Lloyds Banking Group plc (LYG) Q2 2011 Earnings Report, Transcript and Summary

Thank you very much, António, and good morning, everybody. On this morning, I'm pleased to present the group's results for the first half of 2011. In addition, I want to spend some time giving you an update on our continuing balance sheet derisking through the run-off of noncore assets. And finally, we'll talk about capital, funding and liquidity. As you would have seen from the news release this morning, we have provided an enhanced core and noncore disclosure with more detailed cost and divisional financial information, and I hope that you will find this useful. Firstly then, looking at the business performance at the group level. In the first half of the year, the group delivered as a resilient performance in line with expectations with underlying profits up 36%, at GBP 1.3 billion. Underlying income, which excludes ECN movements in last year's liability management gains fell 12% to GBP 10.4 billion, but this is also after including losses on noncore asset sales of GBP 875 million, the sales of which facilitate a substantial Central Bank facility pay down during the half. With costs down 2% and impairments down 17%, this shows that the fundamentals of the business are indeed sound. Aside from these factors, the reduction of underlying income was partly due to a lower net interest margin but in line with the reduction in our average interest earning assets, both for the group and for the core business. Looking at our results on a statutory basis. I'll start with the reconciliation this morning by adjusting the underlying profits that we just looked at by, of course, including last year's liability management gains and the movements in each period in ECN valuations to show the combined businesses results. We saw reduction in the statutory result to a loss before tax of GBP 3.3 billion in the first half. In addition to the reduced profit on a combined business basis, which I just explained, the statutory results primarily reflects the PPI provision as well as further integration costs and the absence of the pension curtailment gains that we saw in the first half of last year. Let's now look at the group's income performance. As I said, group income decreased by 18% but underlying income decreased by just 12%. This increase includes a fall of $470 million in the core businesses, which I will explain in a moment, and GBP 875 million of losses on disposal of Treasury and other noncore assets. But please remember that these losses were largely offset by a related and accelerated fair value unwind, GBP 649 million, which is included low down in the income statement. Excluding those losses on disposal of noncore assets, underlying income fell by 5%, which is almost exactly in line with the reduction in average interest earning assets of 6%. Let me look now at core revenue performance. So if I break out that single core break that I have in the last slide, we can see here in detail the drivers of the 5% or GBP 470 million reduction in core income. The reduction is dominated by an adverse movement in banking volatility in other operating income. The reduction in core customer balances had only a small effect and was offset by other factors, including modest core margin expansion. Having noted the minor trend impact from core balances and the core margin, it's perhaps worth looking at the overall core profitability and performance. As we saw, core underlying income decreased by 5%, which principally reflected subdued new lending demand and continued deleveraging. The core net interest margin increased slightly to 2.35% mainly reflecting the improved funding mix in the core business with the higher amount of deposits and the smaller amount of wholesale funding. The core impairment charge was flat on the year ago and down on H2 2010, principally reflecting a reduction in retail impairment charge driven by the unsecured portfolio and partially offset by an increase in wholesale, which is primarily as a result of lower recoveries on disposals of assets. Core business profit before tax was down 28% compared to the first half of 2010. But excluding liability management and the ECN effects, core profit before tax decreased by just 6%, again broadly in line with the smaller balance sheet. So having touched on the core margin here, perhaps it's now time to look at the group net interest margin in more detail. The net interest margin was 207 basis points in the first half, compared to 208 this time last year and 212 in the second half of 2010. The H1 margin is actually a couple of basis points higher than I had expected, as we enjoyed 2 favorable factors. Firstly, the benefits in the calculation of running off some low-yielding noncore assets. And secondly, the real benefits of a change in funding mix with an increasing proportion of funding income coming from retail deposits, rather than wholesale funding and this mix benefit is happening faster than expected. As I've said on a number of occasions, predicting margins is far from easy. But as I look forward, and I still expect the full year margin to be just over 2%, which implies a second half margin of around 2%, of course. This guidance is supported by the growing impacts of the annualization effects of repaying relatively cheap Central Bank funding and also the increasing weighted average cost of wholesale funding, including that the effects of the fairly expensive H1 issuance will only be fully felt in the second half. In the medium term, the current margin trend will reverse, driven as we have guided before, mainly by increasing base rates and lower new wholesale funding costs and an improving funding mix. The low wholesale funding costs will come both from lower absolute issuance needs and the lower issuance spreads that we expect as funding markets settled over time and recovery sets in. These factors were, as I have said before, support the group margin expansion to our target range of 215 to 230 basis points by 2014. Let me turn now to our cost performance. During the first half, operating expenses was slightly down, 2% in fact. Benefiting from further integration savings and lower levels of operating leased depreciation and absorbing increased employers' national insurance contributions, higher VAT and inflation, et cetera. We've not been able to accrue for the cost of the bank levy during the first half of 2011. However, we continue to expect the cost of the levy for the full year will be approximately GBP 260 million. If this cost had been spread evenly and therefore into the first half, cost would have been broadly flat in line with our previous guidance. I'd now like to spend some time on impairments. The group showed a further reduction in the impairment charge in the first half. The charge was 17% lower than in the first half of 2010 with higher charges in Ireland and Australasia, more than offset by improvements elsewhere in the group, particularly the substantial fall in the wholesale division's impairment charge. Impaired loans increased by just 1% compared to year end and now represent 10.6% of closing advances. The group's coverage ratio over these is just over 45%. Both core and noncore impairments reduced, and I will draw out some comments on this as a look at divisional performance. Retails impairment charge reduced by 12%, driven by the unsecured portfolio and supported by improved new business quality and the stable economy. And even though house prices remain depressed, our core credit performance remains strong, with a number of customers entering arrears lower in the second half of 2010, both in the secured and the unsecured portfolios. As expected, the security impairment charge increased, which reflects the predictive movements in house prices. However, and also as expected, the unsecured impairment charge decreased, in fact about 32%, reflecting the improved quality of new business written over the last few years. In commercial, which we are reporting separately for the first time, the impairment charge decreased by 16% as we saw an improvement in the overall credit quality of the portfolio. At the same time, the stabilization of the economy in commercial property prices, combined with low interest rates, led to an overall reduction in the level of defaults. The wholesale impairment charge materially reduced by 44%, driven by the significant reduction in a noncore businesses impairment charge. The overall charges or percentage of average loans and advances to customers improved significantly to just over 2% in the first half of 2011, compared with 3.11% in the first half of last year. The decrease has continued to be driven by the corporate real estate and real estate-related assets, together with the stabilizing economic environments in 2010, and so far in 2011, a low interest rate environment helping to maintain defaults at a lower level, partially offset the charge overall by increased impairments on leveraged acquisition finance exposures. In Wealth and International, impairment charges increased by 14% on the first half of last year, predominantly as a result of our Irish portfolio where in Q1, we anticipated further falls in commercial real estate prices. I want to look at the headline trend in our Irish portfolio in a bit more detail for a moment. Continuing weakness in the Irish economy resulted in an increase in impaired loans in the first 6 months of the year. Provision coverage levels have been increased due to actual, and as I said, anticipated falls in property values as we have discussed and we now have a coverage ratio of 56%, up from 54% to the year end and 42% a year ago. Although the portfolio is noncore and a dedicated U.K.-based business support credit team is managing the wind down of the Irish book, current levels of redemptions and recoveries are low due to a severe lack of liquidity in Irish assets. A problem which we do not expect will be resolved quickly. As it's topical, I'm now going to comment briefly onto Eurozone exposures. Our Retail and corporate exposures are unsurprisingly dominated by our Irish positions. Our Spanish retail exposure consists of secured residential mortgage lending, about half of this portfolio is to ex-pats and the other half is local mortgages. The performance of these books is fine, with average loan-to-value of about 63% and about 5% of it impaired, with the coverage ratio of about 30%. The Spanish corporate exposure is mainly local lending, comprising corporate loans, project finance and some commercial real estate. The corporate loans and project finance books have seen only modest impairments, 3% in fact is impaired, whilst the CRE book, which is only about GBP 400 million in total is 22% impaired and with a 49% coverage ratio. The corporate exposure we have in Greece relates to loans to Greek shipping companies where the loans are generally secured and where repayment is mainly dependent upon the international trade rather than being linked to the state of the Greek economy. So whilst we manage all of these exposures carefully, they are modest in scale and we are not unduly concerned. Looking now to exposure to sovereigns and local banking groups. More than 40% of these are secured throughout the covered bond or ABS structures. The group has minimal direct exposure to the sovereign risk of any of these countries and this includes national governments and central banks. The other banking groups' exposure shown here are mainly short-term money market and trading exposures or money market lines and repo facilities to some of the major banking group's in Spain and Italy. The counter-parties are a limited number of well-rated financial institutions with whom we have long-standing relationships and more details about this are set out in the news release this morning. Now back to the U.K. and the performance of our U.K. mortgage portfolio in the first half. As expected, the security impairment charge increased, reflecting less favorable house price forecasts. Pleasingly, the proportion of the mortgage portfolio was an indexed loan-to-value of greater than 100% improved slightly to 12.2%. Perhaps more importantly, however, the value of the portfolio with an index loan-to-value growth of 100% and more than 3 months in risk also improved slightly by GBP 0.1 billion and is now GBP 3.1 billion, still less than 0.9% of the portfolio. Let me also give you some insight this morning on forbearance activity. We are of course encouraged by our regulators to treat customers in financial difficulties fairly and forbearance rightly plays a key role in achieving that. Forbearance is also an important role in actually mitigating financial losses from impairment events, as all observers would I am sure agree. We applied forbearance under strict conditions and forbearance in our activity in our portfolio is therefore limited and well-controlled. Where there is forbearance, the asset continues to be reported as past due or is considered impaired, depending upon the arrear status. As a result, we're confident we're currently appropriately provided on accounts where there is forbearance. As you can see from the charts behind me, the number of new cases of forbearance has seen a steady reduction over the last couple of years and in addition, the number of mortgage customers new to arrears decreased again in the last 6 months. All in all, we continue to be satisfied with the performance of these portfolios. Let me now take you through the profile and management of the noncore businesses, which, as you know, excludes Verde. We're pleased with the progress made on our balance sheet reduction plans in the period given the challenging market conditions of the first half. On 30th of June, we updated our strategy to reduce the noncore portfolio and we set a new target to reduce the balance to be equal to or less than GBP 90 billion by the end of 2014. In the first half of 2011, we achieved a substantial reduction with noncore portfolio of over GBP 30 billion, resulting in a portfolio now amounting to GBP 162 billion. We expect the remaining book to account for less than or equal to GBP 65 billion of RWAs by the end of 2014 and as you know, we are targeting noncore run-offs and disposals to be net capital generative over the period 2012 to 2014, and I want to talk more about that specific point in a second. But first, let me look briefly at the continuing derisking of the whole balance sheet. Group risk-weighted assets fell by 6% in the first half, driven by the rundown of our noncore asset portfolio, strong management of risk and appropriate risk criteria for new business. We expect further modest risk-weighted asset reductions in the second half of this year, however, we expect these to be offset by the effects of the implementation of the CRD rule changes and RWAs will therefore be broadly flat from now until the year end. Let me go back to the full noncore business now and look at its overall performance. The 51% fall in noncore underlying income was primarily driven by the losses on asset disposals in the first half of this year, but remember the fair value offset. Excluding these losses, noncore underlying income decreased by 16%. The noncore margin also decreased, primarily as a result of higher wholesale funding costs and the income drag effect from increased impaired assets. We have been rigorous on how we allocate operating expenses to noncore. Only considering the expense as noncore where we have a very high level of confidence that we can manage the expenses down as the assets go down. The noncore impairment charge reduced principally as a result of the material reduction in the wholesale impairment charge and partly offset, as we have noted already, by the increased impairment charge in international, which principally relates to our Irish portfolio. Even though we have seen lower noncore income, the reduction in costs and the impairment charge led to an improvement in the noncore loss of about 25%. I think it'll be useful at this stage to give you an illustration of the capital and funding benefits generated by our management of the noncore book. Despite the continuing impairments and reduced margin in noncore, the rundown of noncore portfolios in the first half has been closed to capital neutral. The capital consumed by the loss after-tax in the noncore business has nearly been offset by capital released by the reduction in risk-weighted assets from their disposal and the adjustment in excess expected losses. But the progress made has further reduced the level of risk in the balance sheet and in addition, we have achieved the obvious and substantial funding benefit from the run-off program. As I said, we target noncore run-off being capital generative over the 2012 to 2014 period in aggregate, although not necessarily in every reporting period. Moving on now to capital, liquidity and funding. I'm going to do liquidity for a second. That's better. Our core tier 1 ratio now stands at 10.1%, reflecting the effect of the statutory loss and partially offset by the reduction in risk-weighted assets of GBP 23 billion. As you can see, the business generated about 50 basis points of ratio improvements in the half, albeit that this was absorbed by the PPI provision that we took in Q1. On 30 June, I also spoke our plans for the capital restructurings, which would reduce the total core tier 1 deduction under full Basel 3 relating to our insurance operations by about GBP 2 billion. That has now been completed. Achieving significant mitigation, equivalent to about 50 basis points of core tier 1 ratio under full Basel 3 and reducing the transitional rules impact from insurance to approximately 20 basis points per annum. We made excellent progress on our term-funding issuance plans. In the first half, we achieved GBP 18 billion of publicly-placed term issuance and in addition, a further GBP 7 billion of term funding via a series of private placements. We continue to expect to issue new funding between GBP 5 billion and GBP 10 billion over the second half of this year, across all public and private, secured and unsecured issuing programs and in aggregate. During the first half, the absolute level of group wholesale funding fell by 1%, despite the strong new term issuance reflecting first half maturities. Successful new issuance also allowed the group to maintain its maturity profile, with 49% of wholesale funding having a maturity date greater than one year despite, as you can see, a significant volume dropping into the less than one year maturity bucket. Turning now to the reduction in government and central bank debt. As we have said already, the group achieved a reduction of GBP 60 billion in the first half of this year, leaving just GBP 37 billion outstanding at the half year. The liquidity support from governments and central banks have various maturity dates, the last of which is on October 2012, and current plans assume the remaining facilities will be repaid in line with contractual maturity dates. As António said, we've also been successful at further improving our loans-deposit ratio. By the end of the first half, our loans-deposit ratio, excluding repos and reversed repos, had improved to 144% and we also further reduced our core business loans-deposit ratio to 114%. This reduction is partly due into the continued customer deleveraging and subdued new lending demand but at the same time, we are successfully growing total customer deposits by 3%, reflecting good growth in relationship deposits in retail and in wealth. And we continue to work towards group loans-deposit ratio of 130% or below by the end of 2014. So let me now summarize the material aspects of our first half performance and also reiterate our guidance for the full year. Our view on 2011 is broadly unchanged. We're continuing to see various pressures on our net interest margin as I have said but despite these effects, we're still targeting a net interest margin of just above 3% for 2011. As a result of increased margin pressures, we've seen a reduction in income and we clearly expect a further slight reduction in core income. As a direct result of cost actions taken in the first half, we continue to expect costs to reduce slightly this year and we remain on track to deliver our target integration cost synergies of GBP 2 billion per annum by the end of the year. We've seen a good reduction in the impairment charge as improvements in wholesale and retail more than offset the further deterioration and impairments in Ireland. Our overall impairment guidance and the division of -- by division comments remain unchanged. Additionally, we're very pleased with the noncore asset reductions achieved in the first half. Together with excellent progress on our term-funding issuance plans and our deposit growth, this has enabled us to pay down material amounts of government and central bank debt. And lastly, of course, we've seen that very strong improvements in the loan-to-deposit ratio, which now stands at 144%. In summary, the group's performance in the first half of this year was in line with our expectations and guidance and we are making good progress towards our medium-term targets. And with that, I will hand you over to Kate who is going to help us organize questions on. Thank you very much.

So everyone -- thanks, Tim and António. We'll open it up to questions. I know it's hard to restrict yourselves, but if we can just do a couple each, it will be the best way to get around the room. Tom?

This is Tom Rayner from Exane BNP Paribas. And just looking for a little bit more color, Tim, on the sort of trajectory now of the margin. I hear you've reiterated the 2014 target range guidance, which suggests we dipped below 2% in the second half or somewhere close to that. I think if you annualize the full year effect of some of the expensive issuance this year, that's another drag as we go into 2012. So I'm just wondering, are we looking at a sort of u-shaped or maybe a smile-shaped trajectory from here to 2014 on the margin?

Tom, thanks for the question, and welcome back. In terms of shape of the margin, I think you're thinking about it in a similar way to me. I'm thinking that the second half margin will be around 2%, and that will give us a full year margin of just over 2%, which is consistent with what we were saying at the 30th of June. The annualization effect of the items that you're talking about, so repayments of central bank funding, which as we all know, was relatively cheap when we had it and the wholesale issuance cost, which will be relatively more expensive, will continue to come through. So I do look at a smile shape in the margin, as you articulate. And what I think will turn that is unchanged from what I've been saying for several seasons now or reporting seasons, which is that we will see movements in base rates, and we will see lower new wholesale costs to us, both coming from lower spreads as recovery sets in but also, very importantly, from lower issuance that we have to make. And I think now that we have much less issuance coming forward, it's easier for us, if you like, to look into that crystal ball of the impacts of wholesale funding on our margin, because the amount of aggregate funding that I have, that is going to change because I'm going to issue new, is reducing, which gives me more confidence in the shape.

Okay. This is my second now. I'll just keep in on margins, so it's really only one.

All right. Just because you've just come back.

Just I wasn't quite sure from one of your slides whether the deposit -- it sounds like deposits pricing has been a positive, because you've been replacing even more expensive Wholesale. But my impression is that deposit competition is pretty intense in the market. So I just wondered if you could comment on that going forward. And also, within the Wholesale, I know you've spitted out Commercial now. But they looked like a fairly good margin improvement coming through in Wholesale, but there was a comment about market value of deposits. I just wondered if you could explain what the trends there were, please. Antonio Horta-Osório: Let me comment to you about the Retail deposits and the deposits evolution. We are very pleased, as we both mentioned, that we have increased Retail deposits by 3% in the first half, substantially above the markets. And we think it is key in that behavior the fact that we have a multi-brand strategy which enables us to segment and have appropriate offerings to the different type of customer segments we have in the different brands. And to give you one example, one of the biggest successes we have was in the ISA campaign through the Halifax brands, where as I've said, we have provided a product that was not linked in price but gave value dates of the day of the application, which was, from our research, one of the key points customers wanted and one of the key points of customer complaints. And therefore, in the total ISA campaign, we had a net market share of ISA ins. So all ISAs transferred into banks. We had a net market share of more than 30%, although we were not being -- leading in price. So we think this is a very good performance in terms of volume. And also, given that the deposits that we've asked in relationships, they are also better deposits in terms of cost, as you were saying, than Wholesale funding costs. So from both perspectives and the multi-brand strategy is key in this respect and having products in different customer needs. We are also progressing well in corporate deposits, where we are providing lending as customers want. But as you know, customers are deleveraging on the back book, and the relationship managers in the different corporate segments have been very focused on getting relationship deposits, which have been also progressing well. So this is the goal of each between the 2 segments, and this is a critical point in bringing our loan-to-deposit ratio down 10 basis -- 10 percentage points in only 6 months. And if you look at our core book, our core business, we are now within the range that we want of a loan-to-deposit ratio below 120%.

Joe?

It's Joe Dickerson from Espirito Santo. I just have 2 questions. First of all, I was wondering if you could -- you mentioned the expense of H1 issuance and wholesale funding. I was wondering if you could quantify, say, the impact in basis points relative to the second half of 2010 in terms of the increase in cost so we can think about that for the margins as we go through the rest of 2011 and into 2012. And the second question I have is just if you could provide some granularity. When I look at the net interest margin in the Retail business, it compressed more than I thought. And I was wondering, specifically, if you could comment on what is happening on the asset side of the margin there and if you're seeing any competitive pressures, say in prime mortgage and what the general trend has been on the asset side, so again, we can think forward on that number. That's all.

Okay. I'll take the first bit. We don't get into the level of granularity, I'm afraid, to answer the first part of your question on the issuance costs and what we did. But with 18 of the 25 having been public, I dare say somebody carrying to do the research could work out maybe like 18/25 of it. When I talk about it being relatively expensive, it was certainly more expensive than we had anticipated at the start of the year issuance would be. But actually, when I look at it today, and I'm actually very grateful that I don't have to be in the wholesale markets at the moment, to only need to do 5 to 10 for the rest of the second half of the year, puts us in a very good position in which to deploy our needs and select the programs and the structures and the tenures and the markets that we would need to access in the second half. So we're -- I don't get into that level of breakout of funding, but it is, of course, fully embedded in the margin guidance that we're giving for the second half in the full year. Antonio Horta-Osório: Sorry. Would you mind repeating the second part, because I lost [indiscernible]?

Sure. The second part of the question was on the Retail net interest margin and specifically, what is happening on the asset side of the net interest margin in Retail. In other words, are -- is there any competitive pressure on the asset side from, say, prime mortgage, et cetera? Antonio Horta-Osório: Our new business pricing continues to be above the stock price. So it has a positive trend there. But when you look at Retail as a whole, we have still a loan-to-deposit ratio of higher than one. And as the wholesale cost of funding is more expensive, that affects the margins we have in Retail overall. But I repeat, we have -- we continue to have better assets, new business pricing than stock price with the positive dynamic there. You have a negative dynamic in terms of savings, in terms of new business versus stock, although the new business pricing is better than our wholesale funding. And to what is impacting the Retail margin as a whole is the cost of the wholesale funding given that our loan-to-deposit ratio is shrinking, but it is still higher than one.

Can I just add. I think I might add something. I'll absolutely agree with what António said, and that gives you the dynamics of what's happening in there. But I understand your point about it perhaps moving more than people had expected. Remember, perhaps, that what we said back in February was that we had a certain amount of the funding cost for the whole business that were left in the center. And so we've actually changed marginally -- wrong word, the word margin. We've actually changed slightly how we are allocating costs around the group in order to reduce the amount of unrecovered cost in the center. So that dynamic also comes through in some of the divisional margin analysis that you have. But my focus is really on the group margin, and that's where we've seen the dynamics in play. We've seen the expansion of margins over previous periods. And now, we're seeing the impacts of the necessary repayments of central bank funding and wholesale funding come into play. So my focus really rests on where the group margin is going, and that's where I'm actually very pleased with the performance we've had in the first half. I hope that helps.

Manus?

It's Manus Costello from Autonomous. In the adverse scenario of the EBA stress test, you reported a cumulative loss on your commercial real estate portfolio of about 14%, which was pretty much the worse of the 90-odd banks that reported. And in today's release, you talked about concerns that the trends in that market are deteriorating again for your portfolio. I wondered why is that portfolio still so fragile relative to peers given the level of impairment you've taken on it already. And what are you doing to mitigate that? And I have the second question as well, which is this. You're noting to post -- you have just posted and you're likely to post a significant statutory loss for the year. I wondered if that was going to impair your ability to switch back on hybrid coupons next year. Antonio Horta-Osório: Okay. Let me start with some general comments and then I'll pass on to Tim. We are very comfortable that the level of provisions that we have in the different credit portfolios is appropriate. Second, we have the largest retail and commercial banking in the United Kingdom, and we see no material change in trends in terms of the different portfolios and in terms of the impairment trends. And we are the largest bank in this country. So I want to be very clear about this, that we are very comfortable, that the level of provisions in the different credit portfolios is appropriate, and we see no material change in the trends. Okay. Now I'll ask Tim to comment specifically on both your questions.

Yes. Thank you very much. I think it's important to remember that the EBA stress test uses a defined set of stress criteria. So then the result, a mathematical exercises that applies those to a book. I can't remember exactly. I'm going to look for a nod on the front row on this, but I think the EBA stress test required us to have a 36%, 38%, 36%? 36% fall in commercial real estate prices to be assumed. So that will be like something that is a feature of the stress test. So I think against that, actually where we came out of that stress test, I was very satisfied with it. I think that the other comment in our announcement was the trends. I'm not quite sure you should take away what we're concerned about commercial real estate, and I actually echo Antonio's comments there. Indeed, commercial real estate was one of the biggest drivers of the improvement in the wholesale impairment charge. So that has been the dominating effect in reducing the wholesale impairment charge. So if there's something left in that that concerns you, we'll be happy to have a chat afterwards.

And on the ability to turn back on hybrid coupons next year?

I don't think that -- the loss -- you'll come back to the statutory loss point. I don't change my position on that, in the answers that we gave to questions on 30th of June, because of course, the loss you are referring to was something that appeared in the first quarter and therefore, was factored into what we were saying in terms of future guidance as part of the strategic review.

Asheefa?

Asheefa Sarangi, RBS. I just have 2 questions. I noticed in the commentary on -- there wasn't really anything on IAS 19 or the need, as of January 1, to recognize the unrealized actual losses of $959 million by that date. Should we be thinking about that coming out of the tangible book over the next few periods? Also with respect to CRD 4, we were thinking that you would fall under the regulation side of the equation. And therefore, you'd fall under the financial conglomerates directive, which would be beneficial for the insurance deduction. But we see no mention of that within your statement. I'm wondering if you could comment on that.

Should we toss for that one? Can I do it in reverse order?

Yes.

On the FCD, I think -- I'm not too read what was written in the CRD 4 draft that came out in the middle of July. And I've seen some commentators, mainly European ones, look at this and say, "Oh, hang on a minute. This FCD group, to a slightly beneficial -- more beneficial treatment of bancassurance, looks like it's still open." My interpretation is slightly different, and I don't actually believe that that route will really have it remain open or if there is maybe route through that may be a continental route through to it rather than a U.K.-allowed route through into it. So at the moment, I'd love to be banking the benefits of that, but I'm not and hence, why I'm especially pleased to further mitigate it since the 30th of June, the Basel III insurance impact on us. As far as IAS 19 is concerned, what you're referring to there, I think, is the expiry of the pension corridor accounting, which is indeed coming down the track at -- in a couple of years' time. Where will we get to in terms of the deficit at that time will have to be brought into the IAS 19 or revised IAS 19 compliant treatment. So yes, mechanically, we fall into net tangible assets, although I hope by then that we had a positive progress and perhaps more than the NTA per share progress we made in the first half.

I'm going to be naughty and just ask one more question. On the German litigation risk that you pinpointed being potentially significant, the number of things that have come through that have been potentially significantly have been between GBP 500 million to GBP 1 billion of late. Is that what we should be thinking about? And I know you've given a direction on timing, but should that be coming through over the next 18 months, do you think?

Yes, this is an early disclosure we've made of a particular item. What we've got here is a small handful of individual complaints that have been made against closed book sales of a German operation. What we've had is a German regional court give here a couple of decisions, which have surprised us. And therefore, we felt it was appropriate at this stage to make a disclosure, although we're continuing to take legal advice on how we progress this. But it's very strong at the moment.

Rohith?

Rohith Chandra-Rajan from Barclays Capital. A couple of questions on impairments, if I could, please. And the first, specifically on wholesale. You highlighted a very substantial drop year-on-year in wholesale impairments, but if I look at Wholesale versus the second half last year, impairments up 20%. And looking at the quarterly disclosure, 2Q versus 1Q looks to have doubled. You mentioned lower recoveries. Just wondered if you could provide a bit more clarity around that. And so that's the first one. And then the second point, also in impairments, really just kind of revisiting your guidance which, as you say, is unchanged at a reduction in the impairment charge year-on-year. Looking at the guidance division-by-division, you talked about modest declines in Retail and Wholesale, sort of fairly flat in Commercial and down in Wealth and International. Just sort of factoring in some sort of 10% declines for Retail and Wholesale and maybe 15% for Wealth and International will get you to a sort of GBP 11.5 billion impairment charge for the full year, so a pick-up in the second half. Just wondering if you could comment on that analysis, please. Antonio Horta-Osório: Okay. I will start and then I'll ask Tim to make some comments as well. The Wholesale portfolio, contrary to Commercial or Retail portfolio, are more lumpy. The Wholesale portfolios are more lumpy. So when we look at them, the results were in line with what we expected. As I said, we don't see any substantial -- any material change in terms of trends. And the fact that it is in Wholesale, you have lumpier movements from quarter-to-quarter. And in terms of our Retail portfolios, where the trends, and in Commercial, are much more clear, you have much more linear trends. In terms of our overall guidance, we are not providing guidance for impairments for the second half of the year. But we have provided guidance for the whole of the year, where we continue to expect impairments to be substantially down year-on-year. Now Tim, maybe you can comment with a bit more detail on this point.

Yes, certainly. I absolutely agree it's -- that the Wholesale, by nature, is going to be lumpy, and therefore, there will be movements quarter-on-quarter. I think I would prefer to continue with quarterly reporting on the basis of improved transparency should be good, but one has to understand that with will come some lumpy movements. I also made reference, Rohith, in my prepared words to some additional provisions. We've taken on some of the leverage exposures, for example, and that's a feature of the second quarter. But if you look underlying this, not only is the actual full year guidance unchanged but also the performance of the core book continues to be very strong indeed. If you had the time to look, I think it's about Page 47 or 48 of the news release, you will see there that there's about GBP 79 billion of core lending in the Wholesale division, and we've taken an impairment charge of about GBP 400 million. And that's a very solid performance indeed. I think the other comment you made around recoveries -- I mean, to me recoveries are always a bonus. And the fact that I have lower recoveries in the first part of this year still means I'm having recoveries. And I guess I look at it, and I say to myself, "What that tells me is I have further evidence that my impairment provisions are appropriate, my marks are sensible, and we are managing the exposures that we have in a very measured and balanced way." We've made good progress in the first half of this year on non-core disposals and as you have seen, and that's given us significant funding flexibility. So overall, I'm very pleased with the performance.

And just in terms of the interpretation of your divisional guidance. So modest equals of 10% reduction, further reduction for Wholesale -- sorry, Wealth and International, maybe 15%. I mean, is that the right ballpark to think about, sort of that sort of level of reduction year-on-year?

Well, I go back to what Antonio had just said, which is where -- the guidance that we give is full year versus full year, and the guidance that we gave in February and that we reiterated again at the end of June stands. And so I think the word modest is the same word that I used back in February, and we used in June. So I'm not going to get drawn into whether the first half can be interpreted as defining the word modest, okay? I'm sticking with where I was full year and full year, and I'm very happy with that. I think where you extrapolated it to though did concern me a little bit. I think you said that if you take all of your maths, and I couldn't write it down fast enough, you ended up at something over about GBP 11.5 billion of full year impairments. I think if you have that in your model, then you're way above where consensus for impairments would be, which is, from memory, is a high 9% in terms of -- is that right about?

9.6%

9.6%, a mid-9. Now, I'm not going to answer your follow-up question, which is am I happy with consensus. But what I can tell you is I have some concerns about people who are modeling it at either end of the spectrum. And if you wanted to interpret it at one particular way then that's up to you. But this is being recorded, and I'm not allowed to say any more.

Chris here on this side.

It's Chris Manners from Morgan Stanley here. Just a couple of questions. The first one was on the capital in the Insurance division. If you -- sort of taking GBP 2 billion out, firstly, how will rating agencies react to that? And does it matter? Secondly, if you're going to be sort of doubling your revenues in bancassurance, won't you need more capital rather than less in the division? And the second one was on funding. Obviously, the funding market is very tricky at the moment. You're saying you need to issue GBP 5 billion to GBP 10 billion for the rest of the year if the markets remain as harsh as they are at the moment. I mean, could you consider not issuing anymore term funding for the rest of the year and running down your liquidity buffer? And if so, how far could you take that down and remain comfortable? Antonio Horta-Osório: Thank you for your question. I will start with the third one and then I'll ask Tim about the first ones. In terms of funding, I think what we did on the first half of issuing GBP 25 billion well ahead of our initial plan while the markets are -- were open, number one and reducing more than GBP 30 billion in terms of noncore assets, number two and number three, increasing deposits by 3% on the 6-month alone is a clear indication of how robust and how much stronger our funding position is. That's why Tim said that we only need, within our original plan, to issue between GBP 5 billion and GBP 10 billion, but we are going to continue within our plan of having capital liquidity in terms of the norm. While we will continue for the future to decrease non-core, we are in a much better position after having decreased more than GBP 30 billion in 6 months alone. We are in a very good position also to choose when to issue even that we only need to issue in our plan between GBP 5 billion and GBP 10 billion. We are continuing to progress further in our deposits, so we are quite comfortable with the funding position. And I think with hindsight, it is even more justifiable what we did in the first half, because as you know the markets have been shut for the past 2 months as you correctly point out. And in terms of liquid assets, apart from having GBP 100 billion, GBP 101 billion to be exact. It's GBP 101 billion of primary liquid assets. We have on top of that secondary assets uncumbered that we can use at any point in time. So the 2 together, if you look at the, we have GBP 218 billion of primary and secondary liquid assets. So we are in a very, very good position.

Should I comment on the Insurance bit. I mean, the restructuring I referred to, you shouldn't read as a repatriation. It's a restructuring of the Insurance and the -- and that division. And obviously, we have considered and discussed and shared it with the rating agencies, and I'm not expecting any issues from that that will cause anybody any concern at all. The capital structure in our insurance groups is perhaps now slightly more in line with some of our competitors rather than being -- all being eventually being provided through common equity. So if you like, we're normalizing something, and that's a sensible way to mitigate the impact of the forthcoming Basel regulations. In terms of what does that do to the catalogs, on Page 70, it's the left-hand side -- 73 of the news release, you'll find the IGD surpluses of the 2 Insurance businesses. One of them is GBP 1.2 billion, and the other's GBP 1.7 billion, I'm sorry. I can't remember which way around they are for widows and HBOS, but there are very strong IGD surpluses in those businesses. In terms of the impact of the growing bancassurance business going forward, I'd have to refer you back to the slide that we -- well, I gave in February, which showed the increased capital efficiency of new product sales, where I mapped out the much reduced capital strain from the redesign of products that has taken place in the bancassurance offering that was launched in July of last year. And that's a good indication of the capital efficiency improvements that we've made in those businesses. So it's a very strong position.

Michael?

It's Michael Helsby from Merrill Lynch. I've got 3 questions, if that's all right. One's a quickie. First of all, on your non-core runoff costs that you've allocated, how should we think about those costs running down? Should we think about them running down in line with assets? Or is it going to be a lot lumpier than that? And also, is there a double counting of those costs within the simplification, because you haven't broke out the non-core when you produced the strategy [indiscernible]? So that's question one. Question 2 is pretty much around the fact that you've got very substantial deferred tax assets, clearly, at the moment. You've got 0 expected loss. I think there's a bit of confusion certainly around, because you've said that you expect a GBP 4 billion expected loss deduction from Basel III. And I'm wondering if there's a bit of double accounting going on with DTA. So should we think about the DTA balance and the expected loss kind of running in parallel together? And then finally, I just wanted to clarify on dividend policy. Now I know this is -- it's quite a way off, but I think it's just quite important. When you think about dividends -- and I know you're not going to be starting them just around the corner, but should we be thinking about dividend policy on a transitional Basel III basis? Or should we be thinking about dividend policy on a fully loaded Basel III basis? Because clearly, it makes a hell of a difference in terms of timing. Antonio Horta-Osório: In relation to the noncore costs, we have only allocated to noncore the costs that are directly related with the business, okay? So the cost that's -- we will run down according to our plans after the end of the rundown to the specific assets, and those include both the rundown of the assets and the simplification program that we have presented to you. So there are no noncore assets, which if you want, are not directly related to the noncore and neither of the assets, okay? So we don't have more noncore costs allocated there, which will then be sticky in the end, which I believe is your question.

I'm just trying to -- are they -- are those noncore costs tied up with your simplification process? I think you just said that they are. Antonio Horta-Osório: Yes. They tie up to our total cost projection until the end 2014 as we have showed on the 30 of June. On your Basel -- on your dividend question, I mean, what we said in 30 of June is what we can say at the moment. We will wait for our -- for regulatory clarity in terms of capital requirements, and we'll resume our dividend policy when we have clarity and our capital requirements will be presently met. So we cannot say now, at this stage. We know how important this decision is. We are absolutely committed to doing that, but we do not have clarity at the moment to be more specific [indiscernible].

No. I think we need further clarity on the regulators and where they're going to be on this going forward. And of course, we -- I'd like to think we'll get that in the coming months. There's a big G20 and Basel group meeting in November and from which we may get some more smoke, hopefully, of the right color. I think your second question was around tax and expected loss. I think you were there -- I'm talking about, again, Basel III impact. Is that right? Well, as I said back at the end of June with the strategic review announcement, we obviously may have an impact in the Basel III transition from any deferred tax assets that are still on the balance sheet when we get into that time. And obviously, that's a factor of future profitability, so one would hope they will be somewhat smaller than they are today, for example. As far as the excess expected loss, there's quite a complicated interplay in here with how fair value works. But I'm -- I absolutely stand by what I said in the 30th of June, and you are right to recall, therefore, that I've added 20 bps, i.e. GBP 4 billion, roughly 1% quarter, one spread over 5 years, 20 bps per annum.

Rob?

Robert Law of Nomura. Could I ask 3 as well as Mike got it? Firstly, in the area of capital release from noncore rundown, I know you don't make that commitment for the current year. Could you comment on whether you expect noncore rundown to be accretive for the current year, and if not, what kind of noncore risk asset rundown you have in mind for this year? That's the first question. Secondly, what are the fair value balances now? I can't see them in the release but maybe they are. And finally, in the area of liquidity. When I realized, obviously, you're restructuring your balance sheet to reduce this, but can you give us some indication of how much higher your liquidity holdings will have to be at this point if you implemented currently proposed Basel III LCR requirements?

Okay. Should I kick off? Let me first comment about noncore. I think we ran down -- I'm trying to remember that. I think it was about GBP 23 billion, GBP 24 billion in the first quarter, and we're saying GBP 31 million for the first half. So you can see that even there, it has an element of different pace in what can be achieved. The amounts of noncore runoffs still to do to our target of GBP 90 billion or less is about GBP 70 billion, and there's 3.5 years to go, which would imply, if you straight-lined it -- and I'm not suggesting you model a straight line -- about GBP 10.5 billion. You can see that on what we've achieved in various halves has been uneven and consistently uneven. And I think what this reflects, Robert, is our policy here, which as we set out in June, is about balancing the trade-off between the funding benefit that you have from the runoff versus the capital either consumption if you take a loss on selling versus release from the RWA disappearing. And of course, all that has to be balanced with the risk that is inherent in whatever asset you're seeing. So I don't, therefore, project a figure for noncore runoff for the second half, but I do expect it to be -- we're not going to do another GBP 31 billion, I mean, if that's what you're thinking, not by a long shot. In terms of will it be capital generative in the second half, I don't know. We'll take a view as each opportunity comes along. What I'd like to do is be -- and this is our target -- is to be capital generative over the 2012 to 2014 period. But as I said in my prepared words, I don't say that I'm going to do that in every reporting period, because it will depend on the balance of liquidity, risk. There's 2 legs to this. Capital disruption through the income statement is just one part of it. Capital release from the balance sheet side is equally important. And even at the end of 2014, we're likely to have GBP 6.5 billion of core tier 1 tied up in our noncore portfolio, really delivering not a great deal indeed. So nurturing the capital asset noncore is a very important part of what we want to do, because it will give us huge flexibility into the future

Yes. That's the reason why I'm asking the question, because obviously, you're putting some kind of floor under the prices you expect to get in the '12, '14 period, but...

We'll still be making recoveries on what we sell. I think that's just one part of your question. Antonio Horta-Osório: I think it's the importance of listening. For example, when you said the first half, where it was broadly neutral, that you have to take into consideration that a big portion of the assets that we sold was not for capital reasons. It was because of the liquidity progress. And those assets, some of them in terms of treasury assets, are much more relevant in terms of risk-weighted assets, you see. So you can see the focus and attention that we are having on the noncore portfolio on the first half of '11, where in spite of liberating more than GBP 25 billion in terms of liquidity, we have been brought to neutrality in terms of capital generation with the secondary target of improving the liquidity position of the group, as we just discussed. So that is the type of mindset that we have in addressing this. And going forward, as Tim just said, in terms of needs of selling for liquidity considerations, we are going to have a very different profile even what we did through, with the hindsights, in half 1.

I think the other part of your question was on liquidity levels. Honest answer, Rob, is I haven't done the math to see what I would need today if I had to comply with everything that isn't coming in for the future indeed, I mean, not so far as 2019 LCR, 2015. So all this is into the future. What we said on 30th of June is we intend to meet these -- meet both the ratios by 2014. And as part of getting there, we would expect to see liquid assets rise, and I would expect overall liquid asset levels to be roughly equivalent to our less than 1-year maturity wholesale funding at that time, which will be less than it is today. So I guess from that, you can derive the tram rails within which the answer lies.

And the fair value balances now?

I haven't got a stock figure for that in my head. What you will find though is that the best place to look for it is within the credit risk disclosures which are broken out by business area, which is about 2/3 of the way through the back of the news release. You will see the impairment -- the fair value item that relates to impairment, which is the principal levels, and that's broken out by division. The other main element is what's left on the fair value of HBOS owned debt, which has a long tail. In fact, if you look in the core businesses, you'll actually see fair value unwind there as a small negative, and that's because that bit -- that is the bit that relates to HBOS on debt. And of course, we regard the debt as supporting the whole balance sheet but only allocate to noncore that which I can manage to 0.

Gary?

Gary Greenwood at Shore Capital. I just had a question on sovereign macro risk related to Ireland in particular. There's a lot of concern at the moment about the risk of a breakup of the euro, and I was just wondering if you have considered the potential impacts on the business if Ireland was to come out of the euro and devaluate its currency and whether you've got any contingency plans in place for that scenario. Antonio Horta-Osório: Right. Well, we were very, very mindful of giving you full disclosure in terms of all the proprietary countries. So you have in there, an asset team made a summary of our disclosures in terms of Ireland, Portugal, Spain, Italy, even Belgium in terms of the different types of securities and in terms of the different types of assets. On the Irish-specific risk that you mentioned, of course, if Ireland left the euro, which we think is a really remote probability, there will be a risk in terms of foreign exchange, as you say, because we have assets in the country, and we don't have deposits in the country, which is a risk that we monitor, that it is included into our risk calculations and that we have -- we are cautious about that it's a risk which you have to consider, but that's not stretchable. You cannot stretch that type of risk. So you can consider different scenarios. You can consider probabilities, but it's a risk that you cannot stretch in any other country.

How would you react to it if it happened? Antonio Horta-Osório: We think it is really an extremely remote probability in terms of that happening. We are not at all thinking that that might happen. Although, in terms of our risk appetite and the different scenarios, we have considered that risk, which we know it exists. But it's a risk that given there are no liquidity -- there is no liquidity in terms of assets in Ireland. There are no transactions, the portfolio is run down according to time and you know that we have already provisioned 56% of the impaired assets -- but that's the risk, which is -- has -- does not have many levers at the moment. Tim, if you want to comment.

No, I agree with that. I agree with that. Antonio Horta-Osório: And that risk is the same to any other country you might want to consider.

Jason?

It's Jason Napier from Deutsche Bank. Two questions, please. The first just goes to the increase in the expected loss deduction and the core capital. I take the commentary that suggests that that's primarily of sell-down of high provision assets. But at GBP 1.2 billion across the capital base, it's 20 basis points of loans. It feels like a big number, and I just wonder whether you could talk about whether there's any inference at all about changes to your own view on terminal loss in the portfolio and whether that deduction might increase as disposals continue. And then the second question, just on deposits. I know for the number outlook, you've -- you sort of confirmed that base rates and perhaps, cheaper wholesale funding will drive you to the target over time. I just wonder whether in the deposit market, you're modeling any kind of change in deposit cost spreads and/or mix. Just noting that your international online balances were up nearly 100% in the last 6 months, it doesn't feel like a market where price can be anything other than very keen if you're going attract those sorts of balances. Antonio Horta-Osório: Okay. I'll start with the second one and then Tim will answer you on the first one. We believe, as you said, that as base rates start to increase, as the eventual increase by the end of the year or into early next year, we will have an impact which is both on our savings balance and also on our current account balances, which have different types, as you know, of reaction to base rate increases. So it's the 2 effects that will drive our profitability going forward from the liability point of view in the next few years. We are having very good behavior in terms of retail deposits, as we just discussed through the multi-brand strategy and attracting specifics to the Halifax brands, significant deposits not being price. And also, as you mentioned, on our international wealth businesses, we have been having very good success, including online deposits and attracting deposits. It's good prices and significantly lower in our Wholesale cost of funding. And the 2 impact together, as I mentioned on this presentation, is what drives the mix change in terms of the costs of liabilities, which we expect to continue.

So to clarify, the international deposits are also cheaper than the stock of Wholesale that you're refinancing now.

Yes. Antonio Horta-Osório: Yes, they are.

Yes, they are. Jason, I'll pick up on your second point. But Rob, on fair value, we put in up -- it's Page 28, rather the back of the appendices, you'll find the annual fair value unwind table that we've set out as usual. And you'll see there in the tail of that, in the outer years, that will recur for a few years, which is the own debt unwind bit, which is bang on what I said it was, about GBP 300 million. I hope that helps. Jason, the EEL I think you're asking about. I think, naturally, over time, we will see an increase in excess expected loss. So you can say, yes, I agree with your analysis, because what will happen here is we will see an improvement in the portfolio. So we will all see a reduction of the actual impairments that we're taking. But by definition, if you're taking less than the excess of the expected loss, that you would see a couple that comes out of the model over what you've actually provided would increase. So yes, I would expect that to increase over time. That's one of the reasons I wanted to give transparency on where I thought it would get to by the time we get into Basel III and the start of transition, which we did in June.

Peter?

Peter Toeman from HSBC. Just coming back to the margin again. I mean, thank you for giving us the numbers on the core business, and the margin is 2.35%. But on your own numbers, by 2014, this could be down to I think 2.10% on your 3-year forecast. Everything you said about deposit behavior, low issuance, cheaper cost of issuance is a positive. So why -- what are the negatives that you've got in your 215-basis-point margin assumption?

Peter, let me help you out. The margin guidance that we gave is actually for the full group, whereas -- which was a range of 215 to 230 bps by 2014, but the group margin would be lower than the core margin. So I don't recognize the 2.10%. The bottom of the guided range is at 2.15% by 2014, and it's a group margin. And I would very much hope and believe that we will continue to see the trend that we see now: the core margin being someway higher than non-core. Does that help you?

It's just that the noncore assets run down, so the trend of the margin should be closer to the core rather than to the mix of the core and noncore.

Yes, over time. Although, sadly, I will still have about GBP 90 billion of noncore assets by the end of 2014. I'd love it to be lower. And inevitably, some of those will be poorer returning assets. Remember that the way -- I actually calculate our margin on a gross basis. So when I refer to income drag in the margin, it's because I'm still reporting our margin over gross assets. And therefore, the more impairment that we take on noncore assets, I get more of the margins drag just because of the way it's calculated. I think it would be -- I don't think it'd be right to switch to doing it on a net basis, because I get an artificial boost in the margin, which really wouldn't reflect what I would think would be a fair and reliable trend.

I think we've got time for a couple more. Mike is on here.

Mike Trippitt at Oriel. I just want to ask you about the average risk weighting. The comments you've made about excess expected loss, does that same calculation drive up the risk weighting now from where it was in the disclosures at the end of last year? And secondly, what happens -- what's your understanding of the Project Verde process if the banking commission makes a firm recommendation for an enhanced divestment? Antonio Horta-Osório: Let me start with the second one on Verde. We have -- as been -- we discussed this in June here as well. In relation to Verde and the ICB, we started the Verde process in March, because in our opinion, the timetable was short or was not, if you want, we did not have a lot of leeway to have -- till the end of 2013 to do not only the negotiating process but also the separation and the completion of the sale. In our opinion, it led us to start immediately, point one. Secondly, related to that, as we finish our integration process, we are liberating a lot of the experienced resources that we can use in terms of together with the buyer, planning the completion of the transaction. So these 2 reasons together were why planned the start of that already. How is it the in terms of Verde? We have received a number of credible bids. We are analyzing those bids. This is quite a tailored-made process from now onwards, with each the other bidders, and we continue to expect to be able to find a buyer by the end of the year as per our original plan. In simultaneous, as you we were saying, we have engaged with the ICB, posted preliminary reports, and we are continuing to engage. We have had a good and sensible dialogue with the ICB, and we are progressing, and we will continue to engage ahead of the final reports. I'm sorry. I can't tell you much more at this stage, but I think we are well aligned in our plans of tying the 2 things together.

But if there was an enhanced divestment, the process has to be delayed surely. And that doesn't comply then with the EU. Antonio Horta-Osório: Well, substantially enhanced has not defined. We have worked substantially since the preliminary report in enhancing the balance sheet and the availability of funding lines to the project for a little there, since we are going to sell. And I repeat, we have had good and sensible dialogue with the ICB, which we'll continue to have until the final report. So we do not foresee delays on our process of finding a buyer for Verde until the end of the year. You also know that the buyer, at his own option can choose to buy less mortgage originated -- less intermediary originated mortgages and have a smaller balance sheet with the limit of the loan-to-deposit ratio of our Retail business. So we do not foresee, for example, that the numbers have been circulating in terms of funding gap which were related to the 2010 numbers and the broad balance sheet will represent a problem for the several buyers we are speaking to.

Mike, the first part of your question was on risk weightings, I think. I think that any impact there would be outweighed by the benefit that we would see from weightings from the lower risk business that we're writing. And we're seeing the first benefit of that is coming through, for example, in the nice -- very nice drop in unsecured retail impairments. And that's an impairment charge rather than risk weighting, but it's indicative of the improving quality of new business that being written. So we saw a small risk-weighted asset reduction in the first part of the year even from our core portfolio. This is a good indication of improving quality.

One more question. Over here.

It's Ed Firth from Macquarie. And sorry, I didn't want to particularly bring you back to wholesale impairments. But just looking at -- on Page 115, if you look at the non-core Wholesale book, your impairments are up almost 40% H1 compared with second half last year. And yet, your coverage ratio was actually down from 46% to 41%. Now I hear what you say about it's a lumpy charge, but we're talking about GBP 400 million of additional provisions there. So could you give us some creative guidance as to what's going on and why you're happy seeing your coverage ratio falling?

I'm very comfortable with the coverage ratio. I believe the coverage ratio is an answer. The coverage ratio is what comes out when you have done a asset-by-asset assessment of what you need to take to reflect the reduction and the value you will derive from that asset. Also in the period, we actually wrote off some of the impaired assets that we had. And those would have carried a weighted average, higher level of coverage ratio. So if you actually look at the level of impaired loans on that same page, 115, they're down from to GBP 27 billion to GBP 25 billion, and part of that would have been -- write off some of it would have caused being recoveries. And some of those recoveries would have been highly impaired assets, where we were absolutely delighted to make a recovery in the period. So to me, the coverage ratio is on an output rather than something I used to measure the overall appropriateness. And that's particularly pertinent in a book like the Wholesale book, where we do that assessment on an asset-by-asset basis, looking at the cash flows we expect to flow from the exposure. Antonio Horta-Osório: And this is the only way you can do it on the Wholesale book. I mean, the coverage is not quite important when you look at Retail or small portfolios where you cannot on a statistical basis, because it's basically derived out of statistical models. And on the Wholesale, you really have to asset-by-asset and see the recoverability of your impaired loans and then when you add it up, you can have a different number. But as I told you, we are very comfortable that the level of provisions on our different credit portfolios is the appropriate one, and you have to take into consideration, as Tim just said, the write-offs, which obviously are done on a 100% basis in terms of loans and impairments.

Great. Well, thank you, everyone, for your questions. We'll end the session now.