DBS Group Holdings Ltd
SGX:D05
If you don’t study any companies, you have the same success buying stocks as you do in a poker game if you bet without looking at your cards.
Utilize notes to systematically review your investment decisions. By reflecting on past outcomes, you can discern effective strategies and identify those that underperformed. This continuous feedback loop enables you to adapt and refine your approach, optimizing for future success.
Each note serves as a learning point, offering insights into your decision-making processes. Over time, you'll accumulate a personalized database of knowledge, enhancing your ability to make informed decisions quickly and effectively.
With a comprehensive record of your investment history at your fingertips, you can compare current opportunities against past experiences. This not only bolsters your confidence but also ensures that each decision is grounded in a well-documented rationale.
Do you really want to delete this note?
This action cannot be undone.
| 52 Week Range |
26.9685
43.02
|
| Price Target |
|
We'll email you a reminder when the closing price reaches SGD.
Choose the stock you wish to monitor with a price alert.
This alert will be permanently deleted.
DBS Group Holdings Ltd
You don't have any saved screeners yet
Over to you.
Okay. Hello, everybody who's dialed in, and welcome to our second quarter briefing for the buy and sell sides. I've got Piyush and the rest of the management team with me. And we have got some analysts with us here. So we'll start with them. Anyone wants to start? Nick?
Can I ask you just -- it's a broad question, I guess. I mean, correct me if I'm wrong, obviously, you did 12.5% ROE in the first quarter.
First half.
First half, sorry, yes. The suggestion from [ Softway ] was this remains broadly stable from here. Is that correct? Then the next question is how do you see that -- if NIM is going to continue to go up, and I assume we have more expectation of rate hikes in 2019 beyond what we have here, what gives to keep ROE flat? Why would the ROE continue to [indiscernible].
[indiscernible] go up.
It will go up? So ROE would still go up from there?
We did 12.5%. We did 13 in the first quarter was because of property number, right? And I said that time that 13 is not the real underlying, that SGD 88 million will be in the property sales which should back out. And I said if you'll take that out then we are about 12.5%. And as we go forward, as NIM flows through, the ROE will increase, which is why if I assume guidance then I can see the breakdown to 13% ROEs composition of that. Now if you take 13 and 14 is all the things that will happen, right? Happens on occasion. But should be viewed -- in fact, if you look at the NIM in this thing, in fact, there's a lot more work around -- on the side of the NIM. Our NIM are positive and it actually runs at 2, 3 years. The first year impact of NIM is -- our loan book, x percent of it can be priced immediately. But a large part of it reprices over 1 year, 2 years and 3 years. So the NIM impact actually flows through to year 2 and year 3 as well.
So from one -- whence rates stopped rising, but others still going...
Yes, and keeps going [indiscernible].
Could you explain that a bit more which part of the book you're getting that flow through which...
Sir, the loans which are pegged to SOR or this thing gets repriced right away. Loans which are pegged to SIBOR repriced with a 1-month lag. We price SIBOR based on the first of the month, we price loans based on that. So there is a 1-, 2-month lag based on these things. But then, there are other loans that mortgage based on [ FHR ], fixed deposit rates and loans based on administered rates, and loans based on invested rates, there's loans based on fixed rate which roll over this thing and then we book them in this thing. So if you look at the whole graduation part, 1/3 of our book reprices now and then, over time, in year 2 and year 3, the rest of the book reprices as well.
Can I ask about the rising cost of deposits? So quarter-on-quarter, it will double the same amount as a loan yield. So where are the cost pressures? I know part of it is because of deliberate liquidity buffer which is a good thing to do but, in the region, are you seeing any concerns...
Robert, first of all, it didn't move up as much as loan yields. So if you take the average of the trade and nontrade loans, you've got 18 basis points lift on the loan yields and you pay 13 basis points on the FD, that's how NIM went up. So -- and you have to expect that, first of all, LIBOR is up by 30 to 40 basis points, so that affects then all of the FD pricing. FD pricing pretty much gets driven by LIBOR rates. But other than trade, the trade goes up but we're not being able to price all of it into the trade. And that's why we didn't do the trade book. It might come back later, but right now, people are counting to price trade. And so the higher cost of deposits are not being reflected in that price, which is why I figured no point going the trade book if you can't make a return on it.
The other effect is I think you see that we have actually run down our commercial paper, and we have basically taken in fixed deposits. So while the cost is about the same, in the -- to the performance summary, you see the cost reflected instead of on other borrowings, it goes to customer deposits. And these are market rates and there's a hit -- sort of a greater rate on the overall customer deposit base. And you are seeing some of the effects as well.
So we ran down CPs at SGD 5 billion, and we keep FDs at SGD 10 billion. So that shifts the line in of the [ futures ].
Could you commentate on the quarterly results for Hong Kong? You did the first half versus first half, but quarterly, I think it was lower.
Actually, quarterly results are -- I mean quarter-on-quarter is very strong. But first quarter to second quarter, the same thing. It's somewhat lower partly...
Properties.
We had SGD 88 million property gain in the first quarter which we didn't have in the second quarter. And then the trading issues for us was off the region. So it's showing up in Singapore. It's showing up in South and Southeast Asia. It's showing up in the things. So we had the trading headwinds of cost.
Hong Kong benefited from the sort of HIBOR rate increase. And that was the HIBOR, when you look at the HIBOR rates, I think the movement of market in the second -- I think in the second quarter versus the first quarter. So they benefited because, yes, the deposit base is around 61% of CASA. So that has actually helped the bigger loan base. It is the higher NIM on the higher SIBOR. It's higher wealth management. It's higher cards. So Hong Kong is doing extremely well.
Just last one from me. Your adjustment on Stage 1, Stage 3 was pretty modest. I just wondered how you think about the process if the world gets worse, the outlook -- because you're supposed to be very forward-looking on this. So how do you think about the underlying process?
So I think we -- so PwC has served the benchmark across other institutions that have announced change to this IFRS 9. So we find that consistent with other banks. About 5% of our book is in Stage 2. But most of the [ ECL ], 95% of that is really in Stage 2 in terms of the [ ECL ]. So it's quite consistent. When the loan moves from Stage 1 to Stage 2, we see a significant lift multiple times of the [ ECL ] in Stage 1 which is the intended objective. So that when it goes into Stage 3, you actually release what you already provided in Stage 2. So one good way of testing how well the model works is every time we have an NPL that comes up, how much are you releasing from Stage 2 to Stage 3, because that is the intended effect. And so far, from what we've seen, the models are all working well. On top of that, we basically have what we call an overlay on our models to say what is the effect is in the environment that may not be completely factored in. So we obviously consider the trade war situation as well, and we [ book ] for additional resource effects. Of course, it's a balance. But I would say we've been very conservative in our estimation.
Robert, to supplement this precise reasoning, I think there is so much to -- in our assumption. Your material assumption which is correct. Because you got to first assume what is the outlook in the macro environment. And if you change your assumption by 0.5%, that has a material impact on what you think of at Stage 2. Then you got to make an assumption of how much a portfolio will migrate because when it migrates, you move from point in time credit losses to 2 cycle credit losses. And that -- but all of these are assumption-driven, right? So it's not -- so credit, one thing which I thought was actually quite helpful to us, so we've actually gotten -- got an external provider who's done this thing for many people around the world. What do you call them?
Derisk engine.
Derisk engine. So we've got an engine which actually takes and computes a forward-looking forecast based on a whole range of variables which includes stock market, what the stock market's indicating, all of these things. You just plug the variables into the engine, and then you come up with the engine-driven thing of what is the likely Stage 2 and Stage 3 are going to be. And then as you said, I want to mention that when we came up with this thing, so this engine-driven process, how does it compare with everybody else, other banks in the world? And I think, from the PwC comparison, we are actually at the conservative end. Across the total of this thing, we were the second highest [ PCL ]. From all the banks that showcase me, only Barclays was higher. And Barclays was higher for some reasons of a few [ ECL ] related to the credit card portfolio, I'm not sure why. We were the second highest across the banks that they were able to show us in terms of [ ECL ] on Stage 1 and 2. So it gives me some comfort that if engine-driven tries to take out assumptions, it comes from a GSIB, it comes from assumptions like everybody. Then the engine design project whether it's a third-party developed tool. A lot of banks use it. And then the outcome seemed to be fairly good result.
So in term of the crops. So whatever Trump decides, it's reflected in market prices. Whatever you think interest rates are, it's reflected. So you don't sit around and try and guess interest rates. GDP, I think that's even more inaccurate in all estimations. So basically, you take the result of the crops, you factor in the prices, it translates into PDs. You do a cycle adjustment by sector, by country. So I think, overall, it's quite a good model and less subject to everybody's opinion on where the world is at this point in time.
If I could take a couple of questions. First, on the cost of fund. If you look at LIBOR, that has not really moved in last 6. There's a case in which it's held the same. I just really -- LIBOR is at 2.3 -- average [ gas stop ] December, January. It stayed there at that level. So when you figure out this cost of funds pick-up, is it more a delayed impact of that spike which happened in December, January?
I asked Michael to put out this data for me. And he's showing me that LIBOR, 3 months at the end of first quarter was 1.93%; end of second quarter is 2.24%. That's 41 basis points of lift. So not sure why you'd say it hasn't gone up. And 1 month LIBOR is up from 1.65% to 1.97%, that's 32 basis points.
So it is pure LIBOR straight pass-through.
So HIBOR is up 53 basis points; LIBOR is up 41 basis points; SOR is up 34 basis points; SIBOR is up 23 basis points.
So it is very straight pass-through from the LIBOR and other things. So if LIBOR stabilizes, your cost of funds should not have.
So our market, this thing, our FD rate have been driven by the market. And by the way, the liquidity is, for some strange reason, in there have been mixed noise. There is no problem with dollar liquidity in the region. So that is the underlying thing. We're just not having a problem funding, either FD or CP. We've made the shift between this thing. We're not seeing any real problem with dollar liquidity.
Okay, so it's more pricing rather than your quantum of money.
And with rate hikes, they normally pass through to LIBOR [indiscernible] impact.
So the LIBOR capped that. The OIS LIBOR has gone up to 40, 50 basis points, it's come down to 30. So it's corrected the LIBOR and OIS spread. But net-net, LIBOR has moved up in this period.
Okay. The second thing is on RWAs. That has been pretty well managed. And CET1 at 13.6. How -- one, what did you do there? And second, how do we think about your ideal core CET1 number and any upside for the round-off?
Sok Hui knows what magic she does.
So in the -- I think we optimize, and this was so we actually market risk, so you have to put through the system changes to do netting more effectively. So in the past, we have a long position, short position within a given method. So we actually use the engine to actually be able to match and be able to net off more effectively. So that actually helps in the RWA. So all the numbers have been shown to the [ MES ] and they kind of understand what the changes we are putting through. So that has been -- that is one of the drivers of RWA optimization.
Is there anything more to come there or you...
We always look at where are the efficiencies. A lot of the efficiencies tend to come from looking data more closely to see whether there are things that are not...
We look at the areas where our RWA efficiency was lower than our competitors'. So one thing that stood out of the marketplace and [ then we go to ] understand what it is that other people do and realize that we're not netting the numbers other people have. The other area where we know that our model is beneficial is private banking wealth asset book. It's a fairly large asset book now for us, beyond $1 million. And we're not efficient, but it takes time for the model to...
Get regulatory approval.
Get regulatory approval.
Probably at 2019?
Yes, probably at 2019.
And on the CET1, 13.6% is that at the upper end of your range?
The upper end of our range. We give guidance at 12.5% to 13.5% is our range. And 13.6% is already there and then we have accretion as you know which is at the upper end.
The other reason for what we booked on in the second quarter was on expanding our ANZ portfolio a lot. So when we run the ANZ portfolio, we assumed the worst of the region. And so we basically put capital aside as we gained similarity of portfolio, whether they are similar versus to our portfolio, et cetera. We were able to bring it in line with our own models and that was the other effect you see this quarter.
That's some good data. The market has come to expect your cost to income ratio to go towards 40%. I hear you on the income side. On an expense front, given what you're doing, is 2019 still a fair expectation that, that side of the equation will work?
Our underlying cost in terms of this thing of the 0.5% a year efficiency, it's still on track. If you look at our second quarter, we tracked our operating expense and the [ jaws ] are flat. So we didn't get positive jaws in there. So the big 12% is the ANZ is 7% of that 12%, right, so you've got to figure exclude that, there'll be 5% on that. And then another 5%, 1% is anomaly because of positive commissions and [ payout ] fees and all we have to pay out as part of the deal. So it's not too far off. Now what happens is like, take this year. As I said that when we get there, ANZ will drag us back. We'd get the 0.5% of the ANZ cost-to-income ratio before we can get on with the drag, so it does flow through on this thing. Underlying thing, look at the growth. But it's like we are in the process of doing this thing. Now we might have to make investments next year for India. So the subsidization in India is that comes through in October and November. Then we do have a plan to scale our distribution quite rapidly. And if you scale up distribution rapidly, that will cost some money to do also. That's why I'd like to be, again driving down some of the underlying approval cost.
Is this on the branches side? You have approval for a big number, so what's the plan?
Unlimited. Approval is for unlimited because after it's subsidized, we can do as many as we want. We gave an indication that what we want to do is open about 20 incremental branches and another 100 points of presence in the country, so kiosk kind of thing. But to support the SME business, the cash management business, the [ asset fee ] business. So if we actually accelerate all of that and do that front load, that expansion, then that will cost us.
I have 3 questions. So on the trading side, you said that it's possible that you can go back to steady state at the second -- coming quarters. Do you also think there's a possibility that you might revert some of the losses that you saw in this quarter like unrealized...
Some we will. For some we will...
So that's...
So but when I said [ 250 ], some of them we took -- so we were -- the bonds for our [ first ] spread is massively out of the money because Singapore rates, some of the credit portfolio was -- we didn't sell the portfolio, we just mark to market. So as the credit spreads come down, then obviously there has to be losses. So to that extent, as the rates come back, it will just reverse.
Right. The second question is on the tax rate. So it seems to have been much higher in Q2 versus previous quarters. Is that a one-off or is there something that's changed?
Yes, it was actually a reflection of the T&M. T&M, there are some components of income. There are tax that are favorable 5%. So you have a loss. It changes your mix, so you have more of the -- your income tax at the 12% to 17%. So the losses that you get are at 5%.
Do you think some of the past quarters when trading income was very weak, we didn't see that kind of same...
That's why I said it's a component of the...
But the traders are not very weak. If we look at the treasury market, we're seeing $50 million loss in treasury markets. And we haven't seen that evolve.
So some are for hedging. But for hedging, if you -- so this category of preferential types of price to 1 type of asset class. So even if it's quite flat, you're taking the gross amount. That's what explains the higher tax rates, and it's related to the treasury income.
Third question is on the wealth management income. So that fell from 331 to 300. Obviously, Q1 was probably a super strong quarter as it was expected. Do you see further downside of that number in the coming quarters because to the extent that trading income has declined similar to past...
So it's one of those deals, this is uncertain. The third quarter has started quite okay. We're not seeing it fall off from the second quarter. But if there's a complete uncertainty around this thing, then I know it will come off a bit because if the equity markets are soft then wealth business suffers. So people, investors often won't go into equity markets. So unclear really on this kind of thing. But the third quarter started off okay for equity markets.
So are -- we're added.
May I add in, Piyush?
Yes, really...
Yes, so Piyush is right. The good news is when the markets are bad, the RMs have some capacity to do a lot more strategic long-term stuff like get more clients, get more sticky AUMs and do more wealth planning. And when you do the wealth planning, you actually do it for long term, and it's very, very sticky. And that's what we're seeing in July of [indiscernible] for July, but what we're seeing is, going forward, there is this that strategic long-term tilt towards getting a clear long-term plan, which are also actually fitting and recurring. And I like the way that that's moving. That doesn't give us that short-term lift that goes up with the beta of markets, but it gives us longer term alpha. The second is it depends on [ volume ] as well, right. So if the volatility of the market continues to go up, then we will see opportunistic ways in which to make money for our clients in these markets. So it really is -- so it does depend. But the key for wealth management growth is sustainable long-term growth, and that comes from continuing to grow your customer acquisition, continue to deepen and widen the relationships. That continues the pace.
Q-o-Q AUM growth is quite strong. Does it mean that net new money was very strong in 2Q because it cannot be driven by market in...
Yes, actually on the market...
[indiscernible]
The mark to market hurt us right, especially for Asians.
Can you give us some color on kind of is the net new money kind of pro-activating kind of some numbers or percentages?
Okay. So we are up 3% for high net worth on net new money.
That's Y-o-Y?
Quarter-on-quarter, yes.
Quarter-on-quarter. On a Y-o-Y basis?
I'm not sure of the Y-o-Y.
I can -- I'll get you the Y-on-Y numbers, it's quite good. But to give you more color on the acquisition. Yes, we are focusing, I mean, the RS scorecards are focused on getting net new money, and both organic and new acquisitions, new client acquisitions, and we're segmenting better, right? We're segmenting both the higher end and the lower end better. The segmenting of the higher end is really sector-based, creating sticky relationships across a one-bank solution. And the lower end is really using digital tool and getting our customers engaged onto our digital tools, so that creates stickiness and more engagement and hopefully a virtuous cycle of usage, transaction and more acquisition.
In terms of the net new money, again, is it mainly local money or are you getting mainly from Asia or are you seeing the West side bringing money?
So it's across the board. We are seeing some local increments within Singapore context. We're also seeing it from Greater China and also from the international local NRI space as well.
[ Hence, the ] question on funding. Looking at your Sing dollar deposit, that actually came down quite a bit in second quarter. Just wanted to understand if you have any color. And also in terms of your liquidity buffer that you have kind of built, it -- do you see that as sufficient as we go throughout the second half?
Actually, in the [indiscernible] segment, we have too much liquidity buffer as it relates to it. We loaded up SGD 5 billion of commercial paper we replaced then we load up another SGD 5 billion, so we got SGD 10 billion. And if the loan book doesn't grow or we don't -- if we can't price for trade, then we might actually have too much money I think, or not. So then it goes back to if the trade book starts repricing and we keep -- can grow it, then we'll have to come to be in the deposit game. If the trade book doesn't reprice, then frankly, we might actually not need to accumulate anymore I think in our buffer. On the first, this thing in Singapore, there's some -- there's $5 billion in total Singapore deposits, of which the bulk of it is about SGD 1 billion or so in fixed deposits, SGD 1 billion in savings, but SGD 3 billion is current account. The current account is just down to SGD 3 billion. We had a lot of money at the end of the first quarter in corporate current accounts. Some of it is sold out. So it was not unusual, this thing. The savings account, that -- of the whole system deposit in Singapore are down for the quarter. And I think there are 2 impacts on that. One impact is I think the savings bonds, Singapore savings bonds, that set down some of the money. There are 2 tranches of that in the quarter. And because the reason the savings bond is quite attractive, I think that took out of some of the money from the system. And the second is there is some conversion. So I think people are converting out of Sing dollar to U.S. which always happens when the Sing dollar is weakening and the U.S. dollar is strengthening. There will be some conversion from one category to the other. But our market share as a whole is flat. So even the rate on market share numbers from this thing our market shares are flat also.
I have a couple of questions. One is employee count. I think this time around, it was the first time in a long, long time when the employee account just declined a little bit quarter-on-quarter. Is that just something that we should ignore or is that a new -- and the second one is you said that trading -- sorry, the trade book is not repricing yet, right? What do you think is the reason behind it?
It's very hard to call. Actually, I noticed [ Bill Winters ] and his team day before also had commented on the [ side ] that trade margins are tight and trade pricing is an issue. I'm seeing the same thing. So as my guys they want to do trade finance at level which I think is absurd. But it's almost like I don't want this business. So it's not just -- this is why I said this dollar liquidity crisis seems to do. So people have been moving, able to fund dollars and then put it out to trade this thing at 20 basis point margins. I'm sorry, doing business at 20 basis points margin is not worth it, but obviously, some people are happy to do 20 basis point margins. It's not clear to me whether people are still you have money and you're doing it for the time being, and it takes time to price up or not, but...
Few questions. Firstly, I mean, you mentioned that structurally, on the treasury market side, you potentially could see income levels decline. And you mentioned that your cost-to-income ratio, you got 44% in this quarter. When you adjust for that, it's actually 43%. I guess the question is whether have you actually responded to that -- or are you looking to respond to that by lowering your cost going forward? If you there is a structural decline and that could result in savings in cost going forward?
No, there's no right way to think, right? I -- we think -- I think you're off about $100 million a year [indiscernible]. The room between the trading and the whole customer mix is SGD 2.2 billion. So that's maybe a 5% structural reduction in revenues we will talk about. And at the same time, our customer revenues are kicking it again. So for the first time since 2015, our customer revenues grew at the same time. So it's not a figure that you've got to fundamentally negate. Can you tighten the margin? You can. And the tightening of the margin really comes from digitizing the process. So as you digitize some parts of the process, then you get expense efficiencies which is right, that's what we do.
Sorry, just a question on [indiscernible] just if you look at CET1. It seems that the RWA optimization can still continue I think to next year. And looking at 13.6%, if this continues, could there be more potential for a second dividend?
I think it also depends on the FX outlook because as dollar strengthens, you have also RWA impact, but if some of this optimization that we talked about, approval of internal models go well, so potentially, the CAR ratio could be lifted, and we'll continue to see how it might guide our dividend policy because we intend to operate more at the 12.5% to 13.5% range.
Just some housekeeping points. How much of your -- so this is for property related. How much of your housing loans is investment? How much of -- what's the rough split in your building construction between developers lending to your company like REITs and then other stuff? And then on your other consumers, the $31 billion, how much is property related, so home equity type lending as opposed to auto loans or other small consumer loans?
You take the loans first then we'll take the other. The first one again?
So how much of your -- your mortgage [ property ] is SGD 74 billion, how much is investment property?
So we are -- 86% of all our loan are homeowners, and that's 81% of the portfolio.
And then the building construction, how will I split that SGD 70 billion -- sorry the...
The -- the -- I don't know land stats. Sorry do you have that?
60% to 70% of these are for investment -- are for income generation.
So the rest is roughly -- it's -- and then on the professional, private, the 31.3...
It has almost nothing to do with home equity. So we might have an additional collateral on home with this thing, but a lot of that has to do with margin and investment management.
It's more the PD side.
Your guidance on not expecting further write-backs in O&M in the second half, it's just being conservative or this is one -- there's not a start of a new trend, you're seeing prices picking up, you're getting things out. Is it possible you might see more write-backs in the second half?
Yes, but the one that -- this is one ship. The second quarter was one ship, which is among these things. But by and large, I would think that while utilization in this sector is picking up a bit, margins are not picking up. So I'm not seeing any improvement in the cash flow. And so frankly, a lot of people are going in as investors into the sector and so on. I think my personal view, are getting in ahead of the deal, because we're not necessarily seeing a lot of pickup in the underlying cash flow of the business. In terms of collateral disposition, the -- we've said this comes to you when you've written down the collateral. So it is possible as ships gets sold, you're getting a little bit there. But most of the ships that we have are large numbers that are relatively low value ships. The ships, people are buying ships for SGD 8 million, SGD 10 million, SGD 20 million. We might get, but it won't add up to a lot. It might be SGD 5 million, SGD 10 million, SGD 15 million, but it won't get this big SGD 60 million, SGD 70 million.
So in that sense, this big ship was a one-off inside...
For this year. There are more opportunities, but this year, this is a one-off case.
If I think about the NPA formation settlement blend, this quarter, more, the NPA formation went up. I guess there was one particular large account and also the settlement went up, which I guess is that ship.
No, also in India we have settlements.
Okay. If I'm thinking about the breakup of that NPA movement, should we revert back to lower number on both of these, or how should we think about that? In net-net, I know it is kind of flat, but where is it coming from? Is there any risk, I'm trying to figure out, being part of the portfolio?
[indiscernible] can answer. The thing is that the NPA formation sometimes is episodic. It's not a -- on an average, yes, I don't see a lot of these things. But if you look at the cases, which are not NPA but I call big cases, there's cases in there if one of them tips over then you see an NPA pick up by a dozen. And the recovery, repayment and the thing are the same thing. There are bunch of cases that's something that we're going to see a meaningful movement on there.
But there is nothing on the horizon either which way which should [ roll ] the number any...
[ Another thing ] is consumer banking tends to create the noise. So consumer banking normally has a large amount for like this quarter, it's like $88 million that's going to NPA because they've crossed 90 days, and then it will be repaid. It's like half. Then they get tend to be repaid. So you find that the repayment are, you get another $88 million. So it's just sort of inflates both sides. So that is one reason why the numbers seems big. So we are considering how else to present the data so that you get maybe real sort of NPA formation that...
[ Yes, it's a good point. And because this amount of the India portfolio. ] India was not that material. Now when we add in the ANZ card in Indonesia, Taiwan, et cetera, et cetera, [ so you have these. ] But NPA, blended, something, [ 20 -- ] 82-something came out of NPA. We showed it on both sides.
Okay, I'll -- Serena, we'll take questions from those that are dialing in.
[Operator Instructions]
Okay. While we wait...
There are no questions in queue.
Okay.
Just curious, is the trade margins going to improve? There's a lot of excess liquidity. How fast can you cut off your expensive, deposits? Like what can be the -- now deposit cost is at 87 basis points. What do you think basis points are? Can you bring down a deposit cost by end of...
A lot of the NPAs are quite short term, 1 month, 3 months. You can [ let it go for a while ], but I can't give you how much the deposit cost achieved. But the -- it's been distinct [ in terms of prices ]. It's not [ material ].
Are you not thinking of letting this [ NPs ] runoff already? Or are you still going to keep the deposit -- I mean, the liquidity debt in the hope to grow loans?
We're taking stock of -- actually, they're not that much incremental liquidity, about a couple of billion dollars in excess right now that we have. So all the -- our asset liability committee meeting tomorrow, this afternoon or tomorrow. [ There's talk of a trailing quarterly ].
It's not like we're losing money because they don't make a lot of money. And therefore, overall, if your asset book is inflated, then it just drags down and then there's a...
Just on the RWA optimizations on the market risk side. What you did -- do you think that engine -- is that like -- are you the first one to do it? Or do you think that's fairly standard practice that a lot of the other banks have also done it in the past?
The netting?
Yes, the netting of...
No, we copied other banks [indiscernible]...
Piyush, thinking about the dividend, there are 2 things. One is this range to around 3.5. And second is the sustainability of absolute dividend. How do you trade this off? I'm trying to figure out that, let's say, even if you believe that maybe you'll end up the year at let's a high 13s, but when you're considering dividend per share increase, you have to also think and you keep it at a minimum through thick and thin. How do you trade this off? And I mean, if I'm trying to think about [ bill ] increase or a dividend per share increase [ for the profitability ], how do I...
So this is -- I don't have a formula for it. It is -- it's a little bit judgmental to sort of balance this thing out. So I look at it and said, yes -- I said 120 could be a 420 that -- I say now, we're just down 120. It's a big hike. We went from 60 to 120. There's no reason to use it again for the next year, we can wait and see. It is that kind of thinking. It's a judgmental call. I think it will already -- I'd probably be at 13.1 or something like that, we have 13.6 because of optimization. You already have a [indiscernible].
Yes, 15 basis points higher. So that's what it is.
At this point in time. But the point also, I want to be consistent. So I paid out and then I scramble so I'm okay, we will -- we'll quite [indiscernible] to make sure that we have a consistent dividend policy. As long as we can afford to keep growing the dividend, we keep growing the dividend. And we have a range of capital that we think is appropriate. So within that, that will balance it out a bit, but there's no formula.
What I'm trying to think is rather than doing dividend, would you then say, okay, let me do some bolt-on acquisition or add something or some bits perhaps?
The idea that I could buy something just because I have the money, [ it won't be asked -- handled -- disappear ]. I mean, I have run the bank for 10 years, I haven't done this. I'm not going to start now. So...
I think there was a comment that a company can go over 20%. So where is the account coming from and is it onshore or offshore, that's one. And on the loan pricing for nontrade corporate loans, are you able to include the spread or is there a diversity that came from the corporate lending?
So the -- [ Sok Hui ], the results.
Okay. So for wealth management, the headcount -- a lot of that was the ANZ acquisition, and particularly ANZ has a private bank in Taiwan and Indonesia, which we obviously bought over. We've rebranded it as Treasures Private Client, and we refresh and we launched it actually in the last month. And actually, we also have started to refresh the value proposition with this Treasures Private Client onshore product proposition, value proposition and priorities and process, et cetera. So part of the trend that we've seen in wealth management is that some of the, Asian, Pan-Asian wealth will start to go back onshore as well. So this speaks well to that longer term strategic move. So the headcount, a lot of it is front office. It's front office, it's salespeople, it's RMs and it's investment councils. There's also some elementals of the back.
But yes, I think -- we're adding in Hong Kong for [ market share ] because that's doing well, and so we keep adding this thing. We're adding in the market here. We're adding in the international side. We're adding in the domestic Indonesia and Taiwan. The good thing with this is that RM piece for -- our sales for RMs [ sell ] very rapidly. So right now, we are modeling RM. It's just the cost on RM comes through. Our RM productivity is very high. So we make $3 million in RM. And as long as it can do that, we don't build the RM anywhere near like that. So it's actually a smart thing to do. So if you can bring the RM integrate, it's just that [ people that do it their measured best ], you don't get out of control.
And on the corporate?
So it's mix. So in some cases, there is -- we're not able to pass on the [ deposits ]. I think it's moving up, but it's not like trade where it's not moving up at all. It's moving up. In some places, there are some pricing [ gap pressure but ] less to do with the [ optical ] quality, it has to do a security. So property-related loans they're harder to move the pricing up. So it's taking some margin compression in that segment. But other parts of these things we've been able to price up. But overall, by and large, it remains [ rolling ].
All right. Relating to your reduction in your guidance for the loan growth, do you think mainly coming from Greater China that you're expecting this? And secondly, what is the -- your comment on the credit risk for this trade loans? I mean, this thing launched in China?
Right now the bulk of our trade book is buying back the Chinese policy banks and so on. Those banks I'm not concerned about in terms of financing. And then last part is the big traders, the commodities are going to China and so on. So there's no credit issue with that book. But yes, the marginal trade that we do tends to be China-centric trade, and it's on both sides. On the commodity side, it's on the manufacturing side, everything. And commodity, you can see the volumes are not coming up. [ So having adjusted to the pricing, it is a challenge ].
A question on digital banking. Early days in Indonesia right now, but if you have to compare and contrast, India have been there for a while, Indonesia has gotten up. But the competitive intensity seems to be very higher in India than in Indo on the digital bank, particularly in the Zip Code where you are in. 2, 3 years down the line, do you think you are able to make way more profitable growth in Indo versus India? And how do you think so?
And the reason for that, actually is I should start with the margin at Indonesia, much better than in India. So I think 3.5% or something on deposits in Indonesia. I make 6.5% on this since I make 3% spread on any deposit I bring in. In India, I used to make a lot. Now I make about 60 basis points. So I make 3% here and 60 basis points here on the deposit. So in India, there's an effort to make it up on the mutual fund and selling in the loan, but it's starting with 60. Indonesia is making 3%, and then I'm making on top of that. What do I make? So it's likely to be a more profitable business in Indonesia as well.
And in terms of scalability?
The numbers in India are a lot -- actually we've learned from India. So in India, we went and said let's acquire anybody. So we were less targeted about that customer selection. And so we've got 2 million customers, of which 700,000, 800,000 are full savings. We've got some customers who are life savings and whatever. And we're finding that the curve in India is actually quite the same. The usual. It's [ 30 crores, 20 crores ]. The 20% of the customers are really profitable. And as you go down the curve, the other customers who were -- the second segment in customers are not profitable now, but we think we can make it profitable. But there is a tail end, half the customers, I don't think we'll be able to make money now. We brought them onboard, but we might use that data. And so right now it's going to be very hard, I think, to make money on that. So I think one, cutting of those customers 6 months ago, and we took out 200,000 of customers, 300,000. And then we've really grown, but we still have this tail end of customers. But learning from that, in Indonesia, we're not going broad-based segments. So we're doing a lot more segmented about customer acquisition. And so we're acquiring fewer customers and the customers we're acquiring a lot more upfront, the profitable kind of customers.
As you think about longer-term credit cost, especially given that you are expecting risks to continue to rise. I mean, at one point, do you expect to see that having some kind of strain on credit cost?
So the -- if you go back to the [ real book spread ], a large part of our exposure is the large corporate exposure. I think they're not that sensitive to a higher interest rate [ testing ]. They have the capacity, they don't want [the line those things ]. The mortgage book I said that it's not fundamentally susceptible. If you look at our 30 and 90 days, that includes sales on mortgages for this quarter. They're lower than they were before, right. So the multiple -- we because multiple [indiscernible]. Then there is the SME and unsecured book. Those are susceptible to higher interest rate. The SME book is the -- [ inventory are inching up under the thing ]. So you will see some flow through into higher NPAs and credit cost over the next year if you ask me. But even that, it's a little segment's [ perspective ]. So -- and in all of those cases, we are heavily secured. [ Learn our lesson, not an unsecured book ]. It's a heavily secured book. So hopefully, we won't have to provide a lot against this but they can see some [ very interesting numbers ]. The unsecured book, you will see some. [indiscernible] unsecured book because of the -- they're add-ons in Indonesia and Taiwan, et cetera. I think there, you would see a pick-up in credit cost as it's under...
So we probably should expect the 20 basis points to trend towards the [ 5% ].
I've been maintaining. They should expect us -- Our long-term average is an average for a reason. We should expect to see 25, 27-odd basis points.
Serena, we'll take the question.
Our first question from Goldman Sachs, [indiscernible].
Just a medium-term question around the virtual banking or different type developments there in Hong Kong. Do you see any impact over the next year or so on the payments or credit card space given that new entrants could come in there, and we have a faster payment, which is also being offered by the industry?
Yes. So I think there's likely to be an impact with all of the virtual payment. But I think in some ways, that might even be good news for us because remember that the big players in Hong Kong should worry as there's large incumbents. They're sitting on the big market share. And so when you have [ Tier 4 ] players, who got like 90% of market share in the market, this incumbent is most likely to be dislocated and displaced. So we fancy ourselves as an attacker brand in Hong Kong, much like the rest of the virtual banks and the fintechs. And so if anything, we might be able to improve our position.
Do you see this flow through to credit cards in any way? In other words, do you have to give back more to consumers in their credit cards?
I think we've always seen that. If you look at the Hong Kong market, the shift from first credit cards to unsecured loans and debt consol has been happening systematically for the last 5 years. So the bulk of the money people make on cards, which is the revolve, that revolve rate has been coming down and the bulk of it is going to a lower earning debt consolidated portfolio. The second piece, which is, in our view, wind up, getting lower MDR as we get more electronic payments. You might have a little bit of that. But contrary to most people's belief that total [ selective MDR ] is not a big source of earnings for the banking system. So if we take a look at -- you take 150 basis points on MDR or something. And then after you [ export the revolve ] cost that you've paid the consumer for using the card and so on, and then you pay off what we paid, Visa and Mastercard for [indiscernible] fees and see how much you actually make. It's not massive. So if that comes down by 5, 10 basis points in volume, it matters. If it's in the few millions of dollars, it's not material.
Piyush, if you take spend as an incrementally from here for next 12 to 18 months, what are the key things where you think you are either behind the curve or it's something that you need to...
I [ already covered it ] last time on data.
So where -- yes, so data lake, that what you talked about, where you are on that and how much of incremental cost should we think about?
Actually, a group of data is a piece of [indiscernible]. It's quite hard. It's even harder than I thought than it was. So it's still struggling to -- but it's not just the technology. The thing with data is that there's a lot of things that show up [ as if a good idea ], but the data governance process, the data cleanup process, getting teams of people to actually migrate the data. The physical migration of data from old systems into a data lake is something we've got to figure or understand back the data, so we've got to employ people to do that, we've got to figure the tools to migrate, you got to check it, and this is going to take us 2 or 3 years to get done [ in the thing ]. So we're actually building some of our expenses that you're adding capacity and people to actually doing this data work around.
So as an -- what does it mean in terms of additional expense? Or is it just that...
No, I said [ guess what ], I won't spend more. But the point, the expenses heightened after I add a team of people to do the data migration, the data management and the data lake.
And despite that, you are still comfortable with the cost income 50 bps?
Of course I try -- I guess I'm going to spend I think an incremental $30-odd million, right, on this whole data thing this year, it might go up to another $10-odd million, $40 million. It's not massive. But yes, it is an expense. [ That's where I'm really spending on so much ].
On the call center, how much are you funding on the [indiscernible]? And the second thing is how much [indiscernible] kind of a network? Is there any [ possibility ] to show kind of a bit more?
I don't have the exact number for [ Service Goods ]. I mean, I don't -- we can [ publicize that this context ]. It's hard to actually -- listen, you have to come up with a guesstimate number because we have a service security control center, that's an obvious one. You can throw the cost of that in there. And then we have this peripheral technologies and so on, the exact technology, but there's a whole bunch of other stuff we do in terms of the data, service and so on. So we don't pull it together, but we can take a stab in seeing what we're spending in there. In terms of branch closures, I don't think we can go much faster. So it's kind of -- so we got to be conscious of the sociopolitical reality of what we can do and face it. So every time you close a branch and then we do this thing, there's a lot of noise in this, mindful of being able to do it in a way that doesn't wind up creating a lot of pushback.
So this is like a repeat question. So you said the short term and the things are a bit cloudy [ in terms of what's going on ]. So over the medium term, what kind of the ROE range that [ it will work ]? I mean, you can see some of that's achievable...
So we gave the guidance in November, which we've repeated in this thing is that line of sight for 13% ROE, we think we have. And it's part of -- well, rate increase, right. So we think between -- we think 12.5% [indiscernible] especially this thing. And then the NIM increase and some of the cost income ratio, this thing, it keeps [ going ] through. It depends on how quickly you can get a cost income ratio up, you could get from 13% to 14%. But that's more uncertainty. The rates and distinction give us a line of sight to 13%.
There's a property slowdown, health, wealth management anyway and people [ have to put the money ] in somewhere, right? So [indiscernible]
[ Some of these things ]. [indiscernible] budget?
No? Sorry, what was your question?
The question was the property slowdown health, wealth management because people can't put money in property so they have to move to financial market.
So I think the wealthy people who live here and potentially rethinking, well, is property a long-term investment asset class, right, beyond living [ in an environment ] for your family, et cetera. The alternative has been to go into the REITs, and we have seen that. We'll have -- because the REIT gives you unlevered 6.2% yield. It's not taxed and it's liquid. If you're wrong, you can take -- cut your losses. If you're right, you can write your profits, cut -- take your profits any time. If you lever that up, you can get 9% to 12% yield. So there has been a shift of kind of what would have gone into a hard asset go into REIT fund. So we're seeing that as a trend, and that continues to pay.
It's a good question. The last time when they had a tightening action for the first couple of quarters, did you see a pickup in those activity because of the tightening action of [indiscernible]?
No, actually, people do compartmentalized their liquid and the illiquid, right. Because in the illiquid, don't forget, you can borrow kind of a lot against the hard asset. But what we have seen with the high net worth Asian clients, for example, they have actually shifted to commercial real estate. For example, in London, they're actually buying using properties because sterling's gone down quite a lot, right? So we're seeing that as one interesting shift, particularly with the greater Chinese wealth creators. And the other is, as I said, into REITs. Within the Singapore context, I think the on-block will create still 2 to 3 per on-block flat, right? You will still see that later in demand that won't go away, but I think they might buy smaller flats.
And sorry, relating to [indiscernible] assets [indiscernible] [ 120% ] of APE commercial property losses. Is that right?
Sorry?
[ On 20% ] APE in commercial property investment.
Where did you see that?
In -- on Facebook.
It's fake news or something...
Or phishing.
Yes, be careful. A lot of phishing going on.
Somebody's sending out SMS in the last 2 days from DBS Bank would say that we produced one software to make you an instant millionaire, [ just click ]. There's a lot of [ fake news ] [indiscernible]. I keep getting -- [ people keep sending me ] 2, 3 of these sites every day.
Okay, very good. Thank you.
All right, thanks everybody.
Thank you.
See you next quarter.