Skandinaviska Enskilda Banken AB publ CEO Johan Torgeby on Q2 2020 Results

Thank you very much and good morning. Welcome to SEB's second quarter results call for 2020. You can find the presentation on sebgroup.com. Starting with macro on page two: we've seen a strong reversal from Q1 weakness in financial markets. The Stockholm index is up significantly year-on-year. Credit spreads have tightened, and interest rates remain very low. On page three, the economic effects of COVID-19 lockdown: Sweden has been less affected than other European countries in terms of mobility and domestic economic activity, though industrial production is in line with Europe due to export dependence. Our economists project a less deep recession and lower unemployment than in Europe at large. On page four, COVID-19 related credit requests: large corporate exposure increased 8% in Q2 to 996 billion, but the pipeline of likely new exposure has fallen from 83 billion to 63 billion, and on-balance sheet exposure is down, indicating clients are using facilities as insurance rather than drawing. Overall corporate credit exposure growth accelerated, and Swedish mortgages grew 7%. On page six, the Swedish FSA has given SEB a remark and an administrative fine of 1 billion SEK for governance and control of its Baltic subsidiary. SEB questions some of the FSA's conclusions and the proportionality of the fine, but has decided not to appeal in order to focus on developing the bank and fighting financial crime. The fine fits within our previously communicated cost target. On page seven, we are enhancing our financial crime prevention program and have launched Samlit, a cooperation between banks and the financial police. On page eight, notable business developments include advising Daimler on electrification, issuing a health bond, being awarded most ESG responsible banking group in the Nordics, and launching a corona solvency company with 5 billion SEK for medium-sized companies. We also launched digital trading and secure messaging. On page nine, financial highlights: encouraging client activity, resilient business, return on equity of 8.7% including the fine, strong capital and liquidity. For the first half, income and costs are similar to last year, but ECLs increased by 4.2 billion and the fine is 1 billion. Net ECL is 35 basis points, cost income 0.47, CET1 17.8%, and return on equity 7.4% (8.6% adjusted for the fine). I will now hand over to Mårten who will go through the second quarter financials in more detail.

I can start with capital returns. The last thing we formally heard was the EBA/ECB guideline to be cautious in repatriating capital, with a date of October 1st. Visibility on loan demand and future credit losses is needed before any firm decisions. Most banks have followed this recommendation and canceled or postponed dividends for 2019. There is debate on whether to extend that beyond October. We have said until further notice we don't pay dividend for 2019, but we will assess when appropriate. Our capital structure targets are not changed; we want to run the bank with an adequate buffer and have a progressive dividend policy. On risk migration: we've taken 4.2 billion out of the expected 6 billion in provisions this year, partly due to migration to stage three. We don't expect as much migration to stage three in the remaining quarters, but we do expect negative risk migration within non-default risk classes, so the risk exposure amount should go up. On NII: we had 14% growth in Q1 and 6% in Q2 year-on-year. The average of 11% is more representative for the full year. Factors include a reduced resolution fund fee of 800 million, the repo rate hike, and balance sheet growth of around 5%. There could be margin pressure and other temporary effects, but a likely outcome is around 10% NII growth.

Thanks, Nick. Well, we start with the — there is to our knowledge no ongoing investigation with a sanctions case against SAP in the U.S. That is not to be mixed up; you never know the relationship with the U.S. regulators is slightly different because they might do something on a desktop, but there is nothing there for us. And also, as we pointed out in Q1, we have now taken away from our risk factors in the Q1 report the things that we think of a significant nature. We did add the FSA holding the Swedish investigation as a factor one should consider, and we've not added any in this quarter. I hope that serves as a good comment trying to weigh the different interests in commenting on this. When it comes to the action plan, as we have concluded, most of the areas that we need to improve have been already identified by the bank and is part of the plan. We also need to remind ourselves that this ended in Q1 2019, so it's a year and a quarter ago these findings were fine. And a lot of things have happened in the last year, including resources, investments in the current business plan, etc. So there will be investments going forward; we will continue to accelerate. You might remember we had pointed to a 200 over and beyond investment last year in the future results, but we are convinced that we can fit that in the current cost framework, reallocating and prioritizing in the best way we can.

Yes, he mentioned — you can see in our disclosure in our fact book that travel expenses this quarter were 20 million compared to 120 million the same quarter last year. So obviously we do see that positive effect. And whether that's temporary or permanent, I think we'll have to wait and see. I'm pretty sure that it's going to come back to closer to a historical level, but a permanent reduction of some sort is likely. And so we see that happening. And on the digital part, yes, I think that will lead to higher cost. But what the net effect of these two will be, I think it's a bit too early to say. So we basically have the same comment that we had in Q1: that there are both positives and negatives, and it's difficult to conclude whether the net effect is going to be positive or negative in the long term.

Okay. I mean, I should be careful here, so let's state what is formally decided right now. This is a board decision: it is not to pay out anything for 2019. Just so everyone, the board can anytime they want ask us in management to do a different proposal, but there is no such ask right now. So if there needs to be some type of trigger changing the environment for this to change. And when it comes to the buffer, you adjust it for the dividend that we still continue to reserve, just so you don't overestimate the capital strength of the bank. We have no difference in our long, medium, and long-term target as we have today, which is a quality and around 150 basis point buffer to the minimum capital requirements. Of course, one needs to take into account in the medium term if the counter-cyclical buffer is likely to be reversed beyond 2021, 2022, so you don't sit in a tight position then. But for now, the board has not assigned any other than that type of management buffer at our disposal. So that is intact, and that's of course the question then: what time and how will one normalize? If this goes well, this needs to be normalization next year. If it doesn't, we probably need to spend this for the benefit of our clients, and that's a good thing, but we also need to have a significant buffer for potential losses if there is a second wave, which we are not assuming right now. Okay, fine.

Yes, Andreas. I'll try to do that. If you look at Q1, I think there are a lot of things that happened just in a couple of weeks when we closed the books in Q1, and I think the oil price reduction actually came in the first few weeks of April. So I think it was very difficult at that point to conclude exactly what would happen. What we've done in Q2 is to go through many of the large corporates that we have on individual name bases and see whether we think that some of them might have problems in the future. So I think now we've had more time to do that exercise, and therefore we've been able to more closely identify the companies we believe that we need to do reserves for. So I think that's the explanation for why the reserves on individual names on the oil sector come more in Q2 rather than Q1. It is correct that the underlying quality is very good. It is very difficult to see corporates with problems outside of oil. And within oil, it's also not just one picture; it's mainly offshore driven, I would say. And within leveraged finance, it could be different sectors, but it's mainly related to retail, for example, healthcare, but it's not very widespread. So it is correct that we don't see large effects in the book as a whole. And as you can see in our disclosure, the average risk weight is coming down and we haven't seen that much risk migration across the book. It's mainly related to oil and gas, and especially offshore. So for now, it's difficult to say whether these are temporary effects and how much government support is helping us here, but I think we're seeing basically what everyone else is seeing, that so far there's no real broad-based deterioration of asset quality.

I can't give you the number, but as you can see, 50% of the model overlay, so 800 million, is related to oil. So we do expect that the future provisions on individual names will also be related to oil to a large degree. It's difficult to say exactly what will happen in Q3 and Q4. And we have a model overlay that's based on an assumption of future problems we could have. If problems arise outside of those assumptions, then it could be different. If it's within those assumptions, then we can use some of the reserves we've done on a portfolio level. So it's too early to say, but we feel fairly confident that given the outlook we have in Nordic Outlook, around the six billion number will be the actual outcome for this year.

Thank you for that question. I think one needs to recognize that there have been highly volatile markets in Q1 and Q2, and so we only had 950 or so of underlyings — we were, you know, somewhere between two and four hundred million short, that's more or less compensated for Q2. So I personally just urge anyone to be a little bit more kind to the analysis than intra-quarter, because this is moving around a lot. We have not seen any reason to change our guidance that over time on average we expect the underlying to be 1.2 to 1.4. And then on top of that, of course, we've always talked about plus minus 2-300. That has of course changed as we saw how much impact we had on following the corona. That just points to that this is a volatile line, but it tends to be reversing — I wouldn't say recovering all the time, but it will be reversing over time, subject to market prices. So same guidance.

Yes, I think that's the conclusion you have to draw. We basically base what we say on the forecast we have in our Nordic Outlook, and in that forecast they don't expect that there will be a second wave of lockdown. So we are on a path of normalization, and the question is how fast that will happen. Based on that path, we believe that by year end at some point this will be back on a normalized level, or at least the level it was a year before that. You should also obviously remember that we've had a structural growth when it comes to our payments over time for many, many years. So even if we're back to last year's level by year end this year, it still means that we're missing a few percent that we normally have in terms of growth. Card issuance goes up all the time, and transactions go up all the time. So you should have that in mind as well. But it is based on the view that at some point in time travel will go back to closer to a normal level.

Yes, there's no target for 2020, but for 2021 it's 23 billion. With the current FX, it will be 23.2 billion.

Okay. First question was around cost efficiency. I think we don't have any further comments on that at this point in time. We are doing extensive work internally on lessons learned from what's happened in the last few months, and when we update our business plan by year end, we will have concluded on how we see the future given what's happened, in terms of both the outlook on revenues and costs, but also what we think will permanently change when it comes to our customers' behavior. So I think we haven't finalized that work yet, and when we have, we're going to disclose that to you and the market as a whole. On slide 4, on the credit facilities and the difference to '08, I think a big part of the difference is the actions by central banks. They've been much more forceful this time around, much faster, much quicker, much more in terms of support to the financial markets, and this support has had very significant effects. So what we saw in March was that the very professional investment grade large corporates were really quick on setting up new facilities, but as the financial markets recovered very quickly because of this support, these facilities have so far not been needed. They are there as some kind of insurance. But I think it's very much driven by that financial markets have recovered much quicker this time around than they did in '08. So I think that explains the difference between what we're seeing so far. We haven't concluded this yet; we'll have to see what happens. But so far, compared to '08, on SME lending, we can't disclose the nominal; it's not a massive impact. I'm just referring to the fact that we have had margin pressure here, and then we've revised prices in June. And obviously this runs through the books quite quickly, so to the extent that we had margin pressure in Q2 here, it should reverse in Q3.

Thank you. There were two main areas in Estonia, and both of those areas of improvements were identified prior to the results being published. However, our plan is longer term, so we have decided within the plan to accelerate those areas that are mentioned by the Estonian FSA, and our aim is to conclude them in time. And it will not change the cost target; we can do it within it.

No, we need to do some work. We have a deadline about a year away or so, and then we need to just do that work, and our ambition is to comply with that in time.

Yep.

All right, Sophie. I didn't say we take advantage of it. I think the rate cuts in the U.S. have had a negative effect on the Q2 net interest income, as we place some money in the Fed as a liquidity reserve. And this is very much driven by the fact that we need to hold dollars as reserves because of the LCR requirements in dollars, which now has been abolished. But by year end when we did this funding, we needed to hold very large liquid facilities at the Fed. If I look at the impact here, I think the negative impact in Q2 from the rate cuts is around 100 to 150 million on NII. And that 100 to 150 million will fully reverse going forward.

Yes, it's 1.6 billion. They are mainly within offshore, and some of them are within leveraged finance. I would say that about 10 companies sum up to this amount. And those two sectors or business lines for us, the largest 10 make up a very large portion of the 1.6 billion.

I'm not sorry if I caught that. So we have today allowances of 10.4 billion. If the negative scenario would materialize, or the probability would increase to 100%, we have to increase that 10.4 billion by 1.25 billion. So everything else equal, yes, provisioning would go up by 1.25 billion in that kind of a scenario. This is very model driven; it has not that much to do with reality and actually what happens in terms of corporates going into default, but I would guess there is some correlation. But the 1.2 billion is not in relation to your guidance on the expected losses for 2020. So if the negative scenario realizes, it doesn't mean that the total would be six billion plus 1.2 billion? Yes, if you allow me to answer: the guidance is in line with what we have in Nordic Outlook, which is the base case scenario here. So obviously if our economies change their view to a more negative macro outlook, then our guidance will not hold because it is contingent on their current outlook.

We haven't given the guidance on that, so I don't know. Then we would have to do work based on that scenario. I mean, we have done that internally; we looked at more severe scenarios than our base case scenario, but for now we only guide on this base case scenario. But as I said before, if it's tilting in any direction right now, it's tilting slightly to a more positive scenario than the base case scenario.

I mean, generally, the underlying is — when we look through the book, we start with looking at the exposure that we feel are larger and potentially more risky or could have a bigger nominal effect on the bank. So by definition, when you go through that, go through the big exposures, and when we look at the riskier ones, those are more related to the oil and offshore sectors. So I would say that when it comes to the underlying level and if you compare that to the model overlays, it's even more tilted towards oil and gas. And so the model overlays are more broad-based than the underlying level. And this is by definition because when you look at smaller companies, it's very difficult to go through all the several hundred thousand smaller companies we have as customers. So therefore, in a very early stage of a recession or negative scenario, you make an assessment on portfolio levels for these companies because it's too cumbersome to go through each and every one of them. So I think by default, in the early part of this kind of scenario, you have this tilt where you can identify the larger corporates on individual name bases, but you do model overlays for smaller companies, and that's where you see a bigger tilt when it comes to model overlay for both CNPC as well as the Baltics, as the corporates there are smaller. On the Baltic, I mean, I think we do see a more negative outlook on macros so far in the Baltics than we see in Sweden, and we have smaller corporates there in general, and therefore we have some model overlay there this quarter. It is too early to say; it's difficult to say what's going to happen. I think what the Baltics went through 10 years ago will be very supportive for both those economies but also people living in those countries because they learned a lot 10 years ago. I don't think they have over-leverage in the last 10 years, so I think you're going to have a generation here now that are very cautious in terms of taking on risk, and I think that's going to benefit us through this kind of a downturn scenario. And then you had a question on CRE, and I think our view on CRE is very much in line with what you've seen so far. When the biggest Swedish companies have reported their Q2, we don't see much yet. We don't see any real effect to be honest. I mean, they've given some leeway in terms of rents, but it's a very small proportion of their income. So so far, basically no effect. But again, we don't know what's going to happen in the next few quarters, but so far so good.

Thank you, Ricardo. On your first question, you're absolutely right in the sense that we've done our model overlay. If in the coming quarters we do not identify individual households or companies where we need this 1.6 billion for, then we don't need it anymore; it will be reversed at some point. We wouldn't have it as a reserve on the balance sheet forever. So we've made an assumption that we will identify companies and households in the future that we haven't yet, and therefore we need these reserves. But obviously we cannot guarantee if this will be the case, and if that's not the case then it will go back. Another way of answering that: if you look at our disclosure, you can see that the provisions for stage one and stage two loans have gone up; the coverage ratio for those kind of loans have gone up, whereas the stage three loans which are the ones we individually identify is pretty much flat. So if there's no migration from stage one and stage two to stage three, then the provision rate for stage one and stage two is higher now than is normally the case. And there's no structural reason to have a higher coverage ratio for those, so we have an underlying assumption that there will be more migration to stage three when we do these model overlays. On your second question, I'm not sure if I fully caught that, but you asked about whether there's been any tactical view on the fact that the NFI is strong and we've taken more provisions now, and I don't really have a comment on that. We've done the provisions we think are necessary given the outlook we had, and we acknowledge that we've front-loaded it since we've taken 70% of the expected level this year in the first two quarters. On the third one, on the one billion versus the previous 500 or 100, I recorded it. I can elaborate on this. This is difficult for anyone who tries to assess. Now, we do not have failures to pay or real bankruptcies in the first six months of this year. So when you don't have that but you're still asked to put aside a prudent reserve for the future, you need to make proxies; these are all statistical estimates. There are three ways of doing them: one is to look name by name and assess a probability, that's what we call the underlying. In no shape or form are we going to be 100% accurate; we will overestimate and underestimate, but we do our best. Then you have a macro correlation assessment, which is just saying if GDP goes down by X, house prices go down, credit should do something. You add that, and then you use more or less your experience and what you think is appropriate as an expert judgment on top — call it a model overlay. What really will happen is that the first real bankruptcies and failures to pay will come in 2021. If you look at any of the large corporates that really drive this, should we have a problem? Rule of thumb, it takes a year from the day you have a problem, because before you even know if you're going to be able to solve it or not, you reserve immediately. And in a year, maybe in the beginning mid-2021, you know if it's unsolvable or not. And most times, if you look back the last 20 years, we solve more than we initially think. We tend to be when we're pessimistic, we're over-pessimistic, and here we just try to be accurate. So what is happening right now is that we're front-loading the reserves in 2020; 70% of this year's current assessment is done. That means stabilization with what we know now; I cannot say if it's a billion or if it's less, but it's clear that 2022 we are of the opinion that when this thing normalizes, we should not be too far off where we previously were. But remember, we have had exceptionally low losses over time, so even when we had the six and the eight and the ten basis point cost of risk, we always indicated for the medium and long run you should have something higher; those are exceptionally low numbers. But that's all I can say right now. So we'll see what actually materializes in 2021, and that will dictate if the results we put on right now are sufficient or not. It's either going to be reversals or we're going to increase them.

Yeah, I mean, if let's assume we are very accurate and it happens like we think, the model overlays right now will be consumed by individual names as we see them. But right now you cannot foresee every single name that will come in mid-2021. So this is a judgment call and an expert judgment, and where credits in the bank of course is spending an enormous amount of quantitative resources to try to get accurate.

Thank you, Jacob. There's no debate about introducing or discussion around introducing negative rates as of now in the Baltics. On corona, there is a lively debate about the required square footage for commercial real estate in office space after corona. I don't have a view; the bank has no view. We are actually doing the work ourselves: what would happen to the required square footage? Should we allow large portions of the bank to work more remotely or have a more flexible definition of geographical space? And just reciting some of the larger real estate CEOs that I've met and heard about, there are many kind of gathering around the number that maybe 10% of office space will be freed up. But I have no clue if that's going to happen or not. And there is of course a tendency in that direction that we will work a little bit more from home, but on the other hand, there might be other things that consume this space. Thank you.

Thank you, Martin. On the competitive landscape, I think it's difficult to assess the future here. I think you're right that if you compared to the financial crisis, it's a different situation in the banking sector, at least up here, that most banks do have a lot of capital. So you shouldn't see the same kind of squeeze in terms of less competition. At the same time, if you look at Q2, we've seen somewhat less competition on mortgages; at least we've seen as an incumbent bank that we have less customers leaving us to smaller banks in this kind of environment. So in the short term, we've seen a bit less competition on mortgages. I think for us as a bank, we are seeing a different competitive landscape outside of the Nordics. We can see that many banks are withdrawing from other parts of the world where Nordic banks have been operational historically, thinking about Asia for example and other parts of Europe where other banks have been more operational and then they're withdrawing. So from that angle, when it comes to our wholesale business, I think the competitive landscape has turned a bit to our advantage. On capital returns, I think for now you should expect that at some point in time we're just going to go back to what we're used to: we pay dividends. And if there's any change to that, like we're going to do that on court level or do buybacks, then we'll disclose that at that point in time. But for now, I think just back to dividend would be a good help.

Then I'd like to thank everyone for participating in this one hour and 23 minute call, and just wish everyone a very good summer. Some of you we will see after the summer, and I hope we can have these in physical form soon. Thank you.