第一太陽能 (FSLR) 2007 Q1 法說會逐字稿

Welcome to the First Solar 2007 first quarter conference call. The date of this call is May 3, 2007. This call is the property of First Solar and any recording and reproduction or transmission of this conference call without the express prior written consent of First Solar is strictly prohibited. This call is being recorded. You may listen to a webcast replay of this call by going to the Investor section of First Solar's website.

I will now turn the call over to Erica Mannion, Investor Relations for First Solar.

Good afternoon everyone and thank you for joining us for First Solar's first quarter 2007 earnings conference call. Shortly after market close today the company issued a press release announcing its first quarter 2007 financial results. If you did not receive a copy of the press release you can obtain one from the Investor section of First Solar's website at www.firstsolar.com. You may listen to an audio replay of this conference call by dialing 888-203-1112 and entering reservation number 9848426. Also webcast replay will be available in approximately two hours on the Investor section of the company's website.

With me today are Mike Ahearn, Chief Executive Officer; Bruce Sohn, President; and Jens Meyerhoff, Chief Financial Officer. Mike will begin with an overview of the company's achievements and progress during the first quarter of 2007. Jens Meyerhoff will provide you with details of the first quarter 2007 financial results and an update to the financial guidance for 2007.

All financial numbers reported and discussed on today's call will be based on generally accepted accounting principles with the exception of the company's EPS presentation, which gives effect to the company's initial public offering, the conversion of its convertible debt during the second quarter of 2006, and share sales during the first quarter of 2006 to prior periods.

Following formal comments we will open up the call for questions. The company has allocated approximately one hour for today's call. During the Q&A period, as a courtesy to those individuals seeking to ask questions, we ask the participants limit themselves to one question and one follow up question. This information will remain available until May 8 at midnight eastern daylight savings time.

Now I would like to make a brief statement regarding forward looking remarks that you may hear on today's call. During the course of this call the company will make projections and other comments that are forward looking statements within the meaning of the federal securities laws. These statements are based on current information and expectations that are inherently subject to change and involve a number of risks and uncertainties. We caution you that actual events or results may differ materially from those in an forward looking statements due to various factors including, but not limited, to the availability of government subsidies, the demand for solar modules, the cost of capital available to our customers and end product users, the field performance and reliability of our products, and the company's ability to successfully replicate production at its German and Malaysian plants, and the company's relationships with customers and key suppliers.

Additional information concerning factors that could cause actual results or events to differ materially from those in any forward looking statement is contained in the company's form 10-K filed with the Securities and Exchange Commission and in the safe harbor language in the press release that was sent out today. The company assumes no obligation to update any statement made during today's call, to revise any forward looking statements, or to update the reasons actual results could differ materially from those anticipated in forward looking statements.

It is now my pleasure to introduce Mike Ahearn, CEO of First Solar. Mike?

Thank you Erica and thank you for participating in today's earnings call which concludes the first quarter of 2007. During the first quarter we increased revenues 27% sequentially over Q4 2006 to $66.9 million, reflecting continued strong demand for our modules. We increased production 5.4% sequentially over Q4 as a result of continuous improvements in module production throughput and conversion efficiency at our Ohio plant. And we approached the net income of $5 million which included production setup costs at $8.5 million and the continued infrastructure buildout required to scale the business to accommodate our rapid growth class.

Underlying these results we continue to make solid progress toward our midterm goal of reducing solar electricity prices to levels that compete on a non-subsidized basis with conventional electricity. In doing our February call I outlined the following initiatives that we're pursuing in 2007 in furtherance of this goal: rapid capacity expansion through our plant's replication process; market expansion both geographically and by market segment; and cost reduction enabled by higher throughput, higher module conversion efficiency, and low manufacturing cost location.

I'd like to briefly update you on these efforts starting with production and capacity expansion. And let me start by explaining at least some change to our nameplate ratings. We've historically referred to our production lines as having a 25 megawatt nameplate capacity. We define nameplate capacity as the minimum expected annual production output per line, assuming it's operating at full capacity. Based on production throughput increases at our Ohio plant and our success to date in ramping our German plant, we have increased our nameplate rating from 25 megawatts to 30 megawatts per line. This brings our total expected annual nameplate capacity to 330 megawatts by the fourth quarter 2008, based on our currently announced expansion plan. Over time, production volumes may continue to exceed the nameplate rating due to continuous improvement in module production, throughput, and sellable watts per module. We've adopted a practice of increasing the nameplate rating in 5 megawatt increments once we're confident that we've achieved the new minimum annual production capability threshold across all of our production lines.

Turning to production, then, during the first quarter all of our production came from our Ohio plant where we operate three production lines with a 90 megawatt nameplate rating. Total production at our Ohio plant was 26.1 megawatts in the first quarter, compared to 24.7 megawatts in the fourth quarter of 2006, or an increase of 5.4%. The increased production volumes during the first quarter resulted from an increase in sellable watts per module driven primarily by higher conversion efficiencies and increases in module production throughput rates.

During the first quarter we also continued to benefit from the additional capacity provided by our R&D [coder]. The improvements in both production throughput and conversion efficiency during the first quarter strengthened our confidence in meeting our 2007 production target for Ohio of at least 100 megawatts.

We start with our German plants progressing ahead of schedule. We started producing modules in April which we intend to ship later this quarter upon completion of our long term reliability qualification and permitting processes. We anticipate being in full production by the third quarter of 2007, and based on these results we've increased our estimated 2007 shippable production volume from the German plant to a range of 55 to 60 megawatts, as compared to the previous 35 to 50 megawatts that we discussed with you in February. We believe our success to date with the German plant validates our copy-smart program and the capability of our replication themes, and provides added confidence in our ability to efficiently replicate plants while achieving satellite plant performance equivalent to base plant performance. This capability is of course essential to our rapid growth and cost reduction outputs.

In January of 2007 we announced an agreement with the Malaysia government to build a four line, 120 megawatt nameplate facility in Malaysia, which is essentially a copy of our German plant. We broke ground at this site in April and expect to substantially complete construction by year end and reach full capacity by Q4 of 2008 in line with our previous communication.

Turning to the market, our product demand remained robust in the first quarter of 2007. We sold 28.9 megawatts of modules at an average sales price of $2.32 per watt, resulting in revenue of $66.9 million. Our sales in Q1 benefited from higher production volumes but also the inventory build we reported last quarter. Based on current estimated projections we expect 2007 German installations to comprise approximately 75% of the total sellable watts, Spain installations to comprise approximately 10% of total sellable watts, Italy installations to comprise approximately 5% of total sellable watts, and North American installations to comprise approximately 5% of total sellable watts. This leaves an additional 5% of total sellable watts currently unallocated in 2007. We currently have demand for these modules in excess of supply and intend to allocate them over the course of the year as we continue to evolve our market strategy.

I'd like to briefly summarize market developments over the quarter by geography now, starting with Germany. The market in Germany continued to be very strong for First Solar in Q1, highlighted by the 40 megawatt PV project using First Solar modules that is currently being developed in the Northern German town of Brandeis. To our knowledge this project is not only the largest but at an average cost of 325 euros per watt, the lowest cost large scale PV system ever deployed.

Our customers in Germany continue to execute well throughout Q1 and we are in the process of exploring additional opportunities for developing multi year pipelines of demand in new market segments in Germany beyond 2007. Despite current political discussions from some in the EEG, we expect that program to continue beyond 2007 in a form that will enable First Solar to execute its business plan. As a major German employer of the solar industry we're taking an active role, along with the rest of the industry in German, in supporting the program throughout these discussions.

In Spain we recently began doing business with a major Spanish company in what we hope will be the first phase of a longer term relationship. In addition, we're continuing discussions with several essential Spanish customers and continuing to monitor efforts to amend the royal decree to remove the 400 megawatt cap and the 100 kilowatt system size limitation. We recently opened a small office in Madrid to facilitate these efforts.

In Italy, a recent amendment to the feed-in law removing the annual limit, increasing the program cap to 1.2 gigawatts, and addressing a number of administrative issues, appear to have stimulated the marketplace. We've made initial product shipments into Italy and we are currently engaged in exploratory discussions with several potential new Italian customers.

In the U.S. we're viewing the solar market in two primary segments. First the large multi-megawatt ground-mounted project segment primarily serving utilities, and second the distributed segments serving a variety of commercial and industrial electricity consumers with rooftop installations. We are currently in discussions with several companies in order to supply modules into these segments and we expect to put multiple agreements in place over the balance of the year.

Our cost reduction efforts are event and program-driven and do not occur in a linear fashion. Nevertheless, as is evident from our previous remarks, we are progressing well on our underlying cost drivers, conversion efficiency, throughput increases, and end capacity expansion.

And finally, we are continuing to strengthen our team in many areas to support our continued growth while adhering to our high execution standards. Most visible, but by no means exclusive example of this, is the addition of Bruce Sohn our new president. Bruce joined First Solar in March 2007 after serving on our board for four years and brings a wealth of operations experience from his former tenure at Intel, as well as a familiarity with our technology, business, and people. In his role Bruce will lead First Solar's daily business execution and the infrastructure build out required for our future growth. Chip Hambro decided to transition his daily role as chief operating officer to a vice president role that will enable him to contribute to our key technology related efforts.

In summary, we're continuing to make good progress in our key initiatives and we remain on track with our mid term goal of reducing solar electricity prices to levels that compete on a non-subsidized basis with conventional electricity over the next few years.

I would now like to turn the discussion over to Jens Meyerhoff, our CFO, who will discuss our first quarter financial results and an update for our guidance for 2007.

Thank you Mike and good afternoon. The first quarter of 2007 concluded with continued strong growth. Revenues for the first quarter were $66.9 million, in increase of $14.3 million, or 27% over the first quarter of 2007; and an increase of $53.3 million compared to the same period of last year.

Revenues benefited by $6.6 million as a result of the finished [inaudible] inventory built during the first quarter of 2006 which turned during the first quarter. Germany accounted for 98.4% of our revenues with no single customer accounting for more than 22% during the quarter.

Gross margins for the first quarter were 44.9%, down from 48.6% in the fourth quarter of 2006 and up from 24% in the same period last year. Gross margins were primarily impacted by annual contractual price declines of 6.5% during the quarter. Gross margins benefited from higher margin throughput and increased sellable watts per module driven by conversion efficiency.

During the first quarter we remained focused on the build out of corporate infrastructure in order to establish our global manufacturing presence, which drove a sequential increase in fixed costs ahead of the production ramp at our German plant.

In addition, stock-based compensation expenses increased due to our November IPO brands which became fully effective this quarter. As a result, our manufacturing cost per watt increased from $1.25 from the fourth quarter to $1.29 primarily due to an additional $0.03 of stock-based compensation. We see this increase in our cost per watt as a near-term timing difference as our expanding increased leads the production ramp expected during the second half of this year, and our continuous improvement programs toward throughput and conversion efficiency remain ahead of plan. However, gross margins and cost per watt will be impacted unfavorably during the second quarter due to the ramp at our German plant. As Mike mentioned, we started production at our Frankfort order plant last month and are therefore incurring ramp costs ahead of our previously communicated schedule.

Operating expenses, excluding plant startup costs, were $16.7 million in the first quarter of 2007, up from $12.6 million in the first quarter of 2006. The increase was primarily a result of increased personnel expenses, including $1.2 million of stock based compensation. Infrastructure build-out and compliance [inaudible] as we prepare for first year of Sarbanes-Oxley compliance. Operating expenses, excluding plant start up costs, were 25% of sales in the first quarter, slightly up from 24% of the first quarter of 2006. Operating expenses include a $4 million of stock based compensation during the first quarter, up from $2.8 million in the fourth quarter due to additional stock option grants made at the first quarter in conjunction with our IPO and recent hires. Excluding stock based compensation our operating expenses stayed essentially flat at 9% of sales during the first quarter, compared to the prior quarter.

Plant startup costs increased by $4.5 million during the first quarter to $8.5 million due to the continued progress made of our German plant and startup of our Malaysia expansion. Operating income for the first quarter was 7%, or $4.8 million, a decrease of $4.2 million over the fourth quarter of 2006 and an increase of $11.5 million over the first quarter of 2006. Operating income for the first quarter included $8.5 million of plant startup costs and $5.8 million of stock-based compensation expenses and reached 28.1% of sales excluding such expenses.

Interest income for the quarter was $4.1 million reflecting an average yield of 5%. The tax rate for the first quarter was 39.5%, slightly up from 38.5% in the fourth quarter of 2006. Net income for the first quarter of 2007 was $5 million or $0.07 per share on a fully diluted basis, compared to $8 million or $0.11 per share on a fully diluted pro forma basis for the fourth quarter of 2006, and a net loss of $5.9 million for the first quarter of 2006.

Pro forma earnings per share has been adjusted for the relevant periods, assuming the company's equity offerings during 2006 that occurred at the beginning of fiscal 2006. On a GAAP fully diluted basis, net income was $0.12 per share for the fourth quarter of 2006. Cash and short term investments increased to $325.3 million during the first quarter. Cash flow from operations during the first quarter of 2007 was $38.9 million, compared to $13.3 million for the fourth quarter of 2006 and $11.4 million cash usage during the first quarter of 2006.

While our net income declined sequentially, the cash generation of our income statement increased slightly as the increase of cost was primarily driven by depreciation and amortization, including stock-based compensation. In addition, our [inaudible] sale outstanding declined to 11 days during the first quarter of 2007, compared to 48 days in the fourth quarter of 2006 as our 10-day payment plans became effective. This provided favorable cash flow of $20 million during the first quarter.

Net inventory declined by $1.5 million during the first quarter. That's a result of finished goods shipment from fourth quarter production, partially offset by an increase in raw materials as we readied for the production ramps in Germany.

We spent $57 million in capital expenditures during the first quarter, against depreciation and amortization of $5.1 million. Cash flow from financing activities were $8.7 million due to loan and grant receipts from our German financing.

This brings me to our guidance for the fiscal year of 2007. We explained to you in our last earnings call that our financial performance during fiscal 2007 will significantly depend on the successful and timely ramp of our German plant. We are updating our guidance for the fiscal year 2007 based on the successes we have had at our German plant during the first quarter. We now expect approximately 100 to 105 megawatts of sellable production volume from our Ohio plant and between 55 and 65 megawatts from our German plant. This ramp in production volume translates into expected revenues of $370 to $400 million for 2007. As previously communicated, we expect 60% to 65% of our annual revenues to be in the second half of 2007 due to the ramp of the German plant.

In the second quarter we will continue to experience significant increases in spending due to infrastructure build out and additional fixed manufacturing costs leading the revenue ramp Germany. This will cause a temporary steep decline in our gross margins as all German startup costs are now classified as cost of sales and could still cause us to incur a loss in the second quarter. We expect significant gross margin recovery during the third quarter and target steady state gross margins by the fourth quarter of 2007 as we achieve full production volume in Germany. At the revised revenue range described earlier, our operating margin should range between 12% and 16% for 2007.

Operating income for 2007 is expected to include stock-based compensation of $26 million, up from our earlier projections due to changes in our forfeiture assumptions and stock price appreciation. In addition, we decided to accelerate vesting of some of our non-executive employees in an effort to harmonize equity compensation between our long-term employees and more recent hires.

With these changes we expect a corresponding decline of stock based compensation costs in 2008 to be approximately 50% to 60% of the 2007 expense level. Production startup costs are reduced to approximately $18 to $23 million. Our 2007 tax rate outlook is unchanged at slightly above 40%. Our estimated fully diluted share costs remain 76 to 77 million shares at the end of 2007. We expect capital expenditures of $210 to $240 million during 2007, in line with our previous guidance.

With that, I turn the call back to the operator. Operator?

Thank you very much. [operator instructions] Our first question of the say will come from Jesse Pichel from Piper Jaffray.

Great, congratulations. Hey, this doesn't count as a question, but you got cut off when you gave your revenue guidance, and I got several e-mails. Can you repeat your revenue guidance for the year?

Yeah, certainly. The revised updated revenue guidance, $370 to $400 million.

$370 million. Okay, great. Okay, my first question is do you see any competitive solar products over the next couple of years that will come close to your cost per watt or your efficiency levels?

Well, I guess you're referring to ThinFilm, Jesse. We don't envision any in volume approaching our cost per watt levels in the next couple of years, no.

And as a follow-up, you're producing above your original nameplate. Does that mean you'll add another customer or contract, or is there ample demand from your existing customers?

No, we're actively in the process of adding additional customers, and we have volume that will allow that to occur on a seed basis in 2007, and then progressively expanding basis 2008 forward; so I do anticipate that we'll have several new customers this year.

That's great. I'll go back into the queue.

Our next question will come from Sanjay Shrestha with Sell Side Equity Research.

Good afternoon, guys. Again, congratulations here on a great quarter. Just a couple of quick questions. First off, same question actually on the last call as well; given the traction that you guys are having in the ramp-up of Germany and eventually with the market demand being the way it is, what are some of the incremental things that need to happen before you guys say that Malaysia is now going to be a 100 megawatt expansion, but a 300, 200, or 400 megawatt expansion?

Well, we have a process that's fairly disciplined, systematic around how we think about plant expansion, and I guess the cornerstone of that is we want to make sure we can repay investment. These are capital-intensive projects, as you know. There's a variety of things that we check off. Part of it is, do we believe there's adequate demand to repay plant investment on a multi-year basis, and demand that fits strategically with what we're trying to do with the company; and then, do we have confidence in our replication capabilities, we check the capacity throughout our chain; and then there's a check on internal operations. So there's a variety of things that we look at, and we have windows of time where we consider expansion, and that's basically the process we continue to follow.

Okay, that's fair. And in this particular quarter, can you guys give a breakdown of how much of the shipment was to the ground-mounted versus the rooftop installation; and given the continuous rise in new efficiency, could you see more of a sale start going to the rooftops for strategic application, and could that then start to have a favorable implication for ASPs as well, versus where it is right now?

That should really count as more than one question, but that's okay. Well I don't have the breakouts for quarter, Sanjay, but for the first quarter, generally we run, have been running 60% ground-mounted, 40% rooftop, and I think you can have some aberrations on a quarter-by-quarter basis, but I feel good with that number for 2007.

Of course.

I do think that we will be moving more, as we look at these new geographic markets, into rooftop in certain situations. The implication, I think, on ASP is driven more by the market segment and the end user segments we're approaching then how it's mounted, and whether it be ground or rooftop. I wouldn't infer just from that mix any particular change in ASP.

Okay, that's great. Once again, congratulations, guys.

Thank you.

Our next question will come from David Edwards with ThinkEquity.

Hi, there. Wanted to know -- you guys obviously continued outperforming both on the efficiency side; can you talk about efficiency levels for the quarter, and where you see them trending before the end of the year?

Well, as you know, Dave, the efficiency roadmap doesn't really lend itself to sort of a weekly or a monthly, maybe even a quarterly, type of precision. We have a set of programs that we continue to execute to continuously improve efficiency, and we've got a roadmap in mind that has us between 11% and 12% over that 2010-2012 time period, which we continue to feel very good about. We're probably feeling better today than ever. But we really don't have it -- we don't have it mapped on a quarterly basis over the course.

Okay, and one more quick question. It looks like ASPs were slightly above what we were expecting. Are you seeing advantages in pricing in the volume that you're selling above your pre-contracted volumes?

I think if you do the math on the price decline, and so obviously that calculates to less than the 6.5% we had in the contracts, so think that impact of pricing right are to some degree, customer mix. We had favorable customer mix in Q4; in Q1, foreign exchange aided the price. Also, our product is denominated in euros.

Okay, great. Thanks a lot.

Our next question will come from Ronan Wolfsdorf with Cowen.

Good afternoon. Congratulations on a strong quarter. I'm just trying to get a sense of how you guys model risk and opportunity in the German market, given the evaluation of the EEG.

Yeah, I think the EEG, as you know, is under discussion politically, and there's a possibility that there will be some changes in the EEG, so the way we think about that is to try to understand the range of potential changes based on what's been discussed and what we think could raised, and try to understand if our business model and our pricing is robust against the potential range of outcomes; and to think about mitigance if they're necessary. Given -- if you're at the extreme end of the range of something very negative happens from our perspective, are there actions we could take to continue business uninterrupted. So that's generally how we approach it. I think more specifically, we don't see in the discussions to date, any changes that have been proposed of that would impact our business in a meaningful way.

Great. Thanks very much.

Our next question will come from Michael Carboy from Signal Hill.

Good afternoon, ladies and gentlemen. Think I'll go back to two questions ago, and get you and try and talk specifically about efficiency -- conversion efficiency this quarter. Could you do that, please?

Sure. Well, the efficiencies are up this quarter relative to last quarter, and we have a set of programs that we've been executing against, really since 2001, some 10 to 15 programs that roll up in terms of efficiency potential to around 15%. These are all programs that have been demonstrated at least at a lab level, at a cell level, with processes that we consider manufacturable in volume and capable of being integrated into our manufacturing lines as they're currently being built. And if you saw our histogram on this, you'd see continued conversion efficiency improvements from 2001 to date at a rate that on average equates to about .5% a year. So we need to be between 11% and 12% over the 2010-2012 timeframe to achieve our specific pricing capability goals that we've set for the company. We're viewing 15% as really as buffer, but over time, something that we'll eventually get.

These efficiency improvements are certainly more event-driven than time-driven, and as these programs are implemented from R&D into our R&D line, and eventually into low-volume production, and then gradually phased in, you see the impact in the total number. But that's coming as the result of a chain of programs that are constantly moving through the pipeline. So I think what we saw this quarter is some efficiency improvement, and we felt it from that process.

All right. And as you had mentioned, the upside this quarter was driven by both throughput and efficiency gains. I was wondering if you could split those respective benefits into two different buckets for me. Of the total upside that you had, how much do you think -- what percentage of that upside, do you think, was driven by volumetric throughput versus convergent efficiency through [inaudible]?

I think if you look at the group of gains that we're targeting as part of our continuous improvement effort, we've guided to modeling around 3% annual improvement, and the results we saw in Q1 were definitely in line with that expectation level.

Okay. Thank you.

[operator instructions] Our next question will come from Scott Davis with Morgan Stanley.

Good afternoon, guys.

Hey, Scott.

Looks like you continue to ramp up quite nicely here. One of the things -- and there was so much information given on the call. I think I have to go back and look at the transcript to see what's said, but you did allude to some additional indications of interest from U.S. customers, I believe. Can you talk a little bit about that, and how real is that? What kind of timeframe are you looking at? Just any additional color there?

Sure. We're talking, I guess you could say, two sets of customers or opportunities, Scott, that break by market segments. So on one hand, you have the central generation or utility-scale projects, and you might recall -- our thought there has always been to be more vertical than simply selling modules, which could range from our own vertically integrated project business to a joint venture or partnering of sorts; but an activity that has us involved beyond simply selling modules for the utility segment. We have discussions underway there with more than one potential partner, if you will.

On the distribute side, where you're mounting generation systems onto buildings, for us that would be commercial/industrial buildings; our inclination there is to work with some integrators that are currently serving that market. California and New Jersey are the primary markets in the U.S. That's because rebates are really needed to buy these systems down for an end user point of view. So we have had discussions -- we're engaged in discussions with several systems integrators to allocate modules to them. Those are real; I think the question for us is, given a limited amount of volume, how do we allocate that? We're trying to build multi-year relationships, obviously; and it's "pick the strongest partner" from our standpoint.

I think we're pretty confident from between those two sets of opportunities, we'll emerge with some pretty good relationships over the course of the year.

Okay. And -- this is a little bit of a strange question, but when you talk about 40% of your business rooftop, 60% ground-mounted, are you seeing any trend towards being able to really serve a small rooftop-type market which might expand your eventual market into more of a residential-type opportunity set; or there some sort of minimum square footage your number of modules can -- current efficiency rate where you can really make things work?

No, I think from a convergency/efficiency point of view, just to give you a data point on that, for a 3-kilowatt system using our 9% modules, you'd need about 500 square feet of roof space. If you compare that to a 13% polycrystalline silicon module, it's about 350 square feet; so more roof space is needed for our systems, but not to the point where space or size typically becomes a constraint. So, we don't feel like we're constrained by conversion efficiency from entering smaller system markets. I think the issue for us is more strategic.

We have a very strong value proposition in some of these larger system size segments where the end users are more cost-sensitive, and so our competitive cost advantage plays to our strength; and being in the larger volume, larger size-typed applications, we can scale more rapidly. A big -- our whole business is modeled or oriented around driving costs down to these competitive levels as rapidly as possible. So, we're trying to balance all the strategic aims as we pick the segments; and I'd say at this point, more than the efficiency delta is what's orienting us in the larger sized applications.

Makes sense. And lastly -- I know you're supposed to ask one question; I'm just thinking of stuff here. When your price increase -- your price decreases, contractual price increases hit, which I'm guessing is something like January 1st and beginning of every year, should we be thinking of gross margins increase through the year as you are able to really -- I know there's a lot of moving parts here with new plants coming up, but gross margins increase -- they'll be at the low point in 1Q, and then they'll increase gradually through the years, you're kind of able to catch up to that price decrease in productivity?

Yeah, I think, Scott, if you look at the trends, number one, it is correct. The prices adjust contractually on January 1st, and that hits the gross margin straight. Then, if you walk through a year, we believe we have enough ammunition in our cost production roadmap to drive the corresponding offset, and so that would better gross margins throughout the year. I think if you look at 2007, obviously, the second quarter, there will be a little bit of [inaudible] as Germany is ramping, and we're dealing with cost of capacity from utilization of that ramp. But generally, you should kind of see a profile of steady-state, where the first quarter is likely to have the lowest margin absent of the ramp penalty, and then growing towards the year.

Scott Davis. Okay. Makes sense; just wanted to confirm it. Thanks, guys.

Thanks, Scott.

Jesse Pichel has a question.

Yes. Startup costs, I believe you said, were reduced to $18 million to $23 million. Can you talk us through why that is, and what you look for in 2008? And I have a follow-up?

I think if you go through the startup cost piece, there's two line items, really. We have planned startup costs and our operating expenses; and then part of the reduction I talked to about is actually those costs moving up into cost of goods sold, and as we commend to production -- so if you look at the total bring-up of the German plant right now, while that is marginally less expensive, there's a decline in the production startup costs in 2007 is primarily driven by those costs; the fixed costs of the plant moving into cost of goods sold early, because we're ahead of schedule with the ramp.

For 2008 planned startup costs, I probably don't want to -- I really don't have a good number for you on that right now that I would like to share. I think fundamentally it costs us usually around $20 million to bring up a plant, $20 million to $25 million in the plant startup costs, subject to the timing; then, obviously, subject to how many plants you're bringing up.

Okay. And my second question is for Mike. Mike, could you just talk around how you think First Solar would hold up there if the traditional solar panel market was in an over-capacity situation in 2010? Because it looks to me that you're already grid-competitive in Italy, which means that you probably are more immune there to an over-capacity situation? If you could just talk through that; those are kind of the issues that everyone's asking.

Yeah. I think there's maybe two ways to look it. Maybe there's three ways. From a fundamental cost perspective, I think our cost advantage is going to continue, notwithstanding the oversupply. If you think about any kind of a reasonable time period and sustainability, I think our fundamentals are still superior. In terms of a shorter term fluctuations or aberrations with supply/demand imbalance, a lot of that for us we believe we've mitigated and are mitigating through these longer-term contracts. And if we don't have long-term contracts, we will have something that -- a proxy for that in terms of relationships, what we're driving, visible demand pipelines over multiples of years.

So by not being predominantly in the spot market, I think our susceptibility to these kinds of things is greatly reduced. I also agree with your point that I think our pricing -- the pricing levels were already at or quite a bit below where crystalline silicon is, even if feed stock were to become available. I mean, the first question is whether anybody is going to ever have to normalize gross margins to a level that drives sustainable business models, and assuming that doesn't occur, it's taken directly in price, and I think we have these other cost advantages and long-term contracts to fall back on.

[operator instructions] And that does conclude our question and answer session. And that does conclude today's conference. Thank you very much for joining us today.

Thank you very much.