Good day, ladies and gentlemen, and welcome to the First Solar Second Quarter 2011 Earnings Conference Call. Just a reminder, this call is being webcast live on the Investors section of First Solar's website at www.firstsolar.com. [Operator Instructions] Also as a reminder, today's program is being recorded. I would now like to turn the call over to Mr. Larry Polizzotto, Vice President of Investor Relations for First Solar. Mr. Polizzotto, you may begin.

Good afternoon, everyone, and thank you for joining us for First Solar's Second Quarter 2001 (sic) Conference Call. Today after the market closed, the company issued a press release announcing our second quarter financial results and our guidance update for 2011. If you did not receive a copy of the press release, you can obtain one from the Investors section of First Solar's website at firstsolar.com. In addition, we have posted the presentation for this call, as well as well as key quarterly statistics and historical data on financial and operating performance on our IR website. We will be discussing the presentation during this call and webcast. An audio replay of the conference call will also be available approximately 2 hours after the conclusion of the call. The audio replay will remain available until August 9 at 7:30 p.m. Eastern Daylight Time and can be accessed by dialing (888) 203-1112 if you're calling from within the United States, or (719) 457-0820 if you're calling from outside the United States, and entering the replay pass code 6647385. A replay of the webcast will be available on the Investors section of firstsolar.com approximately 2 hours after the conclusion of the call and remain available for 90 calendar days. If you're a subscriber of FactSet or Thomson One, you can obtain a written transcript. With me today are Rob Gillette, Chief Executive Officer; and Mark Widmar, Chief Financial Officer. Rob will present an overview of the company's second quarter results and give you an update of -- on the market and our business. Mark will review our second quarter financial results and update guidance for 2011. We will then open up the call for questions. [Operator Instructions] First Solar has allocated approximately one hour for today's call. I want to remind everyone that…

Great. Thanks, Larry. Welcome to our Q2 earnings call. As all of you know, it was a challenging quarter for the solar PV industry, including First Solar, but our business model and our execution served us well, and we're positioned for a significant and better second half. Starting with Slide 6. Q2 net sales of $533 million, that's about 6% lower than last quarter, primarily due to lower average selling prices. Solar PV' policy uncertainties in Italy, Germany and France adversely impacted demand in the second quarter. Net sales were also impacted as we allocated modules to the Systems business which we did not recognize as revenue within the quarter. Second quarter net income was $61 million or 11.5% of net sales, resulting in diluted EPS of $0.70. Return on net assets was 13.1% on a 4-quarter rolling basis representing an EVA [ph] of about 1.1%. Our cash and marketable securities balance of $515 million declined $198 million sequentially as we invested CapEx for new factories and continued construction in several systems projects. In addition, shipments were weighted to the end of the quarter as customers delayed orders until European policy changes were better understood. Despite the challenging European market in the first half of the year, our 2011 outlook remains solid due to our differentiated and resilient business model. We expect significantly better performance in the remainder of 2011 as we build projects from our systems pipeline, develop promising new markets and execute our cost and efficiency roadmaps. As a result, our updated 2011 EPS guidance is $9 to $9.50 per fully diluted share, which includes 450 megawatts at system sales, and some updated assumptions which Mark will discuss later. Significant drivers in the change is a 2% increase in the effective tax rate or about $0.20 per share…

Thank you, Rob, and good afternoon. Looking at Slide 17. Net sales for the second quarter were $533 million, down 6% quarter-over-quarter. The decrease was primarily driven by lower ASPs, which were down 13% quarter-over-quarter as the reduced prices to sell-through shifted our geographic mix. As Rob said earlier, the quarter started slowly due to the European policy uncertainties, which adversely impacted demand for the quarter. These impacts were partially offset by the positive effects of increased systems revenue recognition driven by Santa Teresa, Paloma and PNM. Our EPC revenue mix increased from 8% of total net sales in the first quarter to 11% of net sales in the second quarter. We also increased the module allocation for the System business during the quarter to meet contract auxiliary schedules, which will result in revenue recognition in the second half of the year. The euro increased from Q1. However, the stronger euro had a minor negative impact on net sales as we were significantly hedged, in line with our long-term strategy. Gross margin was 36.6%, down 9.2 percentage points from the prior quarter. The decline was the result of the ASP's decreases, a mix shift both geographically and to more system sales and an unfavorable FX impact. Module gross margin was 40.2%, down 9.5 percentage points from the prior quarter. Operating expenses were flat quarter-over-quarter. Slightly higher R&D expenses associated with efficiency improvements were offset by lower production startup costs as Frankfurt-Oder and Malaysia ramped. SG&A was flat quarter-over-quarter. We focused on managing our expense growth to help partially offset lower gross margins. Operating income was $65 million compared to $129 million in the first quarter. The decline was due to the decrease in total net sales and gross margins. Operating margin for the quarter was 12.1% compared to 22.8% in…

[Operator Instructions] And our first question comes from Stephen Chin, UBS.

It looks like your upside flexibility to the North American region is now only about 50 megawatts, not the 150 megawatts. Is there a reason why you chose not to upsize North America more? And does this -- it just helps with the 2012 visibility and the 2012 profit margin?

Well, yes, I think originally we said between 4 and 6 when we entered the year. And so what we're trying to do, of course, is to meet our customer expectations and manage the business with the fluctuation from the external markets. So for us, we feel the 450 megawatts are appropriate and deliver on commitments and give us balance between 2011 and '12 in the external market that exists. So we feel pretty good about that, and we have enough demand in other locations to ship products. So the quarter was difficult in Q2, but we feel like 450 is the right number. And as Mark said, as we move closer to the end of the year, it just becomes a limiting factor as to how much you could put on the ground in total.

Next up, we'll take a question from Mark Wienkes, Goldman Sachs.

Some of the crystalline silicon companies over the years about higher efficiency cells and plans for high-efficiency modules. Why are or how are your new high-efficiency cells more credible, that they'll turn [ph] into panels? And then how are you thinking about the strategy and timing of rolling that out?

Yes. I would say this, they're more credible because firstly, they're actually done on manufacturing tooling, right? So it's not something that's done in the lab. As well, we also don't talk about a number of different things, and so as I said, we don't do our R&D work specifically targeting hero cells because hero cells aren't what we can replicate in our facilities. So we don't -- we've had a history of and we maintain a history of only talking about things that we think we can implement in our manufacturing process. So I think in general, we're very confident and made confident by the efforts of the team and the people that you saw there and we talked about there in the FiTs [ph]. So we've had a lot of record cells and as you know, efficiencies in the numbers we use, that 11 7 [ph] that I've stated is an average of all of our production lines. So even on our manufacturing lines today, there's places where we make higher efficiencies, and we make sure to report the average.

Up next, we'll take a question from Steve Milunovich, Bank of America Merrill Lynch.

I wonder if you could at least qualitatively discuss your visibility into 2012, both in the U.S. where I think you've said you're now looking at doing a gigawatt of systems business, plus you have the DOE sweeteners, I guess, that's always come through. And then also, outside the U.S. where nobody really knows what's going to happen, but if you have any thoughts about what next year might look like outside the U.S. Just kind of curious if you could talk qualitatively about your visibility into next year.

Yes. The gigawatt, we mentioned, we plan to build in our objective pipeline systems in 2012. So that number is correct. I think we feel good about the growth and expansion, and especially the support of India in terms of -- we started out with the 10 megawatts in '10, and now 200 megawatts this year and we kind of [ph] signed up to 250 between now and into 2012. So that's pretty exciting, I think. I think we'll see some stability in the markets, as we mentioned in Germany, as they've kind of stabilized somewhat. And we hope to see some policy direction that gives us some clarity on the future of France. So I think in Europe, we're still kind of all bouncing back from the changes that took place in the first half and the challenges over there. So we don't -- we're not obviously talking about 2012, but getting the large pipeline applications funded, and into production is definitely something we're excited about, and we'll be doing a lot of that between now and the end of the year.

Our next question today comes from Smitti Srethapramote, Morgan Stanley.

You mentioned that your module -- your guidance assumption depends on current competitive pricing environment. Can you go into more details regarding what polysilicon pricing or Tier 1 Chinese Module pricing that you're assuming in your guidance?

Well, we assume that we will still have a significant cost advantage regardless of what happens with the poly pricing in general. We still look at the other conversion costs in the $0.75 range. So we feel pretty good about our cost position, and we'll continue to execute on our cost roadmaps.

Next up, we'll hear from Vishal Shah, Deutsche Bank.

Question, you've mentioned India and U.S. have accounted for more than 50% of Q2 revenues, so am I correct in assuming that the rest of the markets accounted for less than 200 megawatts in Q2? And if that's the case, how do you see that run rate going into the second half and more into 2012?

Yes, I think, yes, approximately revenue-wise in mix you'd have to look at where we were doing EPC revenues well. So you can't -- it's not a straight map that way. But in general, we -- that's why we mentioned revenue, but -- the split difference. And I think for the balance of the year, we feel like we've worked with our customers to provide pricing and support to drive sell-through, and then we plan, as you know, to start executing on these major projects in Q3 and Q4. So I think as we said, 2011, the second half will be much better than the first.

Yes, our U.S. business will grow pretty substantially...

Next up, we'll hear from Sanjay Shrestha, Lazard Capital.

Great. Guys, quick question on your 17.3% efficiency that you guys achieved, right. What does that now mean in terms of your long term cost target which previously was kind of I think based on 14.5% efficiency numbers? So can we get significantly below $0.50 a watt because of this record efficiency that you guys were able to achieve?

Well, I think what it does is, and then the reason that we mentioned it is that we're proud of the success and the fact that it's our manufacturing tooling. So we don't, as you know, we don't talk about these things unless we feel that we are well-positioned to be able to apply them. And the significance of it is, is that in a manufacturing environment, that we even beat the record that stood since 2001, which is more the experimental side. So what it means for our roadmap, is we're committed to delivering on the roadmap as we planned and still have that same target range in 2014 to get to that $0.52 to $0.63 range. So but obviously, we're stretching and doing what we can. I think I mentioned during the pitch that we're going to continue to see what we can do to accelerate that roadmap further, and when we do roll out, let you know.

Next up, we'll take a question from Daniel Ries, Collins Stewart.

Operating expenses were flat sequentially. We had been expecting some impact from onetime items such as severance. Did those items occur and the run rate expense fall to offset them during this quarter? How should we think of operating expenses 3Q versus 2Q?

Yes, so the expenses within the quarter did include a severance item, which was slightly below, I think, what we indicated on the call. Last time, we thought it could be in the range of $6 million or so and it ended up being below that. But we did -- we were able to offset that onetime effect within the quarter. And so I would expect at this point in time, the expenses for the second half of the year should essentially be flat with our exit rate. We may be a little bit lower in Q3 and higher in Q4 but think of the second half essentially flat with the exit rate for Q2.

Your next question comes from Timothy Arcuri, Citi.

Actually, I had 2 questions on cost per watt. First of all, I believe it's sort of like comparing apples and oranges when you look at the Chinese, how they report cost per watt and how you report cost per watt. And I'm wondering, if you can give us cost per watt calculated the same way that the Chinese do, i.e. stripping out freight and overhead costs. And then secondly, I'm wondering if you look at cost per watt year-over-year, it hasn't really gone down even if you strip out stock-based comp and you strip out the ramp costs. It's down basically $0.01 even though the efficiency is up like 50 bps. And so I'm wondering why that is and why that's going to change going forward.

Okay. Well, yes, we -- usually we don't talk about it, but from what I know in our analysis, what the competitors do is they exclude things like shipping, warranty, recycling, obviously, SBC, capacity ramp, insurance and other production costs and things like that. So ballpark on a comparative basis if you looked at our cost per watt on the same basis, about $0.68 in that range would be the equivalent comparison. And one of the things that, I don't know, maybe didn't come out during the presentation as well is that we, this year, have changed somewhat how we implement some of the changes with the manufacturing team. And we made the decision based on megawatts produced on a full year basis on line to take lines down for more an extended period. So say a week at a time whereas before we would've done it piecemeal more to keep the lines up and running. And the reason we did that was twofold. So one is the market conditions and challenges in volume created an opportunity for it. And two, we feel like for the total year, off of the same lines, we'll able to produce more megawatts. So it's kind of back to where we were I think last year, about this time where we talked about with all these lines we have and the Copy Smart approach, how do we change and make the changes appropriately and continue to bring the cost down? So it will come in more of a stair-step type of approach versus the incremental change quarter-over-quarter, and that's really what we've been working on and part of what you've seen. So it's still down year-over-year, but it's something that takes more significant effort with all the lines that we have and some of the changes that we're making. So you'll see the benefits of the change and effort we've made in Q4 of this year, both in efficiencies and line run rate. So it'll be a little bit lumpier than our past.

Next up is Rob Stone, Cowen and Company.

I wonder if you could just comment on how you see current conditions in terms of linearity of Q3 versus Q2. And do you think with all the noise about European finance and banking, is access to credit a constraint for projects and module sales in Europe?

Yes. I would say, to take the last part of your questions first, access to credit has been an issue, especially if you look at Italy. Because of the uncertainty and all of the debate and discussion that took place, I mean, no one could get funding for some of the projects that were in Italy because of that uncertainty. So there has been pressure on rates as part of this equation. So I think that's kind of a big deal. And the first part of your question was, again?

The linearity of this quarter versus last quarter.

Yes. I think for us we would expect the linearity to be better in Q3 than in 2 and I think as we said, Mark mentioned, I think, and I did, part of the challenge was our customers trying to figure out what the feed-in tariffs and regulations would mean and how they could get things restarted. So as we said, we did a lot more of our shipments late in the quarter than we would normally do.

Up next, we'll hear from Jesse Pichel, Jefferies. Jesse Pichel - Jefferies & Company, Inc.: You're showing 2014 LCOE at $0.10 to $0.12, and that's an impressive number. What would that equate to in terms of a PPA price building in profits for you and profits for the project owner? Because we're hearing about $0.09 PPAs bid into the California RAM market. I just wanted your thoughts around that.

Yes, okay, I think that when we mention a range like that, as I think we've done, we mentioned before, we view that as pretty competitive with peaking -- other sources of fossil fuel peaking. I mean, you start to take to look at all of them, gas and others. So we think that's a competitive range and when -- probably when you're mentioning these single-digit numbers, that's pretty COD-type of numbers so when you get the time-of-day multipliers, they tend to be in the range that we're talking about. So I understand the question and most of them -- and the time-of-day multiplier in California is about 20%.

Satya Kumar of Credit Suisse has the next question. Satya Kumar - Crédit Suisse AG: I was wondering what portion of the 450-megawatts of systems this year you're thinking will get the loan guarantee in your guidance? And has that assumption changed from a quarter ago? In terms of discrete projects, are you assuming that there will be more than one project getting the loan guarantee that's revenueing [ph] this area?

Yes. Satya, this is Mark. For our guidance, we're still assuming the same assumptions as we had before in terms of dependency around DOE funding. And if we're able to achieve the funding for more projects than we previously had assumed, that could provide additional upside in future periods, probably not this year, but could provide some upside as we move forward.

The 2 projects that are to be built this year, Agua and AV, are the 2 that -- Desert Sunlight, for instance, is not one that we expect to build this year, for instance, or Topaz.

Our next question today comes from Brian Gamble, Simmons & Company. Brian Gamble - Simmons & Company International: Question on the potential for increases earlier on the efficiency side of things. And if you want to go options one by one, that's fine. I was just essentially wondering, with the option that you have, how much potential for additional early-term expenses are there within those options?

Yes. Well, I think we don't really talk about the individual incremental changes that we make. And I think that the team is -- we have what we call a go-fast plan internally. So we have the commitment we have to you as our shareholders and externally to the roadmap that we put in place, priced at [ph] $0.52 to $0.63 and the 8% to 10% improvement year-over-year. And that's kind of how we got there. We mentioned today the improvement, and we're looking at things more and more on an installed cost per watt basis. So we've made 30% improvement in the BoS cost, and if we make the 20% improvement or 19%, we talked about between now and then, we will -- we should be able to exceed the $0.91 to $0.98 range that we talked about in 2009. So we made great progress there. And that really is tradeoffs up between some inflation on raw materials, but then the great work would be EPC team and guys [ph] to continue to drive the cost down. And then anything that we do, of course, on a go-fast level in engineering with Dave and his team contributes to further BoS cost reductions. So at 13.5% to 14.5% range, we have internal plans to accelerate it as quickly as we can and hopefully you'll start to see that in our numbers. And that's why we adopted a little bit of a different approach as to how we upgrade the lines and make changes.

Next is John Hardy, Gleacher & Company. John Hardy - Gleacher & Company, Inc.: Obviously, this year, the working capital adjustments have been pretty severe based on the way the Agua financing came through and how that deal was structured. I was just curious if we could talk a little bit about what we know now, for 2012, whether the cash inflow will be a little bit more evenly distributed and what are some of the factors there?

Yes, I think it's a good question. And clearly, the first half of the year, there was a significant cash outflow associated with ramping some of those large projects like Agua. And we'll start to see the cash inflows start to flow in this year and then in Q3 and Q4, which will drive significant operating cash flows for the second half of the year. And then we would expect a more balanced performance into 2012. Essentially, once the annuity stream starts to flow and the gates open up on the progress payments, you largely are matching your inflows and your outflows pretty well. So we would expect 2012 to be dramatically different than what we've seen here in the first half of 2011.

Joe Osha of Bank of America Merrill Lynch is up next.

Further to my colleague Steve's question, you may have seen the national feed-in tariff announced for China just a couple of weeks ago. Wondering at the sort of utility-scale levels, what your thoughts are there and whether that could impact the APAC demand numbers that you're talking about.

Yes, we see this firstly as a very positive step. It's one that we've been really looking for, so the 1.15, 1.15 RMB per kilowatt hour kind of ties to USD $0.16, USD $0.17 [ph], which is a better scenario than we've seen before. What we need to be able to determine now with the officials there is how to get through the process and ensure that certain projects are included in that feed-in tariff. So that's right now what we are sorting through, but we see it as positive. And I think that you should start to see more volume in China and especially that the mention -- I've heard mention of increasing the targets for 2015, and you heard the mention of the range from 20 to 50 gigawatts by 2020. So they're obviously huge goals in the process and we've been awaiting further clarity on something that makes sense from a sustainable standpoint. And this tariff seems to, and now you've got to get through the process of approvals to make sure that your project gets in the pipeline. So that's what we're working on.

Moving to Ahmar Zaman, Piper Jaffray.

Just a follow-up on that. Now that the Chinese government has announced a national feed-in tariff, can you give us an update on your orders project and where you are on that or what are the next steps there?

Yes. We still have the MoU, sort of initial pilot plan of the 30-megawatt. And then we have an MoU for 2 gigawatts as well and for -- with Ordos. And again, it really is tied getting through the pre-approval process and having the site be selected and put into the feed-in tariff program. So that's what we're working on now. So we are hopeful that we can get that done and participate in what we think will be good growth there.

Next up is Colin Rusch, ThinkEquity.

Can you -- a 2-part question, can you talk about the drivers for the ground [ph], for assumed cost reduction, other than just scale purchasing? And then also comment on demand from South America and sub-Saharan Africa that you're seeing at this point.

Okay. Yes, I mean it comes in 3 ways for us, essentially on the BoS, so there's an element of design. So how do we engineer out [ph] materials and utilize the expertise of our guys from the experience that we've had in building plants to figure out smarter ways to do it. So there's the design element. There is the sourcing element that you mentioned. And then there is the productivity element, the speed with which we get the generation. And that's a big part of the economics and the equation. So that's really the 3 ways. And I think last quarter, we talked a little bit about some inflation pressures, which all you got to do is look at some of the commodities out there, whether it's your [ph] copper or other things, or all this [ph]. We get pressure there, but we work through it from the other 2 means, so design and productivity in terms of how we get things installed. So I think as we get smarter and continue to improve efficiency in our own cost per watt, plus the focus on BoS, that we can drive the installed cost per watt down and get us close to the parity levels that we are targeting for 2014.

And then can you comment on the demand in sub-Saharan Africa and South America?

Yes. I think we've done some initial work and there are some tenders in place. I would say that demand in total is something that's still evolving in those regions. There's a lot of interest. There are no formal policies in place to drive a lot of growth yet, but we're positive that both -- Southern Africa, Northern Africa. And in spots in South America will be a good marketplace, but it will take some time to develop, is what I would say. So probably end of 2012 and beyond before we would start to see any real life [ph].

Our next up is Mark Heller, CLSA. Mark Heller - Credit Agricole Securities (USA) Inc.: I actually have 2. One is just a clarification. Can you say what the EPC revenues were in the quarter? And then the second question is on pricing. I understand that you get better pricing in the U.S., but can you talk about pricing trends, I think for your third-party Module business in the second half of the year?

Yes. I'll take the pricing side of the question. We don't talk specifically about ASP because we try to manage our customers and position in the market by providing and driving sell-through. So we want our customers to develop captive projects in the pipeline, so they need to have some visibility out about a year or so to do that. So what happened in Q2 and really in the first half is that a lot of excess inventory in the system with the demand kind of stalling because of all the things we talked about earlier in the presentation. And when that happens, it puts a lot of pressure on the price side. And we work through a rebate system with our customers to fund them and support them so the price pressure in the external market is high from a third-party sales standpoint and that's part of what impacted our Q2 results. So we're continuing to manage and monitor that. We still feel confident in the concept and our approach to being able to buffer some of those fluctuations with our captive pipeline. And as we get nearer and nearer to having those things finalized and funded, especially here in the second half, as we said, Agua will start to recognize revenue on in Q3. That will become something that is very viable for us to do, is to balance those 2 things together. So we want to be competitive in the third-party market, and we want to work with our customers to ensure they develop markets around the world.

Yes, on the EPC revenue within the quarter, it was right around $60 million or about 11% of our total revenue. And you may remember for the first half -- first quarter, excuse me, was a little less than $50 million so we're about $100 million or so of EPC, and as we guided to, there's about $900 million or so of EPC revenue. So you get an indication of the amount of the ramp we'll see in the second half, and that's obviously reflected in our earnings guidance.

And Kelly Dougherty of Macquarie has the next question.

I just want to follow up on the ASP question. Obviously, we know what happened in the second quarter. So I'm just wondering if you think that taking that lump sum then really has gotten us to a level where prices are appropriate to move products in the second half or if you think that Module pricing is likely to move down further before the end of the year?

Well, I think that for much of our discussion, negotiation with our customers, I think the steps that we've taken should provide opportunity for them to sell through. We also work with them to make sure that certain projects are realized. So we'll get into discussions with them on specific projects of size if they want to execute on, especially in terms of developing new regions in the world. So we have a few processes and the rebate that we discuss with them. we have some special deal pricings that we work on with our customers. So I think from what we know, there was a lot of change in the first half. I think that there won't be as near as much change in the second half, but we'll use those 2 tools to enable to them to sell through.

Our next question will come from Chris Kettenmann, Miller Tabak. Chris Kettenmann - Miller Tabak + Co., LLC: You mentioned increasing project tenders in the Middle East. Just wondering if you could give us a sense of timing there and what you're seeing for contract pricing?

Yes. We still don't -- what we have seen that's around the world and especially in that region of the world, is you see prices kind of converging in terms of getting closer to each other. So remember, these are larger applications and prefilled in significant part. The economics are made easier from the total number of available hours of sunlight, so they -- in some places, it's 2,000-plus hours, so the economics are pretty significant. So I'd say in that way, a good opportunity for all. I think that -- the tenders that are out there are for construction to begin in mid- to late '12 and into '13 and '14. So that's kind of the time frame that we're looking at with some smaller ones, kind of prototype and demonstrators, happening a little bit faster.

Next we'll hear from Stephen Simko, Morningstar Securities Research.

What I wanted to ask about was just in the DOE press release that was issued at the end of June, some of the numbers were given out on the AV Solar Ranch project that implied a capacity factor that would be pretty impressive given the conversion efficiencies of your models. And I just wanted to talk or hear a little bit about what is really driving such strong electricity generation at that site, beyond the inverters that are mentioned in the press release.

Sure, specifically AV Solar? Steve, sorry, I missed -- can you repeat the question? You kind of broke up and I kind of lost you in it. But I asked is it specifically for AV Solar or...

Correct. In the press release, the electricity generated on an annual basis is only provided for the AV Solar project. So I was acting asking about in general, if it applied to other projects as well. I'd be fine to hear that, but it's just -- again, the implications of the capacity factor of that project would -- I think [ph] it would be pretty high given the conversion efficiencies of your models. But I just wanted to see if you had any color on what was driving such strong power generation.

So I think -- so this is Larry, I think -- you have to kind of look at each project by project but in some cases with customers, they'll look for maximizing energy yields, which means they may want more modules, installed DC modules, on a site in order to maximize the yield on the site. In some cases they'll want trackers, and we will be using the rate tracker acquisition that we got on some of the future sites. So it also helps to increase the energy yield. So I think it's going to vary based on the site in terms of the capacity factors that the customer wants on that site. So you're going to see that probably vary more than in the past because customers are getting more sophisticated on whether they want trackers or not or what they want the energy curve to look like or the yields to look like. So it's going to vary quite a bit in the future.

Up next, we'll take a question from Harry Shandra [ph] Ariga [ph].

Another [ph] question on pricing. Is there a point in time where you are comfortable enough to price versus the competition so that the disparity is wide enough that you do not have to resort to rebates as an ongoing feature? Or is the rebate a more like a tactical tool for you?

Yes. I think that we -- originally we had agreements in place with our customers in the form of the framework agreements that we had. And that -- we wanted those to remain in place and that's why we chose the rebate vehicle. So that's one. I think that we are working with customers to discuss how we would change our agreement so that we didn't have to go through the rebate process. And we have some ideas that we'll be working on in the second half of 2011. So yes, to answer your question, we will look out ways to simplify the approach and provide pricing in a different way to our customers so they can drive sell-through. But we originally did that because of the agreements we had in place.

Our last question today comes from Ramesh Misra, Brigantine Advisors.

In terms of your [indiscernible] cost of $0.99 per watt, is that on the Series 3 modules? And does that at also include trackers? If not, how would those 2 elements impact balance of the [indiscernible]

No, we do it the same way as we do with the modules, so it's an average number. So it would be across, whether it's Series 3 or 2. And most of our work, of course, now, is Series 3. We actually -- part of our efforts during Q2 were to convert completely to Series 3. So we're Series 3 across the entire product line now. So we want to make sure that we had positioned Series 2 inventory to address any customer issues. But we're Series 3 now. But we look at it, and when we give you a number, it's an average number. And as Larry said, for all kinds of reasons, the costs on a per-watt level for BoS from site to site will vary from land conditions to whether it has a tracker or what the installation is.

And ladies and gentlemen, that is all the time we have for questions today. That also does conclude our conference. We would like to thank you all for your participation, and have a great day.

Thank you.

Thank you.