OT Sort of - This is a freebee high tech newsletter I've subscribed to for a while. I find the content and the investment strategies (option strategies) to be quite good. (I'll post this over in the TI forum as well if not already there.) FWIW.
Semiconductor Update
The High Tech Arena, 1/24/99
By Joe Arena Editor
We have asserted for the past 18 months that the personal computer's importance in terms of driving the semiconductor business will begin to wane in the next few years, and this phenomenon will become more apparent to investors in 1999. As numerous wireless products (e.g., digital cell phones) and PC's costing less than $500 become pervasive in the currently untapped segments of the market, digital connectivity will become the most significant catalyst relative to the growth of the semiconductor industry. As the cost of PC's continues to decline, there will also be less features available driving the cost of these systems down.
Thus, the need to connect to a network for features and applications will become crucial. Obviously, the Internet will play a key role here, and IC (integrated circuit) technology will be instrumental in terms of enabling more and more data to flow through a wire.
Similar to Moore's Law, (i.e., the performance of a microprocessor doubles every 18 months while the price remains constant) IC technology will result in quadrupling the performance of network connections every two years. In other words, an image that now takes 8 seconds to download off the Internet will take only two seconds by 2001. Eight years from now, it may take only 1/30 of a second to download this image. This performance is equal to the same rate of full motion video. When this occurs, the Internet will become ubiquitous for virtually every demographic segment of the population, especially in terms of e-commerce.
No company is better positioned to take advantage of this trend than Texas Instruments, which has virtually bet the entire company on the DSP (digital signal processors) business. These DSP chips will provide the power to drive digital modems, computer networks, and digital cell phones. What is especially noteworthy here is that cell phones and other hand held devices will be used by many consumers who currently do not own a PC to access the Internet.
When it becomes a necessity for this segment of the population to access the Internet, this market will explode. Moreover, the DSP chips which run cell phones are much higher performance chips, and consequently deliver much higher margins. This year may very well mark an inflection point in the chip business in which many chip companies begin to limit their exposure to the commodity PC business and focus their resources on the digital connectivity segment of the chip business.
Certainly, at some point DSP's will become a commodity, but with all the new applications emerging for these chips, it is doubtful that this will occur anytime soon. Moreover, the prolific growth of this business makes it stand out as the one area of the semiconductor industry which has some insulation from the inherent cyclicality. To underscore this point, consider that Texas Instruments DSP business grew about 25% in the second half of 1998 versus the same period in 1997. Contrast this with the overall decline in the semiconductor industry of about 10% in 1998, and it is easy to project that this trend will continue, at least in the intermediate term future.
In addition, there are other new products such as DVD (digital versatile disk) players which will consume enormous amounts of microprocessors. The market for DVD drives for PC applications is expected to quadruple in 1999 to 8 million units. Although DVD players will one day make VCR's obsolete we are not even close to that happening. This market will remain small in 1999, (i.e., $250 million) despite a huge percentage increase off a small base in 1998. But the point to be made here is that five years from now, extrapolating the current growth represents a major opportunity.
How many of us purchased a Furby for our kids this Christmas? (if you could even locate one on a retail shelf) Unknown to many, this sensational electronic talking "virtual pet" comes equipped with a mixed signal, voice processing chip made by, you guessed it, Texas Instruments. The bottom line here is that the day is coming when the number of chips in a typical household will reside for the most part in devices other than the PC. All major appliances in the house will include microcontrollers, embedded microprocessors, and digital signal processors.
On average, currently there are anywhere from 30 to 50 chips in the average home. This number is expected to increase tenfold over the next five years. We believe there is no company better positioned to exploit this trend than Texas Instruments, as there is no company who has placed a larger bet on this trend materializing.
Until the commoditization of the DSP chip occurs, an investment in Texas Instruments also satisfies a key investment tenet of ours. (i.e. buy the company that is selling the bullets, not fighting the war) This risk of commoditization/competition is the reason why Texas Instruments (TXN) is only ranked number 10 on our recent list of top ten technology stocks to own in 1999, as is the recent run up in the stock.
This is also why in May 1998, we chose to short TXN 2000 Leaps puts and use the proceeds to purchase Cisco Systems 2001 56.6 Leaps calls. (as opposed to simply buying TXN or going long TXN Leaps calls) This trade gives us a unique way to play the digital connectivity theme, while at the same time concentrating our assets in only our best ideas.
For those who have been following this trade since its inception, it has now enabled us to turn $0 into $57,875 for each 10 contracts of TXN Leaps puts shorted. Regarding the overall semiconductor business, we feel that the easy money has been made for the time being, and that the more than doubling of the SOX Index since October has discounted a great deal of the recovery which will occur in 1999.
We believe that we are at the beginning of what will prove to be the greatest cyclical upturn in history for semiconductors, but we have come too far too fast and need to consolidate these gains. Sentiment in the industry can best be characterized as cautiously optimistic, and projections vary tremendously, indicating the widespread uncertainty. The main reason why the cyclical upturn now being discounted by the SOX will be of unprecedented proportions has to do with the dramatic cutbacks in capital spending that have occurred. For example, in Japan, capital spending for new fabs (i.e., semiconductor manufacturing facility) has been eliminated until the start of their fiscal year in April.
However, all indications point to the Japanese actually spending the same or even less in 1999 than 1998. This and other similar phenomenon are what have historically created the incredible boom and bust cycles in the semiconductor business, and it will be no different this time around. Semiconductor manufacturers will not commit capital to upgrade and build new fabs until it becomes irrefutable that the upturn is sustainable. By then, demand will outstrip supply by an inordinate margin, causing prices to rise, new capacity to ramp up, supply and demand equilibrium, and then overcapacity occurring again. We believe that this cycle will run until 2002-2003 before overcapacity creates another downturn.
Additionally, our projection is for the semiconductor business to increase 10% in 1999, from $126 billion in 1998 to about $139 billion. Of course, the wildcard to this forecast remains global economic conditions, particularly in Asia. If key Asian economies begin to rebound strongly toward the latter part of this year, we would be willing to concede that this forecast may prove somewhat conservative.
As always, the key to making money investing in the semiconductor stocks is either to time the cycles correctly, or buy when there is blood in the streets, and wait patiently for the market to anticipate the next upturn. However, we believe it is impossible to time the market, and waiting patiently usually forces investors to incur substantial opportunity costs in the midst of a secular bull market in other areas of technology. Therefore, we remain predisposed to having conservative investors desiring a diversified portfolio holding only the best quality semiconductor companies for the long term, such as Intel and Texas Instruments.
For those aggressive investors with high risk tolerance desiring to concentrate their assets in just a few stocks, we would implement a different strategy. Specifically, we favor selling puts on these two companies to generate income which can be put to work building long term positions in our top three or four stock picks.
We are still looking for new subscribers, and welcome all current subscribers to pass The High Tech Arena on to friends and associates.
DISCLAIMER: The information herein has been obtained from sources which are believed to be reliable, but there are no guarantees as to its accuracy or completeness. Neither the information nor any opinion expressed constitutes a solicitation for the purchase or sale of any security.
THE HIGH TECH ARENA
Joe Arena Editor
JRArena@aol.com
|
