November, the only volatility that increased month on month was rates. I get your point about going from 5% to 4%, because Yellen misled, so to speak, investors in terms of the Treasury programme, and the macro data was the same before and after. But I think you maybe want to be a bit tactical in terms of duration: if you think it’s gone too far too fast, then maybe cut a bit, and wait for the next leg up before buying some more.
Leonard, Sona AM: Supply last year was obviously quite short, with everything being tighter and tighter, then we saw the first 10 year issuance coming in senior in November, and it’s done very, very well. I don’t speak for insurers, but is there a part of the investor base that is lacking that product and therefore quite happy to buy quite a lot of it in early January? Not that I particularly like duration, because we’re in a very fickle market, but my suspicion is that some of that supply will be well received. I’m particularly watching the covered bond market, which has been going through some massive palpitations — the reality is, the covered bond market has not been working in the way it should since the biggest buyer, i.e. the ECB, stepped out in July.
Alloatti, Federated Hermes: Would you say they are very cheap at the moment, versus the senior stuff?
Leonard, Sona AM: Yes, I would agree with that, but there are no buyers apart from bank treasuries, and with every new issue they’ll say, make it 5bp wider than the secondary market, which doesn’t really work anyway, so it’s a fake price. It looks like, come January, supply’s going to be quite big in the covered bond world, and no one’s quite sure how to absorb €170bn of new benchmark issuance next year when it’s just bank treasuries and maybe some faster money starting to look into it. Issuance has only been up to five years and I suspect they will try seven now. But if that doesn’t work, and they go back to maximum five years again, against the average duration of their books that’s not a healthy underpinning for the entire capital structure of the banking sector.
Hoarau, Crédit Agricole CIB: Actually what we see in continental Europe is that appetite for covered bonds from pure real money across jurisdictions has been expanding significantly over the past 12 months, to some extent offsetting the decrease in demand from bank treasuries, particularly German bank treasuries, which are not necessarily liquid for other specific reasons, namely concerns around MRR and deposit outflows.
Regarding investor appetite for senior supply at the long end of the curve in January, indeed, I’m very bullish. I fully agree with Jenna that you don’t want to be caught in low quality names, but I suspect that everything that is relatively highly rated in senior preferred or senior non-preferred format in the 10 year maturity is going to be very well received in January — obviously as long as it is priced correctly. There will be a kind of fear of missing out, which will support bookbuilding and primary market execution, particularly for prime European banks that aren’t going to be coming with a 10 year benchmark in euros every day. Dollars is a different story, because people will tend to be more bullish on duration in euros while still having a short bias when it comes to US dollar assets.
Roper, PGIM: The Crédit Agricole senior non-preferred issue that came at the beginning of that 10 year supply was welcome, in that the market wanted 10 year paper from a quality bank. The only pushback is when you look at the spread versus where Crédit Agricole senior preferred might have printed: notwithstanding how well it was received, is the market appropriately charging for non-preferred versus preferred?
Hoarau, Crédit Agricole CIB: If you look on an historical basis, the differential between senior preferred and senior non-preferred has been anywhere from 20bp, 25bp to 50bp, and indeed, in the current market context, it was 25bp-30bp, so potentially at the tighter end of the range.
Roper, PGIM: And when you look at spread curves, Crédit Agricole has issued a 10 year at a similar spread to where it would issue a six non-call five. Again, it’s this preference for duration, but on a spread curve basis, it’s very difficult to want to go out the curve versus sitting in the belly in terms of what you’re being paid for the credit risk.
Hoarau, Crédit Agricole CIB: I hear you, and the 135bp spread for 10 year senior-preferred made it a great trade for the issuer on a re-offer spread basis. But it was also a great trade for investors: people were already in a mood where they are trying to lock in decade-high coupons on 10 years — a record 4.375% in SNP in our case, it was in green format, and it’s trading 10bp inside re-offer. Everyone is happy.
Collins, Brevan Howard: Following up on what Michael was saying, even though I mentioned earlier that I generally like duration, my general view is more that of compression. I think people would go crazy for something like a Deutsche 10 year — it’s not a bank people love, but they don’t hate it; it never issues that kind of deal; and it would be super-cheap. Or a name like CaixaBank. That’s the kind of thing that will do really well — at least in the beginning part of the year. It goes back to the question of what we are getting versus government yields: if you’re looking at something that’s at a very low spread to government bonds, why not just buy the government bond?
Day, BIHC: Michael, what is the outlook for funding volumes?
Michael Benyaya, Crédit Agricole CIB: Looking at public issuances from European banks across all currencies, in 2024 we expect a broadly similar picture to 2023 in terms of total volume, but with a change in the product mix.
Starting from the bottom of the capital structure, we expect an increase in AT1 volumes. This will mainly be driven by the refinancing of calls in 2024 and the first quarter of 2025: we calculate there to be around €40bn of AT1 calls during this period. We therefore have an estimate of around €35bn of AT1 being issued in 2024.
For Tier 2, we expect volumes to be flat on this year, or perhaps a slight increase. The Tier 2 market has been broadly stable year on year over the past years, always around €30bn-€40bn euro-equivalent, and it will remain the same in 2024. All the banks are pretty well optimised in terms of the Tier 2 bucket and it’s just a question of maintaining this — we don’t really see incremental needs.
We actually expect a decrease in senior non-preferred/HoldCo volumes, of up to 20%, to around €150bn. This may seem quite a lot, but if you look at the vast majority of European banks, all of those that had the 2024 MREL target, they are they are well optimised in terms of MREL, having built up strong buffers to minimum requirements, subordinated and total MREL. Therefore, our feeling is that the senior non-preferred market will switch from the build-up phase of recent years to a refinancing market, with lower issuance needs.
Moving on to more liquidity-type products including senior preferred and covered — even though senior preferred is also used for MREL purposes for a good number of banks, especially smaller banks — for senior preferred, the overall picture will be flat to a marginal increase, up to €175bn. What drivers do we see here? On the one hand, the last part of the
TLTROs will effectively be repaid next year, so that could drive up supply a little. On the other hand, we have relatively muted loan growth in banks’ balance sheets. In terms of deposit bases, these are holding up quite well in Europe — there are no massive deposit outflows, no leakage. Therefore, for us, it is more a question of the cost of deposits than deposit outflows, and we don’t think deposit outflows will drive up supply in 2024.
And for covered bonds, we expect around €180bn.
Roper, PGIM: Are you assuming basically flat risk-weighted assets, because loan growth is fairly anaemic?
The deposit piece is interesting, because when you go back to the discussion about central banks, and the propensity of central banks to shrink balance sheets, one of the arguments of QT is sucking liquidity out of the banking system, but I still think there’s a bit of a question mark over the transmission channel and how this will play out.
Benyaya, Crédit Agricole CIB: Happy to hear other views, but in terms of risk-weighted assets, our anticipation is that RWAs will probably increase a little bit, but not that much, and the impact on supply will be marginal. We don’t really see risk-weighted assets moving quickly in Europe. Looking at the latest results, everything appears to be well under control, and we expect that to continue in 2024. We know that there are pockets of risk — commercial real estate is one of them and maybe we’ll come back to that — but the overall picture is broadly stable.
In terms of the deposit base and looking at the results of European banks, again, I think it’s more a question of increasing the cost of deposits rather than seeing deposits leaving the balance sheet of banks. And we are far away from the situation that we see in the US — I’ve never believed that we could see something similar in Europe, because actually there is no nowhere else to go. If you look at life insurance in the major countries, it’s more managing outflows rather than attracting inflows. I don’t feel that retail clients in Europe are willing to go into the money market. Therefore, if you look at the numbers, deposit bases are holding up super-well for European banks.
Leonard, Sona AM: The only country where there has been some deposit reduction is Sweden, where I believe that QT is somehow more linked to deposits.
Collins, Brevan Howard: This is an important point and I’ve been looking into it. What’s important is how the banking system and individual banks are run in terms of loan to deposit ratios. For systems like Sweden and the UK that are at 100%, when the deposit leaves, they need to fund it in a different way. But in Italy, which people always worry about, where it is only around 75%, they don’t. So it’s worth looking, by country and by bank, at how much of the balance sheet is fully-funded versus loan to deposit ratios below 100%.
Leonard, Sona AM: The reason there is this fear is obviously Silicon Valley Bank (SVB) and the speed with which liquidity moves now. They have a slightly different deposit system in the US, but after SVB — and with Credit Suisse, too — the liquidity risk questions from everyone went through the roof. How does LCR work? etc. And we saw lots of bank research analysts doing deposit-watch products, the kind of thing we last saw in the Spanish crisis or the sovereign risk blow-up in 2011. That’s died away a bit, because deposits aren’t moving.
Day, BIHC: We’ve also seen social media amplify volatility and accelerate banking crises — how should this be tackled or addressed?
Delphine Reymondon, EBA: This topic is being considered at the Basel and FSB tables, too. It’s clear that the role of social media and digitalisation has triggered reflections, on whether deposits are as sticky as they were in the past, whether we should amend some rules, in particular from a liquidity perspective, and what it means for contagion risk, which is far higher. In terms of liquidity standards, some might consider it appropriate to amend some outflow rates, for example, or aspects that are not covered as such by the LCR, such as intraday liquidity risk or prepositioning requirements.
But another perspective is also that effective implementation of existing standards and strengthened supervision need to be exercised in full first, before we examine whether some regulatory changes are needed. The Basel Committee issued a report in October where it presented a cascade: you start with the business model of a bank, including the risk management practices and the risk appetite in particular, then you go to strong/effective supervision, and at the end you have robust regulations. When it comes to SVB, undoubtedly the business model, among other things, played a role. As EBA, we have published a lot of guidance for supervisors and for banks via our regular liquidity monitoring reports on the LCR in particular (we started in 2019). Also, the LCR does not stand on its own, there are additional monitoring metrics that exist complementary to the LCR, providing detailed information for example on rollover of funding, concentration by product type or counterpart, mismatches in terms of maturities, you also have information via ILAAP, stress-testing, contingency plans, etc — plenty of tools for supervisors.
The question is, suppose we amend some aspects of the LCR rules, are we certain that this would fill the gap in terms of increased impact of digitalisation or social media? The objective of the LCR is to protect the bank for a 30-day window, so that the bank and the supervisor can take some measures, if the bank is in trouble. And anyhow we would still need to address issues with the usability of the buffer. If the question is, will amending the rules better protect a bank in case of a deposit run? The question remains open for debate.
These are all part of the reflections underway, other aspects are also investigated. In any case, even if Basel were to amend the standards, it would take some time before entering into EU rules. So for the time being supervision is the best tool that we have. Supervisors are already shifting their priorities from capital to liquidity, anyway.
Roper, PGIM: Going back to covered bonds and the ability of the market to absorb what looks like being a very busy year — partly because of TLTROs — at what point does the issuer say, instead of issuing external covereds, I will instead post with one of the central bank’s other facilities for that liquidity?
Hoarau, Crédit Agricole CIB: A tough question and I am not sure I can give you a straightforward answer. Retained covered bond ECB funding is short term funding since TLTROs no longer exist, and covered bonds are term funding, so you’d only go to the ECB to wait for better market conditions rather than arbitrage the cost of retained covered bonds and benchmark covered bonds. The cost of the differential between the two takes into account the differential between the ECB’s MRO and three month Euribor, with the haircut of the retained covered bonds factored in. The calculation implies the comparison between a blended spread level for the retained bonds versus three month Euribor to covered bond wholesale funding versus the same Euribor benchmark.
There’s also a relative value question, namely the cost of issuing covered bonds and putting the collateral at investors’ disposal, versus how much senior preferred funding costs. And this is where views can differ from one bank issuer to another. We also have more and more investors asking us to help them track the covered bond-senior preferred spread differential, something few people did seriously over the past 10 years because covered bonds were trading at silly tight levels. Today, if a strong French bank was to come to the market it might have to pay a minimum of 45bp to ensure that it could safely place a 10 year covered bond, while a 10 year senior preferred from the same bank would be in the very low 100s. This comes back to your comment about the spread differential between senior preferred and senior non-preferred looking very tight. That means senior preferred is attractive for credit investors, and potentially expensive for issuers, but looking at the relative value versus covered bonds, senior bonds are only 60bp back from secured issuance levels on the same part of the curve. And covered bonds looks cheap versus senior preferred. The pricing paradigm across covered, senior preferred, senior non-preferred has been moving around quite a lot over the past 12 months. I don’t think we are yet at the point where we can say, OK, this is the right spread between the two, and I’m quite sure it’s going to take another couple of months, if not a good year before we can say that the relative value scheme across these three funding instruments have settled down for core European names and makes sense.
Leonard, Sona AM: TLTROs could also be cash collateralised, so a lot of banks have been able to reduce their TLTRO amount, by cancelling off the cash they deposit with the ECB. So if they have to pay back €50bn, the impact isn’t that they need €50bn of extra funding; it might be just €20bn.
Reymondon, EBA: We have been having a look at this TLTRO exit from an LCR perspective. We have already published a report in June, and another is imminent. When we published our first report, the conclusion was that it is manageable on average for the EU banking system, but of course, there might still be some individual cases where supervisors would need to have a close look. And it is important to monitor the different movements in the balance sheet of the banks. For example, cash or central bank reserves is going down, so some of HQLA is going down, how is this compensated or not in terms of outflows? There are certain categories of deposits that are exempted from outflow rates, or other deposits with favourable outflow rates, like operational deposits. We are monitoring the different movements. Our liquidity monitoring reports give guidance to banks and supervisors, for example, on how to handle specific aspects of the LCR, where there is some room for judgment on outflow rates to apply to certain categories of deposits, and how to categorise these deposits. We will follow up with supervisors on how this guidance has been applied in practice. We will also see in the coming months if there are changes to the minimum reserve requirements.
As Jenna mentioned, loan to deposit ratios are also playing a key role. All the EU markets structures and competition aspects are different in that regard.
Alloatti, Federated Hermes: When you look at the LCR by currency in December, it makes for very scary reading.
Reymondon, EBA: Indeed, the situation is different by currency compared to euros. We have some detailed analysis in our forthcoming report.
Roper, PGIM: Normally disclosure is a good thing, we have asked for more of it, and banks were very willing to disclose LCRs when they were awash with liquidity. But given how complicated the LCR is, and how reliant it is on various moving parts and assumptions, it worries me slightly how the market is going to react as liquidity is drained from the system and we start seeing LCRs that maybe optically don’t look fantastic. That disclosure could end up being more of a hindrance than a help.
Reymondon, EBA: That is a very good point. What the EU and UK banks have been doing in general is disclosing a lot of information on the LCR, which generally goes beyond the pure regulatory requirements. In the EU, back in time we intentionally took a different approach from the Basel standards, seeking less granular and less frequent LCR disclosures — under Basel it is on a last quarter basis with daily averages, but we said we want last year with monthly averages, precisely to leave some leeway for banks to manage their liquidity needs, especially if they were in trouble, without everything being visible, otherwise they immediately face a stigma. But because there was ample liquidity, and also due to external factors like expectations form credit rating agencies or investors, banks have tended to be willing to disclose a lot. I remember discussions with some banks a year and a half ago, asking what they would do if or when liquidity becomes less ample, because it would be very difficult to go back to less disclosure…
Leonard, Sona AM: And the LCR is just one view of liquidity. I’ve worked in a bank treasury and used to manage liquidity risk. In reality the loan to deposit ratio is another aspect, then the quality of your deposits and the behavioural aspects of them, etc. But it’s also your ability to raise deposits. Because ultimately liquidity is your ability to meet your obligations in the requisite volumes and at the right price. So pricing is a key part of liquidity risk, but it’s not given enough consideration. Ultimately, liquidity risk has got so many facets that any focus on any one of them at any one time is probably wrong. How do we tackle that when it’s actually the most important thing?
Day, BIHC: Moving away from what in balance sheets is driving supply to the expected issuance itself, how well do you anticipate it being absorbed? And what parts of the capital structure are you likely to prefer?
Skornik, Amundi: We are quite in line the figures that Michael mentioned. Overall, I see the main point as being that net supply will be relatively limited, especially for AT1s and Tier 2. That’s quite important, because post-Credit Suisse, the big question that everyone had was, who’s going to be the buyer of this kind of debt? And whether it would find some support. On day one, a lot of hedge funds came in and bought opportunistically. Since then, demand has been quite good overall. But when you look at specialist funds, the demand is not there. With one exception, I didn’t see anyone get significant inflows this year. So it’s quite important for next year that we don’t get too much net supply if we aren’t seeing more demand. We are expecting some demand. On our side we are seeing a bit of a pipeline, with people starting to return. They’re probably a bit late — we’d been advising them to come back earlier — but especially given that they still enjoy some relative value, with other instruments being more priced to perfection, we think that we’ll see some inflows into the funds. So overall we should be able to absorb any supply, especially if it’s net negative. But it’s quite important, at least in the early part of the year, that net supply that isn’t too positive.
Collins, Brevan Howard: As I indicated earlier, I like Tier 2 and AT1 in decent banks. Investors who are looking at AT1 maybe need to remind themselves that it’s not a good idea buying the AT1 of banks who fund themselves at very high levels in senior and/or are not making any money. But if they’re profitable, have low senior funding costs, that’s probably fine.
Skornik, Amundi: There are counter-arguments to this — the credit profile can also improve. Take BCP, for example: their AT1’s levels were very attractive but in line with only “OK-ish” capital metrics. I believe the investor base was also limited. Now that their figures have significantly improved, the demand for the instrument has certainly grown.
Leonard, Sona AM: The most important thing Jordan mentioned there was net supply: if you looked at net supply across the asset classes, you’d see a much smaller number. And if rates are soft to start with or are heading down, then I think appetite will be fine.
Alloatti, Federated Hermes: But who is absorbing all the €170bn of covered bonds?
Leonard, Sona AM: Actually, the difference in covered bonds is that net supply will be something like €50bn.
Hoarau, Crédit Agricole CIB: The question is how you calculate net supply in covered bonds. If you want to capture the impact on the primary market, you need to factor in the fact that there has been no asset purchase programme in place buying covered bonds since July, and CBPP3 redemptions of some €32bn next year means investors will have to absorb a net €70bn-€80bn of euro benchmarks.
Indeed, while I tend to agree that the net supply element is quite favourable for the junior parts of the capital structure, I would be more concerned when it comes to overall supply and overall market absorption capacity. Factoring in what is coming from the SSA world, including government bonds, and what is coming from the covered bond world, and discounting the fact that we will be without the safety net of an asset purchase programme for the first time since 2009, we will be facing quite a new situation in January 2024. That doesn’t mean that we are not positive or constructive towards next year with regard to market absorption capacity, but this is one element that I would put on the risk list for 2024 — it could be bumpy, and there will very likely be further pricing adjustments. And then we need to look at the swap spread, we need to look at what is happening to German Bunds. So it will not be that easy. At the end of the day, many asset managers may say, I’d rather sit on an IG-rated AT1 or Tier 2 from a prime core bank than anything else. It’s easier to sell a Tier 1 than a covered bond today.
Collins, Brevan Howard: One of the reasons we’re seeing this expensiveness in senior non-preferred is because it’s no longer being incrementally issued, there’s just a refinancing need. It won’t exactly be like Tier 2, but like Tier 2, we may see this vacillation — it gets very expensive and then very cheap because of this scarcity element — and the idea that senior preferred and covered bonds look cheap against it could persist for quite some time.
Day, BIHC: What is the outlook for banks?
Roper, PGIM: Bank fundamentals are really strong. You always have the argument, have you reached peak? But there’s plenty of room for bank fundamentals to deteriorate and still remain incredibly strong.
On the P&L, the biggest, or the most sensitive, line item is going to be pre-provision profit. I say that because you probably saw an outsized reaction in equity with the likes of the UK banks when they were missing NIM, but does that feed into credit spreads? It’s still very difficult to construct a story where as a credit investor you’re worried about some of these P&L trends.
The asset quality piece is one that gets the most attention and CRE is clearly the number one risk factor. But actually when you go through bank by bank — away from some specialised CRE lenders — it’s quite difficult to find any with hugely worrying CRE exposures. You’re normally sucked back into the Nordics, for example, but equally, the asset quality of some of the names in question has historically been pristine — look at Handelsbanken: everybody’s sitting there saying this is impossible, but they’ve done the impossible for 10 years.
Then you come to capital and liquidity. Again, the capital positions of the banks are really strong. And when you think about the amount of shareholder distributions that banks expect to make, especially those doing buybacks, they have plenty of flex to manage the capital base, and they’re still going to be generating organic capital. So that’s a comfortable situation. And then we’ve already discussed the liquidity piece. Whilst deposit flows were a big part of what happened early this year, it’s quite tricky to see a bank having a liquidity shock event where there isn’t a business model challenge. Then you’re again talking about some very small banks with specific issues as opposed to there being an industry-wide issue.
The net-net of all of that is that even though people talk about the macro and the impact of higher rates not yet feeding through into things like CRE, it’s still very difficult to get too concerned about bank fundamentals.
Leonard, Sona AM: The implied cost of equity of European banks is extremely high. If you take the consensus price to book value — let’s say everyone thinks it’s going be 0.5x — and then you look at the implied cost of return on equity they’re expecting — for something like Barclays — the market’s expecting roughly 10% RoE. So you end up with a 20% implied cost of equity, which is extremely high relative to history. Then have a look at US banks — take a US regional bank, which are vulnerable if rates keep going up, they have implied cost of equity of something like 12%. So there really is a big problem with European bank equity. It’s like the investor base has been scarred from years of clean-ups, crazy rules on dividend stopping, and all these kinds of things.
Alloatti, Federated Hermes: And also you have Fannie and Freddie in America — there is no equivalent in Europe.
Leonard, Sona AM: But I think part of the UK bank equity volatility during 3Q23 results was around hedge fund or pod positioning in equity. That’s why the stocks moved c.10% on results day and there were no buyers on the break. It’s just not normal, and feels like these equity markets have kind of broken. So when NatWest missed on NIM in Q3 and its stock price dropped 10% in a few hours, credit didn’t really move, and then the next day the stock dropped again another 4%, which felt like positions being unwound, there were clearly too many people long the stock. So that didn’t create an impact on credit, but it arguably could if we see anything a bit strange in full year results that prompts equity movements — we might see a bit more CRE damage, we might see some change in NIM outlook, particularly because full-year accounts are the most stringently audited. These won’t necessarily reflect the fundamentals from a credit point of view, but I tend to watch cost of equity because it could become a lightening rod into liquidity, for example. But fingers crossed it won’t.
Roper, PGIM: Or unexpected surprises. Take Signa: you think you know which banks have exposure, but were a bank to come out and suggest that they have exposure the market wasn’t expecting, even if it is very manageable within the size of balance sheet and earnings, I think the market would take a particularly tough view of them and question perhaps some of the underwriting lending standards, as opposed to being too worried about what it ultimately means for fundamentals.
Alloatti, Federated Hermes: If we go back 16 years ago, a normal half-decent European bank would have a cost of risk between 50bp and 100bp, and that’s before IFRS9, etc. Today, if a bank gave a cost of risk of 50bp, everyone would say, oh, what’s going on there? That’s a bit scary as it seems out of sync with economic reality.
Roper, PGIM: A lot of banks still have management overlays they can also release to cover some of the credit-specific cost of risk.
One area that’s interesting is, if you have banks that have loans backed by property, typically the provisioning or coverage is much lower, because you have recourse to the collateral, which all makes sense up until you start questioning the CRE valuations. If you have a bank saying it’s got an NPL but the loan to value is 50 so therefore it doesn’t actually need to provide against that, you would take comfort if they could actually work through those NPLs — and the system has done a very good job in bringing down NPLs. But now there’s this question mark, with everybody knowing the CRE market is slightly broken, and it doesn’t take a lot to ask, actually how adequate are the provisions against some of those large ticket items if you simply can’t move the collateral?
But then if it then turns from a specific to a systemic issue, I think all the central banks cut aggressively.
Leonard, Sona AM: It’s interesting, because you mentioned the pre-provision profit line, and with NII, it’s really high now because rates are higher. The worst combo would be cutting rates to zero again, because then all that cushion would disappear, and then there might be an asset quality wave. So in a strange way, if we stay at an ECB rate of roughly 2.5%-3%, that would be ideal — not too hot, not too cold, just right, temperature of porridge.
Alloatti, Federated Hermes: With all due respect to the policymakers, even the ECB is now admitting that negative rates were negative for the banking sector, something they never said for three or four years. Reading between the lines, they didn’t actually say it was a tax on the banks, but they said it had a negative impact on the banks. So in theory, they should say, we’re not going to zero.
Leonard, Sona AM: In fact the best combo is, keep rates around 2.5%, demand comes back, because corporates and individuals realise the world’s not ending and they can invest, and then we accept a world where rates are slightly higher, banks are profitable but can lend because they’ve got lots of capital, and then we have more of an American lifestyle where we make money.
Collins, Brevan Howard: It’s almost like a renormalisation, going back to before all the quantitative easing, where banks can make more money. They have much more capital than they did and maybe they don’t grow as much, which the equity markets don’t like, but they are now moving from one level of profitability to another, perhaps higher level, that then persists.
Leonard, Sona AM: But we haven’t mentioned sovereign risk. As long as governments behave and fiscal rules are followed then that’s all fine, but if we have a sovereign risk blow-up, which is another tail risk for next year, then that all changes.
Day, BIHC: Turning back to the instruments again, there are perennial questions around the structure, behaviour and purpose of AT1s, but this past year has only heightened discussions around the instrument. Delphine, what is your current perspective?
Reymondon, EBA: Yes, it’s interesting how these AT1 discussions keep coming back.
In October the Basel Committee came out with a report on analytical work underway on liquidity, IRRBB, and AT1, with the message that it does not mean that there will be revisions to the framework — which is an important aspect of the communication. The aspects of AT1 that are being debated are, for example, that investors perhaps did not appreciate well the risks of the instruments. I am not certain this was completely the case, because the terms and conditions and risk factors are quite clear and detailed, but maybe it is true of the hierarchy of creditors between the different classes of capital instruments, hence the clarifying statements that several jurisdictions published in March, including us as EBA together with the SRB and ECB. We said in this communication that we support the role of AT1 in the capital structure of EU institutions. At the same time, what is clear is that these are complex instruments, with a low trigger — it was defined before the resolution framework — and it is quasi-impossible to cancel the coupon payment. And there has not been enough testing of the loss absorbency of the instrument in a going concern scenario, as mentioned in the Basel report published end of 2022, so the debate over whether AT1 is a going concern or gone concern instrument is still taking place. AT1s are meant to be going concern, this is the way they are structured in the regulation. But this might not be as relevant a question as it was in the past. What might be more relevant now is whether they can absorb losses and create CET1 or not when needed, when you look at Credit Suisse AT1 instruments have generated a lot of CET1. We have seen also recently with Metro Bank that Tier 2, supposedly a gone concern instrument, can even be touched in a going concern, too. So the division between these two concepts might not be as relevant as in the past. What we need to ensure is that the loss-absorbing parts of the capital structure are there when needed, according to a certain hierarchy, and that the eligibility criteria remain stringent. This is why we have the EBA AT1 monitoring reports that we have been doing from the very beginning, starting in 2014, where we recommend what should and should not be included in the terms and conditions, which clauses can and cannot be used. This stringency in the effective application of the eligibility criteria goes hand in hand with support for the asset class. And we are satisfied that when we publish a report, the recommendations are implemented by the banks, well followed by market participants. Also, AT1 instruments can bring additional funding to the banks and access to a different class of investors. In a nutshell, they are complex, they are not perfect, but they have some merits. But we’ll see what happens at the Basel table and if some regulatory changes are decided, and if these changes could come without rethinking all the buffers framework, which is certainly highly complex.
Alloatti, Federated Hermes: This debate has not gone away because it is important to us as investors. You should make the instrument a bit more fixed income friendly. I think there is a bit of a misconception around the extent to which AT1 is bought by equity investors — I think it’s only marginally so, 5%-10%, at most.
I have talked to the ECB, to the PRA, and I don’t understand why there is no dividend stopper. You said you respect the capital hierarchy, but if you say so, then in my view you should fight a battle in order to reopen CRD and introduce this dividend stopper. What you said about no one skipping the coupon is not quite right, it is a misconception, because the coupon is optional. I’m not sure why anyone would, but in theory they can, and it’s very dangerous because we know that tomorrow a bank could decide not to pay the coupon, so why not introduce this dividend stopper?
And then if you want to go further, make the coupon cumulative. I know it’s very difficult, but while the instrument is not perfect, this is not reason enough to stop trying to make it more perfect.
Reymondon, EBA: We also heard these arguments in the past. It is not advisable to touch the AT1 class, especially at the moment, and especially if it is to water down eligibility criteria, so as to maintain credibility of the instruments. Dividend stoppers would probably not have made a difference in the Credit Suisse case. Bear in mind that there is one case where the hierarchy would still be inverted, namely if you hit the regulatory trigger. You might say, it’s very unlikely, because the trigger is too low, but in this case, the hierarchy would be inverted. All these dividend pushers, stoppers, reverse stoppers etc and things we saw in the past were similar types of mechanism that, in a way, created more complexity, and we wanted to introduce more simplicity. Already at the time of CRR1 we received comments that no one would buy the instrument with all the aspects that we designed — like how it would take ages to get a write-up on the instruments for the ones with temporary write down — but in the end, there are still quite a lot of buyers of the instruments.
Benyaya, Crédit Agricole CIB: I agree with Delphine in the sense that today I feel that we need stability in terms of AT1 criteria. The Credit Suisse write-off was a shock to the market, let’s not forget that. Now the market has restarted, I feel we are seeing investors coming back to the asset class, and we need stability. And to me, all the discussions around dividends stoppers, pushers, coupons cumulative or not, I don’t think it will massively change the behaviour of the AT1 instrument.
Skornik, Amundi: It could alleviate some pressure sometimes. During Covid there were a lot of questions was around this — and in practical terms, we actually had dividend stoppers. The question is, so this is a perpetual bond where they can stop the coupon and still pay a dividend forever? Even if a CFO would not do this, why not give investors some comfort that it is definitely not going to happen.
Benyaya, Crédit Agricole CIB: Remember that many European banks declared that they would observe the hierarchy between dividends and AT1 coupon payments.
Skornik, Amundi: But you need to have this confirmation.
Benyaya, Crédit Agricole CIB: Yes, exactly. So it’s not in the documentation; it’s like a management policy.
Alloatti, Federated Hermes: So why not put it into the documentation and the law? Because personally I think the AT1 market is one accident away from being a museum piece. There is demand today, but if something were to happen in the next six months, I don’t think the market would reopen.
Leonard, Sona AM: It depends on the price. That’s my view, always. I was buying right after CS.
Skornik, Amundi: The demand is not the same demand as in the past. The specialist funds, they haven’t seen demand following CS.
Leonard, Sona AM: UBS did quite an interesting thing in the documentation of their recent AT1: their new prospectus was a sea change in explicitly explaining why this instrument can be bailed in. And I quite enjoyed the read. It basically said: you know everything that happened to CS AT1, well we are now making explicit how this can go wrong for you as a holder — caveat emptor. That was the right approach to explaining it — and selling bonds a bit cheap helped, of course.
Collins, Brevan Howard: As nice as it is for a credit investor to have dividend stoppers and whatnot, you need to be able to raise equity in a distressed situation, and so you need to be able to entice equity in any way possible. People just need to know that if you buy capital, make sure your bank is profitable and funds itself cheaply, and if you’re not comfortable with that, then you should be in senior.
Leonard, Sona AM: I’m a bit more laissez-faire. l really think banks should be allowed to fail more often, because when banks fail — where it isn’t the fault of the regulator — then the other management teams realise there is a cost to managing poorly. A lot of banks have gotten away with things in the last 10 years where they should have been bailed in or gotten rid of, which would have made behaviour improve, rather than giving them more rules. And management teams should be punished by losing their jobs or more when that happens. Saving everything all the time is not the right approach.
Alloatti, Federated Hermes: That works in theory, but in practice it’s a political decision.
Day, BIHC: What do the latest issuer and regulatory actions imply for capital instruments call management?
Reymondon, EBA: There were previously more liability management exercises in the UK and far fewer in the EU, and we were regularly asked why, whether the EU regulator was stricter on some of the rules allowing for this. But the situation has changed somewhat and the trend that we see now of having more refinancing before five years is welcome. We have always monitored how supervisors have been exercising the notion of “exceptional circumstances” where you can redeem or repurchase the instrument before five years, and the view among supervisors was that a lower cost would not be considered a sufficient factor to be deemed an exceptional circumstance. In all cases, this assessment needs to remain a case by case basis, which is why the EBA did not want to further define what exceptional circumstances should be. There are plenty of different elements that need to be considered: the specific situation of the bank, its size as an issuer, the frequency of its issuances, the impact on other classes of instruments, access to the investor base etc. What we have tried to do from an EBA perspective is, firstly, to provide more flexibility where possible. We did this for example in pushing back in time the point of the deduction when the bank has received the prior permission from the competent authority to call or redeem an instrument. But very importantly, we continue to carefully monitor all these transactions. We ask supervisors to present to us under which conditions they have allowed some of these transactions and we monitor the treatments applied by the banks and communicated to the markets. We still need to pin things down a bit and to ensure that there is convergence in practices among supervisors.
It’s appreciated that there are more LMEs. They are particularly warranted in the current market conditions, because after all, we may have more stable conditions today, but uncertainty is still very high and we do not know what might happen next year. If banks have windows of opportunity, it’s good that supervisors can find ways to warrant the refinancing — as long as it’s not like one year after the issuance, obviously. But we need to and we will continue to issue Q&As or additional guidance to market participants and supervisors where necessary, again, to ensure convergence and consistency.
Roper, PGIM: Are you’re having similar discussions as well on Tier 2?
Reymondon, EBA: It’s for all instruments — it’s just that AT1 is the one that is more often in focus. We would need probably to work a bit more closely with resolution authorities going forward, because the permission regime for them is more recent for them, compared to supervisors, who are now quite used to our guidance.
Day, BIHC: Jordan, did you have any thoughts on call management?
Skornik, Amundi: I think we have become used to this call, non-call issue, at least on our side. And we have benefited quite a lot from it, having been able to do some relative value trades around it, so I’m happy for the opacity to remain. And it’s funny you mention Tier 2, because while we have long done these relative value trades between AT1 based on resets, we never previously thought about doing the same for Tier 2, but are now doing so — maybe a bit less today, but last year there were plenty of opportunities where people were buying something that to maturity was as good as a bullet Tier 2 but with the optionality for you.
Roper, PGIM: I was more thinking about liability management on Tier 2s, and into a higher cost instrument. How far ahead of call would a regulator or a supervisor be happy to allow a bank that maybe doesn’t have as regular access to capital markets to be able to do that type of transaction?
Reymondon, EBA: I mentioned that it is probably too stretched if after only around one year you already want to refinance. We don’t have criteria prohibiting this, and it’s more for supervisors to assess, but it should not be so far in advance compared to the first call date.
Leonard, Sona AM: Some banks’ CET1 ratios and now so high that, Tier 2 hardly matters as a bucket. I guess it’s cost effective, but at some point shouldn’t they just be allowed to call based on where their CET1 last printed? And their future profitability? To me, going to the school teacher every time you hand in a bit of work just seems a bit much; shouldn’t you let them be grown-ups and say, you manage this?
Reymondon, EBA: There are capital planning and funding aspects to it, too. You might also want different types of instruments to match your funding structure. So I’m not sure banks would always be willing to call Tier 2 instruments in advance even if they have high level of capital in other classes/layers.
Benyaya, Crédit Agricole CIB: And sometimes they are also needed for rating agency purposes. Regulation is one driver of capital, but the ratings agencies are another, and it’s true also for Tier 2.
Day, BIHC: To what extent is CMDI and the potential impact on depositor preference and senior preferred versus senior non-preferred something to watch?
Benyaya, Crédit Agricole CIB: At this stage CMDI is still a proposal from the European Commission and the legislative process is still going on. It will probably happen, but there is still a lot of work to be done before it can become a law that can be implemented.
The proposal today is indeed to make all deposits super-senior in the capital structure — so even going beyond what we have seen in Italy and Portugal, for example. In that situation, senior preferred will be in a different position, because there will be only a single stack, with only senior preferred, and below that senior non-preferred. We know very clearly that there will be a rating impact, because Moody’s communicated that there will be a pretty much automatic one notch downgrade on the senior preferred of the vast majority of banks, losing one notch on the senior preferred.
The current proposal would keep senior preferred pari passu with derivatives, so senior preferred will not qualify as subordinated MREL. Therefore, senior preferred and senior non-preferred will continue to exist with different positions in the liability structure and playing specific roles, senior preferred for total MREL, senior non-preferred for subordinated MREL.
Such an implementation of depositor preference in Europe would, of course, have implications for the senior preferred versus senior non-preferred relative value discussion.
Roper, PGIM: I can see the argument that’s pushed, that senior preferred and non-preferred spreads will have to tighten, because technically you are now ranking junior to depositors. My issue is that to me, senior preferred is a funding tool, is for liquidity, and senior non-preferred was designed as an effectively gone concern instrument, and it’s when you start narrowing the gap between the two that the problem starts: I still think it would be very difficult in a bail-in style scenario to impose losses through senior preferred, partly because of the complications around ranking pari with derivatives, and partly because in certain geographies, especially Italy, there are still a lot of retail bondholders in senior preferred, so you then have the political angle. It’s very valuable for the European banking system to have a funding tool like senior preferred, and if you start using it as a way of recapitalising banks, it then becomes more of a non-preferred style instrument. So I can see the argument as to why you would trade them tighter, but I personally still think that senior preferred should trade decently inside where non-preferred trades today, even with the depositor preference noise in the background. Take a very small bank that’s using senior preferred to meet its MREL: there are a lot of resolution tools available, so you don’t necessarily need to have bail-in — you could have a sale, you could probably liquidate it. It would be very difficult for someone to want to impose losses through senior preferred just because of the impact it would have on a lot of smaller banks and their ability to raise senior funding.
Alloatti, Federated Hermes: But conceptually, it’s possible.
Roper, PGIM: Conceptually, it’s possible. I just think that you’d lock a lot of small banks out of the senior preferred market, and do people want to do that?
At the moment, banks are at times a little disingenuous when they say, look, its senior preferred, but bail-in-able. You get tier two, tier three banks in Spain or Italy using senior preferred where actually there’s not a lot of securities junior to the senior preferred — you go straight to Tier 2 — and the question is, if you have an event at that bank, what on earth does a regulator do?
I would have much preferred MREL to have to be met with subordinated MREL. It’s much cleaner. And then senior preferred becomes your funding tool. Look at the Canadians now: they don’t have the two tiers of senior, just one that’s effectively for bail-in, and after that you have covered, so when the covered bond market isn’t functioning they have a real issue in terms of having an attractively priced instrument to fund.
I like how issuers such as Crelan say that they are going to meet their MREL with subordinated, it’s nice and clean. The way the Dutch and UK banks do it is really clean — maybe it just comes down to having the HoldCo structure.
Skornik, Amundi: At the same time, the current set-up maybe gives the smaller issuers a bit of time to build market confidence to issue senior non-preferred at some point.
Collins, Brevan Howard: Yeah, but things happens when we don’t expect them to, right?
Roper, PGIM: And then you have to transition your capital structure from one with senior preferred by refinancing with non-preferred, so it will actually take a lot of time before you’ve got to the end state where all MREL is met with a subordinated instrument.
Leonard, Sona AM: The ratings point was made earlier, and I believe RBI issued a senior non-preferred to protect their senior rating earlier in the year, which was their first attempt to put a wafer thin layer there. So maybe rating agencies will be part of that discipline.
Roper, PGIM: And if you had depositor preference in Germany, for example, why on earth should it make any difference to Deutsche senior preferred? They’ve got such a big stack of non-preferred, the size of the loss you’d have to generate to get close to the senior preferred is bonkers. So it’s very case by case dependent. For banks that are using preferred and have no non-preferred, you should be trading the preferred as a non-preferred today, and then for banks that have very, very large non-preferred stacks — your large French banks, your Deutsches of the world — I don’t see why depositor preference makes any difference at all aside from maybe having a mechanical rating impact.