伊塔烏聯合銀行 (ITUB) 2017 Q3 法說會逐字稿

Good morning, ladies and gentlemen. Welcome to Itaú Unibanco Holding conference call to discuss 2017 third quarter results. (Operator Instructions) As a reminder, this conference is being recorded and broadcasted live on the Investor Relations website at www.itau.com.br/investor-relations. The audio webcast works with Internet Explorer 9 or above and Chrome, Firefox and mobile devices, iOS 8 or above and Android 3.0 or above. A slide presentation is also available on this site. The replay of this conference call will be available until November 6 by phone on (55-11) 3193-1012 or 2820-4012. Access code 1636828#.

Before proceeding, let me mention that forward-looking statements are being made under the Safe Harbor of the Securities Litigation Reform Act of 1996. Actual performance could differ materially from that anticipated in any forward-looking comments as a result of macroeconomic conditions, market risks and other factors. With us today in this conference call in São Paulo are Mr. Candido Bracher, Executive President and CEO, Chief Executive Officer; Mr. Caio Ibrahim David, Executive Vice President, CFO, Chief Financial Officer and CRO, Chief Risk Officer; Mr. Alexsandro Broedel Lopes, Group Finance Director and Investor Relations officer; and Mr. Marcelo Kopel, Executive Director. First, Mr. Candido Bracher will comment on 2017 third quarter results. Afterwards, management will be available for a question-and-answer session.

It is now my pleasure to turn the call over to Mr. Candido Bracher.

Good morning. It's a pleasure to be here again to talk to you about our third quarter results. I will start on Page 3 of the presentation.

In previous quarters, we have started to present our 6 main structural challenges, which are: focus on clients, digital transformation, people management, risk management, internationalization and profitability. In the last quarters, we spoke about risk management and digital transformation. In this quarter, we will focus on profitability.

So turning to Page 4 now. Here we present some important information about our profitability. And we compare it to 2012 when we decided to implement the strategy that focused on: First, lower risk appetite, which resulted in a significant change in our credit mix over this period, especially for individuals and small and medium companies. Second, focus on less volatile sources of revenues such as insurances and services. And third, a continuous search for opportunities of efficiency improvement.

Implementation of the strategy was a key -- was key for our performance and enabled roles in the value creation above cost of capital to our shareholders even in an adverse scenario for credit operations as we had during this period. When we compare our year-to-date performance this year to what we delivered in the same period of 2012, we see that our recurring net income expanded 76% leading to a 230-basis-point expansion of our ROE and culminated in 163% increase in the value created to our shareholders. Last, but not least, the total dividends and interest on capital paid, provisioned or reserved in the stockholders equity increased 484% and reached BRL 11.4 billion. We will talk about this subject later in this presentation.

So now before I move to the next slide and start talking about our financial results, I'd like to make a few general comments. In this quarter, we observed a relatively calm mood in the market, following the political turmoil of the second quarter of the year. We now see more positive signs in the economy. The base interest rate, Selic, is continuing its downward trend. There is a positive shift in the unemployment rate, which started to go down. And we also see lower inflation and consumer deleverage, which create a positive scenario for credit demands.

Now on Slide 5, we highlight the key performance figures from our third quarter results. The bank posted BRL 6.3 billion of recurring net income in the quarter and BRL 18.6 billion in the 9 months of 2017. This latter representing a 13.9% increase when compared to the same period in 2016. The recurring ROE in the 9 months of 2017 was 21.7%, which represents 170 basis points increase when compared to the same period in 2016.

Focusing on these first 9 months figures compared to the same period of last year, it is important to mention that we had a 3.2% decrease in our financial margin with clients. But this was more than compensated by: First, a 28.1% reduction in cost of credit, a 4.1% increase in commissions, fees and result from insurance, and a 0.9% nominal decrease in noninterest expense. We will comment on the operational trends of our results in the following slides.

So now on Slide 6, we present the same figures in the form of an income statement in which we look now to the right side, the last column. We can notice a small contraction on operating revenues. This 1.3% reduction was mainly due to the impact from the reduction of the Selic rate over our liabilities margin, and remuneration of our capital, and also due to lower credit volumes, which have affected our financial margins with clients. These effects, as we said, were more than compensated by a substantial decline in cost of credit, and increase in commissions and fees and a strict cost control. The combination of all these results led to a recurring net income increase of 13.9% when compared to the same period of the previous year.

On Slide 7, we break down this profit and loss between Brazilian and other Latin American operations. On the other Latin American operations, we have 2 different dynamics which are important to differentiate. First, we have the more traditional markets where we fully own the operations, Argentina, Uruguay and Paraguay. And here, we continue to see a good performance of this.

Now on the recently merged Chilean and Colombian operation, which were merged 1.5 year-ago, we see a more difficult than expected macro economic scenario and the natural challenges of a merger this size. These are impacting the profitability of the business. Nevertheless, I'd like to highlight that we have seen operational improvements with a good evolution on the integration of our clients, of our branches, systems and teams.

Now on Slide 8, we present a chart that reflects our business model in which we break down the consolidated income statement between income earned in operations that bear credit risk including related fees, income from trading operations that bear market risk, income from insurance and service operations and excess capital. What I like to stress in this slide is that in these 9 months, besides the essential focus in insurance and services, is the fact that our credit operations already reached profitability in line with cost of equity. We believe that this should be a sustainable trend for the coming periods, which is quite different from the past few years. And we see there on the bottom, we have a recurring ROE of 14.1%, which is a bit above our credit cost of capital, which is 14%.

Moving to Slide 9 now. We can see that the total consolidated credit portfolios continues to show some contraction. Nevertheless, we note now with satisfaction that individuals portfolio on the first line stopped decreasing. In the Wholesale segment, companies still show a moderate credit demand and are searching for alternatives in the debt and equity capital markets, which result in a contraction of this portfolio.

I'd also like to comment that in this quarter, we sold to the market some BRL 800 million of assets, which resulted in a positive impact in our net income of approximately BRL 60 million.

On Slide 10 now. We see that despite the reduction in the Selic rate and its impact on the liabilities margins and our working capital, the risk-adjusted net interest margin, the blue line, increased 10 basis points in this quarter reaching 7.4% mainly due to lower cost of credit.

Slide 11, we present the evolution of our margin with the market. The financial margin with the market was mainly impacted by the reduction of the Selic rate as anticipated. Looking ahead, we expect this margin level to be around BRL 6 billion for this year as forecasted.

Moving to Slide 12. We see a stabilization in our delinquency ratios, both in the NPL 90 days and the NPL 15 to 90 days. Regarding the 90 days NPL in Brazil, we see that it continues to improve in all portfolios. In the 15 to 90 days NPL in Brazil, we can highlight the individuals portfolio performance that reduced 20 basis points this quarter. On the other hand, the observed increase in the 15 to 90 days figures in Latin America is caused by collection, operational issues in the individuals portfolio in Chile and Colombia and is not related to the growth in 90 days numbers in Latin America, which is associated mainly to the assets in the portfolio.

On Slide 13, we present the evolution of our NPL creation. There was stability in the overall level of the NPL creation in the quarter with a slight reduction of BRL 45 million when compared to the previous quarter. This is the lowest level since March 2014. Operations in Brazil were the main driver of this reduction.

Slide 14, we present the evolution of provisions for loan losses and cost of credit. In this quarter, we saw significant decrease in the provision for loan losses in Brazil, in line with the delinquency downward trend. As a result, the relation between provisions and our average loan portfolio reached 3.6%, which is the lowest level since the first quarter of 2009. Cost of credit also reduced despite an increase of BRL 157 million in impairment charges on corporate securities, reduced to 2.7% as you see in the lower chart.

Moving now to Slide 15. We present our total allowance by type of risk. The Bank's overall level of allowance decreased in the quarter from BRL 37.4 billion to BRL 36.6 billion. This reduction was in line with improvements in our credit quality performance, both in Wholesale and Retail portfolios in Brazil.

On page -- on Slide 16, we come to our coverage ratio. Here it's -- we see that our coverage ratio has reached now 246%, which we consider to be a high figure. Here I'm frequently asked about what is the trend of this coverage ratio? I was asked this question when the coverage ratio reached to 231% at the end of the first quarter, and I said that my expectation was that this rate would drop only to see the increase to 243% and 246% now. So my answer now would be, I still expect this coverage ratio to drop. I think that no doubt the medium-term trend here is for this to drop. This is a very high level of coverage ratio. But what causes this rate to drop is one of two things. Either the companies to reach, we have made precautionary provisions, actually default. And these provisions are moved to the normal provisions of the NPL rate increases, and the coverage ratio decreases with no impact in our result. As these companies improve, and we are able to revert the provisions, in this case, also the coverage ratio decreases with a positive impact in our results. But in the short-term, it may happen that none of this seems to happen. Neither the companies default nor they improve to a point where it can reverse the provisions. And if we couple this with a decrease in the 90 day NPL, we will have an increase in the coverage ratio. This is what's been happening in these past 2 quarters. It may still happen in the short-term, but the medium-term trend, no doubt, is for a reduction in this coverage ratio.

If we move to Page 17 now -- Slide 17. Commissions and fees, we see that they have increased 5.5% when compared to the first 9 months of 2016. We can highlight the good performance of asset management fees in line with the solid growth of our assets under management in the period. The decline in the result from insurance is partially explained by the stay of our group life insurance to Prudential that became effective in April this year.

Now on Page 18, we show the evolution of our noninterest expenses. Comparing the first 9 months of 2017 to the same period in 2016, noninterest expenses decreased 0.9%. If we exclude the impact of some extraordinary events in 2016, noninterest expenses would have grown 1.9%, which is still below the accumulated inflation of the period.

Now to Page 19, I'll comment now on our capital ratios. As of September 30, our Common Equity Tier 1 considering the full application of Basel III rules reached 14.6% already including the impacts of the Citibank retail business acquisition and the acquisition of a minority interest in XP investments. Since our stated aim is to have a 13.1% Tier 1 ratio -- sorry, 13.5% Tier 1 ratio, we have reserved 1.1% of our capital, which may be used to increase dividends or even to make share repurchase going forward.

This is better explained now in Page 20. On September 2017, we have announced changes in our capital management practices aligned to our value creation strategy. First, we have excluded the maximum payout limit that used to be 45%. And we maintained the practice of paying dividends and interest on capital of at least 35% of net income. Second, we set forth that the total amount to be distributed each year will consider 5 elements: First, the actual level of capitalization of the bank; second, the minimum level of Tier 1 capital at 13.5%, which must be composed of at least 12% of core capital, 1.5% may be in the future additional Tier 1 capital; third element to be considered, the profitability of the year; fourth, the expectations of capital use based on expected business growth, share buyback programs, potential mergers and acquisitions and regulatory changes that may change capital requirements; and five, changes in the tax legislation. So the percentage to be distributed as payout may change every year based on the profitability and on capital demands.

The table on Slide 20 shows -- and looking -- our forward-looking estimated -- estimation of what the payout ratios may be depending on ROE and risk-weighted assets growth. So it is important to highlight that we do not intend to have capital surplus in excess of the levels established without the prospect of using it. And that any possible surplus will be returned to our stockholders.

Moving to Slide 21 now. Talking about our forecast. We reiterate all ranges for the year. Having said that, as mentioned last quarter, we believe that we will deliver a credit portfolio growth, which will be around the bottom of the forecast range. For financial margin with clients, we expect that we will end the year also around the bottom of the range. For cost of credits, as mentioned last quarter, we expect to end the quarter close to the top of the range. Commission and fees and result from insurance operations, we estimate now that we will end the year next to the top of the range. And to finalize, for noninterest expenses, we believe, we will end the year around the bottom of the range.

Moving to Slide 22 now. Last week, the Brazilian Central Bank authorized the acquisition of Citibank retail operations in Brazil with the completion of all regulatory approvals, the financial settlement of the acquisitions is happening today. This operation adds around 300,000 clients to our client base and about BRL 9 billion in assets and BRL 6.2 billion in credit portfolio and close to BRL 5 billion in deposits. With this, we reaffirm our commitment to the Brazilian market and value creation for our shareholders. In addition, I'd like to take this opportunity to welcome all new employees and customers who joined Itaú Unibanco at this moment.

So finally, and before concluding and going to the questions and answers, I'd like to extend here a special complement to Marcelo Kopel, who has been our Investor Relations Director for many years and has always shown a level of care in this relation of transparency and quality of information that led to our investor relations having been repeatedly recognized as one of the best in the market. Thank you, Marcelo, congratulations. And I'm sure now that Alexsandro Broedel, who is taking over, will be able to honor this position and take the area to an even higher level of recognition.

So with this, I conclude my presentation, and we may move to the question and answer.

(Operator Instructions) Our first question comes from Eduardo Rosman, BTG Pactual.

I have 2 questions. The first one is on your credit department, right. We can see in the presentation that the ROE of your credit department is finally back to cost of capital levels, right? But as far as I remember, in previous presentations, you mentioned that you expected the divisions post an average returns close to cost of capital, while the value creation would come from other more asset-light areas, right, such as fees and insurance. So I wanted to -- but I wanted to understand, why shouldn't we expect ROE in the credit department to be above cost of equity, at least in the coming quarters or in the next couple of years, right? At the end of the day, you're coming from, let's say, almost from a depression, right? The economy suffered a lot and you reported, let's say, very low ROE in these divisions for a while. Lending mix could improve your margins, right? Risk-adjusted margins in the large corporate segments are probably going to be higher. Banking concentration increase, government-owned banks therefore have capital. So I wanted to understand, why can't we expect, at least for a while, high ROE in this division, right? And the second question would be on REDE. If you can talk a little bit about REDE, right, the acquiring business and the strategy there? Because when we see the results of Cielo and Santander, they are out, right, we can see they already underperformed, again, in terms of market share, of both card volumes and POS machines. So I want to understand what's the strategy there? I know that the company just changed the CEO and the profitability remains healthy. But wanted to understand what is the go there? If you expect to keep losing market share or if market share -- you can recover market share. So if you can give a little bit color on the strategy there it would be helpful.

Okay. Thank you, Eduardo, for 2 very good questions. In the first question, yes, I think you're right. I think, we may expect to see the return on equity for credit above the cost of capital for some time at least. Here it's important to differentiate the Retail portfolio from the Wholesale portfolio of credit. In the Wholesale portfolio of credit, which is taking longer to grow, but we believe, I mean, this is coming, as you know, the Wholesale portfolio is facing a lot of competition from capital markets in Brazil, which are heavy and active year. By the way, if you look in our line of fees and commissions then see that the fees and commissions from advisory have grown considerably due to this growth in capital markets. But anyway, in the Wholesale segment, we know from experience that -- I mean, when it starts to become profitable, capital markets, competition from other banks, from foreign banks kicks in. So I think here it's difficult to expect that we will have a much higher return on equity than the cost of capital. But we may have a high return on equity than cost of capital for some time, especially for the next quarters. In Retail portfolio, I am more optimistic. Here I believe, we can maintain a higher return on equity than cost of capital for a longer period. I think you're right here. And I think this may be one of the sources for next year that will compensate for the reduction in the Selic rate. Your second question on REDE. Yes, you're right, I mean, we continue to lose market share on REDE and we continue to suffer, I mean, from a market risk and new entrants, some quite significant price pressure and competition. And here, I mean, we are making changes in our strategy, as you said, and we have nominated a new CEO to the company. And the new strategy, it essentially consists in clearly separating the 2 roles. The roles of the companies, which are accounts holders visit Unibanco from the companies which are open (inaudible), which are in the open market. So we're having now different strategies for this group of companies. Two of the companies which are account holders we are having, our offers joint with the offers of the products in the accounts. This is already starting to show very good results. And we have higher than we're structuring whole new sales platform to deal with the companies which are not account holders. So it's, of course -- this takes some time to generate its effects. But we are confident that the performance of REDE will improve with these measures.

The next question comes from Domingos Falavina, JPMorgan.

I have actually 2 questions. So the first one, I just want to make sure I'm understanding correctly, what the BRL 11.4 billion in dividends, a part it's only -- or a provision the other part actually still just under consideration for paying. Those -- these BRL 11.4 billion refer to the 9 months? Or is it an expectation for the full year? Because if it is for the 9 months, I'm getting to 63% payout on the earnings? I just want to make sure the number makes sense and if that's the number we can work for the full year? And I'll make the second question later.

Okay, Domingos. Thanks for the questions. Very important that we clarify this. Yes, the BRL 11.4 billion in dividends, which have been reserved for dividends refers to the 9 months. And you said 63%, I mean, we expect the payout to be around 60% for the year as a whole.

Perfect. Super clear. The second question is just a follow-up on the comment you made regarding the coverage. Super clear when you describe the mechanics of the coverage ratio and the fact that they have on certain corporates allocated to them. And either you have them going delinquent, in which case, these coverage will decrease? Or getting better, in which case you reverse. And I understand those are very definitions of 2 out of the 3 provisions you have in there, right? The specific, the generic, but not the additional provision just to have BRL 10.7 billion of additional -- or complementary provision, as you name it, and you could in theory use some of those in previous quarters. So my question is, just like even looking at the coverage ratio, is there a point where you start lowering these excess additional provisions? Or you'll just let this coverage ratio continue to rise?

Thanks for the question, Domingos. The mechanics are described. I think they apply especially to the additional provisions, to the complementary provision. How is this provision made? First, this provision, in Retail, it's made according to models, okay? But as you see by the coverage ratio in Wholesale, which is over 900%. In Wholesale, it is made on a case-by-case basis. So we look at individual companies, which are not defaulting, which risk-weighting is still reasonable, but which have somehow been affected by Lava Jato scandal or are in this value chain. We look -- so we have an expected loss model. And based on this model, we establish this complementary provisions. And then, we follow up these companies quite closely. And it is with these companies that now we look at them, we see some improvement, not enough in most of them, in many of them. But not enough to revert the provisions yet. Orders received, they're still not out of the woods, I mean, and that they may default anytime in the next quarters or so. So we are following them closely, and expecting that as economy grows, they will possibly, I mean, improve and a part of this provisions, we'll be able to revert.

The next question comes from Jason Mollin, Scotiabank.

I have 2 questions. First, on loan growth expectations and drivers. Can you talk about what it will take for Itaú and perhaps the banking sector in general to recover loan growth in the main loan segment? And my second question is a follow-up on Slide 8, the profitability that you show broken down into credit, trading and insurance and services as well as excess capital. We see insurance and services contributing the largest value for the bank. Can you talk about the specifics in that contribution? Credit cards, I think an important contributor is probably the deposit franchise, the deposit spread. I think it'll be interesting to understand the drivers there. It's hard for us to reconcile those numbers in that table from your financial statements.

Thank you, Jason. So thanks for your question. On loan growth expectations. I think here we have essentially 2 different dynamics. For individuals and for individuals and SMEs, may be some dynamic, and in the Wholesale segment and the middle market and large companies, it's another dynamic. In individuals, as I note, we are already seeing some growth. And here, I think that the reduction in unemployment and the reductions in the level of leverage of the individuals coupled with the reduction in interest rates, which we are witnessing will be a powerful tool for growth in this portfolio. For the large companies, I think, they seek to have more of a confidence level in the future. So I think that here, next year elections are still a problem in terms of investment decisions, which would led -- which could lead to a more robust growth here. I think we are starting to see growth. There will be some growth in this portfolio, but I think, still modest for the next quarters. And of course, we have, as I mentioned, the very intense competition from capital markets here. Capital markets are very active in that and equity capital markets this year in Brazil. And so many of these large companies are preferring to borrow in capital markets where they get better rates since capital markets normally can have longer maturities and don't need to allocate capital to these loans.

Do you think -- Candido, do you think that in this competition from capital markets that the fees that the banking sector, including Itaú, will earn? Will -- can it fully compensate the shift to capital markets or partially? How do you view that?

No, I don't think it compensates, Jason. If you see on page -- I'm trying to find it, page here was commissions and fees -- Slide 17, you see advisory services and brokerage, which in the 9 months this year accounted for BRL 1 billion. So this is basically capital markets activity. So this is what we've been making there in a good year, as this year has been. So I don't think it compensates. But I think, I mean, especially, as the companies below the level of large corporate pick up on borrowing, I mean, these companies have less access to capital markets, and middle market and corporates and on so this should provide a growth in income from loans. Now your question on Chart 8. In fact, I mean, it's not possible to reconcile from our financials, I mean, this table 8. I mean, this is why we have devoted so much time and work to making this and seeing our business based on the type of risk we run when we make everything. For instance, just to give an example, in commissions and -- in this column of commissions and fees, you have also income, which is -- does not bear specific market or credit risk, for instance, the over hedge, it's there. Credit card, a part of it is there, but the risky part of credit card is not there. The part of credit card, the credit risk is not there. All the cash management is there. You are right, when I mentioned, I mean, liabilities management. It's an important part of this. All the asset management fees and private banking fees are there. So you have really to collect data from many different points in the balance sheet to compose this chart.

And is there any way to give us, maybe, just a sense of what are the top 3 contributors to that? Is it in the order you suggested or that you stated, not that you're suggesting that? Or is it somewhere else? I mean, how -- what are the largest 2 or 3 contributors to that, that part of line, if I may?

Out of the top of my mind, I couldn't tell you, but Alexsandro will supply this information, later. Okay?

The next question comes from Mario Pierry, Bank of America Merrill Lynch.

Let me ask you 2 questions as well. First one is related to your cost of risk. You mentioned on your presentation that you're already running at 3.6% of average loans. This is already the lowest level since 2009. However, when we look at your loan mix, they -- it's a lot more less risky than it used to be in the past. So I was trying to get from you an idea of where do you think this cost of risk can fall to in the next couple of years, especially in light of the recovering economy, low inflation levels, less leverage consumer? Second question has to do with your excess capital and your international operations, in particular, in Argentina. You show on Slide 4 that you want to keep the same level of profitability in your international operations and quality of operations internationally that you have in Brazil. When I look at your Argentina operations, you are too small in the country. I was wondering if there's already a strategy that you see -- that you can get excited about the growth potential in Argentina and you'd like to get bigger. Or is this something that you're considering divesting in the future because you're too small and you don't have the scale to compete?

Thank you, Mario, for your questions. In cost of risk, I think, it's risky to make any forecast on how low it can get. But it can get lower. Yes, I mean, my expectation is that it will get lower and especially, I mean, if it is -- it is impacted by provision reversions in the future then it can really get significant lower for a short period of time. I mean, we have witnessed that when going out of crisis in the past, I remember, specifically after the crisis we had in 2002, the years of 2004 and 2005 were years with very, very low cost of risk because of provision reversals. So I think, I mean, the trend is for it to go down, but I wouldn't risk saying how low it can reach. But I would say that sustainably, it's not much lower than that. I mean, if you look at a longer period of time. Now regarding our excess capital, and if we should use it to make acquisitions, and you mentioned Argentina. Here we have, of course, we are enthusiastic about the Argentinian market. We are seeing the improvements -- the general improvements there in the situation. It's another bank market we looked at with optimism. Having said that, there are 2 things which make it difficult for us to grow by acquisitions in Argentina. One, is the very high valuation of assets, in general, in this Latin market. And the other one is a tax question. I mean, we -- as you know, we have corporate rate tax of 45% here in Brazil going to 40% from 2019 on. In Argentina, the tax rate, if I'm not wrong, is 30%. So we pay 40% independently; if we [fluctuate] the profits or not. So whenever we are bidding for an acquisition in this markets and we have to make our discounted cash flows, we start at a huge disadvantage against other competitors. And I mean, if we were to pay the prices asked, we wouldn't come -- we wouldn't reach our cost of equity here. And I mean, so this is against what there has been the discipline, in terms of cost of equity that we have been imposing on the bank. Having said that, we do not intend to sell, Argentina, right now. We think, I mean, we have stabilized the operation. I mean, the operation is starting to yield some return to the bank, and we intend to invest there, especially, with the digital technology, which we are developing here in Brazil and elsewhere. And we expect with this to be able to grow our operations there even if we don't make any major acquisition.

The next question comes from Carlos Macedo, Goldman Sachs.

I have a couple of questions. One is a question on your margins and margins took a big drop this quarter. Of course, there are a number of factors that you highlighted in Slide 10. In going a little bit further, I mean, your margins adjusted for risk at 7.4%, as you show it. I mean, last time they were this high was 7.5% in 2014 -- '15 without CorpBanca. And that was also the last time that your cost of risk was as low as it is now. As you said, you might have a little bit of improvement in cost of risk as you go forward, especially, if there's some reversal. But your margin, excluding risk, is likely to come down for manufacturers, not only the Selic, as you mentioned, but also the fact that your portfolio has shifted to be much more defensive over the last few years. Where do you think this adjusted margin can go? Is this mid-6% level? Something reasonable? Do you think it can stay above 7%? Where do you think it can go going out? Second question, going back to something you said before, Candido, where you -- most of the provisions, the excess provisions, as you call it, are on the corporate book, the 900% coverage, and that's based on your models and how to rate the companies. Can you give us -- is there any trigger that we can look at that will lead you -- that will lead your credit models to change their views and allow you to release these provisions? In other words, is the GDP -- the GDP expectations need to be higher? Does the cash flow outlook for the company need to be better? Is there anything that we can point to so that we can have a little bit of private estimate when these reversals will take place?

Thank you for your questions, Carlos. They are both difficult questions. No clear answers here. On the margins. On the margins, I expect in the near future, I mean, that we'll be able to sustain this level of margins above 7% after margin adjusted to risk. And this is -- I mean, as we mentioned, I mean, we have 2 conflicting forces here. One upward, which is the cost of risk, and one downward, which is the Selic rate. And as our portfolio grows -- and portfolio growth -- credit portfolio growth is essential for the profitability for our profits to grow next year. As the portfolio grows, I think, we believe, we may be able to compensate a part of this Selic downward rate and even if there is some spreads compression. And with the improvement in the cost of risk, I think that for still some quarters, we will be able to see this margin adjusted to risk above 7%. Really looking for an honest answer, in relation to the -- what are the -- what indicators you can look at to know, I mean, the trend in our coverage ratio? I would say, I mean, it's to look at the research reports and the economic pages of the newspapers after, I mean, the development of the large companies, which are known to be in trouble in Brazil. I mean, we work with -- there's practically no company which is not in our books. And I mean, I'm sure you can think of many companies, which EBITDA has started to grow again. But which still have a very high net debt-to-EBITDA ratio that are not defaulting. But when you look to the future, I mean, it's not quite clear how they will perform. These are the cases, which we are following closely. Most of them are improving slightly, but not enough yet for us to abandon the level of provisions we have.

Thank you, Candido. Just follow-up, when you say you're going to try to keep that adjusted margins above 7%, does that include potential reversals and provisions? Or is that a number clean of these reversals?

Yes. I think, these reversals are a natural part of our cost of risk. I mean, they have weighted on our margins as we built these provisions. So now they should relieve us a bit as we unbuilt them in the future.

(Operator Instructions) This concludes today's question-and-answer session. Mr. Candido Bracher, at this time, you may proceed with your closing statements.

I just want to thank you all for your interest and for your very good questions, which no doubt, I mean, help us elaborate our explanations about our business. Thank you very much.

That does conclude our Itaú Unibanco Holding earnings conference for today. Thank you very much for your participation. You may now disconnect.