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

Good afternoon. This is the Chorus Call conference operator. Welcome and thank you for joining the presentation of Gruppo MutuiOnline First Half 2020 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 of Gruppo MutuiOnline. Please go ahead, sir.

Thank you and welcome, everybody, to our conference call. This is Marco Pescarmona. We will rely as usual on the presentation that we published on our company website, and we will start from Page 15 of the presentation with the H1 highlights. And well, first of all, we are very pleased with the results of the first half, especially of the second quarter, because, basically, despite the pandemic situation, which affected Italy very heavily in the first half of 2020 and -- 2020, we were still able to deliver good results. And looking at the revenues, in fact, in the first half 2020, we posted revenues of EUR 120.1 million, which is up 11.2% compared to the EUR 108 million of the first half of 2019. In terms of operating income, in the first half of 2020, we reached EUR 27.6 million, which is up 7.6% year-on-year compared to EUR 25.6 million of the first half of 2019. And this corresponds to an operating income margin of 23%, which compares to the 23.7% of the first half of 2019. In terms of net income, here, the difference comes from the tax line. Basically, we were able to post a net income of EUR 20.4 million, which is substantially flat year-on-year. This corresponds to a net income margin of 17% compared to the 18.9% of the previous year. Looking at what's behind these numbers, we can go to Page 16 and see the different contribution of the 2 divisions. And here, you can see that, basically, the performance was mostly driven by the Broking Division. The Broking Division was, in terms of revenues, up 22.7% year-on-year, reaching EUR 50.6 million of revenues in the first half of 2020, which compares to the EUR 41.3 million of the first half of 2019, while the BPO Division was still growing, revenues of EUR 69.5 million in the first half of 2020, which is up only 4.2% year-on-year compared to the EUR 66.7 million for the first half 2019. Of course, we will comment the detailed performance of the division and how, in the end, derive from the performance of the different business line later. So for now I'll speak to the aggregate figures. The Broking Division, in terms of operating income, posted a big increase in operating income because it recorded an operating income of EUR 18.1 million in the first half of 2020, and that's up 47.3% compared to the EUR 12.3 million of the first half of 2019, whereas the BPO Division posted an operating income of EUR 9.5 million in the first half of 2020, which is down 28.9% year-on-year compared to the EUR 13.4 million of the first half 2019. And one point worth mentioning here, it's not on the fact, but it's a bit -- at least EUR 1 million in the performance of the BPO Division, EUR 1 million -- around EUR 1 million of decline year-on-year just comes from an EBITDA allocation of Eagle & Wise, the company does real estate valuations, that we acquired at the beginning of '19 and the PPA allocation only at the end of '19. So we don't have it for the first half of 2019, the amortization of the intangibles, while you have it for the first half of 2020. So that's around EUR 1 million -- a bit more than EUR 1 million. In terms of operating income margin, we recorded an operating income margin of 35.7% in the first half of 2020 for the Broking Division, which is higher than both the 30.6% of full year '19 and the 29.8% of the first half of 2019, while we had a 13.7% of operating income margin for the BPO Division in the first half of 2020, which compares to the 18.2% of 2019 and the 20% of H1 2019. And by the way, here again, just wanted to remind people that while normally we would consider EBIT as the most appropriate metric, now that in recent years we have done a number of acquisitions that basically create intangible assets when we did the purchase price acquisition, which is then amortized, and very big part of the performance comes from this amortization of intangibles. So in our half year report, in the full year report, we also provided EBITDA metric. Maybe over time, we will also include it in the presentation. Because of this PPA amortization, the EBITDA is sometimes a bit misleading. So this is, overall, the performance of the 2 divisions. On the next chart, you have basically the breakdown of the revenue growth for the different verticals. So I would just say that for the Broking Division, we had a very strong growth of E-Commerce Price Comparison. And in general, growth of both Mortgage Broking and Insurance Broking while Personal Loan Broking was down year-on-year. For BPO, you see, well, overall, that the growth was lower. And because of revenue that came from Mortgage BPO, however, with a slightly different mix within the vertical and from Loans BPO, with Real Estate Services more or less flat and sales for Investment Services and Leasing/Rental and Insurance BPO down year-on-year. Now before going into the details of the different business lines, we provide a comment on the residential mortgage market and possibly on outlook. The -- well, first of all, this year, there is also less available information on the market. For instance, we used to provide information taken from CRIF, the credit bureau, but they've stopped publishing it since -- the start after February. And so the data point is not available. We feel like the asset information -- so we'll try to give a flavor of what the market looks like based on that information. But there is clearly more uncertainty even in terms of what is happening in general. Of course, we know what is happening to us, but what is happening to the market is a bit more difficult to gauge now. Anyway, what is pretty clear is that during the central months of the lockdown, March and April, the market was significantly down, in particular because it was impossible to close the mortgages and also the real estate transactions because the notaries were closed. There were a lot of things that were, if not impossible, extremely difficult to do. So that drove the performance of the central months of the lockdown. When -- basically, you saw a recovery of the market due to -- right after the lockdown due to 2 things. First of all, we continue to see significant volumes of remortgages. So year-on-year, remortgages are clearly up in the first half of the year. And also, probably what happened is that the purchase mortgages that have been blocked due to the lockdown were eventually closed in May or June. And so those months were months in which basically an existing pipeline came to fruition, in a way. So that help -- not only in June we closed the mortgages that were expected, that were due for June, but probably we also closed mortgages that were due or expected for March or April. So that helped the performance in those months. And so May and June were okay. And then what is happening possibly is that, obviously, the economic situation is weak and uncertain. And so there is possibly a contraction in the number of real estate transactions, but that was not immediately visible. It's going to be possibly more visible starting from July once all the things that were already in the pipeline pre-pandemic were closed. And so the new volumes depend on the new demand and not on the previous demand. So this is the type of situation we are seeing. In terms of figures, we see that Assofin say that there was minus 30% year-on-year in March and minus 20% in April in terms of mortgage flows, but this is including the remortgages. So otherwise, it's been even more -- a bigger contraction for purchase mortgages. March was down 36% and April 31%. And then we had growth in May and June. But as I was saying, this growth in May and June is probably, well, again, stronger remortgages because of that remain, but also the fact that in May and June, people closed the mortgages that were supposed to close in March and April. So this growth is not real growth. It's just recovering lost ground. And then in July, the market was flattish. And in fact, in July, purchase mortgages were down 10%. And so it's only remortgages that helped. So this is a description of what happened. By the way, in terms of things resuming after the lockdown, et cetera, I think all parts of the value chain are now operational. The only area -- the only thing that doesn't work completely well at all are the public offices, which are not always fully functional, but this is overall a limited impact. In terms of outlook, well, first of all, there's a lot of uncertainty. It was already difficult to make predictions last quarter, and it will continue to be difficult until the end of the year. And we believe that there will be some contraction of the real estate market, not a dramatic contraction, but some contraction, yes, and also prices that were recovering will probably not stop the growth. So let's say, a double-digit contraction of the residential property market. And also in the first half of this year and still in July, we had a good contribution of -- to the overall market growth of remortgages. But remortgages started being strong in September, October of -- actually in August in terms of demand, in case of closing mortgages a bit later of 2019. So in the second half of the year, the comparison also on remortgages will be very challenging. So that will not be a contribution [ degree ]. A lot of that will be able to be a positive net contribution to the market. So this is how we see the market. Final comment on how we see the market. For now, we don't see a disaster scenario where people are so scared that they postpone any economic decision. Despite a certain degree of uncertainty and concern, many families are continuing to behave more or less in an ordinary way, and they are not panicking. So if they need a house, they buy a house. If they need the car -- the car, I don't know, but if they need something, they buy. This is different from 2012 -- 2011. Back then in 2012, people that wanted to buy something, especially property, they just decided to postpone it and they were really scared. Here, there is, in terms of what we see, let's see, more rationality in the behavior of consumers. So this is the market. On the next page, the Broking Division, the outlook. Well, first of all, our performance, especially if you look at the details of only Q2, was above expectations for the Broking Division. And what helps you a lot was E-Commerce Price Comparison, especially during the lockdown and immediately thereafter, but also the fact that Mortgage Broking and Insurance Broking recovered and actually regained a lot of the lost ground after the lockdown. Now what do we expect overall for the remaining part of the year? We expect some growth, but minor growth as the -- basically the consequence of our expectations for the different business lines that I will explain in a second. In any case, we are -- it's all very uncertain. So depending on how both the health and economic situation evolves, things could change even drastically in both directions, I would say, especially, they could go south, if the economic situation deteriorates. So looking at the specific business lines of the Broking Division, Mortgage Broking. Here, we are quite pleased. Basically, what happened was the same that happened to the overall market. It's really the fact that people that were supposed to close the mortgage in March or April eventually closed it in May or June. And basically, we recovered -- also we recovered something that had been lost possibly in March during the second quarter. Also -- but this is more in terms of what we see in terms of demand. Obviously, there is a delay like 60, 90 days. Possibly, this is very -- it is very early to say this. Possibly, well, certainly during the lockdown and possibly also after the restrictions on the opening hours, that we need to go with an appointment, et cetera, for bank branches, those restrictions possibly have shifted some demand to an online channel like MutuiOnline. So there is a chance that in this situation, we are increasing our market share. And what we expect for the coming months? Well, a lot of uncertainty. We certainly expect a significant slowdown of the growth year-on-year because -- well, mainly because the comparison becomes against a strong second half of 2020. So that's the main reason why we expect a deceleration. And then we have to see, hopefully, our market share increase. If that is confirmed, we like to assess the overall market contraction that we expect. But overall, we are already making a comparison against a very strong performance. So we will -- we see a slowdown here. In terms of Consumer Broking, well, this is where really we have achieved only mediocre performance over the last, say, 2 years, whatever. And here, the pandemic didn't help. Basically, what happened was that we saw a steep decline in demand during the pandemic. But while for mortgages, we saw a recovery both in terms of closed transactions and also in applications, here, basically, we didn't see a significant recovery. We saw some recovery but not all the way up as we would -- or a good way up as we would have hoped. Basically, 2 things happened. One is possibly the segment of people that takes a personal loan is a lower segment of the economy, and maybe these guys are more fearful about their jobs or they didn't even have permanent jobs. And so in Italy, layoffs of permanent employees have been suspended by law, whereas if it's temporary contract, they would just let it expire. So maybe some of the people in the target here are just less stable work and so on. But the biggest problem is the finance houses, the lenders got so concerned about the employment outlook that they significantly tightened the lending criteria. So when before we had an approval rate of, say, 40%, and now we have an approval rate of 25%, and that makes a lot of difference. So we have not a full recovery of demand and a tightening of the lending criteria. So here, the prospect has been only modest. On the other hand, this is the positive thing. We did a lot of work to develop the offering, especially by other new lenders. And by the way, lots of lenders that specialize in secured loans, cessione del quinto, which are -- is going to have approval issues even in a downturn because they are secured. So maybe those products and also since the interest rates and many other things will help us to offset the weakness of our panel in this period. By the way, I read the investor report of MoneySuperMarket. And apparently also in the U.K., there was the same type of problem with all the lenders becoming extra careful after the lockdown. Then Insurance Broking -- basically Insurance Broking there. The story apparently is that the contract, during the lockdown, a lot of people didn't renew their policy or they postponed the purchase of the car or whatever. But when they were able to circle -- to go around again, to move again, obviously, they have to insure their car. Or with motorcycles that are only seasonal policy, they had to take insurance in order to use a motorcycle. So basically, whatever was lost in the month of lockdown, and this is our feeling, we recovered in May and June where we had spikes of new sales. And then things obviously slowed down a little bit because there was no longer this pent-up demand. But still we are seeing growth, and we expect to see growth year-on-year. And basically, we think that anyway the lockdown in general has possibly increased the propensity of people to buy things online. So also for this reason, we expect to see growth in the second half of the year. Finally, E-Commerce Price Comparison. Here, basically, we hit a jackpot during the lockdown because basically during the lockdown, there was a combination of 2 things. First of all, the majority of physical stores were closed by law. So you couldn't just go and buy sports equipment in a physical place. You couldn't get out, and the place was closed anyway. So that's the first factor. And second factor was that Amazon was possibly growing too much or having difficulties in fulfilling orders for basic supplies and so on. So they basically decided to stop advertising on the Internet. So for 2 or 3 months, we were possibly -- I don't know the exact figures, but I think we were the biggest advertiser in e-commerce in Italy because on Google, Amazon was no longer present. Amazon is normally the #1. So basically no competition on advertising, and a lot of demand. So that was a driver of the performance. And however, the situation is more back to normal, I would say, because stores are open. They are desperate to bring clients in, so they do promotion. They do all sorts of things. Maybe they're not fully sustainable, but still, they are quite aggressive. And unless -- so e-commerce penetration after spiking is lower. And also Amazon is back to normal. So it's the biggest online spender and so on and advertising is expensive again. So here, you shouldn't expect anything similar to this growth rate for the second half of the year. Also, our organic visibility has gone down a little bit more. So we expect growth, yes, but moderate growth year-on-year, not what we have seen in the first half. So this ends the outlook for the Broking Division, and I hand it over to Alessandro for the BPO Division.

Yes. Hello, everyone. And coming to the BPO Division, as you have seen, in terms of revenues, the whole division improved resilience, basically reaching just above the level of revenues of the same year -- of the last year -- the same period of the last year, but at the same time, we have seen a significant weakening of our margins. Marco has already explained part of that. I'll -- which is due to an EBIT deterioration more than an EBITDA one, but also at the EBITDA level, and you can see the numbers in our report, and -- at the same level of revenues or roughly the same level of revenues, we have had more than a couple of million less in terms of EBITDA. And as we said, there are basically 2 factors here. One is an increase in our division level cost, so what we would call indirect costs, so offices and staff. And this is basically due to some growth. For example, we have opened a second office in Milan. And also in structural -- in personnel costs to manage things and manage complexity at the divisional level. Normally, we would have seen a growth -- a much higher growth in terms of revenues. And therefore, we would have seen an operating leverage effect on this, which we have not seen instead for this half of the year. As we look forward, we have structured ourselves for a higher growth rate. So we believe that it's -- believe possible to come back to the level of operating margins that we have seen in the past. On the other hand, we had also a mix effect. So if you look at our Mortgage BPO, you see a significant growth, but most of it -- actually, all of it, and also compensating for a decrease in the other business, has come from para-notary services. Para-notary services is basically -- agent side also -- a big part of reselling notary activities where our margin is really not as significant as we see in all the other services that we do. And therefore, this mix effect accounts for almost half of the decrease in EBITDA that you see. So if you take away the mix effect and the indirect cost, overall, the division has to be made at the same level of operating margin in percentage terms. But this overall picture is actually a sum of businesses that have reacted strongly to the situation that we had and others that were, in fact, impacted much more. So I'll go in the detail afterwards. So let's say that we expect to see the second half of the year basically in line with what we have seen in the first half. So you should expect something similar to what we have seen in the first half for the second one. It remains positive, our outlook, more in the medium term as we expected. When business will go more or less back to normal, also from acquiring new customers, is obviously very difficult in this period because all the commercial activities for really new big projects are not as simple. They were already lengthy, but obviously, in this setting has been all -- with our people and small broking has been more difficult. So -- but what has been positive is that it is very clear that we have been -- that we are very -- that we have a strong balance sheet, and we have been able to go through this crisis easily, that we have been able to give operational continuity during this difficult period, and this has obviously increased our status with clients. So we are -- we believe that new opportunities will come from the situation. And our competitive positioning is still very strong, actually has strengthened in this month. So if we go into the different business line. What you can see, as I said, is that Mortgage BPO had a significant growth over the semester, and it's even stronger if you look just at the last quarter. But this comes out of, as I've already commented, from 2 different trends. So on one is our core mortgage services that we have seen a decrease year-on-year. And we said we have seen a very significant growth that more than compensates of para-notary services. So the reason why we have the virtuous effect is that on para-notary services, you should know if you've followed all the comments in the last years that we have a very significant market share. So we call over the market. And Marco has already commented on the exposure that we have seen in the last 6 months in terms of remortgages where most of our para-notary services are bought. Instead, on the traditional services, the more complete outsourcing that we do for banks, we have a different market share. And therefore, we are -- our performance is more connected to the single bank, their appetite, their competitive positioning and less then to the overall market. This is the reason why you see this 2 different effect. Looking forward for 2019, we definitely see -- for 2020, we definitely see a continuation of the para-notary services growth. But we also hope to see a bounce back of the traditional services as we expect that the bank that was stopped by Bank of Italy last year will be able to regain its opportunity to restart its operation in -- during the fall. So that will help us get some more volumes also. Real Estate Services BPO is one of those that have been impacted heavily. As you can imagine, though, it was more complex to go and have an appraisal in the property of someone during the lockdown. As we commented in the past, we have had the opportunity to try also some remote appraisals, but obviously, that was just a portion of the growth. So here, we have a stable revenue. But the reality is that we had a larger customer base in terms of banks than in 2019. So in the last month of the semester, we have actually seen a bounce back. So we expect our second half of 2020 to be better than the one of 2019. Loans BPO was actually the good surprise of this period. I wouldn't go as far as the e-commerce comparisons to say that we hit a bit of a jackpot. But definitely here, we were lucky enough that one of the projects that we had been working on in the last part of 2019, which was entering the market of SME loans, exploded because of the crisis created by the emergency -- the pandemic emergency. So as you all know, one of the things that has been done to try to sustain small and medium enterprise is that there has been a very significant fund -- increase in the fund dedicated to government-guaranteed SME loans. And the way the government has chosen to give liquidity to company who was back to using the bank system and giving a guarantee on these loans. So the banks have been overwhelmed by the need to process loans. And so people who were already in this business have seen significant growth, and people who were entering the business like us has been able to gain very quickly volumes. And this is here to stay, maybe not in terms of volumes. Obviously, the volumes are normalizing, but this has helped us enter in this market, which would have been probably longer, maybe it would have required some small acquisition, but we have been able to gain already -- to reach already a decent scale on this business and to win some interesting customers during the pandemic period. Also, we have concentrated ourselves in the origination phase. But as you know, in -- we also work in the servicing phase. And so we are now structuring to start servicing this portfolio that has been originated during the pandemic. And banks have been overwhelmed in the origination. I feel they also will be overwhelmed during the servicing. So many of these guarantees will probably need to be called if, obviously, the economy doesn't have a recovery as fast as the government has expected. And many of these loans were given very quickly during this period. So there might have been some credit underwriting that was not done as -- in a per share way as in past periods. So we will see. But overall, we believe that this business line is -- has done well and will continue to do well in the next months. So as a note -- as a passing note, we used to call this the CQS loans because mostly on what we have done was cessione del quinto. We now call it the Loans BPO more genetically because this year we have everything that is not mortgages and not losing in the credit area. Insurance BPO is instead one where we have seen a significant decrease. And being this -- the smallest of our business line, it was also difficult to contain costs as there is certain level of -- it's difficult to shrink down, especially indirect cost. Fortunately, the situation has been improved starting in July. So marginality that was really hit heavily in the first semester is expected to improve during the second half of the year. Service -- instead, the Investment Services BPO is the one that has remained more stable during this. We have actually have had spikes at the beginning of the lockdown where the final customers of the investment companies did a lot of reshuffling of their portfolios and the positioning of their investments to deal with the crisis. But then this has normalized in the second half with, as you can see, some very resilient businesses and basically has remained stable, both in terms of marginality and in terms of revenues. If we look at the Leasing/Rental BPO, which is done by our subsidiary, Agenzia Italia, here, the automotive business was at ease. As you know, everything was shut down. It was one of those significant economic areas where the government had decided to basically clamp it down. And we have seen a rebound. But obviously, the volumes were not done -- are basically -- were not -- are basically whole in the revenues -- for the whole year. So what we expect to see in -- by the end of the year, if we see some results, that would be lower than the one of last year. Not significantly lower. I would say, probably around 10% lower than last year. Here, also thanks to the dimensions of the business, the size of the business, we were able to control overall cost in a way that the marginality of this one, without disclosing the exact figure, but anyway the marginality of the business was not having impacted as it was instead in insurance and in other smaller businesses, which we record in the other revenues of the BPO. I think this concludes my part of the presentation, and I hand it back to Marco. Hello?

Sorry. I was on mute. So thank you, Alessandro. And I will just add a few comments about our financial position before starting the Q&A. I think we have to go to the appendix here, to Page 40, which is the page before the last one. You have our net financial position for June 30. Basically, you see our reported net financial position is negative EUR 71.5 million, and that's significantly better than the net financial position at the end of 2019. Obviously, we produced significant cash from the operations, and we paid less dividends than we used to pay. And also, we borrowed some money as we had anticipated. And this net financial position -- these are extremely comfortable. Of course, the item M, other noncurrent borrowings, just as a reminder, is the combination of the IFRS liability for leases that we now have to recognize as a debt. And that's around 1/3 of the total amount and is also the liability -- the expected liability for the total call on the remaining 50% of Agenzia Italia. So this is not a banking liability. So one could also read -- and by the way, these amounts are not included in our loan covenants. So when we calculate the covenants for our loans, we don't report EUR 71.5 million, but we report something around EUR 30 million of net financial position. So we have a very comfortable net financial position. Also, this doesn't include the Cerved shares that we have, which as of June 30, we are at 6.5 million shares worth at the time around EUR 40 million. So -- and that doesn't appear in the net financial position. So we have -- obviously, we're super careful because we didn't know what could happen. And so we strengthened our balance sheet. But we think now we are in a very strong position that gives us also, in general, flexibility. So this is the last comment we wanted to make. And so we can open the floor to questions. So please, operator, go ahead.

[Operator Instructions] The first question is from Giovanni Razzoli with Equita.

Two question -- one question and one clarification, please. Is it fair to assume as per the documents that you have sent us that in the second quarter around 60% of the new business in terms of new mortgages was represented by remortgaging activity at the sector level? So I presume that in your case, this 60% is even higher. So this is my first question. The second question is a clarification, Marco, on your last comment. You have -- you said that as of June, you had 6.5 million shares in Cerved, right?

Yes. I think so. I'll check, but yes.

Because I missed the number. I missed your comments.

Okay. Just to make sure that everybody is able to check that information, we provided the figure also in the half year report. I'm looking at the item. This is -- just a second -- this is reported under note #8, which is financial assets at fair value. And here in the comment, it says, we had 6,483,080 ordinary Cerved shares. Basically, we sold some shares, as you know, in Q1, and then -- you know the company pretty well. We have plenty of cash. So we decided to buy back the ones that we had sold and a little bit more. So this is what happened. And regarding the -- I don't think 60% of the market was remortgages. In general, our market share of remortgages is greater -- more than proportionate. So let's say, our market share is overall of -- the mortgage market is a fictitious figure, 8%, probably our market share of remortgage is 11% or 12% and double-check it is 6%, something like that. We are always in that situation because it's more natural to go online for a remortgage than for the first mortgage. But I don't think remortgages accounted for the majority of the market. I would have to check the figure. There's a lot of growth. So they accounted for the growth of the market. But in terms of total size, they are less than that. I'll check and get back to you off-line in this because...

Okay. Because it is probably I misunderstood the document that you usually sent us with the breakdown of the new mortgage.

No. Okay. Let's just comment on the comment on what we have written because, basically, what we have written on the market -- let me explain what it means because it's not clear maybe, well-explained. Basically, in -- back to Page 18 of the presentation. So let's look at originations in May, June and July. Overall, for the entire market, new originations in terms of value were up 12.8% in May, 14.6% in June and 1.8% in July. But if you look only at purchase mortgages, you had a decline of 2.8% in May, 0.6% in June and 10.2% in July. What is the decline in remortgages? Well, it is difficult to calculate because we didn't provide the weight. So let's say, if remortgages are, say -- that could be my memory or guess because I don't have the figure in front of me. If mortgages were onefold for the market, of course, to offset the decline of 2.8% of the other 2/3 of purchase mortgages and reach 14.6% probably in order to grow by 35% year-on-year and so on. I'll try to, off-line, provide the figures that were used for the calculation. But remortgage, anyway, this is not implied in anything that is written here. They were not the majority of the market anyway.

[Operator Instructions] The next question is from Filippo Prini with Kepler.

I've got 3 quick questions, if I may. The first one is on the activity of the loans you would have state guarantee for SMEs. Could you confirm that the business value will be at [indiscernible] state-backed loans you would have did an activity in which you will not count anymore? The second is on the bidding back of the stake in Cerved. Sorry if I miss it, but if you could please explain why the rationale behind the buying back of certain shares in Cerved after that you had sold part there in the next month or previous month. And the last one is on the free cash flow generation for the course of the year. If I'm not wrong, the net working capital being a company that added cash has been cash supportive. It's very difficult to understand the evolution for these lines to be built back. If you can point it out, some specific factor that allowed net working capital to be cash supportive in the first 6 months over the year.