After about 12 months of unabated growth for the solar power group, shares of
have been under some pressure lately.
The culprit has been renewed analyst attention to the potential shortage in these companies' main feedstock, polysilicon; the companies have been warning in recent filings that supplies are tight. I want to focus on Evergreen Solar, which we believe is a buy around the current quote of $14.04 because of a unique supplier relationship.
Note that we held a position in Evergreen Solar in the model portfolio for my newsletter,
The Street.com Stocks Under $10,
, but sold it recently because shares were trading above $15, and we viewed the risk/reward as less favorable. We would consider adding shares back to the portfolio here if the portfolio didn't already have exposure to the solar power space.
Polysilicon is a key raw material used in the creation of solar panels. While not all solar technology is based on polysilicon --
Energy Conversion Devices
uses thin film and
solar technology is based on gallium arsenide -- it is the proven standard with the greatest market acceptance. Typical polysilicon solar cells convert energy from the sun into electricity at efficiency rates of 15% to 20% vs. just 10% to 12% for thin-film technologies.
End-market demand has been booming for solar power, thanks to government incentives and record oil and natural gas prices, but total growth has been curtailed because of limited supplies of polysilicon. Because polysilicon represents nearly half of the cost structure for companies like Evergreen, the shortage and ensuing high market prices have pressured gross margins.
Evergreen shares took a hit last week when the company disclosed in its 10-K annual report filed with the
Securities and Exchange Commission
will not meet its contractual obligations to supply polysilicon to Evergreen. Instead, MEMC has opted out of its Evergreen contract and is taking advantage of record polysilicon prices in the spot market.
This disclosure created a panic, largely because of the incorrect conclusion that Evergreen wouldn't have enough supply to produce at 100% of its capacity in 2006. The company currently operates a 15-megawatt facility in Massachusetts, and will have a 30-megawatt facility up and running in Germany in June. (The megawatts represent total annual manufacturing capacity at each plant.)
The German facility is a joint venture between Evergreen and German power company Q-Cells, called Ever-Q. Without enough polysilicon in 2006, Evergreen probably would fall short of the Street's $100 million revenue estimate.
Evergreen's German plant has a third owner in Norway-based solar company Renewable Energy, or REC. REC holds a 15% stake in the joint-venture plant, a stake it attained in exchange for supplying 250 metric tons of polysilicon to Evergreen Solar in 2006. According to REC's Web site, the company produced a total of more than 1,000 tons of polysilicon in 2005, so Evergreen is only a small percentage of its overall polysilicon end market.
Simple math based on that information should ease investor concerns about Evergreen's ability to secure polysilicon for 2006. One key component to the calculation that works in Evergreen's favor is that the company is an efficient user of polysilicon, thanks to its patented Thin Ribbon process, which stretches polysilicon extremely thin during manufacturing.
This significant competitive advantage in a polysilicon-starved market is best represented numerically. According to analyst Sanjay Shrestha at First Albany Capital, Evergreen will use about 6 to 7 grams of polysilicon to generate 1 watt of solar capacity as the company transitions to thinner wafers for its panels. That compares to about 9 grams per watt for SunPower and an industry average close to 12 grams per watt.
Assuming Evergreen exits 2006 with an average gram-to-watt ratio of 6.5, the 250 metric tons of polysilicon from REC would be enough to generate 38 megawatts of solar power. With the company producing only 15 megwatts in the U.S. and its German facility up and running for only six months this year, the most we expect Evergreen to produce in 2006 is a total of 30 megawatts. That would leave it with a surplus of polysilicon.
Looking into 2007, investors have a right to question whether the company has a sufficient polysilicon supply.
While we don't have any official contract figures for 2007 from Evergreen yet, we do know the company has the aforementioned 250 metric tons on contract for 2006. Based on the Street's consensus revenue estimate of $200 million, the company is expected to produce about 65 megawatts of solar panels, using current market prices.
We expect that in 2007 the company will use 5.5 grams of polysilicon to generate 1 watt of solar capacity, which represents an improvement over 2006; the company's higher-efficiency German operation will represent a great percentage of total capacity. Multiplying 5.5 grams per watt by the industry price of $3 million per megawatt, I figure the company would be able to generate an output of only 45 megawatts from 250 metric tons of polysilicon. To reach the Street's consensus revenue estimate, Evergreen would require 350 metric tons of polysilicon.
In other words, Evergreen will be about 100 metric tons shy of the polysilicon it needs to operate at full capacity. However, unlike MEMC, which is an independent polysilicon supplier, REC has a vested interest in Evergreen's success via its 15% stake in the German facility.
We believe Evergreen will have an easy time locking REC into a longer deal for additional capacity. REC has ample polysilicon, and while the near-term economics of selling into the spot market are compelling, REC is better off building its ownership in the Ever-Q plant, a feat that will be accomplished only by offering additional polysilicon supplies.
Based on company forecasts and available public information, we believe Evergreen will be the most efficient solar operator by 2007, with gross margins approaching the low 30% range. And with the stock trading at just 4.7 times 2007 sales estimates, it's at a slight discount to peers such as SunPower, despite a significantly lower polysilicon risk.
William Gabrielski is a research analyst at TheStreet.com. In keeping with TSC's editorial policy, he doesn't own or short individual stocks. He also doesn't invest in hedge funds or other private investment partnerships. Under no circumstances does the information in this column represent a recommendation to buy or sell stocks. Gabrielski welcomes your feedback;
to send him an email.
Interested in more writings from William Gabrielski? Check out
TheStreet.com Breakout Stocks