DBSDF - DBS Group Holdings Ltd (DBSDF) Q3 2023 Earnings Call Transcript
2023-11-06 10:40:26 ET
DBS Group Holdings Ltd (DBSDF)
Q3 2023 Results Conference Call
November 05, 2023 10:30 PM ET
Company Participants
Edna Koh - Group Strategic Marketing & Communications
Chng Sok Hui - Chief Financial Officer
Piyush Gupta - Chief Executive Officer
Conference Call Participants
Chanyaporn Chanjaroen - Bloomberg
Anshuman Daga - Breakingviews
Goola Warden - The Edge
Presentation
Edna Koh
Good morning, everyone, and welcome to DBS' Third Quarter Financial Results Briefing. This morning, we announced 3Q net profit rose 18% to $2.63 billion and 9-month net profit increased 35% to a new high of $7.89 billion. To tell us more, we have our CEO, Piyush Gupta; and our CFO, Chng Sok Hui. So without further ado, Sok Hui, please. Thank you.
Chng Sok Hui
Good morning, everyone. We achieved record total income in the third quarter and new highs for 9-month net profit and ROE. For the third quarter, net profit increased 18% from a year ago to $2.63 billion, while total income rose 16% to a new high of $5.19 billion. Growth in the commercial book was broad-based. Net interest margin expanded 52 basis points from higher rates, while net fee income grew 9% year-on-year from an increase in wealth management, cards and loan-related fees. Treasury customer sales and other income were up 8%. The growth in the commercial book was moderated by a 38% decline in Treasury Markets income due to higher funding cost. For the 9 months, net profit grew 35% to a new high of $7.89 billion with ROE at a record 18.6%. Total income rose 27% to $15.2 billion as higher commercial book net interest margin and noninterest income was moderated by lower Treasury Markets income. The cost-to-income ratio improved 4 percentage points to 39%.
Asset quality remained healthy with the NPL ratio a little changed from the previous quarter at 1.2%. Specific allowances rose to 18 basis points as allowances were prudently taken for exposures linked to a recent money laundering case in Singapore. Allowance coverage was high at 125% and 216% after considering collateral. Capital remained healthy with CET-1 unchanged from the previous quarter at 14.1%. In August, Citi Taiwan was consolidated, making DBS the largest foreign bank by assets in Taiwan with leading positions in deposits, cards and investments. The consolidation also added $10 billion to loans, $12 billion to deposits, 2.7 million in credit card accounts and over $8 billion to investment assets under management. For the third quarter, the Board declared a dividend of $0.48.
Slide 3. Compared to a year ago, third quarter commercial book total income increased 19% to $5.03 billion. Net interest income rose 23% or $695 million to $3.68 billion as net interest margin expanded 52 basis points to 2.82% from higher interest rates. Noninterest income growth was sustained. Fees grew 9% or $72 million to $843 million, while other noninterest income was 8% or $37 million higher at $499 million. The increase in the commercial book was partially offset by 38% or $103 million decline in Treasury Markets income to $166 million due to higher funding cost. Expenses rose 12% or $213 million to $2.04 billion from higher staff costs and the consolidation of Citi Taiwan. Underlying expenses were 10% higher ex Citi. The cost-to-income ratio was little changed at 39%. Specific allowances rose from a low base to $197 million or 18 basis points of loans. General allowances of $18 million were taken compared to $153 million taken a year ago.
Slide 4. Compared to the previous quarter, commercial book net interest income rose 3% or $103 million from an increase in commercial book interest-bearing assets and a 1 basis point expansion in net interest margin. Loans grew 1% or $5 billion in constant currency terms to $420 billion due to Citi Taiwan, while underlying loans were 1% lower. Deposits grew 2% to $531 billion from Citi Taiwan, while underlying deposits were unchanged. Noninterest income growth was broad-based. Net fee income was up 2% or $20 million, while other noninterest income rose 8% or $35 million from an increase in treasury customer sales. Expenses rose 6% or $107 million. Excluding Citi Taiwan, underlying expenses grew 4% from higher staff costs. Specific allowance rose $83 million. General allowances of $18 million were taken compared to a write-back of $42 million in the previous quarter.
Slide 5. For the 9 months, net profit rose 35% to a new high of $7.89 billion with ROE at a record 18.6%. Commercial book total income rose 33% to $14.6 billion. Net interest income rose 46% or $3.34 billion to $10.6 billion from an 83 basis point expansion in net interest margin. Net fee income rose 4% or $87 million to $2.52 billion as growth in the second and third quarters more than offset a decline in the first quarter. Other noninterest income rose 16% or $197 million to $1.4 billion as treasury customer sales rose to a record. The strong commercial book performance was partially offset by 37% or $358 million decline in Treasury Markets income to $612 million due to higher funding costs. This view of Treasury Markets, which combines net interest income and noninterest income, is the most relevant measure of Treasury Markets performance as there are offsets between net interest income and noninterest income lines due to accounting asymmetry, for example, in equity and FX swap products.
Expenses rose 14% or $724 million to $5.85 billion due to higher staff costs. With total income growing 27%, there was a positive draw of 13 percentage points, which resulted in cost-to-income ratio improving by 4 percentage points to 39%. Specific allowances rose $112 million to $373 million or 11 basis points of loans, while $75 million of general allowances were taken compared to $18 million a year ago.
Slide 6. Commercial book net interest income rose 3% to $3.68 billion compared to the previous quarter as net interest margin was 1 basis point higher and commercial book interest-bearing assets grew 1% or $7 billion due to Citi Taiwan. Compared to a year ago, commercial book net interest income rose 23%, and net interest margin expanded 52 basis points to 2.82%. Combining the commercial book and Treasury Markets, the group net interest income rose 2% from the previous quarter. Net interest margin expanded by a faster 3 basis points to 2.19%. Half of the expansion was driven by the commercial book. The other half was due to a decline in Treasury Markets assets, which reduced the drag on group net interest margin. Compared to a year ago, net interest income rose 16% as net interest margin expanded 29 basis points.
For the 9 months, commercial book net interest income increased 46% to $10.6 billion from an 83 basis point improvement in net interest margin to 2.77%. Group net interest income was 33% higher at $10.2 billion as net interest margin rose 51 basis points to 2.16%. Treasury Markets net interest income was negative for the third quarter and 9 months. As mentioned earlier, this has to be viewed together with noninterest income to provide a relevant measure on Treasury Markets performance.
Slide 7. During the quarter, loans grew 1% or $5 billion in constant currency terms from the consolidation of Citi Taiwan, which added $10 billion of loans. Trade loans contracted due to unattractive pricing, while nontrade corporate loans were lower due to higher repayments. Consumer loans were also lower by $1 billion. Over the 9 months, loans declined $5 billion, excluding Citi Taiwan, as growth in nontrade corporate loans was more than offset by a contraction in trade loans.
Slide 8. During the quarter, deposits grew 2% or $12 billion in constant currency terms from the consolidation of Citi Taiwan. Excluding Citi Taiwan, underlying deposits were unchanged as CASA outflows were replaced by fixed deposits. Liquidity was ample with LCR of 138% and NSFR of 117%, well above regulatory requirements.
Slide 9. Third quarter gross fee income of $1.05 billion was higher than a year ago and the previous quarter. Wealth management fees increased 22% from a year ago to $393 million from higher bancassurance and investment product sales. Card fees grew 21% to $269 million from higher spending as well as the integration of Citi Taiwan. Loan-related fees rose 12% to $137 million. Transaction fees were a little changed at $228 million, while investment banking fees fell 16% to $21 million on slower capital market activities. For the 9 months, gross fee income of $3.06 billion was higher due to a growth in cards, wealth management and loan-related fees.
Slide 10. We have introduced a new slide, which provides a clearer view of group noninterest income. The commercial book noninterest income accounts for the majority of group noninterest income and is highlighted in the red box. It comprises net fee income and other noninterest income, which are customer driven. For the third quarter, commercial book noninterest income rose 9% from a year ago and 4% from the previous quarter to $1.34 billion from sustained recovery in net fee income and growth in treasury customer sales. For the 9 months, commercial book noninterest income increased 8% to $3.91 billion.
Treasury Markets noninterest income is highlighted by the black box and for the 9 months is up 69% from a year ago to $1.05 billion. As indicated earlier, the strong increase from a year ago in Treasury Markets reflects the offset against the decline in net interest income from accounting asymmetry. So the best way to look at the bank's performance is through the lens of the commercial book, and then for the Treasury Markets income, to look at net interest income and noninterest income in aggregate. Commercial book noninterest income accounted for about 80% of total noninterest income in the third quarter and the 9 months period.
Slide 11. For the third quarter, expenses grew 6% to $2.04 billion from the previous quarter driven by higher staff costs. Underlying expenses, excluding Citi Taiwan, grew 4%. Compared to the previous year, expenses were 12% higher, while underlying expenses grew 10%. Cost-to-income ratio was little changed at 39%. For the 9 months, expenses grew 14% due to higher staff costs, while the cost-to-income ratio improved 4 percentage points to 39%.
Slide 12. Nonperforming assets rose 6% or $313 million from the previous quarter to $5.30 billion. The increase was due fully to the integration of Citi Taiwan. Underlying nonperforming assets were little changed with new nonperforming asset formation in line with the recent quarters. The NPL ratio rose slightly from the previous quarter to 1.2%. Excluding Citi Taiwan, the NPL ratio was unchanged at 1.1%.
Slide 13. Third quarter specific allowances of $196 million or 18 basis points of loans were higher than the low levels in recent quarters. The increase of $18 million during the quarter was due entirely to the allowances that were prudently taken for exposures linked to a recent money laundering case in Singapore. Excluding this anti-money laundering specific provision, underlying specific provision was flat to second quarter. For the 9 months, specific allowances rose 3 basis points to 11 basis points of loans or $374 million.
Slide 14. Total allowance reserves stood at $6.63 billion with $2.72 billion in specific allowance reserves and $3.91 billion in general allowances. We have continued to add on to general allowance reserves through prudent overlays given the macroeconomic uncertainties. As at 30th September '22, model overlays to cater for stress situations, which are over and above baseline general provisions, stood at $2.2 billion out of the total GP of $3.9 billion. Allowance coverage stood at 125% and at 216% after considering collateral.
Slide 15. The Common Equity Tier 1 ratio of 14.1% was unchanged from the previous quarter as the impact of Citi Taiwan integration was offset by profit accretion and a decline in risk-weighted assets. The leverage ratio of 6.4% was twice the regulatory minimum of 3%.
Slide 16. The Board declared a quarterly dividend of $0.48 per share for the third quarter, bringing the dividend for the 9 months to $1.38 per share. Based on last Friday's closing share price and assuming that dividends are held at $0.48 per quarter, the annualized dividend yield is 5.8%.
Slide 17. So in closing, we achieved record total income in the third quarter and new highs for 9-month net profit and ROE. Net interest margin continued to expand from higher rates, and the growth in commercial book noninterest income was sustained. The successful integration of Citi Taiwan progresses our strategy of building meaningful scale in our growth markets. As we enter the coming year, higher-for-longer rates will be net beneficial to earnings, while a solid balance sheet with ample liquidity, prudent general allowance reserves and healthy capital ratios will provide us with strong buffers against macro uncertainties. Thank you for your attention.
I'll now hand you over to Piyush.
Piyush Gupta
Thank you, Sok Hui. So again, as usual, a few comments to add to what Sok Hui has talked about. Given, obviously, we've had a few challenging weeks, it's been a challenging quarter from a technology standpoint, and therefore, it's refreshing to be able to record that the business itself has actually been quite robust in the quarter. Sok Hui walked you through the big numbers. But just to highlight, 16% income growth for the quarter reflects solid growth not just on margins but on fee activity. The noninterest income has been -- I guided earlier the second half of the year, we should start seeing strong high single-digit growth, and that's coming through. It's about 9% for the quarter, which is very solid.
NIM, like I said, went up, continued to grow up. The total NIM was up about 3 basis points for the quarter. Some of it is due to the fact that the TM book shrank. And so the negative drag on TM is lower, right, because it's a negative thing. It's about half and half: half from the commercial book, half from TM, if you will. I do think, though, that NIM probably peaked in this quarter from where it is. So given our view, and I talk about outlook, I don't see more rate hikes coming out of the Fed. And therefore, NIM will be unlikely to go up from here.
The little challenge was loans. The beginning of the year, I thought we might be able to get close to single-digit loan growth, mid-single-digit loan growth. Then by the middle of the year, it wasn't coming through. So we said it will probably slow down, get to low single digits. In reality, I think the loan growth has been even more subdued. And therefore, loans in this quarter actually declined, partly of our own volition because we weren't getting pricing on trade, and so we continue to let trade runoff. And partly, overall, a lot of people, because of high interest rates, if they can pay down the loans, they choose to pay down the loans. And so loans were actually a little softer than we had anticipated earlier.
Like I said, fee income was strong. 9% year-on-year is very solid, and that's broad-based. We had 22% growth in wealth management. We had 20% growth in cards, but that includes the Citi Taiwan. Excluding Citi Taiwan, it was 12%, 13% growth in card fee. Loans, everything, fee income is very solid. Our cost/income ratio continues to be well managed at 39%. And asset quality is good. Sok Hui talked about the fact that we've taken some provisions for this recent money laundering cases. People are aware. We've filed some charges against companies who have been associated with the money laundering case for properties, and we just chose to be conservative. No idea what's going to happen with the case, but we just chose to be conservative and provide for all of those.
So I said -- we haven't done well on technology, and so let me talk a little bit about that. Both the internal reviews we've done, the third-party reviews, internal reviews we've done, really point to 4 areas that we need to do better from a technology standpoint. So the first I look at in terms of change management. If you look at the issues we've had this year, several of the issues actually boiled down to either software bugs, which come in the vendor systems we get, or mistakes made by our own people. And it's hard to figure why you're getting more bugs now than we have in the past. Now this is purely my speculation that post COVID, people working from home, I think there has been some pressure on software quality in general around the world. I mean that's only a speculation. I don't know. However, the key thing this raises for us, which we are working on, we can do better, is to actually improve the quality of the change management that we go through.
So what does that mean? That means we've got to, a, tighten the processes. So put more gates and quality assurance on every software before we put it into production. So we're tightening that up. That's a process we're going to change. The second thing that we need to do, though, is we got to do more comprehensive kiosk testing and a more comprehensive production assurance testing. That means today, when we get a system, we test the system. We test the periphery of the system. But sometimes, it's not easy to test in a live production environment because you can't test it close to live data. It has consequences. But one of the things we're going to do is create a very close to live production assurance environment, we should have that ready by year-end, so you can really do a lot more comprehensive testing of anything you put on board.
As you know, we have a micro service architecture. And what that means is, therefore, that we take a lot of different systems and package them together. And that's fantastic. I mean that's the architecture most tech companies have, and it's a great architecture to have. But one challenge it poses is the butterfly effect challenge. So if you have a bug in one part of that ecosystem, you can start seeing problems elsewhere. And so more comprehensive production assurance testing, I think, should be able to have that process.
We've already started this process. We started the process a few months ago actually, improving the system development life cycle, retraining our people into making sure this thing -- creating a new quality assurance team to make sure [ that gates ] before we roll things into production. The production assurance system and the environment itself will take till year-end to get up. But once we have that, I'm confident that we can do a much better job with the change management process.
The second thing that we need to do better is system recovery. So while change management is to make sure that we minimize the challenges we have, the second relates to if you do have a challenge, how quickly can you recover from that challenge. And again, the recent incidents have continued to point out that we're not as quick as we need to be in terms of recovery.
Now the issues actually really are 2 things. One, for some of the systems that we have, we use -- like I said, we use systems from various places, we've got very good talent to understand and troubleshoot. But sometimes the problem in the bug is really deep down in layer 3, layer 4. And therefore, we rely on our partners to try and figure out where the real underlying problem is. That's an area that we have to improve, which is to get more engineering and deep down understanding inside those shops. So that -- obviously, in the process of getting talent on board, it takes a little bit of time to get the right people.
The other thing we need to do is our whole system architecture is what's called active/active or self-healing. In all of these systems, recovery happens by itself because you have redundant systems and all the systems in multiple data centers. So the same application is running on multiple machines. It's running on multiple data centers. And what that means is it should self-heal. If one thing goes down, the rest of the system just picks up the load. You don't really have to do anything with it. But one of our realizations from all of these things is sometimes what happens is that because underlying systems running everything have data application, actually, you do wind up with problems where all the -- across the data centers, across the machines, across the servers, everything can get affected by a bug at the same time.
So then you need to improve your what's called a passive recovery, so how do you actually go to some passive source and to get back. And we are working hard on that. So for all of our critical -- we identified the critical applications. And we're working on building these more warm standby, hot standby and passive recovery to be able to do that in the event that something like that happens. So this is, again, work which has been ongoing for a few months. But over the next few months, hopefully, by the end of the first quarter, we should be able to make a far more robust recovery pathway to be able to get back.
There are a couple of other things. Incident management. Really when an incident happens, how well can we manage it, how do we get on top of it our communications. There's some improvements we need to and can make with the technology risk governance, including beefing up line 2, line 3. We are hiring some new talent, which will be on board in the next couple of months for some of these areas as well.
The other thing that we're trying to do is we -- in terms of service recovery and service availability, we set up some new targets for service availability and recovery. And this is a little different from the MAS regulations -- it's on top of MAS regulations, not different. MAS regulations require us to maintain uptime and delivery at the system level, at each piece of technology. And it's quite obvious, however, that if you have a service, let's call it payments, the payments service actually relies on many pieces of technology. There are like 6 or 8 systems. And while we can focus on each system, what we now want to do is focus on the whole service, across all of those systems as well.
And we're trying to set up some targets at the service level, which means that, let's say, payments -- let's stick to payments for a minute, that even if a system goes down, we should be able to provide an alternate path to be able to make a payment. In the case of payments, we can do payments, in our case, through PayLah!. You can do payments through the mobile bank. You can do payments through our web browser-based app. And one of the things that we're going to work on is to decouple these 3 so that if any one of these goes down, the other pathways can still be brought up, right? Today, all of them are 3 in the front. But in the back, sometimes they're a single common system. So we're going to try and decouple that.
This is also going to take us a few months. But we're quite hopeful, in fact, we're quite confident, that over the next few months, well within the 6-month period, we should be able to make these changes: improve our production assurance, decouple some of our systems, improve the passive/hot standby where we need to be able to do stuff. So like I said before, we're not proud of this. We -- our customers expect and deserve better. But we're also fairly confident that the issues that we've been able to identify are issues that are flexible. And so over the next 6 months, we think we will be able to get our hands around them and get to much better levels of stability, resilience and service as we go forward.
Last thing I want to go quickly about '24. I'm going to try to give a slightly more specific guidance at this point about the next year. It's a little trickier this time because of uncertainty from various things. I'm quite clear the U.S. has been far more robust than anybody expected. But with the rates where they are, and I think rates are going to stay high for longer, you should expect a slowdown in the west.
China, I think we've seen the bottom. I think the measures coming through after the end of July have put a bottom under things like the property market. But nevertheless, I think the recovery in China will still be a little up and down, a little patchy. So I don't think it'll get much worse, but will be patchy. And one of the challenges -- while the rest of Asia is doing well -- India, Indonesia, the big countries are doing well -- I do think we have to keep an eye out on the geopolitics, especially oil prices. So because of the Middle East conflict, I'm hearing binary views on oil. And if oil really gets to be much more expensive, then obviously, many of these oil-importing countries, we've got to keep an eye on.
And therefore, since there's so much uncertainty, we will reserve more specific guidance for later in the year. But the high-level thing that I think is worth being able to reflect on how we're thinking about there are on interest rates. I don't think you'll see any interest rate cuts in the first half of next year. I don't think you'd see an interest rate hike, by the way. First, I think we're pretty much done with rate hikes from everything we're hearing as well as the latest data. I think they'll stick to rates where they are. How much rates drop in the second half of the year is an open question. My own view is we'll probably get a couple of rate cuts, but not more. But I could be wrong. That's why there's uncertainty. I could be wrong because if there is a big issue in the Middle East and it becomes a bigger this thing, then, yes, that kind of event place could promote them to do a bigger rate cut -- set of rate cuts.
However, from a business standpoint, for us, it is not a material. I think our full year NIM this year will wind up at about 2.16%, the full year NIM for the year. And for next year, I think the NIM will be about there. It could be a couple of basis points off. But from a net interest income, I think we'll be flattish or maybe slightly up for the year. So even if NIM is off by a couple of basis points, it will be made up by loan growth. And the correlation between NIM and loan growth actually -- is actually quite distinct. So our model shows that if NIM goes down a little bit more, loan growth will pick up more. So right now, because rates are high, loans are not growing. So I think there is a counterbalancing effect. But net-net, I think it's safe to assume that our net interest income will be about stable. And frankly, when you add back Citi Taiwan, we might even see it up a little bit.
On the fee income side, as I said, we're getting strong high single digits this year. I think we'll get double-digit fee income growth -- noninterest income growth next year. Momentum is good in wealth management. Momentum is good in cards. If the capital markets are a little bit kind, we should get some pickup in investment banking. But even other than that, I think [ fee ] momentum should be likely to be sustained.
From a profit standpoint, I think we'll get mid-single-digit income growth. We'll get 2%, 3% growth ex Citi Taiwan; another 2%, 3% from Citi Taiwan. So we'll probably get mid-single digit. And we'll probably get high single-digit expense growth, Again, 5%, 6% ex Citi Taiwan; maybe 8%, 9% with Citi Taiwan. So our overall profit before allowances should be up.
Our total allowances, we have no insight for -- we're not seeing any major pickup in delinquencies or stress or strain in any of our portfolio. But I'd just like to start the year figuring that we will revert to mean. And we think at this stage, the 17 to 20 basis points is a good way to think about allowances. If it's much more than that because the world gets much worse, we have enough GP cushion that Sok Hui talked about to release it. If it stays this way -- let's say, we'll wind up at about 12 basis points -- 11, 12 basis points. So if the world stays the way it is, we might actually beat this. But I think if you guys are -- analysts are doing the model, it's not a -- 17 to 20 is not a bad number to look at.
So when you put all of that together, I think we'll wind up with holding net profits at about this year's level. This year, as you know, we are fairly confident we'll cross over $10 billion, and I think we'll be able to hold and protect that $10 billion number into next year. Like I said, as we get closer to the end of the year and get some more definitive views on the markets, we might be able to sharpen that a little bit. But that's a good way to start thinking about the future at this point in time. So why don't I stop there, and we can take some questions.
Question-and-Answer Session
A - Edna Koh
[Operator Instructions] Chanya?
Chanyaporn Chanjaroen
Yes. Chanyaporn Chanjaroen from Bloomberg. I have 3 questions. You mentioned a few basis points of NIM. This is for the whole year in 2024 from 2.19%, yes? For second question, you touched a little bit on AML exposure. What kind of charges were you talking about core charges? And...
Piyush Gupta
Say that again?
Chanyaporn Chanjaroen
The AML case. You mentioned you filed charges on property. Could you please elaborate? Did DBS file STRs to the police before the arrest? Are you expecting any regulatory penalty on this case? My third question, regarding the digital outage and DBS -- MAS penalty, what impacts do you see on your system? There are some talks about projects that you have with JPM and [ maybe and ]. Could you elaborate what projects could be affected because of the ban? Peter Seah also mentioned compensation matter. Could you give colors like what kind of level that [indiscernible] of management that we'll see pay cuts and how much?
Piyush Gupta
So on the first one, I think I said NIM will be off maybe a couple of basis points from this year's average NIM, not 2.19%, but this year's average NIM, which is closer to 2.16% -- 2.15%, 2.16%. I think 2.19% -- NIM peaked in the third quarter. But if you look at the average for the year, the first 2 quarters were lower. The last quarter, likely to be a little bit lower. So that's what we should look at from there. We could be around the same levels next year. But like I said, it all depends on what happens to rates. If they don't cut rates, then we have some tailwinds. We have some headwinds from loan growth, but somewhere there.
The second is the allowances. When we file charges for any lending that we do, we file a charge in 2 ways. One, we issue a charge against an ACA for any company that we give a loan to. We file a charge, say, when we've given a loan to said and said company. Or if we've given a loan against property, we file a charge with the mortgage to make sure people can see. So this is standard in banking anywhere in the world. So people can see, is this property unencumbered? Or is this company -- does this company have borrowings from somewhere else? So that's what we say when we say filed a charge. And so we look at our data, it shows that we have filed charges against companies which are linked to this thing, i.e., we have some exposure to these companies, if you will. The total extended exposure in the provisions in this quarter's number are just a tad bit below $100 million of charge and exposures.
The third thing on the outage and the overall impact. So if you look at the guidance, I mean, obviously, like I said, we're going to focus a lot of our energies in the next few months at trying to build the resilience in the areas I talked about, right? So improve our change control, our testing procedures, our unlocking systems. So that's going to take a lot of our energies and effort. But in terms of the specific directions, we really didn't have any M&A or new business venture plans. So you don't really see a direct impact of that. Also, we haven't really closed any branches and ATMs in the last 3, 4 years. We closed some in the early years of COVID. And in fact, in the recent years, we've only built back. We've actually added some [indiscernible].
What we will have to do is defer some of the other product features, new products, new services, et cetera, which we might normally have done. But in reality, because we're going to have to focus on building the resiliency, it is not something that we would have been able to put resources against anyway, right? So we would have to -- I would have started doing it and -- because [indiscernible] is good. It gives us a 6-month window to consolidate, and that's actually a good thing. So it does mean that we will be somewhat later in delivery of some products and services and stuff to the market in terms of what we want, and it's never something you want to do. You don't want to defer some of the stuff. But it's like putting brakes on. When you have good brakes, then you can run faster later. So my thinking is if we can get the resiliency and the underbilling stronger over the next 6 months, then it give us the opportunity to speed up the delivery of products and services in the subsequent part of the year.
Your last question on compensation, frankly, the way the compensation process works is it will all happen between now and the end of the year and early this time. And it's not something that we, the management, decide. The Board gets together, reviews the whole thing and comes to a view on what the compensation, in fact, needs to be. What I can tell you is that from -- we -- as you know, we run a close-knit, tight management team. From a senior management team, we have agreed that we would share responsibility and accountability together for this. But what the actual quantums and cuts and sizes, I don't know, you'll have to wait until we announce. And our compensation is public in annual report filings, which come out at the end of the first quarter. So it will be visible at that time.
Chanyaporn Chanjaroen
Sorry. One question you didn't answer. On the money laundering case, did DBS file STR...
Piyush Gupta
Oh, yes. We filed STR.
Chanyaporn Chanjaroen
You filed.
Edna Koh
[indiscernible]?
Unidentified Analyst
Can you hear me? I wanted to ask whether DBS is considering taking its tech 100% in-house moving forward considering that the October 14 outage is caused by a vendor contractor? And also, as a follow-up, what was the rationale behind the bank's decision to outsource its data center to a third-party vendor? And then another question, like how many data centers does DBS own at the moment?
Piyush Gupta
So actually, almost every big company in the world outsources data centers. Managing data center is a very specialized business, and few companies try to do it because it's managing a large real estate in very complex ways. And that's why the biggest data center providers in the world are specialized, whether it's ST -- Singtel or ST Telemedia or M1 or Equinix. Equinix is the world's biggest data center provider. They provide 250 data centers around the world.
And now increasingly, as there's more focus on green data centers and resilient designed data centers, it's much harder for individuals to do it. These big companies do it. Frankly, even the Amazons, Googles and Facebooks of the world don't -- they also use third parties to provide and put together data centers. So it's not an unusual thing to do that.
We got out of this business of running our own data centers 15 years ago. So we've decided a long time ago that we would outsource and let other people manage the physical data center capability. These people build high-quality data centers. Data centers are tagged as 1, 2, 3, 3 plus, 4. This is -- the current data center is 3 plus. It is the highest tagged data center in Singapore.
When you do your own data center, it's very hard to achieve that level of resiliency. 3 plus or a 4-level data centers are very hard to achieve. And so no, we have no plans to take in data center management. I don't think we are very good at managing physical infrastructures, if you will. Most of the big data center providers run at 99.9999% operations, right? So it's extraordinarily high level of resiliency that they run with. And frankly, it's the first data center problem we've ever had.
Edna Koh
[Indiscernible]?
Unidentified Analyst
Just following up on the digital disruptions a bit. Ravi Menon said over the weekend that there are some deep-seated issues that need to be resolved. What are the biggest deep-seated issues that you see here among the ones that you've identified?
Piyush Gupta
I just went through that at some length. So I can repeat my comments if you want...
Unidentified Analyst
Maybe the biggest ones, if you...
Piyush Gupta
So I said, of the 4 biggest things, I think -- like I said, the change management process, if you look at the bulk of our problems, it's because we run into bugs, both either third-party bugs or our bugs. So each one -- the incident in March was bug related. The incident in May was bug related. The incidents in October -- other than the data center outage, all other 4 data centers were bug or software related, right? So to me, the big issue is how do you make sure that you get good change control because the reality is that as you use a lot of different systems and architecture, you will run into more software bugs. It happens. Your phone gets a bug.
Everything gets a bug. But to get high resiliency, you've got to be able to make sure that you test well, right? And so the -- to me, the first deep-seated, how do you make sure you get the discipline and rigor before you put things into production and improve the quality of the production assurance testing you can do so you don't wind up with buggy software and this thing.
The second I'd point you about that, if you look at recovery, there are 2 issues in the recovery. One issue is we're going to have the deep engineering talent because in at least 2 or 3 of these incidents, the bug itself was so deep that we wouldn't be able to pick it up. So we had to go back and work with the vendor, and they had to put a team to try and figure where the problem might be. And that takes time. You go there -- even though we have SLAs, we've got to go talk to them. They come back to us and so on. So we've got to improve the engineering depth in some of these things so we can try and resolve some of those troubleshooting better. So that's a deep-seated issue.
And the third deep-seated issue is just in terms of thinking about -- our whole thinking is active/active, so redundant and self-healing for recovery. And we've got to figure how we, in relation to redundant self-healing, we add some passive recovery system in the critical areas in case we need those. I mean these are the 3 that I called out. All of these are work in progress. We started to work on this in May. So we've had 6 months under our belt, and I think we can get most of these done within the next 6 months. But each of these is not trivial to get done.
Edna Koh
[indiscernible]?
Unidentified Analyst
Can you please share with us a bit about your views on ROE expectations for next year? And the second question is can you please share your view on the impact from more geopolitical tensions in the Middle East on business in general and wealth flow in the region?
Piyush Gupta
So on the ROE, we guided before that we should be able to stick to north of 17% ROE, and I think that's a safe assumption to continue to work with. That's correct, right, Sok Hui? Yes. On the Middle East tensions, [indiscernible] our direct exposure to Israel, Middle East, et cetera, is quite limited. We've got some bank facilities to U.A.E. and Qatari banks, and I don't think anything is going to happen with that. So I'm not concerned directly from a DBS standpoint about those tensions.
The indirect issues, though, can be a little bit more concerning. It depends on how bad it gets. If you do -- if it stays targeted and contained, I don't think you'll see too much spillover. But this has the potential to become a larger regional escalation. And if it does, then I do think that you have some possibility of oil getting out of hand, and the principal mechanism by which the rest of the world will be impacted is what happens to oil price. So if you see oil price getting up to $150, for example, we know -- we've seen that before that at $150, then most of the countries in the region start having challenges with their trade deficit, trade account, et cetera, et cetera.
And so that will result in some further slowdown in the region as well as some more currency volatility if that were to happen. And so that's something we're going to keep a close eye on. How -- and I can't forecast whether it escalates or doesn't escalate.
Edna Koh
Anshuman?
Anshuman Daga
This is Anshuman from Breakingviews. Can you just elaborate about the -- you said you had exposure about -- total exposure of $100 million for the episode that happened in Singapore. And also, how is it affecting -- how do you see this affecting wealth flows overall for Singapore and the region? If you can just elaborate some trends on that.
Piyush Gupta
Yes. So we had about $100 million in exposures, and we basically provided for most of it, if you will. Most of these are missing property. So a lot of these -- some of these people had opened actually retail consumer accounts with us. We had filed STRs. Consumer accounts is like they transfer money from this thing. And so we didn't have big PB accounts and things. We have consumer and listing accounts. But what they do when they open an account, they use us to finance a property purchase or something, and that results in the exposures.
In terms of total wealth flows, I don't think it will have a material impact. I think we will all continue to tighten the process. But I've said before, I think Singapore's regime has a very robust money laundering regimen in the process. And so you will find people who get through the system but won't change very much. I think net new money flows have continued to be robust through the third quarter. They continue to be fairly solid into October as well. So I don't see the actual money flows and growth of wealth materially suffering.
I do think that like other players, we will continue to tweak and tighten what else we can pick up. In this case, it's obvious that we're going to tighten up again use of multinationality passports. Now you've got to do that carefully, though a lot of legitimate people also have passports from other countries. So we don't want to wind up with financial exclusion either.
So you can do a little bit more on that, trying to make sure you pick up multi-country passports kind of thing. All of this, while these people came into the consumer bank, and so they didn't bring large money flows, but we could tighten up to see what happens when money comes into consumer flows. We can't do the scale because we have 12 million consumer accounts. So you can't do a lot of that. But there are things you could tighten, but I don't expect it to have a material impact on wealth as a total segment to industry.
Anshuman Daga
And I mean what's the reason you don't expect a big impact? I mean if there are other sort of high-profile individuals looking at this, and they will obviously look at the safeguards that are happening, I mean, why is this not a concern in the sense that this flow of money is coming into Singapore and could be disrupted now?
Piyush Gupta
So most of the money coming into Singapore [indiscernible] if you think about the amount of money that came into Singapore in the last 2, 3 years, and this is a small fraction of that money. So we assume that the fraction of the money and people who take money laundering are warned, shot across the bar, don't bring money in, this is like a small fraction, maybe -- I don't know, but a small decimal or percentage of the money that comes into Singapore.
Anshuman Daga
And also, if you can just speak about the loan growth, I mean, that's been a bit of a disappointment. And how do you see this shaping up, especially when the economy is going through a tough time? Obviously, the forecast is it's better next year. But how is that impacting? And some color on the sectors where you would expect a revival for the loan growth.
Piyush Gupta
Anshuman, we're actually seeing loan growth even now. So it's a mix of things. We saw loan growth in this quarter and continue to see growth in the property sector, continue to see some growth in energy, some in commodities. But what's been happening so far is that because rates are so high, anybody who can access money either through equity or through their own cash flows or through tightening the working capital is also paying down.
And so you're seeing whatever loan growth goes away because anybody who can find money pays down the loan because when you're borrowing at 6%, the quicker you can repay, the better it is. And the other piece on loan growth has been that particularly in trade, but also at the [indiscernible] mortgage, there's been a little bit of compression on spreads. And so we've let some of that go because there's no point keeping the loans if you can't make a decent spread on it.
So it's going back to when I said next year is harder to call because it depends on your rate outlook. If rates stay where they are, then I foresee that you might continue to see that. You'll get loan growth in some sectors, and our pipelines are quite robust. But you might continue to see people pay down at the same time because of the high rates. So net-net, you might -- we forecast -- our loan growth right now, the underlying business forecast are 5%, 6%. But when I look at the top-down view on how many people, I figure we might wind up with only a 2%, 3% loan growth. So by end of the year, I have a firmer view.
Edna Koh
Goola?
Goola Warden
Can I ask about -- once again about the loan book? Is it all now priced at the higher yield? And because there's been some pressure, I think maybe just anecdotal pressure, in mortgages, they're all coming off, so do you expect lower yields if the book reprices into 2024, if you could? And do you expect the cost of funds also to come down if that happens?
Piyush Gupta
So Goola, the short answer is no. We have another $105 billion of loans, which are yet to be repriced. And of the $105 billion, about $10 billion get repriced this quarter, another $40 billion next year. So about $50 billion in loans will get repriced. And the loans which are getting repriced are still getting priced up at close to 2% lift. So -- between 1.8% and 2% lift. So that's one of the tailwinds that protects our NIM that we still have a chunk of our fixed rate loan book which will continue to reprice and give us a pricing lift. But it's also true that the through-the-door pricing on mortgage has been coming down as well. So our through-the-door pricing for the quarter was about 3.3% through the door coming in, but it has been creeping down as well. So there is some pressure on NIMs because of the new mortgage bookings. But for us, because so much of the loans reprice from a very low base, you get a lift from there as well.
Goola Warden
And then on the SPs, I mean, you said it was for the AML. But there's some concern over the CRE, the commercial real estate, in the U.S., and I think we talked about China. But what's the size of this? Is it very small? And I know you have a lot of overlays. And is it all covered? Is [indiscernible]?
Piyush Gupta
Yes, it's all covered. So actually, we don't have a big book in commercial real estate in the west. But nevertheless, in our macro overlay this quarter, we added some more for the potential of commercial real estate in the U.S. So I think it's well covered.
Edna Koh
Okay. If there are no further questions, then thank you, everyone, for coming. We'll see you next quarter.
Piyush Gupta
All right. Thank you all.
Chng Sok Hui
Thank you.
For further details see:
DBS Group Holdings Ltd (DBSDF) Q3 2023 Earnings Call Transcript