COFACE SA (COFA) Earnings Call Transcript & Summary

Good day, and thank you for standing by. Welcome to the Coface presentation, IFRS 17 Pro Forma Conference Call. [Operator Instructions]. I would now like to hand the conference over to your speaker today, Phalla Gervais. Please go ahead.

Thank you, operator. Good afternoon, everybody on the call. So I think today, we're going to present you or provide you with an overview of pro forma IFRS 17 full year 2022 results. And I will go through this in this pack. Before we start, I think I just want to highlight a couple of things. The first one is, of course, as we know, IFRS 17 on life on the 1st of January 2023, which means that the next time we talk will be the Q1 ‘23 results will be presented under this new accounting standard. The numbers that are presented today are the so-called pro forma, which is the full year 2022 results with a retroactive application of this new standard. I hope that you will provide you with some usual reference for your modeling. And then that's probably one of the purpose of the call. We have already presented in at year-end results in February, the first-time application, which is the opening balance sheet under IFRS 17 as of 1st of January 2022. Please note that this has been reviewed by our external auditors now with no changes, no change. The pro forma data that we would go through today, which is some quarterly numbers and more importantly, the full year 2022 P&L as well as the closing balance sheet as of 31st of December 2022 under IFRS 17 are, as we speak, under review by our external auditors. I want to highlight as well the fact that we have moved since the 1st of January 2023 from IFRS 39 to IFRS 9, which is related to the investment portfolio, accounting classification. However, on the IFRS 9 doesn't require us to have a pro forma per se on the previous year, and Coface has chosen not to do so. What you see during this presentation is, I think a couple of comments that will be developed during the course of today's presentation that the 2022 pro forma numbers needs to be analyzed in conjunction with the first-time application. 2022, obviously, was a transition year. And during this transition, you will have some short-term differences from one note to another one. You will see that the profit relocation pace is deferred from IFRS 4 and IFRS 17. However, given the short-term nature of our business in midterm, of course, shareholders' equity, earnings and cash, cash being cash convert between these 2 norms. More importantly, Coface will continue under IFRS 17 to report that we rely on the same KPI, but with very limited definition changes that we'll go through. If we move to the next page, which is Page 5. I think this page has been already presented in February, but it's good probably good also to be reminded. So our strategy remains unchanged and impacted -- not impacted by IFRS 17 implementation. So what was our state of mind when this implementation? In terms of principal, I remind you that the short-term nature of our business allows us to apply a simplified approach, which is the premium allocation approach. We have no contractual service margin, no CSM. This is a principle. We have insured or tried to ensure the same KPI, the continuity of KPIs in terms of premium combined ratio, return on average tangible equity, we're not adding any new KPIs, and we're not replacing any of the existing one. We stayed very corian in terms of reserving principles. You will see that later on. First time application, of course, 1st of January 2022 has been presented. And last but not least, we have leveraged our existing Solvency II processes, which means that the best estimate that we are coming up to it under IFRS 17 is exactly the one that we're using for our Solvency II calculation. What comes out of this implementation, first is that the reserving philosophy remains broadly unchanged. You will see that when we talk to you about the level of reserve on the new vintage the opening year and of course, the prior year development, positive or negative, saisosophy. The strategy of Coface is unchanged. The build-to-lease assumptions and the target through the cycle remains unchanged. So with the combined ratio below 80%, return on average tangible equity above 9.5% and payout ratio above 80%. We Cash being cash, cash flow over the lifetime of the policy are unchanged. However, what we will see is that the new rules tends to accelerate the profit recognition, and there will be some volatility, especially from one quarter to another one. Financial leverage unchanged. And what we already see in terms of first-time application as of 1st of January 2022, shareholders' equity slightly has slightly increased compared to IFRS 4, increased by EUR 91 million net of tax net of reinsurance, which represents EUR 0.60 per share. So let's go to the per part of the presentation. What you will see going forward with all the pages where there is an impact related to IFRS 17, and I move to Page 7, which is the turnover. So on -- if you look at the chart on the left-hand side, you have IFRS 4. On the right-hand side, you have IFRS 17. And of course, we are only looking at the full year 2022. So now he will change slightly and the difference is less than 1%. And this, of course, impacts only the earned premium, which is the insurance business where while fees and other revenues stay unchanged. We good earned premium moving from EUR 1.27 million to 56 million. Key various here, we're talking about something which is close to 11 million, which is nothing. However, you have 2 components. The first one is a reclassification and a reclassification from the OpEx line, and this is related to the warm what commissions being the commission that we are paying on our fronting reinsurance business. Under IFRS 17, this will come as a deduction of our gross earned premium. So it's just a reclass from OpEx to premium. Then -- because the duration of our policy slight -- in terms of premium approximate 17 months, which is, of course, obviously longer than 12 months, we -- under IFRS 17, we have to recognize a coverage period extension, which is a timing difference, and we have introduced this mechanism in our premium or condition. And this goes on top of the gross end premium as of full year 2022. Bottom line of this is that the gross earned premium has slightly decreased compared to IFRS 4, less than 1%. I think we can move to the next page. Next page, you could see the same changes in terms of turnover across the region. You can see that the only region where you have this impact more significantly compared to the other ones are Asia Pacific. And this is because -- given the nature of our business in APR in Asia, this is where we have the most fronting business, we ensure fronting business. And of course, the -- we're moving from EUR 151 million to EUR 131 million, as we have taken into deduction in what commissions of this fronting business. Just showing you from a geographic point of view where the nose commissions is impacting. Now let's move to Page 9. I think probably the most interesting page for us in terms of changes. I will start with the chart on the top left-hand side, and you will have the gross loss ratio before we insure and after can ending cost by quarter and for the full year. On the right-hand side, you have the same gross loss ratio on the IFRS 17 pro forma. I will comment first the full year. So on a full year basis, on the IFRS 4, the gross loss ratio last year was at 31.2%, while it would have been at 35.5% under IFRS 17. What drives these differences? I would point you to the chart on the bottom right-hand side. Again, you have the 2 years, IFRS 4 and IFRS 17. And of course, here, we're looking at the 12 months 2022 only. If we -- you're already familiar with this chart, which is the opening year and the prior year development. So I will start with the opening year. The dark blue bar under IFRS 4, you can see that at the end of last year, we have opened the new vintage at 8.2%. Under IFRS 17, we would help open a new vintage actually at EUR 80.5 million but what makes the difference between 8.5%, with the 77.3% is based on the fact that under IFRS 17, we have a discount -- we have to discount the reserving methodology. So to compare apples to apples, you can see that the new vintage would be very similar in terms of reserving philosophy with opening new vintage around 80%. So what happened on the other side, which is the prior year development, you can see that the prior year developments, which are the bodies that we have recognized in 2022. On IFRS 4, will represent 51.6% its reserve release, while under IFRS 17, it was at 44.2%. While we said that we have a different pace in terms of profitable condition, this is what you see here. As under IFRS 17, the prior year developments are recognized faster than under IFRS 4. And the way that we have recognized it faster was through the first-time application, which is an opening balance sheet. And I think that you may recollect that we have a first-time application impact of EUR 91 million net of tax, net of reinsurance. And then we go back to the chart above where you can see clearly the work. So IFRS 4, full year 2022 loss ratio was 31% at 31.2%. Then, if we add the first-time application, which is a faster recognition of the margin developments under IFRS 17, you are adding up 5.5%. Then you have minus 1.9%. This is mainly the discounting effect on the past and methodology differences and timing. And then, of course, you have the impact of the inward commissions as it's a ratio, and we have a lower gross premium under IFRS 17. This is why I think when the highlights that we put, you really need to look at this transition year, which is the P&L but also in the conversions with what we did at the first-time application in the opening balance sheet. This gives you the full view of the transition in 2022 between the 2 norms. Then I will circle back actually to the first draft where you can see the difference between the quarters. And most specifically, it tells you that under IFRS 4, the reserve release related to the prior year mainly happened a big churn happened in Q3 2022 while, of course, this has been recognized under IFRS 17, mainly in the FDA, in the first-time application and the opening balance sheet. Move to the next page, Page 10. You can see these variances for year 2022 by geography. Again, I think if you look at the geography, there's some slight differences from one geography to another, but it also gives you an idea of in which geography, the first-time application, so the prior year development recognition under IFRS 17 in the opening balance sheet has been recognized since [indiscernible]. If we move to the next page, which is only IFRS 17 per quarter. What we want to highlight here is the fact that between the 2 norms, the reserving philosophy was exactly or principle was the same. You do recognize that in Q4 2022 in Latin America under IFRS 4 or under IFRS 17, we have booked up these very, very large claims in Latina in Brazil, more specifically. Same story in Central Europe, where in Q1 with the Ukrainian war, we have booked up reserves, same story under IFRS 4 and 17. In Q2, we have reallocated this reserve IBNR in the region where the risk has been underwritten. Q3, we booked up with the, I think when Russia announces the mobilization and Q4, same story has been reallocated to the region where the risk has been unwritten. Bottom line of this is that between the 2 norms, the messaging in terms of reserving principal and philosophy remains exactly the same. If we move to the cost page, external and internal. We're moving from EUR 851 million to EUR 819 internal cost has stayed unchanged. And of course, you can see this reclass related to the inward commissions of EUR 31 million from OpEx to gross earned premium that has been netted off this in what commission. Mechanically, the gross cost ratio decreased between the 2 norms. If we move to the regions page, again, pretty much the same story, of course, on the reinsurance side. Premium session rates remain almost the same, and this is just the adjustments on the gross earned premium. Claims sessions rate is higher on the IFRS 17. This is due to the fact that we have released reserves on the more reserve on IFRS 4 during the full year 2022 P&L. And of course, this benefit the measures that well... Net combined ratio at 67.6%. The same story that we -- the same explanation we provide you with in terms of gross loss ratio applied to the net combined ratio. So on the IFRS 4, in full year 2022, the net combined ratio was 64.9%. We have this first-time application, which is a faster recognition of the prior year development -- positive development in IFRS 17 operating balance sheet. And this time, it's, of course, net of ration to 3.8%, and we have the impact of the inward commission, not only on the loss ratio, but also on the cost ratio. This led to this list us to the next page, Page 15, with a net income of EUR 240.4 million, knowing that in the first-time application, the net equity has been increased by EUR 91 million net of tax net of reinsurance. If we move to the next page provides you the equity work, but this is on the FY '17, which is EUR 2.229 billion. This is the opening balance sheet with EUR 91 million higher than IFRS 4. Cash being cash, I think the 224 million dividends is the one that has been approved. We are adding up of positive pro forma net income full year 2022. Minus EUR 264.9 million. This is the mark-to-market. So the unrealized loss related to the interest rate increased mainly on our investment portfolio. These numbers exactly the same that we have recognized under IFRS 4. And then you have others, again, part of that is exactly the same number that we have recognized under IFRS 4, lead us to ending equity of EUR 2.18 billion. If we look at the work between the 2 norms on the return on average tangible equity, 15.6 under IFRS 4 at BMN 2022, we have the equity impact as the equity on the IFRS 17 is higher than on IFRS 4. And we have the net income impact this led us to return on much tangible equity pro forma 17 --IFRS 17 of 12.7%. What will the balance sheet. Moving now to Page 18, the balance sheet variances between the 2 accounting standards at year-end 2022. So here, we're talking about 31st of December, a couple of things here, and we have already presented that in the opening balance sheet, the principle is the same. So total asset total liabilities under IFRS 4 was at EUR 8.451 billion moved down to[indiscernible] billion on IFRS 17. As you might recollect that under IFRS 17, there is a lot of netting impact between reserves, other assets and other liabilities. Of course, goodwill insurance investment, factoring assets, factoring liabilities and hybrid debt don't change, not impacted by IFRS 17 changes. I will put it this way. And I think what needs to be noticed is, of course, a net shareholder equity differences which is now for the ending balances, the difference between [indiscernible] to 2.19, EUR 59 million compared to the EUR 91 million in the opening balance sheet. So what would it look like if we have to have, I think, both target through the cycle with IFRS 17 month pro forma for sure. Well, with what we said, the combined ratio at 67.6%, so still below the B2B target of 80%. A payout ratio of cash being cash would have been 94%, which is above the 80% target. Return on average tangible equity at 12.7%, above the 9.5% target and solvency ratio, there's no reason why it has changed. As I said, we are leveraging the same as estimate on the IFRS 17 than for my Solvency II calculation. The key takeaways of this overview. Our -- Coface strategy remains not unaffected by IFRS 17. You can see that the observing philosophy remains broadly unchanged. We are -- we have almost identical opening year if we disregard the discount effect for the new vintage. The bit-rate assumptions and through-the-cycle targets remain valid. And more importantly, I think we have just a faster prior year developmental condition under IFRS 17 than under IFRS 4. But knowing that our business cycle is about 2 years, I think cash being cash in 2 years' time the tons converge. Coface will continue to report and rely on the same KPIs with very limited [indiscernible] changes. You can see that on the course earned premium. I hope that you have seen that the pro forma P&L 2022 needs to be analyzed in conjunction with the first-time application. This just illustrates the faster recognition of the prior development under IFRS 17. We have started to converge during 2022. And more importantly, I think the bit lead through the cycle objectives remain completely valid under IFRS 17. With this, I think we can open up for questions, operator.

[Operator Instructions]. We will take our first question. And the question comes from the line of Michael Huttner from Berenberg.

It's really clear. Really and you're the first of my conference to report, I'm really delighted with it. And also you've provided the exclusive. It's just amazing. I have 2 questions. The first one, so EUR 91 million is the net of tax, net reinsurance impact on the opening balance sheet. So I think EUR 39 million portion has already been used. I don't know how much exactly, but the portion. And there's still a portion left. So I'm just wondering, when will that portion affect earnings? Is it all in 2023? So really, my question is, does it mean that consensus should come down and by how much? I think it's EUR 32 million or something? And the second question is really simple. Is there any change in the asset allocation? So I noticed you said the unrealized gains impact on the balance sheet is the same. But I think under IFRS 9, there are some changes in profit recognition for assets. And I just wondered whether that would affect your philosophy…

Yes. So on the first question, indeed, we have started to converge in year 2022, as you've seen. And of course, what has been taken will not be taken again. It really depends on the development. So we still have policies that are under development. So this will be taken into account in 2023, I believe, for the vintage related to '21 and '22. Still too good to see, it will depend on how -- how our claims are developing. On the second one, which is related to IFRS 9, so a couple of things, and you're totally right to highlight this, and we have presented it in February. As I said, we have no pro forma related to IFRS 9 in 2022. However, now if you go back to the full year '22 presentation, we have shown that, first, we have -- and we know that IFRS 9 will lead to some volatility in P&L. This is a matter of fact, but we have already reduced our equity for instance. At the end of last year, equity represents 3%, and this will go on to the -- it will be -- this will go into equity in terms of mark-to-market. What remains, I would say, volatile in our P&L would be investment funds in real estate. And we have approximately 7% only of -- that might drive some volatility in our P&L, and this is the real estate funds. And you know that real estate funds needs to be looked at a midterm or long-term view. So this might create some volatility in our P&L, but in 2023. This is why you don't see it in any balance sheet because at the end of 2022, we are still applying IFRS 39. Makes sense?

Yes. And how much would it be?

What do you mean how much would be in the...

The difference between IFRS 39 and IFRS 9.

Well, we haven't calculated the P&L impact. It's only -- we have opening -- it's the balance sheet review.

Your next question comes from the line of Benoit Valleaux from ODDO BHF.

One question on my side is regarding your reserving policy. I understand that the level of reserve releases in '22 was mostly related to the first time application. But I just wonder if going forward, you believe that you will still be able to, I would say, manage or smooth a little bit the volatility of earnings. And when I look at Slide 18 on your balance sheet do you have in your 1.4 such as 3 million amount of reserves. I mean, do you still have some, I would say, buffer versus base estimate just to understand. I mean, is a gap between what you had in the previous accounting norms and what you have under IFRS 17, so you are reducing your amount of reserves by roughly 600 million something. So just wanted to waste buffer within that or not? And related to this, maybe you mentioned that obviously, of course, is unchanged, but could it lead you maybe to change a little bit your reinsurance program going forward or not?

Yes. I would take the -- okay. So in terms of -- if we look at the difference between the 2 norms -- you're right, under IFRS 17, we are now calculating our reserves based on best estimate and risk adjustment. Okay. And...

What do you mean...

And the total amount that you're seeing is totaling the 2 items of the 2 components. Does that answer your question? And if you...

Yes.

So this is the 1.4. Basically, in the 1.4 you have more than this, you have we have premium reserves and of course, you have claims with us. It's a leading and a claims reserve is made of best estimate and risk adjustments...

And you do not disclose the amount of risk adjustment?

Well, what we can only -- or it will be disclosed, of course, as we -- when we go with a Q1 conditions, but what we can tell you is the present tile that we have retained for risk adjustment is above 85%.

Yes.

Then I go back to your second question related to the reinsurance, I would say, so far, there's no reason why we're changing any structure or insurance.

if you can say that. I mean you still believe that you'll be able to build some buffer when needed and maybe release some reserves also, as you can -- as you made in the past?

Well, the weather release would be, again, come back to my grades questions related to the development that we see over the course of the quarters. But I think what we should probably look at is the -- this is why I have highlighted the level of the new vintage opening reserves. On the '17 and on the 4, we can see that for full year 2022, it is fully similar and then in '23 onwards, of course, we will see the development of this vintage...

Your next question comes from the line of Michael Huttner from Berenberg.

I have 2 extra questions. So the profit is lower in 2022 and I guess, also in 2023 because only part of the EUR 91 million difference has been kind of opening adjustments has been booked or not booked or whatever. So -- and of course, the payout ratio, as you said, in 2022 would have been under the new accounting, 94% rather than the 80% minimum target that you have. So my question is, should I also cut my dividend given that my understanding of your payout policy is that -- because of the strong growth and inflation and all these other factors, you need to keep a little bit more of earnings back. So if earnings are lower, and I apply 80%, then my dividend would be lower. So that's my first question. And then the second one is you said that the volatility would be higher. And I just wondered if you could explain that a little bit. Many of your peers have said the same, but I don't understand why.

Okay. So a couple of things in terms of 2023 dividend, I think we stick to our target, which is 80% payout. This is what we have promised to the market. Then that it really depends -- it's not only the earnings per se. We also have the solvency ratio. You can see our solvency pretty high and it also needs to be looked at this length when we are paying -- we're coming up with a proposal of dividend. Then if we look at the volatility, of course, you can -- the volatility is really coming from the fact that we have now to discount our reserve and now discounting reserve, you have -- you're using a yield curve which is completely independent to your business. And this will drive volatility from one quarter to another.

Can you explain that in as much detail as you can? And I'm not -- this is really a mystery to me because I've seen so many presentations now. And on this topic, they're all different. So my understanding is that you have 2 mechanisms. The first is you discount the back book reserves, what I call back books that you're opening balance sheet. And that's the -- and the second is, of course, you discount the new reserves, the 80% or whatever. And so I understand from what you said is if the interest rate changes, the discounting of the new reserve changes, so if interest rates go up, so you -- it's no longer 77, it might be 76 or 75, whatever. But what happens to the original discounting that you did in the opening balance sheet with the original interest rate. Do you have to re-measure that? And where does that difference go?

Yes. Of course, it goes, you have to re-measure it as well as you have to add mark-to-market, I would put it this way, the old vintage as well.

And where does that difference on the market-to-market difference go? Does it go to a CI? Or does it cost...

Yes. Part of that will go to CI that you're totally right, okay? And part of that will go to P&L and the P&L side, of course, will drive some volatility.

And how do I know? How much -- how much goes to P&L and how much CI?

I think you see that, I think in Q1, this will be disclosed in our full P&L. Of course, you have a line which is IFR which is interest component -- discounting component. But it's not in the loss line, it would be, as you might know in the financial income line. So it's not even booked at the same level.

We will take our next question, and the question comes from the line of Thomas Fossard from HSBC.

A couple of questions on my side. The first one would be on the quarterly results, apart from volatility, would you highlight in the seasonality impact due to, I don't know, the way you are renewing your book, which is very much clear towards the first quarter of the year. So does that create any seasonality in terms of quarterly numbers when you think about the modeling? That would be the first question. The second question will be related to your mark-to-market of the real estate funds through the P&L. So you've got approximately EUR 200 million of exposure to real estate. If I were to say actually, real estate funds went down by roughly 20%, 25% in the past 6 months. Should I take EUR 40 million of the expected results for 2023? Or I mean...

I hope not. I think I will answer on this one because we have -- if we look at the -- if we have a -- I would say, a good portfolio in terms of real estate fund, which is when we look at real estate investments, we look at which sectors with geography, we are pub mainly investors in prime real estate.

But if I'm looking at, for example, listed real estate funds or if I'm looking at REITs, for example, which are listed, they all significantly done. I mean basically, yes, I understand there is a significant discount to their NAV. But I mean, if you invested in listed funds, which are heavily done, this is a case, why should it not be reflected into negative mark-to-market adjustments in your P&L account?

Well, it will. But of course, the mark-to-market will be in my P&L accounts. So this is -- this will be recognized going forward. But again, I think that we have a good quality in terms of investments, so...

Okay. Okay. Okay. And on the seasonality impact due to the way you're renewing your book skewed towards the...

I think it's a little bit too early to say. But in terms of totality, I would say that it's almost really the case today on the IFRS 4. So not sure why it would change much on the IFRS 17.

Okay. Okay. Okay. Okay. And on the quarterly figures you reported on Slide 9 and the IFRS 17, you're indicating that actually some of the volatility was due to the first time application. So would that mean that actually you're expecting a lower -- I mean stripping out the first-time application, which will not be a recurring effect into 2023 and 2024. So would that imply that actually the volatility will be lower than the one you're showing on a quarterly basis for 2022?

Our core data isn't way well. And I think going forward in 2023, our reserving, of course, reflect the level of claims that we are seeing in our wheel portfolio. So the volatility will come partly to the fact that this interest rate discount effect will slightly change from one quarter to another. But the underlying business per se and the plane per se, will really be reflected. If you look at this Page 9, you can see that in Q3, this is where we have booked up the reserve on Russia. This is the peak that we see. Same in Q4, the book up reserve on the Latin America large claims. What I think we'll tell you this is that in Q3 under IFRS 4, the reserve release, which is the 10 point has happened there, but if you have 2 movements, the book-up is the same, both sides. And then under IFRS 4, you have the reserve release related to prior years that has been probably taken more in Q3 than in the other quarters. That's all it says. So in terms of volatility, that now, I would say, particular wisdom that is changing between the 2 standards.

Okay. But still on a quarterly basis, the incurred claims in the quarter does matter a bit more the IFRS 17 basically, the timing of it when you incur the claims.

Yes.

Right. Okay. And last question for me, [indiscernible] in your preliminary comments, you are indicating that -- and that's only on Slide 4 that actually under IFRS 17, there is a faster profit recognition.

Exactly.

So does that mean that everything being equals, year 2023 and 2024 number should be higher than the one we were expecting on the IFRS 4 if there is faster recognition of the profit? So I mean, if the margins recognition is more front-loaded under IFRS 17 than under IFRS 4 that should imply that actually our numbers would go up.

Well, I cannot tell you when we happen in '23 because it all depends on how the developments on the prior years will go. But what we can say is that, yes, there will be a faster recognition of prior year development, boni or non-boni actually positive or negative on the IFRS 17. So if -- in '23, we are already entering into '17 words. So it will be very difficult for me to say what we will do in '23 under IFRS 4. It doesn't exist anymore. But from what we've seen in this presentation, and this is why we have this first time application is that for the prior year development pattern under IFRS 17, we would have recognized the sales earlier than in IFRS 4.

And to your comment that actually this is a faster recognition of the profit?

Yes.

You have a follow-up question from the line of Michael Huttner from Berenberg.

It's just my last. So the -- you've very kindly talked about the impact on the geographical loss ratio in Slide 10 and Slide 11. And I'm afraid I got confused. And I just wondered if -- just on one of the maybe either Central Europe, which seems to vary a bit. You can explain a little bit more what happened quarter-by-quarter. That would be really, really helpful. I think you mentioned it with Russia and the reserve releases, but I got a bit confused... And then... Yes, and that was my question.

Okay. Let's go to -- I think the Russia case was in a page on Page 11, I just want to see to list the fact that we have exactly the same pattern in terms of reserving between IFRS 17 and IFRS 4, where we [indiscernible] had exactly the same story on Central Europe. In Q1, we have booked up reserves when the war happened. And you can see that we have booked up those reserves as well on the IFRS 17. We have released reserve in Q2 related to Russia, and we allocated back to the region where the risk has been underwritten, which is -- I think in Q2 was much more on Northern Europe, for instance, or Mid-Africa. Then in Q3, I think the 29th of September, you might have recollect that Russia has announced the partial mobilization. And at that time, we have booked up again on the IFRS 4, and you can see the same trend on the IFRS 17, some reserve in Central Europe. And in Q4, we had time to reallocate it back to the region where the risk has been underwritten. So I'm just this illustrating that we have -- the figures of is the same. The way that we are reserving in the region is the same way under the 2... Yes.

So the reinsurers on the IFS 4 got 146 million in the IFRS 17, they get 138. Is that real money? Do they get less under IFRS 17?

No. I think, again, this is just illustrating the fact that on the IFRS 4, as we're leasing more reserves, of course, mechanically, it benefits to the reinsurers. That's all. But if you look at page -- this is where I will put you on Page 14, on the net combined ratio, again, the same way we have taken into the FDA impact, the net of reinsurance side. So what they don't get in the [indiscernible], they get it through the opening balance sheet.

You have a follow-up question from the line of Benoit Valleaux from ODDO BHF.

My last question just on my side. Just a confirmation, please. In the past, you had a kind of seasonal effect regarding revenue and results coming from business services -- is this the case on the IFRS 17? Or could there be some change in seasonality effect? On services?

So this is?

In business. You got the amount of fees in Q1, Q2, I mean at the beginning of the year, and therefore, lower contribution in the next quarters.

Well, I think it's not on the business information business is not at all impacted by IFRS 17. Is that your question, -- this is why I was a little bit surprised by your question.

Yes, but no in terms of recoveries. I mean, all these kind of businesses when you got fees? I hope you mine manage to explained that you have a kind of positive or let's seasonality, positive effect at the beginning of the year. Just wonder if it is the case on IFRS 17 or.

You mean the insurance fees or the noninsurance fees?

Insurance fee. Yes, I mean, as I mentioned, recoveries, for example, or these kind of things.

I think in terms of fees, revenue recognition is not impacted by IFRS 17. But I think the underlying business is not either impacted. So I'm sorry, I missed your question again, but I would take it...

There seems to be no further questions. I would like to hand back to Phalla Gervais for closing remarks.

So if we don't have further questions away, thank you for attending this call. I hope that is clear and it will help you in your quarterly processes. Thank you all. Have a nice evening and talk to you soon in 3 weeks' time.

This concludes today's conference call. Thank you for participating. You may now disconnect.