Swedbank AB (publ) (SWEDA) Earnings Call Transcript & Summary

Hello, and welcome to the Swedbank Third Quarter Report for 2020. [Operator Instructions]. Today, I'm pleased to present Annie Ho, Head of Investor Relations. Please go ahead with your meeting.

Thank you, Keith, and good morning, everybody, and welcome to the presentation of Swedbank's third quarter 2020 results. With me in the room today is our CEO, Jens Henriksson; our CFO, Anders Karlsson; and our CRO, Rolf Marquardt. The format of the call will, as per usual, begin with our opening comments, followed by Q&A. So without further ado, I'll hand over to Jens.

Thank you, Annie. Good morning, everyone, and welcome to the call. I am very proud today to present another strong quarter for Swedbank in uncertain times. Now for the last past 20 years, I've almost every year attended the IMF's Annual Meeting in Washington, and this year, it was a bit different because it was digitally. And the picture that emerged was very clear. It was all about COVID, COVID and Trump. And sometimes, I heard somebody mention Brexit. I have said that in this kind of crisis, the economy goes through 3 phases: mitigation, restart and then budget consolidation following the crisis. But it doesn't mean we will automatically go from Phase 1 to Phase 2 to Phase 3. Sometimes you take a step backwards going back to Phase I, and then you walk up the ladder. This quarter, we've seen signed recovery. But after a calm phase in the pandemic during the summer, cases of COVID seems to be increasing again. Now the IMF also see signs of recovery in its latest forecast, but the IMF still forecast a drop in the global economy by 4% -- 4.4% this year. The crisis is deep. So far, governments and central banks have met it well, but it's important to maintain this and continue to support the economy. The recovery will take time, and it will be difficult. And the effects of the COVID-19 crisis will linger for a long time. Now the economic crisis is created both by lockdowns and the effects of voluntary social distancing. Therefore, the economic risks will remain as long as the health risks are still there. Next year, IMF expects global growth, but again, sees risk of setbacks due to new outbreaks. In Sweden, we see a recovery, but it will take time before the economy is back at the level before the crisis. This year, GDP is expected to decline by 5%, and in 2021 to grow by 3%. And there is still a great uncertainty surrounding the spread of the virus and the economic development. We follow the developments closely and are prepared in case the number of COVID-19 cases rises and it become more volatile again. Now for the bank's customers, the situation began to stabilize in the quarter. We see in our card transaction data that consumption has climbed to level just below last year's. And we continue to be there to support our customers if needed. And we have a close dialogue with our customers in the most effective sectors. A strong trend during this quarter was that savings increased significantly. Deposit flows continues to be strong, and we have approved amortization exemptions for over 70,000 customers. And on the bank's private customers, we saw higher activity in the housing market while companies held back and borrowed less from the bank. Now is what you're looking for. And now let's get to the result. It's another strong quarter in uncertain times. Total income was SEK 11.6 billion during the quarter. Net interest income was stable, even though it was somewhat lower than in the second quarter. But it's the second best in at least the last 5 years if you compare on net interest income. Commission income rose, and this was due to rising activity from consumer spending and rising equity markets. Net gains and losses on financial items stabilized after a volatile first half of the year. Other income improved because of higher income from our insurance business and the card and our saving bank's holdings. The underlying costs are developing according to plan. But the costs connected to the investigations have been lower than expected this year. On the other hand, we increased the number of staff. And the lion's share of the new hires this quarter have been recruited to work on preventing suspected money laundering. During the third quarter, we made further credit impairment provisions of SEK 425 million. The economic development is better but we do not take this into account due to the continued significant uncertainty. Our return on equity during the quarter is 14.3%. This is below our target of 15%, but it's a strong result in uncertain times. And costs as the share of income are quite stable at 0.41. And we have a robust margin of 380 basis points to the Swedish Financial Supervisory authorities minimum common equity Tier 1 capital ratio requirement. Now the Board of Directors is still considering the issue of a dividend for 2019. We want to pay a dividend, but we follow the development and take regulatory authorities assessment into account and pay attention to how the pandemic develops. Swedbank continues to report strong earning and a stable capital and liquidity buffers. And we stand ready to support our customers in these uncertain situation. Now I cannot take -- I must take the opportunity also to say a few words about our work to prevent money laundering. We continue to work very hard to remedy the shortcomings that the authorities and our own investigation have pointed out. Carrying out our AML action plan is a big task. And during the quarter, 28 points were carried out, and in the end of September, 99 points of the 244 remained. And here, we can see the progress that's been made in the AML action plan. Annie, can -- of the 244 points in the action plan, 145 are completed and 19 remain to be started. And according to plan, 63 points will be carried out during 2020, and we will continue to work with 36 points during the coming years. And I'm sorry to disappoint you, but this is the last time I will give an account of our program externally. During the fourth quarter, we will deliver on the Swedish and Estonian's FSA's requirement, and then what remains to be done in the 244-point program will become the part of our regular work. This doesn't mean that our work against money laundering ends, quite the opposite, and we have very high ambitions when it comes to preventing it. I have said that we are at the beginning of the end when it comes to our historical shortcomings. And we've gone through a tough period with investigation. And we're working hard with the action plan, compliance and governance of the company. And we've also paid our fine to the Swedish FSA. But at the same time, we know there will be more work to be done when it comes to past events. In September, we received a notification from the Swedish FSA about an investigation into the events from the time when the issue of money laundering surfaced in media reports and our work with the U.S. before this continues. And I don't have any estimate today of how long the investigation may take. Now we want to be there for our customers. And during the quarter, we continue to make life easier for them with digital services that meet their everyday needs. Just 4 examples: We've launched the new youth app in Sweden, which has become very popular. It focuses on the demand among young people for a simple way to monitor their expenses and account balances. We've listened to what our customers want to do digitally and launch new services, for example, chat and automated Q&A on the website and more digital customer meeting. And our customers, they want simple and secure services. And therefore, we have launched a simplified log in to mobile bank ID with a QR code. In these COVID-19 times, many customers express a need to keep track of their spending, and we've launched a tool that calculates and helps the customers with their everyday expenses. Now this autumn, it feels even more exciting than usual to be the CEO of Swedbank. We have roots that stretch back to the 19th century, and this year, we celebrate 200 years. Now the first savings banks in Sweden opened on the 28th of October in Gothenburg in 1820. Since then, we have always been on the customer side as Sweden has gone from being a poor farming society to one of the most modern countries in the world. One thing I brought with me from the IMF meetings was how much focus there was on climate change, and the IMF proposes large investment initiative as support of the buildup cases. Now Swedbank is already in tune with this. The long term sustainable is the long term profitable and the corona crisis has shown us how true that is. And our tradition to give customers advice on how they can save to build a stronger economy, generation after generation, starts with the young. And that, I will hand over to the young, Mr. Karlsson, who will now dig deeper into the numbers.

Thank you, Jens. Jens is actually 1 year younger than I am. As already mentioned, we are presenting another strong quarter. I will first walk you through the P&L in more detail before asking our CRO, Rolf Marquardt, to speak about asset quality and credit provisioning. I will then sum up with a few remarks on capital and some forward-looking comments. Now let us look at the result in more detail, starting off with net interest income, which continues to be stable, albeit lower quarter-over-quarter. Let me take a step back before I explain the underlying quarterly changes. The 3 quarters of 2020 are the strongest NII quarters since the beginning of 2015. The development quarter-over-quarter can be explained mainly by 2 things: the first one is a couple of one-offs in the previous quarter within group treasury, amounting to a little bit more than SEK 100 million. The second one is market rates and volume development in lending and deposits. As market rates fell during Q3, lending margins were impacted positively and deposit margins negatively. To remind you, we raised our mortgage lift prices and administratively set lending prices somewhat in June that are gradually phased-in since then. As Jens alluded to, deposit volumes continued to increase in the quarter whilst total lending volumes decreased. The average corporate lending volumes decreased, while mortgages increased. These volume trends lead to a structural shift in our funding composition, where wholesale funding is gradually replaced by deposits. As you know, the P&L effect from deposits is immediate, while wholesale funding matures gradually and assets are repriced over time. As the deposit inflow exceeds the lending development significantly, we experienced a small headwind in our NII due to excess liquidity being placed at central banks at the cost. Over to net commission income, which was stronger, card commissions improved due to higher transaction numbers since the initial onset of the pandemic, coupled with the usual higher seasonal effects. FX transaction from foreign travel were still lacking. The asset management business was positively impacted by the rebound in the equity markets. We also saw good institutional and private inflows in our mutual funds, mostly into active equity funds. Turning to net gains and losses, it stabilized this quarter after a volatile first and second quarter. The entire 7.4% shareholding of Asiakastieto, now called Enento, was sold during the quarter with a negative quarterly effect of SEK 110 million. In September, half of the holdings of C shares were converted into more liquid A shares which resulted in a positive revaluation effect of SEK 91 million. And customer activity in was good despite the typical lower seasonality over the summer. Trading and fixed income and FX performed well. Other income was higher. Insurance increased due to higher premium income and lower claims in Sweden, and Entercard had a better quarter as cost of funds and credit provisions decreased. The income from partly owned savings banks was stable in the quarter. Let's look at expenses before I hand over to Rolf. Underlying expenses as well as expenses related to AML investments and COVID-19 are developing according to plan. Our interactions with the U.S. authorities are still at a low level. It remains very difficult to forecast the expenses related to investigations. Although it is unlikely that it will increase to the level that we envisaged when we updated our cost guidance earlier this year. We therefore expect that full year expenses will be slightly below SEK 21 billion. As previously stated, we will give more color on 2021 expenses in conjunction with the Q4 results. I will now hand over to Rolf to talk about asset quality and the credit provisions that were made in the quarter.

Thank you, Anders. We have now been through 7 months of COVID-19, and during the last quarter, we have seen some improvements and signs of increased economic activity. This has, of course, all been supported by continued fiscal, monetary and regulatory response. Today, it's also easier to assess how different businesses and societies have been affected. At the same time, uncertainty is still significant, at least related to the more recent COVID-19 developments. During the last 6 months, we have been very active in working with our customers to better understand how they have been impacted by COVID-19 and how they mitigate the effects. We have made a thorough individual assessment through customer meetings and the subsequent credit risk assessment made by our customer and credit teams. This has been a major exercise covering all customers in affected sectors and a large part of all our corporate customers. The outcome has been more pronounced rating migrations in the second quarter. In the third quarter, rating migrations were limited. The overall level of overdue loans continued to be stable. The conclusion from this review is that overall credit quality remains strong and stable, the oil sector being the exception. This is also the explanation behind the credit impairment ratio of 10 basis points in the quarter and 30 basis points so far this year. The total provisions for the third quarter was SEK 425 million, which is explained by 2 major factors. During the quarter, the macro conditions and the forecast for 2020 and 2021 for our home markets improved more than originally anticipated in Q2. The technical impact of the improved macro scenarios suggested that about SEK 773 million of the reserve should be resolved. Although we see positive signs in our home markets, we believe it is far too early to release any macro buffers, especially with the latest signs of a second wave of lockdowns coming back in Europe and other markets. Therefore, we made the management overlay to neutralize this positive macro impact. The individually assessed Stage 3 provisions during the quarter amounted to 433 -- sorry, SEK 432 million. This is mainly explained by 1 individual oil-related provision in LC&I, which was partly offset by recoveries. Credit rating migrations increased provisions by SEK 114 million, and the sum of other impacts added SEK 234 million. In the past 2 quarters, we have disclosed our exposures in different sectors and our assessment of how these different sectors have been impacted by COVID-19. That is displayed in the left bar in this slide. The exposure amount to considerably impacted sectors defined in this way is SEK 109 billion. The easy one looking at the sector level is that some sectors that are assessed as considerably negatively impacted also contain some subsectors that are less or even positively impacted, for instance, food and electronics within the retail sector. To get a better reflection of the impact to the industries, we have looked deeper into the subsector level, which shows that the exposure to considerably impacted industries amounts to SEK 29 billion. The impact on our exposures in these industries in terms of overdue loans have so far been limited as at the level of Stage 3 provisions, with the exception of oil and offshore. Stage 2 loans have obviously increased due to credit migrations, which mainly occurred in the second quarter. Now, a few words about property management, which you will find in the yellow part of the bar. The core in our policy for property management business is to focus on long-term sustainable cash flows and debt service tolerance, taking into account potential significantly increased interest rates, operating costs, vacancy rates and to apply risk-based amortization requirements. We are also cautious with retail space and other parts of the sector where structural changes can be expected. The level of collateralization is, of course, key, but that is also a natural outcome of sticking to strict lending criteria across the cycle. Residential property accounts for a significant part of the portfolio, and this is a very stable business, supported by growing population, stable rents and low vacancies. The commercial real estate lending is mostly office properties located in the major cities and with low vacancy rates. The average LTV is 53%, and almost all of that consists of individual LTVs below 75%. The situation is similar for industrial and warehouse, which is supported by the ongoing structural change related to e-commerce. What should also be kept in mind is that a large part of these exposures is related to larger property management companies with diversified property portfolios. And then on to our oil-related exposure within shipping and offshore on slide -- sorry, the next slide. I'd like to remind you that our oil-related offshore exposures remain limited and are to a significant degree in runoff. The exposure was SEK 12 billion as per Q3. We have SEK 7 million in Stage 3, and 60% of that have been provisioned for. So with that, I hand over to Anders again.

Thank you, Rolf. Let me now turn to capital. The CET1 capital ratio increased to 16.8%, and the buffer to the minimum requirements stands at around 380 basis points. The profit in the quarter and the pension liability valuation impacted the CET1 capital base positively. The risk exposure amount was stable quarter-over-quarter. We updated our LGD models, which increased risk exposure amounts for credit risk by SEK 21 billion. Consequently, we removed the Article 3 add-on of SEK 16 billion. Furthermore, the Article 3 add-on on PD decreased by SEK 6 billion, mainly due to an increase in average PD during the quarter. As we look ahead, I want to reiterate that Swedbank's -- Swedbank continues to be in a strong financial position, especially from a capital funding and liquidity point of view, and our asset quality is resilient. Looking at NII, over Q4, we expect the development with continued elevated deposit balances gradually to replace maturing debt. On the lending side, we will most likely continue to see subdued corporate loan demand. We also see tightened corporate credit spreads and subsequently a continued preference for funding through the capital markets. We also see signs of competition heating up in the mortgage market. The latter point is also the reason as to why we took the decision yesterday to adjust our mortgage list prices to be more in line with our strategy. Society in Sweden continues to be largely open for business under the same restrictions that we introduced earlier in the year. However, the situation in Europe remains uncertain and new restrictions are being introduced. Market rates will, therefore, most likely continue to decrease in Q4. Looking ahead on NCI, we will benefit from the higher assets under management levels at the quarter end, assuming flat or positive market development. For card payments, so long as international travel does not pick up further, the lack of FX-related card transactions will continue to weigh on income. And last but not least, as we have stated before, we will guide you more on expenses in -- for 2021 in conjunction with the Q4 results. With that, I believe we are ready to take questions. Thank you, and operator, please.

[Operator Instructions] Our first question comes from the line of Magnus Andersson of ABG.

I would like to start off with a few questions on costs. First of all, when I look at your headcount development and compare with Q3 '17, which is the last comparable quarter actually when you included PayEx, the number of employees are up by around 1,500, which is a 10% increase. Also in the beginning of the year, or I think end of last year perhaps, you talked about potentially 300 to 400 employees up in 2020 or up by 800. At the same time, IT costs in the P&L are increasing your impairment of intangibles or amortizations are also increasing. So I was just thinking, I realized that consultancy costs could continue to normalize and come down you by SEK 200 million this quarter. And normally before this AML stuff, you were at below 100. But how should we think about this going forward-looking into 2000 and perhaps second half '21, '22 just in terms of the headcount trajectory? How much is due to AML? And what could you automize -- what reduction should we see? Because otherwise, you might run into a cost problem in 2022 or so.

Thank you, Magnus, it's a very relevant question. And I will not guide you in any precise way for 2021, but you are right. We have had the ambition to recruit new employees, and the reason behind that is, as we have said before, that we need to actually manage the AML situation in the short perspective by adding flesh and blood. It will definitely take time before we can automate things. So that effect will come further down the line. What we also have been using quite significantly during this year is that since recruitment takes time, we have been using external consultants, as you are rightly pointing out. Gradually, we would change that behavior, but we need to get people in. There is also a second element to it, Magnus, and that is the fact that the turnover of staff is usually on a 12-month rolling basis around 10% in the bank. Since the pandemic breakout, it's been dropping significantly. So the anticipated turnover is much lower than we thought in the beginning of the pandemic. So that is also adding to this. And as far as the impairments for investments activated in the balance sheet, they have increased over the years, and you are perfectly right that will increase the amortization year-by-year. So the intention is, at some point, to break this development. We are not in the position as we speak to do it. We need to fix stuff. We need to manage through the AML situation. And as I said a couple of times, we will come back with further guidance forward-looking in conjunction with Q4 results.

Okay. When I look then -- when I think about the drivers into mean more midterm here, 2021, second half, perhaps 2022, and I focus on headcount and IT -- total IT spending, which both have been increasing, what is it that is going to stabilize or even come down? Is it headcount? Or is it total IT spending of both 2022?

I think it is a very good question, and it's too early to answer your question, Magnus. We will come back to it.

Okay. And then secondly, just on asset quality. Unless you had made that management adjustment, you would have had net recoveries already now in Q3. And we see that the losses you have are primarily oil related. You had 10 basis points of loan losses in this quarter. Unless we get the worsening macroeconomic outlook or a severe second wave of COVID-19, can you say something how we should think about your loan loss provisions going into 2021 and '22?

Magnus, it's Rolf here. So I -- I mean, this is the best assessment for the time being, and we can't really guide you. But what is certainly the case that we will watch what is going to happen in the future, and what will be important then is to see the outcome of the development that we now see. What I'm referring to is, of course, COVID-19 situation and how that unfolds, how it impacts our customers. So it is the assessment we will make. And when we have more clarity in that, then we will, of course, take action. But we are not at that point for the time being.

And just on capital and your keeping the door open here for a dividend. What do you think will be the next? When will the -- when will we get the next data point? Do you think it will be in relation to the stability report from the FSA in late November? Is that what we're waiting for?

Well, we don't have a data point as such. It's very sort of straightforward that the Board of Directors, they are still considering the issue of a dividend for 2019, and they are very firm that they want to pay a dividend. But we follow the development and take regulatory forties assessment into account and we pay attention to the -- how the pandemic evolves as said.

I think it's worth adding that I think that everyone awaits ECB's communication around this specific issue, which is supposed to be in December, but it's unsure whether that will happen or not. So I think ECB is another factor to have in the back of your head.

And then just finally, a more detailed one, just on the Swedish Bank tax proposal. If you have estimated any potential impact on you guys and the full impact then from the 7 basis points?

Yes, we think it's -- with the current balance sheet and liability side, it will be SEK 1 billion in the first year, and then I think it will increase to roughly SEK 1.2 billion. But it's tax deductible, Magnus, so you should actually multiply...

It's a gross numbers.

Yes, it's the gross numbers. So that's the best estimate as we speak.

Our next question comes from the line of Johan Ekblom of UBS.

Just a couple of quick ones. It sounds like you don't want to push too much on cost guidance for next year. But just on the lower AML investigation spend this year, do I understand you correct that you expect it to remain lower and that it's not just a shift of those expenses from this year to next?

Well, it's Jens here. It's -- to be honest, it's very hard to say. So we saw quite a large payout in the first quarter due to the delivery of the Clifford Chance report. Then it went down in Q2, and we've seen a sort of on the low level also as Q3. We think that could be a combination of sort of COVID effects but it could also be due to the fact that the Clifford Chance report was so good, and we had all the data available. So to be honest, we don't know. Some things may be pushed to next year, but we answer the questions they ask.

And then secondly, just to come back to your comments around net interest income and the impact of the strong deposit growth and falling market rates, it sounds like what you're saying is that at some point that headwind will turn into a tailwind or at least be neutral. How should we think about that in terms of timing or drivers? What do we need to see to kind of see a stabilization between the funding needs and kind of uses of funds?

It's a good question. There are a couple of moving parts there. But I mean, the first one is market rates, obviously, and there's usually, you will expect market rates to come down further in Q4, both on the back of how the central banks are acting, but also seasonal patterns. The second piece of the puzzle is the subdued or not subdued growth in corporate lending. And the third element is that, as you know, we are primarily funding ourselves through deposits or covered bonds. And when covered bonds are maturing down the line, we will replace them with deposits, which is actually a benefit when it comes to old maturities. Issuance today is basically similar. So the price of it -- we are paying for issuing a covered bond and the price we pay for deposits are the same. And as you also are aware of, when you go further up in the stack, we run a limited portfolio of senior unsecured, but that needs to be carried on due to other regulations. So it's a switch between our mortgage funding and deposits over time, which is positive. But it's a mismatch. Deposits is flowing in faster than we can actually replace outstanding wholesale funding as we speak. And if market rates are moving up, we will, as you know, benefit from that.

And then just finally, is there any news you can share on the payments review you talked about in Q2?

No, we have nothing new to share there. We said in the Q2 report that we'll do a strategic revision of our payments business, and that's ongoing as we talk.

Our next question comes from the line of Nick Davey of Exane BNP Paribas.

Just a couple of questions, please. So firstly, on the cost guidance for this year, it seems to imply a very high cost fee in Q4, especially, I think you mentioned in the comment slightly below '21. So is there anything specific we should be prepared for in Q4 costs? The second one on NII, I appreciate all the comments that have just been made. I think that clarifies it. But just to follow-up on a question from a quarter ago, I just want to try and get again confirm. Do you now think this Q3 NII base is sustainable, particularly for treasury? I know there have been some moving parts on the internal fund transfer pricing model. I just want to understand in the round for Q3 NII, do you think all else equal, this is sustainable? And the third question, I understand on provisions, so holding ourself back from too much guidance into the future. But could I just invite a few more comments on the level of reserving you've now reached in the oil and offshore book and whether you feel comfort that you're now fully provisioned?

Thank you, Nick. On your first question, the reason for guiding slightly below '21 is purely the fact that we have been trying to forecast investigation costs, and it is extremely difficult. So -- and that is why we are a bit soft on that language. I do not foresee any other nasty surprise coming our way. When it comes to your NII question, I think that what I tried to convey in my forward-looking statement is that as long as deposits are still continuing to flow in, and as long as corporate loans demand are subdued, we will have a liquidity surplus that we need to place with central banks at the cost. So it's not a drama in this, but we have a couple of headwinds that we will -- and we will gradually release some of that headwind when we are replacing our outstanding funding with deposits. But that is what I tried to convey in my forward-looking statement. And when it comes to provisions, I hand over to you, Rolf.

Thank you. Nick, so we -- the exposure we have -- the gross exposure we have to the oil and offshore industry is now SEK 12 billion, and SEK 7 billion of that is in Stage 3, meaning that a very large part of that has moved to Stage 3, as you know. And then we have provisioned 60% of that already. And that's the best assessment we have today. We made another provision this quarter. We also made recoveries. We are working actively with that portfolio to remediate the problems that we do have. As you know, the oil price level has been quite stable during the quarter. So it's not only related to that. It's also a question of time for this to unfold. But we have done what we think is necessary so this is the best assessment we have today. And that is, of course, difficult to forecast the future in this sector.

Just a quick follow-up on this question of the fast pace of deposit inflows. Do you see any segments in the large corporate business where you might be able to charge for deposits more than you do today, given that structural problem?

Relevant question. We are charging institutions and large corporates that have access to the capital markets as we speak. And that's the best information I can give you at this point, but it's a very relevant question.

Our next question comes from the line of Andreas Hakansson of Danske Bank.

First question is just more a follow-up, but on the NII, you talked about the subdued interest on corporate volumes. But could you tell us, you had, I think, it was a SEK 12 billion or so outflow of corporate loans in the quarter. Is that -- if I could call it more volatile corporate volumes that came on the balance sheet in the midst of the COVID crisis and that has now been flowing out, has that more volatile volumes are they now out, so we should have a more stable outlook from here, then it could be up and down depending on the market? But this -- the holders that hit your balance sheet so quickly in -- between Q1 and Q2, are those the volumes gone now? That's the first question.

Thank you, Andreas. I think there are 2 reasons for the decline in the quarter. One is that a couple of the liquidity facilities that we provided in Q1 were repaid. So in that sense, you could argue that part of the volatility coming from COVID-19 outbreak is disappearing gradually. The second one is related to the fact that fairly large companies in our portfolio have a rating, and they have access to the corporate bond market. That's extremely liquid with fairly compressed credit spreads. That I foresee to continue as long as we see central banks acting the way they are doing and that the capital markets are still open and liquid. So it's a combination of both, I would say, Andreas.

Is it possible to split it to know how much of the liquidity part or how big that part is?

It is, but I don't have it on the top of my head. I have to come back on that one.

That's fine. Then one more detailed question. Entercard did quite well in the quarter. Was that by any chance done due to reversals? Or was it just a good underlying performance?

It was primarily due to a decrease in funding costs and some reversals or deltas in the provision levels quarter-over-quarter.

But was the delta coming down over the actual reversals?

I don't have that on the top of my head, but I think it was...

Okay, it was probably a small number anyway, right?

Yes, but I think it was delta coming down. That's my best guess.

Then I just want to understand. It sounds like you're talking about the underlying costs now being SEK 4.8 billion a quarter so we're analyzing underlying around SEK 19 billion. Is that what we should be considering to be underlying costs, and then we all have to take a stab on how big we think AML costs are going to be?

Andreas, your guess is as good as mine. I will come back to you in conjunction with Q4.

Our next question comes from the line of Namita Samtani of Barclays.

Just on the operating expenses, are there parts of the cost base that you believe can be reduced permanently in light of the trends we're seeing related to coronavirus, in particular a less need for branches both in Sweden and the Baltics?

Well, first, I just want to say on branches. Let's say, in Sweden, we have 160 local branches, and that's actually a half -- more than half -- reduction of half the last 10 years. So we do not have further plans on closing down sort of local branches. But we, of course, follow the situation and follow what the customers do and we always take small decisions on this.

And to add to it, I mean, what we saw early on in the pandemic was actually an increased need for calling into our contact center. So during that period, we actually had a lot of extra hires into manage the flows. I think it is far too early to talk about any sort of cost reductions due to COVID-19. I think you need to put it in a more sort of -- in a larger perspective.

And then just on the Swedish mortgage market, you made a comment on of competition heating up. So could you just give us a bit more color on what you're seeing?

Absolutely. There are a couple of signs. I think -- but before I go into that, I think it is -- if you think about the situation in Sweden, you have a lot of banks who are well capitalized. We have all experienced large deposit inflows. I think that we all have experienced a subdued corporate loan demand. And the only sort of market that is actually growing in Sweden when it comes to loan demand is the mortgage market. So that is sort of the first ingredient. The second one is that during the pandemic outbreak, we have seen that the participants been sort of cautious. The campaigning were sort of disappearing from the market. The churn, as we talk about, which is back book theft, came down to very low levels. We see signs of churn coming back a bit. And then as you saw a couple of days ago, the second largest player in the market changed their list prices. And as I -- which I think is a game changer. We decided to do that as of today with 1 difference. We keep the 3 months rate unchanged. But the other parts along the term structure is moved more or less in tandem with the competition to put us in a situation where we usually want to be.

Our next question comes from the line of Adrian Cighi of Crédit Suisse.

This is Adrian Cighi from Crédit Suisse. Just 1 follow-up question on capital, please. You're now at 380 basis points capital buffer and above your 100 to 300 basis points even after retaining the full year '19 dividend reserves. We clearly live in very uncertain times, but in terms of potential distributions, what other avenue for excess capital distributions are you thinking about? Or do you see the potential change for that 50% payout? Or is that sort of a contingent on the resolution of the U.S. investigation?

I think you need to look at the 380 basis point in the light of 2 things, or 3 things, actually. The first one is the uncertainty that we have in the environment that we are operating in. The second one is the fact that you have heard the regulators being -- sending, at least up until now, clear signals on their view of dividends. And the third element is upcoming financial regulation, which is, first, it's the IRB overhaul which we deem to have a fairly significant impact on us and other Swedish banks when it comes to increased risk exposure amounts. So it is far too early to talk about excess capital.

Our next question comes from the line of Rickard Strand of Nordea.

Two questions from my side. The first one is on NII. You talked about excess liquidity placed at central banks. If you could give any indication what the impact Q-on-Q is there for Q3. And then the second one is on wholesale funding where you talk about maturing -- gradually maturing over time. If you could give any indication on what you -- sort of what the rollout might look there. And then lastly, a question on AML costs. You have previously hinted about some sticky AML cost of, I think it was SEK 1.1 billion. If you could give any update if that view has changed or not since you gave that indication in Q1, I think it was.

A lot of details there. I'm not sure I have them on the top of my head. I don't have the quarter-over-quarter effect on the liquidity placements with the central banks, but to give you some indication, when it comes to placements with the Central Bank of Sweden and Federal Reserve, there is a limited impact in terms on NII. The expensive one is with ECB. I can come back on some of the numbers. I don't have them on the top of my head. When it comes to the maturities of covered bonds, I think that the first thing you saw was clearly that we changed our funding plan dramatically during this year. I think next year, there is -- since it is benchmarks, there is a large maturity during spring and a couple of others, I don't have that on the top of my head. But it's on tap issuance in Sweden so it should be, I would say, around March to May. And then there will be a second one further down the road. But we can give you the details on that later on. As far as AML comes and the SEK 1.1 billion, SEK 1.2 billion, I see no reason to change that at this point. We have, as Jens told you, been hiring people, both within the business areas, but also within the compliance and the anti financial crime unit within the bank. But there is no news on that as we speak.

Our next question comes from the line of Martin Leitgeb of Goldman Sachs.

Just a couple of follow-up questions, really, and the first one on capital and capital requirements going forward. You mentioned the impact of model changes going forward. Could you give us a view on how much an impact that could be in terms of your capital requirement going forward? And related to that, looking at your comments on mortgage pricing and mortgage pricing coming down, shouldn't there be a pressure from higher capital requirement potentially to lead to an upward price on mortgage pricing as we progress? And the next question, just on NII, I'm just trying to understand the guidance here. It's expectation if deposits keeps staying at the current level that NII would edge low in the fourth quarter.

Okay, I think you have to repeat the second question, but let's take the first and the last one and see if I understood you correctly. I will not be able to give you any guidance on the effects of the IRB overhaul in any precise way. What we have been saying in the past is the fact that there is an ambition to level out the differences in risk weights between the different countries in Europe, which means that Sweden and Swedish banks have been running at a significantly lower average risk weight in the corporate portfolio than some of our European peers. For good reasons, I can argue, that will change. So we will -- it will be an increase in risk exposure amount. The exact number is too early to say because it is a function of us developing models and applying for FSA approvals. So you will see it gradually coming in during next year. On your NII question, please repeat that one as well because I forgot.

So NII was just -- obviously, you talked about the tailwind and the headwinds from -- to NII from today's perspective. Just obviously, given the comments that market rates are lower, all else equal, so it does not pick up in loan demand. And if there's no change in deposit flows, should we expect NII to trend lower from here or stabilize from here? I mean, I was just trying to square for the message here on NIIs. And on mortgage pricing, just to recap the question, if the assumption would be that the average risk weight on mortgages would be higher, potentially starting from 3Q '21 onwards, wouldn't the market rationale be to price mortgages higher to try to compensate for higher capital density of that product?

Okay, if I take that one first, I think if you look at the underlying risk weights, if we were to use internal models for mortgages, they are substantially lower than the risk weight floor. I do not see that there is any indication or reason to increase the risk weight for mortgages. So that is a hypothetical question. As far as NII came, what I said to you is that 2020 is by far the strongest NII year if you compare back to 2015. But in a climate where you see deposits coming into the system in a fairly rapid pace, and you do not see possibility to lend it out to corporates or private customers to the same extent, and you also have the fact that deposits are repricing immediately when market rates are moving since they are floored at 0, or at least the majority of it, and the assets are repricing gradually, and the fact that we will replace our maturing outstanding covered bonds gradually when they mature with deposits, which will be a benefit. What I'm saying is that we have a slight headwind going into Q4.

Our final question comes from the line of Riccardo Rovere of Mediobanca.

Just a quick follow-up on cost, getting back to once again to one of the questions from my previous colleagues, you're running -- if you take out the administrative fine, you're running in the core -- or in the 9 months at roughly SEK 15 billion cost. Now the guidance, the 2020 is set at 21 billion. Sorry, is that NOK? What should drive the sharp increase, given that I would imagine the level of activity, traveling and all these kind of costs, will probably not change much or won't change at all in the last part of the year. And the second question I ask still on cost, sorry to get back. It's not clear to me whether you think some of the, I don't know, AML or, let's say, investments demand supposed to happen in 2020 will only be moved in 2021. That is not 100% clear to me. Another question I wanted to ask you, I've seen reversals in Swedish banking. I was wondering what is driving that, given that the unemployment in Sweden is certainly not going down. It's just a matter of the real estate prices which continue to be extremely resilient. And the very final question I have is on the level of AT1s that you are currently having your balance sheet. If you consider that is okay? Or if you think you may top up something in the future?

Riccardo, as always, many and relevant questions. I will try to capture them. What I tried to convey is that underlying cost is developing according to plan. There are unknowns in terms of investigation costs. And I've been trying to guide on that a couple of times and been entirely wrong, being a bit cautious. That is why we are guiding slightly below '21 instead of the 21.5 that we communicated in March this year. On the AML investments that we are talking about, it's primarily consisting of 2 things. One is actually FTEs, employees coming into the bank working with it. They will continue to work with it in 2021. So that is a structural cost change that we've been talking about. The second part of the investments is very much into IT development. That you can see either through activations or taken upfront. So -- but the continuation of investing into IT systems related to the prevention of money laundering will continue in 2021. When it comes to reversals in Swedish banking, I hand over to you, Rolf.

Thank you, and hi, Riccardo. So that is a quite small reversal, but it stems from 3 different sources. So first of all, it is a small number of smaller customers where we have major recoveries. The second part being a change in models. We have made an annual update of some of the underlying our first-line models. Small impact, but still. And then thirdly, there has been a slight shift in mix related to declining -- partly declining corporate volumes and then replaced by more increasing mortgage volume. So the mix change has an impact on Stage 1 and 2 provisions. So that's the background.

And finally, on the AT1s, Riccardo, we will expect to issue some more AT1s in 2021 to close the current deficit.

Well, thank you, everyone, for good questions, as always. And see you next quarter. Take care. Bye.

Thank you. Goodbye.

This now concludes our conference call. Thank you all for attending. You may now disconnect your lines.