To print article, click
Print on your browser's File menu.
Go back
Go back
Posted 2/1/2001
Related Resources
Keep track of Victor Niederhoffer and Laurel Kenner's picks on their Recommendations page.
Related Site
insiderSCORES.com
Details
 |
|
The Speculator
When the chips are down, buy, buy, buy
Profit from popular misperceptions about the semiconductor industry. Over the long haul, blessed by the magic of compounding and human ingenuity, the payoff could be huge.
By Victor Niederhoffer and Laurel Kenner
Perceptions are, by their very nature, selective.
-- Scott Plous, "The Psychology of Judgment and Decision Making"
Three biases in decision-making under uncertainty make us believe that the semiconductor industry might provide some good opportunities for investment.
 Money Plus. Easy online tools and free Bill Pay, too. |
1. Decision-makers underestimate the power of compounding.
People just don't pay enough attention to the magic of compounding. Asked to estimate the money to be made in 25 years with a 10% return rate vs. a 5% rate, many investors would say double.
In fact, the difference is 983% vs. 239%. That is, a dollar invested at 10% would bring in almost $11; invested at 5%, it would turn into more than $3.
As time goes on, the discrepancy becomes even more pronounced: $117 vs. $11 in 50 years. In 100 years, the return is $13,780 vs. $132.
As economist Paul Romer puts it, "People are reasonably good at forming estimates based on addition. But for operations such as compounding that depend on repeated multiplication, we systematically underestimate how fast things grow."
When we use such long horizons, people often object that nobody is around to invest that long. Our answer is that today's children and tomorrow's grandchildren will be around; and with biotechnology, our younger readers might well anticipate a quantum increase in their own life spans.
We didn't draw the 10% number out of thin air. Random baskets of stocks held for long periods of time almost invariably lead to returns close to 10% a year. That is why we have always advised against short-selling. A rising stock market floats all stocks and has a way of smoothing over weaknesses that might appear in a company's balance sheet, management and industry.
The dramatic difference in the end value of investments with large growth rates and minimal rates goes to the heart of investing. If investors misunderstand how fast 10% will compound, imagine the underestimate involved with a 25% a year rate. But this is, as far as we can see, the consensus estimate for the growth of semiconductor sales over the next five years.
The failure to realize that 25% will double in three years and quadruple in six years may have led people to value semiconductor stocks too cheaply.
2. Decision-makers underestimate the pace of invention.
The ability of humans to invent is much greater than commonly perceived. The old-hearted always think the best days are behind. U.S. Patent Office Commissioner Charles H. Duell said in 1899, "Everything that can be invented has been invented."
And yet, waves of innovation keep altering our life and work.
The first major industrial revolution came at the end of the 18th century as developments in steam power and iron making gradually spread to all sectors of the economy.
The second again made the experts look bad, as electricity, automobiles and plastics dynamized the economy.
The third revolution is still going strong. It is based on the invention and widespread use of semiconductors. This revolution also will be much bigger than predicted, because on its shoulders stands a fourth developing revolution, the extension of life through bioengineering.
By purchasing stocks in the semiconductor industry, the investor can take advantage of the current tendency to underestimate the improvements and diffusion of semiconductors into current and new products that this third scientific revolution will bring about.
3. Decision-makers place too much weight on the recent past and short-term future.
The tendency to place too much reliance on things that happened recently is known as the "recency effect." If a coin falls on heads a few times in a row, people tend to believe that heads are much more likely than they should.
Same goes for a stock that has gone up a few days in a row. People expect the gains to continue and tend to disregard the influence of chance and ephemera in their estimates. Similarly, if a company just reported a down quarter or gave a warning, many people consider it the end of the world.
One example of the recency effect came after the Oct. 19, 1987 crash. The big decline came on a Monday and was preceded by a disastrous Friday. People tended to remember the bad Friday-Monday combinations, and for years whenever stocks declined violently on Friday, they would rush in to sell based on their memory of the recent vivid event. They lost more by selling on Fridays over the next 10 years than the total decline of the crash. Monday became the most bullish day of the week.
If investors would just realize that for random series, the future is independent of the past -- that no matter how many heads you've thrown in a row, the chance of a head on the next toss is 50-50 -- they would be on the road to making better decisions.
Undue bearishness
While the recency effect may be good for helping you remember where your key or book is, it is not overly helpful in the market. And at present, the effect is working to make people more bearish on semis than they should. Right now, everyone remembers the 23% slide by PMC-Sierra (PMCS, news, msgs) last Friday and the 60% drop in Intel (INTC, news, msgs) in the last four months of 2000.
And therein lies the opportunity. All industry forecasts that we have seen call for 25%-a-year growth over the next three to five years, after a slow first half this year. Have we reached a stage where investors are so short-sighted that they will not buy a company that is projected to grow by 10% in the next one or two quarters and 25% thereafter?
Let's assume that a semiconductor company selling at a price-to-earnings (P/E) multiple of 30 today has earnings growth of 25% for the next 12 years. At the end of the 12 years, it sells for a multiple of 15. The earnings in 12 years would be 16 times the current levels, and the stock would show an annualized return of 19%.
Some estimated rates of return based on varying assumptions of initial P/E and ending P/E are shown below, for a 12-year investment horizon.
12-year growth at 25% a year
| 2001 P/E | 2013 P/E | Annualized return |
| 15 | 15 | 25% |
| 30 | 15 | 19% |
| 60 | 15 | 13% |
For guidance on which stocks to buy, we turned, as is our wont, to the source: the officers and directors of the companies themselves, who provide the bulk of the information relied on by the numerous analysts covering the industry.
As noted by Kevin Schwenger in an article at the insiderSCORES Web site, insider buying in the semiconductor industry has picked up in recent months. We found 10 semiconductor stocks with net insider buying over November, December and January.
Cheap chips with insider buying
| Company | % Chg. in 2000 | % Chg. in 2001 | Jan. 30 price | 52-wk high | Net insider buys (Shrs) | Insider price | P/E (est.) |
| Lam Research (LRCX, news, msgs) | -61 | 76 | 25 9/16 | 56 13/16 | 10,000 | 16.68 | 14 |
| Amkor (AMKR, news, msgs) | -45 | 49 | 23 1/3 | 65 5/16 | 20,000 | 15.03 | 15 |
| Amphenol (APH, news, msgs) | 24 | 42 | 48.53 | 70.38 | 6,000 | 34.50-41.94 | 15 |
| Advanced Energy (AEIS, news, msgs) | -54 | 37 | 30 3/4 | 77 7/16 | 276,000 | 16.32-18.10 | 16 |
| Electro Scientific (ESIO, news, msgs) | -23 | 29 | 36 3/16 | 69 1/8 | 700 | 23 | 10 |
| Fairchild Semiconductor (FCS, news, msgs) | -51 | 26 | 18.24 | 49.5 | 2000 | 21.56 | 11 |
| Viasystems Group (VG, news, msgs) | -49 | 20 | 10 | 23 | 345,500 | 6.75-7.55 | 11 |
| Motorola (MOT, news, msgs) | -59 | 14 | 23.01 | 61.54 | 7,250 | 20.38-22.50 | 30 |
| Teradyne (TER, news, msgs) | -44 | 12 | 41.86 | 115.44 | 2,000 | 33.56 | 12 |
| Transmeta (TMTA, news, msgs) | 12 | 11 | 26 | 50 7/8 | 106,700 | 21 | n/a |
Of those, we are recommending Motorola (MOT, news, msgs), Transmeta (TMTA, news, msgs), Teradyne (TER, news, msgs) and Viasystems Group (VG, news, msgs) because their gains this year have been the lowest.
Simple decision
Vic's checkers teacher, Tom Wiswell, liked to stop in the middle of a game, look around at the spectators and say, "I'm in way over my head." After general amusement among the audience (Tom was the undefeated world champion for 25 years), he'd pause and add, "I better simplify."
We are not world-champion semiconductor analysts. Far from it. While we are electronics hobbyists -- Vic fools around with transistors in feedback circuits a few times a week -- chances are that the average reader knows more about the semiconductor industry than we. We know we're in over our head here.
Why, then, do we have the temerity to opine? We have simplified. We believe that we may have found a relatively uncrowded niche where a combination of psychological factors might lead to a reasonable return. And we believe that we are playing on a field where the intrinsic growth of the industry will tend to shield us and others from undue naiveté.
We were equally ignorant the first time we chose to focus on an industry. In December, using similar reasoning to the above along with some ecological models, we recommended purchase of a group of battered Internet stocks that were the subject of insider buying. They have subsequently gone up an average of well over 100% this year.
With so many beaten-down groups going through the roof in 2001, we believe that semiconductors are still relatively depressed. But beware, our ignorance should be taken into consideration. The archetypal semiconductor analyst, of whom there are hundreds, could undoubtedly pick a thousand holes in our argument.
Time to buy
Nevertheless, as of the open today, we are adding these stocks to our recommended portfolio. The beginning of a month is always a good time to buy.
Recent Articles
• 5 battle-tested rules for a mystifying market, 1/25/01
• 3 solid stocks the market rebound missed, 1/18/00
• 24 battered small-caps the insiders are buying, 1/11/00
more...
We recommend the sale of Apple, which is up some 50% from where we recommended it Dec. 28, on the grounds that it has become a little too hot.
Along these lines, whenever we write a column like this we are sure to receive a diatribe from a reader who notes that we don't own the stocks we are recommending, along the lines of "Put your money where your mouth is or shut up." In actuality, we do hold a portfolio of about 25 stocks, most of them concentrated in individual stocks we have previously recommended.
Biases and errors are unfortunately endemic to the investment process. While we are well aware of the limitations of our own decision-making, at least we have attempted to profit from the errors of others.
At the time of publication, Victor Niederhoffer and Laurel Kenner did not own or control shares in any of the equities mentioned in this column.
MSN Money's editorial goal is to provide a forum for personal finance and investment ideas. Our articles, columns, message board posts and other features should not be construed as investment advice, nor does their appearance imply an endorsement by Microsoft of any specific security or trading strategy. An investor's best course of action must be based on individual circumstances.