While perusing the weekly stock-market wrap-up in the Los Angeles Times' business section on Saturday, I was struck by the prodigious increases (can you say, "bubble"?) in year-to-date share-price increases for several solar industry companies---despite the recent corrections/downturn in the markets: Ascent Solar, up 543%; SunPower, up 246%; JA Solar, up 206%; Yingli, up 203%; Trina Solar, up 199%; Evergreen, up 86%.
The biggest uptick belonged to thin-film PV module pioneer First Solar, whose shares have skyrocketed 593% so far this year and traded at just under $207 per share at Friday, November 9's market close.
First Solar, which uses a cadmium-telluride (CdTe) thin-film process manufactured on highly automated production lines, has been in a dealmaking and capacity expansion frenzy of late. Its financials reflect an escalating level of success. The company announced last week that its 3Q07 revenues hit $159 million, more than double 2Q07's $77.2 million and almost quadruple 3Q06's $40.8 million. Income hit $46 million for the most recent quarter, a new record.
First Solar's long-term contracts for a variety of ground- and building mounted solar-power systems amount to more than $6 billion through 2012, according to the company. Factory building efforts will give the company more than 900 MW of ramped production capacity by the end of 2009, when the last of four planned Malaysian fabs come up.
A bright future for First Solar? (Photo courtesy of co.)
The company held a conference call last week (featuring CEO/chair Mike Ahearn, president Bruce Sohn, and CFO Jens Meyerhoff), and Meyerhoff also spoke at the Pacific Growth Equities' clean-tech conference (both of which are available via Webcasts). Here are some highlights of those presentations:
Ahearn said "demand remains robust, in excess of supply," so it's not surprising that First Solar's module production more than doubled from quarter to quarter. The seven operational lines, hitting an average output of 39.6 MW each, are running at full capacity, including the new Frankfurt/Oder facility, which ramped considerably faster than planned because of the company's "Copy Smart" process/factory replication strategy. Throughput and conversion efficiencies have been improved (production conversion numbers have reached 10.5%). Tools will be delivered during 4Q07 to the first Malaysian factory, with the other three factories there to follow in quick order. A total of 16 lines will be producing modules when all the Malaysian capability is up and running in late 2009.
Meyerhoff noted that the gross margin for Q307 was 51%, well up from the previous quarter's 36%, due in large part to the rapid ramp of the new German fab. Manufacturing costs are at an extremely competitive $1.19 per watt (which includes four cents of stock-compensation expenses). Guidance provided for FY07 includes $480 million to $485 million in revenues, total manufactured output about 200 MW, and capital expenditures around $280 million. For FY08, Meyerhoff said they expect to hit $760 million to $800 million in revenues, output in the 370-390 MW range, and capex in the neighborhood of $450 million. The manufactured cost per watt should approach a buck, and the price tag for a fully equipped and operating plant will reach $165 million or so.
The question-and-answer session participants sounded like a who's who of big-time market analysis/investment firms: Morgan Stanley, Deutsche Bank, Cowan, Piper Jaffray, Lazard, Goldman Sachs, et al.; despite their uppity pedigree, the analysts actually did ask some decent questions. Factoids revealed during the Q&A included First Solar's average watts per module (70); year-over-year throughput improvement (3%); percentage of capacity devoted to new markets (<10% but flexible); improvement in cost per manufactured watt in Germany vs. Ohio plant (about 10 cents) and expected improvement in Malaysian plants (about 20 cents); split of business between ground-mounted and building-mounted systems (60:40, with any moves likely to add to the ground segment); and annual degradation of First Solar's module efficiency (about 0.5%, compared with industry average of 0.8%).
During his Pacific Growth presentation, Meyerhoff shared additional pieces of the First Solar story. The company is highly integrated, with capex costs per watt (including the building) at about a dollar. Since the company owns its coating-tool design/IP, this puts it much more "in control of its destiny, in terms of capacity" and less reliant on supply-chain perturbations. First Solar goes from glass to finished module in about 2.5 hours, another component of its "lowest cost per manufactured watt" claim. The company sees its CdTe module conversion efficiencies reaching 12% by 2010-2012, in line with its goal of being the first PV company to achieve unsubsidized grid parity by that same time-frame. On the elemental front, the company has identified "terawatt levels of tellurium availability," so there is little concern of a critical material shortage.
When asked if First Solar intends to move higher up the solar value chain into the system realm, Meyerhoff said the company plans to play a "more active role in what happens beyond the module," especially in the U.S. utilities sector---First Solar's main developing-market target.
While Sharp has announced plans to have a 1-GW thin-film-on-glass plant ramped by 2010, First Solar should be closing in on the gigawatt level by 2009 and will probably control at least 20-25% of overall global thin-film manufacturing capacity. The Phoenix-based company's business plan is solid, tactically and strategically; its technology and manufacturing efficient, proven, and scaleable; and its coffers are filling with money as revenues and profits escalate.
Make no mistake about it: First Solar is kicking butt and taking names, almost single-handedly driving thin-film PV market share---and TFPV's share of the overall solar pie. Given its steady then spectacular performance over the past few years, I wouldn't bet against First Solar's chances of ultimately achieving grid parity before the rest of the entire PV pack---TFPV and crystalline-silicon players alike.
Readers' comments
Comment by GUEST on 2007-11-23 14:25:33 Where is FSLR going to get 600 tonnes of Te to satisfy 2012 production of 6 GW plus? Theoretically, CdTe can't use less than 90 tonnes/GW. Total world production in 2006 was 304 tonnes. Why do i see only me and one other guy on the internet that has an interest in the research that says Te is a serious problem. The 20GW/yr number comes from bad data gathered in a year 2000 research article when knowledge of Te production was scarce. See 2006 Photovolataics article for more accurate data.
Comment by GUEST on 2007-11-23 15:12:15 The CFO's comment of "terawatts levels of tellurium availability" is a flat out LIE. There is not that much tellurium in the whole earth's mineral reserves, let alone readily available. As a matter of fact, FSLR will be killed by competing tellurium demands which strangle it out of its tellurium supply. Dark future awaits FSLR. See this Seeking Alpha article:
Comment by GUEST on 2008-04-04 10:03:07 Don't be fooled by comments made by Seeking Alpha. They are hell bent on making a story out of Te shortage based on such nonsense as average crustal abundance. Also note the relative lack of acknowledgment of increased anode muds from increases in global copper refinery output and sustained average copper price inducement for new investment. The USGS assessment studies, particularly order ones, are severely constrained by data availability in a tiny market segment. There have been major problems with USGS resource and reserve assessments of major commodities like crude oil, let alone small and new market items like Te. As in the case of uranium, measures of scarcity and resource levels change change rapidly with price signal changes. Although anode mud sources for Te would be straightforward for assessment, other refinery sources such as gold processing and emerging market country suppliers could alter the assessment.
Comment by GUEST on 2007-11-26 18:19:42 The last post referencing Mark Osborne has been discredited in other message boards and the poster knows it. His link is a new article, but it references Hoffman's book which uses research done in 2000 (or 1999) when it claims 20 GW/yr is possible. His 47,000 tonnes comes from 2006 USGS report and is only 50% recoverable with future technology AND it's hopelessly locked up in Copper at 100 ppm until the copper is mined. Today we are able to recover 33% of that Te, but only in copper smelters that use the electrolytic method. "The relatively small value of the Te content in Cu ores even at present record prices provides no incentive to change Cu mining and refining processes to accommodate the increased demand." Copper makes up 90% of Te supply.
PROGRESS IN PHOTOVOLTAICS: RESEARCH AND APPLICATIONS Prog. Photovolt: Res. Appl. 2006; 14:743–751 "Improved Estimates for Te and Se Availability from Cu Anode Slimes and Recent Price Trends" Martin A. Green
Comment by GUEST on 2007-11-23 09:38:35 First Solar is the clear market leader in the sector; therefore, First Solar is trading at a premium multiple compared to its peers. Looking forward into 2008, with First Solar's ability to replicate and ramp-up production quickly, I believe the growth prospects will blow-out all of the analysts' estimates. Looking out even further, once First Solar achieves unsubsidized grid parity in 2010 to 2012, the solar sector will benefit from investment rotation from traditional energy sectors - proving solar energy to be the clean, alternative energy of "investment choice". A decade from now, people will wish they made an investment in First Solar when it was only $212.50/share.
Comment by GUEST on 2007-12-01 11:24:48 Any one interested in Cd/Te or Cd/S we have kilos of it that we want to get rid of. We switched to Si after having refiners come to us to go to mines to buy this stuff when they could not get it.
Comment by GUEST on 2007-11-12 22:36:25 With other solar companies trading near $30 i see no logical reason for First Solar to be trading as high as it is, other than that pure stupidity is driving up the price. A good stock perhaps, but vastly over valued.
Comment by GUEST on 2007-11-26 18:31:43 Mark Osborne. RESPONSE: The guest poster has to understand that supply data for these rare earth materials are normally based on general estimates that take into account as much that is known at a given time. Eagerness to ‘discredit’ is rather harsh and unjustified. Conditions of supply and demand in respect to Te will always fluctuate and shortages coupled to price rises invariably lead to new suppliers coming on stream that may hold a premium for a given time but the supply issue is covered. Higher material costs even though supply is secured for Te could be countered with improved manufacturing efficiencies. I do agree that the Te supply issue is an interesting one and one we will endeavour to watch closely.
Comment by GUEST on 2007-11-23 15:11:48 The first comment was posted by the (admitted short) author of the article linked. In other words he's trying to drive hits to his article.
Comment by GUEST on 2007-11-26 12:52:04 Mark Osborne, Editor-in-Chief, RESPONSE: Noting the tellurium supply debate I found an independent report noting recent prices (2006) and trends for price rises and impact on cost-per-watt for thin film solar applications. Although prices look set to continue rising, the overall cost-per watt impact would only kick in if prices increased 20x from 2006 levels. 47,000 tons is the estimated reserves of tellurium covered in the report. It can be downloaded here: http://socrates.berkeley.edu/~kammen/C226/7r.pdf
Blogs 
