Sunpower: Good Planning on PolySilicon

SunPower is the technology leader in solar derived energy products for both residential and commercial applications. The company offers solar cells, solar panels, and inverter equipment. These high-efficiency solar cells and panels generate up to 50 percent more power per unit area than conventional solar technologies, making SunPower the technical leader in its industry.

We have covered SunPower for a few months now. I interviewed Tom Werner, CEO of SunPower back in April, and wrote my first analysis piece on the company (Tapping Into The Sun). I also bought a chunk of SunPower stock right around then, and am happy to report that the investment is already up over 70% in 6 months.

The question today is not if markets are big enough to sustain SunPower and competitors’ phenomenal growth. Rather, it is whether there is enough raw material to continue making solar products to meet the expected demand.

In December 2006, a shortage loomed and was considered a production threat. Polysilicon is the bread and butter raw material for solar panel production. Costs for the material have doubled since 2004 to now $70 per kilogram on longterm contracts. Spot contracts are much higher at $200 per kilogram. Competitors of SunPower such as JA Solar, Suntech Power, and Canadian Solar have dumped much of their IPO funds into securing the raw material. The cost now makes up 40 to 45% of the cost of goods on a solar cell, so the increase is no small matter for SunPower’s cash flow.

Whether this situation continues depends on three questions: 1) how fast will the solar markets in the U.S., Europe, and Japan continue to increase? 2) How fast will solar panel prices drop compared to the price of electricity, and 3) Will something else come along as a substitute to polysilicon as a solar material?

The first two questions seem to be currently addressed by supply and demand. “So much new poly is coming online that it is now becoming a question of when, not if poly will go into oversupply; and analysis suggests that first quarter 2008 could be when supply growth inflects high enough to overcome demand – Satya Kumar, Credit Suisse.”

Demand is also expected to jump by 3 times to 99,500 metric tons by 2010. A substitute, however, is not likely any time soon. Sunpower itself simply noted in April 2007 that its suppliers, DC Chemical and M. Setek, are making steady progress with their new polysilicon manufacturing facilities (DC will produce over 4 years for SunPower at a cost of $250 million beginning in 2008).

However, only four weeks earlier the company was quoted to say that it expects polysilicon prices to increase further in 2007 and polysilicon demand to continue to outstrip supply throughout 2007 and potentially for a longer period. So far, however, there is not much indication of any polysilicon sourcing problem in SunPower’s shop.

SunPower is well positioned to benefit from increasing solar demand as a result of its broad and differentiated product offering. Its vertical integration approach to supply problems continues to provide a defense to polysilicon inflation. In addition to the DC Chemical contract, Sunpower also agreed in July 2007 to purchase polysilicon from Hemlock Semiconductor Corporation from 2010 through 2019 (Hemlock is a JV between Dow Corning, Mitsubishi Materials, and Shin-Etsu Chemicals), showing ample advanced planning. And given demand, it is likely a larger general supply will come online by 2008-09. So while there may be a shortage, SunPower is again already ahead of the power curve securing supplies.

SunPower will announce Third Quarter results on October 18, 2007, and while its stock is probably somewhat over-bought right now, it does seem to be one of those stocks to hold onto and go along for a longer term ride.