Operator: Good afternoon, ladies and gentlemen, and thank you for standing by. Welcome to today's Crédit Agricole 2020 Second Quarter and First Half Results Conference. [Operator Instructions]. I would like now to hand the conference over to your speaker today, Mr. Jerome Grivet.

Please go ahead.

Jerome Grivet: Good afternoon, everyone. To start with, I just wanted to thank you for attending this conference and to say that I've received a few messages asking me to do a short and quick presentation, so I'll try to walk you very rapidly through the document, and then we'll go to the Q&A session. I'm going to start with Page 4, on which you have the main figures regarding our performance this quarter and this semester, just let me highlight a few figures. The net income group share for the group globally which stands at €1.5 billion and, on an underlying basis, at €1.8 billion; and for CASA, €950 million as a stated figure and €1.1 billion as an underlying figure.

This quarter, the elements explaining the difference between stated and underlying figures are besides the traditional DVA loan hedges and home purchase saving plan elements. Two main things regarding the group globally and

CASA specifically: the first one is the cost of liability management operation that we did in June, through which we bought back around €3 billion of debt that we are going -- or that we replaced by cheaper debt, the cost of this operation which is booked in the quarter is about €40 million of costs booked in the revenues -- negative revenues; and the second element, which is even more significant is the cost of the indemnification that we voluntarily paid to the self-employed professionals that are our customers for their insurance policies. You know that their insurance policies cover business interruption but not in the case of a pandemic. And we have decided, nevertheless, to indemnify them. This represented a cost of around €140 million at CASA level and even €240 million at group level.

It's also a negative revenue. And then the last point, which regards only CASA and which is neutral at group level, is the triggering this semester of the Switch mechanism for a very modest amount, €65 million, which is a profit for CASA at the level of the cost of risk. It's symmetrically a cost for the regional banks, and it's neutral at group level. The second and last element I wanted to mention on this page is the high level of solvency that we have reached this quarter, both at group level and at CASA level. At group, the CET1 stands at 16.1% and, at CASA level, at 12%.

If I go now to the following page, I think that the only elements we can mention here, because the rest is going to be covered by the following pages, is that we have had a very specific quarter in terms of profile with a level of activity that declined significantly in April and also in May and a very sharp rebound in June, which continues to be the case across July. Let me go now on the next page, just a few elements to remind you that we've been very active in helping our customers across the board to come over this very specific period of the lockdown. And actually, in France, we've dealt with a €28.7 billion of state-guaranteed loan requests. And we've also provided to our customers more than €4 billion equivalent of deferred maturity of payment holidays for the different credits. In addition to that, we also mentioned this indemnification that was granted to the customers of our insurance activities, voluntarily again.

And then the last point is that in Italy also, we've been active in granting state-guaranteed loans and credit moratoria. Next page, just a few elements on this. We've continued actually to work on the 3 pillars of the group medium-term plan through the second quarter despite the lockdown period. And I think that this proved to be very relevant considering the type of crisis in which we were and considering the achievement that we've managed to make in those 3 pillars. On the following page, just a few elements to explain how we see the economic situation in France and more generally in Europe.

We said that we see -- we said that we are forecasting a V-shaped crisis. And definitely, what we see now is converging with this scenario. There is a very sharp rebound in household confidence and business climate. And we also find that all indicators, hot indicators, that we try to follow, especially indicators that we can follow with our own activities, confirm this very sharp rebound in the level of activity in France. On the next page, and this is going exactly in the same direction, we have seen in our own activities, in our own commercial activities, a very sharp rebound of the volumes of operations in the second part of the second quarter of this year.

As you can see, this has been the case for the number of new customers that we've managed to attract in our different retail networks with a low level of 46,000 new customers in April but a very high level of around 150,000 new customers in June. And this is leading to a sharp rebound in the different areas of activity, be it home loans, consumer finance agreements; be it also savings accounts on -- an off-balance sheet savings accounts; and be it also the sale of new P&C insurance policies in the different networks. Let me go now to the financial figures. And on Page 11, you have the evolution of the revenues of Crédit Agricole S.A. On an underlying basis, between Q2 '19 and Q2 '20, the level of revenues was more or less stable, which is a very good performance, again, considering the fact that in the biggest part of our activities, we are dealing with individual customers.

And for the biggest part of the quarter, actually, the retail activities were more or less stopped, and all our networks were dedicated to deal with state-guaranteed loans and other specific elements linked to the management of the COVID crisis. So this is leading to a level of revenues which are up 2.5% on the first half of the year, which is a very good evolution of our revenues. On Page 12, the evolution of the cost base. It's very well kept under control. On the Q2 only, the cost base is down close to 2%.

And on the full first half of the year, it's up only 4.5%. So very well controlled cost base, which is proving again that our strategy to manage the cost base of the different businesses close to the operational level is the best way to make sure that all efforts can be really efficient to monitor the cost base. This is leading to a cost-to-income ratio on Q2 '20 for Credit Agricole S.A., which stands a 57.4%, down 1.2 percentage points. It's a very competitive and robust level, and it's already below the level of 60% that we were targeting for 2022. Again, the dual effect is positive both on the quarter and on the first half of the year.

Just on this page, the last element that you may want to keep in mind, we have had to book an additional contribution this quarter to the Single Resolution Fund, which is a significant amount, it's €80 million at CASA level, around €80 million, and even €110 million at group level. On the following page, a few elements on the risk situation, starting with a presentation of the quality of our loan books. The level of NPL didn't significantly evolve across the quarter. It stands at 3.2%. The NPL ratio stands at 3.2% for Credit Agricole S.A.

and 2.4% for the group globally. It's more or less stable as compared to end of March. Considering the significant level of additional provision that we have booked this quarter, and I will come back on this question just on the following page, we are reaching very high levels of coverage ratio, 73.4% at CASA and even close to 85% at group level globally. We now have in excess of €10 billion of loan loss provision within CASA's balance sheet and €20 billion, above €20 billion, globally for the group despite the very good quality of our asset books. On the following page, the flows of loan loss provisions that we booked this quarter, you see that we continue to have a level of provisioning, which is significantly higher than the one we've booked on the same quarter of last year.

It's 2.5x the same level -- the level of Q2 '19 for CASA and 2x for the group globally. But you can see also that a significant part of this increase is linked actually to an additional effort of provisioning in Stage 1 and Stage 2 loans at €236 million of Stage 1 to provisioning this quarter for CASA and more than €400 million for the group globally. This additional provisioning of Stage 1 and Stage 2 loans is mainly the result of an updating of our economic scenarios within our model. And you have details on the scenario in the appendix of the document. This is leading, and you can see it on Page 15 to -- you can find on Page 15 that the situation is more or less the same for every single specific business line.

It's obviously massively the case for the regional bank where the level of additional Stage 1 and Stage 2 provision is very high, but it's also the case for LCL for CA Italia and for the financing activities of CACIB. A little bit less the case at CACF because the type of credit risk that we have is less relevant regarding the forward-looking provisioning. In terms of net income, you can see the figure I already mentioned, a net profit underlying of €1.1 billion for the quarter and close to €1.8 billion for the first half of the year. It's a very good resilience. And even if we exclude the significant increase that we had to book in our contribution to the Single Resolution Fund, the evolution of the net profit would be only minus 5% on the quarter and minus 8.5% on the semester.

This very good resilience of the net profit is mainly due to a very good operational efficiency with a gross operating income, which is progressing on the semester and stable on the quarter, thanks to the very good diversification of our business model. The return on tangible equity stands at 8.5% for the first half of the year. And if we spread the IFRIC 21 taxes across the full year, it would be close to 10%. If we zoom now a little bit on the different businesses. Starting with the asset gathering and insurance activities on Page 18, what you can see is that actually, we have had net inflows which were more or less 0 on the quarter, very positive market effect, plus €75 billion of additional assets due to the evolution of the market.

But it's nevertheless less than the negative market effects that we had to support in the first quarter which was minus €125 billion. Despite that, the contribution of this business division to the net profit of CASA is significantly up on the second quarter and close to stable on the first half of the year. Zooming a little bit on the insurance activities. Definitely, in terms of commercial activity, the insurance business division suffered significantly during the lockdown period. But what is very encouraging is that the turnover that we had in June only was globally 74%, above the one that we had in May, which was itself above the level of April.

So the pickup is really here. Maybe 1 or 2 other elements I can mention, the revenues within the life insurance activities were up this quarter, thanks to a reversal of some depreciation that we had to book on certain assets in the first quarter, but the reversal is only partial at this stage. The other revenues were quite positively oriented, and the cost base is very well controlled. On asset management and Amundi activities, what we can say is that the net inflows this quarter were close to 0 with actually a positive movement of inflows on long-term assets and still negative flows, net outflows, on money market funds; a very strong resilience of the P&L with a decrease in the revenues linked to the fact that market indexes were, in average, lower in Q2 than in Q1 but, at the same time, a significant decrease in the cost base. And the last point I can mention is that we have closed this quarter the acquisition of Sabadell Asset Management, and we have announced recently the renewal of the partnership with Societe Generale.

If I go now to the retail banking activities, starting with LCL. Again, we will have the same shape of the quarter with a level of activity which was very low in April and again in May and a very sharp rebound in June. And all in all, actually, loan outstandings outside the state-guaranteed loans are up significantly, 7%, between June '20 and June '19. Revenues are a little bit down. The margin, net interest margin, is actually down mainly because of some valuation effects, i.e., the fact that in Q2 '19, we have had some dividends coming from Visa shares or some revaluation coming from the Visa shares that were, at that time, within LCL balance sheet.

And the decrease in fees and commission is mainly explained by the commissions on payments because of the decrease in the volume of payments in the first part of the quarter and also a decrease in account-related commission because LCL decided to be very moderate in that type of commission during the lockdown period. Operational costs continue to be down. It's been -- it's now the fourth year in a row where we see a decrease in the cost base at LCL, and so the net contribution continues to be very positive despite the sharp increase in the cost of risk. In Italy, we see more or less the same trend, a little bit more pronounced, actually, the decrease in the level of activity was sharper in the beginning of the quarter. And actually, this is leading to a sharper decline in the revenues.

The cost base is also down in Italy. Cost of risk is significantly up. And actually, you may see that we have significantly increased the coverage ratio of our NPL this quarter. And so we have, I would say, voluntarily pushed up a little bit the provisioning of our NPLs despite the fact that the level of NPLs did not significantly increase in order to be ready to take opportunities of potential sales of NPL portfolios. For the rest of the international retail activities, we have more or less the same trend in all countries where we are present.

We have a negative dual between a pressure on the net interest margin because of the decrease in interest rates in connection with the reaction of the monetary authorities in those different countries to the COVID crisis and an increase in the cost of risk mainly driven actually by a further increase in the performing loan provisioning, so bucket 1 and bucket 2. In the specialized financial services division, starting with the consumer credit business. It's a business where the duration of the loan in average is only 30 months. So it means that if you lose 2 or 3 months of production of new loans, this is clearly leading to a decrease in the level of outstandings. So it's been the case at CACF.

The consolidated outstandings are down 1% across the quarter so as compared to end of March. And this explains the decrease in the level of revenues in connection also with the fact that part of the revenues are recognized at inception of new loans. What I can mention in addition to that is that the rebound in June is quite sharp and actually to illustrate that, the production of new loans was only €1.3 billion in April, €2.1 million in May and €3.6 billion in June, so a very sharp increase between June and April. And in June '20, we are only minus 13% as compared to what we had in June '19. The cost control continues to be very good, and so we've managed to keep the cost-to-income ratio below 50% on this division.

Cost of risk is, of course, explained by the situation. On the leasing and factoring activities, we have had also a complete stop of the production of new leasing loans up to mid-June. We've managed to increase the commercial production of new contracts in factoring, but actually the turnover, which is effectively factored, is significantly down. So this explains why, again, the level of revenues was down this quarter. Large Customers division.

Globally, this division is generating an increase in the top line by close to 21% with a very good dynamic across the board with both corporate and institutional. For the custody activities, of course, we continue to benefit this quarter from the scope effect and the integration of Santander Securities Services and KAS BANK. But despite that, we have also a very good commercial efficiency with the signature of a significant number of new contracts and also a good profitability of the management of the liquidity generated by this business, which is sharply up. On Page 26, for the CIB activities, what we have seen this quarter is an acceleration of the financing activities in connection, of course, with the credit demand coming from our customers and also a very good level of activity in our different capital market businesses with -- especially in the fixed income currency division, revenues up 44%. We continue to have a very low level of market risk, and the VaR stands at €14 million end of June.

The cost basis is actually very well controlled. It's currently 3.4% up on the quarter. But actually, we had booked in Q2 '19 a significant write-off of a provision for HR-related costs of around €20 million. So actually, if we restate the figure from this provision that we could write off last year, the cost base at CACIB on this quarter is more or less stable. We have an increase, a very significant increase, in the cost of risk this quarter compared to a level which was still very low in Q2 '19.

And despite this sharp increase in the cost of risk, the net income group share at CACIB is more or less stable on the quarter at a very high level. And the cost-to-income ratio at 43% is very low. On the Corporate Centre, to put it in a nutshell, the structural part of the Corporate Centre continues to improve quite significantly as compared to the second quarter of '19. The volatile part is deteriorating a little bit. It's the reversal of the situation that we had in the first quarter.

You know that the most significant part of this volatile part of the Corporate Centre is due to some intra-group transactions. And to put it in a nutshell, when the spread, the credit spread, of Crédit Agricole S.A. narrows, these intra-group elements are negative and on the contrary when they are -- when the spread is widening, which was the case in Q1. Going now on Page 29, the regional banks of Crédit Agricole. You will see, in terms of activity, more or less the same trends as we have seen for LCL with commercial activities at a very low level in April and May, until mid-May at least, and then a very sharp rebound.

So this is also leading light at LCL to an increase in loans outstanding between June '19 and June '20. Outside the state-guaranteed loans, this increase is about 5%. In addition to that, the COVID-related activities have been very, very important with around 400,000 payment holidays granted plus close to €18 billion of state-guaranteed loans, so a very sharp increase. This quarter, the contribution of the regional banks to the net profit of Crédit Agricole Group is up 18%, and the cost-to-income ratio improves significantly. Going now to Page 31, solvency.

And starting with CASA solvency, the CET1 ratio stands at 12%. It's up 60 bps as compared to end of March. It's the combination of several elements. But what you can see is that one of these elements is the very monitored evolution of the level of RWAs because, actually, the level of RWAs is more or less stable, and it includes close to €2.5 billion of temporary penalization in connection with the state-guaranteed loans. You know that the state guarantee starts to kick in only 2 months after the inception of the loan.

So it means that for all loans granted in May and June, we have a penalization in terms of RWA end of June, which is going to disappear in the course of the third quarter. So outside this element, actually, the level of RWAs would have been slightly down of around 1%. The other elements explaining the evolution of the CET1 ratio are, of course, the results that we've published. Again, I mentioned that this result is net of a dividend accrual that we have booked in the second quarter, exactly like we did in the first quarter on our 2020 profit. OCI reserves, which were significantly negative in terms of evolution in Q1 are recouping progressively.

So it's a positive contribution to the solvency this quarter. There's also a positive contribution from the effect of the quick fix and a negative contribution for some other elements like the acquisition of Sabadell Asset Management, which represents a cost of around 10 bps of ratio. All in all, we have a ratio which stands at 12%. And fully loaded of the IFRS 9 transitional effect, it would be 11.7%. So 12%, it's a distance to the Pillar 2 requirement of 4.1 percentage points.

Going now to the solvency of the group globally, you will see that, again, the evolution of RWAs is very moderate. Actually, it's only up €1 billion between March and June. And we have the same state-guaranteed loan effect which is more important at group level, considering the market share of the regional banks in the distribution of state-guaranteed loans. So this effect represents €7.5 billion of RWAs which are going to disappear across Q3. The evolution of the CET1 ratio is explained again by the high level of profit which is kept, which is retained; the evolution of the OCI reserves; the evolution of the RWAs; and then the benefits that we draw from the quick fix and other regulatory and methodological effect, which is leading to this 16.1% CET1 ratio end of June.

So it's 7.2 percentage points above the Pillar 2 requirements. And of course, the other capital ratio, TLAC, MREL and leverage ratios are at or above our different targets. In terms of liquidity, I think there's nothing much to mention. We have a liquidity position which is very, very comfortable with liquidity reserves above €400 billion. It's an increase of close to €70 billion as compared to the end of March.

We've been, of course, using significantly the TLTRO mechanism that was put in place by the ECB, and we've grown actually €80 billion -- €90 billion end of June at this window. And the LCR ratio is around 155% both for the group and 151% for Crédit Agricole S.A., so a very high liquidity situation. The medium and long-term market funding program is completed, actually, end of July. We've managed to continue to issue in the first phases of the lockdown period. And actually, we've completed now, fully completed, the funding program of Crédit Agricole S.A.

So this is it. As a conclusion, I just want, again, to insist on the high level of profitability that we are able to deliver in the very specific period of time that we are going through. I want also to reiterate the very low level of cost-to-income ratio, so the very good operational efficiency that we managed to provide, the high level of return on equity, especially if we spread the IFRIC 21 taxes across the full year. And of course, the very high robustness of the group both in terms of solvency and in terms of liquidity. Thank you very much, and I am now ready to try to answer to your question.

Operator: [Operator Instructions]. And the first question comes from Jon Peace from Crédit Suisse.

Karl Peace: Could I ask a question firstly about impairments? How should we think about that for the second half of the year and going into next year? Do you think it should fall back to some underlying rate if you don't need to change again your macro assumptions? And in that context, could I just ask about your relationship with Wirecard? Has that had any impact on operations? Or do you expect it to in the future? And then if I may ask another one, do you think any difference to the Switch at the moment either positively because it was a benefit this quarter or even with excess capital now that you might accelerate the retirement?

Jerome Grivet: Okay. Let's start with the impairment. It's true that if we don't have to change significantly the scenario, we will have only to update the Bucket 1 and Bucket 2 provisioning with the evolution of the outstandings, so it means the evolution of the amount and also, possibly, the migrations of the loan between Bucket 1 and Bucket 2 or Bucket 2 and Bucket 3.

But this quarter, in addition to all these updating, we had to integrate a completely new scenario which is explaining, certainly, a significant part of the increase in Stage 1 and Stage 2 provisioning. So going forward, I expect that the main driver of the evolution of the loan loss provision is going to be the evolution of the impaired loans. And there, we don't have much visibility. The only thing I can tell is that we don't expect a sharp increase in the number of defaults going forward, at least in the very next quarter because simply all the state mechanisms that were here to provide relief and support to the economy are still operating. On Wirecard, what we have lost is time actually because we were counting on this partnership to develop and to strengthen our offer aiming at developing our business on e-merchants.

And actually, we have invested a little energy and some operational costs in order to develop functionalities in connection with Wirecard. All this has been actually stopped now with what happened with Wirecard. So we have lost time. It's not a disaster for us because, actually, we have already very strong positions in the payment space. We are the #1 card issuer in France.

We have a very strong market share in retail merchants and so on and so forth. But we've lost time. And we have already started to put in place a Plan B in order to develop ourselves the functionalities that we were awaiting from Wirecard. On the Switch, what I can tell is that, actually, the biggest benefit of the Switch is not really when it's triggered because what you have seen here is that the triggering of the Switch is generating only a very modest compensation between the regional banks and Crédit Agricole S.A. The benefit is that we have a capital relief.

And it's true that considering our strong capital position, we are very comfortable with the idea of continuing the dismantling of the Switch in connection with the commitments that we had taken for the medium-term plan. So technically, we must wait for the clawback to take place before we can do another layer of dismantling. But considering the very low level actually of triggering that we had end of June, normally, the clawback should come as soon as end of this year.

Operator: Next question comes from Jacques-Henri Gaulard from Kepler Cheuvreux. Jacques-

Henri Gaulard: Two quick questions.

When I remember your outlook that you gave at the last call for Q1, which was really, really measured, and when I'm hearing the answer to my colleague, Jon, it really seems things are getting better. So even without proxy-ing ourselves to '21 but maybe more '22, do you feel comfortable simply maintaining the targets of the plan as they are looking ahead 18 months? That would be the first question. The second question is really the performance of LCL on net interest income because if I understand well, the sort of like one-off was the dividend on Visa in Q2 2019. So I would take it that instead of going down 3.6%, underlying, the net interest income would have been more or less flat, which is an amazing performance versus competition. So is there any particular explanation for that really, really good performance at LCL?

Jerome Grivet: Okay.

Let me start with the second question. Actually, I mentioned these are dividends, but actually, it was more a valuation effect of the Visa shares. But nevertheless, it's the same. It's a base effect that we had in Q2 '19 that we don't have any longer in Q2 '20. Now clearly, the net interest margin at LCL is very resilient but -- because of the commercial dynamics because, as I mentioned, loan outstandings outside state-guaranteed loans are up 7% between June '19 and June '20, so this is definitely producing the same mechanisms as the one we've already mentioned, i.e., net interest margin, which continues to be a little bit under pressure because new loans continue to be booked at a lower rate than the average level of older loans, but a volume effect that is compensating.

So all in all, yes, the top line at LCL is quite resilient, but we are of the opinion that this should continue going forward, unless there is something unexpected that takes place. And then the 2022 target, well, we haven't decided that it was necessary for the time being to either read through or review these targets. So the main targets are a target in terms of profitability, 11% of return on tangible equity at 11% of CET1 ratio. A target in terms of cost-to-income ratio, a target of solvency, of course, and also a target in terms of net profit. And for the time being, we are still working on trying to reach these targets.

Operator: The next question comes from Giulia Miotto from Morgan Stanley.

Giulia Miotto: I have two. So the first one on consumer credit originations, so you said that it has bounced back strongly. I wonder when it comes to underwriting consumer credit, how are you thinking in terms of the uncertainty around unemployment once government measures taper off? So this would be my first question. So essentially, will this lead to lower consumer credit loan growth because the outlook is more uncertain or actually, no, you're still comfortable and, hence, you will see good growth? And then secondly, so in terms of loans in moratorium, what's the total for Crédit Agricole S.A.? I think you disclosed the French number.

I'm not sure I have seen the total group number. And how are you provisioning for those? So are you taking any provisions? Or are you still assuming that they are performing at this stage? And also on the credit moratoria, can you comment on how that book has been evolving? So is this coming down? Are you still seeing requests?

Jerome Grivet: Let me start with the first question. On consumer credit origination, it's true that we have had a sharp rebound. We are not back at the level we had in June '19, as I mentioned, but the rebound is sharp. There is also, at the same time, some, I would say, a shift in the mix of loans.

And we are more and more granting equipment loans rather than pure treasury loans. So it means that we have a smaller margin on these loans but also, going forward, a lower level of risk. And you know that in France, we have the [Foreign Language] rate, which is preventing actually the consumer credit lenders to really tackle with the lower end of the market because the lower end of the market would dissolve rates in order to have a decent risk/reward level. That would not meet the [Foreign Language] rate ceiling. So this is why structurally in France, we have a lower dynamic of consumer credit, and we have the whole French population, which is not de facto eligible to this category of loans.

So this is clearly, positively, going forward, going to represent a constraint on the development of consumer credit loans. But on the other hand, again, the development of equipment loans is good, and it's proving to be quite dynamic nowadays. It represented a significant part of the credit demand in June. On the moratoria, the €4.2 billion of capital that we have postponed, the payments that we have postponed, it's €3 billion for the regional banks and €1.2 billion for LCL. There is no specific consequence of such a moratoria on the provisioning.

So it means that we continue to book provisions assessing the quality of the borrower and assessing the existence or not of payment incidents. And then your last question was about do we continue to grant a new moratorium. I think that this is now more or less stopped or at least slowed down very significantly because it was really made to cover the lockdown period. And so we are now in a much more normal situation, so I think that the amounts are not going to increase.

Giulia Miotto: And sorry, can I just follow-up, you gave the amount for France, what's the amount globally?

Jerome Grivet: In Italy?

Giulia Miotto: Yes, as well as...

Jerome Grivet: Well, the rest is Italy. And actually, when we say that in Italy we've provided around €2 billion of facilities linked to the COVID, it's about half-half between state-guaranteed loans in Italy and payment holidays. So the capital of the payment holidays that we've granted in Italy is around €900 billion or €1 billion.

Operator: Next question comes from Jean Neuez from Goldman Sachs.

Jean Neuez: I just wanted to ask a quick question on the dividend extension.

In there, there seems to be possibility for banks or for companies, which are part of the group, to upstream dividend. And I think some Spanish banks have hinted at the possibility of using this. If that's right, do you think that it's a possibility for Crédit Agricole S.A.? Or do you think it's not necessarily the best course of action at this particular time? And when we go to 2021, again, on capital returns, if the dividends would be reallowed, so for example, some Italian banks have also continued to say that they want to proceed with payments in euro of the canceled dividends for last year, is this also something that you would be considering continuing? And then I just wanted to ask on the French retail banking. Again, following Jacques-Henri's question, I think it's quite notable also in fees and in net interest income across all revenue lines. And I just wanted to understand whether you could compare and contrast your performance and whether you think your LCL did better than peers.

I'm just trying to understand whether that's a good base to go forward or whether there was anything exceptional in there.

Jerome Grivet: Okay. So on dividend, our capital position is very clear. You have it on the table, 12% CET1 ratio at CASA, above 16% at group level so, in both cases, far above the targets, far above the different constraints that we have to respect. So we definitely have -- we are in the capacity of paying our dividend.

That's the first point. Second -- and you have the figures in mind. But if CASA had paid its dividend regarding 2019, this would have represented a hit of around 60 or 70 bps on its capital ratio as compared to the level that we have now, so still above 11%. And considering the high retention of results at group level, if Casa pays its dividend at the level it was initially forecast, this would represent around 15 bps at group level considering the retention of 56% of this dividend by the regional banks. So this is the situation.

Nevertheless, we are never going to do anything that is not compliant with ECB's regulation. So it's very -- I think, ineffective to try and comment in length the situation. The recommendation is here. It's not really useful to discuss what could be the situation if it was not here. So again, we have the capacity to pay, that's for sure, both at group and at CASA level.

And you were hinting towards a group assessment. So definitely, at group level, it's a very low impact. But we are not going to do anything that is not compliant with ECB's instructions.

Jean Neuez: Okay. And with regards to the French retail sustainability.

Jerome Grivet: On the French retail, I think this is simply proving that our business model, our model in which we try and develop as much as possible, the cross-selling between the different business lines, we stress as much as we can the interest of the customer in the global relationship, which is a relational approach and not a transactional approach. It's efficient. We know that, actually, this model is more widely developed at the level of the regional banks than at the level of LCL, but LCL is permanently trying to catch up with the parameters and the figures of the regional banks. And this is proving to be efficient, clearly. So we don't -- there's nothing exceptional in Q2 '20 performances, and we don't see any reason why those performances wouldn't continue.

Operator: Next question comes from Tarik El Mejjad from Bank of America. Tarik

El Mejjad: Just a couple of questions, please. In your -- on the cost of risk in your scenarios for the second part of the year and for next year, do you -- how do you factor in the risk of some deterioration in your portfolios after all this government guarantee schemes and supports are unwinding? So can you just -- I'm sure you're factoring that somehow, but can you just explain to us a bit how is that weighted in your risk scenarios? And secondly, in Italy, can you comment on how the Intesa-UB consolidation will impact the competition environment, pricing? And do you take -- have to take any management action to correct the trajectory of your growth given this consolidation? And lastly, just coming back on Jean-Francois' question on dividend, I think ECB put in a press release that if any banks can justify some legal obligation to pay a dividend, then it could be looked at. I get your point that there's no point to challenge the ECB recommendation. But is there a case for you that you have to pay a dividend to the parent to support even more the economy and so on.

That's an argument you've tried to make already back in April?

Jerome Grivet: Well, in the cost of risk, as you know, when we assess the cost of risk, when we evaluate the cost of risk, we are -- we embed in the model different scenarios. So the main scenario is described on Page 39. We have also integrated a more favorable scenario and also a less favorable scenario. So we have a whole set of scenario with different weighting of each scenario. And so of course, this scenario, each of them has a narrative.

And in certain unfavorable scenario, the narrative is that the different fiscal measures that we await from the government are not efficient enough and that also, this is leading to the fact that the biggest part of the additional savings put aside by the individuals in France are kept and are not spent. So this is integrated in, I would say, the less favorable scenario. The second element is, of course, that as you can see, the scenario -- the central scenario that we've integrated is less favorable than the one we used in Q1 because, obviously, we have seen the situation evolved between the end of Q1 and end of Q2. And the third comment I can do -- I can make on this question is that for the time being, every information we get, every data that we collect, is clearly proving that our central scenario is what is taking place now. So it doesn't mean that it's going to be the case up to '21 and '22.

But for the time being, what is happening in the reality is exactly what had been forecast by the government and what is clearly the way the government in France but also in the rest of the world were intending to manage this sanitary crisis. So I don't see any reason why we should, for the time being at least, change our views on that. In Italy, it's true that consolidation has been a motto and question on the table since now several years. It's true that lately a significant operation took place. So as always, we are, of course, monitoring permanently the situation.

As I have always said, you perfectly know what type of operation we could be interested in. It's not that type of big consolidation that has taken place recently in which we could be interested because we are -- we have been, in the past, successfully working on incremental consolidation transactions. Regarding the question on dividend, it's very difficult to comment in depth. Could we be entitled to say that we have a legal and binding -- legally binding obligation to pay a dividend from CASA to the regional banks? I don't think it's really a case that we want to open. So again, I think that what is important for us, besides all the legal and technical elements, is to be in compliance with the ECB's instructions.

Operator: The next question comes from Delphine Lee from JPMorgan.

Delphine Lee: So I just have two questions. First of all, just going back to capital. I just want to understand a little bit your priorities in terms of usage of excess capital. I mean you do have quite a bit of buffer over MDA.

Is the plan, if you have excess capital to, first, reimburse Switch entirely and then consider potentially buybacks or paying a special dividend? Or would it be small bolt-on transactions? Just to understand a bit the thinking on capital management. My second question is on TLTRO. So you have the €90 billion you mentioned. Can you just help us a little bit understand what's the benefit that we should expect in terms of net interest income for the second half of this year, I assume in LCL, mainly? And then my last question is on cost of risk. When I look at your Slide 15, we can see from the Stage 3 that there's been some increases, obviously, from low levels.

But are you expecting these trends to continue? Or is the second half levels in terms of Stage 3 provision going to be much lower?

Jerome Grivet: Well, capital management, to start with, we are not going to enter into M&A transactions simply because we have capital in order to finance them. So clearly, we don't think about M&A because we have an "excess" of capital. So clearly, the 2 things are separated. I think that, as we have always said, on a long-term basis so, i.e., besides the present situation, our goal, our target is to distribute half of our earnings to our shareholders and to use the rest of the solvency that we generate in order to develop in a productive manner our different activities. That's clearly the long-term target.

So it happens that presently, the dividend policy is under constraint. We are not going to reassess our long-term capital management policy simply because there is, for the time being, and we expect and we hope it's not going to last forever, a constraint on one of the elements that is key in our capital management policy, which is the dividend policy. In addition to that, what I can say is that unwinding the Switch has always been in our idea a way of trimming a little bit in an opportunistic manner some excess of solvency in order to enhance our earnings per share without having to modify our dividend policy and without having to enter into a share buyback policy, which would be a little bit detrimental to minority shareholders because of the presence of a majority shareholder that is not going to participate, nevertheless, in this share buyback policy. So we have this additional tool, which is useful in order to fine-tune, I would say, our capital trajectory, which is the Switch mechanism. And we are committed to go up to at least 50% of dismantling by 2022.

So these are, I would say, the key components of our capital management policy. And for the time being, we don't want to modify these key components of our capital management policy simply because there is a temporary constraint on the distribution of dividend. TLTRO, we see the TLTRO as a way for us to globally finance our balance sheet. So we are not in a position where we say we are going to take a certain amount of TLTRO that we are going to dedicate a certain type of asset. We are managing globally the balance sheet of the group.

We try permanently to combine security and, I would say, profitability: security because we want to avoid a situation where we have a liquidity cliff at a certain point in time and profitability because we try permanently to have the cheapest possible way of financing our balance sheet. So clearly, the TLTRO belongs to the tools that can help reduce the cost of our funding. And this is exactly what we had in mind when we decided the level that we have drawn at this window with the idea, of course, that, at the same time, our asset book is going to grow significantly because we are very active to lend to the economy and we have been very active, for example, to develop and to grant state-guaranteed loans. So clearly, the fact that we have drawn this amount to the TLTRO is going to help our P&L going forward. But at the same time, don't forget, for example, that we have put in our balance sheet, several billion euros of state guaranteed loans with a level of customer rate which is around 0.

So at the end of the day, the profitability is the difference between the cost of the liabilities and the yield of the assets. Cost of risk going forward, I already mentioned that the evolutions in Stage 1 and Stage 2 provisioning is going to depend on our capacity to maintain the scenario and the main scenarios that we have used end of June. For the time being, as I said, we see no reason why we should change these type of scenarios. And the second component of the cost of risk is definitely the Stage 3 provisioning, and this depends vastly on the evolution of our counterparts. What happened in the second quarter, which explains a significant part of the evolution of the Stage 3 provisioning, is the fact that, first point, certain counterparts that were already very fragile before the beginning of the lockdown were so fragile that, actually, they couldn't claim any state-guaranteed loan or equivalent help really defaulted because of the lockdown and so entered into Stage 3.

The other elements that we have seen, especially at CACIB, was the fact that certain counterpart that we had across the world proved to be actually a fraudulent counterpart. And so this is -- this represented a quite material component of the Stage 3 provisioning in the second quarter, and we don't expect that to be again the case going forward.

Operator: The next question comes from Guillaume Tiberghien from Exane.

Guillaume Tiberghien: First, I've got a request. Would it be possible to get the NPL ratio for the sensitive sectors? So you provide them in Q2, which is very useful, but could we have the level in Q1, so that we can see the quarter-on-quarter evolution? And obviously, if you could continue to provide that going forward.

My 2 questions are -- number one is on SFS. The margin, obviously, is quite a lot under pressure. You had well flagged it. But can you give us a little bit of an outlook going forward for the margin in consumer credit? The second one is back to capital, your current target is 11%. I'm not asking you to spend your excess capital because you can't at the moment, but actually, it's more about the target.

Should you not reduce it given the reduced MDA? And maybe just to follow a small third question. With regard to your scenarios, you expect -- you explained the different weightings of those scenarios to get to Stage 1 and Stage 2, what are the weighting of base versus adverse versus good scenario that you use currently?

Jerome Grivet: Well, let me start with your last question because the answer is going to be quick. We don't disclose the weighting of the different scenarios. So I'm sorry, Guillaume, but we are not going to disclose it right away, sorry. Going now to your second question regarding the CET1 target, it's true that actually with this margin above SREP, which stands at 400 -- above 400 bps, we are far above the type of buffer that we would want in the long run to keep above the Pillar 2 requirement.

We don't necessarily think that the moment is well chosen to modify the target because, first, of course, this is an issue that needs to be discussed with the ECB. And second, this is an issue that needs to be explained calmly to the market. But theoretically, it's true that all things being equal, the different regulatory requirements have reduced as compared to what they were end of last year by at least 60 bps for CASA. And again, all things being equal, and in a normal period of time, this could trigger a reduction -- a proportional reduction in the level of our CET1 target. So this is an issue which we perfectly have in mind and which we are going to address when the time has come.

Going now to specialized financial services, I think that we are having a structural shift in the mix of the loans that we grant, especially in France. You know that in France, there is a trend, since now several years, the reduction of revolving loans and to the increase of amortizing loans because of legislative modifications. And so this is, of course, putting a certain pressure on the customer rates. But at the same time, this is, in the long run, providing probably a better level of cost of risk. And this has -- this mix has also been impacted in the last period of time by an increase in the car financing business within Agos and within Sofinco.

So it means in addition to the car financing businesses that we do through FCA Bank, there is also a growing demand for car financing loans directly addressed to our Agos or Sofinco branches lately, and this is also a movement that is going into the same direction, i.e., longer loans, i.e., less risky loans but also lower rates. So there is -- there are several important shifts that are taking place in the business of auto consumer credit nowadays, which we are going to assess. The first reaction of the business is to monitor as closely as possible the cost base, which has been the case in this quarter. And the best answer is to continue to monitor the cost base in order to adapt to this potentially a lower level of margin in the medium term.

Operator: And the next question comes from Pierre Chedeville from CIC.

Pierre Chedeville: Yes. I have one question on the insurance. I was listening to the con call of AXA this morning. And they explained that their combined ratio, the impact of business interruption claim was balanced for RV amount by the better claims on motor. So my question is, could you give us any color on that, the balance between business interruption and better claims on motor? And also, you mentioned that you have paid €143 million -- I guess €1 million to professionals for business interruption.

But you mentioned that it was a volunteer act, a mutualist act. But I guess that you have also paid because it was in the contract, so I wanted to know if...

Jerome Grivet: No, no, no. Clearly, and this is an important point because, actually, as you know, in France, considering all the buzz that took place around this issue of business interruption guarantees, there has been an audit by the local insurance supervisor, the ACPR. And actually, the ACPR fully acknowledged that our policies were absolutely not ambiguous regarding this question.

So we have had, I think, no claim or almost no claim from any of our customers on this issue. And this has been a purely, completely, totally voluntary decisions that we've taken. It's a decision that we have taken in our interest because, going forward, this is going to be commercially positive for us clearly, and we expect this to generate and to improve further the customer loyalty and our attractiveness. But regarding the policies themselves, we had absolutely no elements that could lead to some kind of deal as were mentioned by some other insurers. So clearly, no issue at all regarding our policies.

Pierre Chedeville: And concerning motor because the ratio remains quite low, actually.

Jerome Grivet: Yes. The combined ratio continues to be very good. And actually, the ratio between the claim indemnification globally and the premium that we got significantly improved in Q2 as compared to Q1 this year or Q2 last year. That's true.

That's definitely true.

Pierre Chedeville: But was it due to motor claims?

Jerome Grivet: And it was massively due to the car insurance business. Unfortunately, our market share in the car insurance business is not as high, for example, as it is for the home insurance business.

Pierre Chedeville: Okay. But there was less robbery also, I guess.

Jerome Grivet: Yes. But you know that the amount of a premium on home insurance is lower than on the car insurance policy.

Pierre Chedeville: And my second question is just technical, a very quick question. Your buffer against threat is 410 and your buffer against MDA, as far as I understand, is 380. So the difference is 30 basis points, what exactly is 30 basis points? Is it P2G? Or is there something I don't...

Jerome Grivet: No, it's not P2G. It's much more technical, actually. It's explained, I think, on the credit update that we've put online also today. On Page 24, you have the different elements that explain both the distance to P2R SREP and to MDA because it's linked to the Tier 1 capital ratio and not the core Tier 1 capital ratio. So you have all information.

And Emilie, for example, would be absolutely delighted if you have a basic presentation on that, if you wish.

Operator: The next question comes from Lorraine Quoirez from UBS.

Lorraine Quoirez: I have a question on the Corporate Centre. When I look at the Slide 27, you talk about decreasing financing costs, operating expense and so on. And obviously, there are quite a few, I would say, one-off items or specific items this year.

And I was wondering whether you could give us a little bit an indication on how we should think about the Corporate Centre, in particular next year?

Jerome Grivet: Well, the structural path is made of 3 different buckets, as you know. And I think we are now giving some clarity on these 3 different buckets. The first one is what we call balance sheet and holding, and this is the one on which the trend is permanently to do the reduction. So what we expect and what we are working at is to continue to reduce the cost of this bucket. Then there are 2 other buckets which are structural.

The first one is all the businesses which do not sit within a dedicated business line, so typically, the real estate business, which is a small real estate development business. We have also the private equity business, and we have some other minor activities, which are structurally inside the Corporate Centre but which may post a certain volatility in their P&L. And clearly, this quarter, they generated a negative P&L because the biggest part of this bucket is made up of private equity activities. And the valuation of the private equity funds that we have there was negatively oriented. And then the third part is structurally converging towards 0 because it's a series of support functions that work for the group.

So they have costs, but they bill their costs to the business lines regularly. So it means that on a specific quarter, the P&L may be a little bit positive or negative. But across the board, on a long-term basis, it's converging towards 0. So the main item which is the one on which we work harder in order to reduce it is what we call balance sheet and holding. And you see that really, this is improving regularly, and we hope and we are working in order to make it continue to reduce.

And then there is the last element, which is the volatile part, which I explained rapidly, this can be structurally quite volatile with significantly high or low levels. Again, this is more or less converging towards 0 in average, but the volatility, which is linked especially to intra-group accounting elements, may be significant on a specific quarter.

Lorraine Quoirez: And perhaps just very quickly as a follow-up, the balance sheet and holding part, how fast can we expect an improvement in the financing cost?

Jerome Grivet: It's a multiplication of volume effect and a price effect. So what we try to do is to reduce the volume of the debt that is carried by this Corporate Centre. And so every time we manage to upstream dividend from a subsidiary, this is reducing the level of debt that we need to have.

And every time we manage to lower the cost of the funding that we still need in the Corporate Centre, we also manage to reduce the overall cost. So these 2 elements are clearly the key drivers. The average cost has been significantly reducing in the last few years and is going to continue to shrink because of market conditions. In terms of volume, it depends on the capacity of upstreaming dividends. So it depends on the profitability of the subsidiaries.

And it depends also sometimes on the acquisitions that we have made because, for example, when Amundi acquired Pioneer Investments, Amundi needed the capital increase. We participated in the capital increase at CASA, and so this led us to pay a significant amount so suddenly to increase again the cash-out on this Corporate Centre.

Operator: The next question comes from Azzurra Guelfi from Citi.

Azzurra Guelfi: A couple of questions for me on the deposit. When we look at deposit, evolution has been very strong across the different parts of the business.

How do you expect this to evolve going forward in terms of, like, the risk appetite of customers? Will they switch to asset management product? Or do you think that it will continue to remain liquid or, two, as a switch, a consumption more oriented thing? And then on the Large Customers, can you tell us how do you see the business evolving for the second part of the year and next year, especially in the financing? Because I understand that the market component, it's harder to forecast.

Jerome Grivet: In both cases, I'm going to be a little bit disappointing for you, Azzurra, because this mainly depends, in both cases, from the customer behavior. When it comes to especially side deposits, actually, of course, we expect and we hope that the amount of cash that has been parked on the side deposits is going to reduce because this is -- this would prove actually the real confidence of the household. So we expect that at least part of this excess of cash that has been put on the side deposit is going to be used to increase the consumption. But for the time being, we have to wait because, yes, the consumption has significantly increased but compared to the very significant amount of cash that was filed on the side deposits, it's only a small part of this success that has been spent up to now.

And it's more or less the same for the Large Customers division and the financing activities because we are here and ready to answer to any request of our customers that need to be financed. But of course, if, at a certain point in time, there is a weaker business sentiment and a lower appetite of our corporate customers to raise financing in order to develop their activities and to invest, we'll have to deal with it. So for the time being, we don't see any slowdown, but we are going to see a slowdown, which is the usual summer slowdown. So we have no sign of deterioration of the business sentiment. But in both cases, confidence and business sentiment is going to be the key.

Operator: And the next question comes from Kiri Vijayarajah from HSBC.

Kirishanthan Vijayarajah: So my first question is on costs. So at group level, you've done really well on cost. But I just want to better understand the cost outlook in Large Customers. And I appreciate there's been some perimeter effect distorting the first half.

But really, are you working on kind of cost efforts within large corporate as you seem to be doing elsewhere in the group? Or is it more a case that, look, you've got this revenue windfall in the first half so, you know what, there's no pressure to really tackle the cost base in large corporate? So really, the first question on the large corporate cost outlook. And secondly, in car finance, was there any kind of impact from second-hand car residual values? It doesn't look like you've taken anything. You've not flagged anything. So I wonder if that's something you might need to do at the back end of the year. So what's your kind of outlook for car residual values in the car finance segments?

Jerome Grivet: Let me start with the Large Customers division.

It's clear that in this division, as we do it in every division, we have set targets in terms of cost-to-income ratio. And the responsibility of the manager of this business line, as is the case for every business line, is to reach his own cost-to-income target. So of course, this quarter, considering the high level of revenues, there is a supplementary improvement of the cost-to-income ratio. And I'm not pretending that we are going to be able to monitor this business line at a cost-to-income ratio of 43% permanently, but we have been operating the CIB with one of the lowest cost-to-income ratio of all European CIBs in the last few years, and we intend to continue. So we're working on the cost base within the CIB as we do in every other business line.

In the car financing business, it's true that as we are a large car financing operator, we are permanently monitoring the residual value of cars. But actually, the biggest part of the business that we do in car financing is a retail business. So it means that what is important for us is, of course, first, the capacity of the borrower to repay the loan rather than for us to repossess the car. And you know that for retail car financing loans, actually, the residual financial value is much smaller than when it comes to fleet financing. So it means that, actually, the protection that is given by the car is bigger actually than in fleet financing.

But as in every business in which your guarantee is at least partially made of real asset, you need to assess regularly the capacity of having a value of this asset which is at or above the remaining capital. That's a necessity.

Operator: And the next question comes from Omar Fall from Barclays.

Omar Fall: Just two for me. So the first one, sorry to come back to the top line in French retail, but might I just really understand the NII dynamics there at all? If I simply take net interest income ex PEL/CEL divided by loans, it's like a marginal decline in the quarter.

It's basically not that much different to Q4. Yet you've had the high-volume growth on the deposit side, which is higher than the asset side, including current accounts, up 30%. So that's negative. You've got the state-guaranteed loans, which should be negative. And then you've had lower swap rates, which are also negative.

So how is net interest margin is so stable? If you can maybe just give us some more color. And then secondly, could you just discuss GAC Sofinco specifically, please? Because, I guess, that business is like 2 to 3 months ahead of the rest, how do you call it, of consumer finance in navigating the crisis given where China is, and it's a different market. But like, how recovered is that? Is it basically back to a normal run rate of the top line in asset quality? I know you gave, like, some volume changes between June and March, but it's hard to understand those because we don't really know the base.

Jerome Grivet: Okay. So for GAC Sofinco, I'm turning towards my colleagues in order to get the precise information.

I don't have it in mind. When it comes to LCL, so just bear in mind that the outstanding, loan outstandings, were quite significantly up between end of June '19 and end of June '20 and that the mix is also improving because -- I don't have, again, the precise figures in mind, but I think we have them in the appendix. There has been an effort at LCL in order to develop further, first, consumer credit loans; and second, the SME and professional loans which generally have a better customer rate than home loans. So there is a mix effect that has also, I think, played a positive role in the preservation of the net interest margin. But all in all, clearly, the volume effect has continued to be positive for LCL.

For GAC, do you have the information?
Unidentified

Company Representative: Well, we have -- there was a slowdown in activity that was relatively moderate with a 10% decline in commercial business compared to the second quarter. But the activity has been buoyant again since June, in particular, with GAC Sofinco where we had a strong increase in production, reaching again the level that we have had before the crisis. So we're having a pickup in GAC Sofinco. We have to be relatively prudent after the outlook again in China for GAC, but it's -- the lockdown was earlier and it took place also during the Chinese New Year, which makes us relatively confident as to what's going to happen this year for GAC Sofinco in terms of the budget.

Omar Fall: Understood.

Sorry, just a quick follow-up. So I guess the second half of the quarter would have had much better loan growth than the first half, implying that that net margin -- that's actually negative to the net margin because of the averaging effect. And so are you -- you seem quite confident in the outlook. Does that then mean that you need volumes to really rebound very significantly from here to offset that margining effect?

Jerome Grivet: It's true that the level of activity at LCL has rebounded quite significantly in the second part of the quarter. So it means that going forward, on the course of the third quarter, the average level of loans is going to be significantly higher than the average level of outstanding of loans that we had in the second quarter.

So no, we're positive with all the constraints that weigh and continue to weigh on retail banking activities in France, i.e., the cost of regulated savings accounts and the fact that the competition in home loans continues to be aggressive. But with all these constraints, LCL is doing well, has a very, I would say, coherent customer approach and is developing the equipment rate of its customers with the different products that we sell, and we are positive on LCL.

Operator: And the next question comes from Stefan Stalmann from Autonomous Research.

Stefan Stalmann: I have two topics, please. The first one, again, dividends, but from a different angle and I think you mentioned it in the context of the Corporate Centre.

How have you actually handled intra-group dividends that are upstreamed into CASA, like from CACIB or the insurance business, et cetera? How have you handled this in the first half of the year? If you could explain that, please? And the second question goes back to the consumer finance business. I think if you back out the minority interest in consumer from the tax benefit at Agos, it seems as if the underlying business of Agos was basically breakeven in the second quarter. Is that a fair estimate? And also, it looks like the equity income from the joint ventures is holding up relatively well compared to what's happening in the consolidated loan book. And I was curious if these joint ventures are actually using the same IFRS 9 methodologies and also the same timing of applying them as CASA in its main entities, its controlled entities or whether there are, let's say, time lags or permanent differences in the models that are being used?

Jerome Grivet: Let me start with the dividend question. The principle that was used in the first quarter is that all 100% subsidiaries were entitled to upstream their dividend.

So it means that CACIB, LCL, for example, upstreamed their dividends. Contrary to that, all the subsidiaries in which we have minority shareholders couldn't upstream the dividend. So it means that CACEIS, for example, or Crédit Agricole Italia or, even more obviously, Amundi couldn't upstream the dividend this year. So it's a mix of different situations with which we had to deal. In the consumer credit business, let me start with the equity accounted entities.

Yes, of course, they all apply the same type of methodologies that we use in order to evaluate the IFRS 9 provisioning because you know that, in all cases, we have the management of the support and risk functions. So we have the capacity to really monitor that. It's a mix between -- you know that within the equity accounted entities in the consumer credit business, we have the car financing entities where the business continued to be quite positively oriented as soon as the lockdown period ended and where the cost of risk is not very significant because of the type of the business. And there is also mainly another traditional consumer credit business which is Wafasalaf in Morocco, and Wafasalaf has had a significant increase in its cost of risk. And this led to a decrease in its contribution to the equity accounted entities.

So the minus 22.7% that we had leading to the €60 million contribution is a mix between a good performance coming from the car financing entities and a weaker one coming from the Wafasalaf entity. If -- if I really -- I'm going to check that I really understood your -- the other part of your question. What you're saying is that your math is leading to the conclusion that on this quarter, Agos just broke even. Is that the question that you were asking?

Stefan Stalmann: Yes, because the minority interest in consumer finance is basically identical to the minority interest that is associated with the tax benefit at Agos. So it would seem as if the rest of Agos was basically not generating minority interest and, therefore, probably no profits.

Jerome Grivet: Yes. It's clear that this quarter, the cost of risk at Agos was significantly higher, especially because you know that in Italy, the lockdown was quite strict. And you know that in Italy, the collection on home -- on loans with payment incidents is performed by -- I would say physically. And so it means that for 1.5 or 2 months' time, they have not been or almost not been any physical collection on payment incidents in Italy. So it's clear that we have had, probably this quarter at Agos, an extra amount of provisioning led or triggered by this impossibility of collecting physically the repayments on the loans.

So we need to wait until the situation is completely normalized in order to check whether this is a real incident, payment incident, or whether simply, the simple fact, that the collection can take place again is triggering the regularization of the loan. So we'll have to reassess the risk situation at Agos in the course of the third quarter.

Operator: The next question comes from Matthew Clark from Mediobanca.

Matthew Clark: So a couple of questions. First one, could you just help me understand the economics of the government guaranteed loans in France.

I think you said clients were paying around 0 for the loans. If you can fund yourself at 100 basis points below, using the TLTRO 3 as your marginal funding source, does that mean that you're earning a percentage point of interest margin net on these volumes? Or am I missing something? In which case, please tell me where I'm going wrong there. And then next question on the tax rate. Was there anything else this quarter that led to the low tax rate because the affrancamento, €39 million, would seem to only be explaining part of the reason that it's so low. So anything going on there? And then finally, on rating migration within risk-weighted assets, do you have any expectations there for headwinds in coming quarters? Or do you think it's not really an important effect for you? Or if it happens, would that be a next year effect, kind of timed with the end of government support measures? Just any thoughts there on -- in terms of magnitude or timing?

Jerome Grivet: Okay.

So on state-guaranteed loans, the rate is 0, and the borrower has simply to pay a premium which is the price of the state -- of the guarantee. So for the first year, this premium is set, I would say, at a flat rate of 25 bps, and this 25 bps is split 90% for the state and 10% for the bank, which is keeping the premium on its own part of the risk taking. Then -- so it means that for the loan itself, outside the guarantee, the rate is indeed 0. It's 0. So of course, if we were able to say that we are marginally financing a specific state-guaranteed loan with a specific drawing on the TLTRO, we could say that we are going to trigger and to book a net margin of 1 percentage point.

But actually, again, what we do is that we manage globally the balance sheet and the balance sheet is moving permanently, especially in these present circumstances where we have had many, many changes in the behavior of customers. So to put it differently, the TLTRO is going to help decrease a little bit the average -- the weighted average cost of the funding of our balance sheet. But at the same time, the state-guaranteed loans are not going to help boost the average -- the weighted average yield of the assets that we have on our balance sheet. So the answer to your question fully depends on whether you want to have only a marginal reasoning or if you really take into account the reality, which is the global reasoning of -- the global monitoring of the balance sheet. Rating migration, well, we regularly review the rating of our different counterparts.

We have to do it regularly. And of course, every time the credit quality of a counterpart deteriorates, we have to downgrade its rating in our system, and this is triggering an increase in the level of RWAs. On this quarter alone, this led to close to €2 billion increase in RWAs at CACIB. So it's not nothing. It's not nothing, definitely.

And possibly, it can continue going forward. So this rating migration effect is part of the more global set of elements that trigger modification in the level of RWAs. But in itself, probably rating migration will continue to be negative going forward, again, €1.9 billion of additional RWAs at CACIB in the last quarter.

Matthew Clark: Can I just check the -- how do you assess the counterparty rating there? Is it your internal assessment based on, like, interest coverage? Is that the kind of metrics? Or are you relying on external parties to -- external ratings process as well?

Jerome Grivet: It's an internal process that takes into account all relevant information, but it's an internal process with which we rate, and we have an internal rating of all our counterparts. And as you know, within the group, we have several entities that can lend to the same counterpart.

And then there is one of the entities of the group, which is the leader in assessing this rating. So for example, generally for the larger corporates, it is CACIB. But for some smaller enterprises, smaller businesses, it can be either LCL or CACIB or one of the regional banks, it depends.

Matthew Clark: And then on tax. Sorry, I interrupted.

Jerome Grivet: Yes. And then on tax, excuse me, I was just looking at my notes. On tax, well, there is this affrancamento effect, which is generating a decrease in the level of tax of around €60 million gross at Crédit Agricole Consumer Finance. Besides, you know that there are 2 elements that generally trigger a reduction in the level of the tax rate. The AT1 coupons which are tax deductible, but they do not play a massive role this quarter because it's not completely spread regularly across the different quarters.

And if I remember correctly, it was not a very significant amount this quarter. And then we have also the fact that some of our profits in different businesses are booked in geographies where the tax rate, corporate tax rate, is lower than in France. It's the case, obviously, for a big part of CACIB activities, but it's also the case for some other activities. So this is leading to an average tax rate which is structurally lower than the marginal corporate tax rate in France. In addition to that, within the tax line, we can have regularly ups and downs that are linked either to the booking of certain revenues which benefit from a lower tax rate.

It happens sometimes in the insurance business. But also, we have some tax uncertainties, tax litigation, and we saw -- we booked some provisions. And generally, with tax provisions, we are as prudent as we are with credit provisions. So it means that every time there is a tax dispute that is settled, generally, it is settled with a lower cost than the one we had initially booked, and so it means that we can have on a specific quarter some reversal of tax provisions that benefit to the average tax rate. And so there have been some bits and pieces of elements like that this quarter across the board.

So this is leading to the provisional, I would say, tax rate of the quarter because you know that when it comes to taxes, the reality really takes place at the end of the year.

Operator: And the next question comes from Jacques-Henri Gaulard from Kepler Cheuvreux. Jacques-

Henri Gaulard: Yes, sorry, a couple of follow-ups. On the tax, Matthew's question was interesting. There is just this huge €77 million recovery on the financing business, just to have an idea of what it is.

And the second question, what is quite interesting about the performance of the division is Wealth Management, which is doing extremely well, which has almost doubled its profitability versus last year. So again, is it a base from which we can actually work on? And are you comfortable about the return on net equity for 2022 which for this division is, I think, 25% and would correspond to a doubling of current profitability? And lastly, I would be delighted to join the insurance one-on-one pitch-in between Pierre and Emilie.

Jerome Grivet: Okay. On Wealth Management, it's clear that the business is improving since now several quarters. Actually, we had reached earlier, beginning of '19 or even end of '18, a low level of profitability.

So Jacques Prost, the head of the business, is working on improving the cost efficiency, on improving the revenue-generation capacity. And so we expect globally the trend to continue. Of course, there can be some ups and downs on a quarterly basis. The first quarter was very good revenue-wise. The second quarter was a little bit weaker from a revenue point of view.

But it was better also from a tax point of view, to go back to this point. And so all in all, the bottom line improved, but it's not there yet in terms of profitability target. Globally, your second question on the tax rate... Jacques-

Henri Gaulard: The financing division, yes, which is a positive, yes.

Jerome Grivet: We have to dig a little bit into your question before answering.

I don't have the answer offhand. Thank you. I think we have one last question.

Operator: Yes, and the last question comes from Phelbe Pace from Societe Generale.

Phelbe Pace: Phelbe Pace from Soc Gen.

So a quick follow-up question on LCL and on mortgage, specifically, please, that I'd like to squeeze in, if I may. So you seem quite confident about the mortgage volume outlook at LCL in the second half of the year. However, the [Foreign Language] estimates now that mortgage production levels in France should fall by 26% in 2020 if the country's GDP declined by 10% this year, which is, I think, your main assumption. So I'd just like to understand what makes you confident the LCL could maintain good mortgage volume growth in the second half of the year in such a market context.

Jerome Grivet: Well, I said I was confident about LCL.

I didn't specifically say I was confident about the capacity of LCL to grow the home loan volumes. So globally, the business of LCL is behaving quite well. But of course, if there is a certain downturn on the evolution of the home loan market, it's going to be also felt at LCL. But as I said earlier, LCL tried to diversify a little bit the breakdown of its different categories of lending by increasing the corporate volumes, by increasing the consumer credit loans, too, and the loans also aiming at professionals. So what is going to be important for us globally for LCL is the capacity of keeping a good dynamic of loans globally, not only home loans, and also a good dynamic on the sale of other products and services, so not only lending because the key of the business, the retail banking business, in France is really to be able to continue to develop the whole set of activities around the customers.

So I agree that there might be a slowdown in the home loan business, but LCL does not rely only on this business. Well, I think it was the last question. So let me thank you for your attendance to this meeting. And let me wish you also nice holidays, at least for those of you that intend to take some holidays, and it's going to be my case. So have a good afternoon.

Bye-bye.

Operator: Thank you. Ladies and gentlemen, that does conclude our conference for today. Thank you for participating. You may all disconnect.