INTC vs. AMAT:
I've made several long post in the past on this subject, and I think it's timely to do so again. All prices are adjusted for splits, and therefore comparable.
1. INTC profits (and smoothed stock price trend) are a first order derivative of chip demand. AMAT profits (and smoothed stock price, and BTB) is a second order derivative, and as such is inherently more volatile.
2. Therefore, when they go out of favor, INTC tends to bounce repeatedly off a flat support line, while AMAT steadily sinks as long as the news is bad, bottoming at 1/3 to 1/2 it's previous highs. Until the numbers start looking better, all rallies will be false rallies. We've seen a lot of false rallies in both these stocks since October 1997. When the momentum guessers get tired of losing money by buying at the peak of each false rally, then we'll see capitulation. Support lines for INTC are more reliable than for AMAT.
3. The long-term return, measured by percentage return over time periods of at least three years, is identical. This makes sense, since they both dominate rapidly growing, tightly linked non-commodity industries. AMAT is today at three times it's 1996 low (11 to 33). So is INTC (25 to 75). Over 5 years, from mid-93 to today, both have about quintupled (I just love using that word).
4. And, no, I don't believe the market for Intel's chips is becoming commoditized. I think Merced will be bigger than Pentium. If cheap non-Intel chips in sub-500 PCs is the wave of the future, then why is Dell (who uses exclusively expensive Intel chips and ignores the sub-1000 segment) the only boxmaker that is gaining market share and increasing profits? CPUs ain't diapers, and never will be.
5. Given the above, the long-term returns will be the same if you dollar-cost average, or have average luck at timing. If you are a momentum guesser, or follow Mr. Market, then you'll have fewer opportunities to lose money if you buy INTC. The only way that long-term returns are higher with AMAT than with INTC is if you time it right. That means buying near the bottom, and selling near the top or never. That's very hard to do. If you bought AMAT at the high in mid-1995 (30), you're barely above break-even today.
6. I believe that sometime in 1998 or 1999, we will see a 20-30% correction in the S&P 500, caused by a slow-down in earnings and/or wage inflation causing a series of Fed hikes. I don't know when, but I don't see any way to avoid one or the other. That kind of correction pulls everything down. Therefore, I'm positioning myself to lose the least money when the S&P hits 800. I'm hoping that INTC's resistance line at 70, which has held through a steady onslaught of bad news for the last 12 months, will hold. The worst downside I see is 60, which would be only a 20% decline from my 72 entry point (next week I'll get that price).
7. The downside for AMAT? The low P/S ratio for each year is: 1.0 (in 1993), 1.6, 0.9, 0.9, 1.3, and 2.0 (so far in 1998). Using trailing 12-month sales (forward 12-month sales won't be any better) P/S=1 at a stock price of 13. That would be a stock price just above the 1996 low, which is also the 1994 support line. That's the potential downside. Any other analysis is wishful thinking. Anyone who is (or will be) buying AMAT on margin, needs to make sure they won't have a margin call if the stock hits 13. I'm not saying it will, I'm just saying that's the potential downside.
8. The only way to make more money on AMAT than on INTC long-term is to buy AMAT when it is at or below a P/S of 2. Fair valuation is P/S of about 2 to 3.5. Therefore, I've changed my strategy. I won't be buying AMAT, or ZPJAH, till the stock hits 26-28 (P/S=2). Above that price, I think the long-term returns on AMAT won't be better than with INTC, and the potential downside will be much larger.
9. At 32, AMAT is still overvalued. The ugly fact is that EPS peaked in 1996 at 1.68, and won't see that level again till 2000. This is turning out to be an agonizingly prolonged w-shaped downturn. A "W" is composed of two "V" patterns. We are currently on the mid-point of the down-leg of the second "V". In mid-1997, earnings estimates for 1999 were about 2.70-3.00 (if my memory is right). Now, 1999 estimates are about half that. Since forward earnings guesses (the word "estimate" gives the numbers too much dignity) have halved, the stock should be at half the 1997 peak of absurd euphoria (half of 54, at best).
10. Next up-cycle, I'm going to start selling in increments when the P/S=3.5 Or, I might just join lester and never sell. If I convert my LEAPs to stock in Jan 2001, and hold till my 6-year old daughter goes to medical school, I ought to do OK (VVBG). The longer and more agonizing this downturn is, the longer and more profitable the next (real) upcycle will be. I'm still uncertain short-term, and still very bullish very long-term. |
