On Wednesday morning, Yingli (YGE) confirmed its non-cash impairments with the release of unaudited results for both Q4 and full year. For the year Yingli showed 16.69% in gross margin, all non-cash adjustments resulting in $509M loss, or loss of $3.25 per share. From the perspective of business growth for 2012, the company expects the addition of 750MW of vertically integrated capacity, a step ahead of Trina with a total figure of 2.45GW. The technological roadmap presented by Yingli matched similar presentations made by Canadian (CSIQ) and Trina (TSL), projecting increases of multicrystalline conversions to 18.9% and mono conversions to 21% with the introduction of Panda II cells based on MWT technology.
Operationally, the dynamic is a mirror image of Trina's efficiencies at $0.64 per watt in Q4 for non-polysilicon, and rather worse performance in polysilicon cost at $0.33 per watt, after a $0.07 per watt provision. ASP for module appeared to be in the $1.07 range with a 25% contribution from high conversion n-type mono Panda modules, six cents higher than Trina's. Yingli sees Q4 2012 non-polysilicon costs in the area of $0.56 to $0.58 cents, and polysilicon cost in area of $0.17/watt (5.6 g), with total module costs of $0.73 per watt in the best-case scenario; this confirms our expectation that polysilicon is the key to gross margin improvement.
Despite the loss, the company's robust targets for 2012 should eliminate any doubts regarding expected growth. Yingli plans to sell 2.4 to 2.5GW, exceeding its future capacity levels and suggesting full utilization of the capacity with a 56% growth target. If the previously mentioned processing cost is set at $0.73 per watt and ASP can average $0.90 for Q4, Yingli could see 18.8% in gross margin by the end of the year. If ASP can remain closer to $0.90 per watt for the year, Yingli may deliver 2012 revenue equal to that of 2011, a condition that will not be shared by other companies, particularly those without the ability to expand. Yingli is planning to sell 35% of its volumes to China, 40% to Europe (with 25% being Germany), and 20% or so to the US in 2012.
First Solar (FSLR) sent the distress signal to investors the night before, delivering unexpected losses for Q4 and the full year. The company recorded a loss of $4.76 per share for the quarter and $0.46 per share for the year, writing off $393M of non-cash goodwill for its component segment, $164M of cash costs related to warranty expenses, and $60M for restructuring charges. The company announced production stoppage of four lines in the plant located in Germany, as well as discontinuation of the construction of the new plant in Vietnam. Total capacity will be also underutilized, including operations in Malaysia, allowing for retooling for new efficiencies, with expected yearly production output at the 60 to 70% level.
Having $2.7B in revenues in 2011, First Solar is still the leader among solar companies and it appears it will remain so by projecting lower than previously reported - but still impressive - revenue of $3.5 to $3.8B for 2012; conversely the company's module business seems to lose its cost advantage over the crystalline silicon with projected costs of $0.74 per watt for 2012. This average cost is a direct result of operational underutilization, in contrast to full capacity processing at $0.67. In the observation of Canadian Solar reporting a Q1 2012 module production cost at $0.74 per watt and Yingli suggesting costs of $.73 by Q4 2012, it will be difficult for First Solar to compete with the high conversion of silicon modules, while having the same cost dynamic. On average, c-Si (crystalline silicon) modules exceed the efficiency of thin-film by 30%. Assuming the same ASP, or just a nominal difference, polysilicon modules, which already outsell thin-film by many-fold, will not only continue to lead but will take on additional market share.
Another unfavorable condition to First Solar was the increase of warranty costs ($37M) for the sales made in high-temperature climates, described during the conference call. The CdTe modules were, until now, considered to be the best for durability and the conversion without any LID (light induced deterioration) effect, specifically in a hot climate setting. This unexpected admission may suggest that the second largest market after China, India, potentially may not be the best suited for thin-film, if warranty costs start piling up. Faced with $253M expenses in warranty and related charges to date, it is hard for First Solar to ignore the negative repercussions, particularly in a fiercely competitive market where a few cents per watt can make a difference between success and failure.
Furthermore, Solar Frontier, a Japanese CIS (copper, indium, selenium) manufacturer, is considered to be a serious contender in this segment of solar business and is rising quickly in the ranks of thin-film names. Frontier released its new record of 17.8% cell efficiency at aperture area, and the company recently produced a module with conversion of 13.38%, while its commercial modules are rated at 12.6%. In comparison, First Solar projects 12.7% efficiency for 2012. Beyond technology advancements Solar Frontier also has been actively pursuing plant developments in Japan and the U.S., including 100MW of projects in California, while being behind the most extreme increase in Japanese domestic sales of thin-film/CIGS in the last few quarters. Frontier Solar has a 950MW of capacity at home, and it is engaged in plant construction in the Middle East. In contrast, First Solar capacity will be 2.5GW by the end of 2012, but with idling production lines, the company is considering production of only 1.5 to 1.8GW in 2012.
For the longest time Chinese solar manufacturers have had a firm lead on the processing cost of crystalline silicon modules. Now they appear to have conquered this area exclusively for themselves. Without the cost benefit thin-film has no advantage over superior conversion of c-Si modules, which does not translate to a better selling price. Asian competition within the segment will only intensify as South Korean companies now see thin-film and CIGS as the new opportunity, after losing a round in silicon to the Chinese. First Solar, which wants to be more of a developer than a manufacturer under these circumstances, is going to look into areas that are becoming increasingly crowded with other companies, like MEMC's SunEdison, which is installing plants with c-Si modules already in India, Thailand, and soon in Japan. Chinese companies are also looking to enter utility and commercial scale plant development. China Technology Development Group Corporation, GCL-Poly Energy Holdings Ltd. (Hong Kong, China), and a subsidiary of China Merchants Group (CMNE, Hong Kong, China) signed agreements to invest in pv utility scale projects in China worth 500MW, but also in the US, on a smaller scale. GCL Poly, the largest wafer maker in the world, recently entered the US market to build solar plants with NRG and got loan guarantees from Merrill Lynch to fund a number of projects across the USA.
It appears that the only way for First Solar to stay ahead is to work on module efficiency and cost reduction, the routes that are likely to be the most difficult, judging it by the past efforts. In an attempt to improve its lines, another CdTe manufacturer, Abound Solar, shut down its module production today, laying off 180 workers from its plant in Colorado. Abound Solar is planning to achieve 12.5 to 13% module conversion by 2012 with installation of new generation equipment.
First Solar is not currently considering the use of PV silicon modules in it is plants, because, as they stated in the conference call, "our own modules are better", a statement that may have to be revised in light of warranty expenses going forward.