Transcript Post Results Call Half-Year Analyst Meeting 3 September 2020, 08.00am BST EWEN STEVENSON, GROUP CHIEF FINANCIAL OFFICER: Morning or afternoon all, depending on where you are. I’m actually out at our Canary Wharf offices today for a rare day in the office, but we’re going to try to as an executive team to begin to get back into the office one or two days a week, although Canary Wharf for us is a particularly challenging building, given the restrictions on tube travel in London and the inability to cram a large number of people into this 42-storey tower. Just a few update comments following our Q2 results. So I guess we last spoke about a month ago at the beginning of August. On revenues, just as a reminder, interest income will get progressively impacted by the close to zero interest rate environment we now have in the UK, US and Hong Kong. On non-interest income, as I said a month ago, we do expect some of those line items, particularly Global Markets, to slow down in the second half. The second half of last year was dramatically lower than first half of this year, so I’d just ask you to pay attention to that. And on customer activity, I think we’re going to have to watch the path out of COVID globally. Things like the return of mainland Chinese tourists into Hong Kong is important for our insurance franchise, but generally customer activity levels continue to be subdued as a result of COVID, which is impacting some non-interest income lines. On ECL, there’s nothing really to update post what I said a month ago. On operating expenses, I guess just watch for the fact that the quite sharp dollar weakness over recent months, and obviously we’ve got a decent amount of our cost base that will get impacted by that, i.e. a dollar for dollar comparison will be impacted by dollar weakness. On income from associates, in the last couple of weeks you will have seen an announcement out of Saudi British Bank about impairments and goodwill they had from the Alawwal acquisition. That will have an impact on the way that we account in Q3. And then, as we look out more broadly, while the shape of ‘21 is becoming clearer, there continues to be substantial uncertainty as a result of the path out of COVID and how that differs across different markets. I think some of our fears of very severe second and third wave lockdowns economically doesn’t appear to be coming through, so downside risks around COVID have probably moderated and the development of a successful vaccine continues to progress at pace. Equally, I think the sharper v-shaped recovery that people had hoped for doesn’t look like it’s going to come through, so it will be a slower recovery out of COVID. Then, obviously, we’re still waiting for further news around Brexit. We do expect US-China tensions to remain to the fore, particularly in the run-up to the US presidential election. So there continues to be a material amount of uncertainty out there, and I think until later in the year we won’t really have any good visibility on where we’re sitting in relation to distributions at the end of this year and into next year. And as I said a month ago, we would expect not to update you until full-year results. We’re currently doing a lot of work around what the new environment looks like. I certainly think we need to accept that interest rates are going to be a lot lower for a sustained period of time. Consensus now doesn’t really have policy rates moving up or down for about three years, so we’re doing a lot of work at the moment around the cost structure, but it’s too early to talk about that in any detail. -- 1 of 8 -- So I was going to stop there and just open up for your questions. And just as a reminder, I’m joined today by Richard O’Connor and his team in Investor Relations, and also our CFO in Asia, Ming Lau. MANUS COSTELLO, AUTONOMOUS: Morning, afternoon, everyone. I wanted to ask about ECL, please, just as I had more time to read through the interim report post-results. I was looking at the post-model adjustments that you made and they seem to be extremely high. You had, I think, a $4.5 billion reduction in your ECL as a result of the post-model adjustment and relative to your ECL of $14 billion and a charge of $4 billion in the quarter, that does seem like a very big number, so I wondered, could you talk us through why you made such a big manual adjustment and, secondly, more broadly, does it not lead you to question the models underpinning ECL if you have to manually adjust it by $4.5 billion? EWEN STEVENSON: Firstly, factually, yes, there was a very large manual adjustment. As we look at COVID, there is no precedent in terms of historical data going into the models; as we model out through those ECLs, Manus, there are I think a few things. Firstly, the severity of the downturn that we saw in Q2 was something that we’d never seen before. Equally, the strength of the recovery is something that we’ve never seen before and the level of government intervention globally and central bank intervention, and the speed of that intervention, is something that we’ve never seen before. So as we put that through the models there were various outcomes coming out of the models that just made no sense as we applied some reasonable top-down judgement onto those models and some of the outcomes were, frankly, quite extreme. I don’t think that challenges the models themselves. What it does is challenge the fact that the data that we’re using to underpin the models just needs to be updated. I don’t think what we’re seeing is, based on conversations I’ve had with several CFOs globally, anything surprising. I think all of us have had to take fairly substantive what we call underlays into the models as a result of just the unprecedented nature of the shape of the stress event we saw in the second quarter and, therefore, we thought it was appropriate, our auditors thought it was appropriate, the regulators thought it was appropriate that we adjust the model outputs for expert judgement. MANUS COSTELLO: And does that mean, coming out of this then – you talked about the recovery – well, mechanically it lowers the possibility of any write-backs, but does it mean that your potential for reduced ECL charges going forward is diminished? How does that play in the second half and into ‘21? EWEN STEVENSON: Obviously, had we taken the higher ECL charges consistent with the models, there would have been greater degrees of write-back, yes. MANUS COSTELLO: Okay. And just to follow up on it, just what is the actual process for how you make that adjustment? Is it done at a business level or is it at the top level? I’m just confused as to exactly what the governance around that model is. EWEN STEVENSON: It’s done portfolio by portfolio, country by country. It’s as much granularity as we can apply on the wholesale side, and on the retail side it tends to be done much more on a portfolio level by country by product. MANUS COSTELLO: Thank you very much. TOM RAYNER, NUMIS: Thank you. Good morning. EWEN STEVENSON: Good morning. TOM RAYNER: Hi. Two questions please. First one just on the linkage between the forward- looking provisions that you’ve made under IFRS 9 and how that will start to link through to RWAs through changes in probability of default. I’m just wondering if there’s been some sort of lag where IFRS 9 recognises the situation at an earlier point than you actually do in the capital model. So that was my first question. Second question – well, listening to what you said, again, on not commenting on distribution, but clearly when you do decide to restart it’s going to depend both on capital position and the level of sustainable profitability. My question is would you expect, when you do resume distributions, to be starting at a low pay-out ratio with the possibility of then growing or adding via special dividends and buybacks or are you more likely wait until you’re in a position so -- 2 of 8 -- that you’re comfortable to go back to your maximum distribution level, which you can then hold flat, as you had – as you were doing pre-crisis? Thank you. EWEN STEVENSON: I think there probably is a lag effect in terms of credit rating migration relative to IFRS 9 provisioning. We do think that we’ve seen, as per our guidance for the full year, a higher level of ECL charges in the first half than the second half. We think we’ll see more ratings migration and more RWA pressure in the second half than the first half, if you were to just look at ratings migration alone. And in fact, what we’re seeing is actually a slower pace of ratings migration than what we had anticipated currently. Some of that is probably due to continuing government support packages globally and we’ll have to see how credit performs, once those support packages begin to roll off in the next few months, but we do think we’ll see a meaningful amount of ratings migration into the second half. TOM RAYNER: And just on the same point, is it the case that as and when we see actual defaults coming through and a move into stage three that you will then see also a pickup in RWA procyclicality, because that will then trigger the change on the capital model? EWEN STEVENSON: No, we do a forward-looking view on ratings, so in some ways they are anticipating future defaults as well. TOM RAYNER: Okay. Thanks. EWEN STEVENSON: On distributions, I think it’s more likely that it’s the former rather than the latter version of what you said, Tom, i.e. that they will build over time. TOM RAYNER: Yeah, okay. Thank you. EWEN STEVENSON: And I think if you apply a regulatory lens over this as well, I think watch what happens with the Brexit negotiations in the coming months, watch what happens with COVID, because there will be a natural caution – regulatory caution, I think, as there continues to be a high degree of uncertainty around the economic impact of COVID and Brexit in the near term, but the more that we as a sector and the regulator can get comfortable that there is a tighter band of potential outcomes for ‘21 and ‘22, both in terms of profitability and capital, then you can be more confident on forward-looking distribution policy. One of the challenges at the moment is that we’re continuing to face into an environment where you can plausibly predict quite a broad array of outcomes in ‘21 and the flow through into capital. TOM RAYNER: Okay, lovely. Thank you. ALASTAIR RYAN, BANK OF AMERICA: Thank you. Good morning. I’m probably coming at things from the opposite side to Manus, not for the first time. On provisioning, there aren’t actually a lot of bad debts around yet. These moratoria schemes and government guarantees make it extremely difficult to judge whether they’re actually going to emerge, I suppose, as well as whether people have really already defaulted and you just can’t see it because of the moratoria. And that was extended again in Hong Kong, I think, yesterday. How do you approach that, because at some point governments do keep extending guarantees and maybe the bad debts aren’t going to happen, rather than a wave that’s coming when they expire, if they don’t expire, because governments are running bigger deficits and money is cheap? You’ve done more than you needed to, because you’ve taken a view that these things will expire and there will be this massive catch-up, and it’s gone into your capital through your risk-weighted assets as well, so you’ve sort of had a P&L charge and a capital charge against things that may not emerge. Just a discussion really on how we think about that from your point of view. Thank you. EWEN STEVENSON: Yeah, we had a discussion on this internally the other day. Realistically, if we roll forward to Q3 results, which are less than two months away, I still think it’s going to be too early to really make strong judgement calls around this. We are beginning to see globally some schemes roll off, particularly on the retail side. We are able to benchmark that against what the underlying assumptions were and it’s been modestly better, if I could describe it as that, from what we’d thought, but too early really to call. The UK in particular has got very substantive support currently. So realistically, Alastair, I don’t think we’re going to have a strong view on this until we get full-year results and, you’re right, I think -- 3 of 8 -- the likely government reaction will be to try to extend this and push it out as long as possible, but, ultimately, that government support can’t be sustained, given the affordability constraints certain governments are under. ALASTAIR RYAN: Thank you. And then just it’s – I mean, absolutely, it’s too early to tell, but house prices in particular haven’t fallen anywhere when back in March, April it would have been reasonable to expect that a fall would show up in a lot of places. Does that begin to play through or, again, you’d just – you’d review that at the full-year? EWEN STEVENSON: Yeah, in terms of the expected defaults we’d seen, the mortgage book was not the source of concern for us and never really has been, because we were not anticipating extreme falls in house prices, so it’s more on the consumer debt side, where actually, frankly, trends have been reasonably positive, because people are not spending on their cards and they’re paying down debt. And on the wholesale side, you can all see the very different cashflow impacts that COVID has had on different industrial sectors, so there are still a number of sectors that are running very cashflow negative at this point. And again, you’re getting into speculation, for example, in the airline sector, of when and how do governments step in to support national flag carriers. So I think it is too early to tell on a lot of these sectors, and even if we try to think about our own behaviour and how we’re likely to act as a user of services post COVID, I think our own behaviour will change. We’ll travel less. We’ll have over the medium term less need for commercial real estate. We’ll just be spending less on traditional stuff we would have previously thought we were spending on pre-COVID, so how that then factors into various industrial sectors and credit, if that’s translated across the sector, I think is still too early to tell. ROBERT NOBLE, DEUTSCHE BANK: Morning, yes. A couple of questions – the change in Fed policy was the other week to allow inflation to overshoot a steep in the US yield curve and short rates have remained pretty flat. Does that type of shift benefit your balance sheet in any way or should I only be concerned about the short rates? EWEN STEVENSON: Well, a positive yield curve is always a good thing, but relative to what we thought pre-COVID, we’re still looking at an interest rate environment that’s materially lower. You can read that Fed statement a number of ways, including the fact that interest rates are going to remain extremely low for a sustained period of time. I still haven’t got my head around what the trade is that drives inflation in a much different rate environment. So our current planning assumption is that rates stay low for a sustained period through ‘22, ‘23. ROBERT NOBLE: Okay. Thanks. You highlighted in your presentation unsecured personal loans increased in Q2. Is that Hong Kong-driven or UK-driven? And then on the credit card book, I think you said people are spending less, but my impression was the last two months post-Q2 results things have been getting better. Is that not the case? Should we not start to see credit card books grow in Q3 or do we have to wait a bit longer for that? RICHARD O’CONNOR: We are seeing increased spending on credit cards, but from a much lower base than our plan and assumptions based at the end of June. And there’s nothing much to say, Robert, on unsecured personal loans. They are broadly stable when you take exchange rates into account and you can see that the cards are substantially down, so we can take that offline. There’s nothing much to say on that front. But, no, we are seeing an increase in spend, but from a much lower base, and obviously people are spending more on credit cards but less in cash, so in terms of the impact overall interest income, we’re still seeing a recovery from a low base. ROBERT NOBLE: Is it – should the increased spend – should the credit card balance have shrunk in Q2, I guess is the question? RICHARD O’CONNOR: I think that’s right. I think the central plan would be we’ll increase as economies open up, but you can see from the balance that they are substantially down from the full year and we expect a gentle recovery from the base in June. ROBERT NOBLE: Thank you. And then just a quick follow-up on previous questions, why does the credit rating migration seem to wait and not take into account the fact that support packages will roll off? If you expect RWAs to increase because of your outlook, should you not immediately take that into account if you already think they’re going to increase? So -- 4 of 8 -- what’s the trigger? Do you actually need to see actual deterioration in borrowers’ financial position before you actually see RWA inflation? RICHARD O’CONNOR: Obviously, we’re revising the model all the time, but there’s always a lag between companies filing accounts, filing quarterly statements and you can see that in the published credit ratings. There’s always a lag there and you’ll see that in credit rating migration. Also, you’ve seen, obviously, credit rating migration so far being dampened by the fact that companies have raised a substantial amount of cash resource, and that’s been good, both on the debt and the equity side, but as Ewen said, you’re seeing many companies and sectors seeing substantially negative cash flow and we think that’ll feed through in that space, and that feeds into our guidance in terms of RWAs in terms of mid to high single-digit inflation for the year. You saw RWA inflation in the first half, but it was low single-digit, and so we’d expect that to continue in the second half and, indeed, into the first quarter or so of next year as well. ROBERT NOBLE: Okay. Thank you very much. RONIT GHOSE, CITIGROUP: I have two questions please. One is a factual one and one is a more maybe philosophical one. On the factual one, the comment on associates, so Saudi British, it looks like they took a $2 billion goodwill hit in their second quarter numbers that just came out. So do we just put in your share, so about $600 million, of the goodwill charge to reduce the associate line by that? Can you just walk us through the maths please, if you can? That’d be great. And secondly, on a more conceptual or philosophical – we think about capital return distribution and, given everything you’ve said, Ewen, should we start thinking about maybe – historically, you’ve had for the last decade or so a relatively high DPS pay-out ratio and you’ve used buybacks selectively to neutralise scrip. Is there a case here to be made for shifting more towards a buyback heavier mix of capital return vis-à-vis a high dividend? Thanks. EWEN STEVENSON: Yeah, so on the second one, look, I don’t want to get drawn into speculation on what we may or may not do with distribution policy in the future. As I said, ultimately, it’s got to be driven by a view on medium-term sustainable returns and medium- term sustainable growth, and then thinking through cash versus cash in stock, and dividends versus dividends and buybacks, but we’re still working through that at the moment. On the factual question, we own just under 30%, so the goodwill impairment maths is right, but there will obviously be offsetting earnings during the quarter, so the actual share of associate income will be different to that number. RICHARD O’CONNOR: Let me just add to that. We haven’t worked out the presentation in Q3 and we haven’t gone through the goodwill calculation at Group level yet, but we will show you the underlying associate income from SABB and Bank of Communications, and then the goodwill hit, should we decide to take it, will be shown as an exceptional item. EWEN STEVENSON: And that won’t have any impact on common equity tier 1. AMAN RAKKAR, BARCLAYS: A couple from me. One on costs, please – it’s actually quite interesting, your opening gambit on the office in Canary Wharf. I’m just trying to work out if your real estate footprint, be it a combination of your branch network and/or your headquarters property, is that a big part of the cost base? Is there anything that you can give us to point towards what proportion of your annual expense base that might relate to? I guess the reason I’m asking is presumably you guys, as part of your broader review and perhaps as you’re looking to respond to the revenue challenges might be looking to do something on your go-forward cost base. Is that potentially a source of material cost-saving that’s been perhaps capitalised by COVID and the structural shifts that you’re seeing, or basically, ‘It’s not a big expense as it is right now so don’t worry about it’? The second was just a point of clarification. You mentioned the steep fall in dollar that we’ve seen in recent weeks, and I just wanted to clarify that. I think you transitioned about a remark regarding the ECL charge and what that might mean for asset quality, but also in terms of what that might mean for the other P&L lines. Could you just repeat that comment? I think I might have just missed it. -- 5 of 8 -- Third is – you briefly intimated on the Q2 call about a potential funding opportunity and funding cost opportunity, potentially, in Hong Kong. Can I tease that out a little bit more? Could you help us understand what your average deposit cost is at H1 in Hong Kong and what it was at the end of last year? Because when I look at the system-level data, which I think is the only stuff that’s available, HKD deposits basically look like they were already on the floor going into COVID and are still at the floor now. I don’t really have visibility on dollar deposit costs, I guess, that you incur over in Hong Kong, but where is that? Where is the space to take down your deposit costs there, please? EWEN STEVENSON: Ming, I’ll come to you for the third one but I’ll take the first two. On costs, there are two trends out of COVID that we think will enable us over the medium term to rethink previous assumptions on costs to a lower level. How much lower, we’re still working through the detail on. The way that we work, i.e. as a workforce, I think will change as a result of COVID. We think there will be a sustainable improvement in some of the cost line items such as travel and entertainment; we spent $400 million in 2019, for example, on travel and entertainment costs. We expect to be able to basically get, now that we’ve proved video at scale works, we will return to a business model with a lot less travel. Equally, commercial real estate, we do think that previous assumptions on what we call desk-sharing ratios, we will be able to materially shift them upwards with more employees spending more time at home. Again, all of that we’re working through, but on the commercial real estate side the only thing I’d caution – and I’d use medium-term quite broadly – if I look for example here at Canary Wharf, we have a lease contract on this building that runs to 2027, so it’s probably more about optimising other office space we have in London into Canary Wharf in the short to medium term, rather than rethinking this building up to 2027. And it will be very much building by building dependent, globally, but we do think over the medium term that we would be able to save material amounts of money on commercial real estate costs. The other big trend we’ve seen is obviously customer use of digital channels, so much lower use of branches, much lower use of cash, engaging with relationship managers on digital channels rather than face-to-face meetings. All of that, again, we think will allow us to accelerate previous assumptions about how quickly we can get out of physical distribution costs, cash handling costs, number of relationship managers, productivity of relationship managers. Again, we’re working through that. Whether that has a material impact on the thinking for ’21, not so sure about, but certainly from 2022 onwards we think it will allow us to push the cost structure down further than what we may previously have thought. On the dollar comment I was linking it to costs. I know the high sensitivity around single percentage point movements in our cross-structure, but just noting the fact that not all of our costs are dollar-based, so with the dollar depreciation you will see some natural inflation in dollar-based operating expenses in Q3. That’s got nothing to do with our management of the cost base. RICHARD O’CONNOR: Two other quick things. Our commercial real estate costs are about $3 billion out of a cost base of a total of $30 billion, so we can take a big slug out of that over the medium term. No change to guidance. Obviously a large percent of our loan book is in dollars. There are exceptions, including the UK, but no, stick to the current guidance, and obviously that guidance is very sensitive to UK impairments, as you saw in Q2 and Q1. EWEN STEVENSON: Ming, do you want to take the third question on funding costs in Hong Kong? MING LAU, CHIEF FINANCIAL OFFICER, ASIA PACIFIC: Sure, Ewen. The way I would look at it for Hong Kong particularly in that comment on funding costs comes from two things. One I would say is, if you look at the mix of the deposit base, and this is also following industry trends, right? As rates are coming down we’re seeing a shift in terms of the mix in time deposits relative to current accounts. Just to give you a sense, the overall Hong Kong market as a whole was probably at about 50% on time deposits. That’s now come down to about 45%. Naturally for us we’ve historically had a much smaller base on time deposits, so we were probably at around the high teens in Hong Kong on time deposits, and now that’s coming down below 10%. That’s one factor that you need to think about in terms of why we believe, in terms of the costs of funding on deposits, we see that trend continuing down because of low interest rates. Secondly, the other bit impacting that is the absolute rates themselves are also falling materially. If I look at the three-month Hong Kong dollar time deposit rates on new money, earlier on in the year you’re looking at about 200 basis points for Hong Kong dollar and US -- 6 of 8 -- dollar time deposits. We’ve recently now dropped that below 30 basis points, so I think if you take the combination of those two factors on a forward-looking basis, we should still continue to see the trend of time deposit mix coming down, and I still think there’s an opportunity to reduce the actual rates themselves. AMAN RAKKAR: That’s really, really helpful. Actually, it sounds like it must be the US dollar deposit rates that you guys were paying at the beginning of the year that were more expensive though, right? Because the HKMA report three-month HKD client deposits at the beginning of this year at about 25 basis points, so you guys must have been paying a lot for US dollar deposits at the beginning of the year, right? MING LAU: Yeah, you’ve got to remember this is on new funds, so not on rollover deposits. This is for new money. MARTIN LEITGEB, GOLDMAN SACHS: A few questions from my side. Firstly, just touching on your opening remarks and on the outlook from here, I was just wondering, is the prime focus at this stage on costs in order to kind of offset part of the revenue headwind? There was a consideration to potentially address revenues; in some countries where there have been lower rates for a prolonged period of time banks started introducing new charges, charges on current accounts and so forth. Could that be a consideration that one tries to address part of the lower for longer rate outlook with incremental charges elsewhere? Specifically, just on the UK ring-fenced bank, it sounded like at the last quarterly results call that HSBC is being cautious on UK lending, and that includes mortgages. Is that still the case? Just looking at the excess deposit base you have now in your ring-fenced bank, is it the case that you would remain cautious there or could that be an incremental opportunity to accelerate lending there, particularly on the lower-risk mortgage side in order to try to offset some of the revenue pressures? The second question was with respect to the non-ring-fenced bank. I was just wondering if you could update us on the cost and capital progression, how we should think about that from here going forward. Is everything going according to plan and could there be significant step-up, potentially, in terms of RWA reduction from here over the coming period? The final point, just to touch on your comments on Brexit, if your prime concern is here that Brexit will essentially impact economic outlook and progression and potentially the shape of the recovery from here, or is this more a concern from an operational perspective or maybe from a central bank perspective that the Bank of England might become more competitive in terms of interest rates? EWEN STEVENSON: I think there were four questions there, Martin, so correct me if I don’t get them. So on things that we’re looking at to address the shift in interest rates, operating expenses is clearly something that’s well within our control. As I said, COVID has opened up opportunity both on the shift in physical to digital with customers, and also just the amount of money that we need operationally to run the bank as a result of shift in employee patterns as a result of how we work. We certainly need to think about how we pivot harder into sources of non-interest income, away from what now looks like an overreliance on net interest income. Whether that be through how we structurally grow some businesses now or think about growth and the importance of growth in some businesses that are not net interest income driven, but also how we think about the economic return of a customer where previously a lot of that economic return would have come from net interest income and shifting more towards fees. There have been parts of the world – Europe in particular – that have faced into a lower interest rate environment for a prolonged period of time, so there’s clearly a well-laid out playbook there for how those banks have reacted, so I don’t think our responses should be any different to that. The third area you didn’t mention is capital allocation. To the extent that we had underperforming businesses previously, the return challenges on those business have just become more difficult with the shift in interest rate outlook. Linking that to your comments on the non-ring-fenced bank, both on capital allocation and costs, we are definitely going to continue to drive both of them down quite sharply over the coming years. On the UK ring-fenced bank risk appetite – you will have seen for example in the last couple of days some speculation or commentary around us cutting back on higher LTV mortgage lending. We’re certainly comfortably open for business on what we consider to be lower risk lending, but we’re cautious on some of the higher risk areas in the UK at the moment. -- 7 of 8 -- On Brexit, it’s not an operational issue, it’s a macro issue. Brexit clearly will have a compounding negative effect on the UK economy in addition to COVID, and the harder the version of Brexit the more profound that impact will be on the UK economy. I know there’s another round of negotiations coming up. There still remains quite broad outcomes as to where the governments may be able to agree, but it was a comment purely about economics and not about our operational capacity. In some ways actually the harder the version of Brexit the better for us competitively, because we’re a diversified bank that’s not reliant on the UK, unlike some of our peers. We’ve also got a big global network, unlike many of our peers where, as UK corporates seek to reposition their relationships globally we feel well placed to do that. But we will be hit on the domestic side with a weaker macro outlook as a result of a hard version of Brexit. RICHARD O’CONNOR: Just something to add to that. On the ring-fenced bank on mortgages, some of the speculation on the high LTV is due to operational issues, i.e. we need to maintain our service levels, so it’s not just the risk appetite level but the fact that we need to maintain service levels in the face of very high demand. On Brexit, when we do our modelling for IFRS 9 on ECLs then Brexit is highly factored in and is already a factor within those models. As Ewen said, what formal agreement we get will impact economic growth during ’21, ’22 and ’23, so that will have an impact on the outlook for the ring-fenced bank in particular. FAHAD KUNWAR, REDBURN: Just a follow-up on a couple of things, actually. The Hong Kong question was quite interesting. Is the deposit beta that you’re experiencing in Hong Kong better than you’d expected when you spoke to us last month? I know the $25 billion 2021 guidance was obviously given at that point, but are you experiencing a better deposit beta on those dollar deposit or dollar lending rates than you expected at that point, or is it very much in line with the guidance? The second question I had is on Global Markets. I appreciate that the second half will be lower than the fantastic first half, but any sense on what the second half will be versus the second half last year, which – as you said – was significantly lower than H1 ’19? EWEN STEVENSON: Yeah, so on the second question, the second half last year did have some one-off positive events in it, but it’s a much better guide for what we think than the first half of this year. Certainly in terms of trading, I think what we’ve seen since June has fallen back very much more in line with what we anticipated at the start of the year versus what we saw in the first half. On the deposit beta in Hong Kong, Ming, do you want to handle that? MING LAU: Yeah, thanks, Ewen. I’d say on that no material change from what we guided a few weeks ago. The one thing I would say we’re noticing now is probably more on HIBOR rates themselves over the past week with a lot of the announcements on large IPOs and so forth, we are seeing a tightening of the rate. There has been a bit of an increase on short-term HIBOR rates, which would feed through to give us a bit of a benefit in September. FAHAD KUNWAR: Okay, thank you. Ewen, just to be clear, is the H2 ’19 level flattish on that kind of level? Obviously there’s going to be puts and takes, but that feels like the right kind of guide to H2 2019 as it stands at the moment? EWEN STEVENSON: Well, it’s – we’re two months into six months with a business that’s highly volatile, so it’s somewhat speculative. Thanks a lot for your time today, and if you do have any follow-up questions Richard and the team are fully available. As I said, as we look out at the moment we are very, very focused on how we need to structurally shift the business model to address what is an increasingly strong view that interest rates are going to stay low for a prolonged period of time, which is a mix of non-interest income, costs, capital allocation. On distributions, we critically understand the importance of distributions. We understand that we need to get back to paying them; it’s just too early to speculate at the moment, given some of the uncertainty that sits out there, but a lot of that uncertainty, I think, whether it be Brexit, the path out of COVID, US presidential election, etc, we will know a lot more in the next few months. I look forward to talking to you all at the end of October with Q3 results, so thanks for your time. -- 8 of 8 --