Good morning, everyone. I would like to welcome all of you to Credicorp Ltd Third Quarter 2020 Conference Call. We now have all of our speakers in conference. [Operator Instructions] With us today is Mr. Walter Bayly, Chief Executive Officer; Mr. Gianfranco Ferrari, Deputy Chief Executive Officer; Mr. Reynaldo Llosa, Chief Risk Officer; and Mr. Cesar Rios, Chief Financial Officer. Now it is my pleasure to turn the conference over to Credicorp's Chief Financial Officer, Mr. Cesar Rios. Mr. Rios, you may begin.

Thank you. Good morning, and welcome to Credicorp's conference call on our earnings results for the third quarter of 2020. Since our previous conference call, the sanity situation in Peru has improved considerably, and the economy continues to rapidly recover. In fact, the evolution of both scenario has exceeded expectations. Further, our weekly death tolls have registered significant improvement in recent months. Nevertheless, we remain vigilant to track an eventual second wave of COVID-19. In parallel, the Peruvian economy experienced V-shaped recovery throughout the third quarter of 2020. The economics has been encouraging over the past few months. According to the latest official data, economic activity in August was only [10%] below the figure recorded for the same time last year. This compares favorably with the minus 40% registered for the year-on-year comparison in April. This V-shaped recovery has also been observed for several economic indicators. As you can see in the charts, real estate transaction, cement dispatches and vehicle sales has recovered significantly in recent markets. Reactivation is also evident in labor market data, including payrolls through BCP. Next slide, please. Although Peru GDP posted its steepest decline in the second quarter of 2020, it is rapidly regaining territory in LatAm. Moreover, market consensus has gradually improved its expectations for the Peru GDP growth in 2021. Peru is expected to lead economic recovery in LatAm towards 2021 as we leverage a strong macroeconomic fundamentals, high commodity prices and broad government stimulus packages. In our previous conference call, we expect the GDP to drop between 11% and 15% in 2020, and that the rebound in 2021 will situate between 6% and 10%. Our list of latest estimate suggests that GDP will contract around 12.5% in 2020, and will rebound between 9% and 12% in 2021. It is important to note…

[Operator Instructions] Our first question will come from Ernesto Gabilondo, Bank of America.

My first question is on provision charges. So we saw that they started to normalize to PEN 1.3 billion in the quarter. So is this a level we should continue to see during the last quarter? And then my second question is on your reprogrammed portfolio. As you mentioned in your presentation, retail banking reprogrammed portfolio is showing that 6% are value loans, while 12% fell in the grace period. I also noticed that you are presenting the same breakdown for individual, SME-Pyme and SME business. So by any chance if you look at the number of the reprogrammed portfolio at a consolidated basis indicating how much is overdue and how much will be resuming payments in the next months? Considering the overdue loans in each segment and the portfolio that will resume payments, do you think that the presented provision created during the first half of the year will be enough to cover that portfolio? I just want to know if you have already created enough provisions, if you have achieved provisions or if they seem okay. And then my final question is on insurance revenues, which were affected by higher claims in life insurance due to the COVID-19. So I would like to hear your expectations for the insurance revenues in the next quarters.

Thank you, Ernesto. Regarding the questions on provisions, as we expected, we've seen quite an improvement in the levels of provisions during this third quarter as compared to second quarter. We expected to have a number closer to what we did in the first quarter. And the cost difference, both in the macro environment as well as the performance of our clients, will make us feel quite positive towards the level of provision for next quarter. We don't have a precise number today, but we see that, that positive trend continue in the same directions. In terms of the reprogrammed portfolio, this is a long journey, and we've seen -- and we've been helping our clients to reprogram their loans. And at a gradual pace, we've seen our clients both at SME and on individual clients starting to pay their loans at a better level than we expected initially. So in terms of, if we've done enough provisions, I mean, we are in the right trend towards provisioning what we need to do and that positive trend is expected to continue, as I mentioned during the next quarter and especially when during 2021.

Our next question will come from Jorge Kuri, Morgan Stanley.

Two questions, please. The first one is on your operating expenses. Your outlook slide said that operating model is being challenged to conduct structural medium-term measures. What exactly does that mean in terms of expense growth for 2021 and 2022? Your expense would have been, over the last 2 years, of around 6% to 7%. If you may end up with around, I'm guessing, 7% to 8%, does that mean that we're not going to see a slowdown in that level of growth for 2021 or 2022? And especially given the context of the very weak revenue growth, do you think that there's maybe an opportunity for you to try to offset that through being more aggressive in cutting expenses, particularly next year?

What we aim is to have income growing at a faster pace than expenses. And these structural measures that we have mentioned are going to have a gradual impact in 2021 and more visible in 2022. Short-term and continuous improvement measures has been taken in the last years at the same time that we have spent more heavily in the transformation. But these additional more structural measures are going to take some time to mature.

So just to lay it down, in terms of expense growth, does this mean that your expenses will grow similar to the last 2 years, 7%, 8% or more than that?

We expect to be in this range, probably a little bit lower, in line with the increased level of activity in the transactional activity and the offering of new and different products. What we try to manage more than the line by itself is the ratio between fee income and expense growth.

Got it. My second question is on provisions. And so your -- you built a very sensible, I think, amount of excess provisions given the NPL outlook and overall contraction in economic activity. And at this point, do you think it's possible to return to 1.5%, 1.6% cost of risk for the second half of 2021? I'm assuming you're still going to be with probably elevated cost of risk in the first half of the year, given that you're going to start to see a wave of NPLs from forbearance program rolling off. But again, given the significant amount of reserves that you have built, can we see that sort of like towards the second half of next year?

We are going to convert to, let's say, more normal level of provisions at the end of next year. But you also should consider that we gradually are changing the composition of the portfolio to a more retail one. So even with the same line-by-line, business-by-business level of provisions, the mix are going to conduct to a slightly higher structural cost of risk due to the relative weights of the business.

And so we understand what you're saying. So that, say, 2018 and 2019, your cost of risk was 1.5%, 1.6%. Given the mix composition shift, what would be the equivalent cost of risk?

I would say a little bit more than that at the end of the next year. But it is going to be a gradual recovery, as Reynaldo mentioned. It's not an on-off effect. You have a significant impact in the second quarter of this year. And after that, a gradual improvement.

So you think your cost of risk, say, 2022, which hopefully, knock on wood, is a normal year is around 2%. Is that kind of like the way I'm reading through the lines?

We expected that it's probably a little bit lower, yes.

Sorry, I didn't hear that. Sorry, what?

It will depend on how we -- go ahead, Mr. Rios, go ahead. Please proceed.

No. I was mentioned that this figure is probably in the upper bounds but for 2022.

Our next question comes from Thiago Batista, UBS.

Yes. I have 1 question on the margins. You already mentioned in the guidance that the government loans pressure your margin. So my question is, do you believe that when those programs end, the clients will be able to pay the same strata they used to pay before those government -- those programs. So will they trend to return to the normal level when all those programs are ended? And the second question is about Mibanco. When do we believe Mibanco should achieve the breakeven? And also, if you have any guess on when the profitability of Mibanco should return to the high teens?

Okay. First, regarding to the first question, I think when the government programs started to mature the second part of next year, you are going to have probably declines of clients. Clients, who have gone through the process, and they are starting to have normalized conditions. Clients that, even with this support has become insolvent or unable to continue in business. And probably a third category, the clients are going to need additional support probably in the form of longer-term facilities. And I think that the clients in all the categories are -- have understood that these are special conditions. And when the facilities mature, they need to pay market rate condition based in the performance size on credit risk. That's our belief. This is a very special condition with government coverage with special funding. I don't know if it helps you for the first question.

No, very clear. Very clear.

Okay. And for the second question, Mibanco is starting to recover also as we mentioned. The difference is that the government programs and the reprogramming started a little bit later after BCP. In fact, the government programs favor the Mibanco clients, mainly in the second wave. The first wave, there were very few clients who could apply to that. So the recovery is going to happen and it's little bit after the case of BCP. And next year, we are going to have a much more positive year, of course. But the delay is going to be -- the recovery is a little bit slightly delayed in the case of Mibanco for the reasons I already mentioned.

Our next question will come from Tito Labarta, Goldman Sachs.

First question, following up on your margin. I would have expected a bigger increase in your margin. Just if I remember last quarter, you had some frozen installments. That had about a 70 basis points impact, if I remember correctly. But your margin didn't really recover that much this quarter. Is that mostly because more Reactiva loans? Or just to understand the dynamics here. Maybe another way to think about it, if you exclude Reactiva, what would your margin have been without the Reactiva loans?

Actually, I think the margin has pan out as we expected, and we -- as we provide guidance. I will differentiate 2 different things. One is the NIM and another thing is the net interest margin as a figure, as a number. In terms of NIM, the figure is significantly diluted by the huge amount of Reactiva that has almost 0 margins or very low margins and was designed in -- as such because it was a relief program. This is one part of the equation. But thinking in the net interest margin, it has been impacted and is recovering and the dynamics were explained. You have a sudden shock of lower reference rate, more than 200 basis points were a significant part of the cure that has a severe impact in your short-term facilities and you have also some smaller portfolios in the case of -- slightly smaller in the case of BCP and a smaller portfolio in the case of Mibanco. So you have these impacts in the amount of margins. When we start to originate at a faster pace, a process that has already begun, we are starting to recover volumes with higher margins and the NIM is going to expand. The other impact that was positive this quarter was the reduction in interest expenses and the effect of this reduction is going to be carried out in the next quarters because a significant part of the process has already been done but throughout the semester. I don't know if this helps.

Yes. No, very helpful, sir. So then maybe leading to my second question. To get back to the high teens ROE that you expect by 2022, what's going to be the main driver of that? Is it -- are your margins normalizing? How dependent will that be on higher interest rates, on the Reactiva loans coming off your books? I know there's a lot of moving parts but then also provisions normalizing, maybe cost cutting, just to get a sense of what will it take to -- what do you think to get to that high teens ROE by 2022.

Yes. But I would like to emphasize that it's in the second half of 2022, not for the whole 2022. That's a relevant position. And the effects are going to be three-pronged, I would say. The markets are going to be recovering in line with the factors we have been discussing. The provisions are going to be much more normalized and more in line with the new portfolio composition. And we are starting to gain efficiencies due to the execution of several initiatives. It should be noted that the general profitability is impacted in BCP, Mibanco and in all banks of the world for the slower or lower interest rate. This impact is compensated by a number of measures but it makes a difference on our books. Our assumption is that the interest rates are going to be still lower in 2022 locally and internationally. So we expect to regain levels of profitability but with an underlying, less profitable basic margins in several business due to lower reference rates.

Great. That's helpful. So just to clarify on that last point, you can get back to the high teens ROE with interest rates where they are today. Is that correct or do you need interest rates to increase?

I think we can go with a lower interest rate, but applying these 3 kind of levels that I have mentioned, we see our portfolio composition higher efficiencies.

Our next question will come from Jason Mollin, Scotiabank.

You've addressed my questions on margins, provisions, costs. But maybe I can ask a general question on the outlook and the risk to that outlook of 9% to 12% real GDP growth and everything that follows on there. I mean, you've been seeing the reactivation. We've all seen it. It's been pretty impressive but from very low levels. What could derail this recovery? I guess, is it a second wave of COVID and lockdown? What are the risks to this outlook, I guess? If you could help frame that. And how is the group preparing for this kind of scenario, a negative scenario?

Okay. I think 1 significant risk, and I think not only for Credicorp, but globally, is a second wave, not only in terms of infections, but mortality that will lead to some kind of shutdowns, with severe impact in the business activity and the economic activity in general. This is one risk we are monitoring. One encouraging sign but we take very prudently is that already the level of infections in Peru are very high. Different studies conducted using different statistical methods suggest that the level of infection in Peru can be as high as 50%. So even if we have a second wave, probably the impact in total infections and mortality should be attenuated. But this is a significant risk. And the other potential risk is the changes in regulation. We'll have the elections next year. So far, it's not absolutely clear who is going to be the winner. We need probably much more time to have a clear picture. We don't expect significant changes but the risk is already there. The best protection for our business is to have a very strong balance sheet, solid capitalized business, prudent risk management and efficiency that allows us to operate even in less favorable environments and we are working in this direction.

Mr. Correa.

Okay. Thank you. There was a question related with projections in insurance, and maybe it's important to note that Peru has been one of the countries with the highest infection or contagious rate of the world. And that fact has 2 sides. On one side, the high number of deaths, as you have known, had a direct impact in our life portfolio. We have an important portfolio in Peru. And it includes individual life, mortgage protection, insurance, credit life insurance, in general, for credit cards and small business owners and across all the social economic segments. But on the other side, with that high contagious rate, we expect to be reaching a growth of regular rates in a few months. For example, during July and August, we have the highest levels of deaths in our country, more than 2.5x against the others. And during October, the number of exits of deaths was similar and inclusive, lower than the number we had in March, the first month of the pandemic in Peru. So we expect this trend continues in the same direction. And for the net loss, we can see -- there is no reductions in the life loss rate. And in the P&C business, the situation has been the opposite, where the lower economic activity and important part of our clients working at home office decreased the claims frequency and reduced the loss rate.

Our next question will come from Geoffrey Elliott, Autonomous.

Can you help us a little bit more on net interest income? When is that expected to start inflecting and moving up on a recurring basis? Is that -- is that going to happen soon or do we have to wait a little bit longer?

So I couldn't hear the first part of the question very clearly.

The net interest income, when is that expected to start increasing again on a recurring basis once we've taken out the one-offs?

Actually, the process has begun in line with what we already explained. What's happening is that this is starting to gain traction, but at a lower base due to the reduction in reference rate.

So just to be clear, so even though originations are going to be lower than they were pre-COVID, you still think that the 4Q recurring NII can be higher than the 3Q recurring NII?

Just because you start with a lower base of the case, at the level of origination that we are increasing the structural portfolio in BCP and in Mibanco. So it's starting from a lower base. I repeat this point, we are expecting to build up structural portfolio at this moment.

Our next question will come from Brian Flores at Citi.

I want to know how are you thinking about structural NPLs going forward, particularly just second half of 2022.

Well, we'll probably see in the following quarters an increase in the NPL levels because actual problems in loans will appear in all segments, especially in SME and individual clients. But on the other side, we'll see an increase in the level of write-offs. Remember during the first half of the year, due to regulatory constraints, we didn't make the adequate levels of write-offs. So I mean on -- the net effect, because of the number of clients that were impacted, we'll see a shift to higher NPLs. But in general, it will be adequate levels of -- compared with what we are expecting today.

And just as a quick follow-up, is there any particular segment of your portfolio concerning you based on the trends you are observing right now?

There are no surprises in this quarter. We don't expect further surprises in the following quarters. I mean we see constant trends on both the SME and credit card segments of our portfolio. But not important shifts towards what we've seen as of today.

Our next question will come from Carlos Gomez-Lopez, HSBC.

I wonder if you could comment about the legislative initiatives that could impose either lower fees or capital interest rates at the Peruvian banks. We know that a proposal was proposed at the committee level. Where, in your current standing, what do you think the chances are and the potential impact on your numbers? And a small follow-up, do you have an additional process for Madoff in AFP? Is that a one-off? Or are there any outstanding contingencies that we might still see in the future?

Regarding -- this is Gianfranco Ferrari. Regarding the first question, there are some, as you said, some initiatives at the commission levels in the Congress. It's too soon to tell if any of these initiatives may go into discussion to the Congress, and if so, if they're approved and if so, how to calculate an impact. Our vision is that what Peru need is the financial system in Peru is the small financial inclusion and any of these initiatives would generate financial exclusion, which is exactly the opposite impact. As I mentioned before, it's too soon to tell if any of these initiatives will become a law and if so, what the impact would be.

And on Madoff?

This is Walter, regarding Madoff, we have made a provision because there was a recent ruling by the Supreme Court that was not necessary in our favor. So we decided to make this provision and we think that this is the end of it.

Our next question will come from Piedad Alessandri, Credicorp Capital.

I want to know regarding corporate case was announced. If you could give us a bit more detail on the -- those cases you announced on the energy and the airline sector and did you see any segment behaving riskier?

As you probably know, we don't get into specific details in terms of the specific cases of our clients in the corporate world. Having said that, we can confirm you that we have established all the necessary provisions to cover the expected losses in those 2 cases. And we don't expect any new big cases in the future, in the near term.

Our next question will come from Andres Soto, Santander.

My question is related to the deal that was approved 1 month ago regarding reprogramming and guarantees for retail loans, including SMEs. I understand, you guys already mentioned, you don't expect some material impact coming from that. By that, I understand that when you refer impact, you mean that on the balance, it will be a negative one considering that it reduces your net interest income versus the reduction that you will get via the guarantee in your cost of risk. So I would like to confirm that if that's the case that you're expecting that on the balance, that would be a negative one, even if it's a small one? And in the end, this will depend on what is the level of adoption or the enrollment that you guys take up in your client portfolio. Also curious what is the level of -- the rate of adoption that you are assuming, to expect that this is going to be marginal impact?

We expect to have, as we mentioned previously, a moderate impact. It's difficult to say exactly what because it depends on the level of adoption and the specific type of clients that are going to access the facility. But I would like to emphasize that you are going to capture the cost and benefits in different lines and probably in different moments in time. What do I mean? You are going to have an immediate reduction in margins and you are going to have a deferred benefit in cost of risk. I don't know if this helps.

Yes, it does. It does. The question more broadly speaking was -- this is voluntary on the bank side, but obviously, there is a lot of pressure on the banks in general to provide facilities to the clients. So I'm just curious of any percent that you can give us in terms of enrollment of clients in this new program?

Yes. I think we are going to do something that is sensible for us, but good for our clients, something that's an additional facility that some clients are going to benefit from. And in balance, I think the net effect is going to be moderate.

Just to complement what Cesar just mentioned. Because of the amount that was assigned, we don't see -- so these are 2 programs, sir. Because the amount that was assigned, we don't see that this is going to be a massive program. What's assigned by the government, I mean, on one hand. On the other hand, we basically restructured all of our portfolio, all of our clients that needed a restructuring facility. So we don't see a large demand on our client. Having said that, we do -- we're going to work proactively in helping the clients that need it and obviously, trying to balance that with demand to other relatively smaller financial institutions that have not been as proactive as BCP in restructuring their clients' loans before.

Our next question will come from Yuri Fernandes, JPMorgan.

I have 2. The first one is a follow-up in the margins. I understood the message that you should kind of stabilize or even improve a little bit from the current levels. But I have some doubt here regarding first, the overdue loans have increased, as I said, right, like the nonaccrual loans only pay for the increase. In the next quarter, in pricing, it's probably in 2021, in the first quarter in 2021. And you also have a new government program that I understood maybe the incumbent can be super vocal in the program because, again, you have a lot of renegotiations already, the size of the guarantee is not that big. But should that should be a headwind on margins, right? So I just would like to confirm that, yes, margins should move slightly up from the current levels and that we are taking consideration through -- in all moving parts, the nonaccrual loans and also the renegotiated program. And my second question is regarding allowances. I understand the charge-offs are not there, right, charge-offs need to accelerate. But you are in Latin America, the bank is the biggest delta in allowances, right. Allowance has almost doubled versus 2019 on your balance sheet. You have, I don't know, PEN 4.6 billion, PEN 4.7 billion in higher allowances than you had last year. And so far, the data in reprogrammation like in the release -- I think in the second half, you're okay, right? We have a 94% collection in BCP, 84% collection in bank renegotiated loans are coming down from the fee. So the point is how do we discuss 2021 reversions of those provisions not only a lower cost of risk structurally, but I don't know, the bank, the kind of reverting those provisions at some point.

I take the first one regarding margins. Of course, in our calculations, we are considering the non-accrual of the delinquent loans. This is going to negatively impact the margins, but it's part of the assumptions that we have when we provide this guidance. The process is going to be gradual because what's happening at in terms of -- in accounting terms is that you accrue and when you have the actual default, you reverse the accrual and this diminished the income. So this is going to happen gradually at us as what we have seen happening in the last quarter. I think as part of the question, I don't know if you have another complementary question regarding NIM, sorry.

No, no. That's mix. I was just trying to confirm because I understood the mix shift, but I was not sure about the nonaccrual loans, but thank you for your consideration, yes, it's clear for me.

Yes. Yes. And in terms of the levels of provisions for allowances for bad loans, as we have been mentioning throughout the call, I mean, the values have been positive, and the trends have been better than we expected, than we initially expected. Having said that, there's still at BCP, what we call 18% high uncertainty portfolio. These are clients that -- they haven't still started paying their debts yet. I mean, they are in their grace period. And in Mibanco, that number it starts around 43%. So I mean, the -- your question in regarding, if there's an opportunity for reversing provisions will depend on the performance of these high uncertainty portfolio in COVID situations.

And to add start to contribute to the answer of Reynaldo, for example, in the case of BCP in September, we have uncertain portfolio of around PEN 42 billion. Out of them, PEN 34 billion already had the obligation to make a payment during the month. For that reason, we talk about an uncertain portfolio because not every client has the obligation to pay due to the reprogramming process.

Our next question will come from Sergey Dubin, Harding Loevner.

Yes, 2 questions from my side. The first one, just a clarification. You mentioned something about capital ratios calculated according to Peruvian standards versus capital ratios calculated to IFRS, I guess, or some other standards. So can you clarify exactly what the differences are and what's the impact on these ratios? What you're showing on the presentation in -- on Slide 30, are those calculated according to Peruvian, IFRS, what is it? What's the difference? What is the impact? That's the first question.

Okay. Okay. Okay. So, okay. The difference is that in normal times, the difference between incurred losses and forward-looking provisions is very small, so we don't make any difference. To give you an idea, in the case of BCP, at the end of September, the difference between local and international accounting in terms of provision is PEN 1.2 billion, out of PEN 7.7 billion. So in local accounting, you have lower provisions and you have a higher profit. And this impacts the level of capital. I can be probably an accurate, but the difference can be around 50, 60 basis points. In the case of Mibanco, the relative difference is slightly higher. The difference is 100 -- PEN 500 million out of PEN 1.8 billion in provisions. So you have more than 300 basis points in difference when you consider local and international accounting. Does this clarify the question or if you have another position, I'm happy to provide.

Yes, I mean, partially clarified it, but I still have a question. So what you show on your presentation in Slide 30, basically, these graphs show that BCP stand-alone CET1 is 11.5% and Mibanco CET1 is 16.5%. Do these -- are these numbers that you show calculated according to IFRS or calculated according to Peruvian local standards?

No. It's local accounting. It's local standards. So the IFRS number is going to be lower.

Right. So for Peruvian, that's what you show. For IFRS, it would be 11.5%, minus 60 bps. And for Mibanco it would be 16.5% minus 300 bps, correct?

Yes, more or less.

Okay. Okay. That's helpful. My second question is on Madoff. Again, this is one of those issues where I'd like to have some more clarity. It looks like you have 79 -- or 71.9% legal contingency. How did this contingency arise? And the Madoff scandal happened 12 years ago. Why are you just showing this now? What's -- can you give some background, some color around this whole issue?

Sure. Sergey, I'll tackle the Madoff. This relates to Mr. Picard, who is the court-appointed liquidator and us trying to articulate clawback initiatives against some of the investors, I guess, it's about 400 of them. And recent rulings by the Supreme Court of the United States are not favorable to the utilization of the extra territoriality arguments. So we found it prudent to create provisions just to get over that.

So does it mean that you would do -- just to clarify, does it mean that you will be liable? Like you think that in a worst-case scenario, you could be liable to pay PEN 72 million to this court-appointed trustee for Madoff case? Is that correct?

Well, we hope not, but we thought it was prudent to create the provision.

Okay. And I guess it's arose because you advised your clients to put money in Madoff, and then the client is now suing you asking you to make up the difference essentially, right? Is that -- how did that originate in the first place?

Yes. No, no, you've got it completely wrong. This is the clawback provisions. Clawback provisions, the concept here is that the court-appointed liquidator is asking people that have had withdrawals from the funds in months or certain short period prior to the whole scandal being blown up. That those people that withdrew funds from this fund did evaluations that did not reflect reality. Therefore, they apply clawbacks, which means you send the money back to the fund, the fund incorporates into the fund and then redistributes to money back. We have long, long time ago settled with our customers and we bought them back. We bought back the investments at a certain negotiated value to them. So our customers have nothing to do with this. And since we bought back the participations in the fund, the clawback applies to us.

Okay. Okay. Now that you said that that's very clear, but before, it was very vague. So now that you clarified it, it's good. Okay. My last question have to do is the -- sorry, the assets under management in your fund management business. You said that it could be something like PEN 3 billion withdrawal. And you just again explain why that is? Is it because of the new legislation that's being passed? Can you just give some more color around that particular issue?

Okay. I'll take this. We are talking about Prima. Prima is a private pension fund manager. And there is a new legislation that allows clients to withdraw funds. The estimated amount of the new funds that are going to be withdrawn according to this new legislation is PEN 3 billion.

Okay. And is that, again, is this a worst-case scenario that if everyone who can withdraw choose to withdraw? Or is that your estimate of a realistic scenario, or what...

It's our best guess. It's our best estimate. It's our best estimate based on the [indiscernible the previous withdrawal authorization process. It's not the first one.

Got you. And this will be how much of the total AUM? What percentage?

It's going to be -- sorry?

I think it's about 7%.

Around 7%.

Sergey, just an additional clarification, please refer to the 20F where all this Madoff stuff is fully disclosed here, and you can get a lot more information there if you need to.

And now I would like to turn the conference back to Mr. Walter Bayly, Chief Executive Officer, for closing remarks.

Thank you, Shantal. Well, thank you to all of you. This has been a very encouraging and important quarter for us. We now have shown the market evidence of what we have been perceiving in the last couple of months, mainly that we are past the inflection point on several fronts. On the health side, the numbers reported are very encouraging, and hopefully, they will continue this way. The economic activity and the second front is coming back to normal levels. On several fronts, we are already there and we estimate that by year-end or the first quarter of next year at the latest, we will be substantially on pre-COVID levels on a monthly basis. The third and last inflection point refers to the impact of this very severe downturn in our portfolio. The numbers we have shown for this last quarter clearly indicate this inflection point, and we continue to see improvements week by week. We are emerging from the most severe economic crisis in the last 100 years. And we think it is unrealistic to expect the situation to get normalized in 1 or 2 quarters. The recovery path is laid out. And as we have indicated, our guidelines -- in our guidelines, we expect to be back to returns on equity in the high teens in the second half of 2022. Our franchises continue to be strong. The path to change and evolve our business models and distribution to digital technologies has accelerated and continues to be a key element of our strategy. We thank you very much for your continued support. And with this, I conclude our quarterly phone call. Thank you very much.

Thank you very much, ladies and gentlemen. This now concludes today's presentation. You may now disconnect your phones, and have a great weekend. Thank you.