One take-away from the Q3 earnings release season was the aggressive drive to lower production costs. A common metric is the aggregate processing cost excluding the cost of polysilicon. This number represents the in-house cost per watt from wafer to module.
Normally, I do not use aggregate processing cost numbers. They are generally not useful for calculation of gross profits. My calculations are based on the three individual processing cost components of the aggregate numbers (wafer, cell and module). However from a comparative in-house efficiency point of view, these numbers can be useful.
The nine solars discussed in this article are listed below:
- Canadian Solar Inc. (NASDAQ:CSIQ)
- Hanwha Solarone Co., Ltd. (NASDAQ:HSOL)
- Jinko Solar Holding Company Limited (NYSE:JKS)
- LDK Solar Co. Inc. (NYSE:LDK)
- Renesola LTD (NYSE:SOL)
- Suntech Power Holdings Co., Ltd. (NYSE:STP)
- Trina Solar Limited (NYSE:TSL)
- Yingli Green Energy Holding Co. Ltd. (NYSE:YGE)
- First Solar (NASDAQ:FSLR)
The Chinese companies have been highly efficient in processing costs but with the advent of our spectacular module ASP declines, they have been forced to dramatically lower processing costs. Between Q2 to Q3 and Q3 to Q4, there have been some significant changes to processing costs.
Eight cents for TSL between Q2 and Q3 would normally look suspicious but then other peers have achieved significant cost savings during this period. YGE appears to have reduced costs roughly six cents. This would support the TSL claim. Of the savings, I am a bit wary of the LDK cost reduction of 10 cents for the quarter. They just have not shown a history of flawless execution.
The table below summarizes processing costs for 2011 and 2012. For a cross reference to our numbers, the readers can refer to this website.
Processing Cost Roadmap For 2011 and 2012 - Excluding the Cost of Polysilicon
|2011 Q1||Q2||Q3||Q4||2012 Q1||Q2||Q3||Q4|
|FSLR (All-in Cost)||0.75||0.73||0.72||0.72||0.71||0.69||0.67||0.65|
From the author's perspective, the TSL, JKS and YGE numbers hold the greatest credibility.
For comparison, I have included the FSLR all-in cost road map. As the price of poly continues to fall, they will have a very difficult time competing against the Chinese solars. It is possible that they may only have an eight to ten cent lead over the Chinese solars in terms of all-in module costing.
The table below summarizes the potential declines in processing costs for 2011 and for the two year period of 2010 to 2012.
One Year and Two Year Processing Cost Declines for Solars
|One Year % Decline||Two Year % Decline|
|FSLR (All-in Cost)||4.0||13.3|
Even with the very aggressive cost reduction road-map, 2012 could still be a very challenging year if module ASP averages $1 or less. At $1, four of the eleven solars (including TSL) could lose money. As mentioned in previous articles, plus or minus five cents will make the difference between a bull market and a bear market (assuming everything else remains static).
A Note on Our Previous Article
Our previous article was published on Thanksgiving Friday and the hits were pretty poor (around 2,500 so far). So this is a plug for the article which dealt with the accuracy of the Street Q3 EPS estimates to the Helios EPS estimates. These estimates were compared to the actual Q3 earnings reported by the companies. We are pleased to report that we scored a victory of...