DBS Group Holdings Ltd (D05) Earnings Call Transcript & Summary

Ladies and gentlemen, welcome to the Q4 2021 DBS Analyst Briefing. I will now hand the session to Michael to begin today's presentation. Michael, please begin.

Hi, everyone, and welcome to the session for the buy- and sell-side. You have seen the media briefing, so we can go straight to Q&A. Diana, can you open the floor for questions?

[Operator Instructions] Our first question, Mr. Nicholas Teh from Credit Suisse.

Just the 2 questions from me. First, can you talk about the CASA level that you guys have? If we're going to an environment where interest rates go back to the previous peak, do you expect the CASA to stay above that historical 60% level? And if so, is that because you think that the amount of liquidity in the world just stays high? Or has there been a structural change in your deposit franchise over the last couple of years? The second question is just on cost-to-income ratio. Again, we move back to that 2019 NIM level. Do you think that cost to income can get towards that -- or get to the aspirational below 40% level?

Okay. First, on the CASA ratio, our own sense is that, while there will be a reduction in macro liquidity this year, there will still be a lot of money floating around. The Fed balance sheet was up to $9 trillion. And if we look at the last time when they tried to bring down the balance sheet from 4.5, they got to 3.6, 3.7. And then they didn't go down much beyond that. So I do think that while they make every effort to start tightening liquidity, it's not going to be that easy. It's going to take some time. So first of all, I do think there will be a large amount of money in the system. The overall growth of money supply obviously will moderate. So now this time we are saying you will probably get money supply growth, which is more aligned to GDP growth as opposed to far in excess of GDP growth. But nevertheless, there will be a supportive environment for deposits, notwithstanding. Within that, we do think that some of the CASA will flow out. We got to 76%, I said $140 billion of CASA. So some of that is likely to flow out. And in our modeling, we've assumed that of the delta growth, about 25% of that will be repriced and threw out both in Sing dollars as well as in U.S. dollars. So there will be -- definitely be some impact. Now at the same time, our understanding of the kinds of monies we have, the operating accounts and the underlying transaction volume growth that we have, cause us to believe that we will get sticky CASA. So our growth from the high 50s to 76%, my goal stayed 76%, I don't think we'll get back to the below 60% type level either. And therefore, somewhere in between those 2 numbers probably where we will wind up. The cost/income ratio, yes is the short answer. If we just do a different math and you add back $3 billion or $4 billion of income into our numbers right now and keep the cost base roughly where it is, you get that cost/income ratio, which is well south of 40%.

Okay. Got it. But I guess if you do get that improvement in income, you wouldn't kind of, I guess, look at accelerating some of the digital spending or anything like that, that you have in the plans?

Well, part of it is a function of bandwidth. If you look at historically, even in good times, our total spending has, over the last year, gone up from $900 million to about $1.1 billion. And so as possible, if we have a tailwind from rate, we could go and hire a few more people. We want to create a new center. We can do some more activity. But in the big scheme of things, we are talking about an incremental $50 million, $100-odd million, right, relative to the inside, where you're talking billions of dollars of delta.

Our next question, Aakash from UBS.

Sure. So 3 questions. The first one, distribution, just to gather your thoughts on the rate cycle and like what is the base case in your mind. So like you also said from the past experiences, it doesn't seem like the Fed goes all the way when they start raising rates and then they start backing off pretty soon after. So what is the base case for you this time? What do you think happens? Do they have the capacity to sort of go all the way like the 6%, 7% rate hike that has been priced into the market? Or do you think recession sort of fears start setting in once they proceed?

Well, for my money, I think we'll have much better sense by June, by the summer. And the reason for that is the question of how sticky is inflation. And then our view is the inflation is -- I've been saying for some time, if I remember last quarter, I said I don't think inflation is transitory. It's not the supply chain snafu. Actually, last year, global trade volume went up 5%. So the supply chain held up quite nicely. Of course, there were backups in ports in Los Angeles and so on, and that made a difference. And workers didn't come to work. But overall trade volume went up. The real shift came from the demand side. The demand for goods grew dramatically last year relative to demand for services. And so the first question is, does the demand for goods fall off quite sharply this year or not? The second question is commodities, how high is oil, energy and so on? How sticky does it stay? And the third question is wage inflation. How sticky is wage inflation? My point last time was I'm seeing wage inflation creep through everything. We're seeing it in our own business, and we're seeing it in the business of our clients. And wage inflation is sticky. It doesn't go away. So my view is that if inflation comes off the 7% handle and they start seeing spends more in the 4%, 5% handle, they will be moderated if the interest rate increases. But if, for whatever reason, they're not able to bring the inflation rate down, then they don't have a choice. I mean, as the government said, if rates -- I think Raghu Rajan said in India sometime ago that if rates are high, then it doesn't matter what the cause of the rate is. The central banks just don't have a choice. You have to take action. At our -- this thing, my current base -- the thing is that one rate hike a quarter is my base case. I could be quite wrong, but I'm currently assuming that we will probably get 4 rate hikes this year.

Okay. Got it. Fair enough. So following up on this, I think you kind of have a dividend increase. Now from a capital perspective, there's no doubt you're very, very comfortable. My question is more on the cash flow side. So if I look at the consensus estimates for this year, somewhere in the range of $6.7 billion, $6.8 billion for the full year, you need $3 billion to maintain your 13% capital for the RWA growth, 6%, 7% RWA growth is what I'm assuming. So the remaining is $3.7 billion, $3.8 billion cash flow, which you're basically paying everything out, right? Now this is in a year where your credit costs are very close to 0. Now rate hikes come through, yes, that will help you offset that for next year. But if there's any change in the rate outlook, how do -- would you think about paying this dividend next year?

I will leave Sok Hui the question. Not sure -- when you say cash flow, I think cash flow is the question -- the thing, liquidity, obviously, is not a concern. So if you -- assuming this payout ratio, I mean, how much payout am I prepared to make? Is that what you have in mind?

Yes. I mean, I think next year, if your credit costs go back to a normal level, then would you still have that $3.7 billion, $3.8 billion in dividends -- sorry, in earnings to pay for the dividend?

I see. You've not understood fully, I guess, what the guidance that we have given. So we are saying this year, our credit cost is $52 million. We said next year...

This year.

This year, in 2022, we are likely to be in the $0 to $100 million range, right? And that's because we think that our line of sight on our specific provision, that will sort of normalize, but that we also see improvement in the credit quality that will allow us to release some of the general provisions that we've set aside. So our base case assumption is that barring unforeseen circumstances, our allowance line will actually be quite flat. So I think your assumption is that it might not be. I think that's what you're asking.

I meant for FY '23. So FY '22 is very clear. Your credit cost will remain very low, and you have a lot of room to pay this dividend. I'm just asking what happens in FY '23 when your credit cost normalized back and if there's a change in the rate outlook, right? So if your revenue increment doesn't come through.

Well, I guess I understood what you said. You're saying is if our profit falls, and it could be, so you're saying that it's a $6.8 billion, I go back to $6 billion as the cost of credit picks up, I can't rely on interest income growth and there's no other growth in the business, then you won't be able to pay dividends. And of course, if you make those assumptions, then you're correct. I can't retain that dividend. But those are big assumptions to make that after 10, 12 years of regular growth, we stopped growing, that the interest rates don't come through, our double digit noninterest income doesn't come through. And the growth in our balance sheet, this doesn't come through. Now if you want to start with all of those assumptions, then yes, the dividend can't be paid. That's correct.

Sure. So I mean, if you get only 3, 4 rate hikes and then the Fed outlook changed, even do you think in that situation as well, you will be comfortably in a position to keep paying this dividend for FY '23 also? That is what I was asking.

Yes. I guess if we get 4 rate hikes, that total is -- that's worth about $3 billion or $2 billion of income to me if I get 4 rate hikes. And if I go back to a normalized cost of credit, which is at 0, I'll be at $800 million, $900 million. I'll still be $1 billion over where I am right now.

Right. Okay. Fair enough. And then just the last question on rate...

Yes. Maybe to also highlight that if we have 4 rate hikes this year, we're not going to see the entire sort of $1 billion, $2 billion coming through this year. A large part of it will flow through next year, actually. So even on the assumption of 4 rate hikes, you're not going to see a large part of that in 2022. You'll see it in 2023, simply because the rate hike will take time to flow through, right?

So I guess the other way to think about it, my noninterest income, we are now -- it's about $3.3 billion, $3.4 billion, right? And there's a freeze effect. I -- that's been growing double digit. I think we can continue to grow that double digit. So I'll get $300 million, $400 million lift from there. I will -- my volumes and my existing -- if my loan book can grow even 5%, I will get plus $300 million from there. So even if I go to normal SP level, just the underlying growth in my business should be able to take care of it, by and large. And then if I get the rate hikes, whatever else I get is over and above that.

Okay. Got you. Just the last question is on rate sensitivity. And you said $18 million, $20 million, which is a very sensible sort of calculation that you've done. So it's more like a through-the-cycle estimate, cycle sensitivity. So is it fair to say that initially, when your CASA continues to stay at 76%, the sensitivity might even be higher than that? Like we might get, I don't know, $25 million for every basis point initially?

Well, I'm not sure. I think what will happen is that if rates start going up, 2, 3 things happen. One, people start moving from CASA to FD, right? So that will happen, and that will start happening quite quickly. Second is that at the right time, remember, the Fed is buying all the corporate, all the treasuries. I think some of the liquidity in the world will go with banks, et cetera, stepping up to buy treasuries. So I do think you will start seeing some outflows very quickly, and we're starting some repricing very quickly. So you're right. I mean, maybe for a few weeks, you'll see the benefit on the entire pool. But I think the consequent effect on that in both the quantum and rate will start coming through. And that's why we modeled that in.

Okay. So the rate hike that you expect, this is the number we should be using on average or all the way?

Yes. I think it's a good way to check. I mean, if you go back and look at our interest income from 2019 to now, it's quite easy. You can figure out that our interest income is down $2.8 billion. And so you can see the rate cuts from -- as you know, the rate cuts, we're down $2.8 billion. And that pretty much shows you our sensitivity on the downward part of the cycle. And therefore, in the upward part of the cycle and now given that the book has expanded so much more, it's not an unreasonable guestimate.

Our next question, Ms. Melissa Kuang, Goldman Sachs.

I just have like -- just 2 questions. In terms of the rate hike you're assuming for, so if we do assume like 7 hikes, in terms of thinking about asset quality and also loan growth, do you think perhaps we might see a little bit more credit cost than you're expecting and also that loan growth might slow? Or do you think it will more likely be the 2022 kind of scenario that, that will happen? And then secondly, in terms of your dividend packaging, just wanted to check, I mean, looking at the dividend numbers and just analyzing you might keep that $1.44, if we think about payout ratio, is 50% the right kind of payout ratio as you're looking at mid-single-digit growth? So somewhere -- how we can think about what kind of net profit you are actually thinking for 2022? Or is 50% the right number? Or should it be lower? Or can it be higher given the loan growth guidance that you have given us?

So on the impact of higher rates on both loan growth and on asset quality, Melissa, you've got to remember that if you were to take 7 rate hikes or 8 rate hikes, we're starting from 0. And so you're saying your terminal interest rate of 1.75% and 2% and how severe or how steep it goes and do they start putting a dampener on company's willingness to invest and to grow. And my general view is not. This will take you back to where we were in the summer of 2019. In the summer of 2019, you didn't see a disproportionate impact of the 2% rate. Now you could argue in some countries, it might be, it might not be. But I think a 1.75%, 2%, 2.25% interest rate environment is not that material to impact people's investment cycle decisions. In terms of the cost of credit and credit quality, I said in the media thing, I think at that level, the only people to watch out for are the SMEs because for them, debt servicing might start -- if you add that to inflation and wage price increase and cost of goods increase, and now the interest burden also moves up to 2%, you might find SMEs start getting more [ killed ]. As you know, I always maintain that I just cannot understand why the SME sector has been so resilient in Asia for the last 5 years. I keep expecting that SME shoe to drop. And so far, that SME shoe hasn't dropped. It's -- a large part of our financing is secured, but we haven't even had to foreclose. We haven't been -- had to go and get rid of collateral. They've just been relatively resilient. I'll keep an eye out of that. If rates go back up to 2%, the 2019 level inflation is high, do you see more pain in that sector? From our portfolio, though, as you know, with building and construction, retail, tourism, all of the supply chain, textile, garments, all of the typical industries that could suffer, we've been very, very tight with those. We've been managing it very tightly. And so all our own stress testing right now doesn't show any material cause for discomfort. But yes, that is one part of the portfolio that you might see some more stress and pain in. So we will look at that. Your second question, on dividend, actually, a lot of it is opportunistic. So we like to keep some dry powder in case a deal like Citi comes along or we get the opportunity to do something like with Shenzhen Rural Commercial Bank. But I also hasten to add some cities, we probably need the next couple of years to try and consolidate what we've done. So I'm not sure we're rushing to add any more right now. In which case, it could be that we could afford to increase our payout ratios from the 50% revenue. In our modeling, we figured that actually, it could even go up into the 60s quite easily without being constraining the growth of our business. So yes, we'll take a look at it slowly.

Okay. So just in terms of credit cost, so does it mean that in 2023, you're still not going to expect provisions to normalize [indiscernible]? Or do you think by then in 2023, you'll get back to 20 bps type of [ cost ]?

The uncertainty for me is actually not the rate environment but China. So that's why I'm not -- I'm cautious about making a prediction for 2023 at this point in time because China, the existing credit risk is high. We manage our portfolio well. We've not got into any of the real estate stuff. All our supply chains are working well. But between now and '23, there are some party congresses which are being held. There are some Omicron-related -- the country continues to be shut down. The borders between China and Hong Kong need to reopen. So I say we'll have to figure and see if there's any headwinds coming out of China and Greater China before we can start making predictions about 2023 and whether you get back to a normalized credit environment by then or you still have some cushion or not.

Right. Okay. Then just to clarify on the dividend, please. So you think that you can increase it to 60%. I'm just thinking about the [indiscernible] for DBS based on...

Sorry, Melissa, the [indiscernible]. We're not following you.

Sorry. So in terms of the $1.44 EPS annualized, would we be thinking that, that is closer to the 60% that you're looking to pay out? Or is that closer to the 50% in terms of -- because you said earlier that you might -- or you can increase your EPS payout to 60%. I just wondered in terms of your kind of -- when you're doing the budgeting for 2022, would $1.44 EPS be closer to the 60% payout? Or is it still on the low side and the up side in terms of $1.44 EPS for the year?

If you do the math, you'll figure that the $1.44 is closer to 50% than the 60%.

Our next question, Modi from JPMorgan.

A few questions, Piyush. First, does it make sense to look at the sensitivity to near slope appreciation? And does it impact this $18 million to $20 million in any meaningful way?

Yes. So Harsh, that's a good question. And I'd say there is one uncertainty around it. We've been conservative. There's one uncertainty around it, and that is what happens to SORA. So in the past, SOR was directly linked to the exchange rate policy. And therefore, the slope of this thing made a material difference. The strengthening Sing dollar allowed less pass-through. A weakened Sing further allowed more pass-through and could see that quite easily. In the SORA climate, it is not entirely clear how much SORA will be determined and driven by the exchange rate situation. And that's something we're all going to have to sit and figure out and done. In our projections, therefore, we actually model for different levels of pass-through. And we've taken a base case of a fairly conservative pass-through from the U.S. rate into the Sing rate. But it is something which -- it's a good observation. Could we be wrong? Yes, we could. So if the pass-through is materially lower, then that $18 million to $20 million actually, in our tail event, it comes down to $16 million, $18 million, for example.

Okay. Okay. The second one, on the noninterest income, I see the duration is quite low, about 5 years, which is just about $14-odd billion of the book. But if we start expecting reasonably sharp parallel shift in curve, does a lot of it start impacting your noninterest income? And does that kind of offset meaningfully? As in what is the risk of a meaningfully offset from mark-to-market on the P&L and potentially on the balance sheet as well?

Almost none. Because while -- first of all, you said the bulk of the duration of our book is in the 3- to 5-year range. But more important, if you look at the underlying, a large part of our book is held at accrual portfolios and hold-to-collect portfolios. So that doesn't mark-to-market and doesn't come through the P&L at all. That either goes on an accrual, this thing, or if anything goes through the OCI line. At this point in time, the yield curve is quite flattish. So if anything, we've got to think about at what time do we want to start building some more duration in the book? So we don't have that right now, but the downside is the very little.

Yes. So Harsh, you see on our Slide 24 where you see the held-to-collect portfolio is about $50 billion. So that's not going to be impacted. We hold this portfolio for yield. And then on the fair value through other comprehensive income, it's $35 billion with the bulk concentrated in the very short term.

Yes. Yes. A couple more. On the rates, you're expecting 4 hikes or thereabouts this year. How -- have you already started repositioning in terms of your time deposit rates, in terms of your mortgage rates? Are you locking in either of the 2? Or are you saying we will not do, let's say, fixed rate model? Like how are you starting to tactically over last, let's say, a couple of months, changing your product strategy on both sides of the balance sheet?

So if you look at our mortgage through the door, in the last couple of quarters, we are now originating more than 60% floating rate. So -- as opposed to fixed rate. So we changed our pricing in anticipation of rate hikes to be more aggressive on the floating rate side than the fixed rate side. That's the clear -- in fact, we've taken the same strategy across the board. We've incentivized floating rate loans because we anticipate a pickup in rates helping us. On the funding architecture side, as you know, we basically paid down billions of dollars of fixed deposits in the last couple of years because as for our CASA, it's 0 cost money. And because right now, we are so massively overfunded, we haven't really had to go out and raise long-term fixed money at this point in time. So we've not done that.

Okay. Okay. And then the last one, on the operational risk. So is it fair to assume that whenever that 30, 40 bps comes off, that it will be available for distribution?

I guess it is because, I mean, that's the whole intent of MAS. When they've got us to add $1 billion of incremental capital, that is to set aside so much capital, which you can't put to work. And as and when that gets lifted off, then you can either use it to grow or to distribute. I've already said we've got too much capital for growth. So it will be available for distribution here.

Our next question, Jayden from Macquarie.

The first couple are just follow-ups to the ones that Harsh had. On the MAS supervision, you mentioned this on the media call. But what's the expected time line of the independent review? And what's your base case of when they may release this 50% loading?

So I don't know. I mean, I can't put words in MAS' mouth. We've got -- we did 2 reviews. One was from ForgeRock, who is the vendor who owns the software we bought. They sent an expert from the U.S. who spent a month with us. The second was from Deloitte, who's actually a system integrator for this software. Then MAS asked us to -- and so we appointed another third independent review who started the work a couple of weeks ago. That will get done by May. It will take about 3 months to do. But we're not waiting for that. So there's some -- quite clearly, some things we need to tighten up, and we are doing that in parallel. My own sense is that by the third quarter of this year, we'll be able to demonstrate that we've got all of the areas that need to be fixed, fixed. Then beyond that, it's up to MAS, whether they're satisfied, whether they want us to do more, whether they want to take some more time to come to a view or conclusion. The last time in 2010, the ATM outage took 14 months before they reversed the charge. So -- if that gives you any frame of reference. So I think we'll be ready in 6 to 8 months, 9 months. Whether MAS wants to extend it is up to MAS.

Okay. That's really helpful. The next question is just on the NIM. Just to come back to the Sing near slope. I guess MAS sounded quite hawkish lately, and they're obviously seeing inflation come through as well. So you said that your modeling is conservative. Does that mean that you have the plus 1% slope already embedded? Or is it a different level?

Now the way we do it, we don't need to embed the slope of near. We need to take this thing at what swap point of our points and what is the likely flow-through. So we measure -- historically, the flow-through from the U.S.A. to Sing rate can raise anything from 40% to 60% to 80%. So we model at different levels of flow-through into the Sing near rate. We also model -- actually, our book is they benefit in the U.S. dollar book, where, of course, you want 100% flow-through. Then there's the Hong Kong HIBOR book. Today, that's material because Hong Kong -- by the way, CASA is at 82%, even higher than our group CASA. So we modeled the Hong Kong dollar flow-through, and that's generally more aggressive because Hong Kong dollar is the affected currency. And then we model the Sing dollar flow-through at these various levels at 40%, 60%, 80%. And we take a conservative baseline around that. That's how we do it.

Okay. That's helpful. And then my last question, and this has just come up more in conversations with investors, but I don't mean this in any negative way, it's just around succession. I mean, the stock has done so well. The bank is performing really, really well. You've made a couple of, I guess, really timely and opportunistic acquisitions. Is there any sort of latest thoughts around how DBS sees succession at the management level, whether it's a senior position or otherwise?

Well, as I maintained before, we are very robust in our succession planning effort. And so we've had -- identified potential successors for me for years, and we keep sort of developing them and rotating them and giving them mobility, et cetera. So if I get hit by a bus, we have enough bench strength. Also, our team has been so stable over the years. It's a very solidly functioning team. So everyone is expecting the continuity. Now having said that, on the other side, in the last few months, Jamie Dimon has got this never-ending 5 years in the future forecast. Brian Moynihan has just declared he's going to work till 70. James Gorman just said he wants to work until 70. My Board is in no hurry for me to leave. So I've got enough runway ahead of me right now. I don't think there's anything imminent.

Next question, Weldon from HSBC.

Can I just understand the $1.5 billion of GP? In the base case assumption of you said about 0 to 100 million credit cards this year, how much is being written back from that? And then you went right back through the whole thing, I guess, but you had $400 million above MAS requirement. So is that sort of the range that we could see within this year and then next year?

Yes. So I think, again, we predefined that total GP stack is actually $3.9 billion. $1.5 billion is an overlay. The balance is in the normal sort of methodology for setting aside GP. So when we talk about the overlay, we have said we have not released any overlay since we have built it up in the COVID period. So the $1.5 billion has not been touched since we built it. That's a point of clarification. So any release in GP that you see in the year is purely a function of the normal stack as the loan portfolio quality improved, as weaker cases are migrated out, as tenants get short term, as cases get resolved. So the $447 million that you see in the GP write-back is purely from the normal spec. Does that answer your question?

Yes.

I guess let me add to that. So first of all, if you take our total $3.9 billion, $1.5 billion is the overlay, so $2.4 billion is more still lying in the normal stack. That normal stack reversal, I expect to continue to happen this year. And that's because we've been quite conservative and we continue to see improvement all around. So people are repaying. We are seeing upgrades in our portfolio. We've got some -- for potential reasons and precautionary reasons, we've downgraded some names there. Now they're beginning to perform, we're upgrading them. So I do think that we should be able to get some reversal from the $2.4 billion normal stack this year. Then beyond that, depending on how the SP performs, it is also possible that we might be able to reverse some from the overlay if we need to. Though, if I had to bet, I'm not sure we'd wind up having to release a lot from the overlay this year, maybe $200 million, $300 million potentially. If that is the case, we still have enough overlay that we could potentially look to releasing next year. On a going concern basis, I would like to keep some overlays. I don't intend to ever release the $1.5 billion because we always make sure that we have some top-up. But I mean, we have some opportunities not just into this year but next year.

Right. But so in the base case for this year, how much overlays were you planning to release in that guidance that you provided?

We don't have a base case because it all depends on how much I can release from the $2.4 billion stock.

So our overlay release is also guided by our thinking on, I guess, the cross-border opening up. So I think we are monitoring sort of a number of indicators before we decide whether we are comfortable enough to release the overlay as well. We track the [ efficiency ] index. We track airport arrivals. We track delinquencies after moratorium are listed, et cetera, et cetera.

Okay. Got it. And then just another question on trading. I think just now, I guess in Sok Hui's presentation, I think you said that structurally, the treasury market has -- recurring income has improved. But in the net trading income in the income, how does that $1.1 billion translate to the net trading income line in the income statement? Like because this year...

It's not based off the interest income and net trading income. And frankly, it's fungible. So it depends on what the traders want to do. And sometimes they wind up actually taking positions that show up to the yield and the spread. Sometimes they wind up actually working through the derivatives line. So it doesn't show up in the interest income. It shows up in the trading income line. I don't bother about that too much because it's actually quite fungible how they want to treat it. I think what's more important is how robust is the quarterly performance. And as you know, we have now 2 years of $1.5 billion. Last year was helped by the interest rate collapse. So they obviously made some money on the fixed income portfolio. This year has not been helped by that at all. This year has been very active across the whole range of other debt. And as we look at the underlying, therefore, we're relatively confident that getting up to what Sok Hui said, $275 million a quarter is $1.1 billion, it's possible. I think the $1.5 billion for 2 years is still on the high side. So I don't factor -- I am assuming that the $400 million of that might not repeat. But at one point, one kind of range, I think looking at underlying debt, I think that's doable.

Right. To clarify, is the $1.1 billion the trading income that is in the income statement? The one that is this year, I mean, 2021 is $1.8 billion, and then 2020 was $1.4 billion. So it's -- are you -- the $1.1 billion, are you talking about that line, that is the $1.8 billion and [ $1.5 billion ]?

No.

So the $1.1 billion is the combined T&M income that is shown in the net interest income line as well as the noninterest income line. So you can see that in the segmental report that we disclosed, we have a segmental report that is on the performance summary, and that shows the breakdown of net interest income as well as noninterest income. So for the full year, $1.5 billion, the trading income you see on Page 11 of the performance summary, the $783 million sits under net interest income and $726 million sits under noninterest income.

Right. Okay. Understood. But perhaps maybe I just want to try to get -- if there's any guidance on the $1.8 billion going forward. What would you think is a normalized rate for that number?

Well, I don't think about it that way. So I mean, it's hard to do it because I don't hold treasury to how much do you make on the trading line and how much do you make on the interest income line. For me, it's quite fungible and I leave them to do it. So I don't think of it that way at all.

Okay. Understood. And then maybe just one small follow-up on this trading thing. Is there any difference in a rising rate cycle to the trading income versus in the past 2 years where we have seen lower rate?

Yes. Of course. On a falling interest rate, all treasuries do well. And that's why it benefited last year. But this year, there was no falling interest rates or either they're flat. So the $1.5 billion this year is quite different from last year. Last year, we made money on that, and we made money on the sale of our investment securities, right? You can see that last year, we benefited a lot from that. This year didn't have that. In a rising interest rate environment, treasury tends to do -- be more challenged. Because they don't benefit from that, they've got to make it up as well. So a large part also depends on what happens to the other asset classes, how the equity markets are doing because we have a very active equities and equity derivatives book and so on. So it's a balance.

[indiscernible]

Sorry, you didn't come through.

[indiscernible]

Is that the operator or -- operator, we can't -- we're not hearing you. Hello?

Next question is Neel from CLSA.

Congratulations on a good set of numbers. My CASA and inflation-related questions have already been answered. Piyush, I had -- I don't know if -- this is my question on the November outage. On the cost front, are there any fallbacks or clawbacks from your infrastructure service providers for service parameter deliverables? That was the first question I had. The second is in terms of sector stress, you mentioned, yes, you're waiting for the shoe to fall on the SMEs, and it hasn't happened. But within the SMEs, like are they -- like can you think of 1 or 2 areas which you think have a stress point, whether it's construction or property or -- well, obviously, I think tourism-related might be one of those areas. And how much exposure you'll have to those segments? And the last question I had was actually nothing to do with the results, it's more to do with ESG and green loans. How do you see it in terms of your asset base in 2 to 3 years from now? And I'd love to get your thoughts on how you see the environment evolving to a tipping point where the cost of capital for corporates actually start changing or corporates not really toeing the line on ESG parameters. And then outside of the asset book on green and ESG loans, how does the bank look at its own carbon footprint and reduction and targets over the next 2, 3 years? Thank you.

Okay. I think there's 3 questions. One was the infra providers. Actually, there is no clawback because it's all in-sourced. So we don't use third-party providers for any of the stuff. We do buy the technology from people. So you take the access control technology. There are 2 or 3 big providers in the world who produce this. And we have procured the technology from one of them, a company called ForgeRock. They've got some 1,500 clients around the world, U.S.-based company. But after that, we actually implement and install it ourselves. And our -- the first -- the thing, the first couple of reviews we've done, there's nothing wrong with the technology. The full tolerance of the system, the replication methodology, the redundancy, it all stacks up. So no, we have no anticipation of doing any clawbacks on -- from anybody on that. Your second question on the SMEs, the typical ones which we've all been worried for the last 4, 5 years are construction, F&B, retail, tourism and tourism-related. So these are the sectors which are most likely to be under the sink. But like I said last year, we have tightened, I think, our oversight of these sectors. We've got out of marginal names. We've improved our security position. We've done everything that we needed to, to manage these portfolios quite well. And so we are not seeing any pain over the last couple of months. Even I thought after the moratorium got over, you might start seeing some pain. Now the moratoriums are pretty much done. We had $5 billion of loans under moratorium. That's down to some $300 million, $400 million, down to under 10%. And all the companies that have come off the moratorium, delinquencies have not moved up. So they are holding up quite flat. So those are the sectors that we'll continue to look at, but it looks okay from what I can see right now. The last is on sustainability. We're obviously quite actively pushing the sustainability agenda. We grew $20 billion in sustainability-linked loans in 2021. That's on top of $10 billion the year before that. So we're well on track with our target to get to $50 billion of sustainability financed by 2024, '24, '25, which is what our commitment on the total impact. I think we'll exceed that. So we continue to do a lot of that. But the outside of this, we're also doing a lot of capital markets. We did 50-plus-odd deals on DCM, which are also sustainability or green deals worth $23 billion. So that's quite an active agenda for us. And as we're growing our book, this is continuing to be a disproportionately faster growing segment for us, the sustainable and green financing segment. If you look over the next 2, 3 years, as you know, we, 3 months ago, signed up for the Net-Zero Banking Alliance. So we have a commitment to net zero 2050. I didn't do that for a year because I just wanted to make sure that this is not an ideal 40 years from now, somebody after me will worry about this promise. So we did a lot of work to make sure that we had milestones along the way. So we've got some fairly well-marked out measurement processes. Today, we're measuring carbon emissions for about 3,000 of our companies. That's about 1/3 of our book, and then we work with consultants to model the rest of it. That's on track. We've come up with a transition finance taxonomy so we know what we qualify, what we don't. We are currently actively working with our clients to be able to figure an acceptable set of transition pathways, and those obviously vary by sector. So we use the NGFS. We use IEA. We use IMO for different industries for what acceptable pathways. So we have a very good line of sight for what we need to get done between now and 2030, and that's why we were able to sign up for the net zero commitment. On our own footprint, by end of this year, we said we'll be 0 carbon in our own operations. And we've done a bunch of things for that. We've got solar panels on our buildings in Changi. We try to use our own solar park. We've retrofitted some of our buildings like in Newton to be completely carbon neutral. We've changed the entire fleet from fossil fuel to electric vehicles for our cash in transit and for ATMs. We have dialed down on our paper consumption with a bunch of things. But in fact, after doing all that, because we're in Singapore, we rely on the grid. And so because a grid is a grid, you can't actually make a material impact to your consumption of electricity. So you've got to rely on the offsets. So we have an active program of RECs and offsets to make up the difference so we can get to net zero in our carbon footprint by the end of this year.

Our next question, Krishna from Jefferies. This is our last question.

All the questions have been answered. Just one housekeeping one. On the risk-weighted assets, I saw that the growth both year-on-year as well as half-on-half, they are almost twice that of the loan growth. I was just wondering what factors relating to that, if there is any credit risk migration or there are mostly model changes and we should not be paying too much of attention out of that.

It's mainly loan growth.

But I think the RWA grew faster than the loan growth. Is that correct?

Yes. So it is like RWA, typically, the weighting is 50%. So I'm seeing an RWA growth of 9%, whereas loan growth is also 9%. So I'm just wondering that if there is any change in terms of risk weighting itself.

Krishna, I think we'll have to get back to you. I didn't notice that. So we'll take a look at what might have caused that, and Sok Hui will get back to you.

Yes.

Thank you. That comes to the end of our Q&A session. I shall hand over for closing to Michael.

So thank you for joining us. We'll see you next quarter. Bye.

Okay. Thanks all. Bye.

Ladies and gentlemen, this concludes today's conference call. Thank you for your participation. You may now disconnect.