Mario Arruda Sampaio -

Michael E. Gaugler - Brean Capital LLC, Research Division Hasan Doza Sergio Tamashiro - Banco Safra S.A., Research Division

Good afternoon, ladies and gentlemen. At this time, we would like to welcome everyone to the SABESP conference call to discuss its results for the second quarter of 2014. The audio for this conference is being broadcast simultaneously through the Internet in the website, www.sabesp.com.br. In that same address, you can also find the slide show presentation available for download. [Operator Instructions] Before proceeding, let me mention that forward-looking statements are being made under the Safe Harbor of the Securities Litigation Reform Act of 1996. Forward-looking statements are based on the beliefs and assumptions of SABESP's management and on information currently available to the company. Forward-looking statements are not guarantees of performance. They involve risks, uncertainties and assumptions because they relate to future events and, therefore, depend on circumstances that may or may not occur in the future. Investors should understand that general economic conditions, industry conditions and other operating factors could also affect the future results of SABESP and could cause results to differ materially from those expressed in such forward-looking statements. Today with us, we have Mr. Rui Affonso, Chief Financial Officer and Investor Relations Officer; Mr. Mario Arruda Sampaio, Head of Capital Market Investor Relations; and Mr. Marcelo Miyagui, Head of Accounting. Now I'll turn the conference over to Mr. Arruda Sampaio. Sir, you may begin your conference.

Thank you. Good afternoon, everybody. Thanks for attending one more earnings conference call. We will -- this call will have 11 slides, where we would like to discuss the main events in the period. After that, as usual, we'll open for a question-and-answer session. Let's start on Slide 3. Here, we show the company's billed water and sewage volume, which fell 1.8% in the period due to the incentive program for water consumption reduction based on the bonus system in place since February for clients served by the Cantareira system and since April for all the clients in the São Paulo Metropolitan Region served directly by SABESP. In addition, at the end of May, the program was extended to the cities directly operated by SABESP, which are situated within the Piracicaba, Capivari and Jaguari. We call that PCJ river basin, all of which within the area of influence of the Cantareira system. Considering the available data from recent years, the 2.8% and 0.6% reduction in water and sewage billed volume is a nonrecurring, extraordinary event, which normally would not occur if we were not going through the worst hydrological crisis in the history. The bonus program to incentivize water reduction, consumption reduction is the most important measure adopted to reduce demand and, consequently, reduce consumption in the São Paulo Metro Region. Contributing to the efforts to adjust demand to the region's water supply, particularly in the Cantareira system, where currently the permitted withdraw from the system's 3 reservoirs, which are part or within the PCJ water basin, is limited to 19.7 cubic meters per second. Moving on to Slide 4. We will comment on our financial results. Net operating revenue declined 1.5% over the same period last year, affected by the lower billed volume and the BRL 88.1 million impact from bonus paid under the incentive program for water consumption reduction. The 3.1% tariff increase applied since December 2013 partially offset this downturn. Costs in selling, administrative and construction expenses increased by 11.2% in the period. Adjusted EBITDA fell 27.4% from BRL 911 million in second quarter '13 to BRL 661 million in second quarter '14, while the EBITDA margin came to 24% in this quarter versus 32.6% in the same period last year. If we exclude the effects of revenues and construction costs, adjusted EBITDA margin resulted in 31.2% in second quarter this year versus 42% in second quarter last year. On the year-to-date, the adjusted EBITDA margin was 30.2% against 33.7% in the first half of 2013. Again, excluding the effects of revenue and construction costs, adjusted EBITDA margin was 38.1% against 42.2% in the first half of 2013. Net income came to BRL 302 million, a 16.4% downturn on the same period in 2013. On Slide 5, we very briefly discuss the main variation in cost in relation to the same period in the previous year, costs and expenses moved up by 11.2% over 2013. Excluding construction costs, cost and expenses climbed 14.8% due to the increases of 227% in credit write-offs, 49% in tax expenses, 19% in services, 17% in treatment supplies, 13.2% in depreciation and amortization and 12.1% in payroll and benefits. Greater details on each of our costs can be found in our earnings press release. On Slide 6, we present the main variations in the items that affected our net income in second quarter against -- or over the second quarter of 2013, which totaled BRL 302 million. Net operating revenue decreased by BRL 42 million or 1.5%. Cost and expenses, including construction cost, increased by BRL 233.5 million or 11.2%. Although operating revenues and expenses recorded a positive variation of BRL 3.8 million, net financial revenues and expenses, including foreign exchange variations, decreased by BRL 185 million in the period. Finally, income and social contribution taxes declined by BRL 26.9 million due to the reduction in taxable income in the period. On the next slide, Slide 7, we will update you on the hydrological situation in the São Paulo Metro Region and the measures that SABESP has been adopting to ensure continued supply to the population directly served by us. It is worth noting that the crisis is mainly concentrated in the Cantareira system. And it's the worst in history, or more specifically in the 84 years of available records. On this slide, we show the reservoir levels for the systems that supply the São Paulo Metro Region. The data reflects reservoir levels as we approach the end of the driest period of the year, that is from June to August, remembering that the rainy season starts in October and runs until March next year. Specifically, regarding the Cantareira system, the reservoir level was at 12.9% in August 18, that is yesterday, which means we are already using water from the technical reserve. As for the reservoir level in the Alto Tietê system, it is at 17.9%. This level reflects mainly the fact that we are using water from the system to serve part of the demand for the Cantareira system, as we will comment further in this presentation. But before -- let's move on to the next slide, Slide 8, which updates on the measures that SABESP has been adopting and implementing to deal with the current authorized extraction of water from the Cantareira reservoirs sitting within the PCJ water basin, remembering that this figure has been reduced from 31 cubic meters per second in March this year to current 19.7 cubic meters. We have adopted 3 main initiatives that has allowed us to reduce water withdrawn from the reservoirs in the Cantareira system by around 10 cubic meters, maintaining, at the same time, water availability for the population directly served by us in the metro region of São Paulo. In summary, these measures are: first, water transfers between São Paulo Metro Region production systems currently accounting for 46% of this reduction of the 10 cubic meters; decreases in consumption driven by the bonus program, responsible for 22% of the savings; and operational maneuvers, and mostly investments to reduce water losses, accounting for the balance of 32% reduction. More specifically, regarding the transfer of water from other systems to the areas usually covered by the Cantareira system, thanks to past and current investments made in interconnection of all of the São Paulo Metro Region's water supply systems, it has been possible to transfer water such that more than 2.3 million people that use to be supplied by the Cantareira systems are now being supplied by other water systems. More specifically, of the 8.8 million clients normally served by the Cantareira system today, 6.5 million are being served by the system. On the other hand, 1 million are being served by the Alto Tietê system whose population coverage increased now from 3.5 million to 4.5 million. And an additional 1 million being served by the Guarapiranga system, again, in this case, whose population coverage grew from 3.9 million to 4.9 million. The remaining 300,000 covered by other systems are mostly being covered by the Rio Claro and the Rio Grande systems. At the end of this year, additional measures to expand water production interconnection and transport capacity in other systems in the São Paulo Metro Region should allow more than 500,000 clients currently served by the Cantareira system to be served by these other systems. On the next 3 slides, we will analyze rain, water inflow in the Cantareira system reservoirs and temperatures in the metro regions of São Paulo. So let's move to Slide 9. Here you can see that the rain levels were below the average for the month of May, June, July, with May and June falling considerably below the average. In August, as you can see, we're still below average. However, we still have 11 days in front of us. As a result of lower rainfall in the period, water inflow to the Cantareira system reservoirs, as you can see on the next slide, on Slide 10, remained below the monthly minimum records, when expectation in fact was for it to converge to the green minimum line. Our next slide, on Slide 11, we show average temperatures in the São Paulo Metro Region. We can see that from June to July, temperatures were principally higher or slightly higher, but not so slightly, a little bit more than slightly, than in the same period in 2013. In fact, between June 15 and July 15, some days showed temperatures 4 to 5 degrees higher than average. This, along with the increased number of tourists in the World Cup period, certainly played a role in the consumption increase observed in the period consumption of water, as well as in lower adherence to the bonus plan program, which we will discuss later on. These events led to an increased daily reduction in reservoir levels. As a result, in a preventive manner, SABESP has requested and obtained authorization from the state authority, DAEE, to use an additional 106 billion liters of waters from the Cantareira system technical reserve, which we will go through into more details as we move into Slide 12. As you can see, in the Slide 12, we show a graphic representation of the available water volume in the Cantareira system. As you can see, the system total reservoir volume capacity is 1,493,000,000,000 liters of water, 982 billions of which is extracted by gravity and 510 billion liters that we can extract from the technical reserve through pumping. Thanks to the authorization to extract these another 160 million liters of technical reserve, total technical reserve volume available for use should be approximate 285 billion liters. As a result of this increase in availability, combined with the other successful measures adopted by SABESP, we're even more confident that the maintenance of water availability in the São Paulo Metro Region until the rainy season begins. Let's now move to Slide 13. Here we'll discuss in more details the bonus program, which is the measure that most impacts the company's revenue, considering that it encourages our clients to reduce consumption. An assessment of this program in the São Paulo Metro Region in July shows that 86% of the connections saved more than 20%, therefore, reached the 30% bonus, 28% reduced consumption, but did not reach the bonus, and 26% of the connections, in fact, increased consumptions. These results, from the point of view of adherence to the program and consequently a drop in demand, were a little lower than those observed in May and June, mainly due to the unseasonably hot period between June and July 15, which coincided again with the consumption -- increase in consumption caused by the World Cup tourist inflow, as we have already mentioned in previous slides. However, after the end of the World Cup and with the lower temperatures, we've seen in August, the first partial data for this month, in fact, until August 11, already shows increased adherence to the bonus program and water savings, with 52% of the connection saving more than 20% before reaching the bonus, 26% of the connection reducing consumption, but not reaching the bonus, and 22% of the connections increasing consumptions. Well, that concludes our initial remarks. And now let's open for our question-and-answer session.

[Operator Instructions] Our first question comes from Michael Gaugler of Brean Capital. Michael E. Gaugler - Brean Capital LLC, Research Division: I guess my only question really today is going to surround the water supply issue, of course. And I'm wondering -- by the way, the slides you guys did were great in terms of giving us some really good data points in terms of where you are and where things are sort of headed. When I look back at your historical averages, if you have a historically normal rainy season and we look out towards spring of next year, where would you anticipate, based on your models, supply would be, particularly in Cantareira, in terms of total capacity above the technical reserve?

Michael, just a second, please. Michael, I'll try to address the point, but I'll basically start saying that your question is pretty much close to the question we got in the Portuguese call where everybody really wants to know what would be a trigger -- what is the trigger in terms of the reservoir levels and when -- which would lead us to take the bonus program out and so forth. And to answer that one, and to answer your question, is that, today, we don't know. I mean, you can do as many projections as you might think of using the amount of variables we have to deal with. Imagine, we have to deal with temperature, we have to deal with sunlight, we have to deal with humidity, we have to deal with water inflow to the reservoir, considering also that the amount of water in the underground level is saturated or not, not to say we are obviously surrounded by environmental and water authorities. And today, it's giving the scenario that is very hard for us to imagine where our reservoir levels would be spring next year. Obviously, we would love to see it at 100%, but that's not something that we can even say it is doable. So unfortunately, I don't think we do have a precise answer to your question other than just raise the points of what are the variables that we and everybody had to keep track to start understanding where we are and what needs to happen for us to reach where we would like to be. But what we know and could be a good news, obviously, is that we're not running into the dry season, we're running into the wet season. And obviously, if the future -- the curves of water inflow rain they tend to go back to past historical averages. So the new historical average is not the same because of the current curve being built throughout this year. If it goes back, obviously, we are in a very, very positive situation. And again, it comes to the question, can it come back in September, October, November, December? It can go into next year. We don't know. So that is another reason why it's really hard to make any projection such that we think that by x month of spring next year, we will have x amount, potentially x amount of water. What we are doing, I think Rui told me to highlight this, is that we are in a continual effort. And by the way, just so you know, we have 2 colleagues from the metro region executive group that are directly involved here in the water management. And what we do know is that we're building assets to further interconnect the system, and we're finding areas still in the current other water reservoirs that we can add more capacity up to a year from now in such that we can even grow more into the amount of water we can transport from other systems into the Cantareira system. I just mentioned 500,000 up to the end of this year, but there's more to come next year. So that sort of helps us in this equation that has to be put into the formula. So that was a long comment, not to answer precisely your question because it was not doable. Michael E. Gaugler - Brean Capital LLC, Research Division: Okay, it wasn't an easy question. Maybe just kind of as a follow-up. Certainly, we have now fall weather patterns developing in the Pacific that are watched closely kind of on a global basis, including El Niño. And I'm wondering, just in the past, and not something you need to look up, but are there any particular patterns that have formed in the past that have been favorable or unfavorable that we could look to in terms of trying to gauge what the rainy season might look like this year?

Okay. I'm going to put on mute because I'm going to discuss with the guys here in Portuguese. Just a second. Michael, the guys here are basically saying that if we were to look at the historical data, I mean, the trend would always be for us to be running on top or above the minimum curve, what is happening is that we're going exactly below. So any event that has more or less in a recurring basis, which I'm not sure is a good sentence here, but more or less has influenced the pattern can't be addressed as something reasonable for now. On the other hand, specifically on the El Niño, the comment is that the El Niño brings a lot of dry results in the northeast of Brazil, that is easily seen in patterns. And it brings a lot of rain in the southern part of Brazil. And the pattern for the southeast, where we are, okay, it's exactly uncertainty. You can't get a pattern for the El Niño exactly for the southeast. And at São Paulo, it's sort of unnerving. If you look at the map and where the rain stops, it stops just about 300 kilometers below São Paulo. That's basically when it stops raining heavy. So we don't know. It could be that this El Niño could bring us a very wet period or maybe not even change the regular average pattern. So I think the point is that this has been such an off-record event that the record has lost a little bit today of its force for projection capacity. So it's unfortunately not what we would like to tell you, but that's what it is.

And our next question comes from Mr. Hasan Doza of Water Asset.

I appreciate all the slides today. And can you go over Slide #10? The one that talked about your water inflow into Cantareira?

Yes.

Would you explain a little bit as to what caused the -- what is still causing the absorption or the inflow rates into the reservoirs to be so significantly below the historical averages? I would appreciate like what factors is causing this absorption or inflow rate to be so low versus your average levels. That's kind of my first question.

Okay. So let me address your first question. We had a similar question on the Portuguese call. If we -- first, the big reason is if you look at rain patterns as of March, they started to go below, so that doesn't obviously contribute. But what really affected is what happened before, what happened in terms of rain during December, January and February. So the way we look at it is since there was very little rain, a lot of high temperatures, the land became very dry and the water sitting, underground water levels, went very low. So even when rain started to pick up at first, started to pick up in the month of March and April, for that water to saturate the underground water, for it to flow into the Cantareira, it did not happen. So, and then as we moved on into May and June, water rain was very low. So there was really not much water to run into reservoirs. So I'm trying to translate the Portuguese version of the explanation. Hopefully, you understood, but basically is January, February, very, very dry period, drove the underground water levels very low, and the rain from there to now has been insufficient to saturate and bring the inflow to the Cantareira to the minimum levels. So even if you look at the graph here, and you see that as of March, there was a very good improvement in the inflow, okay, which -- not sustain itself exactly because the month after that, the rain levels were very much reduced. So again, as we move forward, what we need to see is not just one big rain, but a series of regular rain that will saturate the underground level. And I'm sorry if the glossary in water, underground water is not so good. I would ask Water Asset help on this glossary. But the point is, as we move forward, we would like to see not only big rains, but good, strong, regular rains for a period. That is what saturates land and lets the water -- the rain coming afterwards, inflow into the reservoirs.

That was going to be my follow-up because if the land continues to remain dry, let's just say, into your upcoming rainy season, even if it's an average rainy season, for example, if the water is being absorbed by the dryland, as you have had in the last 3, 4 months, I mean, the average rainy season, it's going to do you a lot of good because if this factor remains an issue in terms of dryland absorbing a lot of the rain water. So how do you guys think about that factor, this dryland being a factor because it looks like the average rainfall might not be enough because if the inflow rate is being slowed down by the water being absorbed by all the dryland?

Hasan, look, let me explain. Remember that in normal period, and the curve shows this very well, in normal period, this is exactly the period where the inflow reduces a lot, not only because there is less rain, but because the water is absorbed by the ground. So if you look at the graph, it is very clear that this curve right now since March is running fairly close in terms of a pattern, in terms of pattern to both the blue curves and all the other curves. The point is, and again, I'm going back to the period between December and March, beginning of March, is when things really went down because of lack of rain. What you're trying to grasp here from us is giving that if the land is dry, how much time and how much rain is needed to recover the water, underground water saturation in such for the water inflow. What we know that if we only look at the patterns from the last month, it's normal to be low, but we don't -- and again, it's something that, I don't know, 10 days of regular rain, 5 days of regular rain makes a huge difference in 1 huge rainy day. So it's like Michael's question. It's hard to answer because these variables, they all play different roles. Huge rain in one day will probably flow a lot and sit less than it'd be absorbed by the ground, which means that the next rain will probably be absorbed more or less. So there's not a math there, and there's no way we can give you a specific highlight. And there's no way we can tell you that the land is drier now, that it was drier 2 years ago, a year ago, in such that the recovery of the inflow due to the dry period will occur, and given that the past event this year was relevant would take 2, 3 more months than it would on regular times. We can't make that statement also.

Okay, all right, that's helpful. And my last question is for you, and also for Rui, is given that what we have just discussed and also the uncertainties about the timing of the rainy season, the location. I don't need like specific numbers or anything. I'm just curious like from you and Rui's perspective, I mean, how are you guys thinking about contingency plans for 2015? Let's just say, argument's sake, you have another dry year or below rainfall year, and you have already -- your 2 reservoirs, Cantareira and Alto Tietê, are being depleted. So as a company from a risk management perspective, from an overall strategy perspective, how are you guys thinking about contingencies for next year if in the unfortunate situation you have 2 years of back-to-back droughts?

Okay. Give me -- give us a second, please. So for now, I'll start and potentially here, Rui will complement if I leave something out. I think the first thing is we are already doing risk management in the sense that we haven't put a plan in place, and we're just managing what we've put in place. The plan means that we have to do continued improvements in our system network. And as I mentioned to Michael, this question, and also a continued increase in the short-term possibility of water availability, just for you to have an idea. By the end of September, we should have added an additional capacity in the ETA Rio Grande of close to 1 cubic meter, 2.8 cubic meters per second. By October this year, we will add an additional 1 cubic meter, and this is extra water. We're reaching a part of the reservoir where we did not, today, extract water from. So greater water availability. So it's a twofold program. We're increasing interconnection for increased transportation, and we're adding capacity. If we go to sometime in 2015, mid-2015, we will do an additional 1 cubic meter. So by that end of 2015, we should have almost 3 cubic meters of water availability production and availability added in the metro region of São Paulo. That is 10% of the current Cantareira production. That is 30% of what we have to -- we're currently obligated to reduce extraction in the Cantareira system. And all this is exactly the risk management that is not reacting to something, but in anticipation of something. So all this is an anticipation of a remote, although as we've seen possible occurrence of a 2-year or an extended dry period. And obviously, there -- all the other measures we are taking in terms of accelerated investments to reduce water losses, remember, this, today, is responsible for about 3 cubic meters per second. So just that, and there's more to do there. So the risk management right now has really put a lot of effort in anticipating everything that was thought of for the Metro region systems for years to come. So the situation has brought us the fact that we need to have it in place right now, and that's what we're doing. Okay?

The next question comes from Sergio Tamashiro of Safra. Sergio Tamashiro - Banco Safra S.A., Research Division: My question is just a follow-up on Hasan's question. It's regarding the Cantareira production reduction. I see from the slides that, that production reduced from 33 cubic meters per sec to hopefully 20 cubic meters per second, reduction of 13 cubic meters per second. So my question is, out of these 13 cubic meters per second, how much was supplied by other systems? And this is -- from these -- what I understood from your answer is that these 20 cubic meters per second could be even reduced to 17 cubic meters per second. And just the overall production could be maintained the same. Is that correct?

Okay. Sergio, let me do that. On Slide 8 -- that we mixed a little bit here. One thing is what we produced out of the ETA Guaraú, okay, at the right line side. That is supplied by water that is within the Alto Tietê water basin and the PCJ water basin, okay? What happened is our extraction from the PCJ water basin was around 30.5%, almost 31%. That is the max we could extract, okay? But although we also extract water from Paiva Castro and Aguas Claras, what we produced at the end reached around 33%, okay -- 33 cubic meters, I'm sorry. But what we are restricted is what we -- what flows through tunnel 5 from the PCJ into the Alto Tietê water basin. That is where we have to restrict to 19.7. So the number, more rough numbers is we were extracting around average on a year basis 30 from the Jaguari, Cachoeira and Atibainha, and now we're down to 19.7. In other words, we have to save "about 10 cubic meters and 11." We have been around 8, growing to 9, down to 8 something, and now we're up to 10, okay? So that is the first challenge we have to make. Now your other question is of that 10 or 11, whatever, we should -- where is the savings coming from? And then it goes back to our speech where I mentioned to you that from -- in terms of water transfer, water transfer is responsible for -- let me check here the exact number, for 46% of that reduction. So approximately water we transfer from one system, from other systems into the area covered by the Cantareira represents 46% of what we have been saving. Today, we're saving 10, so that represents about 4.6 cubic meters per second. Got it? Sergio Tamashiro - Banco Safra S.A., Research Division: Okay. So out of this 10, 46% is the savings. So the diversion is 5.4, the remaining 54%?

Yes, if that's 54%, 32% we are getting of with reduced water losses, and 22% is based on the bonus program where people are saving water. So people are saving water in an amount of approximately 2.2 cubic meters. Water reduction has been responsible for about 3.2 cubic meters, okay? And transfer from other systems into the Cantareira system coverage area is 4.6 cubic meters. Sergio Tamashiro - Banco Safra S.A., Research Division: 4.6? 4.6, okay, got it.

Yes. That's based on around 10, currently 10. Sometimes, this goes down. And obviously, we're working for this 10 to go up. In terms of population, as I mentioned to you in the speech, of the 8.8 million people, we have 2.3 million being served by the other system.

[Operator Instructions] It appears that there are no further questions. I would now like to turn the conference back over to SABESP for any final remarks.

Okay. Again, thank you very much, everybody, and we'll be back next quarter. And in the meantime, myself, Angela are available for any question and information here in the IR department. So thank you very much. Goodbye.

The conference has now concluded. Thank you very much for attending today's presentation. You may now disconnect your lines.