Société Générale Société anonyme (GLE) Earnings Call Transcript & Summary

Ladies and gentlemen, welcome to the Société Générale Conference Call. Frédéric Oudéa, Chief Executive Officer; and William Kadouch-Chassaing, Chief Financial Officer, will present the group's first quarter 2020 results. Gentlemen, please go ahead.

Yes. Thank you. Good morning to all of you. Thanks for attending this call. And first of all, let me, of course, say that I hope you are all well. We decided in a market environment, which has been volatile and with rumors flying around, to anticipate by 1 week the communication of our results and present as soon as possible the complete and stabilized vision of these results for the first quarter. Let me just mention the only limitation to that is that we will not be able to comment in detail on our quasi subsidiary as we will release their IFRS figures on the 6th of May. We'll enter into figures very quickly. I will leave the floor to William, and of course, we'll answer your questions. I just would like to say a few introductory remarks. First of all, as we all know, this crisis is first, a sanitary crisis. And I just would like to highlight that, of course, we had as a core priority to protect the safety of our clients and our staff. And I think we were able to demonstrate capacity to do that efficiently. Secondly, this crisis will have a very serious economic and financial impacts. I've gone through different crises in my career. I lived the 2008-2009 financial crisis. It's certainly a crisis, which will have more impact as we see and with still a lot of uncertainty. We are working still on different scenarios for the GDP impact. We have a big case, which correspond to a minus 6.8% drop of GDP in the Eurozone, a minus 6.6% in the U.S., which seems to be the one developing. It corresponds, in particular, to a 2-month consignment in these economies. And we have other scenarios, in particular, an extended health crisis scenario, which might be related also to a second wave of contagion at the autumn and which corresponds more to a minus 13% drop of GDP in the Eurozone. I think we need to remain very humble, of course, on these scenarios, but also on the impact of the government reactions and authorities' reaction to this crisis. And while this crisis is extraordinary by its magnitude and diversity, the response of the authorities, central banks, governments and regulators have been also extraordinary. And when I say this, I had in mind, of course, in particular, the capacity to quickly implement these extraordinary schemes. Let me just say, if I take France, just 5 weeks after having started to implement this new guarantee scheme, the level of demand for the overall banking sector is EUR 80 billion. We stand on our side at EUR 14 billion. And at the French level, that concerns 420,000 companies. And what I mean is that this capacity to put that quickly in place will, of course, have an impact, which is still difficult to factor and forecast on the cost of risk and the time. And final point, I'm sure that we will see additional measures, whether it's in Europe, for example, to think about how to finance and stimulate the rebound of the economy. Or whether it could be, for example, to even transform certain of these exceptional loans into semi-quasi equities. So just to say that we are still facing uncertainty as we speak. If I go to Page 5, and I just would like to say that if the concept of being a responsible bank means something and responsible banking, I think it is in this crisis that we can illustrate this. There are a lot of talks explaining. And I think it's true that banks are part of the solution. And we have to commit. As I said, we have committed to protect our staff and clients. For me, it's extraordinary to see what we've been able to achieve. To a certain extent, the fact to be able to communicate and disclose our figures the week in advance is one illustration, and I would like, of course, to thank the teams for that. But beyond, the way we also deal everywhere in the world with our major projects, whether it's business project, we were able to complete an important milestone in the migration of the former Commerzbank equity business at end of March, but also on the remediation projects, and that's very comforting. The second thing is the capacity to deal with our clients. I've mentioned the government schemes, but much beyond, we've been able, and all our teams are on the board with our clients, for example, with all the DCM businesses that we saw. And I think thinking forward that what we are delivering today will help to further enhance the relationship we have with our clients. Our clients will not forget the way we have behaved. And third, with the communities, of course, our societies are heavily impacted in many sectors, as we all know, in many countries. And I feel that we have also the duty to try to help as much as we can. We have designed a global solidarity program with an overall envelope that we estimate at something like EUR 50 million which will be an international one with different donations. And I just wanted to mention that we, as the management team, will contribute to the financing of this program as we have decided and informed the Board that we will give up half of the variable compensation that the Board will decide to attribute to us in due course for the 2020 year. Let me now turn to Page 6. Just to say, I really believe that we are entering into this crisis with a strong risk profile and a stronger risk profile than in 2009, if I take this comparison here. When I look at our credit risk, clearly, we have drawn the lessons of the previous crisis, and I think stick to the strict origination criteria, we will go through, I'm sure, our different exposures in detail and the quality of this exposure. Just would like to take one example, which is one factor that people, of course, look at in a crisis like this, the LBO. We've maintained a EUR 5 billion direct exposure, euro unchanged since several years. But more importantly, when you look at the quality of the exposure we had in 2008, 2009, for example, junior exposures, second lien, things like this, this exposure today is a much better protected, much better quality and predominantly European. As you see also on the slide, we are -- have the portfolio with a low NPL ratio, 3.1%, that's even gone slightly down in the first quarter. It's even lower when you look at the way the EBA measure that below the average in Europe and with a good coverage ratio. The second thing is, of course, the capital -- the capital level. We started with 13.2% core Tier 1 ratio, reintegrating the provision on the 2019 dividend. We spend at the end of March at 12.6%, 12.7% actually, pro forma of disposal that is proceeding this year of our Norwegian and Swedish, Scandi leasing business, which means that it gives us a 350 to 360 basis point buffer and above the regulatory threshold. The MDA, which now stands at 9.05%. You will see that we have also increased our liquidity buffers. William will go on that more in detail. The third element is that we are functioning and operating a more compact business model with good franchises across the board. And effectively, we feel confident because we have good positions, including when I look at the geographies, and here, it's a very difficult thing, all geographies will be, by definition, impacted, let's face it. But overall, I think we have countries which can support the economies. In all, we don't have big exposures. For example, when I think about the U.S. retail operations of banks with consumer credit, et cetera, we have safety nets in Europe, which will limit the impact, for example, on the individual plans. So overall, we feel like we can deal with this crisis. And fourth, this crisis for probably many companies, but for us, has been a way to tackle also our digital and technology infrastructure. And it worked very well, as I said, across the board, and whether it's in developed economies or, for example, with our offshore operations in India, everything functions. And that's good because if we were to come back to consignment, we can do that. And we will probably, of course, going forward, deal in a different way in our functions. Now let's turn to the figures. I will just say a few words on that, Page 7, and then William will go some more into detail. Clearly, we have mixed results. And when I see this, actually, there's nothing really to say on the bulk of our activities. Retail operations, you will see, are doing pretty fine, have remained solid. Of course, they have started to see the impact of the crisis end of the quarter. We have had -- we have also resilient financial services operations. And we had in GBIS, a very strong strict business on the capital markets and good activity on the financing side. We had one, of course, big disappointment, which is on the equity revenues and more specifically, on the structured investment products. We'll come back on that more in detail. You will see that it is a mix of some increase of reserves, the EUR 1.75 million, but also the impact, the negative impact of the dividend cancelation. And of course, the cost of hedging, we will come back to that much more in detail. The second element of these figures are, of course -- is, of course, the increase of the cost of risk. This cost of risk stands at 65 basis points. It's 3x higher than in the first quarter of 2019. We see there the first, of course, impacts of the COVID scenario. Same thing, we will answer more in detail in this cost of risk. What we've been doing in the first quarter is beyond provisioning for certain specific sites, including because of the COVID, is to add overhead provisioning. We will refine our scenarios going forward. I think in the second quarter, we'll have more clarity in terms of what is the most likely scenario for this year and effectively integrate them a more forward-looking IFRS 9 effect. What I just would like to say, and with all the prudence and the modesty of the capacity to predict, but we went through our portfolio. And we looked at what kind of scenario in terms of cost of risk we can expect on the 2 scenarios I've just described. We have our best estimate, at least to date, is for the 6.8% GDP drop in Eurozone and base case scenario of 70 basis points cost of risk for the full year also, something not very dissimilar to what we had beginning of this year. And in this prolonged scenario of low activity, 100 basis points in the downgraded scenario. Same thing, we can elaborate on how we built these figures. And the last thing is we are providing also an indication for the capital. William will elaborate further on all the details, we'll answer your question. But what we try to factor is reasonable assumptions of all the factors which will impact the capital. Of course, the P&L but also the downgrading of counterparts, the drawings of existing facilities, the full impact of the regulatory TRIM effect, the 50 basis points. We factored some increase of operational risk also at year-end. So all this means that we think we are able to tell you that we will remain within the range of the 200 basis points and 250 basis points above this 9.05% MDA level, of course, depending on the functions that we are taking in terms of paying potentially at year-end an exceptional dividend. But that's something we will review, of course, based on the further evolution of the situation. So now I will turn the floor immediately to William, who will answer more into detail.

Good morning, everyone, and thank you for making yourself available on short notice. I will now turn to the Q1 results presentation, which is Slide 9 for an overview. Let me start by saying that, as Frédéric highlighted, we had actually a very good start of the year in across all businesses, I must say, January, February and for retail spending until the period of until mid-March and effectively deteriorated at the moment most countries went into confinements and the market dislocated. This is what you see reflected in this space very much. When you look at the retail operations, both French and International, you see, in fact, a small revenue decrease in French retail, which is fairly consistent, a bit worse than what we had indicated for the whole year, beginning of the year, i.e. 0 to minus 1%, so it's minus 1.2% year-on-year in Q1. Good commercial dynamics. We provided you with a lot of details in the back of this presentation. But you can see that the outstanding, both credit and deposits, were dynamic. That's on the franchises, particularly wealthy clients, Boursorama clients, brokerage volumes, unit-linked in life insurance that was positive. And that was combined with a decrease in cost by close to 4% down, minus 3.8% year-on-year, resulting in a return on normative equity of close to 11% at 10.7% underlying. International Retail Banking, there again, positive until mid-March with a slowing down in production, as you may imagine as well as the beginning of impact on the cost of risk. Revenues were up close to 3% adjusted for perimeter and exchange rate. You have to take into account that international retail, we incur obviously, the impact of the disposals for the headline numbers that we have completed in 2019. That's why there is such a difference between the headline number and the adjusted number. Here again, good commercial dynamics. We've provided numbers on the back of this presentation and well-contained cost, return on normative equity 13.2%. Insurance and financial services, revenues are slightly down. And we have a very resilient performance overall with a return on normative equity of 19.8% -- 19.6%, sorry. Obviously, the picture is much more negative on GBIS, particularly revolving on 2 issues. One on market, as Frédéric said, and I will come back to this. So let me -- let allow me not to comment too much on that page. But we had a very poor performance in equities, minus 99% revenues in Q1. But otherwise, we had a good performance in FIC, resilient revenues across the rest of the business. So the other element beyond the very poor performance in equities, which impacts revenues, is the strong increase in cost of risk that you can see in GBIS for the quarter overall. The return is negative for that pillar. Corporate Centre, we'll comment more in detail later. It's a bit more of the same as far as the fundamentals are concerned, i.e. the cost of funding of the company as well as the OpEx. But we have some volatile elements -- volatile P&L elements linked to interest rates and own credit, which I will comment. Overall, as you can see, good net income underlying is EUR 98 million. The headline number is minus EUR 326 million, which is significantly down from last year as we said. On the next page, I would like to point you to what we have in mind in terms of cost. First of all, I would like to reiterate that for this quarter, we are very much in line with what we had initiated last year and the years before in terms of executing various cost plans, which you know in CIB, in French retail, international trading, in financial services as well as central functions across the board, i.e. the trend of decreasing cost in absolute terms. This is why you see in Q1, 3.6% decrease in absolute -- in cost -- underlying cost from 1 year to another. And so the first thing we would like to say is that there is no issue as to the execution of the savings plan that we have announced to you in the previous years, including the CIB plan last year, and they are well on track. They will be executed, and they will be, as we expected, beyond any context of COVID as a decrease in absolute term of the cost base, starting with EUR 17.4 billion underlying cost base, as you know, in '19. On top of this decrease, and this is very important to have in mind, this is on top of this decrease, we add an additional set of savings of EUR 600 million to EUR 700 million. You have the type of measures that we have taken in the context of COVID: ban on travel and events; very strict consumption as far as consulting, IT services, external providers is concerned; hiring freeze, obviously, touching on variable compensation as you may imagine; and being much more selective on the Change the Bank expenses. To be clear on that point, we will not stop to change the bank. As you know, we have almost EUR 2 billion expense for IT development every year in cash burns, but we would be much more selective focusing on 2 areas, remediation from a regulatory standpoint as well as every project that helps us transforming the company and decreasing the cost base going forward. The rest of the project will be cash canceled or postponed. Cost of risk is the next page. As Frédéric said, we have increase in cost of risk, which is material this quarter. Obviously, we were 21 basis points in Q1 2019 and ended the year at a 29 -- 25 for the full year. This is three-fold more at 65 this quarter, and I will come back on the details, mostly revolving on GBIS, although we're starting to have provisioning up in French retail as well as the International Retail Banking and Financial Services on the back of IFRS 9 provisioning rules. As I will come back on the details on the next page, I will just point you to what is on the right side of the page. We still have -- and this is a good start for us and to -- a good position to start the prices. We have a good number for nonperforming loan, which is still low at 3.1% and comfortable gross coverage rate of 55%. As Frédéric said, we're guiding the market depending upon the scenario that for cost of risk, that would be materially higher than our previous expectations, as you would imagine, between 70 and 100 basis points for the year. Just to give you an order of magnitude, 100 basis points equals to EUR 5 billion to be compared with EUR 1.3 billion in 2019. So this is a hefty number. Next page, give you a bit more detail on how to think about what we had in terms of cost of risk for the quarter. So as I already commented, you have on the pie, the concentration on GBIS, representing 42% of the net cost of risk for the quarter. So as you would expect, as a European bank and a European bank dealing with large corporate, this is where you would find more of the impact of cost of risk to be compared with what we have seen with many U.S. banks, where maybe the increase in reserves was more revolving around retail. I think you should expect that it is more around the corporate world that you see cost of risk increase in -- as far as we are concerned. The split is as follows: on the EUR 80 million, EUR 20 million cost of risk that we had for the quarter. We have, first of all, roughly EUR 400 million of cost of risk, which we consider is a normal cost of risk, i.e., around 32 basis points. In that, we first -- what we have given you as an indication at the beginning of the year that we would expect normal condition, the cost of risk to be between 30% and 35%. So I think we are very much in the ballpark of the small increase we were expecting. We have for COVID-related case, i.e., the provisioning that you would expect under the IFRS 9 rules this time based on overlays mostly, roughly EUR 300 million, EUR 295 million, which is 24 basis points. So this is the very bulk of what's happening here. On top of it, as Frédéric said, we have some specific files including fruits with some counterparties for 9 basis points or EUR 120 million. The next page, I will not comment too much in details. I'm sure you will have questions. So we had the opportunity to discuss with some of you already on that when we provided you a few weeks ago some more details. There are many more details in the appendices and supplements, but we want to reiterate that we feel that we have a strong and diversified credit profile. You have here on the left-hand side the corporate portfolio is EUR 326 billion of the total EAD, approximately 1/3, and you can see that there is no high concentration of problematic sectors. Although, obviously, as you would expect, most of our provisioning is around some specific sectors, such as oil and gas, shipping, aircraft. We have little exposure to LBOs, less than EUR 5 billion. And as we already said that with Diony a few months ago, we have a very strict discipline in terms of structures in LBOs and underwriting commitment so that we can come back on that later. And on retail, we have a very well-diversified geographical exposure. And let me stress the fact that as far as mortgages are concerned or home loans are concerned in countries such as France, which is where we have the bulk of our mortgages, you have the high securities through insurance schemes. So this is a totally different environment than what you would find, for example, in the U.S. Turning on to capital, which is the next slide. So as Frédéric said, we have, for Q1, a ratio of 12.6% at the end of the quarter. Pro forma, we already announced sale of SG Finans, our leasing business in Norway. We end up at 12.7%, which is, I said, more than roughly 350 basis point buffer above MDA. Let me remind you that we have already -- always said that for us, what is comfortable is to be around 200 basis points above MDA, so we are definitely well above. You will find in the appendix the split in terms of RWA increase, which I'm sure is an important element for you. So RWA increased that quarter by EUR 10 billion, so EUR 345 billion to EUR 355 billion. And so what you have here is EUR 4.5 billion of increase in market RWA, and you have EUR 3.5 billion in credit RWA. So rest is linked to regulatory, including some TRIM and some securitization on the CRR2. What is important to remind, to put in context what Frédéric said, is the target by the end of the year to be running the company with a buffer in between 200 and 250 basis points, depending upon the dividend assumption that you take. First of all, as it has been said, we have in our forward-looking buffer totally accounted for the negative regulatory impact. So we don't have -- we still have what we have announced to you as for the TRIM impact and other regulatory impacts totally in the ratio. So we think we have probably the conservative approach on that. But let me say, I think it's a realistic approach to take that into consideration. On the other hand, we did not take as of yet any of the potential benefits stemming from the recent announcement by the ECB nor some flexibilities that have been allowed. For example, you don't have in this ratio 4 basis points we could have taken for the phasing of IFRS 5. This is a totally fully loaded ratio. We did not take as of yet in the computation of the 200 to 250 basis points buffer announced by Frédéric, any benefit from the potential deduction -- nondeduction of software, which could represent us almost 30 basis points. We did not take the benefit of the fast forwarding of SME discount factors that would be another 10 basis point for us nor the discount factor of infrastructure financing that would be an additional 2 basis points. And then going forward, IFRS 9 phasing would represent 6 basis points. So the 200 to 250, again, includes increasing RWAs linked to the context in the year, increase of RWA into TRIM, but none of these potential tailwinds. On the liquidity and funding, which is the next page, let me remind you that we have actually improved the liquidity profile of the company in the first quarter. The liquidity buffer is up to EUR 203 billion from EUR 190 billion at the end of 2019. The LCR ratio is up 144%. To give you an order of magnitude, this is a stress ratio, as you well know. This represents a EUR 58 billion buffer. NSFR is comfortably above 100%, and we remain A-rated company, as you know. But more importantly, I would say we've been able to access to funding, I wouldn't say easily because everyone is aware that the conditions overall were more difficult for the term, particularly in Q1 for all banks, but we could complete an additional issuance of senior nonpreferred. We have been able to access easily to short-term liquidity, and we had record deposit collections. We benefit from the fact that across the board, especially in emerging geographies, we are considered as a very high-quality counterparties for depositors. I won't comment the next page and turn now to the businesses. French retail, again, we will leave it mostly to the questions. We have detailed in the appendices. I've already said, revenue slightly down. Commissions are down but you should have in mind that the financial commissions are strongly up. There were a lot of volumes. We have retail clients, especially on the wealthy side on Boursorama willing to trade or willing to invest in unit-linked, for example. Net interest margin is up on the back of very strong volume of production up until, as I said, mid-March. Discipline on cost, as I said, in the 11% or close to 11% underlying return. I will leave it for questions if you want to go more into the detail. International Retail Banking and Financial Services, I've already mentioned. Revenues up year-on-year adjusted for perimeter and foreign exchange, plus 1.6%. Operating expenses are up 2.6%. But if you adjust it for some one-offs, including contribution to solidarity funds or bank funds or insurance funds across geographies, we have actually a 1.5% increase. So it's slight positive jaw and return of 15.4%. I will spend maybe a bit more time on GBIS, which obviously is the area of underperformance for the quarter. As you can see, our revenues are down 27%, headline number, adjusted for a base effect, which was a sixth revaluation in Q1 2019 as well as disposal in private banking in Belgium and the exit from commodities businesses, you have still revenue down 21%. The performance leads to overall reported group net income negative of EUR 537 million on headline terms, which obviously is not satisfactory, and we've also around 2 things, which I will comment more in detail. One is a global market, especially with a big dip on revenues on the equity side. And as far as financing advisory is concerned, very different revenue development, as for all the other businesses in GBIS, apart from equities and market activities, but yet a strong increase in cost of risk, as I've already commented. So I think what is worth, and in the room, we have both Séverin and Jean-François Grégoire, the Head of our Market. So if you have questions, we will be very happy to answer them, he's to spend some more time on the market. And to begin with, I would like to summarize how we saw the world from our perspective in the first quarter. Because clearly, there's a difference in the way banks performed in the first quarter, depending upon their business model and the type of exposures they have in markets. Yes, there were clearly positive trends in Q1, up until the end of the month, strong volumes in equity benefiting to our flow business. So our cash equity business was up. We listed products, including those we are integrating from Commerzbank, which proves to be, in that perspective, a good acquisition. We're up the prime services. As you know, we have refocused on equity prime, were also favored by the trend. But we may not be a very big player in the space, but we benefited from it as other banks. Good momentum in fixed income markets, particularly in rates and foreign exchange, and clearly less favorable in credit depending upon the category we're talking about. But overall, very good trend in fixed income. Overall, that was compensated by severe and somewhat unique elements. Equity market collapse is the first one, and the speed at which the collapse was done, which obviously is unfavorable for businesses, which for us, particularly on the structured product side, are price driven. Second, extreme volatility with VIX level never reached as well as increasing correlation, which obviously is very unfavorable for the hedging of our positions in structured products. And lastly, sharp decrease in dividend futures. You all are aware that announced dividends has been largely cut across industries, starting with banks, which obviously was not in our numbers. So when you look at what happened effectively in our numbers, which is the page next, you can see that we have strong increase in our FIC revenues. Despite headwinds on the credit side, we had very positive numbers. So excluding one-off activities, i.e., commodities and the capital trading, it is even higher than the 32% you find on this slide. This is an increase of 52%, very good performance, including compared to peers that were published to date with equity as almost 0 revenue after IFRS 9 and was severely impacted despite the strong performance in prime services with the products in cash and flow derivatives. We had to increase reserves -- market reserves. We had a negative impact of -- linked to the dividend cancellation of EUR 200 million, increased hedging costs, which led to a very poor number in exotic products and some default of counterparties especially with some hedge funds, as you have seen, all the banks are announcing it as well. So this is clearly the basis of a poor performance overall in that segment, we will come back to that on questions when you have questions, I'm sure. I'd like to comment finally on the Corporate Centre, which is the last slide on my side. So as I said, nothing much to mention on the usual step-ups normally in the quarter in Corporate Centres. The cost of funding has not increased and actually still is -- compares good -- well relative to budget. Same with operating expenses. We had already mentioned to you that we would have some transversal project, remediation project, but this is very much what we had announced, and that explains the increase year-on-year. So the main element to explain here is the volatile component, what we qualify as a volatile component on the net banking income of the Corporate Centre. And for that, I would just refer you to the fact that the Corporate Centre does primarily 2 things, which may translate into some volatility as we have seen in previous years, by the way, particularly in '14 and '15. But also in other years, they're one way or another. The first thing is, as you know, the Corporate Centre raised the long-term liquidity for the group and also hedges on behalf of the group, the equity participation from any structural rate risk. The hedges we have to take to be -- to transfer fixed into variable for capital instruments, which are EUR 81 billion or EUR 82 billion as well as the equity stakes, such as the equity stake in Commerzbank are accounted for -- account benefit from the hedge accounting methodology from an accounting standpoint, so that creates an asymmetry. So we have -- the account of the hedging mark-to-market, which the underlyings are discounted, that create a mismatch. That mismatch is a pure accounting asymmetry and will converge to 0 over time. So I think it's a pure accounting factor that has no economic implication. The second element of what the Corporate Centre does is being the counterparty of all of our businesses as far as liquidity is concerned. And particularly, there is one specificity with market activity, which are both providers and borrowers of liquidity, which is that the market activities live in a mark-to-market world with the Corporate Centre lives in a discounted world. And so when we raise liquidity through structured products, we have EUR 65 billion of liability in the form of structured products. When the variation on the spread is a Société Générale spread, we have some impact on the liability side, which are over time totally compensated by an equity -- in equities. So over time, that converge to 0 between P&L and equity. And on the asset side, we also have some variation, but it is neutral for the group. This is what you find here is compensated elsewhere. So overall, this is massively accounting asymmetry links to rates and spreads and is either compensated in equities or would be found as neutral for the group and mostly will converge to 0 over time.

Thank you very much, William, just a few words of conclusion. Again, here we are, as we know, facing an extraordinary crisis in an uncertain environment. What we've tried to do is give you perspective beyond the results where we think we can articulate figures with all the prudence that these figures are related to scenarios. And of course, as I said, with one big question, which is, of course, the answers of government and their efficiency. Beyond, let me just say, and I think it's, for me, a very positive that this year is the bank is up and running absolutely functionally for all its operations, clients, projects, et cetera. And we want to move forward beyond this year, we will start thinking, of course, for 2021, 2025. In a different environment, needless to say, this crisis will have long-term impact. We will take that into account to determine the road map drawing all the lessons learned in this crisis, and the crisis is also a way to make further progress. So that's where we -- what we wanted to do. Now let's enter Q&A. As always, with the same discipline, please, 2 questions per people. The floor is yours.

[Operator Instructions] Your first question comes from the line of Delphine Lee from JPMorgan.

So I -- my 2 questions. The first one would be on capital. Just trying to understand a little bit your guidance of around 11% to 11.5% CET1 by year-end. If we pro forma where you are right now of the TRIM impact, you're around 12.2%. So it looks like there's another at least 70 basis points, if you could just -- negative impact. If you could just give a bit of color of how that breaks down between the different components? The second question is on cost of risk with your guidance of 70 to 100 basis points. If you don't mind giving us a little bit of color in terms -- in particular, on your GDP assumption for 2021. And if you could provide as well the different wave streams between scenarios, base case, favorable and worst-case scenarios? And also if you have any sensitivities, I mean, just for us to understand. Because in terms of cost of risk to GDP or unemployment, because when I look at your 100 basis points, it's only 30 basis points higher than your sort of base case despite GDP declining at double the base case. So just if you could give us some color to understand provisions, that would be great?

Yes, Delphine. William will again come back to the different elements that we take into account into the capital calculation and guidance. And Diony Lebot will answer your question on the cost of risk. What you see now, we have not given any guidance for 2021, first. Diony will explain to you that we, in both scenarios, consider that the cost of risk would go down slightly, but would remain relatively high because even, again, if you have more GDP, there's again, perhaps a time lag between -- with the default with the government schemes. And of course, with all the uncertainties I've mentioned of additional further actions that we don't know yet and which might be implemented. So that's why the exercise of 2021 is very complex. And again, Diony will elaborate on the different scenarios. William?

Delphine, obviously, we can't give you all the different bits and pieces, but what we can confirm is, first of all, as I said, you have in this ratio still the expectation of at least 50 basis points of regulatory impact. So we operate, as Frédéric said, as a normal company. We continue our projects, we continue to deal with clients, and we continue to deal with remediation, including remediation of risk models, and we expect the advance at least conservatively, we consider it would be an important factor. There may be some variation around this number, but that's the number we have given to you. That's first of thing. Second, we obviously consider that we should have an increase in RWAs for the year. So as you see, there is an increase in the market RWA for the quarter. We allow ourselves, having learned our lesson to -- for potentially some additional increase, maybe this is not what you would necessarily derive from what we observe today in terms of stressed VaR and VaR computation. But at least, we think it's good to be on the conservative side. As far as credit RWAs are concerned, it is clear that, as you can see already starting in Q1, particularly in the financing and advisory side, there's been a number of drawings by counterparties, which explains mostly the increase in RWA. This effectively, as I'm sure you will ask Philippe Aymerich to comment, a strong production on the corporate side and professional side in France but mostly through guaranteed schemes, which means probably little RWAs. You have to -- you should not match the volumes of outstanding with the volumes of RWA. So we have some of it as well in international retail. And lastly, there will be, of course, RWA. We have computed that RWA increase stemming from downgrade of counterparties and MEDEF integration, which obviously, we take into account.

So as you can see, there are many different assumptions. The new production, as William said, the evolution of the risk-weighted assets before the crisis, or the TRIM. And based on that, we can give this range. And of course, then there is the 50 basis point dividend provision, so depending on the assumption we make on the scenario also of paying or not in what amount of the dividend. So that's why we give this range. Now Diony, can you elaborate on the methodology? How we think? Because I think it's important to spend time on this.

Yes. Thank you for your question. So indeed, there are still a lot of uncertainties, a lot of questions in designing reliable scenarios. First, of course, taking into account the length of the lockdown periods in the various countries and regions and also the time it will take to go back to normal. Also taking into account government and policy actions, which vary from one country to the other. Noting that in France, measures were very quick to implement and a lot of them are already in place. So in designing our scenarios, and by the way, we have not applied them for Q1, these are scenarios we designed to give you a range of on the guidance of 70 to 100 basis points. But as William said for Q1, we based our cost of risk calculations mostly on overlays and really a review of sector-by-sector, country-by-country of the initial impact of the crisis. And of course, the downgrade, we have already proactively implemented. So the base, what we call base scenario, and -- so this is what's today since the initial scenario is an average lockdown of 8 weeks and the 12 weeks to return to normal. This would lead to a recession of minus 6.8% in Eurozone, minus 6.6% in the U.S. to give you just 2 data points. And then we would see an almost V or U-shape type of scenario where in '21, which was your question, in the Eurozone area, we would be at plus 6.6% and for the U.S., plus 6%. So this is the base -- what we call base scenario. And the prolonged scenario, which is more stressed, realized on average lockdown of 12 weeks and 18 weeks to return to normal, and this could also include stop-and-go type of scenarios. In this case, the GDP drop in Eurozone would be close to 13%. And we would have in '21, a plus 10.5%. Again, although this is almost a V-shape type of growth, as you can see, in both our scenarios, we are not back to precrisis GDP levels in absolute terms. So in the projection of our cost of risk, this means that under IFRS 9 and probability of default increases, rating migrations, we would have also to take into account the impact of policy measures and massive relief measures already implemented. And this will result in the delay of defaults. But we still believe that defaults will continue to materialize later in 2020 and even in '21, which means that cost of risk, yes, will be much higher in 2020. This is our 70 to 100 basis points scenario. And as William said, this is EUR 3.5 billion to EUR 5 billion. It's really a significant increase to our precrisis level over the guidance we had begun or, let's say, indicated in a normal context. And cost of risk will continue to be high, lower than that, but high in '21 because we believe that more defaults will arrive as the moratoria will come to an end, they mature. But also because part of the government support and the public support actually results in transforming losses to debt. So at a certain point, the most impacted sectors will see more defaults. So in summary, the 2 scenarios are very much dependent on assumptions, which we will have more clarity on more in Q2, which is the average lockdown period and the anticipated time to return to normal. More impact, of course, moving forward, but also taking into account the government support. And also, as far as we are concerned, taking into account the quality of our portfolio. And as Frédéric said, we have done significant work over the past years to reduce the most riskier parts of the portfolio, drawing the lessons of the past. So we gave you a few numbers and exposure details in our slides. LBO, for instance, is less than EUR 5 billion. We have also in the last year anticipating a downturn, taken much more protections to the -- more risk transfer transactions will be applied to an OTD policy, which makes us confident that our portfolio is well diversified, is sound and can resist the shock compared to the previous crisis, still doubling or even tripling at the cost of risk, which is the 70 to 100 basis points guidance.

So just to elaborate on the methodology to come to that 70 and 100 basis points, I will review all of the sectors, the portfolios and...

Yes. So it's a combination of applying the scenarios and the migration of ratings and the increase of probability of defaults associated to these type of scenarios related to GDP. And also taking into account a review sector-by-sector, portfolio-by-portfolio, country-by-country and adding overlays where we think it is necessary because of the nature of this crisis and more sensitive sectors or applying some shock absorbers given, as I mentioned, the magnitude of government support is provided, which will delay some of the consequences, not in 2020, but also towards '21. So we take an approach, which is forward-looking and taking, of course, a very granular review of our exposures, portfolios and applying all the necessary as quickly in terms of potential add-ons, we call this overlays or shock absorbers and understanding the impact of the crisis towards the 2021.

Your next question comes from the line of Tarik El Mejjad from Bank of America.

Just a couple of questions, please. The first one on capital and dividend. And please, Frédéric, don't tell me it's too early to talk about dividends. But -- so the -- I mean, I wanted to have your sentiment about the ECB actions other than dividends? Because I mean clearly, it's harming the sector when it trades -- it's trading as deep discount, it will more encourage to sector to shrink rather than grow, which was the first incentive from ECB. So what's your thinking on that? And have you discussed that with ECB? And do you see really the action as something temporary or we feel that's arriving in the 1st of October, they actually extend dividend for another quarter or 2? Because if we fast forward, I think in October, this is where registration will be tough, where we start to get unemployment rising much higher. I mean sustainable unemployment, not just furloughing people. So -- and then more specifics to you for the dividend. I mean, thank you, William, for the -- taking us through the moving parts of the capital. But I mean, should we take now MDA as your new guidance in terms of buffer and where you want to stand? And that I assume you're using all the pillar to our kind of forbearance and so on? And the second question is on costs. I mean I understand the EUR 600 million to EUR 700 million are driven by stopping projects that are not harming your overall kind of improvement of efficiency and so on. But how do you think this is really realistic and feasible in the current challenging societal environments where -- I mean, even at Bank of America, we have a firing crew, right? So how is that really -- and my question absolutely goes beyond this because that only offsets the fraction of the cost of risk you will have this year. And underlying, because when we all discuss, you said, I mean, cost of savings underlying will not stop in 2020 and will go beyond. So how can you -- do you think you can still implement some of this essential cost savings to help your underlying cost service?

Yes. Tarik, I must say, it's very difficult to answer your very good question on the dividend. I think, first of all, the supervisors feel that they made the right decision given the uncertainty to ask us to not to pay a dividend and protect as much as possible the capital ratios entering into the crisis. Where shall we stand in October is, of course, a big question mark. And by definition, to be frank, today, I don't know and the kind of perspective and how default will have materialized perspective for 2021. So I think they will look at this. They will look at how capital ratio P&L have evolved and they will make a new decision. It is clear that they know, it was not a easy decision for them, and they know the impact on the market. And so they will take that into account. But of course, with this priority of securing the system. Beyond -- I think, yes, the MDA, the new -- and I think the new reference and if I may, and here we go into even more long-term perspective, which is still unclear. At the end of the day, once you have implemented arbitrary, you have absorbed Basel. Thus, supervisor will be very comfortable with their models. And effectively, they had in mind to then adjust their requirements on the ratios. What they did in the crisis is anticipated by basically 8 months, that's something which was already in the regulation and the P2R in position and also with the P2G. So what they did here is an anticipation of something which was decided, which was part of the framework. So we feel, yes, it's the right reference, knowing of course, this ratio will be applied to higher risk-weighted assets, but based on the review of the modeling. The second point is on the cost. And here, I would like to tell you, yes, we are absolutely clear we can do that because it's overall, I would say, discretionary expenses and expenses under our control. And when we talked about IT project, it's mainly external providers. Regarding the restructuring elements, as William told you, we are pursuing our existing projects, the further restructuring of the network. We finished the plan -- actually finished the plan for 2020. What we've said to our trade unions is we will not launch before September any new projects. And I think, as you mentioned yourself, in U.S., in Bank of America, if I may say, there's a kind of decent approach, responsible approach towards that. But I must say, when I think about 2021, 2025, we'll have to cover that, taking into account the adjustment of the behaviors of the clients. Everybody speaks about the fact that digital penetration will have increased, et cetera. So we'll factor it. And of course, we will see how we have to adjust our setup, but that's more for 2021, 2025, and it's not related to what we've given you, the figures that we will effectively deliver, including in this more complex social environment.

If I may add on capital, just make sure that the modeling on your side is right, and hello, Tarik, for entering for the question. As said, we feel that guiding on the buffer of FDA, so long as the MDA reference is a permanent one, which it is, because the fast-forwarding of CRD5 has been adopted. And the fact that we have a core Tier 1 instrument that allow us to benefit from it has been adopted. The decrease in countercyclical buffers where applicable has been adopted. So this is a permanent. So we consider that the buffer above MDA is the right thing to sell the company capital. Why? Because it gives ample comfort to the investors, the equity investors as for the potential dividend or hybrid holders that there should be no worry at all as to the expectation should be being able, of course, pending the ECB debate to pay the one and of course, the coupons are not a stake. Let me also remind you as fast group as hybrid are concerned that we don't have any call on the hybrid instrument before April 2021. I want also to reiterate what I said, which is that it is a totally fully loaded ratio. You will see some banks taking some benefit on IFRS 9 phasing. We have not done it. So it's totally fully loaded. Again, that would be 4 basis points for us. And we don't, at this stage, envisage -- we don't have in our forecast, what I've said, which is the result of the announcement of the ECB of April 2028, i.e., on software the semi-discount factor, discount factor on the infrastructure and also, as I said, the IFRS 9 traditional arrangements. That is quite important because this is when we have clarity, numbers you may add to the picture.

Your next question comes from the line of Jon Peace from Crédit Suisse.

I wondered given the exceptional circumstances, you could talk a little bit about how has trading been in April? I think some of your peers has said the fixed income environment is still quite robust. Is equities back to a kind of normal run rate? And have you seen any losses from the EMC acquisition? And then my second question was just on dividend. I mean, if we're in a lucky enough position to pay a dividend at the end of the year, how do you think about sizing it? Do you go to the 50% payout policy? Or do you pay out down to taking the CET1 ratio down to the top end of your new 200 to 250 basis point buffer? Just how should we think about that?

Jon, I will let Séverin answer your first question. Your second question, it's, again, very premature because, as I said, we need to have more clarity on the environment for 2021 as we will be able to anticipate the way the supervisor will think about it. So at this stage, we will provision the 50% payout ratio in our computation, but we need to wait. So the money is there, but what we will do and what we will be able to do to be frank, really, I think I cannot say at this stage. Séverin, can we comment on the trends knowing that we do not comment.

Usually, as you know, Jon, we do not comment, but [indiscernible]. But I can say anyhow as the market transition, as you saw, as progressively, stabilized in April after this dislocation we had to manage in March. But we have to say that uncertainties remain very high, and there is now some attention from clients on the investment solutions side. We see still some flows and government activities, but there is some additional accounted them today. So we have to be very prudent. On the EMC, remember, we had taken over and migrated last year part of the EMC franchise, the ETF, which is now within Lyxor and a part which is a small one of structure products, which is not really significant. So there is no specific loss on the EMC acquisition. The big part of the EMC has been migrated at the end of March, which is a listed part of the activity of EMC. So we have no P&L, if I may say, no revenue on this part in the first quarter. And this part has been positively, if I may say, impacted by the price, but we have not benefited from that in the first quarter.

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

I have 2 questions on my side. The first one, so we haven't touched upon any outlook for revenues yet. And of course, the current market conditions are likely to impact transaction fees, for example, but also loan growth on the household side. So how are you thinking about the outlook for revenues? That is my first question. And then secondly, so the government measures, as you said, the French government has been very proactive. But yes, it doesn't -- or on the guarantee side, it doesn't take the full risk. There is still something vested to the banks. And what we are seeing other countries is that when the bank has skin in the game, it usually takes longer for the credit to flow to the economy. So I was wondering if you can give us some color here as well on how well do you think this is functioning from a practical perspective?

Yes. Giulia, I will leave the floor to Philippe Aymerich for the French retail, and Philippe Heim with the caveat that he cannot actually comment so much on the subsidiary. So he will probably provide you with a very brief answer, actually. And knowing that we think it's impossible to provide guidances in this environment. Because overall, I would say the behaviors even of the clients remain relatively unknown or difficult to predict. Whether you will see further mortgage very soon is a question mark on the saving attitude. So that's why difficult to provide the guidance, but perhaps, we can give you some qualitative assessment of what we see today and how, again, the scheme functions. Philippe Aymerich to start with France.

Yes. Thanks for the questions. Regarding the activity, of course, it has slowed down quite significantly since mid-March with the individuals, I would say, except on financial products. For example, we have seen in Boursorama, a very good level of activity, actually, including an increase of the opening of financial accounts. So that's a positive sign. But apart from that, that's true that with seem to be [ doing ] significant slowdown on the activity, even for, we are still doing mortgages, which were initiated before the lockdown. On corporate and individuals, on the contrary, it's quite active, especially with this state-guaranteed facility. So yes, I can share with you some numbers. So at this stage, at the level of -- as a group, we have received approximately 57,000 request from clients for an amount of EUR 14 billion. We have already approved 46,000 files. And one amount of more than EUR 7 billion of -- and of course, we are making sure that these files are processed quickly in order, as you say, to go to the clients and to support the economy. So we are very careful with that. But again, we have created a specific task force to make sure that we are able to deal with this important volume of operations. That's what I can share with you. As you know, this facility is guaranteed depending on the kind of companies, but up to 90% by the state. At this stage, our refusal rate is approximately between 3% and 4%. So it's low, but still we are careful. In some cases, companies were in trouble before the crisis. And this facility is not supposed to support them. I mean we have other mechanisms to support these kind of companies. So I think that -- again, this product didn't exist 5 weeks ago. It was created quickly, implemented quickly and I think that we are careful. And for the time being, we have find the right balance between strong support to the economy and still being careful regarding our risk profile.

Philippe?

Yes. So maybe now on the international part of our network. We operate in many countries. So maybe I will focus on the most important, let's say, set up. The pattern is that we see across the board, let's say, measures taken to support retail finance and corporate clients with this idea of granting moratorium depending between the countries, 3 to 6 up to 9 months, for example, in Romania, both for retail client [ on the game ] and corporate clients, with percentages state guarantees that can again vary a lot. It's up to 90% in France. In, for example, in Czech Republic, it's 80%. In Russia, the state support comes from the fact that the state is financing minimal wages. We have a similar setup also in Germany, as you know, what is called the additional credit. We have something also similar in Morocco. What I need to highlight is that those setup has been put in place, let's say, end of March, beginning of April. So it's a little bit early to give you, let's say, a consolidation of all those exposures and all the consequences of that.

Your next question comes from the line of Jean Neuez from Goldman Sachs.

I have a question on the investment bank, and I have a question on the cost of risk guidance for this year. I just wanted to understand in the 70 to 100 bps cost of risk guidance, what type is -- essentially what still you could give us as to what you think the cost of risk would have been in the event where the corporate guarantees by the government would not have been in place? And essentially how certain or how confident of the way that they work and how they can shield your P&L? I think to an earlier question, there was the -- that the sensitivity between the 2 different GDP assumptions on the cost of risk was not all that high, at least at first glance. And I just wanted to understand whether those corporate guarantee scheme were efficient or how you model them in that assumption? And my second question was on the investment bank. And on the quarters in equities with regards to the mark-to-market and to an extent, whether you think that even without the rebound in activity, whether you think that some of the factors which have led to the hedging losses are reverting and whether you could book them back or whether they have just gone. And also in terms of this mark-to-market differences, I remember in 2008, for example, you had, I think, if I remember well, like a EUR 2 billion gain in CDS when credit grew out. And I just wanted to understand whether in the revenues of today, there is any gains on those CDSs also that might revert if you still have some hedging of your corporate loan book?

Yes. I will -- Jean-Francois, I will let the floor to Séverin to comment on your revenue question. This is very difficult to model what would have been the cost of risk without this intervention because it's easy to say there will be potentially, thanks to the support, no cost of risk. But how much would have been, it's more complex. I just would like to highlight qualitatively 2 things. The governments basically across the world are more or less taking the airline companies in practice. They are nationalizing, putting a lot of equity or all loans, which might become, at some point, equities that fit it. So without this intervention, of course, there would have been an impact of companies probably going bankrupt. And they would -- they have a much higher cost of risk and probably something will be close to 0. And on the other spectrum, if we talk about professionals in France, there is, at this stage, as we've said, guaranteed loans, representing up to 3 months of turnover, subsidies by the government to a solidarity funds. You have the subsidies to finance people who temporarily are not working. And you will have even relief of tax -- of taxes and have social and income taxes and -- net income taxes. So you see that, of course, the governments are doing a lot because they try to save the companies, the jobs and we are the channel for liquidity through these schemes, where we keep a risk, which is the way for the government to ensure that there is a reasonable decision behind the loan. But of course, with the support of the government, and it's true across the world. It's true in Europe, it's true in the U.S., in many geographies. So very difficult to give you a figure, I may say, but you have examples where you can see that, of course, it will mitigate the cost of risk. And as I've said, now there is a big question mark, which is today, it's the urgent things which are being dealt with. But of course, the governments are also thinking about how to help the economy to rebound as quickly as possible, and we don't know that yet. Séverin Cabannes on the revenues.

Yes. Okay. So on -- Jean, I will tell you the answer. So you have this quarter on the equity side, specifically, as mentioned by Frédéric and William, different impacts. We mentioned clearly the dividend consolidation. We mentioned clearly a default on one counterpart, which is [ EUR 155 million ]. And there is another point to have in mind. If you look at the swing in terms of on the market reserve between the first quarter last year and the first quarter this year, there is a swing around, for the global market not only for equity, of around EUR 300 million. Last year, we have release of reserve. This quarter, we have a new reserve constitution of EUR 200 million -- around EUR 200 million for the global market and near EUR 175 million for the equity, as mentioned earlier. This part, which has not been released though in the revenue, and there is a swing with -- compared to larger [indiscernible] due to the market conditions, will be over time visible in the channel. But it's fair to say that it's part of the revenue, which are lost also. And the hedging costs, we had to suffer or to pay during this quarter, which explains the gap between what I said, delta in reserve, dividend cost and counterparty loss that is still to expand the revenue of equity the gap. This gap is the hedging cost with this -- which is cost. But I think the answer to Jean-Francois first question. Regarding the CDS. It's very important to bear in mind that in 2008, and you made a reference to this period of time, we had a big significant CDS book. And we have completely changed the exposure. Today, we are using very limited part of CDS as an insurance, as a coverage, as a hedge on our credit risk in the portfolio of [indiscernible]. And so it's fair to say that we have some impact and positive impact in that case on the revenue of the first quarter, but it's very limited. It's not significant.

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

I have a question on the state-guaranteed loans. Basically, I'd like to better understand the math and how it works on the P&L for the different years. So year 1, year 2 and year 3 because I think I understand that you carry the cost of the guarantee in year 1. And also to understand how RWA moved? So do you actually -- if you have a loan, let's say, 100, do you only take 10 of RWA? And what is actually the profitability of these loans in year 1, 2 and potentially in year 3, just to better understand how this compares in France with other markets? And then the second question I have is, I have to say I do not really understand the explanation of the fair value adjustment in the Corporate Centre. If I'm not wrong, I think you say some is linked to capital markets activity. And then I wonder why this is not booked in the capital market business? And then you also said that some will reverse over time. So how much is going to reverse over time? And when do you think this will reverse?

Yes, Lorraine. I will let William answering the second question. But Philippe Aymerich to answer into really the details of the scheme and how it works.

Hello. So regarding revenues, for year 1, and that's true for the other years, but for the year 1, and at least, we know the cost of liquidity, which is at this stage, a floor at 0%. So for year 1, the clients pay 0%, plus the cost of the guarantee. The cost of the guarantee, which is went like the guarantee up to 90% and for the small companies, the cost of the guarantee is 25 bps. So for the part covered by the state, the 25 bps will go to the state, which then counts. And for the part which is not guaranteed by the state and which is supported by the banks, the 25 bps will go to the bank. And yes, you're right, this -- the 25 bps are paid upfront to the state by the banks. And we will recover this money to a client at the end of the first year. Regarding the following years that's the same principle, except that, of course, we do not know yet the cost of liquidity. So the mechanism will be cost of liquidity plus the cost of the other guarantee. And the cost of the guarantee changes according to the duration of the loan and the size of the company. It goes up to 200 basis points for large companies for a credit of 5 years. So that's basically how it works on revenue side. Again, basically, cost of liquidity which is, of course, unknown within a year and the cost of accounts.

And first on the risk-weighted assets, it will stay like it is, 90% will not be weighted, except at the beginning for the initial 2 months.

Yes.

So the first 2 months are not guaranteed. There is a kind of threshold, by definition, as the loans are there with 3 months, there will be very few defaults in practice. So you will then have the full benefit and it will stay alongside. If a client decided to keep for 3, 4, 5 years, but well, you will have the same thing, 0 for the package guaranteed by the government, and we will wait the 10%. So in itself, I don't think it will change very much the picture. As we expect, some clients will say, "Okay, I want to amortize." And there will be an amortization also -- sorry, it will be amortized. It's not yet -- for the full [ future ], it's not amortized, but then it will be amortized. And what is important also is you have, again, specific trials, large price can be tailor-made sometimes like we saw recently. So it's more or less the same principle, but you could have some small adjustments on large files. William?

Lorraine, thanks for your question. And I'm sorry if I was not clear enough. So as I said, the bulk of the P&L impact on the Corporate Centre stems from accounting asymmetries. A large portion is due to rates, and that this mismatch between the hedges on capital or quasi-capital instruments accounted for, mark-to-market versus the underlying, which is accounted for, which is discounted. That's -- is stayed in the Corporate Centre. And that mismatch, obviously, currently can move if rates go down further. If we reduce -- if rates go up, but this is fundamentally something that vanishes over time when instrument matures. There is another accounting asymmetry, which stays within the corporate center, which is, again, quite significant in the number, which is due to the liability side, i.e., what I said, the stock of the structured notes. Normally, obviously, this is compensated by -- the P&L impact is compensated in equity through OCI. But there is some vibration for methodological reason and also because the duration of the underlying instruments is modeled as opposed to contractual. And as you know, given the structure of the parameters and change in the duration can vary. So there is a small vibration, which is why I say normally to be 100% compensated by equity except for some time difference in the adjustment of the hedges. And then, yes, there is a portion, which I would say in the number I've given you is a smaller portion of, which is due to the cost of funding and the spread paid by our market activities principally, not just market activities, but as they are mostly the one mark-to-market vis-à-vis the Corporate Centre. And so that's why I said it's neutral for the group. Some of it, by the way, has probably been or certainly been considered -- accounted for as reserves based on direct P&L. So that may come back over time. So that's why I mentioned that notion. Let me just finish on saying that I don't think we are unique. We've looked closely at other disclosures. And I think -- as I said, we had this type of volatility in the past. We have seen this type of volatility with others because the spread of every bank have gone up massively in the course of March and then until they started to come down. The rates were down in many jurisdictions, so you can have this type of volatility. And I've seen that some good books, the last liquid portion I mentioned, directing the market activities. So we'll book it directly in the Corporate Centre. So it's a convention.

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

Just firstly, sorry if I missed this, but within the capital guidance for this year, how much in the way of earnings do you include exactly in the 200 to 250 buffer? Because I guess looking at it quite simply in the last 3 quarters of last year, you had something like EUR 5 billion of pre-provision profit, which will be going down a lot this year, presumably even with the cost savings. So I guess, if you then factor in your cost of risk target of EUR 3.5 billion to EUR 5 billion, it's unlikely you would have much this year. Then just a question for Jenny, I guess. What is the current stock of outstanding loans that's under payment moratorium or holiday across the group, please? So not the guaranteed loans, just the payment holidays. And can you confirm that you're not taking provisions or state migration on those outstandings as per the EDA guidelines? And similarly, what is the oil price implied in your cost of risk assumptions?

Sorry, Omar, can you repeat it? Your last question, what is?

Oil...

Oil price. Okay. Okay.

The oil price.

Okay. Briefly, a nice price for the P&L, but we are not discussing what we've said with the intent what we feel are reasonable assumption given the context. Second, I think that we can give you the figure for France. Again, we cannot give you any detail for subsidiaries outside France, as I said, because they will release their own figures in a few days. For France, we can give the figure in terms of loan. Yes, that's the figure, which is mentioned on Page 26 at this stage because, of course, the number could change. We have deferred the EUR 1.8 billion of payments, so basically deferred for 6 months. So it's a relatively small amount, actually. We don't provision, I think, on this risk loans. And on the oil price?

When you look at our oil and gas exposure, which we have given to you, and we have included a slide, you have a very small part, which is directly sensitive to the oil price. They -- the composition of this exposure is integrated companies, which are almost all of them are investment grade, so 22%. So you have LNG exposure, which has almost no sensitivity to gas or oil price because it's relying on the take-or-pay contracts. So you can see the various breakdown. And then you come to the part, which is more exposed to oil price, which is the upstream independent. That's 20% of the total exposure, roughly EUR 4 billion. And within that, we consider that the most risky and exposed to oil price part is our reserve base finance activity in the U.S. and this is $1.7 billion. So we have stressed this portfolio. And it can support the stress of going down to $15 to $20. And also taking into account that for this year, they have hedges in place. So in terms of cost of risk, the impact would be limited in terms of the defaults related to variation of the oil price. Of course, we will see rating migrations associated if the oil price remains at a low level for a very long time. And we are going to see more in the stage 1, stage 2 and of course, the ratings for the company. But overall, we are quite confident with the quality of our portfolio and the sensitivity on oil price in terms of cost of risk. Of course, we already took some overlay provisions on the sector over the years. And including this quarter, we have added to this overlay provisions on the oil and gas sector.

Our next question comes from the line of Guillaume Tiberghien from Exane.

Yes. The question relates to capital again. So you highlight that you'll be at about 200, 250 bps above MDA at year-end and that you're comfortable with that level. The question relates to what level would you see as uncomfortable? So what pain threshold would you be accepting to take if you forecast of 11% to 11.5% equity Tier 1 doesn't materialize this year? So would you accept to fall to 10.5%? Or you think 11% is the minimum for you?

Guillaume, listen, I mean, we -- as we said, we've taken reasonable assumptions across the board, and we think we have the flexibility to adjust parameters if, for any reason, we were not in line with that. So I think we keep really this range in mind. And I think that 11% for me would be the right target for the end of the year in such an extreme environment, and we are talking here about an extreme environment with, of course, elements that we will have to carry. And we've mentioned the downgrading, we've mentioned the drawings on facilities, we've taken assumption of all this, which are, in our view, are reasonable. So yes. I think, for us, the 11% is probably what we target as the minimum threshold.

Okay. I've got a second question on the oil again. In your EUR 3.5 billion to EUR 5 billion of overall provision, how much of that is on the oil and gas exposure? Because if I ignore the little pie, which is, by the way, very useful about the breakdown of the oil and gas exposure. But if I just consider the fact that about 1/3 of your EUR 20 billion of exposure is noninvestment grade, so about EUR 7 billion. What if I wanted to say, I'm going to lose half of that, of that noninvestment grade. So that would be already on its own about EUR 3.5 billion. So I know it's extreme this scenario that I just described. But I've just wanted to understand out of the EUR 3.5 billion to EUR 5 billion, how much of that is for oil and gas?

We don't disclose with this level of granularity, Guillaume. I think that Diony has tried to explain to you that actually, when we look more in detail, there are many of this exposure, which is, for example, backed by contracts, which are not impacted by the oil price or related to gas price and with security there. Second, as we said, for example, even in the reserve-based lending portfolio, it's very important to understand that there are hedges with the producer for at least this kind of environment of 12 months. So to a certain extent, it might be a bit more in 2021 that we might see the default, depending on how long the prices will have. So we don't disclose that granular information, but I think you are absolutely extreme in thinking alongside, we'll see what kind of more information we can provide you. But it's absolutely an extreme figure, and we think very differently. And as we said, we review the portfolio pretty substantially. Knowing that we saw the prices in 2016, remember, the 2016 crisis also on the oil price, and we did not suffer that much with also very low price on oil in the exposure.

Okay. And maybe final question, if I may. Still on the cost of risk, you said that in '21, it will remain elevated, but less than in '20. So are you trying to guide us towards EUR 3 billion being a realistic number?

Listen, Guillaume, we don't give any figures on purpose because I think it would be too risky. But let's say, in our own simulation, what we can say is it will be still relatively high, but that -- we stay there and we'll see when we can say more, but it will take some time, I think. Let's be realistic. A lot is still to happen on the development of the crisis and the government schemes.

Your next question comes from the line of Jean-Pierre Lambert from KW -- KBW.

I would like to come back on the impact of guarantees on the IFRS 9 macro scenarios. How is this -- how are the guarantees incorporate? Is it because you assume a less fall of GDP? You changed the loss given defaults? Or you assume a lower PD migration? Or is it combination of all or you focus on 1 parameter? And then the second question is on the 100 basis points cost of risk guidance, if you want, what is the breakdown between corporates and the rest of the business? Or if you can give some indication of allocation?

Jean-Pierre, I'll leave the floor to Diony on your first question and maybe the second one.

So it's our own piece. Actually, the substitution and taking into account the guarantee. And for the part, which not guaranteed, we apply our normal PHD on this part [indiscernible] as we refer the government.

And your second question, first, we can't comment again on the international retail, unfortunately. What I would like to say France strategy is clearly support companies in practice to save as much as jobs as possible -- many jobs as possible and then compensate for the people who might not work. So it's absolutely fair to say, it's mainly driven by more corporate than individual clients. And we are very different, for example, from the U.S. where people are immediately without any revenues. And we see a very different pattern. And of course, in wholesale, it's more corporate by definition. So I would say it's more geared towards corporate, given the portfolio of activities we have. And we will comment a little bit more in the sixth of May when we will disclose figures on the international retail.

Our next question comes from the line of Azzurra Guelfi from Citi.

2 quick questions. One is on the provision. If you can give us there's a split between what is the macro assumption and what is the specific? And if there is any timing between, like, Q1 and over the rest of the year? The second one is on your cost additional savings guidance. Is it fair to assume that some of these would revert once the economy restarts and the business come back to normality?

I will leave William answering on the cost number but the bulk is savings which will, of course, happen this year. We will, I hope, travel again. I think that we might, at some point, hire again, et cetera. I mean, yes, so I would not clarify these as the necessary structural savings. We will also draw the lessons of how to function in this new world. Now we have learned many lessons with remote trend. So we'll think about perhaps new categories of savings that we might not have had in mind months ago, but the bulk, I would say, will be more one-off. And then we will prepare the 2021, 2025 strategic plan. Diony, on this macro and versus specific, I think it's a little bit of both, I would say, actually, in the way we looked at it.

Your question is on Q1, I think, Azzurra, right?

Q1 versus -- in Q1, in the scenario and yes, we will see probably more IFRS 9 effect in the second quarter as we've crystallized the scenario.

Yes. Exactly. In Q1, actually, we have -- and William gave you the breakdown in the EUR 820 million. We have part of it, which is related to the crisis, and it's EUR 295 million. It's a combination of overlay rating migration. We have also this fraud-related one-offs of EUR 127 million. And the normal, say, stage 3 defaults. And moving forward, indeed, as I explained earlier, we are going to have a combination and a higher impact of the scenario, stage 1 and stage 2, and the reports will materialize later in the year. According to our analysis and the moratoria in place, we believe there will be some time lag effect.

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

Just 2 questions left on my side. The first one, relates to your African operations. I was wondering if you can give a little bit of color on how you think it will be affected by the current crisis. I imagine it could be quite different from your conclusions for, let's say, core Europe either much worse or much better. And the second, more technical question. You had at the end of 2019, still about EUR 160 million of CET1 deductions for expected loss shortfalls. Have you basically used all of those now in Q1? Or are there still some left to offset against further stage 1 and 2 provisions, please?

Stefan, I will ask Philippe Heim answering on Africa.

Yes, Stefan. So a quick word on Africa. So it deserves indeed some highlights. So the situation are at least the following. So surprisingly, this planetary crisis is not so important that is we may have shared a few weeks ago. And is a payment that means to that we have to be humble and wait for some further development. As you know, there is no consignment imposed in certain countries because we are talking of countries of daily subsistence. So this is more a system of late curfew at night. So in practice, end of March, measures have been taken by governments. So to restrict, let's say, the movement of people at night. Our system, our setup is up and running. And we maintain completely opened our branches. You may have seen that in spite of those elements, and once again, we have been impacted by the COVID only in the last week of March. You think that the outstanding books went up by 6%. And as we speak, and this is in the figures developed by Diony, the cost of risk for Africa in Q1 stands at 102 basis points. So it's where we are for Africa.

Thank you. Second question, William?

Stefan, I guess you referred to the expected loss minus provision on funds item. Yes, it's back in the ratio for the equivalent of 5 basis points.

The next question comes from the line of Matthew Clark from Mediobanca.

A couple of follow-up questions for me, please. Firstly, could you just clarify, within your 200 to 250 basis points MDA buffer guidance, are you assuming the 70 basis points or the 100 basis points positive risk to get to that 200 to 250 range? Second question is regarding the fraud charges. Could you just confirm whether you fully provisioned for those exposures or whether there's a risk of further losses related to those 2 specific cases in coming quarters? And then final question for me is just about client appetite for auto calls. Presumably, your -- either your or the private banks you distribute them, clients have taken a bust on these products year-to-date. So does this create a problem for the earnings power of your equities business going forward if clients no longer want to buy these? Just some comment there on appetite for these structured products going forward, please?

Yes. Matthew, so Séverin will answer about the appetite of clients on structured product. Briefly, this -- if you will, this assumption of [ 11% to 11.5% ] is based on the base case. But if you make your calculation, we would remain in this range even with a EUR 5 billion charge risk. But of course, we have more constraints on the dividend that we could pay at year-end. So this is really a...

You can make a very simple calculation, EUR 5 billion, less EUR 3.5 billion is EUR 1.5 billion. If you tax it, it's roughly EUR 1 billion net income that goes, divided by 2. So x3, it's 15 basis points that goes up.

Understood.

You have this appetite on the product. I think you asked this first question, Matthew, sorry, I forgot after...

Fraud charge.

Fraud charges. But I think it's a reasonable level of provision. We'll see how it develops. We might have to add something, but it's on this [ 2% to 5% ]. It's always difficult, but I think it's already a really good level. We'll see how it develops going forward and still a little bit uncertain. Appetite for product, Jean-François Grégoire, Head of Capital Markets.

Jean-François Grégoire. So it's a bit early to say, but we can give a hint. Actually, the main product is the Autocall, I think that is an equity product, it is obviously sensitive to the level of equity. So customers has had some underperformance from those products over this period. But precisely, these products have a protection mechanism that so far are working quite well. And if these products are kept until maturity, if we don't go much below the lowest bond that we have seen, these products will not lose anything. So this protection mechanism, it's precisely in this environment that it will show its strength. Actually, it's precisely the protection that we offer to the customer that is challenging for us to replicate. And we do it at the first order, then we are some vibration, obviously, in our trading books. So it's probable that the appetite for these products will not diminish, but only time will tell. We'll see in a few quarters.

And sorry, could you just clarify what kind of haircut or level of law is embedded in these products? So if I -- I don't know, a Euro Stoxx 50 Autocall for me, where is the protection level set below which I end up owning the underlying?

Usually, it's minus 40% or minus 60% from where it has been triggered initially. So even the products that have been sold just 2 months ago, a few months ago or just at the height of the market are not triggered. So they are still protected.

Your next question comes from the line of Anke Reingen from Royal Bank of Canada.

I just have 2 follow-up questions. Firstly, on your cost of risk guidance. I mean you said that incorporates the government guarantees. But is it correct to assume it also includes the EBA guidance on taking a more forward-looking view? And I know it's early days, but do you think there's a coordinated -- or would you say there's a coordinated approach if it's consistent across the different banks? Or is it very much everyone does his or her own thing at this stage? And then on the dividends, I see that -- I understand there's a lot of moving parts. But this is my understanding that when you pay a dividend, it can incorporate the benefit that could might be deferred and the capital benefit that will be coming out of the new EC proposal yet this week.

Yes. Anke, Jenny will answer on the EBA. To be frank, regarding the bank choices at this stage, I think each bank which try to do its own homework and looking at its own exposure. So I think we'll have that more convergence that is on the scenarios going forward and why we see more. And on your question on the dividend, I would say a dividend payment is I think never 1 year or just exercise, and then we will, of course, have to factor what is expected in 2021 where we stand as we say effectively in practice. In terms of buffer, we've not factored the additional benefit, and we'll check what it means, that it could be a relatively substantial. So I'm sorry, but it's too early to answer that kind of question and I cannot answer at this stage. And also, of course, how the supervisors will look at it. So unfortunately, we will have to wait for probably a turn to have more clarity on all this. And regarding the...

Yes. We think we took into account the guidance of the EBA to distribute taking a forward-looking approach. Also, that applying forbearance of default to COVID moratorium and later as long as they are extended to clients due to the COVID prices. And to your question on coordinated approach, no, there is -- I think at this stage, applying the guidelines to the regulators and I think a lot of banks will refine scenarios in Q2, as we said, because we will have more clarity on the extent of the crisis, the lockdown and all the measures in place.

Okay. Is there any more question?

We have no further questions. I'll now pass the conference back to our presenters for closing remarks.

Okay. Well, thank you very much for attending the call, and have a nice day. Thank you.

Ladies and gentlemen, thank you all for your participation. You may now disconnect.