Moltiply Group S.p.A. (MNL.F) Earnings Call Transcript & Summary

Good afternoon. This is the conference call operator. Welcome, and thank you for joining the presentation of Moltiply Group First Half 2025 Results Conference Call. [Operator Instructions] At this time, I would like to turn the conference over to Mr. Marco Pescarmona, Chairman; Mr. Alessandro Fracassi, CEO; and Mr. Francesco Masciandaro, CFO. Please go ahead.

Thank you, and welcome, everybody. This is Marco. We will rely as usual on the document on -- present on our website and we will jump straight to Page 18 of the presentation with the H1 highlights. And in the first half of 2025, the group posted revenues of EUR 301 million -- EUR 302 million, that's up around 40% year-on-year, and this counts for 55% from Mavriq from the Broking division and 45% from the Moltiply BPO & Tech division. In terms of EBITDA, it's EUR 77 million in the first half, which is up 37.5% year-on-year, and it comes for 62% from Mavriq and 38% from the BPO part. EBIT is EUR 49 million, that's up 50% year-on-year and this comes for 68% from Mavriq and 32% from BPO. In terms of the faster development of the EBIT compared to the EBITDA, this is due to the fact that we finished amortizing some of the PPA coming from the 2,000 acquisitions. And we have not started yet. We will do it at the end of the year. So it will be visible, I think, in the first -- in the last quarter. We have not started yet amortizing the assets from the PPA of Verivox, which will anyway sizable given the size of the transaction. But again, we think it's better to understand the current performance, the actual performance of the business to focus more on the EBITDA because of this or at least normalize the EBIT for the PPA effect. The net income is EUR 22 million in the first half of '25, which is up 7% compared to 24% and the reason why the net income doesn't follow the same strong growth of the other figures is that, first of all, we have higher interest also because when we closed -- when we refinanced the old debt, we had to recognize some costs linked to the closing of the old debt. And also, more importantly, we have almost EUR 10 million of adjustments of put and call liabilities for minorities on some businesses that are performing quite well. One of this is, for instance, Switcho, but there are other things, and these are recognized in the income statement. The only case when we were not recognizing in the income statement, the adjustment of the put and call liabilities was Lercari, which had a different treatment. Now all the future put and call liabilities, you see them through the income statement. Looking at the quarter alone on Page 19. In the quarter, revenues are EUR 169 million, up 55% year-on-year. That's 58% from Mavriq and 42% from Moltiply BPO & Tech. Here, you remember, we are consolidating Verivox from an income point of view from the beginning of Q2. So this benefits also of the consolidation of Verivox. EBITDA is EUR 42 million in the second quarter, that's up 46% year-on-year. And it comes 64% from Mavriq and 36% from Moltiply BPO & Tech. EBIT is up 60%. It's same explanation as before of why -- as to why it's growing faster than EBITDA. And then net income also same story. EBITDA adjustment of the put and call the calculations with the half year when we knew we were able to better assess the performance of our businesses. Looking at Mavriq alone, Mavriq -- on Page 20. Mavriq in the first half as revenues of EUR 165 million, that's up 60% year-on-year. EBITDA of EUR 48 million. That's up 62% year-on-year, and EBIT of EUR 33 million, up 75% year-on-year. You see the margins in the first half are expanding in terms of EBITDA. Then if we look at the second quarter, we see it's -- the margins are instead declining because we are consolidating Verivox, which is quite large and has margins that are significantly lower than our average. In the second quarter of the year, we have revenues of EUR 98 million, that's up 88% year-on-year and an EBITDA of EUR 27 million from Mavriq that's up 79%. And you see the EBITDA margin is 27.4% in Q2 '25 compared to 28.8% in Q2 '24 and so in the first half, the EBITDA margin was expanding, in the second quarter it's declining. This is basically the impact of the dilution of Verivox and we are super happy of the Verivox acquisition. The only bit is that we were starting to show, as we have explained many times an expansion of the EBITDA margins of Mavriq on an organic basis, and this will be masked by this big dilution from Verivox. And I would say this is what we have to say in terms of figures, in terms of comments on outlook, well, we had very, very strong growth in the first half, both organic and also because of the consolidation of Verivox. In terms of comments on the different businesses in Italy, we had credit broking that continued to grow. Growth was quite good also in Q2, but slower than in Q1 as the mortgage market basically peaked at the beginning of '25 and then it's still growing, but slower. Also the remortgages where the peak demand for remortgages was also concentrated at the beginning of the year. Insurance broking is a much more stable business. So it continues to grow in a very nice way, but at a steady rate. And it's been growing this year also with stable insurance premiums, so without inflation in prices of insurance. Telco and energy comparison. Here, we had a good performance, strong performance, I would say. But it's benefiting, this is mainly because of the acquisition of Switcho, which was done at the end of Q2 of '24, so until now, it is -- we are benefiting from the change in the consolidation area. It's not only that we acquire Switcho, but also that we were able to combine it with our existing energy business -- Energy & Telco business and this, of course, had some synergies. Finally, e-commerce price comparison, which is the area where we have been suffering a little bit or at least feeling some pressure is stable year-on-year in terms of revenues with margins slightly down. And no particular news. We continue to have some pressure on the results, not catastrophic as we have seen from the comments, but pressure, yes. And we are looking forward to a decision by the European Commission on the enforcement of the DMA, which will hopefully open up the market and translate into growth for us. In terms of our international businesses, we have strong growth on a like-for-like basis, driven by Spain and also by France. Netherlands is doing okay, but it's mainly in Energy business and Energy in the Netherlands as well as in Germany is not being a strong market. Here, one update is that we have a new local COO and that will allow us to move faster and do more things, and it's -- we think we have a good opportunity there as well. In Q2 '25, we acquired -- actually a few days earlier, but we acquired Verivox and thanks to this acquisition now, more than half of Mavriq's revenues are from outside of Italy, and the outlook is that weight of the international part will possibly even increase as things evolve. Now also, if we look at the group, especially if you look at the foreign business, the foreign business is mainly insurance and energy. We are doing credit, but we are not doing too much in terms of credit in international markets. Of course, we will develop that as well. But -- so it's basically, our Mavriq division is very diversified, both now in terms of geographies and products. And we think that's positive because we can go through all sorts of markets with good stability overall. And in Germany, with Verivox, Verivox had, let's say, solid performance, but weaker than -- so we were down compared to the previous year. And this is in a situation or because of a weaker energy market, even if we start to see some improvements very recently. By the way, we had in our Q1 estimates, and earn-out liability of EUR 15 million for Verivox. We reduced it to EUR 10 million in our current estimates. And of course, this could be a different figure because it's very difficult to know now for a business that has a peak season that is very concentrated in the last quarter, what the end of year results will be. But so far, this is the adjustment that we have made. We have started working together with the team in Verivox. This is really the priority of Mavriq. And we think we are happy of the acquisition. We think it's a very interesting market, and we are working. Finally, in terms of outlook, basically, we have had a big change from Q1 to Q2 in the consolidation area and the most reasonable outlook for now is the following quarters will be a continuation of what we have seen in Q2 in terms of numbers, let's say, of course, adjusting for seasonality. So for instance, typically, Q4 is a stronger quarter, possibly significantly stronger than Q3, but it's easier to make a comparison now and start looking at things and building expectations based on what we have done than on the previous year because of this big change in the consolidation area. And with this, we are finished with the comments on Mavriq and I hand it over to Alessandro for BPO & Tech.

Yes. Thank you, Marco, and good afternoon, everyone, or good morning, if you are on the other side of the ocean. And we are on Page 23. So our H1 financials for the Moltiply BPO & Tech divisions we saw a growth of year-on-year in the first half of 22.6%, growing from EUR 111.4 million of last year to EUR 136.7 million of this year. The EBITDA growth is 10%, raising from EUR 26.5 million last year to EUR 29.1 million in 2025. There was a slight reduction in the EBITDA margin in terms of percentage from 23.7% to 21.3%. This is mainly because if we have to comment the difference in the growth of revenues, which, as you can see, it's very strong and in EBITDA, which, again, I believe it's still strong, but it's half of it. And it's due to the dilution effect on percentage margin that we have for the growth of the paranotary business connected to the refinancing. I have always been commenting about this in the past, but as I mentioned at the beginning, basically at the end of last year, it is currently not only we have the mix effect, but here, we have a stronger effect because there is also a price effect on the notaries because of the new law on the so-called for compensation. So the average ticket of refinancing for -- in terms of notary fees has grown very significantly, while our margin on the fees that we charge to the banks has remained in euro terms, stable. So that has basically created a dilution effect on our margin. Again, these are all variable costs. So there is nothing bad about this. No negative read in this factor. But obviously, this mix effect now is particularly significant as the mortgage business and the paranotary mortgage business is also growing very fast. For the EBIT, we have a growth from 13.5 million to EUR 15.5 million. That's a 15.4% growth rate year-on-year. Here, the growth -- and the EBIT margin also shrinks a little bit from 12.1% to 11.4%. The growth here is stronger at an EBIT level than an EBITDA level for the same reasons that Marco commented at a group level that some of the acquisition of 2020 are now getting finished their amortization phase. So if we now focus on Page 24 on the Q2 financials, we see, again, a growth in revenues that is a strong 25.3%, moving from EUR 56.3 million to EUR 70.5 million in Q2 2025 relative to the year before. The EBITDA shows a growth of 9% growing from EUR 13.8 million to EUR 15 million. The margins, again, for the reason that I commented on the semester results shrink a little bit from 24.4% to 21.3%. Again, no negative news here in our opinion. The EBIT grows from EUR 7 million to EUR 8.1 million in the quarter. That's a 15.5% growth year-on-year. And in terms of marginality at the net income level, we basically have the same numbers that we have in the semester level. So we go down from 12.5% to 11.5% in terms of percentage margin. So now that closes basically the quantitative part. Let's get into the comments. Again, I believe these are very positive results. We have achieved a very strong growth in revenues compared to last year. And also, the EBITDA remains in double-digit growth. And this is mainly -- more than 70% of the growth is organic. We -- the part of the growth that is not organic is coming from our business connected to services in the retirement part, the Mia Pensione, the one that we acquired at the end of 2024. And then there is also a little bit from the acquisition of Evolve, but a very, very small EBITDA contribution. The performance is driven basically by mortgages, lease, which is always a strong contribution and also the wealth business line. Instead, obviously, the claims and the real estate have showed as expected, a gradual normalization after extraordinary peaks of the previous year. If we -- let me just give the outlook for the overall in the division. Obviously, there will still be various trends and differentiated trends among the business lines. But I think it's reasonable to expect that this -- the positive performance year-on-year that we saw in the first semester will continue also in the second half of the year. Obviously, net of usual seasonal factors. So the presence of August, for example, in the second -- in the third quarter. So let's dig a little bit more into the different business lines. Moltiply mortgages is obviously one of the growth engines, and this will continue. We have reasonable visibility on the fact that it's continuing. Obviously, it will not be forever, but let's say, for the rest of -- for the remainder of the year. And here, the growth is both in the paranotary business. I've already commented a lot, but also in some of the new clients that we acquired in previous year, now they are becoming more and more confident in extending credit also through our support. And so they are increasing their budgets and also the results. So we remain positive here. Instead, for Moltiply real estate, we have seen, obviously, at this point at the end of the Ecobonus incentives, and we are seeing a growing business in the real estate valuations also because we acquired new customers and the market is growing, but this growth will not be sufficient to compensate the other effect. But the second part of the year will be better than the first part. In the real estate business, we also report the business that it's cadastral services NPL servicer and this part also instead is, let's say, cyclically weak, and so it's not contributing positively to the growth of this business line. Considering this, we are both looking at new forms of growth, new services that we can offer here, but we are also looking at some opportunities to rationalize the cost base. We have already started, will continue in the remaining part of the year. We're not worried, but it's good discipline to obviously have the costs and revenue trends try to go in parallel. Moltiply loans had a good performance. Actually, it's stable in terms of revenues, but margins are actually a little increasing. Because of some mix effects, we are substituting contracts that had lower margin with contracts that have better margins. So I actually expect this to be even better in the second part of the year. But anyway, substantial stability for Moltiply loans. Claims, we said that, obviously, both revenues and margins are decreasing. You remember that at the end of '24, we also had the closing of a lot of complex and complex claim processing, which also have very high margins. That thing is not there anymore. But we expect to have results both in terms of revenues and in terms of EBITDA higher than the one we've done in 2023. So there is a peak in '24, but there is an underlying growth trend, and we continue to invest in efficiency of the processes and speed of the processes also because we expect next year to start seeing some of the impacts of the comps or insurance for businesses in relation to natural events. So that will bring -- we do expect to have more business. There is a little of constraint, and we have seen that when there was a peak in terms of appraisers and loss adjusters in the market. So it's important to be able to increase capacity. This is a variable cost capacity. So there is no problem in getting ready for it. And that's one of the things that we are concentrating on. Moltiply wealth has a nice growth, and that will continue for at least the next semester. This is also connected, obviously, to good performance in the market, but also to a large IT project that we are doing for one of our main clients, and that will also go beyond 2025. Having said that, it's also possible -- so it's a positive outlook for the next semester. We will also understand in this semester, what will be the impact on our business, on the announcement of the project that it's called the new bank that Azimut, which is our most important client in this business line as announced. As you know, with the -- in partnership with the FSI fund, and it was announced at the end of May and we still are evaluating timing and nature of the impact of this process. And actually, we still need to understand how this thing will go on. But we'll keep you posted. Finally, Moltiply lease confirms as a very stable engine of growth contribution and revenues and EBITDA were high in H1, and we expect that to go on also in H2. By the way, we will have to focus a lot during H2 in efforts for next year as there are some new regulatory changes in the way that the -- basically taxes -- circulation taxes what in Italian we call bollo auto are going to be managed. There are going to be differences between new cars and old cars and existing cars on how this thing gets taxed in terms of timing. So a lot of complexity and each time hear the word complexity in these businesses is actually good news for us because it means that we will have to deal with this complexity on behalf of our clients, and obviously, we get rewarded for managing this. I think this more or less ends my comments. And obviously, then I'll be available for questions. But before that, let me hand it back to Marco for comments on the financial positions.

Thank you, Ale. So on Page 28, we have the net financial position. I would say, no particular surprise here. The reported net financial position is negative EUR 467 million. And basically, that's an improvement compared to the negative EUR 515 million at the end of the first quarter when we had already acquired and paid for Verivox. The improvement is the effect of cash generation of the business and the fact that we did an accelerated book building for 1 million shares that many of you have seen. And the proceeds of which are fully available, in particular, for M&A, which is positive effect under our current full financing contract. The other positive aspect of the current net financial position is that under our financing terms, basically, we go from a spread on the loan -- on the margin on the loan of 2.45% to 1.95% from the second half of the year. So we've been able to move to a much more favorable bracket in terms of cost of financing. Then we still have the same participation in many and so net of that, we are at EUR 353 million. Just as a reminder, in the past, we had loan contracts, which consider the many shares equivalent as an adjustment as a cash-like adjustment to the net financial position. This is not the case of the pool financing, but of course, the parameters are significantly wider. And so we are ready with the new setup as well. And something that is going to have a little bit of impact much of it expected to the net financial position is also the fact -- this is linked to BPO, maybe Alessandro will want to add something. But basically, we acquired a 40% minority in our company that was providing notary services. We paid EUR 8 million in July, and we pay -- but we already acquired it -- there is a deferred payment of EUR 7 million in July of next year. And this was not expected because this was not contractual. This was just negotiated. But -- so we now own 100% of the company. And you see, of course, the cash outflow in the third quarter. And secondly, in the third quarter, and this was said expected and well known. Basically, we acquired 37.9% of Lercari. You remember Lercari is company that does insurance claims where we have the founding family of 49% with put and call on the 49%. Instead of buying 49%, we ended up buying 38%, 37.9%. And the family still keeps and we keep for another 3 years there at 12% that we will acquire at the end of that period. And basically, some of the outflow plus the future liability is -- I mean it's equal to what we were expecting to pay already at the end of Q1 ballpark. So you...

Yes. Just let me comment briefly business-wise. These are both good news. And especially the second part with the Lercari family, we were able to relaunch to the partnership. We are happy to help them still as shareholders and involved in running the company. This has been a very positive partnership in the end, a very successful investment. So we are happy that this has been the outcome, and we look for further growth in an area, as I mentioned before, claims where the opportunities are there, and they are secular opportunities given the regulatory changes that are more interesting than just commercial opportunities in other cases. Thanks, Marco.

Thank you. And I think this ends our presentation, so we can open the floor to questions.

[Operator Instructions] The first question comes from the line of Gabriele Venturi from Banca Akros.

First one, if you could please clarify what Mavriq growth rate would have been without Verivox consolidation. And if you could please clarify -- if you can give us an update on when do you expect synergies and efficiencies of Verivox to get to regime? Second one, in July and in August, we witnessed a strong mortgage market in Italy. So if you can give us some more color if you are seeing a strong momentum also for the third quarter for your business? Last one, if you can give us more color on your Google lawsuit situation given the recent news flow about Google.

Okay. Thank you, Gabriele. I will address the first and the third, and then we will comment together I and Alex on the mortgage market. Now Mavriq, I would say, well, in the first half report, we disclosed that Verivox did EUR 34 million of revenues in the second quarter. Keep in mind that this is a business with significant seasonality with first and fourth quarter significantly stronger than second and third. But so in terms of revenues, EUR 34 million comes from Verivox. In terms of EBITDA I -- what I could say is what we said it's weaker, you could make your estimates. But I would say, if you look at the EBITDA growth of the first quarter, which didn't have Verivox, you could expect, based on our comments, the second quarter still had significant -- very significant EBITDA growth, but lower than in the first quarter. And in terms of when we will see the impact of the synergies and so on? Well, first of all, let's say, this is something that takes time. The contribution to performance of Verivox will change by quarter-by-quarter, based on the seasonality. So like expect normally a strong fourth quarter by market dynamics. So it depends on how the energy market is doing in Germany, especially if people can save by switching energy contracts. It is looking now better, especially in gas than a few months ago. And finally, it depends on how we operate the company, if we are able to transfer best practices if we are able to, I don't know, improve the efficiency or effectiveness of the company. This last part will take time. So there are many things which we are working. We think it's quite promising, but it's not something we are able to do in a few months. It's the same that in Spain and France, we are still working. So I think you'll start seeing things -- for this year, you might see a bit of the benefits but not too much of our interventions because it takes time. And -- but you might see benefits of seasonality of market dynamics, if we are lucky. Next year, progressively, you start seeing also, hopefully, the impact of the work that we are doing. So it takes patience, I would say, to see a good portion of potential improvement, at least the short-term potential improvement you have to wait for a couple of years. And the Google lawsuit, basically, we filed. And then this is in front of an Italian court -- and there are updates, I mean this will have the normal development. We'll meet in front of a judge, we'll exchange memos. And this will have a lot of steps before we have any visibility of a potential outcome, and we will not be able to comment on these steps because it's very difficult to read anything into them. And until we have a decision, we will not be able to say anything and the decision should come normally in an Italian court for something like this in 2, 3 years. But we have to see. We will know better the timing when this gets up to speed. So no news and don't expect any news at least from a court for a couple of years, I would say. In terms of the mortgage market, well, our view is that it peaked in -- but also the market info that we have seen is that it peaked in Q1, and then remains strong, but slowing down in the following months. July is normally also for us a strong month. But I don't know what data you are referring to because I don't think there is anything public, at least regarding July and August. So I don't know if Alessandro has anything to add. But by the way, July is when people close mortgages normally.

Yes. Well, let's say, apart from the news, we had given a positive outlook on mortgages, and we do it not only based on our numbers, but also based on our early signals. So obviously, it takes time before a mortgage closes. And as just -- so we see some weeks, if not months in advance, depending on the different businesses. And it's true that normally July and, let's say, the first days of August tend to be strong in terms of closings, you can expect, for example, people to move especially during the months where it will not rain on your moving. So that's why there is normally a stronger performance in those months. But again, we continue to be positive and -- at least until the end of the year.

[Operator Instructions] The next question comes from the line of Aleksandra Arsova from Equita.

So four questions on my end. The first one is a follow-up on the Google lawsuit and also the compliance with the DMA. So I know it's a separate thing, the two things are separate, but if, let's say, that the remedy is proposed by the European Commission work and Google complies, do you see that the situation will increase or decrease the likelihood for you to win the lawsuit against Google and get some, let's say, damages in terms of, I mean, monetary compensation. The second one is maybe a more broader question on the risk on your e-commerce business coming, not more -- not anymore from -- no longer from Google, but rather from the new artificial intelligence chatbots like ChatGPT or others, I was reading that maybe in the U.S., people are starting to search for e-commerce products, not more on Google, but directly on ChatGPT or other chatbots like ChatGPT. Do you see it as a risk for Trovaprezzi? Or do you see some measures that you can take to avoid this risk or to maybe make more money from this? Then the third one is on insurance premium in the insurance broking business. So I was wondering the growth there was strong also in the second quarter. So just any sentiment on any kind of normalization of the growth in insurance premium in the third quarter or beyond? And the last one is maybe on the fees you got from -- you get from -- especially from banks in mortgages, both in BPO, but also in the broking business in Mavriq. So now that mortgages are definitely picking up, recovering and banks are in a very, let's say, good shape. Do you see any room for renegotiating upward the fees you get from the banks?

Okay. Thank you, Aleksandra. Let's start in order. So Google, let's -- if with the adjustments that Google will have to do with -- to comply with the DMA, we start making much more revenue and much more money in terms of profits, then potentially, yes, this could have an impact. But this is a market where Google Shopping probably has more than 90% -- well above 90% market share. And so I think that even with the -- this is a market also with a lot of network effects. So even with very good remedies, I think it will be very difficult to go back to a situation resembling the original situation when the business started. So I would say, yes, the more -- the bigger the improvement, if there is any to the market now with hopefully, compliance of Google with the new regulation, then potentially the lower the final component of the damages, let's say, but again, because of structural reasons, I don't think it's likely to completely revert. There will be an improvement, but Google will remain because of the network effects structurally dominant. So it will be an adjustment not a change of paradigm. In terms of second question e-commerce and Gen AI. It's true that, particular ChatGPT, but others as well are working and then getting a lot of attention to e-commerce. So -- by the way, I would say, chat bots are actually quite good or at least make a better job than us in many cases at recommending choices. So if you want to know, if you want to -- if you need a 5-kilo or a 7-kilo dishwasher, sorry, laundry machine, and then if you wanted, Class A, Class B, Class C, whatever, ChatGPT can give you good advice. So choosing the type of product and possibly the specific product that you want is something that I see these chat bots being able to do better and better. What is different is, however, once you have selected a specific product, finding the very best offer and the very best offer, by the way, from a reliable provider, someone that will not -- that will deliver, that will accept returns for real and so on. So this is something that I don't see ChatGPT and others able to handle for a long while. So I think that we could see more traffic from ChatGPT and the likes and reduce our dependency bit from Google. But we are still at the beginning of this. And there is, of course, a risk also threat of ChatGPT trying to -- you have read about that in the U.S., they tried to strike agreements not even with e-commerce merchants, but directly with the producers of the products, so that you compare et cetera and then you buy directly from Nikon or from Apple or from whatever the provider is of that product. So of course, that will be a threat, but it's -- I think it's a stretch, especially in a market like Italy or where you have lots of small providers, able to source at very attractive prices and where you have the issue of guaranteeing the reliability of those providers, I think we still have an important role. And again, the diversification of our customer of our traffic could be the benefit. But it's early to say the traffic we're getting from Gen AI is visible, but it's small. Insurance premium, they are now stable. They've already been stable...

Marco -- sorry if I can comment also 1 second on this. And also, we really don't know what the behavior of the providers of the Gen AI will be in terms of tweaking what the Gen AI does. I mean the whole Google thing comes from the fact that at a certain point, Google from trying not to be evil and giving the best possible results started tweaking the results and putting new things in there and changing the model. So as of today, Gen AI is trying to deliver the best and controlling the hallucination. But once they start thinking about how to monetize maybe these things, then the behavior is really unpredictable because it will depend on what kind of path they will go under. So not necessarily, they will do what is the best thing for the product and they might go and diverge into something different, leveraging the fact that people believe that this is the best thing that they -- that the best intelligent answer to the question. We really don't know.

And also the work of Trovaprezzi -- the job of Trovaprezzi is a very difficult job even if it doesn't look like from the outside, but assembling the catalogs of thousands of merchants and just figuring out what is what and if it's the same product, a different product, a different size, different whatever, that's already very difficult. So it's unlikely that this will be doable with the generic approach. And it's a really country-specific, business-specific with a lot of intelligence and customization, merchant-specific and so on. So back third question, insurance premium. We have started seeing stability this year in general, at least in the countries that they have at the top of my mind and we have been able to grow despite no longer having the impact of the premium inflation. And so I think that the outlook is we will continue to grow. Of course, maybe this year, we are growing but maybe a bit less than last year just because we don't have the premium inflation, but we have been able to grow in that environment, and we continue to be able, we think, to grow in that kind of environment. In terms of fees from banks, I would say, in general, banks are printing money across the board. So hopefully, we should be able to -- we've been able to do this a little bit already, but we should be able to improve our fees or at least try to improve our fees. And hopefully, this is -- I don't know if this is happening, but maybe even for BPO, it will be easier because they're really printing money and so complaining and trying to negotiate things when we are making that much money is a bit awkward. So I think we are all happy that our clients are not under pressure, that they are doing very well, and we'll try to help them to grow and be effective with their operations.

[Operator Instructions] We now have a question from the line of Tommaso Nieddu from Kepler Chevreux.

Most of them were already answered, but I would have 2. The first one is on the performance of Spain and France. So if you -- can you please give us more color on that? And what was the contribution on the margin expansion? And the second one, if I may, and sorry, maybe we already talked about it in the last conference call. But can you repeat us what are your thoughts on AI, in particular, on mortgage BPO, if banks could start thinking about or internalize the origination? Is it something you are concerned about or maybe perhaps this automatization of the process is just an upside for you?

Thank you. I'll start with Spain and France. Well, we do not disclose our performance for the individual countries. In Spain, we filed public annual report, but that's at the end of the year. And in France, there is no obligation in such a sense, so we don't do it. We don't do it in Germany -- we will not do it in Germany either. So it's hard to say. To give a precise feedback, I would say Spain is the bigger of the 2 and -- where I would see the strongest performance. We don't know if it's the market, certainly, it's helping the work that we have done. So we see an improvement both in terms of top and bottom line, I would say. But we cannot go into much detail because we don't give that kind of disclosure. Alessandro, if you want to pick up the last one.

Yes. Okay. Well, here let if we look, it really depends on what is the time frame in answering this question. First of all, we see in a lot of our businesses, AI at the same time, has a big opportunity and at the same time as a big threat. Obviously, there are some banks that are doing things with us, let's take out Gen AI for a second. Now our -- what we have been telling banks in the last 15 years is that if you come with us, normally, you would save 50% of your cost base, right? And sometimes we were not able to do -- and still some banks have said no, or even if they said yes, they said they were not able to generate the full advantage of this because to generate a full advantage, they would have to integrate their systems with our systems, so that these advantages could all be generated. So now you have Gen AI, which I understand could generate even larger savings. Still, what are you going to do with the people and how you're going to integrate your systems into these Gen AI models, are you going to do it? So it's really -- theoretically, yes, then practically, first of all, the mortgage underwriting, it's a complex process. Now I will not sit here and tell you that never is going to happen to Gen AI. But let me tell you, I don't think it's going to happen with you dumping all the documents into ChatGPT and telling them, "Tell me -- come up with the underwriting," that will not work, at least not in the foreseeable future. Then you have to get into and really divide in different steps and then do all that. This is what we are working on. This is where we are starting to obtain results. So again, this is the opportunity for us. So obviously, there will be some banks that are going to look at trying to obtain this, but we haven't seen any bank succeeding until now while we are starting to generate some results and sharing, obviously, the advantages that we are able to generate with our clients. That's the rule we are on and what we'll try. Then I don't have a crystal ball on ChatGPT 6 or Cloud 5. So we'll see. But I think everybody that has really tried as we are trying to really scale up beyond proof of concepts has realized that these things are difficult and that the real environment is an environment made of the real ecosystem of IT ecosystem for mortgages, but for a lot of progress, it's an ecosystem where you still have AES 400 mainframes, a patchwork of systems, a very, very complicated environment. So obviously, if you have a completely -- in a theoretical design, well, the old information is just in one system and everything is there available, yes, it's -- I can see that the Gen AI can be very effective. But that is not the ecosystem that is out there. Again, my comment is always, we have built for 20 years of business based on the fact that we were able to digitalize all the documents at the beginning of the process and have the whole workflow in a digital format that helped us being much more effective than people handling paper. And digital documents and scanning documents was much easier to put in place than Gen AI, and it took banks or insurance companies like 15 years to do that. So we'll see what the runway for this thing is. We are working very hard on this, and this is really our top priority.

Gentlemen, there are no more questions at this time.

Okay. Then we thank everybody for participating to our call, and we'll be available as always, for one-to-one's or we'll speak to you at the next opportunity. Thank you.

Thank you very much. Bye-bye, everyone.

Ladies and gentlemen, thank you for joining. The conference is now over. You may now disconnect your telephones.