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

Over to you.

Okay. Hello, everybody. Welcome to the DBS Second Quarter briefing for analysts, the buy-side. This is Michael. I have got Piyush, Sok Hui and the rest of the team being here. So you have heard the presentation just a while ago, so we can go straight into Q&A. So operator, we can open the line for questions.

[Operator Instructions] Our first question, we have Aakash from UBS.

I have 3 questions. The first one is on the loans and the moratorium that you've disclosed in the presentation. It only amounts to like 5% or so of the total loans. So I just want to check if that's all? Or are there any other loans that are under moratorium as well? Because this number actually looks quite small to compare to UOB, which just closed around 16% of the loan book under moratorium. Secondly, whatever this number is, do you think it has plateaued? Or how do you think -- how do you see a change into the rest of the year? And the third related question is, in Q1, you shared some guidance on the stressed loans that you have on the total book, which was around, I think, close to $50 billion, which was 13% of your total loans. So do you have any updates on that number? Has that changed since Q1? So this is the first category of questions on moratorium and stressed loans. The second one is on the brand strategy. So you did mention earlier that you've seen digitization in the last 3 months, which would have happened in the next 3 to 5 years. Has this also led to some change in your thinking around the branch strategy? Do you see the need for less branches now compared to in Q1? And thirdly, on the dividend, what's your thinking for dividend for next year?

Okay. The first question was on the moratorium. Now this is the entire moratoriums that we have. It's -- there is nothing else on moratorium other than that, but $12.5 billion, which is the SME book and $5.5 billion, which is consumer book, so $17 billion, which is what it is. I think one other thing you've got to reflect is that our large corporate book is much bigger. Our percentage on the SME book is smaller than some of our competitors. And therefore, if you start looking as the percentage, you find that our loans under moratorium are somewhat smaller just because our large corporate book is larger, I think. The second question you had was...

Branch strategy?

No, $30 million -- the stress portfolio. Frankly, as I said, we've got nothing to tell us right now that there's incremental stress in the portfolio. We're not seeing any increase in delinquencies in the SME book. We're not seeing any increase in delinquencies or performance this thing in the large corporate book, which is why it's not showing up in any NPLs, but I can see that the delinquencies are not going up. So we're not seeing any incremental signs of stress at this point in time other than the consumer book. The consumer book I pointed out before is, we saw increase in delinquencies in the unsecured consumer space, not mortgages. Mortgage has been flat. It's only the unsecured consumer book that we saw some pickup in stress. Your second question on brand strategy, our brand strategy actually has been consistent now for the last 3, 4 years, which is we are trying to automate and digitize our branches quite rapidly by using things like virtual teller machines, branch teller machines, digital ATMs. So we're trying to effectively shrink branch footprint and automate the branch as much as possible. We will continue to do that. Over the next few years, it could wind up -- we actually need less physical outlets in Singapore. Though, as you know, we don't think we need less physical outlets in other countries. In places like India, we're actually increasing our outlets because it's too tiny already. But in Singapore, yes, it's possible that we could consider that, but the bulk of the shift is what we're doing, which is converting from a manned branches to digital branches. Your third question was on dividend. We've been guided by the central bank to hold dividends at the 60% level up to the first quarter of next year. So that's a given. Beyond that, it depends. If our outlook is correct in terms of what we're seeing happening, then we do think we have the capacity to increase dividend. In fact, as I said originally in the first quarter that if it weren't for being guided by the regulators, we would have been able to maintain a much higher level of dividend. So as long -- now again, tail risk can happen. So I mean it is obviously possible that things get much worse than we anticipate right now. And of course, we would revisit our dividend thinking at that time. But if things go according to what we are seeing right now, we do have the capacity to start raising dividends after the financial year is over.

Got it. Just a follow-up on that, Piyush. So the 5% and the moratorium loans that you have, do you see that number sort of having plateaued? Or do you think that could go up a little bit in the rest of the year?

I think it's plateaued. I think what -- will continue to go up. So in the new loans we gave -- these are moratorium loans, right, we've also given new loans under the 90% government guarantee program. At the end of June, about $2.5 billion had been drawn down, whereas we had approved well over $4 billion. So I think you'll see another couple of billion drawdown in that category, in the guaranteed loans category. In fact, we saw another big drawdown in July. So -- but that, if you remember, is 90% government guarantee. So the risk we take is only 10% on those loans. But on the moratorium loans, they've already leveled off. We saw a big increase in the moratorium loans in the early part of the quarter. And by June, they had leveled off. July, we are hardly seeing any more pickup.

Our next question, we have Robert from Citi.

Just a couple of questions, I think, number one, could you just give more color? You mentioned that on the net interest income, because of the interest rate drop, you -- in the second quarter, you were down $80 million per month. Going forward, you're probably going to be down $100 million a month. So what does that mean in terms of what your exit NIM was at the end of June? And where do you think you will exit at the end of the year? So I know the 1.6% guidance is obviously an average of the whole year, but we're trying to work out where the new normal NIM would be. So that's the first question. The second one is, you obviously just said to Aakash that the loans on the moratorium have plateaued. I just wondered what you're looking at in your assumptions for a peak NPL ratio and, indeed, whether you actually may come in better than your base case. Your base case is $3 billion provisions. Your worst case is $5 billion. You have an implied peak NPL ratio in there. But with so little under moratorium, will you actually come in better than you expect?

So Robert, on the first question, our exit NIM for June was 1.58% right? But as I took pains to point out in the media briefing that the impact of the surplus of CASA that we have is about 6 basis points. And so to that extent, I'm not managing to NIM, I'm managing to income and ROE. That $20 billion CASA surplus [indiscernible] 0, I put it to MAS at 50 basis points. So it makes the income and MAS assets as 0, respectively. So it's good for ROE, it's good for income, but it depresses the NIM. If you add back the 6 basis points, which is discretionary for good businesses, then I'd say the right number to think about apples-to-apples is about 1.64% is very wound up in June. Now as we go through the rest of the year, NIM has to come down partly because, as we've said before, the housing loan portfolio reprices gradually, the fixed rate loans and the FHR loans, the impact of that will continue to come through over the next several quarters. And therefore, on a normal basis, the 1.58, we expect to go down to the low 1.50s at its bottom. The way we've articulated before is that our interest rate vulnerability is about $8 million per basis point. And so with the rates coming -- having come off roughly about 100-odd basis points. If you look at SOR, SIBOR, HIBOR, at $8 million, that's about $800 million in the first year. And then there's some subsequent impact on that in year 2 and so on. So that's roughly what gives you the $80 million going up to about $100 million number a month.

[indiscernible]

Yes. Well we already said that, that if you look at the second quarter exit rate, if you're saying what is the exit rate, you're comparing to June exit. Your second question on the NPL assumptions, it's very hard to say. The way we've done our estimate is both top down and bottoms up. So for the large names, the large corporate names, we've gone name by name that covers like 95% of our portfolio. We've gone by sector. We've gone by names. And for every name, we've tried to figure out what is the cash flow, what is the vulnerability, et cetera. So it's relatively robust. For the SME book, which is where most of the moratoriums are, for some of the names which are larger, we've done name by name. For a lot of that, we've actually just have to do portfolio estimates. And that's why it's hard to predict what the actual NPL and the provisioning number on that could be. We are reassured by the fact that, that book is pretty much all secured. It's 90%-ish secured. And the security mostly is property, and the property collateral values are quite conservative. They're by and large south of 70%. So to me, the tail event, which we have not catered for, what happens if property prices across Asia collapse to 50% of level, then you might wind up with unexpected risk beyond our $3 billion to $5 billion range. But at this point in time, we think we have sufficient collateral cover in that portfolio. At the same time, when you do the modeling, you actually have to model a whole bunch of migrations and what happens in the book. So it's hard to give you a real estimate on out of the moratorium loans, what is our NPL assumption. I prefer to think of it in SP terms what I think the total credit losses might get up to. You've got the most specifics things -- where could -- I don't think we'll -- it's unlikely it will be south of $3 billion. It's possible, but it won't be because I think credit migration will require us to build up ECL. So that's, I think, a given. So it'll be hard for me to see that. When I did the probability this thing in my own mind, I'd say, there's roughly half in half chance that the low end of the range is the top end of the range, right? It will be somewhere in between there.

Thank you. Our next question, we have Harsh Modi from JPMorgan.

One question on NIM. When you think about, let's say, your rate outlook for the next 18, 24 months, how are you thinking about positioning for that as in trying to, say, all right, let's lock in rates right now for next 3 years, as in trying to increase the duration if customers can. Even if it means 20 bps lower yield, let's just lock it in. Is that the view? Or are you trying to kind of push back that repricing to the max? That's one. And second, the 160 bps of NIM guidance, does it include some sort of reversal of accrued interest in second half of the year?

Well I think the problem is, the yield curve is so flat right now and the rates are so low that pushing duration is not very helpful at this point in time. I mean we don't get a lot of pickup when you lock yourself in. Though to be fair, we are doing that in the mortgage book. In the mortgage book because the floating rates in the market is so low, we are doing 5-year mortgages today at 1.50% and so our 80% of our new mortgage bookings are all coming in the 5-year category at 1.50%. And that's less to do with the view on rates, it's to do with the fact that the floating rate in the shorter end is just uncompetitively priced. But for anything else to build duration at these levels is not very sensible. Frankly, if you did what we had done -- if you remember, I've been saying for some time that we've been building -- putting out duration in the last 2, 3 years with a view that rates might collapse, and that's been helping us a lot. We've actually got this $35 billion, $40 billion portfolio. We've built up in our corporate management overlays because of that. That's helpful. But that was built in a time when you could get duration pickup and get yield pickup, right? It's much harder to do that today. Now what we are doing is tactical that if the rates turn up, if the yield curve pops up and we think there's an opportunity to build some duration, we do, do that. We spent a lot of time actively managing our balance sheet. Through this quarter, as you can see, for example, in the liabilities side because we got some close to $40 billion of CASA, we had to let $30 billion, $35 billion of fixed deposits runoff, and we've been very nimble at being able to do that. Similarly, on the assets side, when rates move up and down, if we find the opportunity either through cross-currency arbitrage or just in the cash market, we go and selectively try to add positions with duration, but it's not easy to find those opportunities. Your second question on the 1.60%, no, we've got no interest accrual reversal built into the 1.60%.

Okay. The second one, Piyush, is again a bit weird question. Some of the fee income this quarter was lower activity, that's fine, but the flip side of digitization is a lot of the competitors, the 10 guys, have started to come up with their own distribution, product distribution and what have you, very small, nothing to impact you right now. But in terms of pricing, either on insurance or wealth management distribution, again, at mid to low end, is that something you get worried about? And how are you kind of offsetting that, if at all?

No, I'm not worried about it right now. If it is going to make an impact, it will be over several years. I think the best analogy to look at is Europe. Europe's had this negative interest rate environment now for the last 2, 3 years. Every bank in Europe has increased fee income by rethinking its fee architecture schedules and notwithstanding Monzo is there and Starling and all these guys are there. All the high street banks in the U.K. and Europe and Switzerland, everybody has been able to hold and grow fee income now for the last 3, 4, 5 years. So I don't think it's going to be material in the short term.

Our next question, we have Nicholas from Crédit Suisse.

Just a couple of questions for me. Firstly, on the investment security gains. Just wanted to get a sense of how much unrealized mark-to-market you are sitting on at the moment. And how do we think about -- how much you're willing to recognize or realize in the near term and perhaps in 2021 as well? The other question was just in terms of the loans under moratorium, do you have a rough estimate of how much, I guess, you would expect for these to become NPLs? And even if they do, given that they are quite highly secured, would you expect -- how low would you expect these -- the loss given the thoughts to be?

So on the first question, we're currently about $1.5 billion in mark-to-market gains in the book. And the question of when we realize it is a little tricky because every time we realize it, that means we put further pressure on our NIM going forward because these are really good yielding assets because we've got into these assets at a good time. So we will trade off whether you want to front-end some of the gains or whether you want to let them trickle through. So part of that goes back to your question, what is the outlook on rates? If you're saying that there's a good opportunity to recognize the gains, then you do. Otherwise, it's okay. It's only a timing question. So when does the value trickle in through the book. On the moratorium loans, it's very hard to say. I've answered the question twice. I mean it's -- your guess is as good as mine about how many of them will turn NPL and what's going to happen to that book. The reality is because there are thousands of customers, it's very hard to -- it's a small ticket portfolio, very hard to go back and try and figure how many of them are going to recover and survive after COVID, how many are not going to survive after COVID. Like I said, the fact that we have security collateral in property, that's helpful. But that notwithstanding, we're just being prudential. We're trying to build up provisions ahead of time. The biggest provisions and general [ ounces ] that we're building up, I think, will relate to the SME book and, to some extent, secured -- I mean the unsecured consumer book. That's where I think the biggest challenges will be. And so since we can't actually predict what that number is going to be, we're just being prudential ahead of time.

So because your unsecured -- you can assume that the LGD, which you're asking, is relatively low.

[Operator Instructions] So our next question, we have Nick from Morgan Stanley.

Just a couple of questions surrounding loans, really, for me. First of all, just as a matter of detail. Sok Hui, I think you said in your presentation that wealth management loans fell in the second quarter. I'd just be interested as to what was the driver of that. And then just more generally, I mean, if we think over the next 2 or 3 years, I mean you obviously -- you set out a strategy 2 or 3 years ago where we lifted returns to a reasonable level, you've built up capital to where you needed it to be, the demand for credit or capital-intensive growth is quite low, and so the net outcome of that was obviously a pretty attractive dividend policy. I guess as we go forward over the next 2 or 3 years, your return outlook may be low given where rates have gone. I'd be interested to know what your thoughts are on the capital consumption of growth going forward. Are we in an area where there is more demand for credit? You spoke about digital transformation. And therefore, can we get back to sort of high payout? Or where will the balance be do you think going forward in terms of funding growth and funding payout from the capital you generate?

On the wealth management loans, actually, it's quite this thing. A lot of people -- because of the big challenges in March, people had to pay down margins. So a lot of the wealth management loans come from margin financing. And they had to do margin calls, so people sold assets and paid off their margin loans. And that's basically what accounted for the drop. Most of the drop was in that March-April period. It's actually stabilized after that. On your second question on the dividend policy. Our dividend policy is constructed around being able to support growth in the core business. Our typical RWA growth is about -- has been like 4%, 5% over the last 2, 3 years. And so our capital policy can support that and continue to pay out in a reasonable way over time. As we go forward, if it turns out that credit demand is even weaker and so you don't need the capital, we will not going to continue to increase capital, right? I said earlier that left to ourselves, it's not clear to me that we would have cut dividend at this point in time. And certainly, when the central bank restrictions are over, we definitely have the capacity to pay a lot more dividends, and we're being held that right now. So we would take up our dividend level. How much we would actually take them up to would be dependent on the market outlook at that point in time.

So is your view that the demand for capital going forward will be lower, not higher? You don't expect RWA growth to be as high, say, over the next 5 years. How has it been over the last 5?

So my -- it's hard to say. I think that the impact of COVID is quite unclear. And some of the big mega trends, both the trends around geopolitics and social politics and supply chain shifts and so on are also relatively unclear. So it is something to be watched what happens. The fact is that there is a huge amount of incremental debt built in the system, but most of that is sovereign and government and fiscal debt, et cetera. How much that crowds out private sector borrowing over the next 2, 3 years is also not entirely clear. How much default in bankruptcy there is in the SME sector and how long it takes for creative reconstruction to happen is also not entirely clear in my mind. So I think over the next 2, 3 years, I would be surprised if we see massive credit creation. Our current projection is, back end of this year, we'll go back to our normal rate of credit growth. But like I said, we have to watch and see exactly how things unfold by year-end.

Our next question, we have Melissa from Goldman Sachs.

Just 2 questions. Firstly, maybe just on India. Can you give us some color on how the books have developed in India loans under moratorium? Is there anything we should be worried about? And then secondly, in terms of, let's say, we look at 2022, when your credit costs can come back to a more normal level, but rates are so low, what are you looking in terms of your target ROEs then?

So India has been a bright light for us, frankly and it's only because we went through so much pain. Between 2013 and 2017, 2018, it took us 4, 5 years to clean up our book, but we have a very robust book at this point in time. And so in fact, of all our countries, India is having the most spectacular year this year. We've got no stress in the system, our portfolio is holding up well and we are gaining market share in sensible kind of businesses. So -- and frankly, nothing in more moratorium in India. As you know, we don't have an SME book in India. And our large corporate book has really gone very upmarket. So it's quite robust. On the projection of ROE, the question is really with the collapse in margins, there's going to be some headwind on ROE. So the 2 or 3 drivers we have on ROE, one is going to be how can we increase noninterest income into the future and that's something that's actively occupying us. Like I said, the new world, I think, will create some opportunities for some new income streams, which are fee-generating. We've had some success, but we're going to have to double down on that a lot. Again, it's constructive to see that some of the European banks, not all, have been able to achieve that over the last 3, 4 years. So it's something that we can take some guidance from. The second is our cost structure. We've continued to look at our cost structure. In the short term, we are being very thoughtful. I think it's -- for a company like us, there's enough pain in the system for us not to be creating more pain by beginning to start retrenching people and laying people off right now. But again, there's no question that if you think over the next couple of years, we're going to have to think very sensibly about rearchitecting our cost structures. We think we have the capacity to do some of that as we go forward. And then finally, it goes back to what level of credit costs we have. So I think with the NIM collapse, our 13% ROE [indiscernible] from that and gain back another thing over the next year or 2 remains to be seen.

Our next question, we have Jayden from Macquarie.

I just have a couple of questions. The first is on the fee side. And just looking at -- you mentioned during the presentation some falls in the wealth fees and then the recovery. I'm just thinking, conceptually, how much of those fees are kind of recurring or sort of trailing fees kind of like the service charges you make? How much of it would be there even if there were no sales? Would just like to have some color on that? It seems like the AUM is holding up very nicely and you're seeing flows. And my second question is just on costs. Obviously, there was good cost control. Was there any benefit from the Singapore government's job support scheme? Is that a factor? Or is that something that's not relevant for looking at the cost side?

On the wealth management, as you know, nature of wealth management in Asia is quite consistent. Most of the income comes from activity. So the commission as opposed to trailer fees are [indiscernible], I don't know the number off the top of my head, but last I reckon, our trailer fee number is only about 10%, 12%. So it's not much higher than that. So you really have to have clients doing activity to generate that income and which is why that explains why the fee line, which is linked to confidence. If the confidence levels are high, there's a lot of activity. If confidence levels are low and people don't do anything, then that shows up in the fee income line. It's -- by the way, it's improving, but it's low. It used to be 5%. Now it's north of 10%, gone on to double digits, but it's still the minority of our total income streams. On the second question, which was the job support program. Sok Hui, do you want to take that?

Yes. So job support program for the first half, net of additional accruals, because more people are not taking their leave, is only about 1% of the overall sort of impact in the first half.

Our next question, we have Krishna from Jefferies.

Few questions from my side. Just on this sustainability and ESG initiatives. Maybe you can share some stats. Of the $20 billion loan growth that you had for the first half, how much was it linked to those initiatives because the real estate peers seem to be suggesting quite a bit of green loans. So that's the first question.

I didn't follow the question. What initiatives are you talking about?

The ESG and sustainability initiatives. So maybe if you can share that out of the $20 billion loan growth that you had in the first half for corporate -- for the wholesale segment, how much of it is ESG or sort of green loans? That's the first question. The second question is, in your media presentation, you did say about a few broad changes. So I think on those, you alluded to taxes and taxation changes. So I think banks have generally benefited from government support. Do you see any sort of change going forward either on taxation for your side or on the wealth management segment. Any introduction of the state taxes, et cetera, and how you are guiding your wealth management customers? That's the second question. The third question is on your Hong Kong exposure. I think geopolitics, you mentioned about one trend. So your -- the structural exposure that you have to Hong Kong dollar, is -- does that need sort of a careful management? And what's your thoughts on that? And the last bit is on, you said about the various industries that will change. But on the ground, when we speak to those industry segments, like hospitality, it seems banks are still lending. So I mean will that change? Do you see deals coming through? What's your take on that?

Okay. The first question on sustainability, somebody have the number on what we did. We did about $5 billion last year. And I'm aware of some -- anecdotally that we have done this year. We're not -- I don't have a specific number for you. Somebody is going to have to get back on that. But the reality is we are very active in the space, so both green financing and sustainability-linked loans. As you know, they're somewhat different. Sustainability-link loans, basically, is you agree with the customer on a target set of ESG metrics that they have to hit. And if they hit those metrics, you give them a discount on rate. So that really depends on the future performance of the customer. And the green loans depend principally on what the underlying asset is today. So they're different loans, but we'll get back to you on this specific number. On the tax changes, I think it's going to happen directionally. I have not heard of anybody talking about it today. But I -- it's just my general view, frankly, it's has been for some time, that over the next decade, we're going to see an environment like the 1970s, where the idea of taxing, the rates and taxing corporations and so on, increased tax rates and greater income redistribution have to happen just because of the social issues that we are facing. So my guidance, for example, on the wealth side to our clients has been, right from January, that if they haven't done it, better think estate planning, better think about how you manage in succession and so on. It's a worthwhile thing to do. On the rest of the corporate side, unfortunately, there is a lot more focus on tax avoidance, tax evasion, bets and so on. So I think companies just have to be reconciled to we seem to be doing the right thing. I don't think there's any ways out of that. On Hong Kong, we've been conscious about the Hong Kong challenges for a while. So even in the last few years, we've been reshaping our Hong Kong portfolio. For example, 5 years ago, our Hong Kong portfolio on the corporate side used to be 1/3 SME. Today, it's only 15% SME. And that's been deliberate because over the last 2, 3 years, with all of the challenges, we've been taking our portfolio more to a higher quality level. By the way, that explains why our moratorium loans, et cetera are SME loans. SME business, we shifted a lot more into a liability and transactional business as opposed to a lending business. So it's something we've been thoughtful about. We said our concern is on the Hong Kong dollar itself, I'm not sure if we're alluding to the peg. I don't really think the peg is going to go. I think between Hong Kong and the Chinese, they have enough fire power to hold the peg as long as they wish. If the peg goes, I think it will be driven not by outsider but by the Chinese. In some states, they decide to switch the peg from the U.S. dollar to a basket or a renminbi. That's when it will actually go. Them notwithstanding, we obviously do all of our stress tests on what happens if peg goes, what's our level of exposure, how do we deal with that. A large chunk of our capital -- we have about $10 billion of capital in Hong Kong. About $4 billion or $5 billion of that supports the Hong Kong dollar book. That will be neutralized. If the value of Hong Kong dollar changes, the RWA and the capital will all change proportionately, so I don't think we have an issue. The rest of it, we really used to hedge our U.S. dollar book. So we're looking at ways in which we can actually modify those positions. Finally, in terms of industry, so if you look at hospitality, you said everybody is willing to lend. We are also not that concerned. But I think that's to do with the underlying nature of the sponsors. The hotels in the region, the [indiscernible] the Shangri-La, for example, and some of the other things, I mean, you're banking on large corporate balance sheet. But if in the long term, it turns out that hotel occupancies are down by 30%, 40% on a long-term basis, even the sponsors are going to rethink the business and the business model. So of course, the banks are going to have to rethink their whole financing agenda. You take airlines. Yes, we were part of the sing dollar -- in the Singapore Airlines bailout, we've been there. But we've been there principally because even with 95% traffic gone, the governments are standing behind the airline. At some stage, the governments are going to rethink their strategy and that's why we would also have to be thoughtful about what we want to do. So Mike has pulled out our total numbers already. Our total ESG-linked loans in the first half of the year were about $4 billion, for the first half of the year. On top of that, we also ran bonds. So we did another $4.5 billion of green and social bonds. So our total sustainability-linked transactions in the first half of the year are $8.5 billion. But of the $20 billion loan book that you talked about, it's about $4 billion.

Our last question, we have Anand from Bank of America.

Just a couple of questions from me. Firstly, Piyush, what are your thoughts on the new transition to the new benchmark away from SIBOR. How should we think about it, what time frame, and what could the net impact on them? And secondly, your follow-up question on Hong Kong. We've seen a lot of news around flows from Hong Kong to Singapore in terms of wealth, et cetera. Any trends that you are seeing, any color you can give us, it would be great.

Sok Hui, you want take the SORA question?

So I think like all other jurisdictions, we are also preparing for the transition from SOR to SORA. The MAS has been actively sort of engaged, and I think they just came out with an announcement on how they will sort of administer and sort of be also supporting the growth of SORA, sort of corporate loans and bonds. So in Singapore, it's very nascent, but we have done bonds. We have done loans. We have done some of these derivatives. So I think it's making an effort to have the price discovery, and I think we also have been posting, I think, the SORA sort of benchmark. So I think we are moving at a pace that is as quick as any other jurisdictions preparing for it.

I think to your specific question, the way you should think about it is, collectively, as a regulator in the industry, the intent is to make sure that customers' credit spreads are stable. So they don't wind up having to increase credit spreads or reduce them. And therefore, if you think about it from a banking standpoint, which is not a material impact on our earnings, it'll just change the reference rates, if you will. The Hong Kong question, no, we haven't seen any massive outflow from Hong Kong. In fact, we saw more of it last year when the original social unrest was happening. We haven't seen too much of that in the recent times.

All right. Thanks very much, everybody, for joining us. That will be the last question. We'll see you next quarter.

All right. Thanks, everybody. Thanks all.

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