re: AMAT:
Haven't posted here before, but I've been studying AMAT (and intermittently investing in it) since 1996. I am "self-taught" re the semi-equip industry; that is not my area of professional training.
Some comments:
1. AMAT, like all the semi-equips, is a cyclical. Semi companies are either building new fabs with abandon, or cancelling equipment orders in droves. This is because no one is very successful in predicting chip demand 2-3 years into the future, and it takes that long to get a fab from planning to full production.
2. The semi-equip cycle hasn't matched the general economic cycle, because most chip use was PC-related. It is my belief that, with chips increasingly going into kids toys, kitchen appliances, cars, etc, that the semi-equip cycle will (going forward) begin to align with the general economic cycles.
3. Because of the above, it is possible to have dead money for years, if you pick the wrong time to buy semi-equips. AMAT was at 30 in mid-1995.....and also in mid-1998.
4. I Believe (notice the capital B) that Valuation Always Matters. The semi-equip cycle last bottomed in October 1998. They are in mid-cycle now. My best guess (emphasis on the word guess, because even the best CEOs and analysts in the industry have a poor record of predicting this) is that the cycle will top in 2002. Definitely not this year. Unlikely in 2001. 2003 is a possibility. So, being in mid-cycle, the stocks should be in the middle of their valuation range. Being cyclicals, P/S is the yardstick, not P/E. But valuations, beginning in Fall 1999, moved far above anything ever seen before in this industry. AMAT typically can be expected to trade in a P/S range of 1-6. At its peak this year, it hit a P/S of 15. Now a bit below that.
5. Another way of valuing AMAT is to use peak cycle EPS times expected PE then. My guesses are (for a 2002 cycle peak): 5$ X 25 =125. Look at the stock chart, and draw your own conclusions. I bought out-of-the-money longest-term calls in October 1998, and finished selling them in January 2000. I have no position in AMAT now, except that they are a component of the Nasdaq 100, which I am shorting.
6. Is AMAT a gorilla or king? The short answer is: I'm not sure, but I'm leaning toward gorilla. Within each cycle, the switching cost in huge. Once a semi company has committed to equipment from AMAT, it would be impossible to switch to NVLS (their closest competitor). The equipment is too complex, and all the equipment in a fab has to work seamlessly together. But, with each new cycle, the leading-edge semi companies get rid of most of the stuff they bought last cycle, and could switch to a different vendor. So, theoretically, Intel could decide, when the time comes to switch to copper from aluminum (or to 300mm from 200mm), that they will buy from NVLS, not AMAT. However, in practice, over the last 20 years of cycles (which usually last 2-4 years each), there has been little switching away from AMAT. There has never been competition on price in this industry. AMAT's track record argues for the gorilla label.
7. AMAT seems to be doing to the semi-equip industry what CSCO did to the network equip industry. They are about 10 years behind CSCO, but on the same path. With each cycle, AMAT gobbles up another sub-sector of the industry. It buys weak small companies at the cycle lows, integrates their products with existing AMAT equipment, and has gradually built up an end-to-end solution, unrivalled in the industry.
8. There has been a gradual shift, with each successive cycle, for the semi-equip companies to assume more of the burden of developing, servicing, and upgrading the equipment in a fab. The semi companies used to do this, but they have moved slowly toward long-term service contracts with the semi-equip companies, to outsource this. The growth of fab-less semi companies, and the maturing of the semi-equip companies, means this trend will probably accelerate. It will make switching costs even higher.
9. the above is my personal opinion, which numerous people on the AMAT thread would violently disagree with (especially the part about valuation). Caveat emptor. |
