Danske Bank A/S (DANSKE) Earnings Call Transcript & Summary

Welcome to the conference call for Danske Bank's financial results for the first 9 months of 2020. Thank you all for taking the time to listen in on this call today. My name is Claus Ingar Jensen, and I am the Head of Danske Bank's Investor Relations. With me today, I have our CEO, Chris Vogelzang; and our CFO, Stephan Engels. Slide 1, please. In today's call, we will present Danske Bank's financial results for the first 9 months of 2020. We aim to keep this presentation to around 25 minutes. After the presentation, we will open up for the Q&A session as usual. And afterwards, feel free to contact the Investor Relations department if you have any more questions. I will now hand over to Chris.

Thanks, Claus, and good afternoon, everybody. Thanks for dialing in. The first 9 months of 2020 were a challenging period, not only for society but also for banks as the outbreak of the coronavirus pandemic had a huge impact on the global economy. Consequently, we saw a steep decline in GDP across almost all countries. However, as the lockdown measures have varied from country to country, the economic impact has also been different. In this context, the Nordic economies have held up relatively well. And despite overall lower economic activity, our underlying business remains strong, characterized by good customer activity and stable income driven by our Nordic markets and large corporate customers. However, as the corona crisis continues to evolve, we also continue to have clouded visibility regarding future economic developments. It's also been a busy period for us. Across our business units, several initiatives have been launched. At the same time, we've supported our customers in handling the effects of the corona crisis. At Banking Denmark, we have introduced FlexLife, a fixed-rate mortgage loan type and a new setup for servicing retail customers planned for launch in 2021. Banking Nordic was able to continue the onboarding of many new retail customers despite the corona turmoil. At Corporates & Institutions, they have been busy securing liquidity facilities for our large corporates. While at the same time, providing advisory services and access to primary debt and equity markets. There was thus an increasing number of transactions with a strong sustainability focus and finance by green loans. In Wealth Management, we have now completed the remediation of the Flexinvest Fri case and solved the issue with the authorities, while at the same time, improving net sales in the institutional segment. Both Danica Pension and asset management have been busy launching new investment solutions for our customers with a strong focus on sustainability. Financially, we're today presenting a result for the first 9 months with good evidence of steady progress in most of our income lines, lower costs and with a net profit for the first 9 months of DKK 3.1 billion. On that basis, on October 27, we announced a revised net profit outlook for the full year. We now expect net profit for 2020 to be in the range from DKK 4 billion to DKK 4.5 billion, based on improved developments in the financial markets as well as lower costs and continually good progress in our underlying businesses. We saw slightly higher net interest income, and both fee and trading income maintained a good momentum despite higher overall activity last year. In order to align our pricing more with the market situation and following pricing initiatives for our corporate deposits, we've today announced lower thresholds for our retail customers in Banking DK. The initiatives we have announced today are specifically designed to ensure that we continue to have a relative attractive offer for our customers. Our expenses came in higher than last year, due primarily to costs related to the Better Bank transformation, continued compliance remediation and of course, the Estonia case. In the third quarter, we saw that our cost management initiatives are starting to yield results, and we also continued to launch a range of further initiatives across the organization in order to ensure a good trajectory towards our 2023 ambitions. To further accelerate the process, we have recently announced the discontinuation of up to 1,600 positions across the group. As a first step, who have initiated voluntary redundancy agreements for staff in our operations in Denmark. Furthermore, we continue to make progress on product simplification, not only as part of our cost program but also as part of our overall simplification aspiration, and we are well underway to meet or exceed our targets for 2020 of reducing products by 30%. Adjusted for the one-off income item in 2019, profit before loan impairment charges came in 11% below the level in the first 9 months of the last year. Credit quality remained strong despite low impairment charges being the most significant reason why the results for the first 9 months of 2020 is lower than the results for the same period the year before. The relatively good performance of the Nordic economies I mentioned earlier led to very limited credit quality deterioration among our customers. Most of the model-driven impairment charges recognized in Q1 due to coronavirus have now been reversed. However, they have been replaced by increased post model adjustments as visibility regarding future economic developments remain clouded. Our oil-related credit exposure suffered from difficult market conditions throughout most of the period. Stephan will comment on this in more detail later in this call. Let me now update you on the redesign of our organization we have announced as part of our continuous execution of and progress on the Better Bank agenda. Towards the end of August, we announced the next big step in our efforts to decrease complexity and simplify the organization. The new design will ensure a better quality of processes and make us a more competitive bank. The announced changes include the merger of Wealth Management and C&I into a new large Corporates & Institution unit. Furthermore, we are consolidating Banking DK and Banking Nordic into the new Personal & Business Customer units. We just -- we thus bring the number of business units down from 4 to 2 with the aim of breaking down internal silos and utilizing knowledge and expertise across the organization. Furthermore, the introduction of the new commercial leadership team was announced to build a talent pipeline and to improve the knowledge exchange across the organization and of course, very much in order to drive revenues in the commercial agenda. We also continue to make tangible progress in terms of our anti-money laundering efforts. Currently more than 2 million ODD reviews have been conducted year-to-date. About 75% of the reviews for customers categorized as low-risk have been processed automatically. Our internal investigation into the nonresident portfolio at now closed Estonia branch is progressing with some of the tracks completed in Q3 as planned. We expect to finish our -- all our investigation in the fourth quarter of 2020, as previously announced. And then I'd like to hand over to Stephan for the financial results in more detail.

Yes. Thank you, Chris, and over to Slide 3, please. Let us look a bit more closely at the financial results for the period, which overall reflects steady progress. Across our markets, we have seen generally lower economic activity. However, key economic indicators such as housing market activity and consumer spending hold up well, supported by low unemployment and few bankruptcies. The better-than-expected macroeconomic development is evidenced by our Nordic lending activities, increasing 4% in local currency terms from the same period last year. The major government support packages that were not fully utilized were phased out at the end of the summer and replaced by more bespoke facilities aimed at specific business sectors. The very recent increase in contagion has led to new support initiatives from the Danish government, however, not of the same magnitude as we saw in March. The new restrictions that came into effect last week are not causing the same degree of lockdown that we saw also in March. Customer activity, including housing market activity is holding up well despite clouded visibility regarding future economic developments. Excluding one-off effects, total income was in line with the level in the same period last year. Overall, total income came in at DKK 31.4 billion, or 4% lower due to a one-off gain from the sale of Danica Pension Sweden in 2019. Net interest income increased as a result of good business activity and despite negative currency effects. The development was driven by a strong development for deposits, which increased 22% in the period, primarily because of corporate deposits at market rates, whereas the increase in retail deposits, primarily on ordinary transaction accounts was due to changes in consumer spending behavior during the period. Given the continuous increase in deposits and what seems to be a permanent negative rate environment, we have taken initiatives to adapt to the current market conditions. As announced today, we will adjust the threshold for the charging of negative interest rates for retail deposits at Banking DK to DKK 250,000. And in September, we announced that commercial customers in Denmark will be charged negative interest rates on the entire deposits. Both adjustments will take effect from January 1. Fee income, excluding one-off effects, maintained the good momentum from last year, driven by Wealth Management and despite lower remortgaging activity at Banking DK, lower -- and lower activity as a result of the corona crisis. Fee income in Q3 was affected by the compensation related to Danske portfolio play and was lower than the preceding quarter, which benefited from extraordinary items at Wealth Management. Trading income was up slightly from the level the year before on the basis of good customer activity at C&I and a reversal of negative valuation effects. The level of trading income in Q3 reflects continually good activity, however, not to the same extent as in the very busy second quarter. Expenses came in lower in Q3 than in the preceding quarter, not only due to seasonality effects, but also as a result of cost management initiatives. Compared to the same period last year, costs continue to be affected by transformation initiatives under our 2023 agenda and high cost for the Estonia case and compliance remediation. We now expect costs for the full year to be around DKK 28 billion. The outlook encompasses the planned cost for the Better Bank transformation, including severance payments, and in addition, full year may also include a small extraordinary amortization of intangible assets subject to final decision closer to year-end. The lower expected cost for 2020, a good starting point for the cost target for 2021 of around DKK 26 billion. Loan impairment charges, which, to a high degree, explained the difference to the financial results for the same period last year, amounted to DKK 6.3 billion for the period, but was unchanged at DKK 1 billion from the preceding quarter. The accumulated charges for the first 9 months account for between 85% to 90% of the charges we expect for 2020. Overall credit quality remains strong with very limited losses caused by the corona crisis. 2 items explain the development. Firstly, model-driven charges recognized in Q1 have been reversed and replaced by post model adjustments; secondly, an adverse development for our oil-related exposure. I will come back to credit quality later in this presentation. The net result for the third quarter in isolation was DKK 2.1 billion, which was lower than in the second quarter, whereas the net result for the full 9-month period amounted to DKK 3.1 billion against DKK 10 billion last year. Slide 4, please. Now let us take a closer look at the underlying development in net interest income. Compared to the same period in 2019, NII saw positive effects from good customer activity in the form of volume growth and deposit margins benefited from the Danish rate hike in March. The positive effect was, to some extent, offset by headwinds from the Norwegian krona, in particular, but also from the pressure on lending margins resulting from the low interest rate environment. Adjusted for currency effects, NII was up 3%. NII was stable from the preceding quarter as positive effects from FX and [ days ] Compensated for lower lending volume. As part of the pricing initiatives I mentioned earlier, we are now adapting to current market conditions in Denmark by lowering the threshold for the charging of negative interest rates on retail deposits from DKK 1.5 billion to DKK 250,000, and altogether, removing the threshold on corporate deposits, which was previously DKK 200,000, both with effect from January 1, 2021. In isolation, both initiatives will contribute positively to NII with an annual effect of approximately DKK 0.5 billion, all else equal and subject to changes in customer behavior and deposit balance developments at Banking DK. This should counterbalance some of the ongoing NII pressure. At the end of Q3, we had completed a significant part of our funding need for the year. Overall funding spreads have decreased from the level in the same period last year, due primarily to a change in the issuance mix and a notable decline in spreads for nonpreferred senior debt. However, the average spread for new funding exceeds the level for redeemed funding this year, confirming the negative impact on NII. Slide 5, please. Lending at Banking DK was down 2% year-on-year due to the declining demand for credit from municipalities, in particular because of the various effects resulting from the corona crisis support measures. Lending to commercial and retail customer was lower for the same reason. However, lending volume stabilized in Q3 from the level in Q2, driven by better housing market activity. The pressure on lending margins was most evident at Banking DK, where we see the effect of our customers shift towards longer-term fixed-rate mortgages with lower margins. At Banking Nordic, we saw growth in business activity, with lending up 1% or 4% in local currency, as a result of strong growth in retail Sweden and Norway, where the inflow from the partnership agreements continued in the first 9 months of the year. We also saw higher lending in Commercial Finland, NII at Banking Nordic improved from the same period last year, benefiting from the volume growth as well as an improvement in lending margins. C&I saw higher customer activity during the first 9 months, reflecting an increase in short-term credit facilities provided to support customer during the corona crisis. Overall lending at C&I was down due primarily to collateral value adjustments. However, with General Bank, lending was up 1% year-on-year. Specifically for Q3, we saw a decline due to lower utilization of short-term credit facilities from the very high levels in the preceding quarters when we granted a significant amount of credit facilities. NII and C&I benefited year-over-year from higher lending and significantly higher deposit volumes at market rates. The elevated level of deposits as well as some of the short-term lending facilities provided to corporate customers should be seen in the context of current uncertainty and are, therefore, not necessarily of a permanent nature. Slide 6, please. Let's have a look at fee income. Fee income maintained the good momentum and was in line with the same period last year. Overall, we saw a positive development for investment and pension activities, whereas activity-based fee income was lower due to the effects of the corona crisis. Adjusted for the extraordinary items with pension and insurance, fee income in Q3 was in line with income in Q2. Investment fee income in Q3 saw an adverse impact from the compensation booked for the Danske portfolio play case. At both Banking DK and Banking Nordic activity-based fee income declined from the same period last year due to the lower level of activity that resulted from the corona crisis. However, quarter-on-quarter, activity-based income was stable. Fee income at Banking DK benefited from better housing market activity in Q3 than in Q2, which generated higher fee income from lending and guarantees. In the first 9 months of 2019, remortgaging activity was historically high, which explains the decline year-on-year. Fee income generated by our capital markets activities at C&I came in higher than last year due to strong customer activity in the primary markets. However, usual seasonality explains the decline from quarter to quarter. At Wealth Management, fee income from investment and pension and insurance activities increased from the level last year due primarily positive effects at Danica Pension and a positive development for assets under management. Slide 7, please. Now let's turn to trading income. In the third quarter, despite being the summer holiday period, we continued to see relatively high customer activity, albeit not as strong as in the unusually good Q2. Positive developments and value adjustments, or xVA, also contributed to the result. Trading income was slightly up year-on-year as significantly higher income at C&I compensated for lower income in the other business units. The higher income at Corporates & Institutions was the result of stronger activity among our FI&C customers and improved market conditions as well as positive value adjustments. At both Banking DK and Banking Nordic, trading income came in lower due to lower investment activity among customers and at Banking DK lower remortgaging activity was also a factor as remortgaging activity was unusually high last year. At Wealth Management, trading income year-on-year as well as quarter-on-quarter was lower as a result of market development and a decrease in the yield curve that affected the investment result in the health and accident business in Danica Pension. Slide 8, please. On the expense line, we now see the first effect of the cost management initiatives we have launched during the year. In Q3, cost came in at DKK 6.7 billion, down 4% from the preceding quarter, driven by lower staff costs and seasonality, whereas the planned costs for the Better Bank transformation, compliance remediation and the Estonia case expenses -- impacted expenses for the first 9 months, which came in at DKK 20.4 billion, up 6% from the same period last year. In order to execute on our 2023 plan and to ensure strong progress towards a lower cost base, which is essential for our ability to remain competitive, we have recently announced further cost savings initiatives in the form of voluntary redundancy agreements in Denmark. During the next 6 to 12 months, we expect to discontinue up to 1,600 positions across the group. As a first step, we have initiated voluntary redundancy agreements for our staff operations in Denmark, and we maintain a strong focus on further cost-saving initiatives such as product simplification and nonpersonnel expenses. As we communicated on 27th of October, we now expect expenses for the full year to be around DKK 28 billion. This includes costs for the planned transformation of DKK 1.5 billion, including severance payments. And in addition, full year may also include a small extraordinary amortization of intangible assets related to IT software, subject to final decision closer to year-end. As I mentioned previously, we also reiterate our 2021 cost target of DKK 26 billion, including a transformation budget of around DKK 0.5 billion. Slide 9, please. The impairment charges for the third quarter reflect a further reversal of DKK 1 billion of model-driven charges we recognized in Q1 as a result of the outbreak of the coronavirus pandemic. In total, we have now reversed DKK 1.5 billion of the DKK 1.7 billion recognized in Q1 as a result of more positive developments for all key economic drivers and thus continually strong credit quality among most of our customers. However, as visibility regarding economic developments remains clouded and based on our timely approach, we have increased sector-specific post model adjustment by 1 point...

I think we have a technical issue. Please stay online. Thank you.

So I'm not sure where we lost you. I'll start with the impairment charges. In the third quarter -- the impairment charges recognized in the third quarter equivalent to a reported loan loss ratio of 22 basis points, however, with significant fluctuations between business units. As a result, the allowance account, which stands at more than DKK 23 billion in total, now includes impairments recognized for anticipated credit deterioration of DKK 2.2 billion. The oil-related exposure, the offshore service and drilling segment, in particular, continues to be challenged and led to additional impairments of DKK 0.5 billion in Q3. This exposure continues to be challenged by difficult market conditions and low activity that makes restructuring increasingly difficult, and we see pressure on collateral values affecting impairments. Total impairment charges for the first 9 months, which amount to DKK 6.3 billion are in line with our guidance and represent most of what we expect for the full year. Under current assumptions, the impairment charges booked year-to-date should represent 85% to 90% of what we expect for the full year. Slide 10, please. Our capital position remains strong with the reported CET1 capital ratio of 18.2% at the end of the quarter. The CET1 capital ratio was up 0.6 percentage points, driven mainly by a lower market risk REA and the effect of net earnings after dividend accruals. The fully loaded CET1 capital ratio was 17.9%. Total REA came in at DKK 19 billion lower than at the end of the last quarter due to a DKK 20 billion decrease in market risk as the higher risk levels we saw in Q2 reversed in Q3. Credit risk REA was up DKK 1 billion as we are starting to see the effect of implementation of the EBA guidelines, which was countered by lower lending volume and FX effects. Counterparty risk was unchanged quarter-on-quarter. In Q4 2020, we expect an impact on credit risk REA from model updates related to further implementation of EBA guidelines of DKK 20 billion to DKK 30 billion, all else equal, with further increases expected in 2021. The leverage ratio was unchanged at 4.4% according to both transitional and fully phased-in rules. Slide 11, please. And then the revised version of our financial outlook for 2020, which was announced on 27th of October. The outlook for NII is unchanged. We expect net interest income to be around the same level as in 2019 as margin pressure and higher funding costs will offset continued volume growth. Net fee income is now expected to be slightly lower than the level in 2019 as improved developments in financial markets has partly offset lower remortgaging activity. Fee income is subject to uncertainty regarding assets under management, customer activity and market development. Expenses now expected to be around DKK 28 billion, including planned costs for the Better Bank transformation. In addition, full year may also include a small extraordinary amortization of intangible assets. Loan impairment charges in 2020 are expected to be significantly higher due to the impact of the coronavirus pandemic on the economic outlook, with most impairments recognized already in the first 9 months of the year. And finally, we now expect net profit to be in the range from DKK 4 billion to DKK 4.5 billion due to the improved developments in the financial markets and continually good progress in the underlying business. Slide 12, please, and over to Claus.

Thank you, Stephan. Those were our initial comments and messages. We are now ready for your questions. [Operator Instructions] And a transcript of this conference call will, as usual, be added to our website and the IR app within the next few days. Operator, we are ready for the Q&A session.

[Operator Instructions] We have a question from Mads Thinggaard from ABG Sundal Collier.

Mads from ABG here. The first one here is on costs. Stephan, I think you mentioned before that the transformation costs you expected for '21 was still DKK 0.5 billion and also with DKK 26 billion for 2021 guided still. And then I was just wondering a bit about the kind of lower -- I mean the drop in guidance -- cost guidance for 2020 and what implication that has for the underlying cost base in 2021? Is it -- I mean, can you kind of help a bit about, it's probably still DKK 22.5 billion. But I mean, is it -- how much is from the FTE cut? And how much is from the improvements, I think you pointed to in Q2? That was the first question. And the second question, I was -- I mean now you're talking a bit about a potential small intangible write-down in Q4. It doesn't sound like it's a very big one. You still have quite some intangible assets. I think it's DKK 9 billion. If you were to do a larger adjustment, would you consider adjusting the kind of profit for the dividend decision? Would you adjust for intangible write-down for that base? That was my 2 questions.

Mads, you did very well in putting many questions under the heading of 2 questions, but I'll try to at least attend some of them. I confirm that the DKK 26 billion guidance for '21 has DKK 0.5 billion for the transformation budget as has the 2020 guidance, DKK 1.5 billion also for the transformation budget. The question I heard you saying is like what does the performance in 2020 tell about 2021? I think what you can say about 2020, obviously, is that since the guidance on AML remediation Estonia remains the same as well as the transformation budget remains the same, that underlying cost's obviously developing a bit better. I think that gives us, first and foremost, reason to believe that we can achieve the '21 -- 2021 target. With any more specific detail, I would still like to wait until we have the full year results and then go to the guidance in 2021. The intangible asset stuff, first of all, it's something where IFRS also sets certain rules. So you are not completely free, obviously, in what you want to do there. What we are doing is reviewing mainly IT and software to find out where we need to adapt following the reduction from 4 to 2 business units. So that may shorten the life cycle of one of the other stuff, and we have also retired quite some product this year. So that may also give reason for proper early write-off. We expect, let's call it, a couple of hundred million in 2020, which then, as I said, would be on top of the cost expense line and still wouldn't change our net profit guidance. I hope that answers most of your 2 questions?

If you were to make a -- I mean, this is a bit theoretical, if you were to make a much larger adjustment, would you then also adjust the dividends that you base your dividend -- or the net profit, you base your dividend decision on?

I think, Mads, this is now more speculation than anything else. As I said, IFRS sets limits to what you can do. And I think the IT stuff is something that makes sense and can be done, and I think we should stop the discussion there.

Our next question comes from the line of Sofie Peterzens from JPMorgan.

Yes. It's Sofie from JPMorgan. I was wondering if you could give any update on the U.S. investigation. I realized that in the report, you're just saying that you don't have much additional information, but there were some press articles, a few weeks ago now, where the CEO allegedly was in Washington discussing with the U.S. Is there anything you potentially could share of these discussions with us? Do you have any view on a potential time frame? Have you had any additional information requests? And what are kind of the next points that you are waiting for? So that would be my first question. My second question would be on the dividend. I recognize that we have a dividend ban in Europe, but have you had any communication with the Danish authorities on potentially being able to pay a dividend on 2020 earnings? And kind of what's your view on a potential dividend next year?

Okay. Thank you for the question. This is Chris. I'll answer the first one. There is -- beyond what I've just said and which is in the report, there's essentially nothing to say about this, the Estonia case. I mean we have worked through almost all the things we have to do, and the ball is, in that sense, in the hands of the authorities. With regards to my alleged visit to the United States, I think we are on a continuous basis in discussions with all the authorities. And I think it will be very wrong to give more information on that. And you -- for the...

Yes. On the dividend issue, again, you are right, the dividend discussion is still on. We are accruing 60% of net profit so far. I sense a bit of a discussion that is also kind of flowing a bit back and forth depending on what the corona contagion wave version 2 is doing and what government programs are doing. I think it's far too early to really judge where the jury will end up. I have also heard quite some comment across Europe that banks with a stronger capital ratio may be allowed at least to pay part of the dividends. But again, I think it's far too early to really tell. And we stick to what our dividend policy says, and then we need to see where we get to in the spring next year.

Our next question comes from the line of Kazim Andaç from Deutsche Bank.

A question on the impairment outlook. Just to get a bit color on how should we think about impairments in the coming quarters? Given increased restrictions across Europe and their, like, impacts on growth. Do you plan to revisit your macro assumptions here? How should we think about the impairment outlook going forward? And the second one is on the growth and margin outlook in your core markets. You continue to lose market share in Sweden, in particular mortgage segment, while you have the highest negotiated price among the major players. If you can elaborate the strategy here and perhaps the bank's performance in other markets on mortgage and business lending? And lastly, if you -- please provide us on how back book and front book -- loan book repricing developing again in mortgage and corporate segments? Any color, much appreciated.

On the impairment outlook, I think all what we are -- one general answer, and then I'll try to give more detail. I think in general, what we see right now on the macro side is better than what we have modeled in our IFRS 9 models. That is why you see these continuous releases throughout the last 2 quarters. I think it's a bit too early to really give the 2021 outlook on impairments. But so far, as I said, we have quite a good chunk of post-model adjustments. And we will -- once we are having the full year in -- and again, have a little bit of more visibility of how the second wave of corona will look like and how government programs will develop going forward, we will update you on that part. But so far, I am of the opinion that the second wave of corona will probably go along with less, far less intrusive lockdowns that we have seen so far and also with far less disturbance of the international supply chains. So in that sense, let's see. On general, I'm not sure whether I got your mortgage question completely right. Mortgage business is something that is pressured this year, specifically on the margin side because many of the customers are going for longer-term fixed rate stuff, which obviously has lower margins. That's kind of a bit across markets. In general, we still have good growth in all our Nordic markets on the loan book. In general, I think that's no surprise if you look at how the interest rate development was over the last years, back book margins tend to be better than front book margins. And I expect that to be the fact and the reality for quite a while since the interest rate environment from my point of view is not going to change, at least not to the better.

Our next question comes from the line of Jakob Brink from Nordea.

Just coming back to costs, please. So just -- I guess, I was a bit surprised about your update the other week where you're now guiding for the lower end of the DKK 28 billion to DKK 29 billion range. I think also, I can't remember exactly what you said, but that you've been hinting towards the upper end of the range before. So what has changed and also on that matter, you showed seasonality between Q3 and Q4 is around DKK 1 billion. And as you pointed out or as -- looking at your presentation, you have still around DKK 700 million left of transformation of Better Bank costs for Q4. So that would be a DKK 1.7 billion seasonality and hence, your cost would end at the upper end of the range. So what is it that it's so much better than normal? And what is it that is -- or not happened in Q4 this year? I guess that was my first question. Then just coming back to margins. In your margin bridge -- or NII bridge in the presentation, you have a small, I think, DKK 3 million, positive, Q-on-Q impact from lending margins. But looking in the fact book, basically, every segment is down, apart from LC&I, which, of course, is a smaller part of the group. So how should I look at this, please?

Are you already? Okay. I honestly do not remember that I have been hinting or that we have been hinting at cost between -- at costs being at the higher end of DKK 28 billion to DKK 29 billion. So wherever you got that from, at least it was not intentionally from us. I think what the consensus has been for most of the -- at least last 6 months that I am here or 7 or 8 is that it was kind of in the midpoint between DKK 28 billion and DKK 29 billion. Then you asked about seasonality for Q4. Now traditionally, Q4 has a number of IT projects and other stuff that normally gets finished, delivered and then also built. So there's a bit of that seasonality, I think which is the usual stuff. The second part is, as I mentioned before, we expect to book quite some severance pay in Q4. So the transformation budget now is roughly DKK 800 million so far for the first 9 months, and we expect that to go to DKK 1.5 billion, which is the originally expected volume. As I said before, a mix of IT, other project and stuff that we do under the Better Bank transformation as well as severance. So I think that should pretty could explain the difference between where we are right now and where the last quarter should end up.

But just on that -- sorry. Can I just follow up on that? Yes. So exactly, this is my point. So if you have DKK 700 million left of transformation or severance for Q4, I guess, the transformation or DKK 700 million remaining is earmarked for severance pay. And then you have the normal seasonality. If you don't -- or if you have that, you end way above the DKK 28 billion. So why is the normal seasonality not there in Q4 this year?

Now, first of all, it's not DKK 700 million severance pay. So part of the normal seasonality will be part of the transformation budget because a number of the projects or the good part of the project that we are doing this year is part of the transformation budget. So that is one part of it. I would expect severance to be like DKK 500 million for the full year, keeping in mind that we have booked already something like DKK 237 million. So I think that should put that into perspective. And then other than that, some of the better performance to get to the DKK 28 billion is obviously also reflected in underlying cost.

So you said DKK 500 million, leaving around DKK 250 million for Q4 of severance. Is that correctly understood?

Yes. Yes. It's DKK 500 million in total, and we have booked something like DKK 237 million, if I have it correct, and the remainder would then be booked in Q4. If we will -- if we would have higher numbers, I would guess that we will probably also book a bit more and then exceed the transformation budget in that sense, but that should still keep us well within the range of our net profit guidance.

But how can you only book DKK 250 million, if laying off 1,600?

Yes. We are not laying of 1,600 this year. We are laying 1,600 off over the next 6 to 12 months. And IFRS, as you know, requires you to have certain trigger points met before you can book severance pay.

Can you please repeat your question on the lending margin, Jakob?

Yes, it was just in the presentation, you have this slide showing NII from Q2 to 3. I think it says a positive lending margin impact of DKK 3 million. But if I look at the fact book on every single business unit, except for LC&I, it's down quarter-on-quarter, i.e. the margin -- lending margin?

Yes. That's not exactly a picture I can recognize. It's -- first of all, it's very small numbers, right? So it's down in Banking DK and Banking Nordic, but it's up in C&I and it's also up in Northern Ireland.

Maybe we take this offline since the amount is not -- probably not driving share price really today.