Good day, and thank you for standing by. Welcome to SEB's Financial Results Third Quarter 2026 Webcast and Conference Call. At this time, all participants are in listen-only mode. After the speaker's presentation, there will be the question and answer session. To ask a question during the session, you need to press star one one on your telephone keypad. You will then hear an automatic message advising your hand is raised. To withdraw your question, please press star one and one again. Please be advised that today's conference is being recorded. I would now like to hand the conference over to our first speaker today, Jonas Ahlström, acting President and CEO. Please go ahead.
Good morning, everyone, and a warm welcome to SEB's Q3 telephone conference. While I've met several of you before in my previous roles in the bank, for those of you who don't know me, I've been with the bank for 20-plus years, most recently as Chief Operating Officer. But I've also headed the Corporate & Investment Banking division and prior to that, Baltic Division. And several of you I met a long time ago having been both a buy-side and a sell-side analyst. These are now my first quarter results in the capacity of acting CEO. While the past few weeks have involved a period of transition, quick adaptation, and a couple of external communications, focus across the organization has remained on serving our customers and executing on our long-term plan. Against this backdrop, I look forward to sharing our results for the quarter.
Let me start with a few highlights from the quarter. Starting with lending growth, we saw growth across all business divisions, reflecting continued good credit demand across the bank. We also had a strong quarter in terms of fees with a record Q3 for net fee and commission income. FICC, however, faced challenging market conditions which weighed on a net financial income line. On costs, our strong cost discipline continues to yield results, and we now expect to come in around SEK 500 million-SEK 600 million below our 2026 cost target. This includes the $33 million announced last week related to the resolution of historical matters with the DFS. Last week, we received a decision from the Swedish FSA related to our internal ratings-based models.
While the conclusion of the review does not point to weaknesses in SEB's credit risk assessment, but rather in the model's regulatory compliance, a temporary increase of REA will be implemented from December 31st. Looking ahead, our focus is on getting the new models approved. We expect to start submitting these models in the third quarter of 2027 at the latest. As for the impact of the decision on our capital position, Christoffer will go into this in more detail later on. In light of continued loan growth in combination with the capital impact from the Swedish FSA's decision, the bank will pause its share buyback program. Turning to the next page, Sferical, our shared AI system set up built on NVIDIA technology is now live. Last week we celebrated the launch with an event here in Stockholm with NVIDIA founder Jensen Huang.
As previously mentioned, Sferical provides SEB with secure access to advanced third-party independent AI infrastructure and compute capacity in Sweden. It enables us to develop, test and scale AI solutions faster while maintaining control of our data, models, and operations. Combined with the knowledge sharing across the consortium, it strengthens our ability to bring AI into the business both responsibly and at scale. in connection with this, we also wanted to provide an update on the AI triangle that we have shown before, particularly the running the business dimension. We are already seeing good progress on how AI is being used across the bank and wanted to share a few examples. We are using AI-assisted investigations to improve efficiency in financial crime prevention. In software development, agentic engineering is helping our teams become more productive with leading teams increasing their output significantly.
We have also largely automated a high volume workflow in collateral management, having reduced manual handling by some 90%. in business and retail banking, we are using AI to support our client-facing staff in our call center operations known as 365. Customer calls are analyzed to help staff determine the appropriate next step for the customer. Approximately 70% of customer calls are now analyzed within hours. Previously, it was fractions of a percent, twice or four times per year, and this is just the beginning. Turning to the next page, we want to highlight some business developments from each division. Starting with Corporate & Investment Banking, as previously mentioned, Q3 was a record quarter in terms of fees and commission income.
Also, SEB was ranked number one among tier 1 institutions in the Prospera Nordic Equity survey and maintained its number two position overall and closing the gap to the number one position. Additionally, we ended the quarter with record assets under custody of almost SEK 23 trillion. We also continued to receive external recognition. SEB consistently ranks among the leading providers in industry surveys such as Global Custodian and Global Finance, and most recently ranked number one in the Nordic Prospera survey. Over time, we have built a strong position in post-trade services. Today, we are the leading provider of sub-custody services in the Nordics. We are also the largest custodian and depositary bank for non-bank-owned asset managers in the region, with a market share in excess of 50% in both segments. For Business and Retail Banking, this is an update of the numbers we showed last quarter.
Looking at the share of new mortgage lending in Sweden captured by SEB each month, we have seen positive data coming since May this year. In August, the most recent data point, SEB managed to take 19% market share of new mortgage lending. This is an encouraging early sign of our efforts to regain market share in the Swedish mortgage market. Within the mortgage process, we have also continued to improve and redesign key customer journeys supported by AI-driven solutions that help increase efficiency, shorten processing times, and improving the customer experience. I also want to highlight that the legal merger of AirPlus and SEB Cards is now fully complete. It has been a multi-step journey across platforms, markets, and operations, and the teams have delivered according to plan. Looking to Wealth and Asset Management, there are a few things I would like to highlight.
First, we have continued to see positive flows with net flows of SEK 16 billion year-to-date and excluding the pension reform impact in Lithuania, that number is SEK 30 billion year-to-date. We are also seeing good development on the lending side with volumes up 6% compared with last year. It was also pleasing to see the continued recognition of our asset management business in the annual Prospera survey of institutional investors as we was ranked the leading external asset manager in the Nordics for the fifth consecutive year. The ranking is based on feedback from investment decision makers across Sweden, Finland, Denmark, and Norway. This year, SEB achieved a score of 4.09, the highest score recorded by any participant in the survey since its inception. The result reflects the strength of our investment offering, our teams, and our long-term client relationships across the Nordic region.
Turning to the Baltic division, both consumer purchasing power and corporate investment activity supported broad-based lending growth with continued strong development both in mortgages, SMEs, as well as large corporates. In other words, the structural growth story of the Baltic economies continues, and it is also pleasing to see that we see some support on margins from rising interest rates. We also wanted to highlight a recent AI initiative now live in the Baltics. Since July, an AI-powered voice bot has been handling calls during night shifts, helping customers with card issues and internet banking access blocks. It has significantly reduced customer waiting times while creating operational efficiencies. The next step is to expand the service into daytime operations and explore additional customer journeys where the technology can be leveraged further.
Turning to the next page, I will not spend much time on this as we already covered the lending growth to some extent. To recap, we see continued lending growth with FX adjusted lending up 2% quarter-on-quarter and 5% year-on-year. We are now seeing growth across all of our divisions. Moving on to one of the bank's favorite slides, and definitely one of mine. Our focus on positive operating jaws continues. This is supported by continued cost discipline, which is delivering results as reflected in our expectation to beat this year's cost target by some SEK 500 million to SEK 600 million. Over to you, Christoffer.
Thank you, Jonas. If we turn to the financials and we will go to the summary on the next slide. Net interest income is increasing both compared to the same period of last year and compared to the previous quarter by 3% and 2% respectively. The increase reflects a combination of the positive impact from stabilizing and increasing interest rates, as well as continued support from growth in lending and deposit volumes. We can see that the underlying trend of strengthening corporate loan demand remains intact, and on the private side, mortgage volumes improved in the quarter. SEB also increased its market share in both corporate and mortgage lending in Sweden during the quarter. Net fee and commission income posted a solid Q3 up 7% from Q3 of last year, recording the highest Q3 fee and commission income on record, as Jonas mentioned.
The decline compared to the previous quarter reflect a seasonal slowdown within most capital market related activities compared to a very strong second quarter. While both net interest income and commission income grew from Q3 last year, total revenues were largely unchanged due to a decline in net financial income. This is primarily related to the challenging market backdrop that Jonas also referenced. The decline compared to the same quarter last year was around SEK 600 million and impacted primarily our FICC businesses. We will come back to this a little bit more in detail when we go through our usual net financial income breakdown. Net other income contributed negatively this quarter mainly reflecting a negative contribution from an associated company held in our central functions.
Turning to operating expenses, SEK 8.2 billion for the quarter increased moderately from Q2. The increase relates primarily to the impact on our staff costs from equity-linked remuneration, and that is resulting from the higher SEB share price. We also had somewhat higher technology costs that were offset by seasonally lower costs for external consultants and marketing. The underlying cost development is steady, well within our full year cost target, reflecting our ongoing cost consolidation. As we flagged already in Q2, we are tracking well to come in below our cost guidance for the full year and now considering the seasonality of our cost base, along with investments that we expect to book in the final quarter of the year, we have now the visibility to communicate that we expect to beat our full year cost target of SEK 33.4 billion by between SEK 500 million to SEK 600 million.
The main drivers contributing to the faster than expected cost consolidation during the year are related to effects from our external hiring pause, which is helping us consolidate the cost base. Also lower operating expenses within AirPlus, which have accelerated the cost efforts to offset some of the lower income trajectory, and also lower pension costs which have been supported by the surplus in our defined benefit pension scheme. Net ECL in the quarter of SEK 179 million continued to reflect a solid underlying asset quality across our credit portfolios. This quarter we released portfolio overlays amounting to some SEK 60 million from business and retail banking exposures. Overall, our model portfolio overlays stand at SEK 800 million. The ECL in the quarter corresponds to 2 basis points of our total credit exposure.
Levies of SEK 857 million is in line with our full year guidance of SEK 3.5 billion and a tax rate of 20% also in line with our guidance. Net profit SEK 7.5 billion and earnings per share of SEK 3.85 and an ROE of 13%. As usual, we provide the ROE also adjusted for the effects from our over-funded pension scheme. This quarter that ROE is 14.4%. Turning to the next slide and net interest income. The development in the quarter is a continuation of the trend that we have seen over the past few quarters as policy rates have turned from declining to steadying and in some geographies rising.
These dynamics resulted in the first quarter of this year marking a turning point for net interest income in line with our expectations of that 3-6 months lag from the last cut to the full effect on our balance sheet. The same dynamics are now working their way through the balance sheet in the other direction. As a result, the effect of the first ECB rate hike in June, further supported by continued lending growth, is largely now reflected in our Baltic NII. The impact of the second hike that we had in September is still being realized. For the Baltic division, this added up to an increase of NII of SEK 200 million in the quarter.
Within CIB, the increase in NII was mainly driven by higher volumes, particularly those volumes that were added towards the end of the second quarter which now had a full effect in Q3. This resulted in an overall increase of around SEK 200 million. Margins in CIB were largely unchanged on the whole, and within markets the NII declined somewhat compared to the previous quarter. For business and retail banking, the NII increased by SEK 100 million, and that is also volume driven as policy rates were unchanged and STIBOR largely flat over the quarter. Mortgage margins were also moving sideways. Within Treasury, net interest income declined due to increased funding costs and higher funding volumes in the quarter. Finally, the weaker Swedish krona had a positive impact on NII around SEK 70 million, and the day count contributed about SEK 100 million.
Turning to net fee and commission income on the next slide, the development in the quarter reflected the typical seasonal pattern with a decline in capital markets related income compared to a very strong Q2. Nonetheless, it is a record high fee and commission income for the quarter, both within CIB and for the group. The most important drivers of the year-on-year increase include the securities related commissions rising 13%, supported by mutual funds and custody fees. Advisory fees, lending fees, and other commissions were largely unchanged. The nature of our card business within business and retail banking is also impacted by the Q3 seasonality due to the high share of corporate activities. Compared to the same quarter last year though, card and payment fees increased by 5%.
We can also note that this is the first quarter where AirPlus is comparable on a like for like basis following the exit market closure at the end of Q2 last year. Turning to the next slide and the net financial income. As mentioned earlier, the lower contribution from CIB in this quarter is primarily a reflection of a broad-based repricing of global rates markets during the quarter, where rate expectations shifted quite dramatically from lower to higher rates in a quite rare way. This market movement coincided with a period of muted customer activity over the summer period, which then impacted our client flows as well as our market-making activities. As you well know, we have a strong standing in the rates markets in the Nordics, and we are a leading market maker across many products.
During the quarter, we have seen long rates move away from many of our customers who typically seek to lock in lower rates in their financing. Against this backdrop, we have continuously provided markets for our customers, and that has impacted us both via the valuation effects on our inventories as well as the reduction in the flows. Importantly, our risk management has served us well throughout these difficult market conditions. If we look at rates markets in October or the weeks we have had so far, they have remained challenging. Of course, as we get closer to year-end, customers will typically start to pull back on positions and reduce risk. Early days, but we will come back as usual, in our pre-close, if there is any further developments to report on. As I mentioned in Q2, we are in the process of disposing of our shareholding in Euroclear.
This quarter, we booked a second part of that capital gain resulting from the divestment amounting to EUR 36 million. Going the other way though, this quarter, we have had negative marks on other large holdings, including positions in our FinTech and GreenTech venture portfolios, offsetting the gain from Euroclear. The net contribution from these holdings was immaterial in Q3. Turning to the next slide and the development of our capital position. We closed the second quarter with a management buffer of 250 basis points. We added 37 basis points from net profit in Q3, and our continued credit growth consumed 35 basis points, and the FX effect due to a weaker Swedish krona reduced the buffer by 22 basis points.
Positive asset quality trends and other effects added 20 basis points to the buffer, which then took us back to the 250 basis points also at the end of the quarter. Turning to the next slide, I would like to make a comment on the Swedish FSA decision that Jonas mentioned at the start of the call. As announced last week, the Swedish FSA has concluded a review of our IRB models, our internal rating-based models. Due to the deficiencies identified, the Swedish FSA will implement a temporary increase in our capital requirement of SEK 10 billion as of the 31st of December this year. As communicated by the FSA, the findings relate to regulatory compliance and not to the risk level in our credit portfolio, nor to the capital position in relation to our credit risks.
We already hold some SEK 10 billion in extra capital since 2023 in the form of a Pillar 2 requirement that is equivalent to around 100 basis points. This add-on will now be moved to Pillar 1 as part of a change in practice at the Swedish FSA. These two effects will be implemented simultaneously. Looking forward, we expect to start submitting models no later than the third quarter of next year, so within the next 12 months. Thereafter, the Swedish FSA will process the applications. The additional capital requirement will be lifted once the models have been approved and implemented. As this is a temporary capital effect, we will continue to operate our customer franchise as usual. We will also continue the work with our capital efficiency actions to support mitigation. This includes obtaining external ratings on more counterparties, which is a very efficient tool.
It also includes continuing our readiness for SRTs, which we have talked about for some time, including other capital efficiency measures that are already underway. Since the Swedish FSA's decision last week, we have revisited the bank's capital plan together with the board. Based on these discussions and taking into account the credit demand from our customers, our dividend payout ratio of around 50%, and our plans for mitigating actions, we expect to close the year with a management buffer of around 150 basis points. This is based on FX rates as of the end of the third quarter. Finally, to conclude on the last slide, our financial targets, which remain unchanged, which was also part of our communication from last week. We continue to aim for a payout ratio of around 50%, a capital buffer of 100 to 300 basis points, and a return on equity of 15%.
With that, we are open to take questions. Over to the operator.
Thank you, dear participants. As a reminder, if you wish to ask a question, please press star one one on your telephone keypad and wait for a name to be announced. To withdraw your question, please press star one and one again. Please stand by, we will compile the Q&A roster. This will take a few moments. Now we are going to take our first question, and it comes the line of Andreas Håkansson from Nordea. Your line is open, please ask your question.
Good morning, everyone. Can I say it is a pity that Johan is not around so we can thank him for the last 10 years. I think we all would agree with that he did a very good job. Anyway, on to the capital side. If we do the pro forma impact now, you are at 120 basis points and you expect to be at 150 basis points. We have seen that the FX move in this quarter was some 20 basis points, and we have seen just a couple of quarters back that you have FX swings around 40, 50 basis points in a quarter, which means that that could take you very close to the 100 basis points bottom of your range. In the board discussions, where would you like to potentially start to change the view? Would it be volume growth?
Is it a dividend, or how should we think about the capital that is getting quite tight now? That's my first question.
Good morning, Andreas. Thank you for your kind words regarding Johan. Christoffer, you can comment on the FX. Those are the numbers, Andreas. I think the way we're thinking about this is that we have the range of 100 to 300 basis points, and within that, we continuously then work around the capital distribution growth and capital retention. So those are the ones that we're triangulating, and when we are towards the lower end of the buffer, we will think about making our way upwards. We are at the higher end and expecting to stay there, we're making our way downwards. Now, the expectation that we're providing over 150 basis points is based on what we now can see in terms of the credit demand for the remainder of the year, in combination with some of those mitigating actions that are underway.
We'll have to, of course, review what other developments could happen in the quarter, but those are the variables that we're working with.
Okay. Thanks. Second question, just you said that you had a benefit now that helped you on the cost-saving side from the pension foundation. Could you tell us how much have you increased that takeout, or have you increased it, and how much more could you increase it before you hit the maximum, so to say, on a yearly basis?
Yeah. There are two effects from the over-funded pension scheme. One is the cost effect, and that is also linked to the size of the surplus, and that then comes through to lower pension costs in the P&L. The other aspect that you are also referring to the takeout, that is effectively upstreaming capital from the surplus into the CET1 ratio effectively. The second part that you are asking about, the upstreaming or the takeout, we have historically now the last couple of years been around SEK 1 billion when the surplus started to grow, then we were around SEK 2 billion, and the maximum there is somewhere below SEK 3 billion. Depending on the size of the surplus when we come to the year-end, we will take that decision.
On the cost side, how much have you been helped, and can you do more on that side if you do early retirements or whatever?
Last year, 2025, the positive pension contribution P&L wise was just over SEK 900 million. This year, they are tracking towards SEK 1.2 billion or so. That is the positive P&L effect. The surplus there is there to cover pension costs. If there are other pension costs that would occur, then we could revisit those and the opportunities to cover those.
Perfect. Thank you.
Thank you.
Thank you. Now we will proceed with our next question. The question comes from the line of Namita Samtani from Barclays. Your line is open, please ask your question.
Morning, and welcome, Jonas, and thanks for taking my questions. My first question, just to fund further potential lending, is there going to be this continuation of issuing wholesale funding? I just wondered why deposits cannot be used to fund this growth. Surely on the corporate side, at least, I would imagine SEB is still a dominant bank there, and you would have one of the best deposit franchises. But I also see deposits decline quarter-over-quarter in Corporate & Investment Banking, so any color there would be helpful. Then secondly, just on the IRB models, you write the recently announced 100 basis points add-on is related to regulatory compliance and not to the level of capital in relation to the bank's risks. Could you just explain what regulatory compliance exactly is? Because I do not really expect there to be regulatory compliance issues at a bank like SEB.
Just on the previous 100 basis points add-on on your CET1 requirement, what is that related to? Is that regulatory compliance too, or is that to do with the assumptions and your models not being prudent enough? Thank you.
Thank you very much for your questions, Namita. So if I start with the deposits, it is correct that they are dropping a little bit in this quarter. If you remember, we specifically flagged at the end of Q2 that we had quite a sizable deposit coming through towards the end of the quarter that we were expecting to come out again. So that explains the movement in the deposit base. As far as the wholesale funding activities are concerned, there is no change. We have been active in funding markets in the third quarter and have taken up some pre-funding as well. When it comes to your question on the IRB models, yes, it is correct that they are not related to an expectation or an indication from the Swedish FSA that there is an underestimation of capital or capital insufficient to the risks.
It has to do with regulatory compliance.
This comes back to a number of new supervisory expectations that have been introduced over the last couple of years related to the technical standards of how to design, validate, document, process these models. One practical example is data. The amount of data, the amount of historical defaults is one technical standard. Here we have very limited historical defaults, and therefore, we need to work our way around the deficit of data attributable to the fact that we have limited historical defaults. So those are the kind of technicalities that we are now working our way through to ensure that we are in compliance with those supervisory expectations that have also been developing over the last few years.
That is helpful. Thank you.
Thank you so much, Namita. Now we are going to take our next question. The question comes line of Nicolas McBeath from DNB Carnegie. Your line is open. Please ask your question.
Thank you, and good morning, Jonas and Christoffer. I wanted to follow up a bit on Andreas' question on the pension surplus. If you could please confirm how large the surplus is now and why the ROE headwind is actually lower in this quarter than the previous quarter. I would have assumed that the pension surplus has widened with higher interest rates and rising stock markets. If you could start there, please.
Morning, Nicolas. We don't disclose the size of the pension fund surplus. The adjustment that we're making for the ROE is both on the cost and the capital side. Those are both two adjustments that are made, and the outcome of that takes you to the 14.4% versus the 13% reported.
All right.
Was that answering your question, Nicolas?
Yeah. That's a good start. I was wondering, because you mentioned up to SEK 3 billion in potential contribution in Q4, is there any reason why you shouldn't maximize that contribution given the position you are now capital-wise? If so, is that a SEK 3 billion tailwind then that we should take into account for capital in Q4? Would that also be included in the 150 basis points buffer that you alluded to for your CET1 ratio margin to the requirements?
The decision in and around that takeout is done at the end of the year, then based on where we stand on the size of the surplus. The capital effect will not be seen in the fourth quarter. It will come through during the course of 2027, gradually over the year. It's linked to the pension costs.
All right. I also wanted to ask, because we've been talking about this before, but why are you not making a more comprehensive action with your pension surplus? Have you investigated what is the alternatives maybe to do, selling the assets and liabilities maybe through a pension risk transfer or what is the status there? Any reason why not more proactively seeking to deploy this capital more efficiently than today?
Well, the main regulatory framework around this is called Tryggandelagen in Swedish. These are the rules that stipulate what can be done with the means inside of the foundation. The purpose for the foundation is to cover costs for pensions, as we talked to you about earlier. In that sense, of course, having an overfunded pension fund that allows us to compensate ourselves for pension costs is a valuable asset to have. We're also providing you with the impact on both costs and capital as a result of that. Also bearing in mind, if we go back a number of years, this was more in the order of closer to not being a surplus at all, and we've had historically, of course, when we've had a deficit in the fund. So now we're in a comfortable position.
We are adhering to those strict regulations in and around this. We are proactively looking at ways in which we are compensating ourselves for pension costs, which is exactly what we are doing.
Yeah. I understand the compensation and the cost benefit that you get. If you get SEK 1 billion in cost benefit and you have a surplus, I think it is around SEK 40 billion, that is not really a good compensation or a good return on that kind of surplus. I was just wondering, and also if you were to sell it, both the assets and the liabilities through a pension risk transfer, then there would be no risk at all for the bank, right? That would actually reduce risk. Is this something that you have investigated? Is it at all possible or is there some kind of regulatory prevention that makes it impossible?
Well, I will just come back a little bit to the historical context of this pension fund. It has been there for a long time. It has been in a deficit, and now it is in sizable surplus. It is part of how we run the bank and finance our pension costs. I think there is a history of both has been in deficit and in surplus. Now we are in surplus and that opens up opportunities to work with that surplus and that is what we are doing.
All right. Thank you.
Thank you.
Thank you. Now we are going to take our next question. The question comes from the line of Magnus Andersson from ABGSC. Your line is open. Please ask your question.
Yes, hi. Good morning and welcome into your new role, Jonas. Just starting with the SFSA order, I was just wondering whether you can say anything about approval timeline at all. You said you will submit new model applications by Q3 2027 at the latest. It is a while until then. Do you think this for any reason should be potentially faster process than usual or is it just complete uncertainty? Secondly, related to that, do you expect to get gradual approvals as you talk about models? Will there be a gradual release of your 120 basis point buffer impact or do you think it will come at once? That is the first. Secondly, just on costs, will this increased regulatory burden or by temporary incur any extra costs for 2027, 2028?
Also related to costs, will you be prepared at some point to quantify the potential impact on your cost base from the usage of AI as you talk a lot about it? Thanks.
Thank you, Magnus. If we start with the approval process, what is in our control is to get the models to the Swedish FSA as quickly as possible, and as we said, no later than Q3 of next year. We are in a good dialogue with the FSA, and they have told us that they will process those applications as soon as practically possible. It is complex material, so we cannot expect a quick turnaround, but there is a good discussion there. There is also a constructive discussion to your point around getting certain models approved and then moving on to others. That is the dialogue that we are having.
Good morning, Magnus. I think on the cost side, the link to IRB, that is already reflected in our 2026 run rate. When it comes to AI, I think more broadly speaking, we are moving from piloting and experimenting with a large number of use cases to focusing on really scaling fewer use cases and extracting the benefits. I think you should see that in the light also our communication on costs today and this work continues going forward.
Okay. I know you said before it is reflected in your 2026 cost base, but this was hardly an expected move. At least, I think it surprised a lot of us on Wednesday last week. It will take much longer than you originally thought. But you are saying that there will be no additional impact on the cost line from this beyond 2026?
The work to get the approved models has been underway, and what we mentioned was really that the remediation started earlier on. We said starting this year that one of the prioritized areas has been related to ensuring that we have regulatory compliance, resilience, and AI technology. In and around that, the remediation activities from our IRB models, which are already the deficiencies from the SREP 2023, were part of that plan.
Yeah. Okay. Just finally, the Baltic add-on you have, are they related to the same issues that you have in Sweden?
The Baltic IRB add-ons is something that we've had with the ECB. In terms of data deficiencies because of historical low defaults, yes, that applies to those challenges as well. There are two different approval processes, one with the Swedish FSA and one with ECB.
Okay. Thank you.
Thank you.
Thank you. Now we are going to take our next question. The question comes from the line of Martin Ekstedt from Handelsbanken. Your line is open. Please ask your question.
Good morning. Can you hear me?
Yes, we can.
Excellent. Thank you. First one capital related question. How much of these temporary IRB related add-ons do you expect to reverse when the new models are approved? How much could remain to structurally higher risk rates? You mentioned probability of default parameter calibrations before on this call to historical default ratios. Could we view this gap as a rough indicator for where your new capital requirements might end up? Could we run the numbers on this in our models, for example? Thanks. That is the first one.
Thank you. If I start with that, effectively, once the models have been approved, we do expect there is going to be a net positive contribution to our capital position. As we said, the final outcome on REA will be subject to what those models ultimately produce. We expect it to be a net positive effect.
Okay. You mentioned the probability of default versus historical default ratios before, and there should be a gap there, right? Is this a good indicator of what is being calibrated now in the new models?
I think what Christoffer meant was that it was one example of a data issue we have been struggling with in terms of having a broader set of default data given our credit loss track record. You cannot from that directly infer any link to the PD output, just to be clear.
Okay. That is not the thing that is specifically leading to this recalibration now.
No.
Okay. Then moving on to a business question, if I may.
Yes, sure.
Yeah, sure. I wanted to ask about corporate lending and how much of the demand you are seeing now is driven from clients funding new investments and working capital versus just refinancing and temporary drawings on existing trading lines and the like.
Well, to put it broadly, I think we have seen during the third quarter, and in particular comparing to early in the year, healthy growth across a number of client segments in CIB. But if you look to the more traditional industrial companies, it is more of a refinancing activity. We have yet to see demand really pick up there when you think about larger CapEx investments or M&A activity. So it is growing in a number of segments. But more refinancing in the traditional corporate segment.
Okay. Excellent. Thank you. That's it from me.
Thank you.
Thank you. Now we're going to take our next question. The question comes line of Gulnara Saitkulova from Morgan Stanley. Your line is open, please ask your question.
Hi. Good morning. Thanks for taking my questions. Just a follow-up on the prior questions on corporate lending. Given the good momentum and the sizable regulatory capital headwinds, does this change how you think about the allocating capital and risk appetite across the corporate franchise? Are there particular sectors or geographies where you're looking to become more selective, or do you still see sufficient opportunities to grow at attractive risk-adjusted returns without changing your risk appetite?
To put it briefly, I think our focus remains unchanged. We are looking to grow profitably with our clients, and we have continuously an eye towards risk reward. We look forward to growing in a healthy manner, but also triangulating the three components that Christoffer mentioned are in the core. From our perspective, business as usual is the communication to staff. We are looking to grow in a good way.
Maybe just to add to that, Gulnara, this is a temporary add-on. We are building this business for the long term, so we will manage through this temporary add-on. We are not adjusting our franchise activities during the meantime.
Thank you. Can I also follow up on the retail lending in Sweden? In August, you mentioned you captured 19% share of the new lending. Could you remind us of your market share ambitions in Swedish retail lending and elaborate on the key initiatives that you are undertaking to gain market share in this highly competitive market?
Thank you. I think what we have said before is that we do not want to lose share in this market. I think it is satisfying to see the market share figures that we are currently observing, and I think they are a result of some diligent work with process changes. This is both more basic process redesign, but also deploying AI into the process. We have been becoming faster in how we respond to customers, and that is now showing in the numbers, which is very satisfying.
Thank you.
Now we are going to take our next question. Just give us a moment. The question comes line of Johan Ekblom from UBS. Your line is open. Please ask your question.
Thank you. Just to come back on the IRB models. I guess two questions from my side. Firstly, the IRB overhaul has been going on for a number of years. I think there were deficiencies identified back in 2023. How did we end up here in 2026? Because I think the broader question that was raised last week is, why were these deficiencies not sufficiently addressed? We have seen your two large cap Swedish peers, basically in the last two days announced that they have had their models approved. Can you talk a little bit about the governance around your models and your risk and why you think we ended up in this situation?
I guess secondly, when we think about the remediation or how we get back to normal order of operations, do you think at some point before you have the full new model approvals, you will be able to guide as to the ultimate capital impact, right? I fully take your point on board that the add-on is kind of the worst case outcome. When can we start to narrow the range of outcomes as we look through this temporary period?
Thank you, Johan. If I start with your first question and putting this in a little bit of a context. You are right, we had our first identified IRB deficiencies in the SREP process 2023. We, along with most other large peers in the region, got a Pillar 2 add-on at that point. In 2024, we then submitted a new set of models to remedy those deficiencies. In 2025, we had, first of all, more changes in supervisory expectations in the meantime, whilst also we had our internal modeling investigation in the Baltics. Concluding after that, and a little bit referencing Magnus' question as well, we saw that there were things in that feedback where we recognized that our new set of models would not be sufficient to meet with those changing regulatory requirements.
We withdrew our 2024 applications, and those are the ones that we are now planning to submit again in 2027. That is where we got there. Also worth mentioning, in 2024, we had our retail models approved and our mortgage models approved. That was done in 2024. But that is where we are and that is how that played out, Johan.
But it just begs the question, how did you interpret the regulatory expectations so differently from your peers? I realize it might be an impossible question to answer, but I think we are all scratching our heads. I think someone mentioned before, this is not what we would have expected from SEB, and I think that is what is leaving us all a bit puzzled. But yet, let us leave that there. But in terms of-
Well, maybe worth mentioning there, Johan, as well, I think is that the nature of our business with the complexity of a large corporate portfolio with multiple segments and multiple types of business could also be one of the reasons how various supervisor expectations are both interpreted and implemented.
Yeah. Then the second part of the question on the-
On your second question.
Yes. Sorry. Go ahead.
On the outcome. It is going to be something we will have to wait until we actually get the approvals on the model. What we are expecting now is to say that there will be a net positive outcome. How big that will be, it is going to be difficult to steer on early doors until we do have more clarity on the approvals. As soon as we do have more of that, of course we will be returning. But at this point, it is too early to give an indication.
Approval would, best case, be 2028?
Well, as I mentioned, we are-
If you submit Q3 next year.
If we submit in Q3 next year, and with a good and a constructive dialogue, and we get the feeling that the Swedish FSA will deal with this as soon as practically possible, as I mentioned. It is complex. It is not going to be a quick turnaround. But we sense that there is a good constructive approach from the FSA.
Thank you.
Thank you.
Thank you. Now we are going to take our next question. The question comes line of Shrey Srivastava from Citi. Your line is open, please ask your question.
Hi, and thank you for taking my questions. Two from me, please. We have mentioned that this issue to you seems to be one of regulatory compliance. But if I look not only at your corporate risk weights here, which if you include the add-on actually now screen in line with peers, but also in the Baltics, if I compare you with some peers, just using the 2025 Pillar 3. On some products where you would expect to have very similar risk profiles, mortgages for instance, your risk weights are often less than half that of peers. So I would just like to ask, what exactly is it that makes you so confident that this is going to be a positive on the corporate side going forward versus peers? What are you doing differently? My second question is on the treasury NII move, in the quarter, which was very significant.
You mentioned a little bit that's due to higher pre-funding volumes, and also higher funding costs. If we had to look at the evolution of this item going forward, how much of that can we expect to come back in the coming quarters and how much should remain? Thank you.
On your first question, when we look at the impact from the add-ons, those are temporary capital add-ons attributable to compliance matters in the construction of the models. As we see them and as the regulators see them, they are temporary, which means that once they are addressed, we're expecting the outcome of the underlying models to represent the risk weights without the add-ons. The reason for the risk weights underneath comparable to peers, that has to do with the nature of the exposures, the risk profile of the exposures, historical default rates, as we've discussed a little bit earlier on this call. It's the quality of the underlying counterparties that leads to the risk weights. Now, those models still need to be fully compliant, and until they are, that add-on is there. But the quality of the counterparties is what drives the underlying risk weights.
On your second question there, on the treasury, that's correct. We refer to the pre-funding and some of the funding costs in the quarter. Will this come back? Well, to some extent gradually we should see some improving trends from treasury from here on. Of course, subject to market movements, but gradually that's the reason to take back.
Okay. Thank you. Just a quick follow-up on the first question, please. You mentioned the quality of the counterparties. If I look at the Baltics, some of these much lower risk weights were on mortgages where I can't imagine that the counterparties would be that much different to peers. If you could just zoom in on that specific business and why the risk weights-
Yeah. It's hard to comment on others' risk weights. I think what we can conclude is that if we look at a number of the players in the Baltic markets, many of them are on the standardized approach from the get-go. So we are or have been on IRB, and now we have the equivalent of standardized pending approval on the new models. So I think in that comparison, hard for us to comment on how others arrive at the risk weights, but our risk weights now represent the equivalent of standardized. And prior to that, prior to the add-on, they reflected the outcome of our IRB models.
Shrey, it's Pawel here. I think when you look in the Pillar 3 reports, I think you have to remember that all the banks are currently in discussions with regulators on where REA and models will eventually end up. So looking at Pillar 3 numbers from, say, a year ago and trying to compare all of them and see who has the lowest or highest risk weights, it doesn't maybe fully capture the picture. But let's continue the discussion and please send over the data that you have, and I can have a look at it.
Brilliant. Thank you very much. Thank you.
Thank you. Now we're going to take our next question. The question comes from line of Sofie Peterzens from Goldman Sachs. Your line is open. Please ask your question.
Yeah. Hi, here is Sofie from Goldman Sachs. Thanks a lot for taking my question. Sorry to go back to these capital questions, but you paused the share buyback. Is it a pause? Should we expect the share buybacks then to eventually come back in 2028? Or is it fair to assume that the share buybacks are more like a cancellation and then you start receiving new share buybacks in 2028? Also, how should we think about the special dividend? I think our consensus has a special dividend of SEK 1.7 per share. Is there any risk to that? That would be my first question. Then my second question would be around the rate outlook in Sweden. What is your expectation? It seems that market views on interest rates are quite varying from one hike to three hikes.
What is your view and, can you just remind us on how we should think about the net interest income, kind of outlook and volume growth if we have under different rate scenarios in Sweden. Thank you.
Thank you, Sofie, and good morning. If I start with your first question on capital and capital distribution, I will come back a little bit to the earlier remarks. Our buffer range of 100 to 300 basis points, and when we operate towards the lower end expect us to add to the buffer. When we are at the higher end or when we are expecting to stay at the higher end, that is when we would take those kind of actions that you refer to. I think you can look at earlier in the year. Look at Q1, Q2, the buffer was somewhat below 300, but still at the upper end of the range. At those times we were undertaking buybacks. That is probably also the time when loan growth had not really started to take off yet.
Now we are at a slightly different point in terms of credit growth, but we will continue to triangulate where we are in the buffer with capital distribution, capital repatriation, and growth. Those will be the variables that will decide, Sofie, when the buybacks will be revisited.
Yeah. Good morning, Sofie. On your second question regarding the interest rate outlook, our economist, in their most recent forecast, are looking for one rate hike this year and two more rate hikes next year. We can, however, note that the market is pricing in overall slightly more, so we will have to see how that plays out.
As for our sensitivity then, Sofie, we've provided some numbers on the size of the equity, the zero cost deposits, and those other volumes. From there, you should be able to work out a little bit what you're expecting and what the deposit beta will be, and then you see the impact they have from there.
Okay. Sorry, just one more quick question on the capital. You mentioned that the retail models had been fully that we should not expect any headwinds from there.
That's right, improved and implemented.
Okay. That's very clear. Thank you very much.
Thank you, Sofie.
Thank you.
Thank you. Now we're going to take our next question. The question comes line of Riccardo Rovere from Mediobanca. Your line is open, please ask your question.
Thanks. Thanks a lot for taking mine. Just sorry to get back once again on the capital. Kristoffer, you mentioned some capital optimization measures at some point in the call. You also mentioned SRTs. This is something that you have been talking about for quite. Can you explain why is it taking so long to put these kind of measures on the ground? All banks in Europe are using SRTs. For you, it seems to me that it's taking very long time to put this. What could be the impact of these capital optimization measures? The other question I have is on the cost. It's not clear to me whether the new cost guidance of SEK 33.4 billion and then managing SEK 500 million into SEK 600 million, does this include the settlement you had with U.S. authorities, or we should consider that in a different way?
Good morning, and thank you. If I start, Riccardo, with your first question and on the SRTs, I think, as you know, this is not something we've done in the past, so we are starting with this. When we mentioned this on the call, that was really when we were starting to prepare ourselves. One of the reasons this has not really been in focus is also the regulatory treatment from capital perspective, where the Swedish FSA differs a little bit from the European regulator on this. Therefore, it has been a lower priority for us. Nonetheless, we've been gradually doing the work and preparing. What is it that takes time? It is getting more of the processes in place, the data collection, the reporting.
This is effectively setting up an infrastructure to operate and to run the processes and the products, and then the dialogue with the regulator as part of that. So we don't consider this an unusually long process. It's something that we've been working with, and I think we've suggested that towards the end of this year, we should be in a position to start undertaking transactions. On the capital impact from this, it's a little bit too early to say what that could be. But of course, when we are planning for capital going into next year, this will be one of the variables that will be taken into account. Do you want to add anything?
Yeah, just perhaps to add, Riccardo, I think there are also other tools in the toolbox. One such example is credit risk insurance, which we have used selectively on a small scale already. So we're looking into that. But please also remember that you give away some income when you do these type of actions. So it's also a risk-reward equation that we need to be mindful of.
On your costs, Riccardo, the new guidance effectively includes everything. It does include the charges attributable to the settlement as well.
Okay. That is very clear. Just a very quick follow-up. Christoffer, you said that you think you might be in the position to have some benefit out of these capital optimization measures with the end of the year. I am not sure I understood it correctly what you said before.
What we said, Riccardo, is that we are working on mitigants that may already give us some benefit in Q4, and that we will come back with, of course. Those are the ones that we are now revisiting and seeing what pace that we can take and what makes sense to do given the capital position and the growth in our business. Then through the normal course of business, we will continue to work with these measures going into next year.
All right. Okay. Thank you.
Then maybe, Riccardo, also a comment on the cost there. As usual, we will come back with a cost target for 2027 with our full year results. But I think just bearing in mind, of course, that now that we are giving ourselves a little bit more room to maneuver, this is also an opportunity for us to see which ones of our prioritized areas that we would like to ensure that we might be able to accelerate or generally just strengthening the franchise. This is something we will come back to give more details at the end of the year.
Thank you. Thanks. Very clear.
Thank you.
Thank you. Now we are going to take our next question. The question comes from the line of Ilija Novosselsky from BNP Paribas. Ilija, please ask your question.
Hello. Thank you for taking my questions. Just one kind of related to one of the questions asked previously. We have seen that corporate competition in Sweden and the region is quite high. You have had some of your peers across the other Nordic countries coming to take share in Sweden. How does the situation about the corporate exposures change your strategy? I assume that you are ready to compete no matter what happens, but are you going to be more selective now if your peers are more aggressive? How would you respond to that? Basically, I am asking how should your strategy change in an environment of very tough competition? Thank you.
I think, as you say, there is intense competition and it has been. Our strategy remains unchanged. We look to grow with our customers.
Stressing that this is a temporary add-on, so we are not changing our strategy as a result of that.
Of course, but there can be a short-term strategy, mid-term strategy, long-term strategy, and so on. Would it change which corporates you are targeting over the next one year or until the models are approved or not?
I think we are focused on the long term, on the building and nurturing our long-term relationships and supporting our clients.
I think it would be a little bit uncharacteristic for us to adapt in customer relationships briefly as a result of that. One could, of course, look at the broader direction of travel in Europe around harmonization of risk weights and capital output floors being phased in. Will that impact pricing more generally in markets over time? But right now, this is a business that is more set by market conditions than where we are competing on those fronts.
I think operating with that, I think that will have to be our last question that we answer, as I think we are running out of time, unfortunately.
I think Swedbank's call starts in 10 minutes, so we will give you some time for that. If you have any further questions, please reach out to us at the IR team, and we will try to answer any outstanding questions.
Thank you very much for dialing in, and thank you for your questions.
Thank you.
That does conclude our conference for today. Thank you for participating. Have a nice day.