September 15, 2006
Solar Power Wafer Demand
(c)
copyright View from Silicon Valley, 2006. All rights reserved.
Solar power is definitely "hot."
The hype came one day last week from Applied Materials, who does not seem to add much value to solar but apparently felt the "buzz" from a press release with solar in the title
would bump the stock. The same day, we got a blurb from Dow Corning proclaiming a technology to help bridge the "huge gap" between wafer demand and supply. We even got "The economy and
Silicon Valley" on news radio implying solar was driving up overall semiconductor and wafer demand.
You will not be
surprised to learn we still see solar as over-hyped. Since our May 13 piece on SunPower, SPWR is down -20% (vs. NASDAQ ~-4%) but still trades at 14 times sales and a forward P/E of 96x. But hey,
if there's a wafer squeeze maybe their "investors" will be proven right?
As reported last month, wafer area shipped is growing faster than semiconductor dollars. So maybe it's a perfect storm? Solar and semiconductor investors
are about to see an explosion in earnings due to a wafer squeeze?
Inquiring minds will be relieved to know we
uncovered some hard truths with just a little research (emphasis on little) on solar-vs.semi wafer demand.
It started
with an e-mail to the folks at SEMI:
"Do SEMI's figures on wafer area shipped also include wafers in-demand from solar
power products?"
SEMI answered our 5:25pm e-mail the next morning at 8:51am, with:
"The SEMI wafer shipment data
does not include solar wafers.
Check in with XXX at Sage Concepts (XXX@aol.com) and ask what he covers in the solar
wafer market."
Wow! A fast answer AND a recommendation of where we might find hard data.
We
e-mailed Sage Concepts the same morning:
"Do you have any hard data on wafers, or wafer area, shipped for solar power applications?
(Any details on how such data is collected would also be much appreciated.)"
Their response arrived 22 minutes later:
"A
report containing all of the information in which you are interested is listed on www.solarinsights.com ."
Who knew semiconductor research could be so easy? Why did neither of the full-time technology research professionals
who were so quick to criticize our work manage to try this themselves?
Excerpts from the free executive summary
include:
"While the semiconductor industry has created a rather financially unhealthy supply chain, the photovoltaic
industry by contrast has almost no supply chain. For 30 years the photovoltaic industry relied primarily
on scrap silicon from the semiconductor silicon production chain. Fortunately for the photovoltaic industry,
its new millennium growth spurt occurred at a time when the semiconductor polysilicon industry was in a period
of vast overcapacity. Thus a plentiful low-price supply of polysilicon was available for the period ranging
from the late 1990s to the year 2003.
"The year 2004 brought an end to this era of cheap plentiful polysilicon.
In 2004 semiconductor polysilicon demand increased 15% and photovoltaic polysilicon demand increased by 45%. These demand
surges created a dynamic where demand outstripped supply. With a polysilicon shortage thrust upon the year 2005,
the two polysilicon demanding industries are left to compete for the available material. The winner of
this struggle is obvious. The semiconductor industry, it owns many of the plants, and it has a financial
structure that allows it to outbid the photovoltaic industry on price.
"This scarcity has evidenced a basic flaw
in the development strategy for photovoltaic development. The photovoltaic industry has no viable supply
chain for the raw material that it has selected for accelerated growth. It basically has been lucky to
be able to develop to its present stage without its own supply chain.
"If it is to succeed in the continuance of its
growth pattern, the photovoltaic industry must develop its own polysilicon supply chain. This means either building
it’s own facilities or having it’s suppliers build new facilities that will be dedicated to supplying raw material
for the photovoltaic industry. The planning that goes into the construction of new polysilicon facilities indicates
that the price of this new polysilicon material will be higher than the photovoltaic industry
has been anticipating.
"In the near term the photovoltaic industry will not be able to meet
demand at even the volumes of the low-growth (21 % CAGR) business as usual scenario (2006 and 2007 time period).
By 2008 this situation should be alleviated to the point where the photovoltaic industry will be able to start to return
to a more robust growth scenario, but not approaching the more than 30% growth of the last 10 years."
You
can read the entire summary at: http://www.solarinsights.com/summary.htm.
Conclusion:
In short, the headlines have it backwards. Solar power
demand is not squeezing semi company wafer supply. Semi companies are squeezing out solar power demand.
As noted above, when semi and solar companies compete for wafers, semi companies "win."
Any semi company "victory" may be hollow, however, unless they can raise prices to offset increased
wafer costs. Since wafer area growing faster than dollar sales tells us dollars per wafer (DPW) is already
falling, few companies figure to be able to pass along increased wafer costs.
* * * *
The above is not intended
as advice to buy, sell or hold any stock, bond, real estate nor any other financial product or service. Buy and sell at your
own risk (just like we do.)
