MEMC (WFR) has been in the news lately thanks to its recent quarterly report, which was — just as expected — sending the stock down nearly 20%.
It was thanks to the researchers at MIT that we discovered something special about silicon described in our February 6, 2006 report. We highlighted “fast silicon” and the obvious investment play MEMC, which at the time was changing hands at $28/share. WFR dutifully climbed to $67 over the subsequent 14 months.
Admittedly most of the enthusiasm from MEMC is as a supplier to the booming market for solar energy. However, our thesis was and is based on continued secular increases in demand for silicon and the increasing value-add possible at the wafer level.
Silicon is by far the leading material for making semiconductors. It has allowed phenomenal increases in density and offers very low power dissipation.
However, silicon has one significant drawback. Turns out that electrons move fairly slowly in silicon vs. other materials such as gallium arsenide. Engineers have been forced to make a trade-off between high-speed/high power consumption materials and slower, more efficient silicon.
Obviously there is strong desire to have the best of both worlds — high speed and lower power dissipation.
By changing the lattice structure of silicon by a fraction of an angstrom, electron mobility is improved tenfold. The basic fact is that by turning raw silicon wafers into highly engineered substrates, great advantages can be created and consequently increase the value added by a supplier like MEMC.
Most view MEMC as a commodity play on what has been a tight market for silicon supply. It’s true that these conditions create an even better short-term story for MEMC, but to us the longer-term opportunities in more engineered products is far more interesting.
There are big competitors out there. That market prices for silicon will be volatile. And that other, even more advanced technologies, could replace silicon in the very long-term.
However, we also know the opportunities in silicon for MEMC are at least as good as those in other parts of the semiconductor value chain with less competition and real switching costs in the more complex segments of the market.
WFR 1-yr chart: