Good day, thank you for standing by, and welcome to the Crédit Agricole Q2 and half-year results 2021 conference call. At this time, all participants are in a listen-only mode. After the speakers' presentation, there will be a question and answer session. To ask a question during the session, you will need to press star one on your telephone. Please be advised that today's conference is being recorded. If you require any further assistance, please press star zero. I would now like to hand the conference over to your speaker today, Jérôme Grivet, CFO. Please go ahead, sir.
Good afternoon, everyone. Happy to share this hour and a half with you for the presentation of our Q2 and H1 results, which, as you have seen, are very good results. I should say that these results are good. It's not a surprise because, as Philippe Brassac explained it several times yesterday in front of the board, in the context of this very specific crisis, where public authorities have put in place several measures in order to protect the economy, it's not a surprise that banks have good results.
Nevertheless, I think that our results are specifically good with a net profit, and you can see it on page four for the group, which is not only up around 85%-90% as compared to last year, but also significantly up as compared to 2019 for the same period, both for the quarter and for the first half of the year. It's the case for the stated figures. As far as the underlying net income is concerned, the improvement is also very significant, plus one-third on the quarter, plus close to 45% on the first half of the year, but also a significant improvement as compared to the underlying figures in 2019, + 28% for the quarter and + 21% for the half of the year. On the following page, you will see the figures for Crédit Agricole S.A.
For the quarter, close to EUR 2 billion in terms of stated net income and close to EUR 3 billion for the half of the year. Restated from some specific items that I will comment later on, we still have EUR 1.6 billion of net profit for the quarter and EUR 2.5 billion for the half year. It's up around 45% in both cases as compared to last year, and also up 25%-30% as compared to 2019. This very good performance has been reached by a very, I would say, virtuous combination of a high growth in terms of revenues, +12% for the quarter and +10% for half year, which is much steeper than the increase in the cost base, +8% and +4%.
The gross operating income is sharply up, around 20%, and then the cost of risk is very significantly down, -70% for the quarter and -60% for the first half of the year. The underlying cost-income ratio for the quarter stands around 55%, which is a further and significant improvement as compared to Q2 last year. The solvency is at a high level, 12.6% for CASA end of June. On the following page, you have the main messages that we want to insist on in this quarter, but I think I can summarize that in saying that the economy is recovering very rapidly after the different phases of lockdowns that we have had.
In this context, the level of activity that we have had in all business lines has been very booming, very buoyant. This is leading to this very sharp improvement in our profitability, both through an increase in the top line and a significant decrease in the cost of risk line, thus preserving and improving and strengthening further the capital position of the group and of Crédit Agricole S.A. If I go now to page eight, you will see an illustration of what I was just saying regarding the economic situation in France, where you can see that seen through our own prism, which is the number of payments that we process, we have seen, indeed, in the different lockdown periods, a significant decrease in the level of activity.
The two comments I can make is that every time after these lockdowns, we have seen a sharp rebound, and then lockdown after lockdown, the decrease is smaller each time, proving that the economy is progressively integrating in its behavior the consequences of the different lockdowns. This is leading to the indicators that you have on the bottom part of the page, which illustrates the level of confidence of both the households and the business leaders, as well as the PMI. On page nine, some elements illustrating the fact that this very, I would say, fast rebound of the economy is translating into a very good level of activity for our different businesses. This is illustrated by the customer capture, which was significantly up in H1 2021 as compared to H1 2020.
Indeed, we are now close to the number of new customers that we've had attracted back in 2019. It's also illustrated by the volume of new loans that we've put in place in Q2 2021, up 15% on Q2 2019. It's also the case for some additional product lines, like the volume of new consumer loan that we put in place, up 62% in Q2 2021 as compared to Q2 2020. In the insurance area, it's the example of the number of new policies that we've underwritten in P&C activities, up also 62%, close to 63% between Q2 2020 and Q2 2021. This is leading to this very sharp increase in the top line. You can see on page 10 that this increase is not only the case as compared to Q2 2020, but also as compared to Q2 2019.
Indeed, both for the quarter and for the semester, the increase is a double-digit increase or close to a double-digit increase, even if we exclude the scope effect. Interestingly also, this increase is very well spread across all business lines, with the exception of the large customers division, which you remember, had a very buoyant quarter in Q2 2020 and is now at a more normalized, I would say, level of activity, but nevertheless, a good level of activity.
What I wanted to explain on the following page is that actually this good performance in terms of revenue generation that we've posted in Q2 and H1 2021, both compared to 2020 and to 2019, is actually part of a much wider achievement, which is the fact that in the last five years, quarter after quarter, both for Q1, for Q2, for Q3 and Q4, we will see later on this year, but both for Q1 and Q2, we've managed to improve the top line year after year after year. This is the result of a very balanced breakdown of our revenues at CASA level. First, between fees, net interest margin, insurance revenues, and other revenues.
This very good resilience is the consequence of a revenue generation which is made by three quarters of revenues in connection with stocks of contracts and stocks of products that we have in our balance sheet. Only a quarter of the revenues are linked to transactions that we have regularly to rebuild year after year. A very good resilience and a very good revenue generation capacity illustrated on this page 11. When it comes to the evolution of the cost base, indeed, when we first look at the evolution of the cost base between Q2 2020 and Q2 2021 at 8.3%, it may seem a little bit too dynamic, even though it's still well below the evolution of the top line.
If we take into account 3 elements, which are the base effect, Q2 2020 was a very low base in terms of costs, considering the fact that Q2 2020 was earmarked by a lockdown, the most severe lockdown for the biggest part of it. Second element, the scope effect, and third element, an effect linked to the increase this quarter of the provisions that we book for future variable compensations. Actually, the evolution is very moderate. Actually, if we compare Q2 2021 to Q2 2019, excluding scope effect, so in two years' time, the increase in the cost base is only 3.6%, which is very moderate. If you assess those figures for the full quarter, you will see evolution which are in the same region, around 4%-4.5% if you compare to 2020 or to 2019.
All in all, we continue to have a very strict monitoring of the cost base in this period of time. This is illustrated indeed by the following page, which is page 13, where you see two interesting elements. The first one is the fact that in the last five years, in Q2, we've managed to post a positive jaws between the evolution of the top line and the evolution of the cost line. Indeed, in five years' time, we've managed to generate a total of close to 10 percentage points of positive jaws effect. This is leading to what you see on the right-hand side of this page, which is a decrease by close to 5 percentage points in five years of the cost to income ratio of the second quarter since 2017.
Indeed, if we had looked back a little bit earlier in time, the decrease would even have been sharper. Let's go now on page 14 to the topic of the asset quality. Nothing much to say indeed, because the asset quality remains very, very good across the board for Crédit Agricole S.A. and for the group globally. You can see that the NPL ratios are stable as compared to March 2021, and you may recall that in March 2021, they were decreasing slightly as compared to end of last year. Indeed, we continue to have a very good quality of our credit portfolio. Considering the provisioning that we've did this quarter, this is indeed leading to a slight increase in the coverage ratios that we post.
It's close to 75% at the level of Crédit Agricole S.A., and close and even a little bit above 85% at the level of the group globally, considering the very high 102% within the regional banks. All in all, we have more than EUR 10 billion of reserves in CASA's balance sheet and EUR 20 billion all together for the group. In which, at CASA, we have around 27% of this amount, which is facing sound loans, so performing loans. At the level of the group globally, the proportion of provisions covering performing loans is even above one-third. So a very prudent provisioning approach. Despite this prudent provisioning approach, what you will see on page 15 is that the cost of risk this quarter declined significantly. At CASA, it's - 72% Q2-on-Q2 and -34% Q2-on-Q1.
This decrease is explained both by a very sharp decrease in the evolution of S1 and S2 cost of risk, but no write backs on S1 and S2 provisioning, simply a slight increase, and also a sharp decrease in the S3 provisioning because of the evolution of the quality of the loan book which, again, was very good. At the level of the group, you will find the same kind of evolutions, a decrease of 63% of the total cost of risk Q2-on-Q2, and -17% Q2-on-Q1 with more or less the same trends. A very sharp decrease in S3 provisioning because there is a fewer number of defaulting loans, and an effort of S1 and S2 provisioning, which is maintained at a rather high level, especially within the regional banks of Crédit Agricole.
It is not linked to a change or a significant change in the macroeconomic scenario, but much more to the continuous effort of sectorial provisioning within the regional banks. Let's go now to page 16, where you see the global evolution of the P&L of CASA. What you can see in a nutshell on this page is that both on the quarter and on the semester, the evolution is very positive. It's +45% around in terms of underlying profit as compared to Q2 2020 and H1 2020, and it is still +30% and +25% if you take the 2019 reference. What is interesting is to analyze the way this evolution is made. If we take the example of the quarter, you can see that the net profit is increasing by around EUR 500 million as compared to last year.
There is an increase in the gross operating income of around EUR 450 million. There is a decrease in the cost of risk of around EUR 650 million. Then you have, of course, a very sharp increase in the level of taxes, + EUR 400 million, an increase in the other items, amongst which you have the minority interest, + EUR 200 million negative impact. This explains how this improvement of EUR 500 million has been obtained. What is interesting really to keep in mind, it's a balanced combination of the improvement of the gross operating income and the decrease in the cost of risk. Lastly, on this page. You can see that this improvement is also very well-balanced across the different business lines.
On page 17, again, a look back in the past at the appreciation of our return on tangible equity across the last four and a half years, so 18 quarters. We've managed to post a return on tangible equity across those 18 quarters, which was permanently at least two and a half percentage points above the average of the 10 biggest European banks posting a return on tangible equity. Again, this quarter and this semester, at 13 and 13.6%, the return on tangible equity that we publish is very significantly above the average of this sample of European banks. Last point on this page, as we've said, in February, once the ECB's authorization will be granted to us, we will conduct the second share buyback operation that we had announced beginning of this year. The first one being well on its way.
It's completed at around three quarters for the time being. On following page, I don't have the number here, 18, some highlights regarding our contribution to the transitions of the society. We have decided to join the Net-Zero Banking Alliance, the Net Zero Asset Managers initiative, also we will join, later on this year, the Net-Zero Insurance Alliance in order to fully comply to our commitment to be best in class in terms of accompanying the transition of the economy globally. At the same time, we also want to deploy our effort in terms of social inclusivity, and we will have dedicated initiatives regarding the integration of young people amongst the labor world in France going forward.
Lastly, on this page, we will continue to integrate, as much as possible, climate change opportunities in our business models in the different businesses of the group, retail banking, asset management, insurance, CIB, and consumer finance. On page 19, an illustration on which we regularly insist, but I think it's very important to, again, explain it's the growth engines that we have for the future. As you know, our model of universal banking is playing on three different growth engines. The first one is the scope of the services and businesses that we are able to develop, and we regularly extend the scope of our offers to our clients. The second element is the equipment. Once we have a client and considering the scope of activities that we are able to propose to those clients, we try as much as possible to develop their equipment rates.
The difference between the gray and the blue bar illustrates the improvements that we've managed to obtain in this regard in the last period of time. The third axis is the improvement and the increase of the customer base itself by the acceleration of our customer capture, be it the customers that directly sit within CASA's balance sheet, i.e., within LCL or within Crédit Agricole Italia, or also, of course, within the regional banks of Crédit Agricole. You have here some indication regarding the improvement of the different equipment rates in 2021. Let me go now on page 21 precisely, in order to give you some highlights regarding the different business lines of the group, different business divisions, starting with the asset gathering and insurance business division.
On page 21, maybe just one key element, which is the further development of the assets under management within this business division, EUR 2.3 trillion of assets under management. Indeed, there's been an improvement both in asset management, life insurance, and wealth management activities. The P&L of this business division continues to improve significantly as compared to last year. On page 22, insurance activities. It's been a very good quarter in terms of commercial activity for the insurance business division, both in savings and retirement, where we have net inflows of EUR 1.6 billion.
What is interesting is that unit- linked product represent 40% of the gross inflows. Around 100% of the net inflows, because what you can see is that actually we've been a slight negative flow, a net outflow for EUR products, and they now represent more than 25% of the outstandings, which is, as you know, a key target that we have for this business division. In P&C, the premium income is up 10% Q2 on Q2, so it's a very sharp improvement. Generally, we have between 7% and 8% increase over time. 10% shows a very sharp acceleration. In protection businesses, personal insurance revenues are up even more sharply, + 23%. It's mainly driven by the lending business, both for home loan and for consumer credit loans.
In this context, the net income group share of the business division is up and reaches a record level above EUR 400 million for the quarter, and EUR 700 million for the half year. It's including positive market effects, but it's also the consequence of this very good level of commercial activity. On the following page 23, Amundi. Amundi has already published its results, so you perfectly know all the figures, but I think that it can be summarized with a few ideas. The first one is that this quarter was earmarked by strong inflows in medium to long-term assets. There was some outflows in money market funds, but in medium and long-term assets, we have had significant inflows. The second highlight is that Amundi is reaching a new high in terms of assets under management, close to EUR 1.8 trillion.
Revenues are very sharply up, both management fees and performance fees are very significantly up. Actually, Amundi has reached a record in terms of performance fees this quarter. The gross operating income is very significantly up, despite the fact that the provisions for variable compensations were, by definition, up in this context. Amundi is posting also a record level of underlying profit. If we go now to the large customers division, so CACIB plus CACEIS. Globally, there's been a good level of activity, probably less buoyant than the one we have had in Q2 2020, but nevertheless, a good level of activity, and which was definitely above the Q2s that we've had in the past in 2017, 2018, and 2019.
In terms of custody activities, revenues were slightly impacted by a very ample liquidity position, which is costly, despite the fact that CACEIS is billing the biggest part of the cash it receives from its customers. There is, nevertheless, the remaining part that is supporting the negative interest rate conditions that we have now. When it comes to the fee part of the revenues of CACEIS, it is up significantly in line with the evolution of assets under custody and assets under administrations, which are up around 12% this quarter. Let me go now to CACEIS on the follow to CACEIS, excuse me, on page 25. It's a mixed set of figures in terms of revenues, as you know, because again, Q2 2020 was so high in terms of activity in the world of both syndicated loans and bond issuances, that it was difficult to replicate the same performance.
Actually, what I have seen is that most banks are in the same situation this quarter. Nevertheless, globally, the activities of CACEIS were good, with good performances in structured finance, in trade finance, in transaction banking, and also in investment banking activities and equities. In this context, the gross operating income is down, but considering the fact that the cost of risk was very benign this quarter, and actually, CACEIS is even posting a write back on loan loss provisions by EUR 40 million. The net income of CACEIS is up 15%, above EUR 450 million for the quarter, and above EUR 700 million for the half year.
In the specialized financial services division, both in consumer finance and also in leasing and factoring activities, the quarter was earmarked by a very high level of production of new contracts and new loans and new leasing put in place this quarter. This is leading to an improvement and an increase in the outstandings. An increase also in the top line, a very good cost control, and a decrease in the cost of risk. All in all, again, an improvement of the contribution of those activities to the net profit of Crédit Agricole S.A. I'll finish with the retail banking activity, starting with LCL, where, again, I'm sorry to repeat the same words permanently, we've seen a very high level of activity with a good level of production of new loans, leading to an increase in loans outstandings. A good level also of customer assets.
A good level of customer acquisition, with close to 90,000 new customers attracted in Q2, and an increase in the equipment rate of these customers with the different additional products that LCL is selling to them. LCL also launched its new project, LCL New Generation Network, which is going to lead to the regrouping of 250 branches in a network of 1,600 branches in order to better meet the customer requests in terms of advice and quality. In this context, the top line is up significantly, + 8%. The cost line continues to be very well monitored, plus only 2.2%, so the gross operating income is up 21%. Cost of risk declines significantly, -63%, and so the net income group share is up 78%. Italy. In Italy, of course, there is the integration of Creval this quarter, and I will come back on this issue later on.
Let's start with Crédit Agricole Italia, the original perimeter of Crédit Agricole Italia, because it's a quarter on which, on this perimeter, the activity was very intense, with a strong increase in the production of new loans, a very strong increase in the management of customer savings, and a very good development of the level of revenues, which are up 12% if I exclude the Creval scope effect. At the same time, expenses, again, excluding Creval, are decreasing a little bit, -1.5%, and the cost of risk is down close to 60% on the permanent perimeter of Crédit Agricole Italia. In addition to that, Creval is adding up some revenues, some costs, and all in all, a net contribution of EUR 7 million for the last two months of the quarter because the acquisition of Creval was completed beginning of May.
The integration of Creval is not only leading to this contribution of EUR 7 million of net profit for the quarter. It's also, of course, the integration of Creval's balance sheet in Crédit Agricole Italia balance sheet with the first consolidation. This is leading to an increase of our RWAs. I'll come back on this point later on, but it's an increase of EUR 8.5 billion of additional RWAs. There's also the recognition of a net bad will, which is provisory, but which represents, for 100% of Crédit Agricole Italia, EUR 378 million. When translated to CASA, considering the minority interest within Crédit Agricole Italia, this is EUR 285 million of net bad will. The difference between the gross bad will and a first appreciation of the provisions that we will book in face of this amount of bad will.
Let me note also that this bad will is not, for the time being, prudentially recognized. It's going to be the case only end of this year, when we will have the definitive PPA performed. Let me take the opportunity of this page just to comment globally the Group's presence and performance in Italy. We've managed to post a net profit in Italy globally of EUR 385 million this semester. It's very well-balanced between the four main business divisions of the Group. 35% for the retail banking activities. Also 36% for the specialized financial services business division, 19% for the asset gathering business division. 10% for the large customers division. Again, as we've already mentioned, a comprehensive set of activities generating a high level of profit in Italy. International retail banking activities excluding Italy.
You know that we've had several difficult quarters after the start of the breakout of the pandemic because the political answer in those countries was, amongst other elements, to decrease significantly the interest rates, which is putting a very high level of pressure on our top line. At the same time, we had, of course, to book some additional provisions. The situation is now stabilizing, and actually, revenues this quarter are up close to 5%, and cost of risk is declining significantly. This business division is getting back to a more normal level of profitability, close to EUR 40 million this quarter. It's back to a more normal level of return, around 15% of return on tangible equity. It's much more normal for this business division.
Corporate Center, nothing much to say, despite the fact that for the stable components of the Corporate Center, we continue to have this progressive improvement. We have some relatively volatile elements. One negative one, which is the fact that the net tax profit is reducing for technical reasons this quarter. At the same time, the private equity business, which sits within this business division, is posting better revenues than in Q2 2020. These elements are compensating each other. Let me go now to the regional banks of Crédit Agricole on page 33. You will see more or less the same trends as the one we have noted regarding LCL, with a very good level of activity, very good level also of customer capture, with close to 650,000 new customers attracted in the first half of the year. Loans outstandings and customer assets are up quite significantly.
Revenues are rebounding +4%. Costs are very significantly up. You may notice that, but it's really in connection with the fact that last year, the regional banks posted a very sharp write-back on the variable compensation reserve because of the dividend of CASA that was not paid in Q2 2020. If you compare the cost base of the regional banks in Q2 2021 to the one they've posted in Q2 2019, actually, the evolution is very moderate, around 1%. The contribution of the regional banks to the net profit of the group, EUR 740 million, it's up 18% as compared to last year. Let me go now to the solvency on page 35. The CET1 ratio at CASA to start with, 12.6%. All in all, it's a decrease of 10 basis points as compared to end of March.
Actually, you can see on the right-hand side that the total RWA increased by around EUR 9 billion, out of which 8.5 are explained by the consolidation of Creval. It's really this one-off effect and nothing more. On the right-hand side of the page, you can see that this explains also, and more than explains actually, the slight decrease of 10 basis points of the CET1 ratio, because the retained earnings represent around 20 basis points of additional solvency. This EUR 8.5 billion of RWA consume around 30 basis points of solvency. The difference is -10 basis points on the ratio. The other ratios continue to be at very good levels. Of course, the 12.6% CET1 ratio at CASA is far above any requirement and far above the 11% target. We've already commented that.
On the following page 36, the CET1 ratio of the group globally stands at 17.3%. It's stable as compared to end of March. It's 840 basis points above SREP requirement. Again, the evolution of the RWAs is almost completely explained by the integration of the EUR 8.5 billion of RWAs coming from Creval. Again, this also explains the biggest part of the consumption of the retained earnings that we have this quarter at the level of the group. In terms of liquidity, nothing much to say. We continue to have very ample liquidity reserves, above EUR 460 billion. Maybe just one point I can mention on this page, which is that our total outstandings, TLTRO outstandings, have increased a little bit by EUR 10 billion this quarter.
It's the result of, first, an additional drawing of EUR 5 billion that we did on our own, I would say, plus also EUR 5 billion that we found in the balance sheet of Creval. Market funding program is well on its way. 72% of the program of Crédit Agricole S.A. has been completed as of now. At group level, it's close to EUR 18 billion that we have raised end of June since the beginning of the year. It's really not an issue as usual. I think I can really stop here and try to answer now to your questions.
Thank you. As a reminder, to ask questions, you will need to press star one on your telephone. To withdraw your question, press the hash key. Please stand by while we compile the Q&A roster. The first question comes from the line of Jacques-Henri Gaulard from Kepler Cheuvreux. Please ask your question.
Good afternoon, Jérôme. Two quick questions. The first one, when I look at your large corporate investment banking, it is really good when you compare to the objectives you had in 2019. If I annualize your cost base, which was supposed to be EUR 2.8, you're there. Your revenue objective, if you annualize them, again, you're at EUR 5.3, you were at EUR 5. I was wondering to which extent even the targets you have, and then you have the level of provision, which is at virtually zero, whereby you're not actually completely too conservative on your CIB forecast for 2022. Which leads me to another question, which is, I'm surprised you haven't actually confirmed the EUR 5 billion net income. I was wondering what drove you to be cautious. The second question is also linked to CIB.
If we could have from the ground, maybe, a little bit of color about how the ESG transitioning is going on with your clients, with the whole Fit for 55. Do you find them well-organized? Do you find them a bit in a panic? Any color you can actually give would be helpful. Thank you very much.
Thank you, Jacques-Henri. Well, normally, we don't have to confirm quarter after quarter all our targets. I think it worked the other way around. If I had some doubt, or if I felt necessary to infirm one of our targets, I'd definitely warn the market from that. For the time being, as I regularly said it, we think that we are on track to reaching our different targets for 2022. For the time being, we don't feel the necessity of updating these targets, but we will certainly have to do it at a certain point in time. Regarding specifically CIB, it's more or less the same. It's true that we continue to have the proof that the model that we've put in place on CIB, since now probably around 10 years, is working.
It's a model that has been challenged initially because it's not, as I would say, market-friendly as some models more invested into equities, for example. We deem that this model, which is really focusing on financing our customers with market instruments and also balance sheet instruments, is really coherent with the DNA of the group. It produces a very stable level of profitability across time, which is very important for us. It's generating a reasonable level of profitability, which is again, coherent with our global model. Of course, once we will review globally our targets, which may happen, I don't know exactly when, but somewhere before end 2022, obviously, considering the fact that the present targets are aiming at 2022, we will give some additional color on all the businesses, including the CIB.
Regarding your questions on ESG and CIB clients, I think that Philippe Brassac stated to the press a few hours ago, actually, we intend to have a dedicated event regarding our ESG commitments probably in the fall this year. Of course, this would be a good opportunity to give some color on the way our customers behave regarding ESG challenges. What I can tell you is that this ESG question, and especially the climate change issue, is taken more and more seriously by all our clients. It's no longer a feature that you add up on top of a business model that is designed outside this preoccupation. It's now completely embedded by us in our business model, but also by our clients. We have also now around 8,000 business clients that have been granted a transition grade, a transition note by our internal tool.
It's really going to be a more and more important feature. We'll have this press event somewhere in the fall regarding ESG issues, and we'll be able at that time to give colors on many aspects of this issue in all our businesses.
Thanks a lot, Jérôme.
Thank you.
Thank you. The next question comes from the line of Jean Neuez from Goldman Sachs. Please ask your question. Your line is now open.
Hi, Jérôme. Good afternoon.
Hello.
I wanted to ask you first on the achievements so far in your business plan to follow up on Jacques-Henri's question. You had a plan to be between 25% and 30% of revenues in asset gathering, which is the most capital light of the businesses you have, I guess. Now it looks like it's going to be probably a bit over that. That will mechanically change your ROE in a sense, and your capital generation higher in terms of the mix, compared to risk-weighted asset consumption. I just wanted to understand, whether you think you have more growth opportunity than you had at the time, in which case you'll be able to reemploy that, or whether, in a sense, not to compress your ROE going forward, you're going to have to rethink the capital distribution slightly.
Leading me to my second sub-question, which is, I have to say, I'm ashamed. I had understood that the share buyback would have compensated the entire share issuance from the scrip, but now, rereading the past communication, I realize as is written today, that it's also including the switch too. In a sense, why not expanding this and given all the capital you have, to the entire scrip dilution if you want. Lastly, I just wanted to know, because today you produced some good results in Italy with Creval integration and so on, whether you could give us a sense as to, in 2022, what you think pre-provision profit, guidance could be for that business.
Many questions, François.
Apologies for that.
No, no problem. No problem. I'll try to make it short in the answers. Regarding the first question, i.e., the breakdown of our activities between the different businesses. We are not going to constrain the development of one business line simply because it's a little bit ahead of the curve in terms of development. We are happy that the business in which we are probably ahead of the curve is a capital light business, and we are happy with that, and we are not going to try and force a rebalancing, which wouldn't make sense if it does not correspond to business opportunities, real business opportunities, I would say. We have always given indicative breakdowns of our sources of revenues in the future, and we think that this indicative breakdown makes sense.
If the asset gathering business division is ahead of the curve, no problem for us, of course. Again, I don't want to take the opportunity only of a second quarter result publication to reshuffle any kind of objectives. We'll have, at a certain point in time, a dedicated moment to give some longer term prospect, for the whole business of the group. It's the name of the game. We have, for the time being, targets for 2022. We are on track, which is a positive, and we will someday give, again, longer term prospects, taking into account, of course, everything that has happened in the course of the present medium-term plan. What I can say is that what has happened indeed is satisfactory, at least in our viewpoint. It's the case in terms of business development.
It's also the case, of course, in terms of profitability, as I have explained, taking a certain long-term horizon since beginning of 2017 in terms of profitability. Regarding the share buyback, I don't know what you understood at the time, where we announced the mechanism that we had put in place for the payment of the 2020 dividend. What is true is that we are doing exactly what we had announced. We had announced that we wanted to eliminate first all the extra shares created, on the basis of the take-up of the minority shareholders. This is what we are doing now, and as I said, I think we have completed 77% of this program, and it's going to be ended probably end of August, beginning of September. I don't know exactly when, depending on market conditions, but this will be done.
Then we will add to that, an extra share buyback operation that is going to take place in Q4 as soon as we have, of course, the authorization of the ECB. This corresponds Of course, when we presented this program back in February this year, we've made in our models some assumptions regarding the share price, which by definition are not exactly met, but nevertheless, globally, the operation is exactly as we presented it. This operation, the second share buyback, was aiming both at generating in combination with the switch dismantling, as we've said, and we are committed to fully dismantle the switch mechanism before end 2022, and we'll do it. In combination with the switch dismantling, this was aiming at improving, enhancing the earnings per share at the end of the day. All things included, all operations included.
It's going to be the case at the end of the day, once the second operation will be completed. With pro forma, I would say, the dismantling of the switch mechanism, it's pro forma because we cannot tell in advance when we will trigger the dismantling. This will enhance globally, more or less, the earnings per share by around 1%. The second goal, of course, was to rebuild the tangible book value per share, which is also going to be the case by end 2021 as compared to end 2020. Why don't we conduct an additional or a wider, larger share buyback?
I have been asked about share buybacks since probably five years, and I have always been very reluctant about share buybacks for 1 simple reason, which is that the majority shareholder, the SAS Rue La Boétie, in my best understanding of its intentions, of course I cannot talk on its account, but it is my understanding that they do not participate in the share buyback. If I repeat regularly share buybacks operation, there will be a permanent increase in the stake owned by the SAS Rue La Boétie. Which is not, at the end of the day, probably the best thing for the minority shareholders. I will conduct the second share buyback operation before year-end, and then I will stick to my normal dividend policy, which is to pay a cash dividend.
As I said also earlier, I want to be precise on that, we consider that we haven't fully paid the 2019 dividend up to now. We consider that end of 2021, end of 2022, when we will set the amount of the dividend that we intend to propose to the general assembly meeting to pay, we will round up the numbers in order to progressively repay this unpaid 2019 dividend. This is the way I'm going to answer to your request of somehow boosting, I would say, the dividend yield, if I understand you and your concern correctly.
Thank you.
For the end of June, we've booked exactly the normal dividend policy for half of the attributable results. It's already EUR 0.39 a share that is booked, reserved in our books. End of year, we will see exactly what we can do in order to probably round up the numbers, in order to boost the dividend that is going to be paid in 2022. If it's not enough to fully absorb this unpaid 2019 dividend, we will repeat the operation end of 2022. It's not a commitment, it's an intention. You know that generally, of course, we stick to our commitments, but we try also to stick to our intentions.
Okay. Quickly on Creval.
On Creval, excuse me, your question was regarding the NPLs, right?
No, I just wanted to think what you think is achievable run rate of pre-provision profit in the quarters to come?
Excuse me. It's way too early to tell because actually, we are the owners of Creval since the beginning of May, but actually we are in the driver's seat since only mid-June. You know that the general assembly meeting that changed the management took place only on June 18th. We are now conducting different due diligence operations within Creval in order to fully assess the potential of the integration of Creval. We are going to give more color on that, in September, October, November. I don't know exactly when, but I'm very confident that this is going to be a very accretive operation. I'm very confident that we've been conservative when booking this provisory allocation of the purchase price end of June. I'm very confident that this operation will prove to be a very positive operation for the group.
Okay. Thank you very much. Very clear on that.
Thank you. The next question comes from the line of Azzurra Guelfi from Citi. Please ask your question. Your line is now open.
Hello, Azzurra.
Hi. Hello, good afternoon. A couple of questions for me. One is on French retail, which has shown a very nice rebound, especially on the revenue. When I look at the NII, what are the drivers that you expect for the second part of the year? Could this be a level of sequential growth from now, thanks to the momentum in lending and maybe a little bit better margin? If you can give us some color on that. The second one is looking at the slide where you show the market share gain. This has been increasing since the beginning of your plan. What are the area where you see more momentum for growth in the next couple of years, and what are the one that contribute more to the profitability improvement? If I may, a quick one on the provision.
You have stage one and stage two provision. Some of it might be used, if there is a deterioration of economic condition. If you will have still some left, which probably is the base case, what are the rule to release them and what are your expectation for the release of this? Thank you.
Thank you, Azzurra. French Retail is a competitive market. It's a difficult market. It's a competitive market. As an example of that, you are aware that, in Q2 this year, this is probably the lowest level that we've seen historically for the average price, not only with us, but on the market for new home loans. Clearly it's a very competitive market. It's absolutely key on this market to be able to have a very intense global relationship with your customers. This is where actually we think that we have an edge as compared to some weaker or less complete, less comprehensive, competitors. In terms of prospects, what is for sure is that, for the NII.
The NII is, at the end of the day, it's the difference between the yield that you get on your loans, on your assets, and the cost that you have to stand on your liability. For the time being, we are seeing a further pressure on the average yield of our assets because again, as I've mentioned, we've reached, I hope, a bottom this quarter in terms of rates for new loans, but this continues to put a certain pressure on the yield of assets. At the same time, we've benefited, and we are still going to benefit in the coming quarters, hopefully from some positive elements like the fact that we have a resource at a negative rate, which is the TLTRO.
This is not going to last forever, so it's very important to have good momentums in terms of volume development, and volumes of good quality, of course. We don't want to relax our credit standards, of course. It's very important to keep a good momentum in the development of the volumes, and also it's very important to be able to generate a growing amount of fees within your revenues. The NII was significantly up this quarter, and definitely, it's benefiting for the time being from this very specific monetary conditions, in connection with the ECB policy.
Even more importantly, fees are up more than 6% this quarter, and this is also a very positive element because fees, of course, it can be a little bit more volatile, but it's here to stay also because if our customers are loyal to us and we do everything we can in order to keep them loyal, products and services are going to add up, and the fee base is going to continue to grow. It's very important for us to continue to develop this model, which is leading to your second questions regarding market shares. What we are looking at in terms of development of products is by definition, all the products that improve the loyalty of our customers. What we've noted is that P&C insurance policies improve the loyalty of retail banking customers.
You know that I've been running the insurance business of the group for five years before joining this position. I've seen it from the side of the insurance companies. It's absolutely true that the customer loyalty, considering the way we do our business of P&C insurance, every time we manage to sell a P&C insurance policy to a retail customer, it improves his loyalty. Even more every time when he has a claim on his P&C insurance policy, considering the fact that we manage the claim in a very specific manner, we improve further the customer loyalty. It's very important for us to continue to develop P&C insurance policies. Of course, life insurance, because it's the continuation of the management of the assets of our customers. P&C insurance is very important. Creditor insurance as well.
Real estate is important not for all our customers, of course, by definition, because not all our customers can afford to invest part of their savings in real estate solutions. Every time we have a customer that wants to invest part of his savings in real estate solutions, we must be able to provide the answer. We are working on continuing to improve the capacity of selling real estate to our customers directly. In terms of provisions and potential for releases, you know that of course, Crédit Agricole Group fully applies IFRS regulation, and we don't want to be outlier from this point. At the same time, we have been in the past very reluctant in terms of provision write-backs. This explains why at the end of the day, we have a higher coverage ratio than most of our European peers.
My main preoccupation is not to wait for the moment when I will be able to release S1 and S2 provisions. I know that it's maybe not as aggressive as some of our competitors, but clearly we feel comfortable with the idea of having ample provisions, even ample S1 and S2 provisions. We are not desperate and waiting to write back those provisions in order to boost our profit. Okay.
Thank you.
Thank you. The next question comes from the line of Giulia Miotto from Morgan Stanley. Please ask your question.
Yes. Hi, good afternoon, Jérôme. Thank you for taking my question. Two, please. The first one on Italy.
It's actually more strategic than the quarterly performance. Creval is now on track, and you're looking at that new asset. Would you potentially be interested in acquiring further assets, as it seems that there are other potential banks or parts of banks coming to the market? That's my first question. The second question, instead, looking at trends for loans and deposits in the quarter, so quarter-on-quarter. Deposits still keep growing more than loans in France, but also at the group level. Do you have any evidence already from the July data that this trend is inverting? When do you expect it to invert? Perhaps, what opportunity do you see from potentially lower deposits for your asset gathering businesses? Thank you.
Thank you. Let's start with Italy. We are pragmatic, and we are prudent. Pragmatically, we are focused on the integration of Creval, and this is going to take probably another two or three quarters because, as I've said, we are conducting due diligences in order to fully assess everything within Creval. We'll have, of course, to organize the legal merger of the two entities, Creval and Crédit Agricole Italia. We'll have to industrially operate this merger. We'll have to migrate Creval's operation on our platforms and so on and so forth. This is going to take time, and I think that part of the successes that we've had in the past in this viewpoint are linked to the fact that when we conduct that kind of operation, we are very focused.
I can tell you that Crédit Agricole Italia's management is really focused on these operations, which are complex. Even though we are very confident, it needs to be really well prepared. We are prudent at the same time, which means two things. We don't want to absorb too many things at the same time. Also, of course, we continue to monitor the situation. Because, again, I'm looking back on what we are in Italy now, EUR 385 million of net profit in the first half of the year. We have many, many interests in Italy, and we need to permanently assess the situation and be able to protect and defend our interests in all business lines and not only in retail banking. We are monitoring the situation. Of course, we are aware of some movements that are taking place or may take place going forward.
For the time being, we are focused operationally on the integration of Creval. Maybe one last point on this question, there's no need for us to grow further inorganically in retail banking activities. We have a nice setup. It's efficient. It's working well, as the figures of Crédit Agricole Italia this quarter illustrated. There's no need for us to go further. Regarding deposits and loans, it's true that this quarter, I think that globally for the group, including the regional banks of Crédit Agricole, customer loans increased by around EUR 28 billion all in all. Customer deposits, at the same time, increased by around EUR 30 billion, if I have the correct figures in mind. It's true that the difference between loans and deposits shrank a little bit by EUR 2 billion.
I think that this is going probably to reverse when our customers will have make up their minds on what part of this excess of deposit they want to consume and what part of these deposits they want to invest long-term. What we see now is that consumption accelerates. Part of this extra deposit is really going to be consumed, but not in such a proportion for the time being as to completely, I would say, wipe out the extra savings of the pandemic. There's probably going to be part of it that is going to be long-term invested, and of course, both Amundi and Crédit Agricole Assurances are ready to make their propositions to their customers in order to be here, and to attract these inflows.
For the time being, consumers haven't made up their mind, and when you see all the uncertainties regarding the future evolution of the pandemic, I cannot blame them.
Thank you.
Thank you.
Thank you. The next question comes from the line of Omar Fall from Barclays.
Hi there. Two questions. Firstly, could you help us in modeling the insurance business please? Since it's quite volatile. That EUR 400 million net number, is that sustainable on which to base our forecast, or do we have to make an adjustment for the financial margin? Is that +EUR 187 million positive market impact year-on-year that you quote just a normalization from a depressed return a year ago and the current run rate is normal? Or is the current quarter elevated? On the P&C side, I guess you will have had an elevated combined ratio with some of these weather events. Could we have the EUR million impact, if you've got it and it's material? Secondly, just wondering on the comment you made on rounding up the dividends for future years to include the 2019 dividend, if I understood that correctly.
Number one, do you consider all of the canceled 2019 dividends as being available to be rounded up? Secondly, why do you suggest that this will be done over time when you're in a position of significant excess capital today? You could pay EUR 2 billion today and you'd still be above 11%, and the ECB doesn't seem to mind your peers paying more than 100% of earnings. I guess, the previous question kind of asked it a bit more elegantly, but there are fears that, well, maybe you're holding capital so that you can make a larger acquisition or larger transaction in Italy. Thanks.
Okay. Insurance modeling. It's going to be difficult to fully walk you through all the insurance modeling issues in a few minutes' time. Of course, we'll be ready to, and probably will have at some point in time, to reorganize a dedicated workshop on insurance activities, because definitely it's a little bit less familiar to bank analysts than traditional banking issues. No, these elements that we quote as a positive market effect has not to be taken out of the normal run rates of the insurance revenues. It simply is an element that we point because this quarter, contrary to what happened last year, we have had some positive market effects in two aspects.
The first one is the fact that there are some legal reserves that we have to book within our books if the value of certain assets decrease, the market value of certain assets decrease below a certain point. There is, for example, provision pour garantie plancher. It's a provision for the floor guarantee that we provide on life insurance policies. There is also the provision pour dépréciations durables. It's a provision regarding long-term depreciation of certain assets. All these provisions have to be booked when some assets go below a certain value, and they have to be covered fully by the insurance company. It's not shared with the policyholders. We have had some specific provisions of that kind one year ago, and part of it is coming back in this second quarter.
We have also some positive re-evaluation of assets that are marked at market value under IFRS 9. This EUR 187 million number that we've quoted is the addition of many events of this kind. Some of them are being shared with policyholders, some of them being 100% for the insurer. What is for sure is that this quarter, considering the fact that partially because of this positive market effect, we have had a more higher level of financial revenues. We have been able to take only a proportion of the financial margin that we are allowed to take on these financial revenues. We have been able to book some additional provisions that are allocated to the policyholders, but that will not necessarily be paid by year-end.
You know that this accounting between participation benefits, so the yearly profit-sharing rate and participation aux excedents, PPE, which is reserved in our books but not necessarily paid immediately and kept for the future, is definitely made only by the end of the year. What we've done this quarter is that we've provisionally, again, increased this PPE because we had a very high level of financial revenues. It's not definitive before year-end. This is globally the mechanics that operate within the life insurance business. In the P&C insurance business, it's true that this quarter we have had a significant amount of weather events that impacted a little bit the P&L. You know that there are two mechanics that really filter the impact of these weather events on the P&L of our insurance company. The first element is that part of it is reinsured outside the group.
Of course, a part of these weather events was taken by the reinsurance companies. There is a risk-sharing mechanism between the insurance company and the banks that sell the policies in order to align the interest in terms of underwriting the policies. Part of the cost of these weather events is shared by the regional banks of Crédit Agricole and by LCL through the level of fees that they get from Pacifica, which is the P&C insurance company. All in all, this quarter indeed, the ratio of claim costs as compared to premium within Pacifica increased quite significantly compared to Q2 2020 from, let's say, around 62% to 72%. It's significant, but Q2 2020 was very low because of the lockdowns.
In terms of combined ratio, which includes the commissions that Pacifica is paying to the distribution networks, actually it's been very stable, around 97%. Actually the profitability of Pacifica was more or less unchanged as compared to last year. It's not a very significant impact that we've had this quarter, despite the fact that all these mechanisms, reinsurance and sharing of the risk with the distribution banks has been operating. All in all, is this level of profitability a run rate? I'm not pretending that every quarter we are going to post EUR 400 million of net profit for the insurance division, but it's not an abnormal level whatsoever.
Very clear. Got it.
Okay. Coming to your question regarding the dividend. First, the unpaid 2019 dividend was EUR 0.70 a share. What we've said when we've presented the 2020 dividend scheme earlier this year was that this 2020 dividend scheme was covering around EUR 0.30 of unpaid dividend coming from 2019, because the normal dividend in 2020 would have been around EUR 0.50. It means that we deem we have another EUR 0.40 per share to pay to the shareholders in the future. Why don't we pay one-off this amount? We don't like the idea of decreasing the level of the dividend. Paying an extra dividend or paying a very high dividend on a single year is not an idea that we like. We really want to be as stable, as recurring as possible.
What I can tell you, nevertheless, is that we do not intend to keep, in the long run, extra capital in order to have a reserve, just in case we have a large M&A operation to finance. I have always said that it's better to be close to the target in terms of solvency and to ask the shareholders to provide the necessary funding of an M&A transaction, because then it creates, I would say, a positive inducement to present a very convincing operation. It's far better, and it's far more efficient.
Just as a follow-up, I take that on board, but do you see that there's maybe a clash between that view? Because if you don't want to pay one-off more sizable dividends, yet you're in a position of significant excess capital, how do you then get down to a more normalized level of CET1 if you're not banking on that additional capital for M&A?
Yeah. Of course. Just keep in mind that up to a few days or weeks, the level of capital where we stand was a compulsory level, considering all the restrictions put in place by the ECB. The level of capital that we have now, and it's going to be the case up to the end of Q3 this year, is again, a compulsory level of capital under which it was not possible to go, unless precisely we had concluded very large M&A transactions. Because it was the only way to spend capital accordingly to ECB's recommendation. Now we are going to be starting in October in a more normal situation.
Regarding our intentions for the fourth quarter, we want to stick to what we had said beginning of this year, again, it's the idea of being very predictable. We want to stick to what we had said for the rest of the year. Then end of this year, we will assess exactly this rounding up the numbers exercise I was mentioning.
Thank you, Jérôme.
Thank you.
Thank you. The next question comes from the line of Delphine Lee from JP Morgan. Please ask your question. Your line is now open.
Good afternoon, Jérôme, and thank you for taking my question. I just want to quickly clarify on what you said on the full year 2019 dividends, the EUR 0.40 remaining. You mentioned 2021 and 2022 to pay that. Is the intention to pay the full EUR 0.40 amount by the end of 2022, i.e., by the end of your business plan, or do you intend to spread that out over longer time periods? Could we expect EUR 0.20 per year, or is it going to take much longer? Also just sorry to come back on the share buyback discussion, but I understand the regional banks don't want to increase their stake too much, and that's why you have a preference for dividends. You've just issued quite a lot of shares, because the scrip that they have taken.
I don't understand why you wouldn't buy back a lot more of those shares, you're only buying back the shares taken up by the minority shareholders. In proportion, that could be considered. I don't know, you've always run well above the 11% target, and I was just wondering if that's a real target, or you actually aim to maintain some kind of buffer for prudence on top of that 11. Just on capital, and that's my last one really, is just regulatory impact on TRIM. You haven't taken much this quarter. Can you just confirm how much is left for the year, please. Thank you.
Okay. Many questions, finally, all related to capital somehow. Our intention is not to spread this EUR 0.40 repayment on the next 40 years, to be a little bit caricatural, Delphine. Of course, we don't want to do it in too many years, but probably if the situation is convenient, we can do it in two years. On the back of the 2021 dividend to be paid in 2022, on the back of the 2022 dividend to be paid in 2023. Regarding the share buyback, actually, we are about to finish to repurchase the shares issued with the take-up of the minorities at the end of the present share buyback operation. It means that the second share buyback operation is going to apply to part of the new shares created for the sake of the regional banks. That's for sure.
We are not preserving this number of share. The EUR 500 million that we've announced is going to apply, by definition, to shares somehow, I would say notionally, intellectually, created for the sakes of the regional banks. What I was saying is that in my understanding, but really, I'm not the shareholder, I'm only the listed company in which the shareholder has a stake. In my understanding, the regional banks don't intend to sell shares during this share buyback operations. It means that mechanically, as they keep their shares, they are going to be diluted within the capital of CASA. I think that it's not in the best interest of minority shareholders to see the majority shareholder seeing its stake mechanically growing and growing permanently. This is why we prefer to pay a cash dividend to our shareholders.
Actually, if all these events regarding the dividend payment restrictions hadn't happened, we would be much lower than the level where we stand now. That's for sure. There is no management buffer above this 11% target. That's absolutely certain. Again, we are in an environment where we have restrictions on payments, and we are also in an environment where there are still some uncertainties in terms of regulation, because we haven't talked about Basel IV yet. I think it's not useful to talk too much about Basel IV because we're still awaiting for the draft proposal of directive from the commission. We need to know exactly where it's going to land before being able to really project our capital trajectory in the long run. Then TRIM. TRIM is part of these regulatory uncertainties.
In my understanding, but it's varying regularly quarter on quarter. We still have, let's say around EUR 6 billion-EUR 7 billion of RWA to take in the coming six quarters. Let's say between EUR 6 billion and EUR 7 billion in the coming six quarters. The calendar is not very precise because you know that it depends on interactions with the ECB on specific reports, and of course, we try to discuss, and then there is a whole phase of interactions. The calendar is not very precise, but this is more or less what is left in front of us. Okay?
Great. Thank you very much.
Thank you. The next question comes from the line of Lorraine Quoirez from UBS. Please ask your question. Your line is now open.
Hi, Jérôme. Thank you for taking my question.
Hello.
Just a quick one from me actually. It's regarding to the accounting of TLTRO. Can you remind me how is it accounted for in LCL, and for how long we'll have the impact of that, and how much, please? Thank you.
Yeah. It's accounted for, I would say, over time. It's accrued, actually. It means that every quarter we book a liability within LCL, but also within CACF, within Crédit Agricole Italia, within CASA, and of course, within the regional banks of Crédit Agricole, which bears a negative interest rate of -1%. This was the case up to end of June this year because we had certainties about the fact that the condition related to the extra premium was going to be met, and actually, it's been indeed met. For this first period of premium, which is now over, the accounting was really straightforward. Since July the 1st, we have started the second premium period, which is conditioned by another metric, another period of check of the evolution of the loans. We will see with the accountants how we do for this new period.
For the time being, what I can tell you is that the portfolio of loans, which is the condition for being granted the second premium, is up 3% or 4%. We are, again, quite comfortable with the idea that we are going to be able to receive this second premium for this second year.
Right. In total, it should be over three years, right?
No. The second period is also for one year. The first period was between June 2020 and June 2021.
Okay.
The second period is between July 2021 and June 2022.
Okay. Perfect. Thank you.
Thank you.
Thank you. The next question comes from the line of Tarik El Mejjad from Bank of America. Please ask your question. Your line is now open.
Hi, Jérôme. Just a couple of follow-ups. First, on the provisions. Clearly, you seem cautious on releasing the provisions and favoring higher coverage ratio versus these releases. Isn't it automatic if you have to update your models? I guess you have to if the environment is improving, and that would lead to releases, or is it your choice not to adjust the models and be voluntarily more cautious in your assumptions? The question number two, sorry to come back on capital because, correct me if I'm wrong, but I sense some cautiousness versus your rhetoric in Q1. Basel IV, yes, is uncertain, the Pillar two requirement relief that you promised to update us in full year would probably offset all Basel IV or most of it. You're left with the massive excess capital.
If your intention is to increase a few cents per year to keep the progressiveness of DPS, that would definitely take a long time if you don't use this excess capital to grow faster or non-organically. I just want to know if the tone has changed, just the impression, or you sense that you want to be more cautious and progressive. I hear you about the fact that being restricted to pay dividend has the accumulation of capital, but it's fair as well to have a catch-up to basically, as it was abnormal to ban you pay dividend, it's also maybe unusual for you to deviate from your strategy of progressiveness and run down the excess capital faster than what you had in mind pre-COVID. Thank you.
Okay. Let me start with the second question, actually. We didn't change tone as compared to the beginning of the year. What I'm just saying is that before accelerating the unwinding of this excess capital, I have first one tool and one usage, which is the dismantling of the switch mechanism. It's the commitment, as I said. It's going to be important to enhance the earnings and the earnings per share, and it's to be done before year-end 2022. This is going to consume 60, 65 basis points of capital. It's important, of course, to keep the necessary amount in order to do it exactly when we deem relevant to do so. Second element is that, of course, we have these uncertainties regarding Basel IV.
Alongside with Basel IV, there is this capacity of covering Pillar two requirement partially with AT1 and Tier two, and we don't forget this component of the package. Of course, it's better to know the full picture before drawing a trajectory of depletion of capital, if I may say so. Really, don't misunderstand me. I'm absolutely not changing tone as compared to the beginning of the year regarding this issue. In terms of provision, of course, it's automatic, but it's automatic with many assessment that you can make. The first assessment is the macroeconomic scenario that we use in order to calculate the expected credit losses, one year expected credit loss or maturity expected credit losses. Of course, we have to reassess permanently the macroeconomic scenario. For the time being, we are positive, and of course, we have a positive scenario.
We don't want to be over-positive, even though there are some indications saying that most of the official public institutions that publish macroeconomic scenarios are more optimistic than what we are. Nevertheless, we are not over-optimistic in our own in-house scenario that we use for the provisioning. Second, you may have in mind that S1 and S2 provisioning is made of two layers. The first layer is the central forward-looking, which is based upon our global macroeconomic scenario. We have local forward-looking layers that are added up to the normal provisioning, S1 and S2 provisioning, either in connection with certain specific sectors or with certain specific geographies. In this front, we have certain capacity of being more prudent than only the central forward-looking scenario. Okay.
Thank you.
Thank you. The next question comes from the line of Stefan Stalmann from Autonomous Research. Please ask your question. Your line is now open.
Hello, Stefan.
Yes. Good afternoon, Jérôme. I wanted to ask you, please, on the stress test for Crédit Agricole Group. I mean, the group drawdown was more than 6%. I guess we can back solve roughly for something around 5.5%-6% for CASA as well. With that kind of stress test drawdown, is 11% really the realistic CET1 level to shoot for? In that context, do actually Pillar 2G considerations play any role at all at your level, or is that only an issue for the group? Maybe related to this question, a very general question. It always strikes me as surprising that a group like Crédit Agricole has such a severe stress test drawdown compared to many other banks. Is there anything that you would highlight that you think is responsible for that?
Are there maybe any issues arising from the group structure that penalize you compared to other banks? Any color there would be very helpful. Thank you very much.
Very good question, Stefan. Thank you. Thank you very much. Well, first, it's true that we have a depletion, which is huge and which is bigger than the one we had previously and bigger than the one of most of our peers. Considering the starting point of our solvency at group level, it's not really an issue. As you know, we end up this severe scenario stress testing, with a CET1 ratio, which is close to 11%, which is the second-best in the European space of systemic banks. The best being exactly at 11%. It's not really differentiating. We can consider that amongst the space of European systemic banks, we end up with the best resilience to this very severe stress scenario. Why does it happen that we are so severely hit? I think the main reason is the scenario itself.
The scenario is hitting very hard on France, very hard on professionals, very hard on exactly all the businesses in which we are dominant in France. Of course, the scenario is hitting hard on our group because it's hitting hard, and in my opinion, absurdly high, specifically on the businesses in which we are very present and in which we don't intend to exit, of course. These are our main priorities, and we are going to stick to these priorities. Of course, we think that the severe scenario of the EBA is not very credible, especially this idea of saying that all public support is lifted and not applicable, doesn't work like that in the real life. We don't buy that scenario. Nevertheless, we've applied it.
A second point, there are some specific elements that hit us particularly, which are either methodological or which are linked to our good resilience to the scenario itself, I would say. In this second category, it's the fact that at no point in time we across the MDA threshold. It means that during all the period, we continue to pay AT1 coupons. Of course, compared to banks that stop to pay AT1 coupons as soon as the first year of the scenario, it's an additional capital depletion. For us it's around 18 basis points of depletion in the period. Second element, and it's purely, in itself very interesting, but we have had in 2020, which is the reference of the stress testing both for the starting point and for the run rate of activity, you know that we have had some specific one-offs.
Like for example, the fact that we've decided voluntarily to pay and to indemnify some of our customers that have an activity interruption insurance policy with us that was not supposed to be called for a pandemic, but we've decided to, nevertheless indemnify our customers on this ground. To pay voluntarily a significant amount was around EUR 300 million for the group globally. The ECB, the EBA said that we had to do as if we were to continue to repeat the same every year of the stress scenario. It's an example, but this is also a cost of around 16 basis points of CET1 depletion at the end of the day. As this stress testing exercise is not a negotiation, we have to apply at the end of the day to all their requests, we couldn't modify that.
We consider it a little bit unrealistic again. We have some bits and pieces like this, and we have also the fact that our business model is precisely in the heart of the adverse scenario of EBA. Going back to your question, there's nothing regarding our structure, our organization that is contrary to the methodology and that would really penalize us. How it's going to play on the P2G, we don't know exactly. The ECB is to a certain extent, a little bit of a black box regarding the connection between the results of the stress testing exercise and the stress exercise. We'll know a little bit more about it in the fall because we have this stress exercise, a meeting that is organized, I don't remember exactly, probably in September or October. We'll see a little bit more how it works.
Regarding CASA, there's no CASA stress testing. Of course, CASA is integrated in the stress testing of the group, but we don't publish any CASA stress testing specific result, and there's no request or requirement on CASA specifically regarding the stress testing. Thank you, Stefan.
Thank you very much for that, Jérôme. Thank you.
Thank you.
Thank you. The next question comes from the line of Kiri Vijayarajah from HSBC. Please ask your question.
Hello, Kiri.
Hi, Jérôme. Good afternoon. Just a couple of questions on Creval, if I may. Firstly, on the net bad will on Creval, correct me if I'm wrong, but you haven't included any kind of DTA benefits in there. Really what's your thinking there? Roughly how much could the DTAs be worth eventually, on some of those adjustments you're taking? Then, related point, now you've got your hands on the Creval business, have those PPAs come in line with what you originally expected when you were trying to price up the original tender offer? Things like, the extra provisioning needs on the loan book, et cetera, have they come in better or worse than anticipated? I guess the macro has kind of yo-yoed around a bit since the beginning of the whole process, when you launched the tender offer for Creval.
Yeah, just your view on how the PPAs have come in versus original expectations? Thank you.
Thank you. You're right, we didn't book any DTA so far because in order to be allowed to book DTAs, we have to have a legally binding decision of merger in place. The legally binding decision of merger is going to be taken after general assembly meetings, both of Creval and of Crédit Agricole Italia, that are going to take place in Q4. It's going to be only end of Q4 that we are going to be able to book the DTAs. Within Creval, off-balance sheet DTAs amount to around EUR 180 million. In terms of findings regarding the Creval situation, well it's too early to tell, but what I can tell you is that everything we see is confirming the very good potential of this acquisition in terms of complementing very nicely and very efficiently our setup in the northern part of Italy.
It really is a very attractive acquisition.
Great. Thank you.
Yep.
Thank you. The next question comes from the line of Matt Clark from Mediobanca. Please ask your question.
Good morning. Sorry. Good afternoon, Yves.
Good afternoon.
Can I come back to your comments earlier that the EUR 400 million net profit for the insurance business is not an unusual or not an abnormal number? Would you also say the same would apply under IFRS 17? Can we still expect to see that level of profitability come through some quarters in the future? Anything else you can say to help us understand the impact of IFRS 17 on the group, both from a capital and profitability perspective? Actually, my other question's been covered. Thanks.
Well, thank you. It's a very good question. Definitely IFRS 17 is going to be a headache, both to apply and to explain and to live with in the future. That's for sure. We were not, as you can imagine, really candidate for IFRS 17. Nevertheless, we are going to apply IFRS 17 when it's applicable. Starting in 2023, but with the dry run that we have to do in 2022. We will have a clearer view pretty soon now on the effect. You know that under IFRS 17, the total profit generated by insurance activities at the end of the whole day is going to be the same. The issue is that it may be spread quite differently over time, and that's the difficulty.
We have still some arbitration to do with the first-time implementation of IFRS 17 between what is going to bite on our solvency at the level of the group in order to protect as much as possible the profit generation capacity. On the contrary, what is going to preserve better our solvency at group level, but at the price of deteriorating a little bit the revenue trajectory and delaying a little bit the acknowledgment of revenue. These are questions that we are working on presently quite hardly, and we are going to be able to give more clarity on that probably at the end of this year. As of now, I am not really able to give you some precise color on that. What I told about the EUR 400 million is really under the present accounting standard.
Okay. Just to follow up on that, I think in the current business plan to 2022, when you came up with that, it was still expected that IFRS 17 would have been imposed by 2022, but you chose to exclude that effect separately. Does that mean that really when we think about the order of magnitude, IFRS 17 could be the order of magnitude that would have threatened your group targets?
It was much more because in 2019, we had a much less clearer view of all IFRS 17 impacts, and it was too difficult for us back in 2019 to really be able to give the, I would say, the sensitivity of the 2022 P&L target to the implementation of IFRS 17. We preferred at that time to say that we were setting those targets under a constant accounting standard for the insurance activities. At the end of the day, we were right, because indeed, IFRS 17 is not going to really be the standard before 2023.
Okay. Thank you.
Thank you. We will now take our last question. The question comes from the line of Pierre Chédeville from CIC. Please ask your question.
Hello, Pierre.
Hello, Jérôme. Just two question. Regarding insurance, we've seen growth in penetration rate in your domestic markets and even if there are room to progress further, notably at LCL and Italy. My question was regarding the international development, because I understand that you were keen to develop this insurance expertise elsewhere, and I wanted to know where you are regarding this international development and if you are also exposed to exceptional floods in Germany, Belgium or Eastern Europe?
Excuse me. Exceptional floods? What do you mean by that? Oh, okay. Excuse me.
Sorry for that.
The answer is no.
Okay.
No impact of the floods.
Okay. The other question is regarding provision. We have all understood that the level of provision is exceptionally low. We are also understood that even if you are reluctant to make write backs, you will do so somehow in the future. My question was, in your opinion, when do you see the level of provision or a normal level of provision, we would say around 40 basis points, coming in 2023, 2024. How do you see that in your view? Thank you.
Okay, thanks, Pierre. Well, in insurance activities, it's clear that we want to accelerate the international developments of, I would say, our know-how in terms of bancassu rance capacities. We have in the recent past taken several initiatives. Actually, Creval is the result of one of those initiatives, because initially we started with a life insurance distribution partnership with Creval. This was the starting point of our love story with Creval. We've also concluded a partnership in Spain with Abanca. To be clear, I'm not pretending that the Creval scheme is going to be repeated in Spain. That's very clear. We have also a partnership with Novo Banco in Portugal. We have also a global insurance partnership between CACI and CACF activities across Europe.
CACF is distributing CACI, which is one of our insurance companies, products across the whole Europe, alongside with personal finance products, consumer credit loans or auto financing loans. We have a full range of initiatives in order to develop our insurance activities across Europe. We definitely would like to accelerate on this front. We're working on several potential partnerships that could enhance the development of our activities. Of course, we are going to work on several partnerships that are not going to generate anything. It's something on which we are not going to communicate. From time to time, we may conclude some additional partnerships in order to boost our capacity of distributing insurance policies. We very much believe in the strength of this model. We think that we master this model of bank insurance distribution.
We want to develop this model across Europe. Regarding provisions, you rightly quoted this figure of 40 basis points, because this was precisely across the cycle figure that we had in mind when we published our medium-term plan, 40 basis points of cost of risk. It happened last year, we were above, and this year we are already significantly below for the time being. When are we going to reach back this 40 basis points amount? I don't know exactly, but on the single Q2 quarter, we were at 24 basis points on the perimeter of CASA. If you take a rolling four-quarter average, then you will end up with this 40 basis points figure.
This 40 basis points figure across the cycle is not irrelevant, I would say, and I expect that we are going to vary around this figure, and possibly we are going to be below this figure for the coming quarters. Then at a certain point in time, but I don't know exactly when, we will be back at 40 basis points.
Okay. Thank you very much.
Well, thanks. I understand this was the last question. Again, thank you for your time. Thank you for your attention, and I wish you all a good holiday. Personally, this is what I'm going to do in the coming hours now, and most of the team in the coming days. I'm glad that we are going to meet again in the fall after the vacation. Bye-bye.
Thank you. This concludes today's conference call. Thank you for participating. You may now all disconnect.