伊塔烏聯合銀行 (ITUB) 2011 Q3 法說會逐字稿

Ladies and gentlemen, thank you for standing by. This is Itau Unibanco Holding conference call about the third quarter of 2011 earnings. At this time, all lines are in a listen-only mode. Later, there will be a question-and-answer session and instructions to participate will be given at that time. (Operator Instructions). As a reminder, this conference is being recorded and broadcast live on www.itau-unibanco.com/ir. A slide presentation is also available on this site.

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 those 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 Sao Paulo are Alfredo Egydio Setubal, investor relations officer, Sergio Ribeiro da Costa Werlang, executive vice president of Risk Control and Finance, Caio Ibrahim David, chief financial officer, Rogerio Calderon, corporate controller and head of investor relations, Marco Antunes, accounting director and Mr. Rodolfo Henrique Fischer, Itau BBA's vice president.

First, Mr. Alfredo Setubal will comment on the third quarter of 2011 earnings. Afterwards, management will be available for questions and answers. It is now my pleasure to turn the call over to Mr. Setubal. Please, sir, go ahead.

Good morning for those who are in the US, good afternoon for those in Europe. It's a pleasure to be back to talk about our third quarter results. And for those who are following through the Internet, we are starting on page two with highlights.

The first one is the result of the bank. We have a recurrent net income of BRL3.9 billion with an annualized ROE of 23.5%, an increase of 18.8% when we compare to the second quarter of this year. The net income was BRL3.8 billion when we reduced the extraordinary gains in this quarter.

The second point is the loan portfolio growth. The growth in the quarter was 6.1% when we compare to the third quarter of 2010. The number was 22.8%. We have to note here due to the devaluation of the real against the dollar, our dollar portfolio increased in reais. So if we offset this impact, the growth of the loan portfolio has been 3.5%, so in line with the growth of the credit portfolio of the financial system in Brazil. So, it was a good quarter in terms of portfolio growth.

The third point is the financial margin with the clients. The financial margin with clients achieved BRL11.8 billion. That means an increase of 5.3% from the second quarter. With this growth in the credit portfolio, so we were able to increase a little bit the financial margin with the clients in this environment that we are operating, especially in credit when we are going into lower risk segments.

Also in the financial margin, we see the financial margin with the market BRL1.1 billion in the quarter with a growth of BRL446 million when we compare to the second quarter. It's important to notice that, of course, the market was much better for the treasury in this quarter as was released by CETIP. We sold part of our stake with other shareholders, who are new investor, and this sale represented a gain of BRL151 million that is inside this BRL1.1 billion. So we have this extra gain in the treasury in this quarter related to the sale of CETIP shares.

The fourth highlight is the non-performing loans and results for loans and lease losses. The expenses was BRL4.97 billion with a decrease of BRL136 million when we compare to the second quarter. That was a quarter that we made a lot of provisions, especially related to small companies, if you remember.

And this quarter, we reduced the level of provisions because we don't think these are necessary due to the anticipation of provisions created by the expected loss model that we are using since the end of 2010. So we made prior provisions, so we don't need in our models to make new provisions in this quarter.

The 90-day delinquency ratio increased 20 basis points to 4.7%. Here, is important to notice, is the real still had the devaluation that increased our credit portfolio, as I mentioned. This number should be 4.8%. So, our guidance was for 4.8%, was a little bit lower, especially because of the impact of the real devaluation in our credit portfolio.

We continued to see a good performance in the delay, so we don't see further deterioration in terms of non-performing loans in this quarter. Of course, we have the reverse impact because the real is undervaluation against the dollar, so, of course, this will have an impact in the opposite direction than we had in the third quarter and maybe a marginal deterioration in individuals. So we expected the number for this quarter to be around 4.9%, considering these two factors.

The fourth highlight is related to efficiency. Our efficiency ratio achieved 47.5%, with a decrease of 80 basis points when we compare to the second quarter and 310 basis points when we compare to the third quarter of 2010. So we continue in the trend of increasing these efficiency ratios both from expenses control and revenue growth.

Non-interest expenses increased 5.7% quarter-over-quarter. Personal expenses grew BRL135 million, so we were able to continue the trend of reduction they accumulated expenses when we compare to 2010. In the first nine months of this year, the growth was 12.5% and we continued to believe that by the end of the year the total expense -- non-interest expenses will increase between 8% and 10% as the guidance we gave.

In terms of revenues, the growth was lower than expected. We were expecting more growth from in terms of revenues. That's the reason why the efficiency ratio is not at the level that we expect it to be at this time.

This lower growth was related to the pace of changing the mix of the credit portfolio. We are in a faster pace entering into lower risk credit portfolio, so we are going to have lower revenues, lower spreads also because banking fees in third are growing lower than we expected. Our guidance is 14% to 16% growth in banking fees for this year. And but we don't believe that this number will be achieved, so the new expectation that we have is for the whole year, 2011 of 13% growth in terms of banking fees and tariffs.

Going to page four, highlight some numbers. The operating revenues with a growth of 12.5% when we compare 12 months. Loan losses increased 18.8% to BRL10.5 billion and decreased 1.6% in this quarter. Non-interest expenses, the growth of 12.5%, as I mentioned, reducing from [19] in the -- by the end of the second quarter in the accumulated basis. And recurrent net income with a growth of 13.2%, achieving BRL10.9 billion in the nine months.

Highlights also on page five. Net interest margin, 11.6%, for reduction when we compare September 2010 and an increase of 10 bps when we compare to the second quarter of this year. So second quarter was 11.6%, then we achieved 11.7% going into a lower risk portfolio.

Banking fees, almost BRL14 billion, with a growth of 10.7%. And as I said, we expected now 13% for the year. And the ROE in the quarter, 23.5%, but due to the second quarter this was lower than that. The nine months was 22.4%.

On page six, the net interest margin [effects], as I mentioned, were increased in this quarter, was good news. On page seven, we can see the dynamics of origination of the credit portfolio. We already renew this year 57% of the -- with the portfolio and we still have increased last year 26%. So we are renewing very fast the credit portfolio.

On page eight, just numbers. The total assets of BRL836 billion, stockholders' equity of BRL69 billion loan portfolio of BRL382 billion, and total funding from clients, including here the AUAs, BRL869 billion.

On page nine, we see the results in this quarter. As I mentioned, important growth in terms of operating revenues of 6.7%, achieving BRL19 billion. An important impact of margin with the market of BRL1.1 billion in this quarter, BRL2.7 billion in the year. The second quarter, you remember that it was not so good.

And when we compare with last year, we see that results -- this line, is under the numbers of 2010 with a reduction of 2%. So with the reduction on expenses for loan losses and better control of non-interest expenses, we had this recurrent result of BRL3.9 billion.

On page 10, we see the loan portfolio by kind of client. We continued to see a very important growth in terms of mortgage. The portfolio is still small when we compare with other segments, but it's growing fast, for almost 15% in this quarter and 79% in 12 months. So it's gaining importance.

In terms of very small, small and middle market, the growth was only 2.2% due to the fact that we reduced the portfolio for very small and small companies, as we mentioned in the second quarter conference call, but we are reducing this portfolio due to the risk and the level of provisions that we were obliged to make in the second quarter.

So we continued to grow in the middle, very middle companies, but we reduced the portfolio in the other segments. And corporate also increased of almost 9%. Of course, here also with impact of the growth of the dollar portfolio due to the devaluation of the currency.

On page 11, the NPL ratio of 4.7%, should be 4.8% if we didn't have the devaluation of the currency. In terms of companies and almost stable, but we had deterioration in this quarter in the case of individuals when we increase the NPL ratio over 90 days, increased from 5.8% to 6.3%.

On the NPL ratio 31 to 90 days, we see numbers -- good numbers, no we don't see any big problem related to the increase of deterioration in the credit portfolio. We believe these numbers for the fourth quarter will remain something between 4.8% to 4.9%. We don't see big problems in this area. The coverage was 156% and the total provisions, BRL24.7 billion. Of this total, BRL5 billion is related to the expected loss model, the provision that we made before.

On page 12, we have the evolution of the expenses, [maneuver dissolve rate] we can see the reduction in the third quarter compared to the second quarter. And also on the right side of this page, we can see that renegotiation under the resolution of the Monetary Council, is 3.9% of our total portfolio. On page 13, the total funding and working capital and so on, BRL1.1 trillion.

On page 14, we can see that we improved the balance between funding and credit portfolio in this quarter. We increased the deposits, so we have moved [increased] level of the pre-deposits to credit. When we consider in this number the reserve requirements in the Central Bank, we have 75%. So, funding continues to be not an issue in Brazil. To fund the -- the deposits is not an issue to fund the credit portfolio growth.

On slide 15, we can see the banking fees. The growth in the quarter is 3.2%, in 12 months, 10.7% and here we expect it now 13% for the year.

On page 16, we can see the non-interest expenses, 5.7% in the quarter, 12.5% when we compare to last year. And we continue to be very strict in terms of controlling the expenses and we maintain our expectations for 2011 of growth of non-interest expenses between 8% and 10% when we compare to last year.

It is important to notice in this slide the evolution of the employees of the bank. We continue to reduce. We achieved almost 110,000 in March of this year and we finished with 106,000 in September of this year. These 6,000 are outside Brazil and 99,800 is in Brazil, which shows us a very good efficiency in terms of number of employees to the size of the operation that we have here in Brazil.

On page 17, the efficiency ratio. We continue to grow the level that (inaudible - background noise) of 41% in 2013. As I mentioned this year, we are in a lower pace than expected, especially because of the lower growth of our revenues.

On page 18, the evolution of BIS, very comfortable position, 15.5%. We still have some approvals pending on the Central Bank. When these numbers are approved, the subordinated debt, this -- remember we've increase to 15.7%. So, I think we had a very good quarter. Of course, the second quarter, as we mentioned, was below the potential of Itau Unibanco for specific reasons that we talked about in the last conference call.

But we were able to release a very strong BRL3.9 billion recurrent number for the third quarter. Our portfolio growth was for both the average of the system. We were able to grow the net interest margin. Delinquency and bad provisions clearly show us a stabilization in the levels in the short-term.

We are very confident in our models, in the expected loss model that we are working since the end of last year. So we are in line with the best profits. And when we compare to IFRS and Basel III, so we continue to be very confident that our models are working very well when we are talking about the provision.

In the terms of expenses, I think the numbers are showing a very good control and we continue to target this 41% by the end of 2013 and we expect to grow the expenses in line with the inflation in the coming year.

So in general, I think it was a very good quarter. I think that the research analysts and almost all of the research that we were able to read during these two days after the release was very good in showing that we are on a very good track in terms of results. So, that's our first consideration and now we are open to questions.

Thank you. Ladies and gentlemen, we will now begin the question-and-answer session. (Operator Instructions). Our first question comes from Mr. Saul Martinez from JPMorgan.

Hi. Good morning, everybody. I guess my question is on asset quality. And just generally speaking, I want to get your sense for how confident you are that fourth quarter will be the peak of the cycle. I think if you wanted to be a skeptic, you can see -- you could show that NPLs did increase in the quarter, you've seen a bigger increase, obviously, on the retail side, and new NPL formation did worsen in the quarter as well. New NPLs before charge-offs did grow.

So, what other -- other than a pretty small decline in the early stage delinquencies gives you and your models comfort that the NPL cycle and asset quality is very -- the asset quality cycle is very close to reaching its end?

Saul, it's Rogerio. You're right, we are pretty confident on the trends of the provision and the expenses and the NPL. This was actually more than proved along the year. We increased the provision based on the expected loss model by the first quarter. And since then, we didn't change anything. The criteria is absolutely the same one, very consistent and not showing any need for additional provisions.

On the contrary, by keeping the same level is because we are actually economically using a part of the provision provided by them. Some other sources, that's -- it's confidential on our perspective.

You mentioned the early delinquency based on the 31 to 90 days with a very slight improvement, it's true. If we take the system, normally the system uses the 15 to 90 days, this is even better. It's actually showing an improvement of 20 basis points for the consolidated portfolio, 40 basis points down for companies, [reflects] actually 5 basis points now on individuals, but it's been improving.

We don't see any other change that could give us any sort of movement on the other direction. We are pretty confident that this is what is going to happen. Also remember that we have been changing the mix of our portfolio towards a lower risk portfolio. So we should see an improved -- further improvements in next quarter's on this.

And then, just on that latter point, Rogerio, your post-provision net interest margin, as you show on one of your exhibits, has declined. Obviously, as the NIMs have come down your loan loss provisions have gone up. Can you give us a sense of what the outlook is for your NIM with clients after provision for credit risk next year?

Do you think that that -- do you think loan loss provisioning levels could start to moderate as a percentage of loans? And hence, even if NIMs are flat to down, you could actually see an increase in your NIM with clients after the provision for credit risk?

That's what we see. So we are not expecting any major movements on this, but we do believe that net interest margin, risk adjusted after bad debt provisions, should be resilient, flat or even improving as time goes on since we have this benefit due to the bad debt expenses.

On the other hand, it's pretty difficult to be assertive when talking on the net interest margin after bad debt -- before bad debt, sorry, because we have at the same time movements in the spreads. You know, the interest rate is moving down in the country with some perspectives of keeping it the same -- at the same direction. So the future yield curve tends to create the spreads -- lower the spreads during the next [futures]. And while we also have this impact from the mix of the portfolio.

So should be, may be resilient. But definitely, the best way to look at the future is by looking at the risk adjusted net interest margin that we are much more confident on the resilience of the figures. Thank you.

Okay, great. Thanks a lot.

Excuse me. Our next question comes from Mr. Jason Mollin from Goldman Sachs.

Hello, everyone. Thank you for giving me the opportunity to ask my question. Maybe you could just talk a little bit about Itau's expectations for interest rates, particularly given what you just said, where you think the Selic will be heading next year, to what level if you're saying they're going down, how far? And then, maybe you can talk a little bit about the path of the net interest margin, the relationship or sensitivity of your margin to that path.

And my second question would be just to understand a bit of your outlook for cost synergies, particularly the efficiency ratio you talked about on your metric of still sticking with the 41%, I believe it's by the end of 2013. But you did talk about lower than previously anticipated revenues. If it makes sense -- if you could give us some of your views if it makes sense to stick with that. It's only been a couple of quarters, but to stick with this efficiency ratio expectation if, in fact, we're starting from a weaker position. Thank you.

Okay. Jason, this is Sergio. Interest rates next year, our expectation is that the Selic rate in Brazil should reach around 9.5% by sometime mid-year next year and remain like that -- at least with a more plausible scenario that we see right now.

As for the linkage with the margins, I'll pass on to Rogerio.

Jason, of course, we have several different consequences of any movement in Selic. But we have a direct consequence, an obvious consequence, on the cash position we have invested in federal bonds bearing Selic -- or in directly bearing Selic or directly bearing Selic.

This is a [notional] corresponding to around BRL60 billion in our balance sheet. We show that figure on page 13 of our MD&A. So considering any reduction of 100 basis points, we should observe a contraction of the margin by BRL600 million in a year. This is pre-tax. After taxes, BRL400 million, BRL450 million. What means around BRL30 million to BRL35 million per month, around BRL100 million per quarter. Considering that the only consequence of Selic would be on this position.

Of course, when we look at Selic, we have several others, like I mentioned before on the other question. We had the spread with some level of sensitiveness because of the future yield curve and we also have positions built up in terms of treasury. And the fourth consequence, and by far the most important one, is that with the lowering level of Selic we tend to observe a removal of pressure on the [income] from the individuals, what could benefit the delinquency, removing from pressure and, of course, helping the risk adjusted net interest margin to recover and to increase. Those are the most important consequences of Selic. Of course, also in expenses we have impact out of this.

And in terms expenses, you're sticking with your 41% [that you'll receive]?

Sure, sure. In terms of -- as Alfredo mentioned at the beginning, we are actually improving very much our expenses line. We decreased our expectation of growing expenses from 10% to 13% and just decrease it to 8% to 10%, will be inside this range in low single-digit growth this year what is pretty positive.

Revenue generation was a little bit lower than expected. What could cause some difficulty on delivering the first stage of our efficiency ratio, however, we keep pretty confident and we reiterate our commitment on the 41% by the end of 2013. We have no reason to believe this is not going to be delivered.

Very helpful. Thank you.

Excuse me. Our next question comes from Mr. Marcelo Telles from Credit Suisse.

Hello, everyone. I have two questions. One follow-up on asset quality. When you look in terms of your short-term and NPL ratio indicator, there was in the quarter let's say at a 10 basis points improvement. But then, you expect, I believe, a 20 basis points deterioration in the fourth quarter.

I would expect that given the short-term delinquency improving that you would actually expect maybe some stability or maybe some improvements in the fourth quarter. Am I reading that right, or what would explain this difference between what your short-term delinquency indicator is showing versus your expectation for the fourth quarter?

And my second question is regarding the operating expenses for 2012. Do you think that you're going to be able to grow OpEx below revenue growth next year? Can we expect another good year in terms of operating expenses as you are seeing in 2011? Is there anything that would make it more difficult to achieve, for instance, the increase in the salaries that we had at the end of September. They were relatively high for the sector.

And just one -- third question as well. This is a very specific question. I was wondering if you could share with us what was the provisioning coverage for the 90-day overdue credit in the renegotiated portfolio? I know for the third quarter you mentioned you were -- the provisioning coverage was 141%, but I didn't see, at least in your release, the same number for the second quarter. Thank you.

Marcelo, starting from the end, I just received your message asking for -- including this other information in the [G&A]. Thank you very much. We are considering it. We are preparing it to share with you guys.

Going from the third to the second question, the OpEx, like I just said, we do believe we are going to deliver this 41% by 2013. The improvements from the current level to 2013 is similar. Between 2012 and 2013, we should see long-term expenses growing in line with inflation, definitely not more than inflation. And probably, it's going to be a similar level, maybe even a lower level than what we are expecting for 2011.

So, in 2012 we should see around 7% to 8% growth at the currently stage of our budget process is the perspective we have in mind. I just want to highlight that we normally announce our prospectus for the following year by the beginning of the year. It's going to happen in February whenever we mention -- we present the financial statements for the whole year.

So far is a preliminary figure, but we have very good movements on this efficiency process this year in 2011, that will contribute integrally for the year in 2012. That will help in some cases even more than -- 2011 we'll notice this in 2010.

First question on NPL ratio as a whole. What we are observing in terms of our delinquency movements, Marcello and all of you, is exactly what we had anticipated in our expected loss model. That's why we created additional provisions, mainly address -- complement to the model of expected losses. This happening the first quarter.

In the second quarter, we get at the same level the provisions. If you look inside the movement of the provision, what we have in the second quarter was an increasing of the generics provision, not the specific provision, and what we have now is an increase of the specific provision without any increase in the generic provision.

At the same time we have NPLs, any NPL reducing -- reducing the pace of the NPL, this growth, 20 basis points, was exactly what we had anticipated. It would be 30 basis points, not 20 basis points, where this was just exchange rates benefit, that will be back on the fourth quarter, that's exactly why we are anticipating 10, maybe 20 basis points at the fourth quarter.

And more, more, much more important than this, this should be the peak and then NPL should be -- delinquents should be flat or down more. Pretty confident that the models never changed, the consistency of the models are actually have been proved. They are pretty efficient. We are seeing exactly what we have anticipated.

A couple of other indicators show that the 30 to 90 is improving 10 basis points. The 15 to 90, that is the system-wide more used figure, is actually improving 20 basis points on NPLs. So we are at the end of the spread cycle, it should improve now. Of course, our coverage tends to reduce when the period is reached and then increases when the peak is -- after the peak when NPL decreases. This is the normal cycle.

I just reaffirm -- I just reiterate we don't have any longer any cyclical provision. We don't see a cyclical provision as the correct way to account for provisions in line with IFRS and BIS III. So the unexpected losses, the anti-cyclical provisions, are considering in our capital formation not in our income statements.

Rogerio, if you'll allow me just one more question. I don't want to take more time from the other analysts to ask questions, but if you think about the outlook for next year, and I'm thinking about your expected loss model, at least the data -- the economic data we've seen so far in terms of industrial production, GDP growth, employment, they've always -- they have come in terms of more on the negative side at least versus a market expectation. So we're starting to see some deterioration in this economic indicators.

How is that captured in your model? Do you have a different outlook for the economy next year, or you are betting a lot on the fact that you are becoming a lot more conservative on your credit granting policies? So how much do you think it's related to a more bullish view on the economy and how much would be because you guys are betting on this -- getting out of the weaker segments and moving to low risk segments?

Marcelo, this is Sergio Werlang again. Just to comment on the macro -- and here I'm going to comment on the macro overview, which is compatible with the macro scenario that I just mentioned in a previous answer. And then Rogerio is going to comment on the impact in our balance sheet.

Our current macro view is that there is going to be still a further deterioration on delinquency rates for personal credit. It's not a large one, but still there is going to be some deterioration on the average NPL of the system to more or less the middle of next year when we see some stability and then starting to go down as we see the full impact of the cuts in interest rates taking effect into the economy. Now, let me pass on to Rogerio.

Thank you, Sergio. I just want to add a couple of other things that's impacting our portfolio other than, of course, the macro scenario. The mix of our portfolio is moving towards a lower risk. We should present a better performance than the system because of that. Also, the interest rates reduction, as I mentioned some minutes ago, helps a lot on the individual's income.

It's true that unemployment rate could increase, but is going to be in a good level. And considering that the macro scenario and these two movements in our portfolio, we believe this is -- the trend should be the one I just mentioned.

Thank you. Thank you very much.

Excuse me. Our next question comes from Mr. Jorge Kuri from Morgan Stanley.

Hi. Good morning, everyone. I have two questions. The first one is, sorry, again on asset quality. And I guess I'm just trying to put together what Sergio explained and what Rogerio mentioned.

I guess I agree that NPLs move in waves, they just not jump and down on a quarter-on-quarter basis. And if you look at the NPLs this quarter on the consumer side, they accelerated quite significantly up 50 basis points versus 10 basis points in the second quarter, so we're still seeing an acceleration.

Given that these things move in waves, I think it's fair to assume that they will continue to deteriorate for several quarters more, which, if I understood correctly, that's what Sergio said. Now, I'm trying to reconcile that with your view, I guess the top-down view, that we're only going to see NPLs up in the fourth quarter and that's it.

So, two questions there. First, can you just give us a little bit of sense on exactly what's happening on the consumer side, what products or segments or particular clients are suffering the most? And then just what the expectation is if NPLs are going to go -- are going to peak in the fourth quarter or next year, just to clarify that.

And my second question is on margin. If Selic rates are at 9.5% by mid-2012, as your macroeconomic department just explained, average Selic rate for 2012 is going to be around 10% versus 11.5% on average in 2011. So that's 150 basis points lower Selic. Can you be a bit more specific at what the impact is going to be on your margin?

I appreciate the risk adjusted explanation, let's just leave that aside for a second. Just on the margin per se, the 150% decline in Selic, what does that do to your net interest margins? They've been at 11.7% this year. What do you think they're going to be next year considering the 150% -- 150 basis points lower Selic and your more conservative lending mix? Thank you.

Jorge, Rogerio. Well, beginning from the impact in margins, on page 13 of our [MD&A], there is a statement in which we show BRL60 billion -- around BRL60 billion to BRL61 billion of our net assets that are sensitive to any variation in the Selic.

So it's just 150 basis points on average times BRL60 billion would represent around BRL800 million net effects BRL500 million, BRL600 million in the whole year, you should consider. And that's what I just mentioned, that this is one of the movements that Selic causes. There are several others among them. The most important is a potential improvement in the risk adjusted margin.

Talking on the 50 basis points this quarter. What we are observing -- and that's really very important. What we are observing is actually what we have anticipated in our expected losses -- expected loss model. And the movements quarter-per-quarter, they can be different, but what's important to say is that it was -- it's true, it was 50 basis points in the third quarter. But this is the -- 50 basis points was the total for the year up till now and could be another 10 to 20 basis points. As we mentioned, this should be the end of the cycle.

So it puts under the perspective of the whole year, you're going to see that our behavior in terms of NPLs is actually better than the system and similar to the other big players in the system, although we have risk portfolio on individuals because the level of Consignado, for instance, is much lower in our portfolio.

The 50 basis points increase in the third quarter, we don't have any specific line of products that shows initial, it happened across the border. Of course, it's set by the Consignado and the mortgage, but those two portfolios are very low and they actually serve as a sort of a subvention to any reduction of the orders.

But as they are very slow, they don't help the net interest margin as a whole in the individuals portfolio. So the 50 basis points in personal portfolio was the same 50 in credit cards, so it's pretty much the same for each one of the lines. And what is important to say is that it was 50 basis points in the quarter, but it was 50 basis points in the year up till now.

All right. And just to clarify, so you expect NPLs from consumers to peak in the fourth quarter or in the middle of next year (inaudible - multiple speakers)?

Fourth quarter.

Fourth quarter.

Fourth quarter. Sorry, I missed that.

Okay, fourth quarter. That's the peak of NPLs for consumers, right?

Absolutely.

If I do the math of the BRL500 million on your margins, that -- I'm trying to do the math really quickly here just -- and this is why I wanted your help. So, from the 150% decline in Selic -- 150 basis points, sorry, decline in Selic for next year, that translates, if I'm doing the math right, to around 50 basis points reduction in your net interest margin. So does that sound more or less correct?

I should make the calculations. It's around -- I think net of tax would represent around BRL600 million -- BRL500 million to BRL600 million. I did the math when I talked to you. We can check and get back to you.

All right, thank you.

Excuse me. Our next question comes from Mr. Daniel Abut from Citi.

Good morning. Rogerio, just to follow-up on the whole discussion about your cost income ratio program -- improvement program that you're reiterating the commitment to reach 41% by the end of 2013. It seems to me that earlier in the announcement of this program you were planning to rely more on revenues than on cost. And as you started to see revenue being softer, you moved the end process toward more reliance on cost.

You will recall that you started the year guiding cost growth in the 10% level then you reduced it to 8% to 10%, which you are comfortable you will reach this year. And then for next year, softly, you are starting to guide more in line with inflation. I think you said, Rogerio, 7% to 8%. So it's good to see that you're relying now more on cost because it's likely that revenue will remain soft next year.

So I wanted to talk a bit about revenues into next year. I know that you're not giving guidance yet, but at least softly you could comment what your expectation will be for loan growth next year, assuming, as most economies are assuming now, that the economy grows somewhere in the 3 point something percent real GDP growth next year?

And given what you're seeing in fee income, that you already reduce your expectation for this year to 13%, Alfredo said, what could be a good expectation to have for fee income growth next year? I want to get a sense of how soft, again, revenues could be next year, that will make even more important to be more stringent on costs as it seems to me that you are.

Daniel, you are right. We are entering a new scenario where revenues will be more difficult due to environment and the mix of products that we are aiming to achieve. So we have to work hard on both sides -- try to compensate revenues growing more volume, if possible, and controlling expenses by the level of inflation, to really achieve this efficiency ratio and to keep good levels of ROE for the shareholders.

So we have to work both ways, of course. We expect that credit growth for next year to be around 15%. So considering inflation and the growth of the economy, so 1.5%, around that in terms of nominal GDP, we think is feasible to achieve. Of course, we have to follow-up the macro scenario to see the real trend in the coming quarters. But up to now, we think it's feasible to achieve this level of growth in terms of credit portfolio.

And on fee income, Alfredo?

May be a little bit lower, but similar. For this year, we expect 13%, maybe around 13% to 14%, but I think it will be a little bit lower than the growth of credit portfolio.

Thank you, Alfredo. That's useful.

Excuse me. Our next question comes from Mr. Mario Pierry from Deutsche Bank.

Hi. Good morning. Let me ask two questions as well, the first one on the asset quality. If you can just give us some color on how your SME portfolio behaved this quarter. I think there was a bulk of your pick-up in delinquencies. Last quarter was related to the SME portfolio this quarter. I have only seen the data for the overall corporate portfolio. If you could break down for me the deterioration in the large corporate and the SME segments?

Also, on the asset quality, I think your coverage ratio of 156% is the lowest I have seen probably in the last 10 years. Just trying to get a feel from you, at what level do you think that this ratio could go before you start to worry about it?

And then my final question would be on costs. We have seen your headcount come down from 110,000 in the first quarter to 106,000. Should we expect more headcount reductions, or is this is the level you're comfortable with? Because I do expect you to be opening branches next year as well. Thank you.

Okay, Mario. A couple of questions here, so let's just start from the movement in SME. The corporate portfolio growth was 9% including, of course, the exchange rate benefit. Net of this would be 3.5%, 3.7%. The very small, the small and middle -- mid-sized companies growth posted was 2.2%. If we break the micro -- small companies, actually we had a reduction in the total portfolio. And if we take the mid-sized companies, there was a growth. The figures are similar.

The growth for mid-sized companies was 4.1% and the decrease in the balance for micro and small companies was 2.2%. So if you look at the SME, taking as a whole, the growth was 2.2%. But what you have was actually -- in a de-acceleration there's more companies. As a result of that decision, we took since November of -- asking for more collaterals from the clients. By doing so, there was a de-acceleration in this portfolio.

Just highlighting that we are talking on is small, very small companies. The mid-sized companies, we keep, very, very strongly growing. It's pretty positive results and returns on this -- risk adjusted returns are very positive. Second question you --? Excuse me?

Yes, sorry. I meant the delinquency ratio on your SME portfolio rather than the overall growth. So, what happened to the delinquency ratio on the SME segments this quarter?

I don't have this figure here. I was trying to get it. I think was -- we have a deterioration of around 20 basis points during the quarter. And the corporates as a whole was actually benefit by the large corporates managed by Itau BBA. But I can't confirm to you. There was a bit of deterioration, but a small one.

Talking about the coverage, I think this is particularly the most important point because I think there was some misunderstandings on this. The coverage ratio is always -- declines when NPL is growing. And always, when NPL starts to decrease, we have an increase of the coverage over 90 day. This is the cyclicality of the process.

We decided as from December that we don't have any anti-cyclical provision in our income statements. We treat any anti-cyclical risk and expected risk. This is different from the scenario we are looking forward we treat as capital. This is in line with IFRS and the BIS III. Those are the best practices we can find.

So, through the glasses of expected loss perspective, this is what drives our provision. Of course, it will look at the future and anticipate NPLs growing in the future. We increase provisions, the coverage increases whenever the NPL is coming, confirming what we had anticipated, the coverage declines. That's absolutely expected.

We were not expecting a different behavior. A different behavior would be -- a wrong behavior if you look from the cyclicality of this. I just draw your attention to the -- if you want to look at the absolute coverage, it's very important to mention that if you look at our coverage on a specific provision, that is the provision based on [impairs], this reduction was actually 2 percentile points.

If you look at our coverage by removing the class H of our provision, because class H is 100% provision, so there's no sense on increasing additional provisions on 100% provisional credits. So if you remove that, the coverage during the whole year, was actually flat.

I think you had a third question on headcount. What we are seeing is a consequence of integration of some areas, particularly the consumer credit areas, and we should see -- we may see some additional reductions over time, but not strongly. Most of them were already observed in second and third quarters.

Okay, thank you.

Excuse me. Our next question comes from Mr. Jonathan Prigoff from Equinox.

Hi. I just wanted to see if you could touch more on the different types of individual loans and which ones are growing faster. Specifically, we've seen personal loan growth pick up a lot. I was wondering if you could touch on that. And I know vehicle loan has been slower due to what's going on in the market and maybe due to some regulatory concerns, but maybe you could help us understand your expectations for those other buckets.

The personal credit growth was mainly accelerated during the last three quarters -- two or three quarters because they were a little bit behind during the process of integration of branches from Unibanco to Itau. Now, with the single platform, we can explore more adequately the level of risk, so we are catching up. That's the reason of the personal loans increasing in recent quarters. Actually, we had lost some of our normal market share on this end, we are catching up. That's the reason.

Looking at the auto loans, you are right when saying that auto loans is not growing. But the reason is not -- it's not actually very related to this macroprudential measures. It's actually related to the maturity of this portfolio. We have a very strong risk appetite on this business, is a very contributive business, and we keep originating around BRL2.3 million every single month on this portfolio. That what happens is that with the maturity of the portfolio there's no growth because we have the same level of contracts expiring.

If I look historically, as I recall, vehicle loans were a big engine of growth and kind of helped fuel the growth that we've seen over the past few years here. Given that, as you've said, maybe that market has kind of matured for you. And given the pull-back you've seen at the smaller end of the SME business, what's the engine of growth going to be going forward? What are the projects that you think are going to help you get a multiplier effect on nominal GDP growth?

Alfredo just mentioned that we believe the normal pace onwards now is no longer three to four times GDP. We are talking 1.5 times GDP now. Corporate loans should be one of the important drivers, mortgage another one. Credit card business is still underpenetrated in Brazil, should account for it. So I'm referring to the lines of portfolio Consignado, the payroll type of credits. I'm referring to those that should grow a little bit ahead of this average 15%.

On the other hand, some others -- mid-sized companies, again. On the other hand, auto loans should be behind. Even under a positive scenario, it's difficult to figure out any growth bigger than 5% to 8% in the auto loans. So some could be behind this average, some others ahead, and the total 15% is what we expect.

Thank you. One more question, if I may. Just to take a step back about individual delinquencies, what do you think is actually causing the delinquencies? A few months ago, people were talking a lot about the debt service requirements being high, and I think you and a lot of the other banks made a good case as to why, at least by historic standards, debt service requirements are not that high for Brazilian individuals.

But yet we're seeing this up-tick in delinquencies without a huge increase in unemployment. So what's the -- what do you think? Is it just due to interest rates, and once rates come down that should correct itself? Or what's the --? Are more people having affordability issues with their levels of debt?

It's related to inflation and interest rates. The increasing of this two components in the individual weaken, that's cause of this. Of course, we noticed this and we provided for it by the beginning of the year based on the expected loss. What we see now is exactly what we had anticipated. Nothing new, nothing different from that.

Based on the methods, you anticipate provisions before NPLs. That's the reason we don't see anything and just to reiterate [Bubble] we don't share any view on Bubble. We believe Bubble is impossible to be created in a scenario of so high interest rates and so short duration credit, to talk in summary.

Let's go to the last question because the time is expiring, please.

Thank you.

Excuse me. Our next question comes from Mr. Victor Galliano from HSBC.

Hi, thank you. A couple of questions here. Just on costs, Rogerio, if you can focus on -- you're talking about in line with inflation OpEx growth, but can you just remind me what you're also planning in terms of further expansion? Is there anymore -- in that in line with inflation cost growth, has that built in some increase in the branch network, et cetera?

And my second question is on margins and really focusing in on this revenue line. So, the kind of core scenario -- of the soft guidance, shall we say, for 2012 is 15% loan growth. If you go to 15% loan growth, what can we expect in terms of net interest income line? I presume we're looking at best case scenario in this environment of lower risk credit, 12%? Something along those lines? In line with kind of fee and commission guidance that you gave us, is that right?

Yes. Actually, when we look at the top-line, we see credit growing 15%, we see a resilient risk adjusted margin eventually improving a little bit. So let's consider around 15% growth in top-line. Fee income should be a similar level, should not reduce too much. It's important in relation to the total top-line.

But I just want to mention that we are going to present our perspective for 2012 within three months in our conference call on the results for the entire year that is scheduled for February. Those are all preliminary views on this.

Back to the other question on costs. The 7%, or the inflation alignments in our cost line, it's actually -- what we have behind is the improvements in cost, the efficiency in cost could stunt our growth. We expected to open branch -- on average 150 branch for next year. Actually, next year we have some -- a little bit bigger figure for next year. It's scheduled to be 170 branch, but on average we open around 150 branch. And also, we have scheduled a remodelation of branch totaling 600 branch for next year.

All those costs should be funded by this further improvements in cost in order to have the full line growing in line with inflation, let's consider now around 7%, and we'll be back to this and following up on this within three months.

Okay. So just to recap, on the branches, you're expecting to increase about 150 this year, 170 next year?

No, it's actually --. We have already opened 50 branch this year.

Okay.

And we have scheduled another 90 branch toward the end of the year. So it's going to be around 130, 140 branch this year. The normal level anticipated for this year was between 120 and 150, we are going to be inside this range. Next year, it's going to be a little bit higher than the normal 150, around 170.

Remodelation, we had this refurbish -- remodeling of the branch this year with a total a little bit below 500 and we're scheduled to have 600 next year.

Okay. Okay, great. Thank you.

Thank you, Victor.

Excuse me. Ladies and gentlemen, this concludes today's question-and-answer session. Mr. Setubal, at this time you may proceed with your closing statements, sir.

Thank you. For all -- we still have some questions that we could not attempt, we are going to call these -- for those who are in line. Sorry, but the conference was a little bit too long, so we have to stop here. But thank you for your time.

I think we had a very good quarter. I think we are on the track of good quarters ahead. So we continue very confident about our models, about our business. We are very confident in [that]. So we will see you again in the conference call next year about the year-end results. Thank you.

Thank you. That does conclude our Itau Unibanco Holding earnings conference call for today. Thank you very much for your participation and have a good day.