Good afternoon, ladies and gentlemen, and thank you for standing by. And welcome to today's Credit Agricole 2019 Second Quarter and First Half Year Result. [Operator Instructions] I must advise you that this conference is being recorded today, Friday, 2 August, 2019. I would now like to hand the conference over to your speaker today, Jérôme Grivet, Chief Executive Officer. Please go ahead, sir. Jérôme Grivet: Yes. Good afternoon to every one of you. Thank you for having a wait up to now, up to today before taking your vacation. And I'll try to make this presentation as swift and quick as possible in order to give you your freedom for the rest of the day and probably the rest of the month. Let me start with Page 4, just a few headlines, a few key messages I just wanted to pronounce and start with. On this second quarter and first half of the year, Credit Agricole S.A. posted a very high level of profitability with its results. The operational efficiency continues to be a key motto for us, and we continue to have a cost-income ratio outside the contributions to the Single Resolution Fund, which is below 60% in the second quarter. The cost of risk remains low, and actually indeed much lower than our medium term assumptions. And last point, the solvency continues to shore up with an increase of 10 bps, both from the group and for the S.A. On the following page you have some key indicators about the market environment, nothing much to comment. I think that everybody has in mind what is happening presently on the markets. And so I can go now to Page 6, where you have the main figures for the result, net profit, both for the group and for Credit Agricole S.A., for the quarter and for the first half of the year and in stated figures and underlying figures. Just as a reminder, the figures for the first quarter and - for the second quarter, excuse me, and for the first half of 2018 were actually good figures which were boosted by a good level of activity and revenues, but also by a very specific risk situation especially at CACIB, where we had in the second quarter a significant loan loss provision reversal, and by a low level of corporate taxation, while in the beginning of this year, 2019, so first half and second quarter of 2019, we continue to have a very good level of activity and a good level of revenues. I'll go back on this matter a little bit later on. But obviously, the cost of risk has reversed within CACIB, and we have now a much more normal level of corporate taxation. This is why all these figures of profitability are slightly down as compared to the same period of last year. Nevertheless, we managed to post for the first half of this year a very high level of return on tangible equity at 11%, which is exactly the commitment we have taken in the last medium term plan, an 11% which is, as you know, calculated as twice the first semester of this year, which is a prudent way of calculating it because it includes twice the hit of IFRIC 21. On Page 7, a few highlights in terms of activity. In retail banking in the different networks that we operate, both in France and in Italy, a very good commercial momentum with not only outstandings up but also a very significant level of customer capture, with 140,000 new customers for the regional banks of Credit Agricole, for LCL and Credit Agricole Italia in the first half of this year. In asset gathering, we continue to see an increase in the level of assets under management and inflows in medium and long term assets. In specialized financial services, a good level of activity especially in the car financing businesses that we have, despite the situation of the car making industry in Europe. And lastly, in the large customers division, a very good level of activity. And CACIB continues to record some key positions in its different key businesses, like this quarter the position of number one in euro syndication in the EMEA region. You know that green financing was a key component of our medium term plan. And this quarter again we continued to boost our different positioning in the field of green financing, both through CACIB, which remains and continues to be the number one in the world arranger in green bonds, but also with Amundi, which is more and more one of the most recognized experts in green financing amongst the world of institutional investors. If I go now to Page 10 in order to analyze a little bit more in detail the way our results were made, as I said, the underlying net income group share of the quarter at CASA level is a little bit down at €1 billion 242 million, and at €2 billion 38 million for the full half of the year. I just want to insist on the fact that, for the first half of the year, asset gathering business division, retail banking business division, and specialized financial services business division post an increase in their contribution to the net profit of Credit Agricole S.A. with, at the large customer division, the slight decrease is fully explained by this phenomenon I already mentioned, which is the fact that the cost of risk is going back to a more normal situation, which is a charge and not a product. Despite this situation, I think that we have another way of actually looking at the evolution of the profitability of Credit Agricole S.A. on the quarter. I already mentioned that, in the second quarter of 2018, we benefited from a specific characteristic of the situation with the cost of risk at CACIB with a low level of taxation. And if we input these elements to the level of profits that we have in the second quarter of 2018, actually this would translate into not a decrease in the overall level of profitability, but an increase of around 7%. If we go now to the revenues of the second quarter, so on Page 11, what you can see is that, on an underlying basis, revenues are slightly up on the quarter and on the semester, 4.6% on the quarter, 4.4% on the semester. But if we zoom a little bit more precisely on the business lines only, the increase is much more significant, 1.9% for the business lines on the quarter and 1.4% on the semester. On the specialized financial services division, well, you can see that both on the quarter and on the semester, actually the level of revenues is a little bit down, €50 million on the semester. Actually, I think that the level of revenues does not fully translate the level of activity, because as you know the growing part of the revenues is generated through joint ventures, car financing joint ventures. And as you may have seen already on the documents that we've published, the contribution of this joint venture is very significantly up between 2018 and 2019. So actually, even in the specialized financial services division, actually the level of activity is up despite the fact that it doesn't fully translate into an increase in the revenues. In the large customer division, it's also interesting to note that, on the first half of the year, revenues are slightly up despite the fact that markets were a little bit difficult in this first half of the year, and we have seen some competitors posting quite significant decrease in their level of revenues. As far as the cost base is concerned, on Page 12, you may see that, on the basis of the quarter, the cost base is up around 2%. But this includes around €10 million of one-off costs that are in connection with the different and significant structural transactions that were undertaken by several business lines, namely consumer credit with the partnerships that were renegotiated with FCA and Banco BPM at Agos, within the insurance business division with the new partnership that was concluded in Spain with Abanca, and within CACF with the two partnerships that were - or that are underway in Spain with Santander and in the Netherlands with the public offer that was launched recently on KAS Bank. So part of this increase in the overall cost base is linked to one-off costs. On the semester, the evolution of the cost base is 1% globally and 0.9% for the business lines. So what you can see is that, for the business lines, both on the quarter and on the semester, we continue to have a positive jaw before - between, excuse me, the evolution of the top line and the evolution of the cost line. Let me go now on Page 13 to the cost of risk. And what you can see, I already mentioned it, is that we have seen a slight pickup in the level of the cost of risk on this second quarter of 2019. It's completely explained by the reversal of the cost of risk situation at CACIB, where again we had, in the second quarter of 2018, a loan loss provision reversal when we have this quarter a loan loss provision charge, which is the more normal situation. So on these figures, which are an average on the last four quarters, what you can see is that at CASA level globally we have now a cost of risk which stands at 25 bps, and at group level it is at 19 bps. In both cases, we are at a level which is significantly below, again, the assumptions that we have reassessed when we presented our medium term plan in the beginning of June this year. We had made the assumption of a cost of risk of 40 bps at CASA and 25 bps at group level globally. If we zoom a little bit more precisely on the cost of risk in the main business lines contributing to the overall cost of risk, what you can see is that, within the consumer credit activities, the cost of risk remains more or less stable, around 120 bps, so nothing much to mention, and no sign of significant trend towards any kind of deterioration. At LCL, it's a very stable level of cost of risk, around 16 bps to 17 bps very constantly in the last 8 or 10 quarters, so nothing significant. And I can add that it's exactly the same situation within the regional banks, where it's even a little bit lower at 13 bps. At CA Italia, obviously the trend - the decreasing trend is slowing down, but we continue to experience a decrease in the overall level of the cost of risk. It's now at 62 bps on the last four quarters. We are probably reaching the bottom, but no sign of deterioration either. And lastly, on the financing activities of CACIB, it is still at a negative level on the basis of the last four quarters. But as I said, on the basis of the single last quarter, Q2 '19, it's now a positive cost of risk. It's, as already mentioned, mainly the consequence of a single file on which we have made provisions. And again, we see no sign of a trend towards a deterioration on the financing activities of our corporate and investment banking activities. Let me go now to some more details on each business line, starting as usual with the asset gathering and insurance business division on Page 15. As I said, the total assets under management have reached a record level in the last quarter at €1,964 billion of assets under management. And actually, it's also a record in each of our different business lines within this business division for among the life insurance activities and the private banking and wealth management activities. And the contribution of this business division globally to the net profitability of Credit Agricole S.A. continues to be up 2% on the quarter and 1.5% on the first half of the year. If we zoom a little bit more precisely on the insurance activities on Page 16, what you will obviously see is that we have had again a very good level of activity, be it in savings and retirement or be it in the risk products, protection and P&C. Protection and P&C revenues are again up close to 7% second quarter on second quarter. And in the savings and retirement business, a very high level of net new inflows, €3.3 billion on the quarter. In the life insurance activities, we continued to see a slight improvement of the breakdown of the outstandings between euro denominated products and unit linked products, plus 50 basis points over one year at 22.2%. And in the different P&C and protection businesses, we see improvement in the number of policies managed in all the different sub-businesses. As far as the P&L is concerned, the net profit is up 2.2% on the quarter, 2.6% on the first half of the year, with a good level, a very good level of revenues, and operating expenses that are significantly up. But actually, it's mainly explained by a pickup in the level of tax expenses. And the strictly operating expenses, at plus 6.7%, are perfectly in line with the development of the business. I already mentioned the fact that we have signed, in the course of the second quarter, a very important partnership with Abanca in Spain and Portugal. It's important because it's a new country for the insurance activities. And it's important because after the partnership with CreVal in life insurance activities, it's now a partnership in non-life insurance activities that we have introduced. On Page 17, Amundi; Amundi has a record level of assets under management despite a small level of outflows - net outflows this quarter. But actually these outflows were only driven by money market funds, and we have had a positive level of inflows in medium to long term assets. And the P&L of Amundi continues to improve with, all in all, a contribution which is 4% up on the quarter and 3.5% up on the first half of the year. At LCL, same situation as in the first quarter of this year, good level of activity with customer savings up 6% and loans outstanding developing very well at 9.9% June on June, but also a very good development of the equipment rate of our customers with additional products and services, and a good development of the customer base, plus close to 30,000 net new customers on the first half of the year. When we look at the P&L, we continue to see revenues up 1.5% and costs down 4.6% if you exclude the contribution to the single resolutions fund. Cost of risk, which is down 10%, at a very low level, I already mentioned it. So net profit, which is up 6.5% on the quarter and even 11% on the first half of the year, and the cost-income ratio, which continues to decrease, being now below 65% on the quarter. Italy, more or less the same story. Customer savings are more or less stable, but it's the combination of off balance sheet customer savings, which are up, and on balance sheet savings, which are more stable or even a little bit down compared to June '18. I already mentioned the fact that we are progressively getting rid of costly customer deposits that were in the balance sheet of the three banks that we acquired 1.5 years ago. It's now coming to its end. And loans outstanding are up 3.5% if you exclude the disposal of NPL that we regularly realize. Despite - or in addition to this good level of, I would say, balance sheet activity, customer deposits and loan outstandings, we continue to have also a good commercial trend with 12,000 new customers, net new customers on the first half of the year, and a continuous improvement of the efficiency, commercial efficiency of the three banks that were integrated last year. In terms of profitability, so revenues are up this quarter 1.3%, costs are down 4.9%, and as the cost of risk itself is slightly down, 1.9%, there is an increase of 7% on the net income group share for the quarter and 7.5% for the first half of the year. Credit Agricole Group in Italy continues to be a very significant contributor to the profitability of CASA, with a net income group share close to €350 million on the first six months of the year, plus 19% as compared to H1 '18. International retail banking activities outside Italy continues to generate growing profits, plus 12% and plus 16% on the first half of the year, close to €75 million for the first half of the year. Nothing much more to mention on this aspect. Going now to specialized financial services, starting with the consumer credit and car financing activities, the gross managed loan book is now a little bit above €90 billion. It's a significant growth of 6.2% as compared to June '18, and it's especially dynamic within, again, the space of car financing partnerships. I already mentioned it. It's been also quite buoyant in Italy, in Germany, and in Portugal, a little bit less dynamic in France only. As far as CALF is concerned, a positive evolution of the leasing alone a book and a good level of activity in the factoring business in the second quarter, which is going to help boosting the revenues in the second half of this year. In terms of profitability, we have a situation which is a little bit contrasted between CACF and CALF. Globally, CACF is up in terms of profitability 2.5% on the quarter and 6% on the first half of the year thanks again to the contribution of the equity accounted entities, plus 21% and plus 24% on the quarter and the first half of the year. As far as CALF is concerned, the net profit is down for some specific and punctual reasons, be it one of a gain in the first half and the second quarter of 2018, a special dividend that is not going to be repeated, and also a level of revenues that is a little bit less dynamic in the second quarter of this year in connection with the fine-tuning of the risk parameter of Cash in Time, the new product that was launched last year. But the prospects are positive, and we expect a far better second half of the year for CALF. Large customers division, what you can see, as I already mentioned, is that on the first half of the year, the overall level of revenues is a little bit up, which is, again, a good performance in markets where we have had some turbulences in the last quarters. It's a little bit contrasted between capital market activities and investment banking activities, in which revenues on the quarter were a little bit down, and corporate banking and asset servicing activities both experienced a positive evolution of their revenues in the quarter, plus 1% for the corporate banking activities and plus even 6% for the asset servicing activities. In terms of P&L, on the quarter, revenues slightly down, costs slightly down, and cost of risk, of course, with this reversal of the situation, so all in all a net income group share which is down 17.5%, but still at a very high level for a single quarter at €461 million. And for the first half of the year, the profitability is slightly down, 8.6%, but close to €700 million for the first half of the year, so a good level again of profitability. I also want to mention that the risk-weighted assets in this business division is down between end of March and end of June. And thus, the return on normalized equity for this year comes in above 10% at 10.6%. Corporate center; you may have seen that the contribution of the corporate center decreased, or actually increased in terms of negative figures, both on the quarter and on the first half of the year. And we wanted to give a little bit more clarity on this part of our P&L. Thus, we are breaking down the corporate center into four different boxes, actually, which are indicated on this Page 23. The first box is the management of the balance sheet and the holding of Credit Agricole S.A. And this is generating a level of loss which is rather stable, around €250 million a quarter, and which is progressively reducing as we manage to reduce the overall debt of the corporate center or the cost of this debt, but it's going to reduce slowly. The second box is made of the different businesses which doesn't lie into one of the four main business lines, so it's typically the private equities activities, the real estate activities, and some other activities like editing activities. They generate a modest but positive level of profit each quarter, which was a little bit more positive last year than this year. The third box is made of support functions that are pulled together and managed by Credit Agricole S.A. for the account of the whole group, be it the subsidiaries of Credit Agricole S.A. or the regional banks. So this box is structurally generating a level of profitability which is zero, but there might be some positive or negative small numbers on a quarterly basis depending on the speed at which the costs are being built to the different users of these pooled resources. And the last category is made of some other elements which are volatile, which are typically the consolidation adjustment that we have to make when we consolidate the figures of different entities at CASA level. And in this fourth box, we have had an abnormal, I would say, high level of positive contribution in the second quarter and first half of 2018 which is not repeated this quarter and this first half of the year, which explains the decrease in the contribution of the corporate center to the level of profit of Credit Agricole S.A. If we go now to Page 25, the regional banks, you can see that actually the profile of the regional banks on the second quarter and first half of this year, 2019, is quite close to the one we have seen at LCL, with customer savings up more than 4%, loans outstandings quite dynamic too at 6.6%, and revenues up at 1.5%. Cost of risk is quite significantly up on this quarter, but starting with such a low level that it's not really significant actually, and it continues to be very low all in all at 13 bps, and a contribution of €563 million to the net profit of the group on the quarter, and €1,228 million on the semester, up 2.8%. If we go now to the solvency at CASA level first, you can see on Page 27 that the CET1 ratio of CASA increases by 10 bps between the end of March and end of June. It's the result of a good level of results, good level of profit, the normal level of distribution with the dividend that we provisioned plus the 81 bps that we have to impact on our capital. The unrealized capital gains, unrealized reserves are up 11 bps. The business lines' RWA growth consumed 7 bps of solvency. It's focused on retail banking activities and consumer credit activities. And the other category consumes 18 bps of solvency, the biggest part of this category being the cost of the finalization of the acquisition of ProFamily by CACF, which represents a hit of 10 bps on our solvency. You can see on the right-hand side of the page that RWAs are up around €2 billion between end of March and end of June at €323 billion. And again, this is concentrated on LCL, CA Italia, and CACF. At group level, we continue also to build up the level of solvency. CET1 ratio stands now at 15.4%. It's, again, the combination of a high level retained earnings, a slight contribution of the OCI reserves, small hit linked to the business lines growth, and some technical elements, plus the impact of this acquisition, which is included in the last bar of 10 bps. At group level, the level of RWA increases by 0.8% between March and June, plus €4 billion. TLAC ratio is at 22.7% of risk-weighted assets and 7.6% of leverage exposure excluding, as we always do, any element of eligible senior preferred debt. And MREL ratio is approximately 34% of risk-weighted assets. In terms of funding, market funding, at group level we've been issuing €24 billion of debt at the end of June, and at CASA level close to €12 billion of debt at end of July, so it's 69% of the funding plan of - market funding plan of CASA. This is excluding the €1.1 billion AT1 issuance that was made in February, which was outside the funding plan. In terms of liquidity, I am finding nothing much to say on Page 30. And I think that I can conclude now on Page 32 by just reiterating the main messages of high profitability all in all, with a return on equity of 11%, of business lines posting an increase in their revenues and a positive jaw between the evolution of their revenues and the evolution of their cost base; a cost of risk which is more normal than one year ago but still low, and thus a dividend policy which is completely confirmed; and the first achievements of what we announced in the 2022 medium term plan with a significant number of external operations or partnerships that were either signed or closed on this quarter. Thank you very much for your attention for this presentation. And I think that we can now take your questions if you have some.

[Operator Instructions] And the first question comes from the line of Jon Peace from Credit Suisse.

My first question is just on the insurance business. And do you expect this sort of higher tax expense to persist, or should it be more normalized in the second half of the year? And my second question is on litigation. I believe you potentially expected a refund of the tax paid on the Emporiki disposal in the second half of the year. Is that still likely to come through, and in which case would that allow you to sort of accelerate the repurchase of the Switch, or should we model the Switch repurchase fairly evenly over the rest of the business plan? Jérôme Grivet: Thank you for your two important questions on insurance and tax level and globally, more generally on the tax level. It's really the situation that we had in Q2 '18 and H1 '18 that was a little bit abnormal globally, because - or I would say at the insurance business level, because the insurance managed to book some capital gains which were bearing a low level of corporate taxation. So this may happen going forward in the future, but this is not the normal situation. More normally, the revenues generated by - the financial revenues generated by the life insurance company are subject to the normal corporate tax rate. So the situation again in Q2 and H1 '18 were a little bit abnormal. And in addition to that, within the corporate center, we also had some exceptional or nonrecurring tax products which helped to reduce the overall level of corporate tax rate that was supported by CASA beginning of '18. So the normal situation is much more the situation where we are now, where the corporate tax rate is, let's say, on average between 27% and 29% before the decrease in the French corporate tax rate really starts, which we continue to expect and continue to believe, but we'll see. But this is more now the normal situation, I would say, to conclude. On Emporiki and the Switch, because this is going to lead to some additional elements on the Switch, so Emporiki, we don't know when the final decision of the complete data in France is going to take place. We expect this decision to take place some when, but we don't know when because actually it seems that the decision seems difficult to be taken. So for the time being, we are not forecasting a precise date for the settlement one way or another of this litigation. But as far as the Switch mechanism is concerned, let me just take the opportunity of your question to give you some more details. We have two windows each year to trigger a dismantling, a partial dismantling of the Switch mechanism. One is in the beginning of the year, so in January, and the second one is in July. So what we say is that, with the level of solvency that we have reached now, with the capital planning that we have in mind, this is putting us in the situation where we could consider to trigger the first, I would say, de-levering of the Switch beginning of 2020, so using the first window of 2020. This is what we said we deem possible as of now. What would we do if we have an unexpected one-off like the one you were referring to is another story. But what we say is that with, I would say, the normal course of business that we see now, with the level of solvency that we have reached now, we are in the position where we could consider starting to unwind the Switch mechanism using the first window of 2020.

And the next question comes from the line of Tarik EI Mejjad from Bank of America.

I have two questions, please. The first one is on the volume growth in France. I'm always impressed to see these numbers. I mean 11% growth in corporate and 9% in mortgages. So how do you manage and think about the risk of booking such low margins in the mortgage side and also managing the cost of risk on the corporate side, so if you can shed some light on that. And then second question on the RWA growth in CIB. It's comforting to see that you're going back to normal growth in RWA after the spike in Q1. But can you explain actually the comment you have in the slide for a lot of customers about the non-operating RWA optimization that you have done? What is that related to and if that actually has anything to do with Basel IV and sort of frontloading a bit the impact, or - sorry, or offsetting a bit the impact? Jérôme Grivet: Okay, volume growth of credit in France. You are asking two questions. One is about margin and the other one is about cost of risk. In terms of margin, it's true that we have been booking credit with a lower customer rate than expected once we started the year, at least. But at the same time, we continue to see a slight decrease in the overall cost of our liabilities. So in terms of margins, at least for the time being we don't see a significant, I would say, modification of the conditions on which we are working. So it's true that we continue to see a slight additional reduction, but clearly it's completely offset by the volume effect. And actually, this is helping us to generate an increase not only in the level of revenues, but also in the level of net interest income. In terms of cost of risk, future cost of risk, it's clear that we are not releasing or easing our credit standards. And actually, we continue to be very cautious. Be it on home loans or on corporate loans, we continue to use the same kind of credit standards, the same kind of credit granting criteria. And the consequence of that is that, as you have seen, we have absolutely no increase in the cost of risk. No increase in - the cost of risk as compared to the outstandings continues to be around 16 bps to 17 bps at LCL, as I was explaining. And we continue, actually, to have - you know that under IFRS 9, we have to assess the different buckets of contribution to the cost of risk, the overall cost of risk. And at LCL, bucket 1 and bucket two contributions to the cost of risk are almost nil. So it means that neither on the bucket 1, which are the new credits, or on the bucket 2, which are the credits which are supposed to be significantly deteriorated, on those two categories we see absolutely no sign of deterioration. So clearly, we are comfortable with the quality of the risks we underwrite now. In terms of RWA at the CIB, what we say is that we are regularly - and this is what we were mentioning. We are trying to optimize as much as possible within, of course, the framework of the regulation, being perfectly compliant with all regulation. We try to optimize the risk-weighting of the different risk that we have or the different assets that we have. And what we were mentioning by mentioning the non-operating risk-weighted assets is that, CACIB being a legal entity, it has its own balance sheet, and we can do some cleansing in the way the assets, non-operating assets again, are computed in terms of risk weighting. So nothing about frontloading Basel IV or front easting Basel IV. It's simply the normal cleansing that we regularly do on the way we calculate our risk-weighted assets.

Very clear, Jérôme. If I can just follow up on the first one, in terms of volume growth for mortgages, do you see some new wave of renegotiations, given the drop in commercial rates in the last few months or not? Jérôme Grivet: No, not for the time being. Not for the time being. I am not pretending it's not going to happen forever, but the volumes of renegotiation in the second quarter at LCL, where the outstandings in home loans are close to €80 billion at the end of June, were around €400 million on the second quarter. So it's very, very low compared to the highs of several billion that we had in a single quarter when the - at the peak of this renegotiation wave in 2017.

And the next question comes from the line of Jacques-Henri Gaulard, Kepler Cheuvreux. Jacques-Henri Gaulard: Just two quick questions. The first one, I was quite curious, in your press release you mentioned a €10 million one-off impact of consulting fees related to the group's structural operations. I was wondering what that was, simply, and the type of consultants you ended up using and why would those be one-off in particular. And the second question, to bounce on my respected colleague, Tarik, on risk-weighted assets, you have indeed stabilized the level at about 320 bps. Is it a fair assessment to say that now you're going to keep the risk-weighted assets there until the end of the year, or are you looking at capital differently, considering how much, I would say, earnings power you have that can actually play the numerator more and be a bit more relaxed about the denominator? Jérôme Grivet: In terms of fees, it's more advisory fees than consulting fees, because consulting fees is something we regulatory have because we have a lot of internal projects where we use consultants; too much, from my viewpoint, but nevertheless it's kind of a must. And so we wouldn't even mention that type of fees, which is, again, part of the normal course of business. I was much more referring on advisory fees in connection with the negotiation of the different partnerships I mentioned. And namely, to put in a nutshell, it's around €5 million fees at CACF and €5 million fees at CACEIS, which were both engaged in two quite significant and quite time-consuming negotiations, at CACEIS for KAS Bank and the discussions with Santander, and CACF discussion with FCA and discussions with Banco BPM for Agos and [indiscernible]. In terms of RWA, well, the only thing I can do is to ask you to refer to what we said on June 6 a few weeks ago, because the way we see things is that, within our capital earnings and capital generating capacity, we have to cover, of course, all the regulatory constraints. And we estimated those at 30 bps between now and 2022 for trim and different things like that, and 60 bps in 2022 for Basel IV. We have, of course, to cover our commitment to unwind half of the Switch mechanism, so it's around 60 bps before 2022. I should have started with the very point, which is that we have to cover the dividend policy, to which we are going to stick. And then the remainder is here to, if it makes sense, if it's value creating, to fuel the growth capacity of the business lines. But it's really - the way we see things, it's that we have taken some commitments and we are going to stick to those commitments. And then inside, of course, we are going to use the remainder - the rest of our capacity to, in the most efficient way, in order to try to fuel our growth and to boost our earnings.

And the next question comes from the line of Giulia Miotto from Morgan Stanley.

A couple of questions for me. So the first one refers to the macro environment, or rather the deterioration since you had your investor day. So if I remember correctly, the assumptions on rates were for 85 bps for three months LIBOR by 2022 and 170 bps for a 10 year swap. Of course, in the meantime we have moved even further into negative territory, and the expectations are potentially for this to be cut further. If you had to mark-to-market your revenue expectations then for the current environment, would this change? And if you can quantify that, it will be very helpful. And then the second question is more specific on the quarter on fees. So if I look at fee income, both in France and Italy, it's either flat or up only 1%. And so I was wondering what's driving that. Because I was hoping strong loan growth could lead to some stronger cross sell, but it doesn't seem to be coming through at the moment. Jérôme Grivet: Thank you. Let me start with your second question, if you allow me, because what I can say, you are pointing to two specific situations, in France, LCL, and Italy. But if I'm going on a more global basis, the fee part of our revenues have increased, actually, between Q2 '18 and Q2 '19, so slightly, only slightly. But it was 35% in - I think it's more H1. It was 35%, the total revenues. It's now 34%. This is for the net interest margin. And the fee part has remained stable at 40% of the total revenues. Then you have the treaty and other revenues at 14%, also stable, and the insurance revenues, which were 11% last year H1, which are now at 12%. So what you can see is that what had been reduced is the net interest margin part of the revenues, but the fee part has remained stable between H1 '18 and H1 '19. So if it comes to the two specific situations you are mentioning, so LCL and Credit Agricole Italia, at LCL it's very clear that French banks have collective taken some commitments end of last year in order to freeze their overall price grid for all the services that they sell for 2019, and to reduce the fees that they take on the fragile part of their customer base. So this is obviously putting some pressure on the overall level of fees for 2019. It's not massive. It's not significant. But indeed it plays a role. And the second reason explaining the stability of the level of fees at LCL is the fact that we have seen a reduction in the fees linked to off balance sheet savings, especially all investment in the equity-linked products in mutual funds and so on and so forth, simply because customers have been quite in a wait and see attitude in the beginning of this year. But the fact that we continue to see the equipment rate of our customer bases in insurance products or in payment services is going positively to play a positive role for the evolution of fees within the networks. But I think that, as far as fees are concerned, as far as the breakdown of the revenues are concerned, it's also important to have the assessment at a global level. Then if I go now to the macro environment and the rate situation, it's clear that between June 6 and now, we have had the Sintra speech of Mr. Draghi and a rather sharp modification of the monetary policy prospects across not only Europe but also the U.S., as we have seen with the Fed decision yesterday or the day before. It's very difficult to, as you say, mark-to-market our financial trajectory through a modification of the rates scenario, first because I don't know yet what will be the relevant rates scenario going forward. What I see is that rates have decreased between the spring this year and now. That's for sure. And what I see is that the narrative of the policymakers is much more dovish than what we had in mind. But I still don't know exactly what is going to be the scenario. The second unknown is that this type of rate environment in which we are is triggering some significant modification in the behaviors of the different players, behavior of our customers. And clearly the decrease in the rate curve, in the rate structure, is triggering a growing appetite for loans. So there is a link between the level of rates and the volumes. And the level of rates has also a link with the level of cost of risk. So you have to assess all the pieces of the puzzle and not only the top line and the effect on the top line. And even on the top line, even on the net interest margin, it's very difficult to try to identify what is going to be the - what are going to be the changes in behavior in the different segments of business in which we are engaged. It's true that mechanically for retail banking activities, on the one hand, and for life insurance activities in euro on the other hand, this rate environment is going to be a more challenging environment than the one we could have in mind only two or three months ago. But for some other businesses, it's not the same situation. And even for those two businesses, it's clear that we are not going just to stay like this, waiting for the risk to decrease and to continue to keep our P&L. And for example, in the life insurance activities, we are progressively modifying the way we structure our products in order to adapt them to this environment and also in order to try to propose alternative products to our customers in order to ease the constraint. So it's very difficult to mechanically adjust or input a new rate scenario in our financial trajectory and to say, yes, it's going to be modified like this and this. What is for sure is that we continue to commit to what we have said, i.e., the level of profitability, i.e., the capacity to adapt the structure of each of our business in an environment that may be different than the one we had in mind, and this is exactly what we are going to do.

And the next question comes from the line of Flora Benhakoun from Deutsche Bank.

I have two questions, please. The first one is regarding the consumer finance business, where I see there is clearly some pressure on revenues compared to the evolution of the consolidated loan book. I know the performance of the equity affiliates. But just focusing here on the consolidated loan book and the revenues, there has been now three quarters in a row of clear pressure in the revenues per loans. So just wanted to understand why, because supposedly this is a business that is not going to see margin pressure like a pure retail business would see. The second question is regarding the PPE reserve, which I see you have increased this quarter. Just if you could elaborate on this, and also give us your view whether you think you have now contributed enough to this reserve considering the recent fall that we have seen in bond yields, or whether you think that most likely you will need to provide again in the coming quarters. Jérôme Grivet: On consumer finance, I think that, as far as the consolidated loan book is concerned, you have two or three different elements explaining this pressure. The first one is that we continue to see a modification, a shift in the breakdown of the consolidated assets. It's the continuation of the effect of the decisions that were taken a few years ago now by the French government in order to penalize much more and to prevent customers engaging in revolving loans and to promote much more amortizing loans. So we continue to have a modification of the thoughtful view of loans between revolving loans and amortizing loans. So of course, the amortizing loans yield less for the consumer credit entity than the revolving loan. The second element is that, in the last part of 2018, actually the refinancing costs of this business, especially in Italy, where you know we have a significant part of our consolidated business, have increased because market fundings in Italy have increased. And what we see in Q1 and Q2 this year is that the margin at inception for new loans consolidated in the loan book of CACF is slightly improving, precisely because market health has started to decline. So of course it's only for the new loans, and it's going to take a little time before it translates into the net interest income and into the ratio between net interest income and the outstandings. But clearly, the situation is improving and the game, of course, of CACF, the goal of CACF is to keep as much as possible of this improvement of the refinancing costs in order to improve its profitability rather than competing with the other players in the pricing of new loans. And maybe the last point is that, as the competition is fierce, we have had also some quite significant, I would say, commercial costs to support, and they are embedded in the net interest income. They are deducted from the net interest income. So all these reasons explain why, in the last three quarters, we have had quite a stable level of revenues as compared to the evolution of the outstandings. But I see, we see an improvement as far as the new loans are concerned. For the insurance activities and the PPE, you know that the PPE is not necessarily especially in an income in this quarter, the result of an explicit decision. Actually, you know that in the life insurance activities as far as the euro-dominated products are concerned, what we do is that we assess each quarter the level of financial revenues that are generated by the bought-for view of assets. Then we make the first breakdown between what is going to be the part of the revenues kept by the insurer, the insurance company, to cover its costs and to cover its cost of capital and to generate its profit, and the other part which is attributed to the policyholder. And then within the part that is going to be attributed to the policyholders, we have a second breakdown to make, which is between what we call the PAB, participation benefit, which is the profit sharing that is going to be attributed on this precise year, and then the PPE, which is more what is left, actually. And so the PPE is more, I would say, the result of the rest of the decisions rather than really a target that we have in mind. Then of course, on the last quarter of the year, this is when we finally set and definitely set the profit sharing rate for the customer that is going to be attributed for the account of the year. And then this is the time when we set definitely the PPE. So we'll see at the end of the year what the PPE will be. But clearly, for the time being, it wouldn't make sense to allocate a more important part than what we did to the PAB, i.e., to the part of the revenues that is going to be attributed to the policyholders this year. So it's much more this way that we see things.

And the next question comes from the line of Guillaume Tiberghien, Exane.

I've got two questions. The first one is on the LCL revenues. If I remember, in Q3 '16 you realigned the way you were funding the business. And that triggered a minus €300 million impact, which was meant to have a €50 million positive impact this year in the P&L. So number one, I wanted to check that this €50 million is still there this year. And number two, should we expect, given the rate environment, maybe that you would have to set up another realignment of your funding? And the second question relates to the RWA again. Your target at the capital markets day for 2022 was €360 million. If I deduct €17 million for Switch and €24 million for regulation, you should be at €319 million in '22, and you are at €323 million today. So actually, you talk about fueling growth, but I am actually wondering where will the reduction come from to achieve the 2022 target. Jérôme Grivet: Let me start with LCL revenue. So what we did actually in 2016 was, as you say, to exactly do what our customers did, i.e., repay part of the costly debt of LCL to reduce the overall cost of funding of the balance sheet. We have absolutely no reason to modify this going forward, because actually the structure of the funding is being made massively by customer deposits and hedges that we've put in place since that time, being priced at the low level of rates that we have seen since this time, doesn't trigger any necessity to repeat the same type of operation. And definitely what we had mentioned as the positive effects of this operation are here in our balance sheet. And what we see, this contributes positively to the evolution of the revenues, in that the cost of the hedges continues to reduce a little bit quarter after quarter. This is exactly through that that the benefits of this operation is going to be taken. So clearly it's within - of course, many, many pieces have moved since we've put in place this operation. So it's very difficult to track this specific benefit within the P&L. But clearly, this has significantly helped our net interest margin within LCL, and this is partly why we have this good level of activity and good level of revenues at LCL. In terms of evolution of RWAs, I haven't noted all the figures you were mentioning. But what we say, to put it in a nutshell, is that there will be growth, growth of our capital base, growth of our capacity to cover RWAs, and the growth of our RWA is going to be offset by mitigation actions that we intend to develop once we know exactly how the Basel IV transposition is going to be done. So this is exactly the way we see things. We have calculated in the best of our knowledge what would be the impact of Basel IV in terms of CET1 ratio. And once the rules are going to be know, we'll try to mitigate as much as possible by optimizing the structuring of the different categories of credits that we make and to adapt to the rules of Basel IV. But this is exactly what we said. And again, I think the motto of our capital management policy is to stick to our dividend policy, to cover all the regulatory costs, to unwind half of the Switch, and then the rest is more or less free to help us develop our business.

And the next question comes from Azzurra Guelfi from Citi.

Just two questions on Italy. One is on the margin. When I look at the revenue, it seems that you are suggesting that there is margin expansion in Italy, and I was curious to know what is it driven. Is it funding cost? Is it specific repricing or some changing in the mix? And the other one is on your sovereign exposure. Can you tell us what is your sovereign exposure to Italy and what is your policy around it? Jérôme Grivet: So two questions on Italy, if I may say.

Yes, I'm just paternalistic. Jérôme Grivet: The first one, on the BTP, I think we have - I don't have the precise figures in mind, but we haven't changed anything. So roughly we have around €6 billion of exposures within the life insurance company that we have in Italy, CA Vita, and we have around €8 billion within the books of Cariparma Italia. So the exposure that we have, it hasn't evolved in the last full year I think. I think at least since one year. In terms of margin, actually Credit Agricole Italia is working very hard on the improvement of its margins. And they have managed to improve actually quite significantly the rates at the inception of their credit, be it for home loans and for corporate loans. And so it's a matter of structuring the loans. There is a progressive shift between floating rates and fixed rates, especially in the field of home loans. It's also a matter of picking up the right customers and picking up the right corporate customers. But they are definitely working on the improvement of their margin at inception of the loan. And what I can tell you, without providing precise figures, is that it has improved quite significantly between Q1 '18 and Q2 - H1 '18, excuse me, and H1 '19.

And the next question comes from the line of Jean-Francois Neuez from Goldman Sachs. Jean-Francois Neuez: This is Jean-Francois Neuez from Goldman Sachs. I just wanted to ask a question on Switch. So Switch pays a low interest rate, as you've been able to discuss in the past on many occasions when it was set up. I just wondered, to spend a notional €100 of capital - well, any quantity of capital on the switch versus buying back your own shares, how do you choose? Because in reverse, in a sense, when you're buying back the Switch, you're buying an asset on 16, 17 times P/E, whereas your shares are maybe 8 times now. And I'd obviously observe that share buybacks are not very common in Europe and a bit innovative, but I'd also observe maybe that the Switch was in any case quite innovative also at the time. So I just wondered what's your policy and how you choose to buy back the Switch versus other capital usage. And just as a follow up on Italy, very interested in this margin development, because Cariparma or Credit Agricole Italy looks to be in a very special spot in terms of being able to improve the margin. And I just wanted to know whether you'd be able to shed some more color as to whether you think competitors are getting there as well, or whether this is just isolated and specific to your client base or the way you manage the loan book or the deposit costs over there. Jérôme Grivet: Well, let's again start with the second question. We are a very small player in the field of retail banking activities in Italy, and we are not located across all the country. So with our small market share, we have the capacity to pick and choose our customers. And in addition to that, we are especially picking and choosing the customers within the customer base of the new banks that we acquired 1.5 years ago. So we have this capacity of either picking the right customers for the loans that we grant and also, I already mentioned it, getting rid of some costly deposits in order, again, to improve the cost of funding of the balance sheet. Probably some bigger players with much wider market shares couldn't do exactly the same. So I think we are taking advantage of what we are in Italy. In addition to that, I could add also that we have the good rating of Credit Agricole Group, and we benefit also from this good rating and the strength of the group in developing our activities in Italy. So for all those reasons, I think we are, as you mentioned, on a rather sweet spot in Italy, but it's not necessarily replicable to a bigger player. And nevertheless, it's not my duty to give advice to our competitors. On the second point, which is the Switch, we've been discussing that in depth several times. I think that it's not only a mathematical debate. It's clear that Credit Agricole has suffered a long time ago - it's probably now no longer the case. But it has suffered from having the image of being complex, complicated, having too many specificities in its culture. The Switch is one of the very few remainders of this image. And so I think that unwinding the Switch mechanism has two advantages, if I want to enter into your reasoning. The first one is to reduce further what is - or what may be seen as the complexity of the group, so reducing potentially the cost of equity that the market attributes to us. And this also a benefit for us. And the second point is that buying back our shares is reducing the capital at group level when unwinding the Switch is transferring back some capital from CASA to the regional banks, thus maintaining the same level of capital at group level. And you know that CASA benefits from the high level of capital of the group globally, notably for the cost of its funding. So I think we could also find a benefit in improving the earnings per share at CASA level in unwinding the Switch rather than in buying back our shares. And the last point maybe will be that buying back our shares, as a controlled listed company, would reduce the free float, which is not good news for the minority shareholders.

And the next question comes from the line of Stefan Stalmann, Autonomous Research.

I have two questions left, please. The first one kind of relates to the previous discussion of potential share buybacks. You have another employee share increase coming. And it's not particularly big, but it's quite a recurring feature over the years. Have you actually thought about handing out these employee shares out of treasury shares that you buy back in the market? And the second point relates to the consumer finance joint venture with BPM. If I understand this deal correctly, you have maybe paid around €300 million or so for a distribution agreement, which shows up as an intangible asset now. The first question on this is it fair to assume that that generates a return on investment, or are you putting this into place just to preserve existing profits in consumer finance from this joint venture? And the second question related to this is are you going to amortize this intangible asset over 15 years? Jérôme Grivet: Well, actually, what we are doing is that we are buying an entity, which is ProFamily. We are not paying this amount as a one-off fee, I would say, to buy the distribution agreement, or at least we are buying an entity which is the owner of the distribution agreement. And so this is an entity that we are buying, and of course the price at which we buy the entity embeds a high level of goodwill, actually. But this is not exactly the same as buying simply the distribution agreement. And I think I may want to check to be completely sure that there is no part of this amount that is going to be amortized, but I think that it's purely a goodwill and not a distribution agreement. So, we'll check and we'll go back to you with the precise answer. And the price is indeed not €300 million, but €310 million, to be precise. It's what we announced when we communicated on the signature of the deal. On your other point, it's true that it's the second year in a row and the third year since 2015 that we are having this capital increase for the employees. It's going to be completed in the course of the third quarter, and it's going to represent around 5 bps of CET1 ratio, so €150 million at CASA level. For the time being, the amounts are too tiny to imagine the type of offsetting that you are mentioning. We'll see going forward if the appetite of the employees increases. And we hope that it's going to increase, because if we do it now on a regular basis, it's precisely with the idea of having our employees quite used to the mechanism and ready to invest, if they wish, regularly. But if this takes a more important size, we may consider some mechanisms to offset the impact in terms of dilution. But for the time being, the dilution is very, very, very tiny.

And the next question comes from the line of Jean-Pierre Lambert from KBW. Jean-Pierre Lambert: I have three questions, please. The first one is just a simple one. For CIB, you indicate a reversal of provisions on personnel cost. Can you disclose the amount, or was it disclosed in the exceptionals? The second question is the fees and commissions. You indicated that effectively the 40% of revenues has been stable between the first half last year and this year. But if we look at the reported figures, we see actually a fall in fees and commissions. And I know the intricacies of the group, but we see that client fees have come down 7% and fees on mutual funds down by 20%. Can you give an indication of the trends of the total pool, the subcategories? And then also on the fees and commissions, we see that there's a net outward payment for payment services, which is a bit surprising because you are a provider of payment services. So that's on fees and commissions. And the third question is about the insurance business, your strategy of partnership. Where can we see the contribution of the partnership in the P&L? Unlike consumer finance, it is probably in the revenue line. But how much are these partnerships contributing overall to the revenues? Jérôme Grivet: On the first one, the answer is very simple. We didn't disclose neither any name nor any amount, and we will not. So I think it's quite clear. It's a covenant policy that we have on this matter. Let me go now to your last question before coming back to the second one. On the insurance partnerships, actually it's only starting, because the first partnership is the one with CreVal, which started the beginning of 2019, so beginning of this year. The net inflows at CreVal since the beginning of this year was around I think €250 million in the first six months of the year. It's going to generate revenues within our Italian insurance activities, so it's going to be seen progressively as a contribution, a direct contribution to the P&L of Credit Agricole Assurance. But it's going to take time before it moves the main and the big figures of Credit Agricole Assurance. So it's maybe an operation on which we will give you some, I would say, more commercial figures before you see really the lines moving on the P&L of Credit Agricole Assurance globally, considering the size of Credit Agricole Assurance. But we see it as a real growth engine for the future. And it's going to be the same for the partnership with Abanca, which hasn't started yet because it's only been concluded. So we now have to establish the company and to start the operation, which is not going to be the case before 2020. Regarding your question on fees and commission, I think it's a quite technical question, if you allow me to answer like that. And I think we'll have to dig a little bit into your question before we are able to give you a precise answer. But keep in mind, for example, you are mentioning payment services, that actually we are providing payment services to our internal banks and entities. So one thing is to be a very big player in the field of payment services. Another thing is to generate revenues which are directly taken outside the group. The way we manage our payment services entity is that it provides services to the regional banks, to LCL, to CACF, and to other entities of the group without generating a profit in itself. But we are going to go back to your more technical questions on the accounting of fees and commission maybe after this call, if you allow me.

And the next question comes from the line of Lorraine Quoirez from UBS.

Just two question for me. The first one is could you please disclose the amount of excess liquidity you have at the ECB presently? And the second question --. Jérôme Grivet: Excuse me. We don't hear you.

Can you hear me now? Jérôme Grivet: Yes, it's better. It's better.

Yes. Could you please disclose the amount of excess liquidity you have at GECB? And the second question is I would like to know whether there is actually some discussion at the regulatory level regarding the capital allocation at group level. Jérôme Grivet: On the capital allocation, what is behind your question? Excuse me?

Is there a discussion at the regulatory level on the capital allocation at the group level? Jérôme Grivet: You mean capital allocation between the group, the SA - and the subsidiaries? Well, let's start with the second question. As every year, we have a high level discussion with the ECB in the threat process. And in this process, the ECB is expressing its requests in terms of capital requirement for different entities of the group, for the group itself, obviously, but also for different entities of the group. So the process for this year is not completed yet, but it has already taken place twice in 2018 and 2017. And every time it has taken place, we have published the results, at least the part of the results of this dialogue that was able to be published, which is the Pillar 2 requirement. And you may have seen that between 2017 and 2018 there hasn't been any modification in the Pillar 2 requirement that was assessed to the group and to Credit Agricole S.A. by the ECB, so nothing much to mention. And obviously, probably by the end of this year, we are going to get the new SREP requirements, and we will publish again these requirements, as we always do. Excess liquidity, well, this moves every day. But I think the latest figures that we have indicate it is €75 billion of central bank deposits, be it at the ECB or with other central banks, because as you know we have a complete network of treasuries across the world. And we may deposit also amounts at the Fed, at the Bank of Japan, Bank of England, or whatever.

So it's not possible to have the part that's only at the ECB level, right? Jérôme Grivet: No.

And the next question comes from the line of… Jérôme Grivet: I think we still have three questions. And if you don't mind, I think we'll try to stick to those three questions.

And the next question comes from the line of Kiri Vijayarajah from HSBC.

Yes, a couple of questions from my side. Firstly on asset quality, I see you're letting your coverage ratio slip slightly at group level and I guess a little bit more meaningfully in Italy. And I'm really just wondering from the perspective of kind of the new NPE provisioning requirements, the EBA backstop kicking in at some stage next year, so for the kind of 2018 vintages of the unsecured NPEs. So do you think there's a risk, or are you coverage levels on the unsecured side kind of comfortable from that perspective when the EBA backstop kicks in? And then secondly, just on the - more a hypothetical question on the potential kind of tiering of ECB deposits rates. I'm just wondering would you have to share the benefit of that with the regional banks? Because I guess they're ultimately the source of the excess deposits and the excess liquidity, or could CASA keep hold of all of the NII windfall, if you can really call it that? So just curious how that would work as well. Jérôme Grivet: Well, let's start with your last question. It may be a little bit disappointing to you to learn that actually, as all other French retail banks, the network of the regional banks has a deficit of liquidity between all the customer deposits and the loans that were granted. And it's the case for all French banks, actually all retail banks in France, have a loan to deposit ratio which is between 110% and 120%. So actually, the excess of cash that we deposit to the ECB much more comes from the treasury activities of CACIB, comes from CACF, comes from the wealth management business, but not from the different retail banking activities. And probably if a tiering takes place, we would have to share the benefits of the tiering with the customers of CACF, for example, that presently we bill from the cost of the deposits that we put at the ECB because they deposit some cash with us. So I don't expect a significant improvement in our P&L coming from the implementation of a tiering, depending of course on the type of tiering that can be put in place ultimately. When it comes to asset quality, I think that one of the most important elements to really assess the asset quality is the NPL ratio. The coverage depends mainly on the category of each loan that is nonperforming. But what is really the indicator of the evolution of the quality of the loan portfolio is the NPL ratio. And the NPL ratio at group levels stands at 3.3% - at CASA level, excuse me, is at 3.3% and it's stable. And at group level, it's 2.6% and it's stable. The slight evolution that you may have seen in the coverage ratio depends actually fully on some operations that we can undertake when we sell NPLs. And if we sell NPLs with a high coverage, then it may marginally reduce the coverage ratio. And on the contrary, if we sell NPLs with a low coverage ratio and a low level of provision, it may improve a little bit the coverage ratio. And generally, typically we sell the NPLs with the highest level of provisioning because these are the NPLs on which we think it's impossible to have a positive evolution of the relationship with the customer. So this is why we can see on a quarterly basis some slight evolution of the coverage ratio. Of course, we are working seriously on the consequences of the backstop of the EBA in terms of NPE provisioning. It's far too early to really assess the consequences, but we are working on that. And be sure that we are not going to be caught by this backstop without having prepared our adjustment, I would say.

And the next question comes from the line of Omar Fall from Barclays.

Just two question for me, please. So actually just on LCL, or even at group level if you like, if you're not able to give us a sensitivity to rates, what would be just as helpful is if you could tell us how much of your deposits you reinvest via replication portfolios or government bonds, and roughly at what average duration, please? Then secondly, just looking at the contribution from the equity accounted entities, and specifically FCA Bank and GAC-SOFINCO in China, I know that the market is very healthy, but the pace of growth of them in the first half is really quite spectacular, 24% at FCA. It's almost doubling in China. So what exactly is happening there? Because it's 10% of the earnings that we don't talk about much, so it'd be great to know if that's sustainable or if there's something else going on there. Jérôme Grivet: Well, at LCL, as is the case in each retail bank, actually we don't reinvest the customer deposits in bonds or assets of that type. Actually, we use - after the application of our ALM conventions, we use the funding provided by the customer deposits to fund the loan book, actually. And so each time we grant a home loan, it's partially financed by term deposits. It's partially financed by sight deposits. It's partially financed by the proceeds of regulated savings accounts. And it's partially funded by some market funding. So it's a combination with a weighted average cost. But we don't invest these customer deposits in bonds, really. So it's not exactly the way you can analyst, I would say, the situation of banks in some other countries maybe.

I think, well, if we ignore the bond side, certainly you have swaps against your stock of deposits that are not - particularly on the current accounts that are not at the overnight duration. They're at their effective duration. Jérôme Grivet: This is what I was mentioning. What we do is that we take the volume of customer deposits, let's say the sight deposits, and we consider that they progressively amortize on a certain duration, which is assessed with ALM studies and regularly checked on an historical basis. And then we use the part of the sight deposits that we deem that is going to stay on our balance sheet for the next 10 years. We use it to fund the 10 year part - 10 year maturity part of the loan book. And so every month we progressively amortize the initial amount of sight deposits. And so there is notionally a production of new sight deposits that we, again, amortize on the same duration. And what we did in the last three or four years - considering precisely that the outstandings of sight deposits were increasing very, very rapidly, we identified the more volatile part of those sight deposits that we amortize on a much shorter duration in order not to consider that we have "invested" these sight deposits into long assets. Well, it's exactly what we do, I would say, on a monthly or even a weekly or on a daily basis. This is exactly the ALM management that we have to do in a retail bank, because we have non-dated deposits and we have dated assets.

And of the sight deposits that are deemed volatile and therefore invested at a shorter duration, how much is that? Do you know? Jérôme Grivet: Well, I don't have the precise amount, but actually what we did is that we have taken, on a historical base, what we deem the stable amount. And we think - we assess that the stable amount of the sight deposits by the customer evolves in proportion with the GDP, I would say. And what is the amount of sight deposits exceeding the GDP growth in the last two or three years, we consider it's more volatile and we consider we may not use it on a long duration.

Interesting. Thank you. Jérôme Grivet: As far as car JVs are concerned, I think that's - well, those two JVs are working well. We could maybe someday organize a dedicated event on this business, because it's a very important and interesting business. It may seem a little bit strange to see a good evolution of this business when we are talking about a difficult car market both in China and in Europe. But I think that there are at least two or three reasons why a good car financing business can develop nicely, even if the car market is a little bit more difficult. The first reason is that when we're talking about the car market, we're only talking about the new car market, and we are also interested in the secondhand car business. The second element is that, especially when the time is a little bit more difficult to sell new cars, carmakers are keen on trying to promote car financing offers in order to boost their car sales. And so this is why we have a lever, in order to propose our services and to develop the activities. So this is exactly how we manage to develop nicely the business even in a difficult car market.

And the next question comes from the line of Nick Davey from Redburn.

Two question, please. The first one, I think when we spoke at the full year results in February, you outlined an aspiration to start repaying TLTRO perhaps even a bit early. And I just wondering whether, given what's going on with interest rates, you stuck by that opinion or whether you might hold TLTRO balances stable or even increase them as some way to try and mitigate the margin pressure going on everywhere. And the second question would be around the new disclosure in the corporate center on slide 23. And the question would really be around this sort of €1 billion, €1.5 billion of negative revenues from the balance sheet and the holding at CASA level. And I remember we talked about this in the previous investor day. But I just wondered if you could give some context in the very long run about, I don't know, any division of that revenue between the acquisition debt cost and then just the everyday running of the liquidity for the group, any kind of aspiration of where you think those negative revenues could mitigate - or could be mitigated in the long run. Jérôme Grivet: Okay. Well, clearly we are talking mainly about the acquisition debt, because as you know CASA has been built over time through many, many acquisitions. And actually, there is a significant amount of debt that still lies within the balance sheet of CASA. So part of it has been repaid when we did the Eureka transaction back in the beginning of '16. But still, we are having a level of debt which is - I don't have the precise figure in mind, but it's around €20 billion. That is the remaining part of the debt that we have. So every time we manage to upstream dividends from our subsidiary, and actually the policy of upstreaming dividend is a 95% dividend policy from the different entities, we manage to reduce progressively the amount of this debt. But regularly, we participated in capital increase of different subsidiaries, like the one of Amundi when it acquired Pioneer Investment, or some other operations in order to help certain subsidiaries to develop their business. So we are not reducing this overall amount of debt as quickly as we would like to, but nevertheless we do. And then the second element is that every time we can, we try to reduce the cost of this debt in taking advantage of different - of the market conditions and the improvement of our rating. But we will maybe later on provide some guidance on the possible evolution of this financial cost that is supported by the corporate center. Going to your first question on TLTRO, actually we haven't been paying back significant amounts of TLTRO II. And clearly, since we started reflecting on that, the circumstances have a little bit changed. So we are actually reflecting on TLTRO II/TLTRO III, which is the new mechanism that has been announced by the ECB. And we are regularly reassessing the situation. TLTRO II is not very costly, minus 40 bps, at least if you can show that your loan portfolio has increased, which is obviously our case. But it has to be repaid before the end of Q1 '21. So it's now going progressively two weeks the moment when it has only a remaining maturity which is below one year, which is less attractive in terms of ratios. So this is a paramount fine-tuning, given at the same time the cost of funding on the market and the needs that we have in terms of funding. So it's embedded, I would say, in the funding plan publish every year and on which we give some elements of development every quarter. And nothing much more to say on this, the TLTRO.

Can I just follow up quickly on the acquisition debt? So you said it's not falling as quickly as you'd like. What are the impediments to it falling, or at what dates can you accelerate the speed? Jérôme Grivet: Well, the only way that we have to accelerate the speed of decrease is to upstream dividend from our own subsidiaries. And then the difference between the dividend that we pay and the dividend that we receive is the part that we may use to reduce our debt. But regularly, for many different types of reasons, as we are organized with different subsidiaries, we have - or we may be faced with the necessity of funding our own subsidiaries, be it, again, to fund their development or to fund their acquisitions. That was the case for Amundi. So it's - every year, we try to have a positive balance between the dividends that we receive and the dividends that we pay at CASA level, okay?

Thank you. And this concludes our Q&A session for today. Please continue. Jérôme Grivet: Thank you very much to all of you, and I wish you again a nice summer and good holiday. Bye-bye.