LDK Solar: A New Business Model 11 comments
-
Font Size:
-
Print
- TweetThis
LDK Solar (NYSE:LDK) is a relatively new company, positioned well in the supply chain. This article is an in depth analysis of their existing business model and the effect of vertical integration as they begin manufacturing polysilicon in 2008. It offers very concise and presents compelling arguments why LDK Solar will be one of the fastest growing, publicly traded companies for 2009.
LDK Solar's current business model is described on their web page as follows:
LDK Solar Co., Ltd. is a manufacturer of multicrystalline solar wafers, which are the principal raw material used to produce solar cells. LDK sells multicrystalline wafers globally to manufacturers of photovoltaic products, including solar cells and solar modules. In addition, the company provides wafer processing services to monocrystalline and multicrystalline solar cell and module manufacturers. LDK's headquarters and manufacturing facilities are located in Hi-Tech Industrial Park, Xinyu City, Jiangxi province in the People's Republic of China. The company's office in the United States is located in Sunnyvale, California.
LDK Solar has had phenomenal growth, making their first shipment of wafers in April of 2006. Becoming an IPO in 2007, and achieving 1 GW capacity in 2008. Earnings have followed, with the
FY08 Q2 quarter yielding $1.29 and meeting or beating analysts projections by an ever increasing margin for the last 4 consecutive quarters. LDK also has a low PE and PEG ratios less than 20 and 0.4 respectively at the time of this article.
These impressive statistics are generated from their wafer business model described above. LDK Solar is changing, expanding into a polysilicon manufacturing behemoth on a scale of 16,000 metric tons by the end of FY09. The strategic move will allow LDK to insure a polysilicon supply line for their wafer operations and or sell up to ultra pure 11N silicon grade polysilicon into the market. This move adds incredible value to their business model in a remarkable way while securing their own supply of polysilicon. Revenue from polysilicon operations will add a significantly to their bottom line in the form of profit or savings. There is a wealth of information presented in the most recent Q2 investor presentation and it is a must read for any serious investor.
Let's take a look at this in detail by showing tables and charts based on analyst's projections, LDK projections, and estimated polysilicon costs. I'll explain in detail later on how some of the numbers are generated by providing definitions, formulas, and links where appropriate.
click to enlarge
[Warning: These are mathematical projections based on current conditions that I believe are correct. Market conditions change. No guaranty is implied. You are responsible for verifying data and calculations. I do not make any claim that LDK Solar will perform in this manner! This is simply an output graph based on a mathematical model described under section “Definitions” below! See disclaimer for more information.]
Definitions:
Year
Manufacturing year end. This model's EPS and Estimated Stock Values are valid only after Q4 is public.
Tax
Tax rate applied to LDK. Current tax rate is 12.5% and likely to continue as China has granted tax breaks and lower costs for electricity to encourage rapid business growth. They may change impacting projections.
Shrs(Mil)
Shares in millions.
Poly$/Kg
Market price per Kg. This is an estimate based on renewableenergy.com. It is assumed that the article is referring to the spot market rather than wholesale. Estimating wholesale prices to be ½ of the numbers quoted and “long term” being 6 years, we arrive at a projected decline in prices of around 23 dollars per year. If this article is referring to wholesale prices instead of spot market prices, all related polysilicon earnings are to be doubled. And if more countries jump into the effort to reduce greenhouse emissions, this estimate may be low as the polysilicon market will remain tight.
P'COM$/Kg
Polysilicon cost of manufacturing per Kg. reported from Q2 presentation, pgs 22, 23. LDK Solar stated that they estimate a cost of manufacturing of $22.00 to $32.00 with production from the 15,000 mt facility. This facility will not be complete until the end of 2009. The first year thereafter (2010) will be the year that these numbers apply. Prior year P'COM numbers presume a non-linear rate of reduced production costs as LDK leverages the shear scale of a 15,000 mt plant to their advantage.
P'Mfg[MT]
Manufacturing capacity in metric tons based on LDK estimates. For 2008, the 1,000mt facility average is 225. For the 15,000mt facility, capacity is expected to reach 6,000mt at the end of 2008. This level of capacity can only be utilized for ¼ of the year of which they will be ramping up production.
Thus:
or 975 would seem correct. For 2009, the numbers are given. For 2010, we assume the 1,000mt and 15,000mt facilities will be operating at full capacity. Information can be found relating to polysilicon capacity in the July analyst meeting slides pg 11.
P'Rev(Mil)
Revenue from polysilicon. This excludes any wafer revenue.
P'Margin(Net)
Profit margin after taxes.
EPSPoly
EPS generated from polysilicon.
EPSWafer
For 2008, the EPS is given by the analysts consensus. Since EPSPoly is calculated, the difference must be EPSWafer. A constant can be determined representing the EPSWafer per GW by taking the 2008 EPS Wafer and dividing it by the 2008 year end GW capacity specified in the Q2 presentation. The constant is then multiplied by the wafer capacity projections (Q2 presentation, see pgs. 8, 27) for the remaining years respectively. It is a simple matter of scaling. This is conservative as their exists a certain efficiency in scale that is not taken into account. It also appears the analysts may have forgotten to factor in the polysilicon manufacturing, perhaps theorizing that the polysilicon plants would not be producing in the time frame that LDK has stated. In any case, I've used the analyst estimates as the reader will place more faith in these numbers.
EPS
Earnings per share.
[Note: I've color coded many fields in this table. A green field means the data is either from analyst's data, Q2 report data, or mathematically derived from very logical arguments above. A yellow field is highly subject to market conditions or estimated tax rates. I've placed what I believe to be reasonable in these fields. The polysilicon that LDK will be producing is ultra-pure 11N silicon grade, and thus likely to fetch higher value than what I have considered an average from the link provided regarding this topic.]
Analysis:
I find these graphs astounding as one can readily see that in one short year, the polysilicon manufacturing may contribute more to earnings than the wafer part of their business. For the wafer business, this level of earnings will have taken 3 years. For a stock that has great earnings history and an outlook that implies around 300% growth next year and 200% the following year, it is without a doubt a bargain if projections come true. Trading multiples of 30 to 80 are certainly not out of the question as we see LDK Solar growing into one of the largest suppliers of polysilicon and wafers in the world in 2009. As you will note, the chart above for estimated stock value is a calculation of just 22.5 which results in astounding numbers! Imagine what the earnings will be in 2011 as the maximum capacity is fully utilized coupled with investor excitement pushing PE factors north!
LDK Solar has a long list of customers that keeps growing. LDK reported that they were near wafer plant capacity sellout for 2008-2009 in March of 2008. It is virtually certain that the plant capacity for these years now completely sold out. Many of these contracts are pre-pay, take or pay type contracts which further strengthens LDK Solar's future revenue stream in a no loose scenario. A quick look at their customer contracts and duration in the table below is impressive. These customers are based worldwide. The global diversification of their customer base adds a layer security for LDK as they are not as susceptible to any one countries alternative energy policies. Details of these contracts can be found at LDK Solar's website.
The effect of the prepays is not without notice. From the FY08 Q2 presentation we see prepay dollar amounts increasing. The plant expansions are valued at approximately 1.2 to 1.8 billion per several posts at yahoo MB for LDK. It is likely that with 6 more quarters to go before the plant expansions are complete, that prepays may end up paying for all expansion activity. That means little or no interest on plant expansions with LDK Solar having successfully funding expansion activity directly from the customers who require LDK Solar products in exchange for wafer production.
These projections may indicate that LDK Solar will not maintain a low profile for long. Will Wall Street take notice? Very likely! LDK Solar will soon be opening their 1,000 metric ton facility if not already. They re-confirmed this with even more certainty in their Q2 as I recall. This event will signify success and will likely serve as a catalyst, a signal that LDK Solar is open for polysilicon business. It is unlikely a company in 2009 and 2010 EPS and Revenue with projections this high can hide in the shadows for long. Can anyone really afford to not have LDK Solar in their portfolio with such huge potential? LDK Solar represents a buy even if the projections given here are only 50% correct! Where exactly is the down side with a full load of customers either signed or waiting to sign with LDK Solar and margins such at these?
FAQ:
Q. How did you derive future polysilicon cost per kilogram and how can I be sure it is correct?
A. I'm not sure it is correct. There is a link below to renewable energy talking about polysilicon prices. It is unclear if it is wholesale or retail. Wholesale is usually about half of retail. The contracts in the industry seem to define long term as 6 years out. Since the context of the article is about Solar PV, I presume the quality is 5N purity. I took a best guess aiming low in my opinion, by assuming wholesale costs at ½ of that which is posted, and a linear rate of decline over 6 years as worldwide polysilicon manufacturing is ramping up to meet demand. This extrapolation comes out to be around $23.00 decline per year. Actually, the number is probably higher as 11N polysilicon is silicon grade and likely much more expensive.
Q. The 2008 EPS number is low in my opinion and you based your arguments on this data, wouldn't the overall data be incorrect?
A. Yes. There are 2 other methods to derive a likely EPS target. We could double FY08 first half numbers of (.45 and 1.29) but this would not be correct as the Q2 had a one-time gain that we cannot use in any extrapolation of future data. We can back that gain out of the Q2 and use it in this manner. We then arrive at 2 x (.45 + .82) = 2.54 but this would not account for the 11% guidance given by LDK for Q3. We could presume the 11% guidance will be repeated in Q4 which would give us a more linear rate of growth that matches the effect of the ramping of the factory wafer operations. These numbers would then be (.45 + .82 + .91 + 1.01 = 3.19). I actually believe this to be correct and reasonable for the wafer side of the business. Thus, I have been conservative, allowing the reader to place faith on analysts which they typically do. The overall effect then is that this table and graphs are likely conservative on the wafer side of the business. These charts are bullish enough to tell the story.
Q. I've looked at your projected earnings and calculated the market capitalization for 2010 of LDK. Are you implying LDK Solar would be valued at around 30% of Google's market capitalization is today. How can this be correct?
A. I've only provided the math and extrapolations. Data has been taken from the Q2 presentation and analysts estimates. LDK Solar has demonstrated meeting or beating every target projection. Xinyu, China has backed LDK with reduced taxation and electricity costs. LDK has stated that they are targeting a full 30% of the market. Look at the table found here to gauge what a 16,000 mt polysilicon manufacturing facility really means in comparison to others. If you believe these projections are incorrect, you have to find an error in the calculations, prove the polysilicon estimates are incorrect, or state that the input data from the Q2 presentation and analysts projections is in error. If you take the position that the plants will not be at the capacity projected at the time stated, you only shift the time-line but not the EPS value. Plant capacity is booked. With low labor costs and aforementioned financial breaks, it is unlikely any other polysilicon manufacturer will be able to compete. This projection is actually conservative as with the earnings increasing at the projected rates, a PE of 22.5 is low. PE numbers of 30 to 50 or higher in late 2009 and 2010 are likely as the projections give way to fact.
Q. I've noticed that you have calculated plant operations at full capacity. What if they do not run at full capacity?
A. It is unlikely that LDK will operate at anything less than full capacity. They have already stated that their plant capacity is booked solid for around 2 years. Operating at less than full capacity means they will be delinquent on contracts and risk client relationships. Secondly, excess poly can be sold into the market at today's higher prices rather than the lower prices of tomorrow. There are many people in China looking for higher paying jobs. The Solar PV sector would be a highly desirable job to have in China's economic climate and I believe there are more than enough people to fill positions.
Q. LDK stated they would be using less polysilicon in their business and that they have increased yield from wire saw operations and polysilicon slurry recovery methods. How will this impact your projections?
A. The polysilicon side of the business is based on projections that LDK has given us. I do not know if their projections on P'COM$/Kg. include these cost saving measures. Since they are in the position to know, I presume they have factored that into the projections.
Q. What about the investments with Sonoma, a crucible manufacturer. Won't this help?
A. These investments are wrapped up into the EPS for 2008 in a manner of speaking. They are effectively Grand-Fathered in-whole or in-part.
Q. What about “efficiency in scale” which according to the last Q2, was responsible for a major increase in earnings.
A. This on-going effect is not factored in. That would tend to result in the table also being conservative.
Q. Are you a CPA? How can I trust this information to be accurate?
A. I am not a CPA. I do not pretend to be a CPA. You should not trust me. I am a common person with a lot of experience working in a high-tech engineering field with experience in manufacturing, management, quality control, and have built relational databases that have saved one company about a million dollars by stratifying warranty returns data and feeding that information back into Engineering design and Quality. I have an Associate's in Electronic's Engineering and have worked in the field for over 20 year's. I program computers for fun using Linux, C, and C++ although my current job relates more to customized electronic hardware. I am very logical in my thought processes having worked in this environment for years. Life experience counts for something. Review the data for yourself! I want you to!
Q. Why hasn't Wall Street reacted to LDK.
A. I suspect it is due to the fact that LDK is a very new kid on the block, and not well covered in the US publications. It is in China where it is more difficult to gain access and get interviews. However, the projected earnings, and the world view of China as an ideal manufacturing environment is changing for the better in a big way. I believe the incredible growth of LDK will be hard if not impossible to ignore. As LDK becomes more widely acknowledged, and recognized, PE ratios will likely increase dramatically, especially in 2009. There also seems to be disbelief that manufacturing plants can be built as fast as LDK claims. Yet, the Q2 clearly shows many older pictures of the rapid progress. LDK Solar is committed to rapid scaling of their business. They have to, due to their customer's needs. I believe the analyst's have compared the rate of construction in the USA to that of China. Different construction requirements relating to work hours and inspections apply. Or, maybe they know something that I do not. You be the judge. Investing is always a risky business!
Disclosure: I currently hold less than 500 shares of LDK stock, at the time of this article. I am not receiving compensation for writing about LDK Solar. This article is based on forward looking statements which may change. Projections do not necessarily come true. This article is not a recommendation to buy, sell, or trade any stock whatsoever. The author is not responsible for an individual’s choice to base any investment decision on the data presented. The author advises independent research if you choose to invest in any stock or mutual fund.
Links:
- http://media.corporate-ir.net/media_files/irol/19/196973/july_Analyst_Day_ALL_Slides_FINAL.pdf
- http://media.corporate-ir.net/media_files/irol/19/196973/aug_12_LDK_Investor_Presentation_Q2_Final.pdf
- http://www.renewableenergyworld.com/rea/news/story?id=52696
- http://www.ldksolar.com/index.html
- http://www.ldksolar.com/Company Profile.html
- http://www.ldksolar.com/6-4-07.html
- http://www.ldksolar.com/8-25-08.html
- http://finance.yahoo.com/q/ae?s=LDK
- http://finance.yahoo.com/q/ks?s=LDK
- http://www.ldksolar.com/3-13-08.html
- http://www.ldksolar.com/Company news.html
- http://www.ldksolar.com/7-1-08.html
Related Articles
|
Find People on SA
-
The Seeking Alpha 100
See allThe top 100 stock
market authors
selected for publication
Seeking Alpha 100 » -
Fastest Climbers
Authors with the largest
increase in followers
in the last 7 days-
1
David Fry
-
2
Bill Gross
-
3
Philip Davis
-
4
Tyler Durden
-
5
TraderMark
Fastest Climbers » -
1
-
Top Commenters
Top commenters,See all
as ranked by their peers
Top Commenters » -
Top Instabloggers
The 100 most activeSee all
Instabloggers.
Top Instabloggers » -
Top StockTalkers
TopSee all
StockTalkers.
Top StockTalkers »
This article has 11 comments:
sweitzman@gmail.com
-scott
A. I suspect it is due to the fact that LDK is a very new kid on the block, and not well covered in the US publications.
Huh???? How about all the media coverage it got back last year? TONS of coverage but all the wrong kind...
IAM PHREE... cute. That implies a value of zero.
I was hoping to accumulate more before the big guys jump in. And jump in they will.
Thanks, whatever your name really is...
If we narrow our focus on investment return: the pertinent question is why are so many people still very afraid of putting money in LDK? It takes too much time to answer this important question. Suffice it to say that if one follows the elites on Wall Street, he or she must have a significant loss due to the collapse of Bear Stearns and others.
Why do I trust LDK? The simple answer is Mr. Xiaofeng Peng, the founder and CEO of LDK Solar.