SunPower Corporation (SPWR) 2007 Q3 法說會逐字稿

Good morning and good afternoon, and thank you for standing by.

And welcome to SunPower's third quarter earnings release conference call.

Your lines have been placed in a listen-only mode until the question-and-answer segment of today's call.

This call is being recorded.

If you have any objections you may disconnect at this time.

I would now like to turn the call over to Mr.

Tom Werner, CEO of SunPower.

Sir, you may begin.

Thank you for joining us today.

Today we will report on our third quarter 2007 manufacturing and technology channel and financial results and provide guidance for the fourth quarter 2007 and 2008.

We had another solid quarter as our third quarter performance beat our top line and bottom line performance.

Our Q3 2007 revenue was $234.3 million, up 35% from our Q2 2007 revenue, comparable to our 2006 full-year revenue of $236.5 million.

Our Systems revenue was $157.7 million, a 51.6% increase compared to Q2 and our components revenue was $76.6 million, a 9.9% increase from the prior quarter.

Note that SunPower panels used by our Systems business segment accounted for within the system financial results rather than in the component segments results.

Later in the call, Manny will go into more detail on our EPS performance.

Note that we beat EPS guidance in Q3 and therefore are going to exceed our guidance for the year.

And Manny will go into more detail in that after I finish.

Now let me start with strategy.

I'd like to highlight our Solar America initiative contract which we signed in September.

We were awarded a $24.7 million three-year contract by the Department of Energy, the largest of any company.

We believe this reflects confidence in our vertical integration strategy as a means to ring in efficiencies out of the value chain, lower installed system cost.

After three quarters of working as an integrated company after our PowerLight merger, we are more confident than ever that our integration strategy is already creating value to our customers and shareholders.

Now let me walk you through the value chain to reflect on the progress we're making at improving efficiencies and lowering costs.

First let me cover silicon.

We remain in a very dynamic environment where the next generation of major new polysilicon supply capacity is beginning to ramp to significant production.

In the third quarter, one of our silicon partners, [Emsetec], continued to ramp their first polysilicon manufacturing plant which will ultimately support their increased delivery of (inaudible) goods to us.

We can report substantial progress at Emsetec.

They have received Japanese government approval to install and implement their plans to plan 3,000 metric tons capacity.

To accomplish this, Emsetec is converting to producing their own TCS gas.

This transition requires their plant to be decommissioned briefly.

We expect Emsetec to recommission the plant in Q4 with full capacity brought online with in-house production of TCS gas to support the expected reactor capacity.

In Q3 we provided Emsetec sufficient delivers of polysilicons to ensure that our ingot production continued in adequate volumes.

We expect to continue to support Emsetec with polysilicon in Q4 with diminishing quantities as they begin production of their own TCS gas.

Now let me speak to DC Chemical.

They've made significant progress in their polysilicon plant and it is on schedule for meeting their contract delivery dead line in the first quarter of 2008.

It's also important to note that DC Chemical is starting their factory with their own TCS gas production and that they have significant experience through their subsidiary [Soldiv] producing TCS gas.

Now let me discuss progress on the silicon agreement front.

Our success in partnering with Emsetec and DCC to expand the ranks of polysilicon manufacturers is part of our portfolio approach to rationalizing our long-term silicon supply chain from polysilicon to ingots and wafers.

In that last quarter, we have taken several major steps to expand our silicon supply.

First, we signed a two gigawatt ten-year agreement with Hemlock that begins in 2010.

Additionally we signed a 53 megawatt agreement with Fokker, a three year agreement, it's our third with Fokker and it begins in 2009.

Now let me speak to progress on turning polysilicon into ingots.

We signed an outsourcing agreement with a company called NorSun to produce ingots with polysilicon that we procure primarily from Hemlock.

Further down the supply chain in terms of wafering, we signed another outsourcing agreement with First Philippine Electric, a local Philippine partner with extensive relevant manufacturing experience.

As a result of this JV, we will have more space in our second fab and now plan to install 12 lines.

That's two more than originally planned.

In general, we continue to focus on working with partners or developing JVs around areas of value chain where vertical integration is less of an advantage, where partners can add value.

Now let me speak to manufacturing.

In manufacturing, we expect our name plate capacity for Fab 2 to grow from 330 megawatts annually to 466 by the end of 2009 due to two adjustments we've made to our plant.

Two extra lines that I mentioned earlier in Fab 2 will add 80 megawatts thanks to space created with that joint venture with [Ersilic].

And we expect to increase throughput on each of our lines to 40 megawatts from 33 megawatts beginning with line seven, which will begin ramping production at the end of Q4.

So to summarize, Fab 2 has lines 5 and 6 currently ramping to 33 megawatts annually.

Then we will add ten more 40-megawatt lines.

One of these lines, line 7, begins production at the end of this quarter.

Five more will ramp next year and the last four will ramp in 2009.

Our manufacturing success in Fab 2 has now reduced our silicon utilization to less than 7 grams per watt, paying dividends by offsetting short-term contract pricing which has has increased 15 to 20% over 2006 pricing.

We believe that through R&D, supported by our SAI contract, we will be able to combine additional efficiency gains in cells and panels with thinner wafers and achieve a silicon utilization of less than 6 grams per watt within a few years.

On the efficiency front, we are pleased with the success of our generation 2 solar cells now in mass production.

As reported in Milan, we have achieved median efficiency of 22.4% cell efficiency in production.

When assembled into a solar panel, India National Lab has measured our panel efficiency to exceed 20%.

You would think of this as something like breaking the four-minute mile equivalent in the solar world.

On wafer thickness, we've been running sample lots of our production lines with 145 micron wafers.

Let me end with the conversation or description about our channel strategy.

As our solar technology continues to draw strong demand and command a premium in the market, our strategy is to ensure that we have a diversified approach across customer segments, applications and geographies to match the high value placed on our efficiency and aesthetics.

For smaller scale applications, we see both retrofit and new homes performing well.

Residential retrofit we now work with more than 125 dealers, 25 states, to deliver solar systems on residential and small commercial roofs.

We have trained more than 400 installers from these dealer companies and have established a premier dealer network offering training, financing services, lead generation and logistics.

For new homes we were happy that our partner Lennar received a major industry award at our national conference Long Beach, recognizing their leadership in implementing solar as a standard feature across California.

We are also working with over a dozen other home builders in California, representing a major share of the market and finding substantial interest in solar as a differentiator and sales velocity improvement tool in a down housing market.

In large commercial systems, we are on schedule to deliver our first systems under our multi-site contracts with Macy's and Wal-Mart, as well as having sold several new systems in the last quarter.

We have sold systems to AC Transit, Agilent Technologies and the San Jose Tech Museum of Innovation, as an example of our SunPower Access power purchase program and how it's expanding.

And this is where our customers buy power from their solar energy system instead of buying the system outright.

In power plants, we are on schedule for construction of the Nellis Air Force base solar power plant, which will be the largest photovoltaic plant in the United States.

We signed a contract in Spain at a location called Olivenza for our largest EPC agreement to date and the size of that contract is 18 megawatts.

This facility called Olivenza will be the first application of our next generation tracker.

This is the European version of the tracker that we use at Nellis, which generates up to 30% more energy from the same solar panel as a fixed tilt application.

Additionally, we're excited to see Korea moving forward as well with another two megawatt power plant at Jeonju.

Expansion of our channel strategy is where we have the most tangible evidence of the value of integration SunPower PowerLight completed at the beginning of this year.

Market visibility we have through direct customer contact has affected our market entry and expansion plans, technology road maps and our growth strategy.

Combination of high efficiency solar panels from SunPower, high energy delivery from the systems technology that originated with PowerLight is a potent combination for our customers which allows us to win business and plan for aggressive cost reduction over the next few years.

We will be able to serve more of our systems business segment's solar panel supply from SunPower as we grow our manufacturing in the Philippines, which will improve our systems margin, as well as meet the needs of customers with space constraints on both their roofs and in power plant sites.

I would like to turn the call over to Manny Hernandez who will report details of our Q3 reports and provide guidance for Q4 2007 and a couple comments 2008.

Manny?

Thanks, Tom.

Good morning everyone and thank you for joining SunPower's earnings conference for the third quarter of 2007 which ended September 30th of this year.

I'd like to remind everyone that during the conference, management made and will be making statements that are not historical in nature.

Please consider those statements as forward-looking pursuant to the Private Securities Litigation Reform Act of 1995.

Those statements are based on our current expectations and are subject to certain risks.

Please refer to our press release and our SEC filings for a more detailed discussion of those risks.

I will now give you a summary of our 2007 third quarter financial results for the combined company.

Revenue for the 2000 [sic - see Press Release] third quarter was $234.3 million, up approximately 35% from the prior quarter revenue of $173.8 million and up approximately 260% from the year ago third quarter revenue of $65.3 million.

The component segment of our business accounted for $76.6 million of the third quarter revenue, approximately 10% increase from the prior quarter revenue of $69.7 million.

Our systems segment accounted for $157.7 million of the quarter's revenue, up approximately 52% from the prior quarter revenue of $104 million.

Our systems segment represented a greater portion of the total company's revenue this quarter, representing 67.3% of total company, versus 59.8% last quarter.

The faster than anticipated completion of certain projects in North America and Spain, most notably the Nellis Air Force base project that Tom mentioned, contributed to a very strong systems performance during the quarter.

As Tom noted earlier, the eventual sale of SunPower manufactured components which are allocated by the company to the system segment is reflected as revenue of the systems segment.

In the 2007 third quarter, approximately 25% of panels installed in systems projects were SunPower manufactured panels.

That's up from 21% last quarter.

On a GAAP basis, SunPower reported operating income of $6.8 million and net income equivalent of $0.10 per share.

This figure includes non-cash charges for amortization of purchase accounting intangibles of $6.9 million and non-stock -- non-cash stock-based compensation of $13.4 million.

On a non-GAAP basis, adjusted to exclude non-cash charges for amortization of intangibles, stock-based compensation and their related tax effect, SunPower reported a total operating income of $27 million and diluted net income of $0.33 per share.

This compares with the prior quarter's operating income of $22 million and $0.25 diluted net income per share.

The total company non-GAAP gross margin for the 2007 third quarter was 20.4% compared to gross margin of 23% in the prior quarter.

This third quarter gross margin was influenced largely by the higher mix of revenue in the systems segment with achieved gross margin of 18.7%, while the component segment achieved gross margin of 24.1%.

Couple of notables on gross margin.

Our systems segment achieved gross margin within our guided range and was largely influenced by the Nellis project, which accounted for approximately 39% of the systems segment's revenue for the quarter.

Our component segment also achieved gross margin within our guided range and was influenced by our need to procure additional polysilicon to support our ingot partner Emsetec, who this the process of ramping its first polysilicon factory.

This resulted in a slightly nonlinear supply of silicon during the quarter, which impacted our opportunity to produce more units.

Absent this issue, however, we would have been in the higher end of our guidance.

The key take-away here is that the vertically integrated and diversified company, we are able to mitigate issues like this and still deliver great results.

Briefly on the combined balance sheet, we ended the third quarter with cash including short term investments and restricted cash of approximately $495 million, which reflects the financing transaction that we completed during the quarter.

Note that during Q3 the company completed a financing transaction where we issued 225 million in convertible debenture and sold approximately 174 million of common stock.

Our DSO improved to approximately 32 days, while our net inventory closed at approximately 49 days.

Capital expenditure for the year is now estimated at $210 million and full year's depreciation is estimated at $29 million.

As we have noted in prior calls, our business results may reflect quarterly shifts in mix between systems and component segment revenue.

Also from quarter to quarter, we expect shift even within the systems segment, due to size and type of project and percent completion factors that could lead to non-sequential or marginal growth in revenue margins or earnings.

The second half of 2007 is essentially shaping up as we guided you last quarter, specifically, that we expected a very strong Q3 followed by a slight decline in Q4, influenced largely by previously committed project schedules.

Notwithstanding our better than expected Q3 results, however, we are holding our previous guidance for Q4.

The following is our guidance for the fourth quarter and for the fiscal year 2007.

For Q4 our total company estimated non-GAAP results are as follows: revenue of $210 million to $220 million, gross margin of 24 to 25% and earnings per share of $0.33 to $0.37 per share.

For Q4 '07 by segment, our non-GAAP estimates are as follows.

Systems segment revenue of $120 million to $125 million.

Component segment revenue of $90 million to $95 million.

In the fourth quarter, we expect an increase in planned allocation of SunPower panels to the systems segment due to strong internal demand such that approximately 35% of panels installed in our systems projects will be SunPower panels versus 25% in the third quarter.

We expect our component segment gross margin to improve by as much as 200 basis points to 25 to 26%.

This gross margin guidance comprehends the continued ramp of lines five to seven with line seven coming on towards the end of the quarter.

It also includes some additional procurement of polysilicon to support Emsetec.

As noted by Tom earlier, supply of polysilicon to Emsetec will diminish during the quarter as they begin poly production with their own source of TCS gas.

This should no longer be an issue going forward and we should get back on track as far as hitting our gross margin model of 30% early next year.

Our systems segment gross margin for the fourth quarter is expected to increase over 600 basis points in the 23 to 24% range, benefiting from OEM sale of balance of SystemTracker system with our deliverable in the quarter.

Collectively, we expect total company gross margin to improve by approximately 400 to 500 basis points from 20% in Q3 to 24 to 25% in Q4.

For the fiscal year 2007, we now expect the following non-GAAP results.

Revenue of approximately $760 million to $770 million.

Earnings per share of $1.20 to $1.24.

These earnings are based on our non-GAAP tax rate of 11.5% for the year.

Lastly, as noted in the press release for fiscal year 2008, we now estimate total company revenue of $1.1 billion to $1.25 billion, and a non-GAAP EPS of approximately $1.90 to $2.05 per share.

Let me now turn it over to Tom to lead us through the Q&A session.

Thanks, Manny.

We're going to go to questions in a moment.

First, I would like to introduce who will answer your questions with me.

We have Howard Wenger, our VP of Global Business Units, Peter Aschenbrenner, our VP of Marketing, Julie Blunden, our VP of Public Policy and Corporate Communications and Mike Armsby our VP of Finance.

So that we have adequate time and you don't beat me up for going too long on the call, if you would please ask -- limit yourself to one question and one follow-up and if there is more questions that you have, if you could get back in queue we would appreciate it.

We'll do something a little different than we've done previously on calls.

There were three homework assignments we took from the last call and we'll briefly cover those and then we'll go to questions.

The first two of those homework assignments, the first one was on our cost reduction road map our commitment of 50% cost reduction by 2012, what does the profile of that look like.

I'll just say that it depends on which part of the cost reduction you're looking at.

In terms of the downstream channel cost reduction, that will tend to be more back end loaded, meaning that will be in the out years.

Whereas in terms of cell and module cost reduction, that is more front end loaded.

So as we get into '08 we'll probably start to give you some specific measures to track us to the 50%.

But it's mixed is the answer to your question, to that question.

The second one that we tabled was what are your silicon costs doing in 2008?

Because we have talked about long-term contracts coming on line in 2008.

We've spoken to our silicon vendor partners and they are not comfortable with us giving any guidance on silicon cost so unfortunately, we can't give you precision on that, although we appreciate the question.

The third question I'll turn to Manny and then we'll go to your live questions.

The third question was how much synergy or benefit do we derive from big scale projects.

So for the purpose of responding to that question, we're using two megawatts or bigger and it's actually a great question because we do get scale and economies from big projects like this from a profit contribution standpoint, not just gross margin.

You do benefit from lower overhead support because you don't have to increase the support for a bigger project necessarily.

And you also get operating expense leverage for huge projects like this.

As you can appreciate, since this is largely a negotiation tool, we'll give you a range.

The benefit for large scale projects is anywhere from 0 to 5% from the historical deals we've done, it could be as much as zero if you want to use the benefit to more competitively bid on the project for example.

Thanks for the question.

Okay.

So Michelle, we'll go to questions in the queue.

Thank you, sir.

(OPERATOR INSTRUCTIONS) Please limit all questions to one question.

Steve O'Rourke you may ask your question.

Hi this is Steve O'Rourke from Deutsche Bank.

Thank you.

Just I'll ask one very quick one and then a follow-up.

What will your internal sourcing for PowerLight be in Q1 and going into 2008?

Does it hold to the prior comment you made on percentages or will it go up?

It's Manny.

The answer is yes.

The goal or the stated goal of the Company has been that 30% of installed systems would use SunPower material this year and we're certainly trending that way.

Estimated to increase to 50% in '08 and even higher approaching maybe 70 to 80% in the ensuing years.

One caveat to that is we obviously as a company want the flexibility of being able to redirect modules appropriately depending on the business at hand.

But those ranges are good for planning purposes.

And Tom, if I could ask one other.

Can you kind of articulate your longer term strategy for selling energy rather than simply selling systems.

I know you're moving in that direction.

But where might it make sense to start doing that more aggressively around the world in the next year or two?

Selling energy --

Electricity.

reduces the friction in the selling cycle because customers are used to paying for energy.

And for customers like Wal-Mart and Macy's as two examples, they're buying power in some cases and not necessarily buying a solar system so that when they justify it to their board, they justify it on the economics of buying power.

And then they have the added benefit of they're doing something good for the environment.

So in the enterprise channel, call it, it's a natural fit because of the nature of how those projects are justified internally.

There's definitely -- by the way, that's gone as well on power plants and that's pretty much the norm in Europe.

It will be interesting to see our ability to move that into other parts of the market.

We believe we will be able to do that.

We believe we're uniquely situated to do that because of our scale and because the amount of transactions we do in the other parts of our business.

Howard, did you want to add anything else or is that --

No, I have nothing further to add.

Okay, Steve.

That hopefully answers your question.

Our next question comes from Sanjay Shrestha.

Lazar Capital Markets.

Good morning, guys.

Couple quick questions.

First, you talked about the component margin getting impacted on a sequential basis.

How us some more detail on how much of that impact was related to having to fulfill poly to support Emsetec and how much of that was related to the ramp-up cost?

Sure.

I'll turn it to Manny in just a second.

He can give you a rough split.

As you can imagine, as you grow as rapidly as we're growing, that you're not going to have perfect linearity.

Absolutely.

And delivery as you correctly asked the question, it does split into linearity of delivery as well as buying a little more poly, so Manny, do you want to --

Hi, Sanjay.

Just to give you a bit of a range, using the high end of our component margin guidance as the benchmark, I would characterize the impact of 2/3 silicon related, 1/3 ramp related.

A quick follow-up, then, guys.

Given what just happened here in the near term, out of the poly availability that we're talking about for 2008, how much of that is -- obviously you guys have a portfolio approach.

There's a lot of cushion there.

I understand that.

But can you sort of talk about how much do you actually expect to get from Emsetec and DCC during 2008?

Yes, so when we plan silicon the way we do it, Sanjay, is we use an expected value so we have a supply multiplied times the probability, that gives you an expected value.

Of course we add those up.

And I'll repeat.

We have sufficient silicon to match our guidance based on that.

And as we have said previously, some of Emsetec DCC is north of half of our supply next year.

Okay.

Perfect.

I'll head back in the queue then guys.

Thank you.

Satya Kumar, you may ask your question.

Thanks for taking my question.

It's Satya Kumar from Credit Suisse.

I just wanted you to explain the dynamic here.

You're not actually increasing your megawatts of silicon available for next year but you are increasing your revenues and earnings for next year.

What's driving that?

Hi, it's Manny.

The benefit of having an integrated company now is we actually have opportunity to upside one or the other side of the house.

So some of our tweaking for our guidance is largely just a mix between component and system.

Okay.

Quick follow-up to that.

You recently talked about having 100 megawatts of cumulative orders in Spain.

Can you talk about what part of the orders have been posted so far and also if you can also tell me what your megawatts of component production was in the quarter.

Thanks.

Okay.

So we'll take that as two separate parts.

Manny can talk to megawatts production and I would ask Manny to cover once again Q3 and as long as we're answering it, we'll give a flavor for what Q4 looks like.

Then we'll turn it to Howard who can answer the first part of your question.

So Manny.

The megawatt answer is flat from last quarter, if you recall last quarter was 24.5 megawatts, so we were practically flat in Q3.

Going forward, though, with the lines ramping and better visibility on silicon, our estimated output is 34 megawatts or at least 40% increase from the prior.

And this is Howard.

I'll respond to the 100 megawatt question in Spain.

We are in various stages of completion of those 100 megawatts from booked orders that have not started, so zero percent, through to 25, 50, 75% completion on those projects.

I would say that the majority has not been fully constructed.

Thank you.

Thank you.

Rob Stone, you may ask your question.

Please state your company name.

Cowen & Co.

First question, the way you guys describe the transfer of production from component to systems seems like it's a little harder to calculate than if you would describe it has the proportion of your production that's going to the Systems business since it's hard to know how much the systems business may be doing in megawatts of product coming from third parties.

So could you restate the targets for supply of megawatts from the components to the systems business that way?

So Rob, we thought we worked it out with all of you folks on which way to do that.

We'll continue to evolve with you.

We're certainly not trying to be difficult.

So I'll let Manny try to add clarification and perhaps with the follow-up calls we can make sure we're honing in on the best way to communicate.

Okay, Rob.

I can -- the way we are providing answers to that question now where we say 30% of systems installed product utilize SunPower modules, then the reciprocal is therefore the derived third-party module.

I understand that.

But that presumes that you know what the total megawatts is of the systems business.

It's much easier actually to track the megawatts that you're producing.

In other words, I'm asking the question a different way.

If we were looking at the total megawatts that you produce, what portion of that is sold externally versus transferred to the systems business.

So Rob, we got it.

Let us table that for now.

We'll come back to it perhaps later in the call.

Certainly we can do it as a follow-up.

I understand you're basically asking for the split of the output.

So give us a little bit of time on that one.

Okay.

A follow-up, if I could.

On the systems business, can you give us some flavor for how much relative proportion balance of systems components, the mounting and tracking systems contribute to overall revenues?

Sure.

And we had a dynamic at the beginning of the year where we called that a significant one-time event.

And we're looking back on that, feeling like that wasn't a good characterization.

It is part of our ongoing business and you should consider it as part of the way we do business.

Is part of our channel strategy.

Having said that, I'll turn it to Howard who can give you a sense for that split.

Yes, Tom.

It's approximately 5 to 8% of our total business in the systems channel.

I'm sorry, I missed that percentage.

5 to 8% in the systems channel.

5 to 8% of systems.

Thank you very much.

Vishal Shah, you may state your question.

And please state your company name.

Hi, Lehman Brothers.

Thanks for taking my question.

Tom, one of your largest competitors in Japan recently announced plans to start mass production thin film based solar cells and he said that the move was being made to keep them competitive against rising prices of silicon.

Just wanted to know what your thoughts were with respect to thin film technology given the tightness of the polysilicon market and if you have any plans going forward.

Thank you.

Sure.

So I have several thoughts on this.

One is that it was a really good question earlier about selling energy and moving towards a selling of energy so when we make a strategic decision about high efficiency crystalline-based silicon versus thin film we're using the metric of levelized costs of energy or just think of it as the cost of energy and we believe that our solution will be the most effective for our customers on levelized cost of energy or the cost of energy.

If we felt differently of course, either organically or acquire our way into thin film and we're not planning on doing that currently.

The other comment I'd make is we do expect that silicon will come into balance and that there will be a significantly increased availability of silicon in the back half of 2008 and 2009.

So it's a little bit of buy high sell low for us at this point if we jumped into thin films in reaction to a tightness in silicon, because we think it's going to improve, certainly for us as a company is improving over the next 6 to 12 months.

So we wouldn't want to overreact to something that we believe we're solving another way.

So that would be the two aspects of how we think of that.

All right.

Thank you.

Thank you.

Our next question comes from Mr.

David Edwards.

You may go ahead and please state your company name.

Hi.

Morgan Stanley.

Question for you about Spain.

Understand that a couple weeks back the Spanish government released a proposal for what the program should look like starting in 2009.

Now we're hearing that the negotiations may actually bump up the volume and bring up the tariff price.

Wondering if you guys have any comments on that.

Yes, we do.

Howard will make most of them.

I would just give you the context that there's a reason why we have a 50% cost reduction plan by 2012.

It's that we expect that -- and it's built into most of the incentive systems that the incentives will go down in time and it's our part of the deal that we'll reduce costs such that the economics to the customer remain excellent.

Now, of course your specific question is the dynamic in Spain.

I'll let Howard speak to that.

Thanks, Tom.

First I would like to say that we're happy with what the Spanish government is doing because historically in our industry, if you look back five or eight years ago, policy and incentives were very stop and go kind of proposition.

Now since our industry has reached critical mass in my opinion, governments are paying a lot more attention to being very proactive about ensuring continuity in markets.

And so we're very happy that Spanish government has taken a stand and said okay, we're going to increase the cap from 371 megawatts to 1200 megawatts and we're actually thinking of going to 2,000 megawatts.

That's good and that's actually given a fair amount of clarity in the market already.

And so with respect to the tariff, there have been a lot of discussions and so forth and numbers thrown around.

That's still in play.

It's moving in the right direction from our perspective in terms of the reduction and so we expect that market to continue to be very strong for us going forward.

Great.

Thanks.

Thank you.

Our next question comes from Mr.

Al Kaschalk.

You may go ahead and please state your company name.

Wedbush Morgan.

Tom, I just wanted to follow up on the silicon question and if you could give a little bit more granularity as we roll into '08 and specifically the front half versus the second half.

How much are you using spot versus more mid-term and long-term contracts?

Yes, Al, thanks for the question.

We purposely said short term versus spot so I'll answer your question, spot is a very small percentage of what we buy.

Below 5%.

Now, let me further answer your question, however.

The way we characterize short term is within a year and buying from existing partners and therefore it's significantly different dynamic than spot.

Tends not to be a one-time transaction at a very high price.

And that percentage varies by quarter and that still is a very small number as well but that's less than 10%.

Okay.

So if I may just then, in terms of margin improvement, and I know this is really kind of granular here, but in terms of what you're setting up for '08 margin expansion opportunities, how much of that is due to the silicon pricing dynamic, in other words, your mid-or short-term coming online in a greater level?

Let me just paint broad picture then drill down to your question.

We have significant improvements in our factory in the Philippines.

We are getting more watts per wafer out of our first generation technology through manufacturing improvements and of course our second generation technology has median efficiency that is almost two points higher.

So we get more watts per wafer.

We're also using thinner wafers.

Therefor we end up with less grams per watt.

That's a huge driver in terms of cost reduction in 2008.

And of course, as we scale, then we absorb the overhead of the second Fab as well as the pre-op.

I don't recall if Manny specified the pre-op for the quarter but the pre-op number is significant and that goes away as we ramp that facility.

So we get the benefit of that as well.

You're right, silicon is an important part of the cost reduction.

And it's roughly 25 to 50%.

It depends on the time of year and of course it depends on the execution of our partners, but 25 to 50% of our cost reduction in solar cells in 2008.

Great.

Thanks.

Great execution by the team.

Thank you.

Thank you.

Michael Carboy you may go ahead and ask your question.

Please state your company name.

Signal Hill.

Good morning, ladies and gentlemen.

Let's focus on the upside of PowerLight this quarter.

Normally construction schedules are things that are pretty well known and understood.

Could you give us a little bit of insight into what suddenly accelerated the programs, Howard, and sort of how do we deal from a planning perspective with the opposite side of the coin when perhaps schedules may drag out?

It's a good question.

We're doing a number of things to shorten our basically book-to-build time and one of them is in the area of technology innovation.

So for example, in the Nellis project, which we realized a good amount of revenue in the third quarter, we implemented our new power tracking technology which does not rely on ground penetration.

So we're actually modeling it on our leading roof technology which also has no roof penetration.

We kind of transported that idea to the ground.

So it enables us to install much faster, much more predictably and we're doing it, a lot of that construction in a manufacturing facility and taking it to the site.

And that improves quality control as well.

So in terms of going forward, the mix of our projects, the size of our projects, where they are geographically, all bear on construction schedules and I believe that we guided in the past to -- from book-to-bill on the order of six to nine months and we're looking to shorten that time and I think in the future we'll be able to provide more information as we go forward and implement more of our leading power tracker technology in the field.

This is Peter Aschenbrenner.

Let me add very quickly to that.

The Nellis project, which was in the last quarter, largely, that was the first project that utilized this new T20 tracker technology.

We were a little bit conservative on your schedule because of that, our planning schedule, because we -- it's the first time we had fielded it.

And as we got into it we realized that we were exceeding or hitting our original estimates and being able to install on the rate of a megawatt a week which is unprecedented.

And that's really what helped accelerate that project schedule, which was a big piece of Q3.

Sorry to extend this.

Michael, appreciate the question.

The other thing I would say is great execution by Dan Sugar and his systems team and the team in the field.

They did an exceptional job.

So we're really thrilled with the work that the folks in the field did.

We have 18 questions in queue and so we're going to be little less eloquent in our answers so we get through them.

Next question, please.

Thank you.

Pavel Molchanov, you may go ahead and ask your question.

Please state your company name.

Raymond James.

Just a big picture question for you guys.

Right now something like 95% of the PD market is in OCD countries.

When or I suppose how do you think that changes whereby emerging markets become hopefully over time the major drivers in this business?

We'll have Peter take that question.

Yes, we believe that the majority of the market will continue to be in the OECD countries for the foreseeable future.

What we do believe will happen is that the percentage of the total market that those say five or six top markets comprise will diminish over time.

So if you look at the top five or six countries today, they are something in the order of high 70s or 80% of the total market and we see that falling over the next three to five years to something closer to perhaps the mid-60s.

So the new markets will grow and grow very fast but they won't outgrow the OECD markets.

And a quick follow-up, who do you think that that second tier of markets that's going to be ramping up?

I think it's probably the logical candidates, India, China, perhaps Brazil, that's the quick answer.

Got it.

Thanks very much.

Thank you.

Corey Tobin, you may ask your question.

Please state your company name.

Corey Tobin from William Blair.

Good afternoon.

Two quick ones.

Anything on pricing with respect to average price per watt or just general comments on pricing and then any change to the timing of achieving the long-term margin model that you've discussed in the past?

Thanks.

Howard will take the first part.

Manny the second.

The demand is very strong.

We expect ASP to remain steady on a global basis in Q4 and into 2008.

That's the short answer.

And long-term gross margin milestone or model Corey is the same.

Hoping to achieve it fourth quarter of '08 or first quarter of '09.

Thanks.

Thank you.

Jeff Osborne, you may go ahead.

Please state your company name.

Thomas Weisel Partners.

I just had a follow-up on the pricing question.

Do you have any thoughts on the pricing for next year?

Do you think it's down 5% or a little less than that?

And then just for Manny, any thoughts on tax rate and CapEx for '08?

This is Howard.

I'll take the first part.

Pricing as I mentioned before, we think will remain steady going into 2008.

Of course it depends on geography and mix.

And so in the more mature markets like Japan, it is the second largest market in the world but has the lowest ASP, contrasted by Spain and Italy, emerging markets that are better incentives, higher ASPs.

So it really depends on the mix.

But I do want to echo something that Tom mentioned previously which is we are driving to a model to be competitive with the electric brand and we expect ASPs overtime to decrease to increase our competitiveness.

Are you saying you penetrated Japan?

I thought you hadn't.

You weren't selling anything materially to that market.

That's correct at this time.

Okay.

Great.

Then for Manny, any thoughts on tax and CapEx.

Tax rate for 2008 non-GAAP is 22.5%.

And GAAP tax rate of 26%.

As far as capital expenditure for 2008 our early estimates are at the $250 million to $300 million range.

Note that the following capital activities will be covered by that.

One, is five more manufacturing lines.

Three module lines in the beginning of the expenditure for Fab 3 which is our third facility.

Great.

Thank you.

Thank you.

[Mark Hiller], you may go ahead and please state your company name.

Thanks, Merrill Lynch.

Manny, I just had a clarification.

I guess I'm just still trying to figure out the margin dynamics for Q4.

So you're lowering your component gross margin guidance because of you're sourcing I guess more polysilicon from the spot market and your systems gross margin is going up sequentially because you're sourcing more solar cells or panels from SunPower into PowerLight.

I just want to confirm that that's the case.

Those are both accurate, plus, I would only add that on the systems side they're also going to benefit from OEM sale of certain balance of systems that's going to help the margin in the quarter as well.

And implementing the new products that Howard spoke about.

Okay.

Got it.

And in terms of the gross margin, have you guided to a gross margin for 2008 and where -- is the improvement more going to come from the poly side or is there some other part of the supply chain that gross margin improvement is going to come from?

This is Tom.

We'll get to more detailed metrics on 2008 as we get closer to 2008.

And in other words, probably in the next earnings call.

In terms of the breakdown, we provided overall guidance on where the cost reduction comes from.

I think of a target of 50%.

Where 40 or 50% of that 50% comes from downstream scaling silicon.

And when I say silicon I'm thinking of polysilicon ingot wafer so the combination of those things could be 20%.

Conversion efficiency, 15%.

And then scaling the factory will give you another 5 or 10% in terms of a rough overview, you should add up, if my math was correct you'll add up to 60% because we're giving ourselves some room that not all the plans will come to fruition.

And that will get us to our 50% target.

That's a long winded way of saying silicon is 20 points and as I said, that's a combination of long-term partner contracts coming online starting in the first quarter of next year and then throughout the year.

That's also higher efficiency even on our first generation technology and that's also using thinner wafers so it's a combination of those things.

Thank you.

Thank you.

Stuart Bush, you may go ahead and please state your company name.

Hi, good morning, it's RBC Capital Markets.

My question is about Germany.

It looks like the government there is going to be reducing the tariff levels significantly more than it expected before starting in '09 with a one-time down charge.

Are you factoring into your estimates for '08 a pull-through of demand into '08 from '09 and so the second question is what percentage of your sales are you factoring in to be through Germany for that year?

We'll split this question into two.

I'll let Howard comment on percentage breakdowns and regions and Julie Blunden will speak to the market dynamics.

This is Julie.

The German government has given everybody lots of visibility that they intend to change or evaluate and likely change EEG across the board including PV and certainly everybody has got good understanding of when that transition would happen.

No sooner than 1/1/09.

So I think as with all adjustments in program rules, there's likely to be some folks that are looking to get in before the adjustment changes so it's entirely possible that you'll see an uptick in Germany just due to that factor, regardless of what the results end up being as they resolve the -- exactly what their tariff level will be post transition.

But we'll have pretty good visibility on that.

The Germans are pretty good about that.

I don't think there will be any real surprises there.

This is Howard.

Regarding the amount of product that flows into Germany from SunPower it's less than 10% of our total so we're not very dependent on that country at the moment but it's a very important market still.

It's number one and we do expect it to continue to be a viable market.

Especially for our product where space is a premium on roofs in Germany.

Great.

Thanks a lot.

Thank you.

Michael Molnar you may go ahead and please state your company name.

Goldman Sachs.

Good afternoon, everyone.

Just a couple quick ones.

If we look at the three buckets of end users as you described, retrofit, new homes, and large scale commercial its' called, can you give a rough breakout of your sales for each of these buckets for the systems and component segments?

And secondly, what bucket do you expect to drive U.S.

demand the most over the next one to three years?

Okay.

So the way we would split that, we have a slide that we present most of the places we're at that has customer end segments.

We combine retrofit, new homes, are collective residential market.

Commercial public, call that Microsoft, Macy's, Wal-Mart, Tiffany's, Agilent, et cetera.

And then the power plant business.

And in terms of split, I think Manny will have that in just a second.

In terms of forward-looking where we think -- in the U.S.

we expect those sales to be biased and I'll let Howard answer that and we'll go back to Manny.

In the U.S.

we have concentrated efforts and programs to deliver our products and services in all of the channels, meaning new homes, retrofit homes, small commercial, medium commercial, large commercial, power plant and we're leading in all of those areas.

And so in terms of emphasis, we're going to continue to emphasize all of those channels, if you will, and segments, because they're all growing.

And in terms of balance, I would expect it to be -- lean a little bit towards the commercial and power plant, maybe 60% and 40% residential as we go forward.

Hi, Michael.

For the third quarter itself, these are approximate numbers but close.

Residential accounted for about 30% of our business, commercial about 30% and power plant about 40%.

All right.

Thank you, guys.

Thank you.

Pierre Maccagno, you may go ahead.

Please state your company name.

Needham and congratulations on the quarter.

Can you talk a little bit about the thickness of your cells.

My understanding is that thicknesses that are between 100 and 200 microns, those are very difficult to put into modules.

They break quite easily.

If you go to 100 at that point it becomes flexible and somewhat easier.

So do you have a special method of reducing the breakage of wafers in the range of 100 to 200 microns when you install them in modules?

Yes.

We are currently in production of all of our solar cells are 165 microns.

We believe this is leading edge but we certainly don't know what everyone else is doing.

And we are experimenting on sample lots with 145 microns.

And the ways that we're doing this, a credit to the R&D team and the manufacturing teams, R&D because the architecture of our cell is actually preferential for thin wafers.

We get higher efficiency, the thinner it goes.

We also because it's an all back contact and back architecture cell it's inherently more effective when assembling in full module.

It's all plainer soldering and the back contact on architecture also adds stability to the product.

In terms of manufacturing contribution to this, on thinner wafers is actually the biggest challenge is not necessarily cutting the wafer, it's actually handling it in the Fab.

And our employees on the manufacturing line are all part of SBC teams that implement improvements and one of the focus areas has been wafer handling and they've designed into the equipment ways of handling thin wafers and you're right, putting a very thin wafer into a cord spode for diffusion for example gets very difficult as it gets very thin and our teams have innovative ways to handle wafers, different configurations and different carrying methods that allow us to do that.

In terms of making it into a module, we spoke quite a bit a year ago about investing in automation in terms of module assembly and the -- both the manufacturing and R&D teams have done a really good job of characterizing that automation in terms of thin wafers, they've actually done some element analysis and modeling of characteristics of thin wafers such that the automation is compatible with very thin wafers and we in fact are assembling modules with sampling with 145 micron wafers because of the automation work that's been invested in over the past year.

Thanks and a short follow-up.

What is the interest rate on the debt?

The interest rate on the last convertible was 1 3/4.

Okay.

Thanks.

Our treasurer here is Jay.

3/4 of a point.

.75.

Okay.

Thank you, Pierre.

Thank you.

Jesse Pichel, you may ask your question.

Hi, Piper Jaffray.

Do existing installers with the PPA model have any competitive advantages versus PowerLight and will PowerLight need to own the installation assets on the ground to move towards the IPP model and I have a follow-up.

Okay.

Sure.

So we think as the market moves to PPAs and particularly the commercial market that favors SunPower because of the scale of our company, and also that PowerLight was actually the first to use a PPA structure many years ago, so we have quite a bit of knowledge and experience internally.

And of course, we use the scale advantage we should get very competitive terms.

There are other market dynamics in terms of when you set up a PPA, the financier has very strict requirements and Jesse, I know you're familiar with quite a few of them.

Very strict requirements of who they want to work with and the track record of that company and the fact that we have the -- we ourselves have done the most installations of any company in the market.

That gives us another advantage.

Howard, did you want to add a couple more comments or --

Sure.

Just to say that the barriers to entry actually are relatively low when it comes to the solar business because there are a lot more companies out there.

But the barriers to success are actually quite high.

And just echoing Tom's comments, when you start especially in the PPA model, when you start to peel it back, the owners of the systems and the financiers of the systems are looking not only at the technology which is important, the reliability of the systems, but the credibility and the balance sheet of the companies behind it.

So we're finding increasing success actually as the opportunity expands and each of the opportunities gets bigger, like the Macy's and the Wal-Marts.

We're actually finding that we're enjoying more success.

So you don't need to own the guys on the ground then?

Yes.

Sorry, Jesse, if you could repeat that.

We didn't answer your question completely.

The other part was will SunPower end up having to be the owner.

Not necessarily.

There are structures where that will be the case and of course --

And a question for Manny, if I could.

Why are the balance of system components on the systems sales recognized apart from the overall system?

It seems like you had a margin pickup I guess the first quarter as an integrated company on the same issue, which I assumed was the recognition of the power tilt or the trackers, apart from the overall system.

Could you explain that?

The sale of what you call balance of system is essentially no different than selling a component.

So it's recognizable upon sale, title is transferred, no further obligation.

So that's why it's -- however, it's still negotiated, targeted and seeked by the systems guys that are trying to implement the system.

So we either win it through an EPC where we implement it all the way or if we can't do that for whatever reason, we want to win it as a seller of material as well.

So you're recognizing, you're selling just the material then next quarter which is improving your margin?

Yes.

Would be right.

I think Jesse the important dynamic there is that it can actually be turns business even within a quarter because it's -- the sale cycle is much shorter.

Got you.

Thanks a lot.

Thank you.

Michael Horwitz, you may go ahead and please state your company name.

Hi, everyone.

A little bit of follow-up.

On the PowerLight business how much, going forward in the competitive landscape, how much of an advantage is it for you to be using your own product, rather than somebody else's and how do you view larger EPC companies coming into that market?

So it's an interesting question.

One that was thought about quite a bit before we merged with PowerLight and get to think about it a lot more because the sales force in the projects teams are doing a very capable job internally now of motivating the group in the Philippines to ramp faster.

Because they want more SunPower product.

And we'll start seeing the benefits of an integrated company in terms of product design over the next couple years where because you're vertically integrated you can design the cell and module to be best suited for the end product and we'll be able to cite some of that certainly in 2008.

So you have a better selling proposition and you're going to have a customized product over time.

That obviously gives you an advantage over a less vertically integrated EPC because you also control your own destiny and the folks in the sales and projects teams are very effective at raising priorities internally.

Howard did you want to comment further?

The only thing I would add as a weapon in the field of battle in the landscape that the SunPower provides a terrific weapon in that for example, if we're with a customer and we have SunPower versus a competitor that has a conventional technology, we can field the one megawatt system on the roof and the competitor can field, say a half megawatt or 600 or 700-kilowatt system.

So the magnitude of the savings to the customer and the impact on their carbon footprint, if that's a priority, and the magnitude of financials greatly improves for the customer.

So it's a terrific weapon for us.

A quick follow-up.

In terms of margin in that business over time, is the Nellis project something that we can look at and derive any hints at how you're going to develop that business and the margins around that business going forward, given that they was a rather large project?

Yes, the Nellis project benefited -- the short answer to your question is kind of.

But I think moreso Olivenza and I'll elaborate really quickly.

The Nellis project had the benefit of a new system design and to some degree, SunPower modules as well, albeit that was a subset of the project itself.

There was some rather heroic efforts by our folks in the field because they were the first ones to use that product and they were able to let's say recommend some design improvements that are being implemented and we'll capitalize on at Olivenza and then subsequent project.

So I think Olivenza is going to be a better example of getting to model in the systems business which again is 30% gross margin when we're vertically integrated.

Great.

Thanks, Tom.

[Paul Lema] you may go ahead and please state your company name.

Soleil Securities.

Good afternoon.

Two questions.

On polysilicon first, could you tell us what quarter in 2008 Emsetec will be running at a 3,000-ton rate and then the same question for DC Chemical.

And as a follow-up, are there any new polysilicon projects in China moving forward that you find particularly interesting or intriguing?

Yes.

So let me clarify with Emsetec.

We buy ingots, not polysilicon.

We buy ingots and wafers.

So there is a layer of isolation between us -- isolation, I should say, between us and the actual polysilicon ramp.

And I understand that the group -- Emsetec fields calls from analysts.

So probably best you get their ramp plan from them.

I think it's very aggressive and you should think of it as largely starting now and it's significant, what I said, that the Japanese government has approved their expansion of TCS gas and reactors.

And let me be more precise about that.

They have four reactors online today, I believe, four-ish.

Again, they should clarify that.

And will be ramping to somewhere in they the low teens.

They had to get approval from the Japanese government.

That's done.

So I think my impression is it's certainly south of a year to ramp up to that capacity.

DC Chemical is first of all our Chief Operating Officer is meeting with them next week.

I understand they have an earnings call in about a month.

So again, you can get better information from them.

They've been very aggressive in terms of construction and are finding favorable test results in terms of gas production, largely because they already produce the gas but now they're scaling it rather substantially.

So think of that as being very aggressive and I think happening in 2008 as well.

In terms of China, I think one of the dynamics in silicon that I didn't necessarily call right was that some of the new entrants to polysilicon in China and Russia and other parts of the world have been a little slower to develop than at least I personally thought they would be.

However, there are some Chinese companies coming online with polysilicon production.

As far as I know, mostly below 1,000 metric tons.

There is at least one of those that I understand is considering going public in the next six months.

That company, yes, we find very interesting as a potential partner.

But as they work on their S-1 and if my information is correct then of course you'll become aware of who that is.

Thank you.

Thank you.

Tim Luke, you may go ahead and please state your company name.

It's Tim Luke at Lehman.

Thank you.

This is for Manny.

In thinking about the shape of margin improvement next year, I was wondering if you could give us some feel for how you think that will develop with the different elements of the ramp of the lines and possible improvement in supply of raw material and then whether you have some seasonality associated with PowerLight at the beginning of the year and how we should think about putting those elements together broadly.

Thanks.

Hi, Tim.

We have not really provided guidance for 2008 margins, certainly on a quarterly basis, although one of our stated goals to achieve our gross margin model of 30% potentially in the fourth quarter of '08 or first quarter of '09.

So this is not going to sound too sexy.

You could literally reasonably improve our margin over time and that should work out as well for planning purposes.

Okay.

Reasonably a steady improvement, notwithstanding seasonality?

Correct.

There's so many dynamics that are contributing to puts and takes, as we ramp the lines, they're improving our utilization.

On the other hand, you're ramping more lines so you've got pre-operating expenses.

Silicon can start to come down.

So all those all thrown in, we should be able to ratably improve margins quarter-on-quarter.

And Tim, with the significant growth in the company projected for 2008, just by virtue of the fact that it is growth, we will be back end loaded in the year in terms of scale.

Thank you, guys.

Good luck.

Thank you.

Collin Rusch you may go ahead and please state your company name.

This is Colin Rusch from Broadpoint Capital, formerly First Albany.

My question is about the MPV of ongoing costs.

I'm estimating land cost, O&M, inverter replacement insurance and overhead to be close to 50% or $0.50 per watt on an MPV basis.

Could you give us some insight on how those costs are accounted for in your sales agreement and how you expect them to evolve relative to system price and margin going forward.

Howard will speak to that.

Let me just do a little housekeeping.

We're well aware of the fact that we're into an hour and fifteen minutes.

We have seven people in queue.

So we'll be brief with our answers.

Howard.

Those costs that you mentioned O&M, land, inverter also?

Inverter replacement.

Inverter replacement.

Okay.

So those are all factored into the levelized costs of energy and are embedded in either in the case of the PPA, access per kilowatt hour that's offered to the customer or in the case of an EPC and then a service agreement into those contracts.

So you calculated all those at $0.50 per watt MPV.

I think it really depends on the size and scale of the project.

I would be happy to follow up later on that one.

That sounds great.

The second one is just about the sales cycle going forward.

We've got the construction cycle.

But how long is it taking you from initial contact to closing a deal with a lot of your customers?

The very end market dependent, residential is much faster obviously and that can be certainly within a quarter.

Commercial and power plants can range rather dramatically.

Howard?

That's right.

So for the larger projects, with commercial customers and for the power plants business, it can be any where from three months to two years.

Just depends.

Great.

Thank you guys.

Paul Clegg, go ahead and state your company name.

Paul Clegg from Jefferies.

Just kind of a macro question.

Do you see any other geographic markets or market if you will that's in short term say in 2008 it had the same type of meaningful impact of the type that Spain has had this year on the global market, maybe Italy or South Korea or some market we should keep our eyes on for a big surprise performance in 2008.

We're quite bullish on the Italian market.

We believe that that has probably the best chance for a meaningful impact on the scope of the current business in 2008.

And could I get you to make a couple comments on California and the strength you're seeing there as you step down the PBI into step 4 and you see the incentives come down?

Sure, this is Julie.

We've had a lot of success in California this year.

Obviously the fact that we're in the fourth step of a ten-year program in the second year of the commercial cycle is evidence of the success to date.

There's no doubt that the -- we've slowed down a little bit on the commercial side from the really heated pace that we saw earlier this year which is entirely expected and exactly how the market was designed.

So I think that California is ironing out all of the operational details to make a program really successful.

That's slowed down.

That's really in reservations and you shouldn't see that hit reported numbers, not years necessarily, just installations themselves being connected until well into 2008?

I think it's fair to say that if you look at the CPC data that's up on their website, there's a substantial backlog.

People will be working through that for a little while.

Thank you.

[Mark Mailey] you may ask your question.

Please state your company name.

Hi.

Any chance you could break out the residential segment into new construction and retrofit and maybe give a sense of the trend there?

While I buy time for the finance guys, the way to think of new production homes is attach rate.

You have an increasing attach rate but you have a decreasing sales rate generally in the market that everybody knows very well informed of.

And when you put those two things together you've got a significant growth in new production homes and residential retrofit fits our product portfolio excellently so we've got great growth in that as well.

As Manny mentioned it's 30% of sales and in terms of a breakout.

Of the 30% total residential, the new homes or new homes market, portion of that is about 3 to 5% of the 30.

Okay.

And then on the efficiency side of things you talked about going up to 40-megawatt name plate.

How much of that is driven by improvements in cell efficiency?

The question is if the module efficiency going up, how much is driven --

Of the line capacity going from 33 to 40-megawatts per line, how much of that is driven by just improvements in cell efficiency?

Sorry.

Got it.

It's a combination of improved UPH or units per hour or throughput of the equipment and efficiency.

I don't have the precise math.

I can give you an educated guess and take that as a homework.

I think we'll have to do that.

And it's probably a little less than 50% the conversion efficiency and a little more than 50% is the throughput of the equipment.

Thank you.

Thank you.

Our next question comes from Brian Gamble and please state your company name.

It's Simmons & Company.

Guys, wondering with the success of the 145 micron thick wafers on the test lots, what's your estimated time for being able to move those to some sort of commercial scale?

I think we would go into production of 145 micron assuming our sample lots go well, we'll see some production certainly in 2008, probably in the first half.

And then you talked about Germany and about Spain on the subsidy side.

Could you talk a little bit about the U.S.

energy bill and what you're hearing most recently as far as how that's going.

Julie will speak to that of course very -- do so in great depth.

Please do try to influence that.

Our industry group is SEA and you can find them easily on the web and you can make your voice heard and you can also contribute to them.

So please help us put quote unquote more wood behind that arrow.

Julie?

Hey, Brian.

I actually just got back this morning from DC where I spent yesterday on the hill.

You know, this is a very exciting week on the hill.

A lot going on.

The energy bill is very much in play.

I think the speaker has put a substantial amount of her personal time and effort into ensuring that we will have an energy bill this year.

It doesn't guarantee that we will.

But I have to say, all of the activity on the hill suggests that people are trying really hard to find compromise between the Senate and House bills and take something that is veto proof.

And Julie, from a compromising standpoint, what is that going to mean for solar.

Are we going to have to take a hit to get something through.

Or is there something agreeable on both sides that is being pointed to right now?

Yes, I have to say I don't think solar is on anybody's plate as part of the chess game.

Would that that would be the case.

But it's a real big question is whether there's a tax title certificate as part of the energy bill or whether they'd split it up into pieces I think.

So as long as there's a tax title, I think we've got a good shot of having solar get a, something that looks pretty similar to either the House the Senate or the best case a complement of the two.

Thank you very much.

Okay.

We have four more and we'll stop at the fourth one and we'll do these quickly.

Thank you.

[Kaelin Wilford] you may go ahead.

Please state your company name.

Thanks.

[Chellis Management].

Thanks for taking my question.

The -- just a quick question I think for Manny.

In light of the capacity add to twelve lines through '09, do I have that correct, it was twelve you said?

Twelve lines will be the new capacity of the Fab 2, which was originally estimated as ten.

It's now 12 lines.

Okay.

Is there any update recently a, kind of a guided CapEx per watt number that would carry us between now and into the end of '09?

The answer to your question is no.

We have talked broadly about a 5% plus reduction for each new line.

We'll take that I think as our third homework assignment.

Because it's something we're comfortable communicating.

We just need to make sure we've got it right.

I just think it would be -- it had kind of fallen out of the pile but I know that maybe in my notes I have a specific number.

Can you comment on the trend, whether it's flat or down or

It's definitely down and it's however got a counterweight by a lot of equipment in Euros and of course the dollar has depreciated against the Euro so it's offset, but we've had better design and we of course buy more quantity and so -- and of course, we've done some things in manufacturing to utilize equipment more effectively.

So the combination of those things ends up with a definite decline, largely in the range of what I said.

Okay.

Great.

Thanks a lot.

Thank you.

Mark Roberts, you may go ahead.

Please state your company name.

Wachovia Securities.

Good afternoon.

First question I have is do you see these residential solar developments as large revenue generators for the future?

And if you do, do you have any percentages on revenue generation?

That would be great.

This is Howard.

The answer is yes and yes.

The new home part of our business actually is the fastest growing segment of our company.

It's a small part, as Manny noted right now, but it's the fastest growing and we do believe that it will be a significant part.

I just want to make one comment is that we're working with some of the largest home builders in the U.S.

like Lennar, Centex, DR Horton, and they're making solar standard.

And at first, they were contemplating one community at a time, 50 or 100 homes.

Now they're looking at multiple communities.

And even at the -- in higher -- in the state of California making solar standard.

And the other question I had, the Olivenza project -- someone said the revenue from that project would be as high as $125 million.

How much do you think this project can add to revenue and how are you seeing SunPower fit into the competitive landscape?

So really quickly, Howard can maybe take a shot at revenue.

When we talk about 2008, we consider the portfolio projects.

We of course know what's in our pipeline.

So Olivenza certainly was part of our guidance.

I don't know that we'll go further into that.

I think when you add it up, you get to install plus backlog of over 100 megawatts, I think it's a 100 and something megawatts.

That's a substantial part of the Spain market.

Our competitive position is excellent.

And it's because of the great EPC capability that we have, the great system design and of course the efficiency to manage with our cell modules and it fits well in that market and we'd note that that's also true in Korea.

Korea is even more land constrained and has a higher value on efficiency.

In terms of the revenue impact of that specific project, I don't think we're going to guide further on that.

Thank you.

Thank you.

Satya Kumar, you may go ahead.

Please state your company name.

Thanks for taking my follow-up.

Let me make it very quick.

Tom did you mention that pricing would be flat in the first half of next year.

I just wanted to make sure I caught that right.

I mentioned that I expected it to be steady going into next year.

So that's -- we're not giving guidance beyond the first quarter.

We can think of it in Q4 of '07 and entering into 2008.

(inaudible) I guess, right.

That's my guess.

I didn't quite understand.

It's steady.

He's just trying to calibrate.

It's flat.

(inaudible) on 5% would you call that steady or --

Think flat.

Think flat.

Okay.

Thanks.

Thank you and our last question comes from Michael Carboy.

You may go ahead.

Please state your company name.

Signal Hill.

Thanks very much for the follow-up here.

Howard, can you give us a quick rule of thumb for PowerLight system install costs or I should say price per watt?

It depends on geographies and some system size so the only thing I can give you a range of somewhere between, I would say, $4.55 per watt on the low end up to $8.00 per watt on the high end.

Okay, thank you.

That's helpful.

Sure.

Thank you.

At this time I'm showing not further questions.

Mr.

Werner I will now turn the call over to you for any closing comments.

Sure.

A twenty second wrap-up.

I thank you very much for you time.

There are lots of moving parts in our business.

We believe we're managing them well.

We're very positive about our future.

We appreciate your time and attention.

Thank you.

Thank you.

This does conclude today's conference call.

Have a nice day.