One Guy's Investments

MEMC Electronic Materials (WFR)

Bought June 10, 2005 at $16.25

[note: see my other writeup on a semiconductor industry company, Formfactor, here]

MEMC Electronic Materials (WFR) has been a hot property lately – a week or two ago they were even yelling about it on CNBC. What put this bland little company on the hot seat this summer?

One word: polysilicon.

Polysilicon is what semiconductors are made out of, eventually. You see, MEMC is a supplier to the semiconductor industry, they make the actual highly polished, ultrathin wafers of precisely tuned silicon that Intel and the gang use to punch out the little tiny doohickeys that power and control our computers, phones, cars, and smart toasters.

Still awake? Well, it turns out that polysilicon has had a wild ride as a commodity during the past six months or so. At one point this Spring, spot prices jumped 25% in just a few days. And this isn’t one of those “China commodity” plays like steel or oil – no one in Shanghai is using polysilicon to build malls or factories, and it’s not being manipulated by the NYMEX crude oil futures traders.

There are a couple reasons for this increase in the price of polysilicon – first, semiconductor manufacturers are actually starting to make things again, recovering from the huge inventory surpluses that smushed their stock prices last fall.

And second, it turns out that the folks who make photovoltaic cells and similar equipment for solar power systems rely on polysilicon, too … so an industry that has been boom or bust on the volume of semiconductor manufacturing for years is now forced to serve two growing industries, with solar power growing very quickly thanks to some big subsidies from places like Germany, New Jersey and California.

The increased polysilicon demand is very real, for more information on what it is and why it's critical to this industry, read the briefing from SEMI (Semiconductor Industry Association) here, and a good article explaining the recent market trends and shortage from EE Times here

So why does that matter for MEMC Electronics?

Well, unlike the other major manufacturers of wafers, MEMC is almost completely vertically integrated. That’s right, they make their own polysilicon at a factory in Pasadena, Texas. And polysilicon has not gotten more expensive to make, according to the company, it’s just in higher demand. MEMC needs almost all of the polysilicon it makes, it seems, so it’s not as though they’re going to ride this boom in the spot market to boost their sales. They’re still focused on designing, manufacturing and selling wafers. So how do they get value out of this vertical integration thing?

One (more) word: margins.

The other wafer manufacturers in the world have to buy their polysilicon from someone else, so unless they wisely hedged against the rising price by buying their supply well in advance (remember Southwest doing that to great acclaim last year with jet fuel?), they’re going to have to raise their prices or suck up losses. MEMC can either raise their prices, too, and reap the benefits of higher margins, or they can cut prices and try to pick up some more customers – either way, that sounds like good news.

Now this might be overplayed – MEMC has already advanced significantly, up close to 50% in the last six months. But the trailing PE is under 13, and the analysts who follow the company guess that the coming year’s earnings will put the PE at about 11. None of the other major wafer manufacturers trade on the US markets, so it’s hard to compare, but next to any other company that relies on overall increased demand for chips this PE has to be at the bottom of the list.

Why did I buy it?

I believe in the silicon economy. I think that the increasing prevalence of semiconductor chips in consumer products is a continuing trend, which will mean greater overall demand for semiconductors. That greater demand means that the few manufacturers who supply them with high quality silicon wafers will have some pricing power and should see continued good business. I see this industry as a long term good bet, but I wasn't comfortable betting on a specific chipmaker. I think the semiconductor services and supply companies show more promise for those of us who can't predict who will design the best new chip (or in the case of Intel and AMD, win the next lawsuit). Applied Materials seems overpriced to me even at this point, and I generally prefer to invest in smaller companies that are either undiscovered or underappreciated or that have potential for more significant growth. The other stock I particularly like in this area is FormFactor (FORM), which supplies high-tech and patented testing devices for semi manufacturing.

WFR is cheap, and very competitive within its industry, which means the downside should be somewhat limited even if the industry goes through more turmoil (though I hope the turmoil has worked itself out of the system over the past couple of years).

There are good writeups in Smartmoney from last fall and in the most recent print issue (July? Not available online yet). This is a "cheap growth" story, in my opinion.

Why would I sell?

My primary concern with MEMC is that other competitors in Asia might succeed in building lower cost platforms and increasing their polysilicon supply to the point that MEMC's edge in their integrated business plan is moot. I don't expect that to happen, but the major competitors at the top of the list of wafer suppliers are building capacity for more polysilicon and new chips. I expect the demand to continue to keep up with and at times outpace supply, especially with the move to the larger, 300 mm wafer as the standard for the industry that is now underway.

It is likely that I would sell if MEMC becomes overvalued, just because the history of the industry makes one extra interested in taking profits, but I don't see that happening. I foresee continued solid performance, strong growth from this point in line with increased wafer demand -- but not a meteoric rise, or, one hopes, a meteoric fall.

The only reason I might sell at a loss or breakeven in the next three years is if a price war develops in the wafer market and MEMC's competitors are able to overcome MEMC's advantage in lower polysilicon prices -- but with price increases looking more likely this summer, I'm not very worried. Even with a price war, if MEMC's positioning looks solid I would consider buying more if it appears that they can maintain better margins than their competitors going forward.

Technorati tags: wfr, memc, , semiconductor