DBS Group Holdings Ltd (D05.SI) Q3 FY2025 Earnings Call Transcript & Summary

Hi, everyone. So welcome to the 3Q results call. You've heard the media, so we'll go straight to Q&A.

[Operator Instructions] So first question from Jayden from Macquarie.

Can you hear me okay?

Yes.

Great. I appreciate it. A couple of questions. MediaCore, you were fairly upbeat, Su Shan, I think on the outlook for demand for loans. Just any thoughts on sort of what's changed versus prior quarters? Any commentary you can give us in terms of industry or geography. My second question is on the NIM and the sensitivity. Can you provide us with any sort of updates as last quarter, you gave us very helpful sensitivity on how the NIM would perform with respect to various base rates. Those are the 2 questions I wanted to ask.

Okay. So let me walk you through the non-trade loan growth expectations. Like this year, we're looking at nontrade loans to grow in the mid-single digits next year. And structural growth, you see it in tech data centers, et cetera. So TNT will remain strong. Real estate, there's quite a lot of government-led sales still in Singapore. There's quite a lot of big government projects as well here. And so the public sector and I guess, real estate in selected cities, Singapore, possibly Australia, we'll see some good momentum. And alongside -- I talked about TNT alongside the whole GenAI adoption as well, we're seeing some structural growth there. In energy, although there seem to be, at least from the headlines and roll back in renewables, actually, we are seeing in Asia, still a lot of due flow M&A and transactions in the renewable space. And as rates come down, I think the LBO market will come back, and that should also see some growth there. In food and agri as well, you're looking at sort of inventory financing for some of the big guys as well. Housing loans, mortgages, we also had a -- we had a good Q3 this quarter on new launches. So that should filter in next year. And also, I think for wealth, as rates come down, there should also be some of these wealth loan growth as well. Do you want to add anything Kwee Juan. Kwee Juan is our Head of Corporate institutional banking.

No, I think it's as in net area. The TMT [indiscernible] looking at pipeline.

Okay. Then you want to talk about the NIM sensitivity, right? So we have net floating assets of about 11 -- sorry, 110 billion. So if rates come down by every 1 basis point, we will lose $11 billion there. And then we have $160 billion of net floating assets. Sorry, the other way around. We have $110 billion of net Oh, sorry, sorry, sorry. We have $160 billion of net floating assets. So for every 1 basis point drop, we will lose $16 million. This is for next year. And we have about 50 billion net floating liabilities. So for every 1 basis point, we will lose 5 million. So plus 16 million, so it's minus $16 million, plus $5 million.

Okay. Thanks very much, Su Shan. Appreciate it.

Yes. Sorry for this year, I didn't tell you for this year, right? So this year, it's $120 billion net floating assets and $50 billion net floating liabilities. So it's plus $12 billion minus $5 million.

Next question is from Yong Hong from Citi.

Yes. I have 3 questions. First question is on deposits, strong growth quarter-on-quarter and year-on-year. Just wondering where the source of money is coming from. And if they are corporate or are the consumer monies or these are -- both of these are so driver? And a follow up to that is given that we are in a position of excess liquidity, what are thoughts of managing down cost of funding more aggressively maybe even lower than the 1% sing dollar FT that we are seeing. I have 2 more questions after this.

Okay. So the deposit growth was actually point wide. We had -- from the retail side, quite a lot also because the treasury bills matured and they came back. We had SME deposit loan growth. We had wealth loan growth. We had corporates as well with a fair amount of -- so it's quite widely spread and across foreign currencies.

And maybe...

It was also helped, I guess, in Singapore, we have the CDC vouchers, right? So people were spending a little bit less in Singapore. So that also mean some more sing dollar CASA retention as well.

Yes. And given the position of excess equity, any thoughts of managing down your cost of funding more aggressively maybe even lower than the 1 percentage sing dollar FD base that we are seeing right now.

Yes. This is Phil here, Corporate Treasurer. So if you look at our deposit margins, they are basically above 1%. So we don't have any need to cut back on deposits. In fact, this is a very good deployment for us. As all the wealth deposits come in, the GTS deposits come in we can put it to work and we can make a very good margin on that.

Yes. Because what the treasurer -- we are always open for deposits, right? The fact is philosophy, you must always bring in deposits. And we can always reinvest in HQLA is a good high ROE, decent NIM spirit.

Yes. The context of my question is given the deposits grow is so strong, I don't think they are coming in for that 1% kind of deposits a bit. So based on your deposits coming in do you see any room to manage down your cost of funds, especially if you think that the deposits coming in are not going after the rates?

No, we actually make sure that we always have a positive deployment margin on all our deposits. So if you look at our FD rates versus MES bills, for example, we would always make a spread on that. Now the NIM might get diluted because the marginal deposits versus the marginal deployment will be less than our current average NIM. So you get a NIM dilution, but it's ROE accretive, and it's NI accretive.

Yes, yes. No, I understand.

Deposit cost is already down 20 bps.

Okay. Okay. Got it. And maybe just moving on to AUM. Any breakdown on the net new money and the market impact? And any color on where this net new money is coming from our sustainable that is? And whether there is a result of natural flows into Singapore? Or is there something that we have done in recent years that is using results?

Okay. So when we talked about the net new money, that was only at the high end, but we should include -- if you include treasures and above, then the growth was actually stronger. So for the total wealth stack, which is treasures TPC and PB, AUM grew 7% quarter-on-quarter, 18% year-on-year to $474 million. That's really across the board. As we said earlier on, we have wealth centers in pretty much most of our core markets, right? So Singapore, Hong Kong being the 2 big onshore offshore hub. China, we have a new World Hub as well. India, where we're also trying to focus now more well. Taiwan, after we bought Citi, it's a strong mature wealth business in Taiwan. And Indonesia onshore wealth also is growing. So it's across the board. But Hong Kong is seeing very good flows from Greater China, Hong Kong area. And China onshore is also seeing very good flows because, I think we were not involved in the whole trust debacle over the last few years. So we have a good reputation there. We talked about the need for insurance and life planning by the Chinese, Mainland Chinese. We talk about Stock Connect, we're able to give them good wealth management products or solutions, SIPs, et cetera. So that's good organic growth there. Tse Koon is here, he can weigh in.

Yes. I think just aligned with what Su Shan has said, generally, the wealth growth has been very robust and very well spread, right? So first thing, we benefited from the fact that -- and we continue to benefit from the fact that we have a full wealth continuum. In other words, we are able to bank customers whether from their first 1,000 to their first 100,000 the first million, even to those with billion with us. So I think we are 1 of the few that had the full continuum of wealth from private bank, ultra-high net worth to those kind of almost emerging to private banks right into the affluent, even right into the retail wealth. We have also been able to connect internationally very well. So customers who use us to manage their wealth, both onshore and offshore, right? And our wealth management customers come from more than 120 nationalities. So being anchored in Asia with 2 booking centers in Singapore and Hong Kong, that's been very, very strong. Our full suite of products and services, some of which Su Shan has alluded to, is very, very comprehensive. So these have actually been a key contributor to how we have been able to deepen and broaden our wealth franchise.

And my final question is on dividends. You $0.15 extra quarterly dividend per year. I think based on the latest guidance still expire by year-end. Any guidance, is that a similar quantum would be...

Sorry, that's not correct. So who is that.

So, Yong, you're referring to $0.15, the capital return dividend to the capital return distribution we talked about lasting for 3 years until the end of 2027. So $0.15 that amount is third quarter goes on until the end of 2027.

Okay. Okay. I think we basically ordinary or a similar quantum.

So we said, $8 billion. $3 billion for share buyback $5 billion for 3 years of the $0.15 step-up. So $0.15 times $0.04 $0.60, $0.60 time 2.84% is 1.7%, right? So 1.7 3 years is $5.1 billion. So that $5 billion goes is return over 3 years.

Dividend is a different category from capital return dividends. So they'll be confused.

Yes. Okay. Now I was just wondering whether this $0.15 could be returned in a different form, but I think that is clear from Susan. And on buyback, if the share price continues to be quite strong, can we be flexible to allocate the capital for buyback to do something else? Or there should be a reserve for a certain period of market downturn.

I think we always assess the opportunity, and we said we'll do over 2 to 3 years. So if by the end of 3 years, we have not fully bought back the shares, we could always think of other ways to return the capital.

Okay. Got it. And you -- and the general mechanism is that should not exit the trailing 30-day pre price or something like of the buyback. I don't see any general or there's no orders not disclosed before?

No. We just said we'd be opportunistic.

Next question from Harsh from JPMorgan.

I have more questions on the payout and buyback. So I see the trade-off between a program buyback that every day, irrespective of price, at VWAP or something, you do consistent buyback versus trying to be opportunistic. So what has been the discussions? And how do you see -- and then what I'm trying to get to is, is there a possibility to move to a program buyback? How do we stay with the current stance?

So we are not in the camp of doing program buyback every day, whatever the price, but we do have an alignment on how we'll go about doing it. We look at standard deviation of the price, and it's a bit higher, more than 1 standard deviation over a period then we'll do less. If it's like much lower, we'll do a lot more. So I think we calibrate it that way internally, and that's how we have been running this program.

Sorry, I got muted. Yes, so thanks for that. But then the problem with doing a buyback in an opportunistic basis, is the daily trading volumes during down days, can the buyback amount be half of the daily trading volumes. Otherwise...

Also of an internal rule that we in general, will not exceed 10% of the daily turnover on the exchange.

Exactly, exactly. If that's the case, then if the weakness lasts for, let's say, 3, 4, 5 days, then you can't use up this entire amount if you are just opportunistic.

We have 3 years, right?

Yes. But -- it is -- again, if I look at the YTD performance and the stock continues to do well, which we all hope it does, then that -- then this question on can you really use that buyback? So either you have to win that rule of 10%, that during down days, you can go as high as 30%, 40%, 50% of volumes, if need be, or you need to do program buyback. Otherwise, this buyback amount becomes a bit question mark.

We'll take that back. I think we have been discussing this at the level. And this whole [indiscernible] program buying every investment bank wants to do that because our stock is liquid, right, has what they behind the right? So I think it's just a bit more prudent that we have -- we keep the option to exercise more discretion on coming in on bad days because the market will be volatile next year, right? And we want to be able to have that flexibility.

And I understand that, Su Shan.

We'll take it away.

The risk if things become -- remain too good. And second question on NIM. Thanks for a very detailed guidance on your assumptions on rate cuts, fee rate cuts, flats or sing dollar appreciation makes sense. But let's say, if out of blue, if we end up getting, let's say, USD rally in that case, some of these assumptions on the SORA pass-through and sing dollar appreciation may not materialize. So how are you looking at adjusting your hedging strategy in that scenario? And what are the various pros and cons if that ends up happening?

So your question is, just so I get it right, the Sing does not appreciate, it depreciates instead of appreciates, right?

Yes. And that may be just because dollar goes up, right? Because this year dollar was down massively. Yes.

Okay. Okay. And U.S. rates.

Still same, let's say, rate cuts because the dollar depreciation this year was largely due to risk premia increase, right? And for whatever reason, the risk premium dollar decreases despite the rate cut, we end up getting dollar value, which we are starting to do dollar is up 3%, 4% from the lows. So that's the question that in case of an outcome, which is different from your base case assumption, how do we think about impact on NIM? And how is the bank preparing for that eventuality?

We think it's always good for us. When we report because we have a fair amount of U.S. dollar come and Hong Kong dollar income and losing income. So honestly, a weaker thing is not a bad thing for us at all that we like it. And so you're thinking in the past, when you have a weak Sing, you actually have higher Sing rate cost of the forwards, right? That seems to have broken down now, Harsh. So we've taken to checking the MES bills as a guidance on Sing rates because SORA overnight server is also too volatile, very hard to predict. We're up and down 40% sometimes overnight. So it's very hard to use that as a gauge of the real cost of money in Singapore. So we use the MAS build instead. And if you think the Sing dollar depreciates, I don't think Singles can collapse, right? So it should, if anything, go up from here. So then if you have higher sing dollar rates and lower Sing that's both a double tailwind for us.

Right. So net-net.

I hope, that comes true.

That 100% -- so that's on our -- on our commercial book, you'll make more money maker than Sorokin of stay steady and sing dollar appreciates or translation gains get. And -- but then is there any risk on the hedging book that we should worry about it is basically wash, either direction.

There shouldn't be any risk on our hedging book because we put it as a dynamic book, right? We put it on every things rolled off, things get put on. We haven't really been able to put on quite a lot yet because the yield curve is not in our favor. So it's a dynamic hedge. So we'll see next year, if what you say come we might catch more. And I think we've shifted a lot of the hedging more to U.S. dollars. -- because Sing dollars are not attractive to hedge at these levels. So as they roll off, our hedges have been more in the U.S. dollar category.

Right. And sorry, the last question on this is the hedges rolling over. So basically, you have a very good track record of being able to replenish those hedges which are rolling over. So we should assume something similar going forward? Is that what's baked in your guidance for NIM and NII next year that the bank will be able to catch these sales quite well over the next 3, 4 quarters? Or the -- or have you lengthened the hedge duration so that you have visibility.

Okay. So Harsh, next year, we have $78 billion of fixed rate asset maturities, the maturity, the maturing rate was 3.3%. And we are looking to do 2/3 of that to be replaced.

Got it. And Yes, if I may just 1 last question just on Wealth Management. Net new money growth this year, if you could give some granular details on which countries residence, is it more North Asia, is it more South Asia, any particular market that will be very useful?

Thanks for the question. This is Koon here. Our net new money has been very, very strong. I think it has been for the last couple of years. And actually, as we dive deeper in, it's actually very, very broad-based. It's across various markets. So just as an indication, the private banking customer base comes from more than 100 nationalities, right? So we have 2 booking centers in Hong Kong and in Singapore, we're agnostic. It depends on where the customers choose to book. Some of them have bookings in both, but they are very, very broad-based, right? We have customers from Southeast Asia, of course, being here in this part of the world. We have customers from Northeast Asia. We have customers from South Asia. We have customers from the Middle East, we have customers from Europe, both Western and Eastern Europe. So very, very broad based.

Not this year, has there been any particular market which has seen a faster pickup in net new money growth compared to, let's say, last year?

No. Not -- I can't single out any market in particular. And at the same time, just also to elaborate a little bit further. Our net new money, the growth has been very, very robust through the entire wealth continuum, that has got both the private banking clients, the high net worth and ultra high net worth as well as in our segment we call Treasures, which is the affluent customer base. And it's the same trend we've seen is very, very broad-based.

Next question from Aakash from UBS.

Can you hear me now?

Yes.

Sorry, was there -- this is Aakash from UBS. The first question is just on the NIM decline this quarter. I wanted to understand a bit better because in July, the exit NIM was $1.95 billion and so which was already down 10 basis points from the previous quarter average level, right? And then after July, we had 4% deposit growth, which were parked into HQLA, which are NIM dilutive. And loan growth was much smaller. So a majority of it went to HQLA. And then we also had further declines sort of overnight rate declines in August and September. How did none of these -- any of these affect the NIM? How did NIM stay at 1.96% for the whole quarter? That's the first question.

Okay. Because -- so you're looking at the quarter, why did it stay so flat when Sora overnight went down so much, right?

Yes. And SQLA as well, Su Shan, because SQLA would be NIM dilutive.

Okay. So 2 main reasons. The first is the group NIM declined less than the commercial open because the market's trading book because of the lower accounting asymmetry, our cost of funds also was lower. There was some pushing back from that. And there was also a rebound in HIBOR as well. So that also helped the overall NIM for Q3. And then also IRS and fixed rate asset deployment as well, which was about 1/3 of the commercial book assets. Do you want to add anything, Sok Hui.

I think you're right. Your observation is right. So I think with the commercial book, we gave you guidance last run, right, in July, it will be about 1.95 and you have been able to hold it, and that's partly reflecting that market trading has also helped to cushion the pressure.

Okay. Okay. Got it. Then the second 1 is, I also wanted to understand the hedges a little bit better because I think your NII guidance for next year, which is slightly down if -- I think people expect that if hedges are going to roll off next year, this guidance does look a bit more optimistic. So to have a high level of confidence. I think if you can share more details on the hedging book diarisaps that you have, right? -- because we only see annual numbers. So we know it was $59 billion at the end of 2024. What is that number today? And what is the breakup between the different tenures and how much of it is expiring next year? If you can share that information, I think that will be very helpful.

Okay. Aakash this is Phil, Corporate Treasurer. So I'll just give you a couple of data points. So we've got about we don't look at IRS and bonds separately. When we talk about the hedges, we're talking about both the funded and the unfunded hedges collectively. So when we've given you that $200 billion of assets -- fixed rate assets, that's the totality across bonds and IRS, right? So that number is about there still. So of that, about $78 billion rolls off next year. And we actually have assumed, and it's consistent with assumptions Su Shan just gave that we're probably going to redeploy that at about 50 basis points lower than the maturing yield that those particular assets have. So it's built into the forecast. -- and it's all consistent with all the guidance that we've Su Shan has rearticulated.

So This INR 78 billion is primarily the fixed rate mortgage book? Or is it the IRS swap?

This is a mixture. It is the IRS, it's got the mortgages because remember, mortgages in Singapore are not long-dated unlike say in the U.S., right? In the U.S., you take long-dated's mortgages and you sit on it. In Singapore, typically, it's 2 years. So the average weighted life of a mortgage is actually owning 1 year. If you think about it, right? So the mortgage book does roll off. The IRS do roll off, but the numbers I've given you incorporate all those effects roll together.

So you should think about the NII as having a few components, right? Floating rates will also come off. Fixed rate hedges will also come off, but we will also benefit from the deposit volumes that we have told you about all the deposits that come in, we can deploy at slightly more than 1%, and that's going to cushion the effects of the lower rates. And I think if you look at the treasury book, I think the funding cost will go down quite a bit with the rate cuts. So that's another relief from the headwinds in the commercial book. So net-net, as a package when we say slightly down, we're not saying like a little -- I mean it's like high something like $0.3 billion, $0.4 billion down that kind of number.

Okay. Understood for explaining that. The third question is just on broadly liquidity. And Su Shan, how do you interview with Nicolas on the podcast yesterday, very interesting, very inspiring. I think 1 of the comments that you made was liquidity in general has helped financial assets everywhere. -- right? Now some of that liquidity also found its way to Singapore and must have helped with the wealth management net new money that DBS has seen. I just wanted to get your thoughts on how do you think about this inflow? What part of it is cyclical? What part of it is structural, would you extrapolate this net new money strength into the coming years? Or do you think maybe you can normalize this pace of net new money that you didn't see.

Well, I look at M2 and I look at the market cap of different markets, and I try to find where when the money flows might go next. And it's entering because Hong Kong, China has been very strong, performing indices this year, but a lot of the U.S. flows has stayed in the U.S. and the money that the wealth creation has stayed primarily in the U.S., right? When you think about the U.S., it's now 70% of the world's global capital markets is very concentrated. So the wealth creation and the wealth effect in the U.S. is very real. Money supply and the banking deregulation in the U.S. is going to be helpful. And so yes, I see liquidity remaining strong. Do I see it coming to Asia? I see liquidity certainly in China and Hong Kong because China rates are so low. No one's buying real estate. So the money is going to go somewhere. And that's why you see very strong growth in things like banker insurance, life policies, -- and as people get more and more of the seated as well. And people are starting younger, right? You have the Robinhood phenomenon where younger people in the 20s, early 30s, starting to invest as well. which is why we're building this digital wealth for retail as an offshoot of our current more mature wealth business. The team has rolled out digital wealth products online for precisely to meet the needs of the young 20-something year old millennial or Gen Z investor. So I see North Asia liquidity remain pretty robust. Singapore depends whether we have bought sing dollar assets to absorb the -- there's some government land sales, there's some government projects there's new launches of property, the stock market is getting better. So let's see. I hope the liquidity in Sing dollars gets recycled. I think people are trying to build a proven asset ecosystem in Asia as well. There's been private equity deals, certainly in Japan and some nascent venture and all that in China because of the AI and humanoid robot pool that we're seeing. So we see some interesting flows. That's a bit of the formal flow now back into China because the market has done so well. So I see that happening. But I don't know. Like every other of my peers, I do worry about the high valuations in the U.S. on these AI companies we're very cognizant of not being caught up too caught up in that. But whilst we invest in AI ourselves, we're also cautious about the high valuations that we see. Yes, I think North Asia, the equity remained good. South Southeast Asia, yes, large corporates are doing okay. The midsized SMEs are doing less okay, and people are saving more on spending less -- but if they can recycle those assets, as I said, into the stock markets, into reads into more rail assets, then hopefully, that sort of domino effect that the velocity of money picks up. The progress the velocity of money hasn't been very high in Asia, that's the problem.

We move on to Melissa from Goldman.

Just back on this that you're talking about liquidity and North Asia being a better area in terms of -- do you think then in terms of DBS, the net new money coming in and also your deposit growth that you have been seeing can still be sustained from this year to the next? Also, you mentioned...

Melissa, we're now #4 in Hong Kong, yes. We're the fourth largest bank in Hong Kong. So we've been doing very well in Hong Kong and well, wealth centers in Queens Road, Central and all that are doing well, and we are also growing in China onshore wealth as well. So as long as it's any of our 6 core Asian markets, we're happy. We're not Singapore only.

Okay. And then just to clarify. Earlier, you said in terms of taking deposits and putting in securities, you get roughly about 1% in terms of NIM or that's at least a base point at which you ideally do this kind of gapping scenario. Would you say the ROEs on this, it's roughly about 20% to 30% in doing this and versus like your book itself is about 17%.

But Treasurer tells me that ROE is 50%.

50%.

High credit.

The risk rates are pretty low.

Government securities.

Government securities. So if you're collecting deposits and deploying it, all you get a small charge.

Okay, okay. So most of these that you do will mainly be got securities and then for your other books, then you will kind of have a spread, right? Because if you look at your security book, there is a good spread between govies and corporate bonds as well.

Corporate bonds are done by -- the corporate treasury team doesn't take views on corporate bonds. So the corporate bonds are taken in the dealing room. On the go, they can respond.

Okay. That's good. Then I have another question on tech and data center exposure that you were saying that you've been growing. If we look at by geography others, it's now quite high, almost 19%, 20% of corporate loans. And it's been like only 16% in 2022 on the other category.

Data centers is about $8 billion or $8.6 billion.

Okay. No, I was talking about just in geography wise, the other geographies, like you are seeing Hong Kong, Greater China, Southeast Asia and others, and the other category has been growing quite fast, as well.

So I think that -- So as you know, the phenomenon is redriven the institutional bank -- we do see the data center phenomenon being 1 whereby it is growing in Southeast Asia, as you saw it, just not Singapore, we're seeing some in -- we're doing some in Thailand. We did some in Australia. Also U.S., we are participating in the growth in the US lending house. As you know, the U.S. market has got quite a big development there. So the others would be on

Right. I was just wondering.

We do these very structure. It's a healthy structure, 15-year hyperscalers only. No build on spec.

There are triple net type leases. So all variable costs are passed on to the tail.

Right, right. Okay. I just wanted to get that because in terms of that you've been growing, right, in terms of -- it's the biggest hype now, but what if we're down 5 years later in terms of asset quality, are we quite confident that it will be still okay that in terms of pulling this exposure prices?

It's really the underlying offtaker that because the -- we believe that in the AI space itself, your big tech companies are the ones for investing and are providing services because they have an ecosystem. Customers take off their AI solutions. And they do need a fair bit of data center, which is why you're seeing so many deals being cut all the hyperscalers. And these data centers that we are lending to are actually supplying to these hyperscalers where variable costs are counter. So we're quite confident of that particular piece of portfolio. We don't go up. We do the other 1 that's a GPU as a service by the customer base could be any corporates out there but looking for an AI use case. Those are not the ones to me in. So therefore, from a risk perspective, it's pretty much in line with what most of the TMT analyst days on who in AI winners. It's really those who already had the boat today in deploying AI execution, like to Microsoft and what...

Okay. Just lastly, I know it's too far away 2027. You've mentioned before that 2026 in terms of dividends stepped up to $0.24, you're comfortable with that for next year. But in terms of 2027, now that we're a bit nearer there, do we have confidence to say that we can do it? Or we still have to wait and see?

I think you have to wait and see. Give us some time.

Last question from Wee Kuang from CGS.

Yes. Okay. I just wanted to ask, I think moving forward going to 2026 in your guidance outlook, you mentioned that there is a possibility of write-back in GPs. I just wanted to understand what is the guiding print when you look at writing back GPs. How much are you -- how much do you allow itself to go below in terms of, I guess, management overlay or whatever the guiding principles that you may have?

I think it will to calibrate it based on the external environment at time. Remember, I told you overlays are also based on stress scenarios. We have been very prudent. We built up $1.8 billion during the COVID period, and we have actually not released that. So I think we do assess the external environment, and we'll take a call. Whatever is you'll kind of -- we think we have more than adequate buffers for us to be able to release some if the external environment is okay.

Do you consider the current environment is stable? Because I'm just thinking out loud that I mean it has built up, even though, I guess, if you just look at it retrospectively, things have kind of been relatively stable, yes. So there is kind of enough that you have on your balance sheet to write back some.

No, it's safely better than April 2, after liberation day, we took -- remember, we took to an liberation day we for the tariffs, and we haven't released the COVID GP buildup yet. It depends. I think lower rates are benign for borrowers and for real estate and for most corporate borrowers for retail borrowers as well. where we still see some headwinds are in the consumer unsecured loan side. We're looking out for things like unemployment rates, et cetera. Also SME they're not recovering yet. The larger corporates, the financial institutions, the high net worth families are okay, they're doing very well. So it's a bit of a bifurcated recovery, if you will, right, unfortunately. So we remain quite prudent in our asset book. We we tend to be -- we have target markets and we use a lot of data and cash flow analysis to make sure that our clients are okay. And I think we will stick to that discipline.

Just 1 follow-up question from Melissa.

Just 1 follow-up. What is your exit NIM that you had this quarter? And also in terms of the HIBOR, has everything been factored in? And -- or should we see perhaps some minor lift from HIBOR moving up next quarter?

The exit NIM in September was 1.95 mills. October was 1.92 because of SORA.

And next, will we see some benefit from FIBO next quarter?

Unlikely.

That's all the questions we have. Thanks, everyone. We'll speak to you next quarter.