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
Thank you. Over to you.
So hello, everybody, and welcome to our second quarter session for the buy and sell side. We'll go straight into Q&A. I'm going to ask those in the room. Does anybody? Yes.
I'll just ask -- I just wanted to clarify or make sure that I get it right what you're saying on loan MAS charge.
So [the other split here is ] Obviously, below the cycle average loan discharged, this looks very cheap.
And then as we go forward, you're obviously, talking about -- there's a possibility you'll get some recoveries. So you're saying over the next 2 or 3 years, you can also keep loan MAS charge below the -- through the cycle average.
Actually we've got 2 or 3 things happening here. One, we were very conservative in provisioning. And so we really built up some fairly strong buffers all around, including this quarter by the way. We took some large charges in our general provision just because of the environment. And so we did some management overlays to buffer up. So our provision has been actually been very conservative. The second thing is that our past provision was also conservative, particularly in the offshore marine sector. I think the situation, like you said, will be structured, they are performing okay. So I think over the next year or 2, there will be opportunities to take some recoveries there. But I think the important thing is that, I think, structurally we made some meaningful improvements in our risk management capability as well. So we feel a lot more confident today that we have our hands around our portfolio and our processes are a lot more robust than they were 5 years ago. So I think in a structured way as well, related to minor provisions it's actually better than it used to be. So there is some cushioning from the past, but I think, on a [ going- concern ] basis as well , I think, we have some upside.
Through the cycle it's 25 basis points?
On SP's we've given 25, 27 basis points guidance in the past and yes...
[indiscernible] and risk management changes are they more consumer linked or are they more SME, corporate?
Mostly corporate. If you look at the bulk of our credit issues over the last decade, they have been corporate issues. And what we've done is the last 2, 3 years it's hiked up a lot all around corporate credit management. We got much tighter in terms of our concentration -- object of concentration remains, we've got a lot more diversified, we have far more robust, early warning signals. We actually managed and balanced our portfolios. So for example in the first half of the year, we've sold down about $2 billion of loans. We know we should do that. So we now actually anticipate as we sold [$1 billion ] in the second quarter alone. So we actually get ahead of the curve, we rebalanced the portfolio, we reduced our exposure to individual names. So it has a lot of different things.
That's what you did -- that sale because you were worried about these names or because these were hitting, let's say, x percent of your...
Hitting x percent. So not even...
But we started -- tightened up on how much exposure cap we want. And so we used different businesses so maybe we should participate or something or sell down something. So we're a lot more active in managing our portfolio than we used to be.
Just one more question. Just want to understand your NIM guidance a bit better. Asset yields were down, understood. You've paid back about SGD 3 billion of revenue to quarter, rate SGD 7 million on CP's ending the quarter. How much more can you do that then? Or if that setting the limits and the entire asset yield commission comes down. As in, if I want to just to break that up in next 3-, 6-months or further?
We can do more. So our total CP borrowing is about SGD 16 billion, SGD 17 billion. We've checked, there is enough market appetite for us to -- there's another SGD 10 billion in the commercial paper market. So it's a question of trade off. How much more -- so the fixed deposit we've paid off were high-cost deposits which come from some segments of the market. So they did a payoff. Then you've got to then make the tradeoff between, if there is a core customer deposit, do you want to pay it off or not. So you calibrate that a little more carefully. But in terms of the immediate to issue commercial paper, [we have a lot of appetite] . We could do a lot more. In terms of the asset yield side of the thing, I think, if rates come off and obviously, pricing will come off a little bit. I do think that there's still that -- some part of the fixed rate loan book and mortgage that's going to be pricing. The tail end of the book so you'll get some lift from that. But you'll also start seeing some drops from SIBOR and HIBOR link zones [in these]. But unclear what happens to HIBOR. HIBOR, has been extraordinarily strong. And given the uncertainties in Hong Kong, it's not entirely clear to me whether HIBOR stays strong, doesn't stay strong. I'd say it's tough to call what happens to HIBOR more than anything else.
Sir, just a couple of housekeeping type questions again. So first of all on the loans, just looking at the year-to-date break down, obviously congrats you're doing much better than the system right now. But you've up growth here in others and you've got growth in professional and private individuals. Those together is about SGD 4 billion. Just want to know what's driving that? Those are not usual patterns that I look for that [indiscernible]
Well, so -- maybe Sok Hui.
So you're looking at Page 18 -- 16, the performance summary. You have a question on the industry cuts and the others? That's your question?
Others and also professional and private individuals in terms of year-to-date from June to December.
So June -- so year-to-date, yes...
We have about SGD 5 odd billion for the book and SGD 2 billion in others, SGD 2 billion in professional and private, not the usual categories that I would look.
So the others are mainly sort of Singapore, step up on the sort of [one-off ] category in the others and where we have actually seen growth. So that's a safe category. Professionals and private individuals have given more financing. Maybe you have heard about the global income bond we launched in the first quarter, so that's been quite attractive. And therefore, on the back of that, some additional financing for that product.
So basically, leverage on private bank [ income ]. So Wealth Management, private banking leverage loans go into that category. And [stack board and government-linked goals is with the other] category.
And so [ on the mortgage ], could you give some guidance on the SME. What are you seeing a pickup in? Because I'm assuming it's primary then secondary still pretty soft that we measure.
Actually, they say that it's soft. Refinance and building under construction has worked and retail picked up relative to first quarter. But still hasn't picked up as much. But do you see it has picked up with our launches and all the refinance actually picked up again.
Sir, and on NPLs, again just on slightly granular, I know there was some question on this earlier but [the movement] in Southeast Asia, the movement in --
The big NPLs are only in South and Southeast Asia, actually.
Sir, just in --okay-- just in Indonesia in particular, there's also some obvious issues.
No, no.
No.
In 1 case, It's a government company, single company. It's government-backed and we've got a strong network support from the government so don't expect it -- in the way, I took some provisions just because [indiscernible] but we don't expect to go under. But it's a large government-linked company related to the NPL.
But how much [probability would you take] on that percentage?
Very small. About 20%
20%.
That's it?
This poll for quarter end left us confident in this.
Yes, but then neighboring countries over there are all comfortably expanded in trade before.
It is not like a guarantee, this level of comfort. And this is -- still I'm assuming. So it has -- but under IFRS 9, when it wasn't all that's required, it's actually nothing. In fact, it's the nature of collateral and the nature of support. So how much you rely on the government. The total exposure to the system in Indonesia is SGD 2.5 billion. They are like SGD 50 million.
That's why I'm -- I was assuming maybe you'd take.
The rest of the system hasn't been doing much in NPL, and they haven't marked any provision on it, 0. So from our standpoint, given the strong government support, the fact it's actively being restructured right now and we restructure an acceptable promotional terms from what we can see. We probably don't have to take provisions on it so we might as well take something just in case. I won't expect it to go on.
Can you talk about your stress tests that you talked about earlier? I mean actively Hong Kong and China looks risky, but you said you're quite clear on your stress test there's no issues.
Well, we've been -- this time we've been very proactive about addressing weaknesses. So we've been doing this obviously now for 6 to 8 months. So we tested the entire tech supply chain. There are some 400, 500 companies altogether who form part of the tech supply chain in the whole system. And so we tested what happens if tariffs go up to whatever 25%, 50%. We tested what happens the revenue [indiscernible] and so 7. We tested the whole scheme and then we took off from that, which companies are likely to continue to exhibit more stress on stream when the situation happens. And then we proactively look at ring fencing exposure, working down exposure, et cetera, for those kinds of companies. So I'd say we've been quite active about rebalancing our portfolio for the last 3 quarters. That's partly a reflection of all the stress testing and portfolio actions that our portfolio deals that we've done.
[indiscernible] in there then we have option in that [indiscernible]. Is it just the warranty from using the index [indiscernible] in Indonesia [indiscernible]
We also had a problem in the last 2 years for which we have put into then [ 300 and 400 ]. At this point in time, I don't think we have anything else.
You mentioned that you're committing exposure in terms of your underwriting [ you have ] on the corporate side. Will that translate to RWA, kind of lower RWAs and kind of how soon that...
Actually, it has. If you look at our ES, ECL of the thing, including last quarter, this quarter as well, what's happening is that our overall RWAs and the CCL improved because we are reducing our exposure into weaker names and the order of quality of our book is improving. And so then we are actually putting orderly management positions to top it up right, because of weakness.
Can I ask you to speak up? Because they're recording this for the transcript.
So just to clarify, under the new IFRS 9 you start building up provisions before they go bad. So when loans do become bad and you see sort of a flow-through, through NPL, you actually release your GP because you have set some of these effects for specific means that have not gone bad previously. So on our provisioning methodology we have restructured. So that's one reason why GP will go down. The other reason, which we explained the last quarter, is we're actually managing up some of these names, which have not reached NPL, just if you think they are sort of weak, so we manage that up. So if you see the SGD 58 million charge for this quarter, you would have to understand that we have taken much more in terms of buffering. Because otherwise with all these sort of write-backs in GPs, it actually offsets it together with a SGD 58 million charge. So the general provision allows us that we have set aside for this economic uncertainty is actually quite large.
I'm asking, just in terms of guidance, I guess stuff has really come off in some of your corporate loan book price itself. But is that one bit that you're talking -- think you perhaps that affecting moves in terms of the mortgage for, I mean for advance that is competitive in the end, that has kind of pick up in terms of your linear pricing for the back book and front book. What is the difference in price and how fast can you see it help you in terms of [indiscernible].
In terms of the front book, we compete tactically. So we don't drop prices with it. We are the only game in town. So we don't specifically take a reduction in prices, so we compete tactically in some segments of the market. But we try to hold pricing up. So so far, we've been able to hold market shares. Means that you know the front book, so which means that we've been able to actually compete effectively without dropping a lot of pricing. Our overall portfolio NIM, [indiscernible] are holding up. They've been going up and they're working up. What happens going forward is hard to say. If people start dropping pricing dramatically then of course we have to compete notebook value. So my forecast is based on our best-case guess for what the market exactly do for the next couple of quarters.
I have a question, follow up question on the ECL mentioned earlier. So far we have not seen a decline in the [ NPL ] and so far we have only seen increases in that. Please help us just understand how that works, like what drives the decline in our LDR. Secondly, on your risk management. Is this also IFRS 9-driven? So your better risk management is then being driven by the IFRS 9 framework in itself or is it something that DBS has done specifically, just for that [indiscernible] industry on the back of this? And then you put this offset by a weaker than expected environment, growth continues to be slow and maybe this recovery that you're expecting might not come through for the next year or after?
Do you want to talk about it?
So the MAS methodology that ensures that every bank keeps at least 1% based on the methodology. And any difference is actually taken through the result. So at this point in time, we are at the about 0.85% compared to the old methodology. Our total side of ECL. So that's actually very high relative to actually more banks that you see. So like you said, indication of seeing coverage. And as a proportion of NPL as seen at GP, we're 100%, right? So that's -- and in terms of the risk management, our IFRS 9 methodology is built upon the credit process.
So our credit process actually requires our officers to sort of flag up early where they see those weaknesses, either in the sector or in specific means. So as they move from sort of performing to amber, to red, to [indiscernible] before they reach NPL, the IFRS 9 expected credit loss, that's up substantially. And to the extent that they are managing the same names and release, the sort of ECL then drops. So that's how the process works. So they are linked, yes, and -- to our credit process.
But the other way around. So actually, it means sort of -- because they added so much GP this quarter. So 1% minus the GP the smaller number, is how it really comes down.
So you can offset that? Because if your ECL is lower than 1% then you have to drop it, but if it's more than 1%...
So if the -- if my ECL was all 1%, my other level would be 0. The both responses would be 0. They commented that they don't like to keep RLAR. The only requirement we have to put RLAR there is if you're under 1%. If you're 1% and over there's no RLAR. So it's a net number, so depending on what you put in goes into RLAR. If you bump up this number [indiscernible] number comes up. And on the other thing, we actually have [indiscernible]. We have improved the whole risk management processes. IFRS 9, our experiment is not dependent on IFRS 9 or IFRS 9 outcome comes from IFRS 9.
A couple more questions. The Wealth Management income this quarter is also super strong. It's at like almost the same level as Q1 and Q2, which I didn't feel like that quarter. Generally will be that quarter and you have a lot of volatility. So was there any run-off this quarter or [indiscernible].
Actually, it is actually quite strong. And it's also a variety of different products. So as far as units, that's went up a lot so the people bought funds [ and they made us ] quite actively. People did a lot of currency links up. People did more insurance. It was very broad-based.
Would you expect this to kind of...
So far in July, we very [indiscernible].
Third question, please, on digital banking. I think one of the key advantages in the virtual banks have is that they don't have branches, right. So they have lower CI. Do you think this will also make you rethink your brand strategy or your steady state cost to income ratio target like over time? If you don't -- if you realize that you don't need as many branches, you can probably bring your cost to income even lower than it is today...
Yes, sir. Of course we can. We start a bank from ground zero, you can start a bank much more cheaply, do you have to pay for cost of customer acquisition? So one thing you got to figure that all of the people are saying that they don't need branches, yes, but the amount of money that's spent acquiring customers is a lot more than the cost of a branch reconstruction, for all of these guys. There are only money left, right, in terms of them trying to get all the customers. So you can add customer acquisition costs into the CI and they think it will be low-cost R&D. All of these guys lose money. And if they lose money they spend more than 100% of the revenue they're trying to find.
But yes, particularly because we are not -- we are also conscious of the nature of the bank that we are. We are a public bank business in Singapore. And therefore, there's a limit to how sharply we want to come back on brand footprint. I think that's all we have to pay. And we do that consciously. So what we've done and we got good at is maximize the efficiency of our branch network. So we don't actually cut in the number of branches. Though we've cut down few branches. But mostly, which it is a footprint in the branches, there are -- and we do branches half the size of the old branches. We'll give you the head count in the branches. So branch head count is now already down by about half. So we continue to drive efficiency from the cost to the branch structure, not necessarily reduce the number of branches.
Just following up on that. Piyush, obviously you have a lot ahead and done of the whole digital journey and agenda. But do you think that the newest company stake could possibly generate new revenue streams likewise, with strong sales backing them, like alternative things coming into play and chipping away at your current market share?
It's possible. So far, I've not seen any evidence of a new revenue stream from any digital bank anywhere in the world. Neither the U.K. bank, there's nothing in the U.S. nor any other banks. The only person who got a consistent strong revenue stream is Ali. And they have -- it's not a new revenue stream. They have gone and tapped into a lending market. And one day the market will have people there, so -- versus China.
The size of that market of small countries are de minimis. And then on top of that, in China, they are unregulated they can lend what they want. In Singapore, MAS doesn't let you lend more than 4x, so what do you do? It's a great tiny market that you can lend into.
So you can't lose up. It is go for these segments. Micro -- like I said, moneylenders, micro SMEs, which are very few indeed, you can argue there might be some opportunity there. It's not entirely scarce to be in, as I'm looking and it's been 5 years, I'm still looking around the world to look for some really good use case. Some of you have been able to track in front, but the way I see the private banking is money raises where there is money. You move money around and you lend money. That's all every banking revenue stream that you look at fundamentally comes from this raising money, moving money up and charge the fees along the way, right?
So if you take it to fundamentals what is banking about? So you've got to try and raise money from segment or raise it for GP. Can I lend it to segments or lend it more intelligently or better? And can I make money by being more efficient in the way I move money around and raise capital [indiscernible]? And when we started to look at that prism, it's not obvious to me where are the big opportunities to create with the [ use ] and [ countries ].
And a thing on the Indian banks kind of catching up on the radar in terms of the whole tech movement and transformation, like could you share some thoughts around that as well? Like you were obviously a little ahead in the game, but it seems like kind of getting there.
Well part of them. I think around the world in the last couple of years, almost every bank has woken up to the sight that we took, so were [ 2 or a year ] and ahead of other banks. As much as, you can see JPMorgan, Jamie just said they're spending $11 billion a year. If you see the Bank of America, you can see a whole slew of banks, including the Indian banks. So a lot of banks are investing money to catch up. I think the 2 or 3 challenges -- in fact, everybody will -- have to progress on those at those in fact.
But there are 2 or 3 things which are not that easy to do. One is changing of your whole tech stack. In my [indiscernible] 10 years to be able to change the tech stack. It's not that easy to replace the core banking to move to micro system, to move to the big tech like small. Small is big. And even the most of the Indian banks I looked at, they're still not even building, testing the architect in the whole tech stack, so I think it'll take time.
The other thing that is changing the way we work, so to be able to drive more culture so you can actually complete with the big tech companies is also not ready. So I think there is assertive progress in the industry overall, including with the Indian banks. I think for all of us, including us when you compare us with the -- out of Amazon, Alibaba offerings. We still have a long way to go.
Just wondering on your noncorporate -- sorry, the corporate loan book, I think recently there have been some green financing and then the real estate here say they are about 20 to 50 basis points lower. So is there a sort of margin profitability tradeoff where you get somebody based on the risk-weighted assets [ waiting along ].
There is another -- a tradeoff.
Okay. So there's no tradeoff.
No, there is a tradeoff. We're giving it up. So we're giving it up to drive this thing so we don't dwell on cost of funding. Right now, it doesn't go on cost of capital. And RWA calculations don't change based on green or sustainable. And therefore, when EBITDA average ends greatly, I tell people to understand some things bearing the cost in data and unfortunately, it's a couple of billion dollars in [indiscernible] the recited business you find cheaper. I'm happy to do that to promote the agenda, but so far, it's a tradeoff.
Okay, and seems you're seeing that net cycle, your overall risk management culture is very cautious and hardly come down. Just getting what in their conversation differ on these, but do you think that there is a room that is in all this capital countercyclical buffers and all are lower, and therefore, there can be some room on the capital side?
I think you got all the rules, so I don't think MAS has discussion to amend these countercyclical buffers in terms of regulation. And you don't have a Singapore tax of capital requirement. It's already above [ ARPU now ]. So that will stay. But I think our ratio as it is now, I think that's like comfortable with the sort of ratio guided to.
And just a housekeeping thing, so you're in the margin. Your margins I think year-on-year is up -- for the second quarter is up 6 bps, right, 1.85%, 1.91%. Your Hong Kong margins are up over 22 bps, right? Singapore should be up because SIBOR has gone up. So where exactly -- which part of the book the overall market has...
Dollar book and the other currencies were down.
U.S. dollar...
We got about 2 basis points in Singapore; we got about 1 basis point in Hong Kong, we had a drag from the dollar -- U.S. dollar book and other local currencies. But we also had some technical stuff so that washed out a little bit.
Okay, I'm going to ask the operator to check with those that dialed in if they have any questions.
[Operator Instructions] We have Conrad Werner from Macquarie.
I just want to add a quick follow-up on the outlook for the mortgage book. I think there was one comment in the media session talking about a net flat outcome for that for the full year. But then I think later, there was also mention of adding growth of SGD 2 billion to the book year-over-year. Could you just maybe repeat the official guidance on the mortgage outlook through the end of the year, please?
2 billion was effectively the delta from first half to second half. The first half was minus 1. We did the second half. It will be plus 1. So the full year will be close to 0. Between 0 and half of it.
There are no questions at the moment. [Operator Instructions]
Yes, [ Robert. ]
Just think about extreme situations, what are your critical levels on capital, return on risk-weighted assets, whatever are the metrics you've used before at SGD 0.30 dividend before it gets revised or reviewed? And maybe some critical numbers if you would have to really look at that.
So our operating range, we guided to about 12.5% to 13.5%. So now with quarterly dividends, so it's more stable. You remove all the noise. But I think you should expect us to upgrade sort of somewhere in that range. And our guidance continues to be that -- it will sort of increase our dividend in line with a stable outlook. So that continues to be our dividend policy.
Is that sort of ROI or similar KPI that you -- if you fell below a certain level, you rethink this?
Right now, I'm biased to existing fees if not reduce it. So you and I, we see different parts of the universe, Robert. No, I mean clearly if you go back to [indiscernible] recovery plan. So the recovery plan essentially says that if we breakthrough our capital targets, and if it's at 11.5% or 12% capital adequacy, then we might go back and rethink our dividend.
Actually, what's the exact number is? Does somebody know what the company spend...
No, we do intend to operate at level. But I mean anytime we go below 12.5%. I think the guidance range is 12.5% to 13.5%. So as we go slightly over to sort of 12.5%, but I think...
No, but a lot of these things. If it slightly turns high before we go to dividend, there's a lot about the tools in our toolkit, and we can sort of manage our RWAs. We can do a bunch of -- get the less efficient asset. There's a lot of things we can do before we start to attack the dividend. One of the reasons we've been cautious about the dividend policy is I don't envisage the -- ever having to cut it, so we've been very comfortable about how we increase it, so we don't have to go down and cut it.
[indiscernible] talk about [indiscernible] Vietnam and another market and what is the outlook there and when you intend to go in?
We don't have specific plans right now. It's something we're looking at. And it is only for the management bandwidth. We feel positive on Vietnam. But between now, there's more investment moving from China to Vietnam. But also underlying growth is good. So we're positive. Question is pacing out there. So how do you...?
[indiscernible] chart digital bank...
No, we already have a branch there. So -- and the business is doing well. Our business in the last year has done very well. It's all a corporate business. We don't have a SME retail business. So the idea would be to use the digital -- exactly as we've done in other countries. We use digital to expand into an SME and retail space if we wanted to.
Just [indiscernible] question please. [indiscernible] offsetting 0.6% so you're above the top of your rank based upon I think it's projections -- dividend projections et cetera, et cetera. Your dividend looks self-sustained. So I can't imagine why we didn't come down to the top end of the range. So even if we've got a period of economic uncertainty, is your view that you are quite happy to continue to operate just a little bit at the top end of that range. So if you sit there, let's say in 2 quarters of time, and we're at 30, to 8 or 14, would you say we'll actually have to recalibrate it then or are you quite happy [indiscernible]
Well, we didn't have a reason to hold or stay in that range.
So you would have to bring it down to give it a balance?
So the reason we've given guidance to our top end of the range. We want to be a safe bank. But at the same time, we got to calibrate between being safe and returning what we can to shareholders, right? And so we calibrated the range so that, if you're well over the range, you don't get to [indiscernible].
Okay. So we should take that range today...
And so our dividend policy, we will continue to grow our dividend in a sustainable way over time.
So we should believe in that range.
Even that you booked in Hong Kong but the [indiscernible] range in network. Does it impact your profitability at all?
When people comment...
[indiscernible] taxes.
The Hong Kong/Singapore taxes.
What is the shift of AUM? What do you exactly mean by that when people worry about Hong Kong? You said that...
No, no, no. If it were, I don't think it happens. If it were, the shift depends -- the macro shift will now change but who the beneficiary of the shift. So if you're operating in Hong Kong, we have strong market share, right? So you have big banks and large -- and look at all of the [ Swiss ] banks, they're much bigger than us in Hong Kong. But if money were to shift to Singapore, we have double the market share. So most people have more money to Singapore, they look at DBS twice as seriously as they look at us in other parts of the world.
But your product suite and ambition would be things that doesn't matter [indiscernible].
It is more psychological, that kind of thing.
You talk about a very interesting bit on customer acquisition in the fintech. In India, you had to look for a long time. We're going for the technical partnership. How is your customer acquisition cost shifting in India? And [indiscernible] utilization, but is it getting to a point we are feeling it is cheaper to apply a customer even after we have some cost on branching around that level. What you do for your delivery?
No, no. So they're only branching enough to be able to create brand presence. And together on the regulatory change that I need to have some physical presence since I want to market in a particular area. And sir, is this the [indiscernible] I want to market in digitally and corporeally to have some presence in Bhopal. So if a customer signs up from Bhopal, I still have to go to override as I don't have a branch in Bhopal, we can't make a change in customer regulation, which is a strange regulation but it is what it is. So it allows me to address that. And second is you need some brand relevance if we got the digital thing going in Indonesia works better if people see the DBS chain somewhere in the city, they would know that we exist. So we have very limited plan on the consumer side. The other thing, we're trying the digital for SME. SME you do need some points of presence to [indiscernible] documentation, et cetera. That's what we're [indiscernible] in India. The customer acquisition cost [indiscernible] our customer acquisition costs has come down from the digital cost. SGD 66 is now down to about SGD 10 made per customer. So it come down very dramatically over time. But you've got to understand what is the quality of the customer? So a big learning for us in the first 2 years is we were getting a lot of customers. But when you look at them how the customer never going to be profitable. So we don't have to repeat -- pivot it and sharpen up the learning from Indonesia, so we scaled down the amount of customers from what we get. But we're now getting much better customer profile. That's what got us to close to our new hurdles that [indiscernible]. But then you have cost per customer acquisition is higher because my fixed cost is spread over a smaller number of customers, so the customer cost per acquisition goes down, but a newer customer goes up. But you will trade off the quality of the customer as well when you track acquisition cost.
You should count ambition loan mix in India make 2, 3 years down the line [indiscernible].
It's way early. The 2, 3 years is not enough to move the needle. So we today, we're booking now on secured loan of about SGD 8 million to SGD 10 million a month from India and Indonesia, right? And that's growing. It's doubled from here to here. But you have SGD 120 million. Double that, it'll be SGD 200 million, SGD 300 million. I can't get worth SGD 300 million. So it's a small percentage in the big scheme of things now. But the good thing with this is it's growing exponentially. So we projected -- SME -- SME is just started. So SME we just started. It is very early. So we only got 2, 3 months of SME experience on that one.
Can I ask one -- on the GE increase this quarter, could you help us understand which part of the ECL [indiscernible] how can you provide more JV? [indiscernible] did something else change? Because the last time [indiscernible] rights of [indiscernible] How does that happen? How does that work exactly?
So typically, you would see -- in last quarter, we had SGD 100 million write-back right? So we said that you actually get ECL moving into -- when you have NPL moving as ECL, right? Then you have a weak case, for example, that goes out of the system into a reduction. The last year, these 2 elements were there in the first quarter.
On top of that, remember fourth quarter, the markets were looking really bad. So first quarter, based on our methodology, actually we released -- in some sense, driven also go by the equity market. So we actually released some contract to the very bad environment in the fourth quarter where everybody was expecting the worse with the rate hikes and all that. So there was a SGD 100 million write-back in the first quarter.
In the second quarter, you have the same 2 elements. Again, we released our ECL because NPL -- some have moved NPL, and you further managed down some of the weak names in the ECL stage. So having moved this to the change in the second quarter is the full weight, right? So this is second quarter everything sort of looked much more strong from a macro perspective, from a geopolitical perspective. So it would be up element in the second quarter. [indiscernible]
[indiscernible] most of us are model-driven. But we retain the flexibility to put some over [indiscernible]. So the -- we don't -- if you [indiscernible] somehow, the macroeconomic behavior and the model are not capturing everything, then we actually add something about that.
So it's not different from what most bank would describe in the management overlay. So for example, the UK banks have all put in management overlay for Brexit because of the uncertain outlook. So that is how you manage an ECL [indiscernible]. You need to split [indiscernible] justify the document.
And the auditors has got to sign off on it. So what is the underlying and why you need to do it. So you look at the [indiscernible] numbers, it's got a big over lift from Brexit. So nobody knows what it might be, but it is just to put something there to be [indiscernible]. So we just took note of the good macroeconomic headwinds from trade and taken up our [indiscernible] and adjust it for this.
Do you have any intentions of disclosing more details around the ECL framework [indiscernible] banks are doing [indiscernible].
So we will sort of -- I think at year-end, there's normally ECL profit last year. So last year when we publish our ECL, actually ours was the most comprehensive in this part of the world. So this year, we'll see what advantage additional disclosures will be helpful.
One more question on NIM. So previously, you mentioned that one great cut you are expecting NIM to impact it by strong bps and additional cuts perhaps 1 to 2 bps. And you also mentioned that you had the room to take in more strategy for funding up to about 10 billion. So does that 1 and 1, 2 bps reduction in NIM taken into account the potential funding using cheaper CPs versus say at least?
No, so I estimated the current book. Of course the management base keeps seeing what we need to do with the book, including hereon high [indiscernible] funding cost. There's more of that.
I'm thinking of last quarter's LDR it seems to be one of the highest in the last few quarters. But NSFR, LDR looks pretty stable, which means there is probably more room for you to get LDR [indiscernible]. Can that move us significantly more because given your dilemma buffer on LDR.
Of course. But I'm also conscious of the fact that I don't want to wind up being more way aggressive on liquidity. There's no idea what's going to happen if things suddenly go [indiscernible].
Just on the cost side. You said there's a bit of flexibility on the cost. But if I'm looking in the longer term, is this of a level that you let cost to income go to -- if we see more pressure on revenue than you expect.
Yes, I don't have given it too much thought, but it affecting that for a period of time, a year or so our cost/income ratio has to go up, then we will dip it. It's not something we want, but I'm quite serious. We don't manage to say okay let's scale down for quarterly performance, right?
So are you seeing more pressure in mix shift in deposits towards savings going down everywhere? I know you think we held there. I think in other markets, it's been quite a clear shift.
Anyway, still in Singapore. So we had market share. But the system there's a clear shift from savings to fixed deposits and [indiscernible]. I think it really come off. You'll see that hardly works. I went and modeled it, the last 2 [indiscernible] increase cycles for the last 25 years. It always happens. And the rate increase cycle is starting a shift from savings to deposits. And then the rate increase cycle changes starting the shift to where it was. So of course we could -- the behavior could be different than [indiscernible] granted. But so it's been going up. But the pressures already moderated. So I think it's not actually [indiscernible].
All right. So we'll see you next quarter. Thanks for coming.
All right. Thanks, everybody.
Thank you.
Thank you. We will now disconnect. Thank you.