Is There Value in Silicon Solar PV?

Several “analysts” advise us to avoid the recent leaders in the Solar Photovoltaic [PV] industry, as they will crash and burn because they are “growing too fast”. The cost of rapid growth “will end their ability to generate profits”. They also point out that manufacturers may have to “prepay” for some of their supplies, as well as fund large costs for additional production lines to meet increasing demand. To one who looks for “growth stocks”, throw me in that briar patch!

Other postings discuss how new PV technologies under development will undercut the prices of Silicon PV. Claiming that thin film, e.g., CdTe, and more recently CIGS (Copper Indium Gallium di-Selenide) can be manufactured “far cheaper” than the Silicon/Polysilicon panels that dominate, by MW production, the current sales and installation statistics.

These articles are full of useful, helpful information. I put quotes around the term “analyst”, as there are a variety of definitions for the people doing this work. Some work for private consulting firms where their income is based on providing research to leading companies in the industry, or provide research to investment companies or marketing companies. I think of “analysts” as those who work for brokerage firms who make inputs to the consensus forecasts found on First Call or your favorite broker or website.

Recently, a student “analyst”, Michael Lu, posted an excellent article about LDK Solar (LDK), in which he researched, in Chinese, the tax records that show how profitable LDK is, and will be, as their big investment in expanding their Silicon factory creates more product output soon. He also shows how income taxes in China encourage new companies and technologies. First, with startup cost write-offs; then when they begin manufacturing and selling products, they are exempt from national income taxes for 2 years, etc – see his article. So LDK and other startups do not pay the full 25% Chinese income tax for a few years.

Another “analyst” recently posted “Who Will Crack the CIGS Nut in Thin Film?”. Neal Dikeman provides a very conservative list of the monetary costs to startup a CIGS PV manufacturing company, along with an estimate of 3 to 5 years to turn that investment into product, with positive sales and income. The costs and time for Silicon PV manufacturing are similar, if not more expensive. He says that most of the CIGS companies are at a very early stage of startup (while Silicon companies grow and mature).

Below is a table of forecast sales and earnings for several solar stocks which I obtained from Schwab and/or Daily Graphs. I am told that Schwab got this information from Reuters Research and Wall Street Horizon. As a retired person, with modest funds, I have made the assumption that the “analysts” whose consensus earnings forecast are listed below, include the cost of money, and all line items provided in each quarterly and annual report as expanded on during the quarterly conference calls. These forecasts should take into account all the “scary” information about the hidden costs that are an integral part of “fast growth” companies.

My general conclusions about Solar PV companies then includes:

• The consensus earnings and sales numbers provided by Schwab & DG are researched numeric estimates, compared with the scary opinions of some “analysts” who have an unknown agenda. Particularly when they provide no facts, nor numbers that can be verified.
• New technology may lead to reduced manufacturing costs, and lower selling prices, but the cost of funding a new startup and the time required to set up suppliers, buy equipment, hire and train staff, gives the current Silicon leaders in MW production and sales significant time to react to the potential competition. Checkout Advent Solar who laid off manufacturing staff in March while continuing their R&D activities.
• There are advantages in both cost of labor and income tax rates for those manufacturing in China or Asia, as most of the current volume manufacturers are doing.
• The number trends provided for LDK should also apply to most other current leaders who do their manufacturing in China.

Now, to the focus of my posting – comparison of several of the leading Solar PV companies sales and earnings forecasts. These values were obtained from Schwab & DG as of July 11 and to the best of my knowledge are the consensus of several “analysts” who are paid to just present the facts, after talking to the technical and financial folks at each company, and probably visiting each company.

Most of these companies report that they made, sold and delivered 25MW to 120MW of PV last quarter and they expect to have 200MW to 1GW capacity by the end of 2008, with double that by end of 2009. Most of them also claim to have sold most of their 2008 production and some of their 2009 production. It is unclear whether some or all of these companies require a deposit(prepay) to offset any deposit(prepay) they incur from their suppliers; but the forecast numbers should include the appropriate cost and cash flow reality. I included Energy Conversion Devices (ENER) in these tables as it has gained in popularity on SA, even though it had negative earnings last year and expects to earn $0.01 in calendar 2008. I normalized their earnings and sales figures (their fiscal year ends June 30), so everyone in this chart is compared on a calendar year basis.

Here are actual and “analyst” forecasts for sales and EPS for these companies.

Detailed quarter by quarter forecasts are available for sales and earnings, but space limits me to annual numbers only. I expect the 2009 estimates to change every quarter, as each earning cycle is reported, and conference calls provide new information. Therefore I will not comment on the 2009 numbers. The year over year forecast sales and EPS growth numbers are ALL way above average for S&P companies.

In good economic times these growth numbers would be worthy of P/E’s far above the current stock prices. I won’t comment on this data, but I think everyone can find a case for your favorite company. Even the larger, slower growing companies in this group have sales and earnings growth that leads the general market, and they are generally supported by many “analysts”, so their P/E is at a premium, as shown in the next table.

Note that the first 8 companies in this list make Silicon PV products, and lead the PV market in MW volume sold and shipped. Note that ENER is two years behind the leading Silicon companies in MW volume and in EPS. Their ’09 forecast of $1.47 is equal to the actual ’07 EPS for TSL, which is forecast here at $4.47 for ’09.

Here is a table showing capitalization vs price, EPS, P/E, forward P/E, P/S and forward PEG, where P/S is simply capitalization divided by ’08 sales (Price/Sales), and PEG is forward P/E divided by % EPS growth this year, as listed in this table. This table is sorted on the forward PEG in the last column.

These numbers are based on consensus sales and earnings numbers forecast by many analysts. Based on Trailing (current as of 7/11) P/E, 8 of these companies (all Silicon PV producers) are at 32 or less, and I would deem them attractive based on forecast sales and earnings growth from the first table. These eight companies also have a 2008 forward P/E that is very attractive; less than 22. These same eight companies have a Price to Sales ratio that is less than 4; with five of these less than 2. Finally, six companies have a forward PEG less than 0.2, which is almost unheard of; and the first ten companies are less than 1.0, where most “analysts” recommend growth companies with a forward PEG less than 2.

Another recent posting suggested only three of these companies, with addenda later. In "Temporary Market Bottom? 3 Solar Stocks That Look Like Bargains", David White used earnings forecasts from TD Waterhouse, which are within 5-10% of the numbers listed here. His forward PE and PEG numbers, however, used his own ‘Target Price Estimate’ instead of the current July 11 price used here to compute forward P/E, and forward PEG.

Sunpower (SPWR) ($0.37 forecast) and Evergreen Solar (ESLR) ($ -0.10 forecast) [forecasts are from Bloomberg.com: Earnings] are releasing actual earnings for the June 3 quarter on July 17, and the remaining companies will likely release earnings the 2^nd and 3^rd week in August, when we will witness the “crash and burn” or continued growth shown here. I assume we will get lots of news from InterSolar 2008 at the Moscone Center in San Francisco this week, July 15-17, which is combined with SEMICON, where Applied materials (AMAT) will present their new SunFab equipment.

Disclosure: I have held positions in the first ten this year, and currently have modest holdings in CSIQ, TSL, and SOL.

Comments

[rana]

these PEG numbers are not how we usually calculate PEG, so please try and use the standart method.
if a company made 1 cent last year and will make 1$ this year and 2$ next year and share price is 50$ with 50 million shares. we get PEG of 0.05 using the above method. if we will apply the concept of PEG of 1 the stock price will be 1000$, while next year we will get PEG of 5 which will take the price to 200$. (all is theoretical and is given for example purposes).
in order to smooth the results and get better picture of the longer term picture we use the 5 yr. expected growth which gives us very different results.
we still get very low numbers compared to other sectors, but no way near these results.
overall good post as it's very informative and comparative.

[Tonny]

why always donot you cover the stock "CSUN"?

[Jim Brown]

Rana - If you google PEG, you will get:
The PEG ratio, Price/Earnings To Growth, is a valuation metric for determining the relative trade-off between the price of a stock, the earnings generated per share (EPS), and the company's expected growth.(Price/Annual Earnings)/%Annual Growth. If a company is growing at 30% a year, then the stock's P/E could be 30 to have a PEG of 1. PEG ratios between 1 and 2 are therefore considered to be in the range of normal values. A crude analysis suggests that companies with PEG values between 0 to 1 may provide higher returns.

A lower ratio is "better" (cheaper) and a higher ratio is "worse" (expensive). A PEG ratio that approaches two or goes higher than 2 is believed to be too high. This means that the price paid appears to be much too high relative to the projected earnings growth.

They also say that some people use high future growth rates , as much as five years out, which can be set arbitrarily high. I used only the numbers in the tables, and did not attempt to project out in this new industry.

Tonny - CSUN only produced 9MW of 17.2% efficiency cells in Q1 and their last 4 qtrs eps were $-0.09, $-0.11, $-0.06, and $0.01 and they just closed a Convertible Sr note offering to increase production and improve product efficiency. They are behind the curve of these other Silicon companies in product and earnings growth.

[alpha24seven]

Much discussion about analysts. What you should probably know about analysts that most believe the best analysts won't ever be heard. Why? because they all work for hedge funds and hedgies don't broadcast research. There is no way the smart analysts can repel the massive money and better life a hedge fund will provide for them. The best proxy you can use is short interest. If there is big short interest and the pps has been beaten that is probably a result of the best analysts money can buy. Relying on stock analysis by some junior analyst from a known firm (even GS) or guys on SA will get you killed.

[alpha24seven]

Another thing. There is a coal mine canary named Nanosolar. Watch them carefully. If -- and I say if they can do what they say than all Si based solar will be killed -- dead...

The only thing to do is see if you can get borrows on the entire Si PV basket.

Completely crushed. But then again, so will a whole slew of other technologies like coal and natgas etc. Far and wide reaching destruction at the hands of just this one company.

[rana]

you use the early stages of growth which is unsustainable.
do you really think that CSIQ will have 600% earnings growth in 2009 and 2010?
i only commented on the reason we don't use 1 year data, just like the reason we calculate moving averages and exponantial MA or the reason we use logarithmic scales and not linear.
if it's not understood you can use the unsmoothed method, but it will probably cost you somewhere down the line.
even here with CSIQ, if you think under 1 is good measure of value, you can pay 500$ for this stock and stay well under 0.5.
do you really think it's logical?

[User 139507]

Well thought-out, well researched and well argued. Almost makes me wish I could purchase a solar etf minus the whale, fslr.

azalphainvestor

[Jim Brown]

rana - totally agree early growth is unsustainable, but it separates the potential stars from average startups. Due diligence seems to be appropriate, i.e., keep your eyes open and ears to the ground.

alpha24seven - I am aware of 8-10 CIGS privately held companies. And CIGS on glass seems to have highest conversion numbers per the NREL. Some use evaporation, electroplating, printing (like Nanosolar) on foil, thin film silicon and thin stainless. Since they are private we only hear the positive news, and successes, and potential capabilities. Perhaps they will provide real information at InterSolar this week. But then we can't trade their stock, and we are stuck with consensus estimates or guesses on those that are public. I expect it will be six to 18 months until we learn of actual sales, along with hopefully: sales prices per watt, conversion % and production capability(and $ in and out) for them all. And perhaps some will go public.

[Global Warming Examiner]

All of the silicon solar companies are growing capacity rapidly. Some analysts seem to think that this growth will eventually push the group off a cliff as governments remove or reduce incentives for solar systems. This assumption is based on the fact that as production increases, solar incentives will become more costly for governments to fund.
What is most likely to happen is that governments will reduce solar system incentives. This is not a problem for the industry as the costs of solar are decreasing. The cost of alternatives to solar like natural gas are increasing. That means governments can still get solar power to grow in the future with smaller incentives. Other uses like solar panels on electric cars could create additional demand. Hopefully solar will reach a point where it will grow without incentives.
There are a lot of companies that are researching and even producing solar from thin-film technology. But they can not meet all of the demand from the market so silicon based solar still has tremendous room to grow. Many of the solar companies are sold out into next year. Article on my web site: "Solar Cell Manufacturers Have Room To Grow" shows how small a contribution solar is currently making to world electric production. Shorting these companies, especially the ones that have raised guidance is very dangerous. I am long CSIQ and TSL.

[frflyer]

Alpha24seven I would not assume that Nanosolar or anyone else will kill off the silicon solar market. There is so much room for increased demand for solar that there can be many players with different technologies. Silicon cells still provide the most wattage per square foot, with twice the efficiency of thin films. Where square footage is an issue, silicon is the best choice, where it is less an issue, thin film will be the better choice. And that doesn't even take into account solar thermal power plants or concentrating PV power plants, which are a whole other ball of wax. Yes there will eventually be shake outs, like in any emerging industry, but I believe it's too early to predict how that will play out.
And silicon is still one of the most abundant materials on earth, easily mined or collected. With enough capacity for refining the raw material and making wafers, economies of scale may keep silicon cells competitive for a long time to come. I'm impressed with Nanosolar too, not bashing them.

[flashflash]

What do I do with the short intrest table?
Is a stock with less short intrest a safer buy?

[Ames Tiedeman]

JASO looks like a low cost producer. What scared me is they are so dependent on Spain.

[Ryan Hall]

Hi,

I haven't been able to find an answer to this question, so I hoped that you might be able to help me.

I often see the price of solar energy listed in price per kilowatt hour. I don't understand the time frame used for this. If solar pays for itself 15 years down the road, then over 15 years, solar's price/kwh must be better than conventional energy (15-25 cents/kwh), right? So for what time frame is the price of solar used? One year? Five years?

Ryan

[10Q Detective]

Excellent overview. But, let me repeat my mantra: When cost of capital begins to exceed expected IRR, the solar game is up!

industry.bnet.com/ener.../

My Best,

David J Phillips - Editor, 10qdetective.blogspot....
Contributing Energy Analyst
CNET/BNET