Banca Sistema S.p.A. (BST) Earnings Call Transcript & Summary

Good afternoon. This is the Chorus Call conference operator. Welcome and thank you for joining the Banca Sistema Nine Months 2020 Results Conference Call. [Operator Instructions] At this time, I would like to turn the conference over to Mr. Gianluca Garbi, CEO of Banca Sistema. Please go ahead, sir.

Thank you, and good afternoon to all participant. Ilaria Bennati, our CFO; and Carlo Di Pierro, our Investor later Director, are also connected to this call. Let me start by saying that since the end of February, we have proactively implemented a set of measure to mitigate the COVID-19 impact on our employee and on our customer of the 3 business division. And since last week, we have increased the percentage of people working from home. But we were able to test this way of working already some months ago when we had the lockdown in Italy, and the restrictions were stronger than today. So despite the negative macro environment caused by this pandemic situation, we have so far registered a good commercial performance with the recovery on the factoring turnover in the third quarter, which has increased 10% on year-on-year compared to the same period of the previous year, and a robust purchase of CQ portfolio. Then we have finalized the acquisition of the gold and jewelry-backed loans business unit, and we have completed the integration of this subsidiary of ProntoPegno. But let me now go to the number that are included in the presentation that is available. So on the first slide, you see the number. So factoring turnover, which is equal to EUR 2.183 billion, has increased 4% on a year-on-year basis. And as I mentioned before, is a 10% increase of the -- of the third quarter of 2020 versus the third quarter of 2019. If we exclude the tax receivable and the receivable related to the football industry that have registered also regain of growth in the last quarter, we have registered altogether a 6% growth on a year-on-year basis. The CQS and CQ business outstanding have reached EUR 931 million, which is an increase of 21% on a year-on-year basis. The gold and jewelry-backed loans outstanding have reached EUR 75 million from the EUR 13 million that was at end of the first part of the year following the acquisition of the business from Banca Intesa. The net interest income is equal to EUR 52.8 million, which is a reduction of 10% on a year-on-year basis, mainly influenced by factoring less late payment interest collection and accrual. The lower funding cost on a year-on-year basis is at 0.6%, which is in terms of the quarter the same level of stock than the previous semiannual result. The total income is equal to EUR 72.1 million, which is almost stable. The cost of risk is equal to 36 basis point with the EUR 7.2 million of loan loss provision, which is in line with the expectation. The total operating cost have increased, but this increase is mainly due to the higher contribution to the Resolution Fund, as we already mentioned in the previous quarterly result and of the Atlantide consolidation in 2020 plus the nonrecurring expenses of integration costs from -- coming from the acquisition of Banca Intesa activity of gold and jewelry-backed loans. And then Ilaria Bennati will explain more in details the trend. At the end, the net income is equal to EUR 19.5 million, which is down on a year-on-year basis, as we expected, as I already mentioned in the previous quarter. And this is mainly due to the lower late payment interest contribution, as I mentioned before. Moving to the next slide, which is Slide #3. As you can see, the factoring outstanding trend has shown an increase on a quarter-on-quarter basis. And the growth is explained by the volume, which has been registered on the first half of the year. The tax receivable represents 21% of the total outstanding, in line exactly the same number, same percentage that we had at the end of the first part of the year. And for the same reason that I explained in July, the new volume of tax receivable is influenced by the rescheduling of all tax deadline. And so this amount of receivable is certainly in lower than what we will have expected, mainly to postponement of payment by the company of their tax. As you can see, the turnover on the CQ was higher on a year-on-year basis that the performance of indirect component versus the direct component because as you may remember on -- during the lockdown period, there was a difficulty in collecting signature for new contracts. So now the activity has moved back to normal and hopefully in the few -- in the next months, we will now have a new lockdown again. If I go back to the comment at the end of September, we have -- we are in line with the expectation excluding the period of the lockdown. The contribution of the business unit that has been purchased was not full in the third quarter because the broking jewelry-backed has to consider for 80 days rather than 90 days. Then we have also to consider that historically, in the month of July and August, there are not many auction and auction every time that we do auction, this is an income for the ProntoPegno. And so that is something that has to be seen in the following months, a good result of the business. So I'm underlining this because this jewelry-backed loans are, as I said, will produce their result in the following months, also thanks to the revamp of the auction that we will be able to do from now to year-end. Now I will leave the floor to Ilaria that will explain more in depth the result. And at the end, I will comment the capital, and I will take questions. Ilaria, go ahead.

Thanks, Gianluca. Good afternoon to everybody. As usual, I'll start my comments from the balance sheet on slide 4. Assuming the table, total assets is up 4%. And in particular, the Govies' portfolio is down quarter-on-quarter at EUR 991 million with an average duration of 16 months. The sale of Govies allowed us to register trading gains also in this quarter. Loans and amortized cost is now EUR 2.8 billion and is up quarter-on-quarter, mainly thanks to CQ assets and gold-backed loans. In particular, factoring receivables are instead down 7% versus year-end due to higher collection over the quarters. CQ loans are up 14% versus year-end and also quarter-on-quarter, as Gianluca mentioned, and gold and jewelry-backed loans are significantly up quarter-on-quarter. Again, as Gianluca mentioned, thanks to the acquisition of the business unit from Intesa Sanpaolo. And now the amount of credit stands at EUR 75 million. Goodwill increase is related to the acquisition of the business unit just mentioned. On the liabilities side, due banks quarter-on-quarter, increase is driven by bank financing agreement related to the pawnbroking acquisition which was already in place and which would be refinanced at the beginning of next year. ECB funding is stable at EUR 690 million. Due to customers decrease quarter-on-quarter is mainly driven by the decrease in repos as a consequence of the sale of a portion of the Govies' portfolio and that has more than compensated the increase in term deposits. Debt securities decrease quarter-on-quarter is due to a reduction of the CQ securitization. We now move on to the next page, and we discuss the P&L. 9-month interest income is down 11% year-on-year, and the drop is mainly driven by lower contribution of factoring in the -- especially in the third quarter. As a consequence, we observe now a reduction in the contribution of factoring to total interest income, which is now 66% while it was 72% in the same period last year and historically even higher. The reduction of income from factoring is mostly related to LPI from legal collection. As anticipated, during previous calls, the benefit of the update of accrual rates would be lower and lower over time, as the statistical model has now stabilized around the current level of accruals. So that was not a surprise for us. The update of the model performing in Q3 have provided levels, which were just slightly higher than the previous ones. And as a result of that, the income effect of the accrual update has only been EUR 1 million compared to the EUR 4.8 million related to last year. Overall, the 9 months’ contribution of NPI from legal action has been EUR 16.1 million compared to what it was in the 9 months 2019 at EUR 24.3 million, so significantly higher than current levels. The segment where we've reduced the stable performance steering the NPI segment has been extra collection where the EUR 7.7 million of collections compared with the EUR 8.5 million in the first 9 months of last year. There has been no sale of LPI in this quarter. From this quarter, we've started to see a significant contribution of the pawnbroking business also thanks to the acquisition. The contribution to total interest income of the business line has been EUR 1.9 million, of which EUR 1.1 million is originated by the business unit acquired. Let's now move on to discuss margins. As you can see from the top-right table, adjusted income margin is down year-on-year, driven by both factoring and CQ. CQ margins seem to have stabilized there on the levels seen in the last 2 quarters, while factoring margins have not benefited from the usual seasonality effect registered over the years where the third and the last quarters usually had shown a sharp increase in marginality. So we didn't have that effect this quarter for the reasons we discussed, which is, as I said, mainly related to LPI accruals. We now -- we move on to discuss LPI now. The stock of LPI, which is represented in the bottom left table, stands at EUR 153 million, of which EUR 101 million is part of the accrual perimeter. This is down quarter -- slightly down quarter-on-quarter mainly due to cash collections. The amount of LPI currently recorded in the balance sheet is EUR 50.2 million. We now move on to comment the total income performance. So if we turn to Slide 6. In the first 9 months, total income is roughly stable year-on-year, thanks to the contribution of other income, which is treasury activity and the sale of a factoring portfolio. This activity -- this income has almost entirely compensated the drop in the net interest income. If we look at the net interest income, the decrease is mainly driven by the decrease in factoring income described before while total cost of funding, which is now equal to 0.6% is lower year-on-year and stable quarter-on-quarter, also thanks to a good balance between diversification of the funding sources, cost and duration. Net commission component is lower year-on-year as a consequence of lower origination of factoring commission products. The other income item also includes EUR 2.1 million gain from the sale of factoring portfolios. We have registered also a strong contribution of the Govies portfolio in the first 9 months, which is in total worth at EUR 8.6 million. Of that, EUR 3.6 million is included in the net interest income and EUR 5 million is included in the other income. Gold and jewelry-backed loans business has contributed to total income with EUR 3 million, of which EUR 1.6 billion is generated by the business unit. Cost on slide 7. As you can see, total costs are up year-on-year. The main driver of the cost increase in the last quarter has been the integration of the business unit starting from July. This accounts for EUR 1.4 million operating costs and EUR 1.3 million integration costs. If we then add up the EUR 300,000 integration costs already incurred in Q2, we get to a total impact on cost for the acquisition for the first 9 months of EUR 3 million. If we then break it down in the various components we have, that EUR 1.6 million is the total integration costs, of which EUR 1 million is the registration tax on goodwill, EUR 1.4 million is operating cost, of which EUR 1 million is personnel expenses related to the new 58 headcounts. Then another important driver of the cost increase is the additional contribution of the EUR 900,000 to the Single Resolution Fund, which we had already discussed in the first half. In the top-left graph, you see the total operating costs have gone up by 9% year-on-year. But if we normalize it for the nonrecurring integration costs and for the contribution to the Single Resolution Fund, we get to an increase of 4% year-on-year. Then in order to make a like-for-like comparison, if you deduct the impact of the 100 -- sorry, EUR 1.4 million operating costs, which I mentioned earlier, related to the business unit acquired, we reached flat operating cost year-on-year. We now move on to the next slide on funding. Funding mix keeps on being balanced between wholesale and retail. The retail component is now 54% and is up quarter-on-quarter. As mentioned already, the cost of funding is lower year-on-year, stable quarter-on-quarter at 0.6%. The wholesale component decrease is due to the reduction of the Govies portfolio as we discussed. And as said already, ECB funding is stable at EUR 690 million. Retail funding is up in the quarter due to the term deposit stock, in particular, for the foreign component. After managing the stock of term deposit down in the first half, we have readjusted the interest rate offer in order to resume the funding from this channel. We are now receiving stable inflows every month, which would move the redemption profile. The flow -- the inflows we received have an average duration of 12 months. After the end of the quarter, in October, we have reimbursed the senior bond. And as mentioned already, we do not intend to tap the capital market in the short term. Finally, we move on to the next slide, Slide 9, and we discuss asset quality. Gross NPL, as you can see, is down quarter-on-quarter, thanks to the reduction in past due. Net bad loans represent 1.1% of total loans and stable quarter-on-quarter. Cost of risk at 36 basis points is slightly lower quarter-on-quarter and stable year-on-year. As mentioned already in the previous call, loss loan provisions for the 9 months are higher year-on-year. And part of the increase is a consequence of the update of the impairment model to account for reversing of the macroeconomic context caused by ongoing -- the ongoing health emergency. Other than that, we haven't registered any impact due to the COVID emergency on our assets. I'll now hand the floor back to Gianluca.

Thank you. We are now on Slide 10, talking about the regulatory capital. As you can see, we were at -- in terms of core Tier 1 at 13.7% in -- at the end of June. Then with the acquisition of the business from Intesa, this went down to 11.7% because the cost of the acquisition was worth 200 basis points. And now from 11.7%, we are up at 12%. Same consideration for the total capital ratio that were at 17.3% at the end of the first half of the year after the acquisition -- immediately after the acquisition at 15.2%, and now it's up from the number to 15.4%. Let me say that based on the economic and business environment and also looking at the data coming from the Association of Factoring Company, we are very pleased of the performance that we had in the last 6 months, that prove business, which is resilient, even though as anticipated during the previous call, the contribution of late payment interest is -- has been less. But this is something that we were expected. But this doesn't mean a reduction on the return on asset, but simply postponement of some of the revenue for the following year. Bear in mind that delinquent interest has an accrual rate of 8%. So more time means more return for the future. So I'm also very pleased for the acquisition finally of the pawnbroking business from Banca Intesa. Clearly, this will show the contribution to the P&L in the following year, but the commercial trend is very strong, let me say. Unfortunately, because that means that the macroeconomic situation is worsening, but we still have a lot of people, new clients and existing client that every day come to our shop all around Italy. I have noticed a negative trend on -- of our stock in the market. But at the same time, we have also noticed that has been published some short selling on our stock from Quant Fund as well that could be seen as a negative signal -- negative effect in the short term. But overall, I think that the fact that our stock has become so liquid that attract the attention of this type of hedge fund that short the stock will be -- is a positive signal means that the liquidity is big enough to have also this type of investor. And in the medium term, long term, this will be something positive for all the investors that would like to invest in a small cap like Banca Sistema because thanks to the high liquidity on the stock they probably have an easy way to enter an exit on the stock as well. I do expect that in the next 3 months from now to year-end, the performance in terms of net profit will be in line with the first 6 months. Of course, as you know, our business is also a business where the last part of the year can have an important impact. So it can go up or down, mainly dependent on some one-off transaction on the factoring or some speed up in collection in of the receivable. These are 2 elements that can change the trend. But if no extraordinary event will happen, I do expect that the net profit of the first 9 months will be projected, therefore, the rest of the year. Now I will leave the floor to question for all the participants that would like to have a question.

[Operator Instructions] The first question is from Manuela Meroni with Intesa Sanpaolo.

I have some questions. The first one is on the factoring business. Factoring turnover increased by 4% in 9 months and 10% in the third quarter but factoring outstanding is declining by 4% quarter-on-quarter and year-on-year. So I'm wondering if you can explain the divergence between the trend of turnover and spending. Again, on the factoring business, you're accounting for EUR 1 million revenues for the update of the LPI model. Can you tell us what are the moving parts here? So higher or lower collection rate, different in values or what else? And can we assume that the impact of the model update in the coming next year will be negligible considering that your historical series are now more complete? And then again on the factoring business, I'm wondering if you can share us an outlook for the factoring business in the next quarter and for 2021, both in terms of actual and in terms of LPI collection? Then on the football company, we have heard that some football companies are in crisis due to the lock of revenues. Do you expect to have some impact from that on your existing portfolio? Or you are safe on that? And last question, sorry for making the same question. Yesterday, the European Economic Committee asked for some revision of the calendar provisioning and the new definition on default. Could you share with us your view on that? And if you have any update on this matter compare what you said in June -- in September?

Thank you for all the question. Let me, well, start to ask to some of the questions and then maybe leave to Ilaria to complete maybe my answer with the additional comment. Being in different locations, clearly, I'm now able to split. But in terms of the factoring outstanding versus turnover, I will leave then to Ilaria the answer. In terms of LPI, you correctly said we are now at the level of accrual that is in line with the expectation. So we do not expect for the future an important adjustment like in the past. So that is, I would say, something that is positive. So we will have less volatility in our P&L on these items. The volatility that remain in place, clearly, is the time of collection. As we take today, 62% of accrual of the 8%, we are not indifferent anyhow on the cash component. So if we are able to collect faster, this will have a positive impact on our P&L. I do not expect that due to the current situation, we will see some major speed up coming from court. Unfortunately, probably we see the opposite. So postponement of cases in court, which will imply for us less collection, but probably more return because of the more time. As of today, based on our model, we do not have any statistical -- and probably because there was a short period, evidence of major difference in terms of time value because otherwise, we will have adjusted the time value accordingly. We have to see based on the current situation, if the close of court of the slowdown of court cases will have a statistical impact in the model. In this case, we may have to consider a different time value, and this could have a negative adjustment that, in any case, I remain positive that can be readjusted with a positive accrual evidence. Just to in general, but maybe Ilaria can be more precise in terms of outlook. We have today more than EUR 150 million of LPI. We have taken -- we are -- the perimeter of accrual, if we exclude the city that are in distress where we are not allowed to take accrual or other cases where we are not taking accrual, so the perimeter is EUR 100 million. And we have taken today only EUR 50 million in the P&L. So that means that we have other EUR 100 million of LPI that didn't go to the P&L. If we consider that the performance in terms of collection has always been above 80% of the 8%, this means that in the following year, on accrual on a cash basis, we will have this treasure of EUR 100 million that will go sooner or later through the P&L. In terms of the problem of football company, I understand the general comment. At the same time, bear in mind that our risk is protected because our credit is towards the broadcaster and towards the league. And when we look at the numbers of the football company, we haven't seen a major change in the number. The company, the football club that are performing very badly are probably the top one, which are outside of the parameters of our client. Because we would top 3, 4 football team, we never had any discussion for different reasons. One is because they are used to a cost of funding that is not in line with our return. So we cannot be competitive. Second, because we see them as more risky because usually, this most important football club, they invest a lot of money upfront in order to try to be in the European League. But if they fail as it happened to some of this club and they invested a lot of money but then they fail to go in the European League, they don't have the revenue coming from the television right, which by definition will create a drop in their P&L. This is independently from the COVID or non-COVID. This has always been the case. From a risk perspective, the best club are the ones that stay on the bottom part of the first league. And they may go down to the second league because they receive what they call in technical term a parachute, so they receive a subsidy, thanks to that. So there -- thanks to this subsidy, they reduced their cost, and they get more revenue. So we are not concerned in terms of the football company that we are financing. In terms of calendar provisioning, clearly, there are continuous update. Even last week, there's been from the regulator additional Q&A. So it seems that there is a moving -- keep the moving target, clarification and so on. We try to keep up to date ourselves on the situation. But based on what we see today, we do not imagine to have major impact coming from the business that we are involved. If they postpone the calendar provisioning, I think that is good for the market. It's good for the banking sector. Probably it's the most wise decision that should be taken from politician or from regulator, but we are not the one that are asking for it because we don't see that as a major impact to our activity. Now Ilaria, would you like to answer to the turnover versus outstanding on the factoring?

Yes, sure. As you rightly pointed out, Manuela, the outstanding year-on-year has decreased by 4% despite the solid growth in turnover. This is entirely due to the amount of collections over the quarters. We have received a collection amount, which was higher by 17% versus September last year. So just to give you some figures. Last year, in September, we had cashed in EUR 1.8 billion. At the end of September this year, we had cashed in EUR 2.16 billion, so an increase by 17%. The increase is mainly related to fiscal receivables. While on the other side, commercial receivables have shown a stable pattern of collections. Just to give you again some figures, last year in September, we had collected EUR 133 million of fiscal credits. The amount in September this year is EUR 400 million. So an increase, which is triple digits. So this is the only explanation to the decrease in outstanding.

The next question is from Luigi Tramontana with Banca Akros.

Just one question on my side. To me, the main positive surprise in the results was the evolution of your capital ratios. You gained around 30 bps quarter-on-quarter. Can you please give us some explanation of the trends explaining this increase?

Ilaria, do you want to take the question?

Yes. Sure. In terms of trends, basically, this -- the 12% ratio is -- has a net impact of the positive effect of the reduction in the secured CQ RWA consumption, which occurred in June this year and the negative effect of the goodwill related to the acquisition of the pawnbroking business, which occurred in September. So the 2 effects almost entirely compensated one each other, although they have occurred in 2 different quarters. So the 12% is the result of that together clearly with the increase in the capital driven by the performance of the business.

Plus, let me add. If you look at the overdue, there's been a reduction of overdue and overdue as an RWA of 150%. If we are able to reduce the -- what is technically called overdue and we are able to do it, this go back to the original RWA that can be 0%. It can be 20% or 50%. So these things had also positive impact on the capital ratio.

So yes. Sure. This is the element I was missing. How did you reduce so significantly your past-due loans?

We have in order to preempt any possible problem of the future coming with the new definition of default, we are -- our collection team is tackling those position where we may see a past due. While in the past, we did consider it in the collection strategy as I would say, neutral, now we are taking more, more attention to everything that come to past due. So we offer a restructuring of position in terms of repayment plan. And we may offer some settlement in order to faster collection. So we take action proactively more than in the past, keeping in mind our goal to reduce as much as we can the past due, and in order to free up capital. So it's, I would say, become a more strategic activity rather than operational as it was more the case in the past.

Your next question is from Filippo Prini with Kepler.

Two brief questions. The first one is on the margin of the CQ business. Could you tell me which is the exact contribution of this business line to the interest income of this quarter? And still on CQ, if I understood correctly, maybe you are facing some headwind from the effect of early repayment, can you give us some visibility on the end of this negative effect on your margin?

So I will take this questions, Gianluca.

Yes, please, Ilaria. Go ahead.

Okay. So in terms of margins, the CQ margins have seemed to have stabilized around the levels shown for the past couple of quarters, which means around a level between 2.6% and 2.8%. And this is expected to continue in the future. In terms of contribution to interest income, the contribution of the CQ business has been almost flat with respect to the last year. To give you precise figures, we had interest income for almost EUR 17 million from the CQ business this year as opposed to the EUR 17.3 million last year. So as you said -- as I said, almost flat. It's true that the effect of early prepayment is affecting profitability. Otherwise, we would have been able to show a higher contribution to P&L, considering that the amount of outstanding has indeed increased. We believe this is -- this year is the peak of this effect because most of the portfolios that we had purchased from external originators have reached the time which is 4/5 of the contractual life. The time where the prepayment effect is most -- is more sharp. So probably 2020 and 2021 will be the financial year, so where this effect is sharper. Going forward, these effects should reduce in size.

Ilaria on your legal point that you're able to refinance only when you reach 4/5 of the period. And those are the coming back to the original portfolio when we build most of the position. So we are ending to this period in 2020/2021.

Okay. So very briefly. So maybe you should be able to some tail effect next year and it will be clean in 2022?

Correct.

Maybe -- yes, and maybe not entirely clean, but take us into account the fact that we are provisioning against this effect. So the impact -- the future impact on the income statement should be smoother over time.

[Operator Instructions] The next question is from Christian Carrese with Intermonte.

Gianluca, I was -- my first question is on fourth quarter outlook, in particular, on net interest income and the cost of risk. As far as net interest income, I would like to understand that the ruling part you expect in the fourth quarter, in particular, on the contribution from the pawnbroking business. It was EUR 1.6 million this quarter. There is any evidence that you could improve in the fourth quarter, also taking into account the possibility maybe of a positive mark-to-market of the gold, we this will be more spread over 2021, but if you can share with us your thoughts on this line? And in terms of spread, what do you expect in the fourth quarter, still some pressure on maybe an improvement. So if you can give us some ideas on that. And on LPI as well, I think that you don't expect an increase in LPI from a legal action. So if you can give some numbers on that? And also, if you can give us an update on October turnover. So how was the business this month? Do you expect to specifically look down maybe in some areas could have a negative impact both on legal action and turnover production? The second point is on the cost of risk. We saw a good -- as you said, you took action on past due. So there was a positive trend there. Cost of risk was in the area of EUR 2 million per quarter, more or less net of the model update in the second quarter. What do you expect for the final quarter in terms of cost of risk? The third quarter -- the third question is on common equity Tier 1 ratio. So basically, on a pro forma basis, 11.7% of the common equity Tier 1. So 12% mainly due to, as you said, no were waiting for the reduction of past due or there is also a positive impact from IFRS reserve on Govies portfolio? And see if you can share with us today an annualized capital gain on the portfolio? And finally, on the strategy, do you see in the current scenario, any alternative business that you can look at, as you did in the past with pawnbroking with CQS with low capital absorption and higher return?

Okay. So many question, even though you classify it 3 questions, but I counted more. I will leave most to Ilaria because they're relating to number. But I will take some -- in terms of turnover October, the October turnover as of today is higher compared to the October of last year. We continue to see the component of tax receivable being lower than from the past. And going to the end of the year -- can you still hear me? Yes. Okay.

Hello?

Hello. I don't know. Good. So the -- in terms of turnover from now to the year-end, I do not expect to see some of the big ticket of tax receivable that we have seen in the past simply because this area of tax receivable has deeply offset by government measure that are provided to companies. So the area of tax receivable is something that I do not expect in these days to increase significantly as it was the case of last year with some big ticket, which today, we don't have any visibility on that. In terms of collection of LPI, as I mentioned, I don't see due to the probably even further restriction potential and new lockdown, any possibility that the collection of LPI, which is the part of collection that go through court to speed up but eventually to slow down. So while as Ilaria mentioned before, we have received a very good income in terms of payment of receivable from obligor. These are mainly related to receivables that were not in action legal action. But for the component of receivable in legal action, we do not expect that the trend of positive -- the positive trend will be from now to year-end. That's the reason why I said that our expectation of the profit -- net profit from now to year-end will be in line with the first 9 months because this extra one-off transaction as well as faster collection from now to year-end, may not happen as it was the case of the past. This means that the return on asset is still very strong and continue to be resilient. But from a P&L, as we had the discussion already in the previous quarter, this could be postponement of a part of the P&L for the future, probably even higher because with 8% rate, if we wait for more, it's potentially even positive for the long-term return on the bank. Many of these things will also depend on what is really going on from now to the -- in next few days, few weeks, a new lockdown. They are going to close court like they did do through in the lockdown period. We don't know. But certainly, we do not imagine more activity from quarter that will justify an increase of collection of LPI.

So on this point on factoring, so basically assuming no pickup in the NPI from legal action...

Yes.

And some positive seasonality in the fourth quarter in terms of volumes and maybe in terms of margins, less tax receivable should improve the spread. I would expect net interest income to go up in the fourth quarter compared to the third quarter?

If you exclude the LPI component.