I thought this would be helpful- anyone interested can get a this news letter upon request from- JArena@aol.com
To: IQBAL LATIF
From: J.D. Kelley Sunday, Aug 23 1998 7:40PM ET
Ike,
Attached is a letter from someone like yourself that I follow very closely. He has been very good on INTC, TXN CSCO MSFT adn the Sox.
A somewhat bearish opinion but backed with substantive thought, If you are interested I can ask that you be added to the mailing list.
J.D.
Subj: The High Tech Arena 8/21/98
Date: 8/21/98 12:15:08 PM Central Daylight Time
From: JArena3773
Whither Intel?
By Joe Arena
How many times have we all heard the hackneyed cliche, don't fall in love with a stock? Unfortunately, even many analysts on Wall St. make the mistake of extrapolating past earnings growth trends and expecting them to continue. Such is the case with Intel. Although Tom Kurlak's comments this week were arguably made with Merrill Lynch's best interests in mind, directionally his analysis should be considered valid. Although Kurlak's earnings estimates for this year and next discount the worst case scenario for Intel, it should be increasingly evident to all investors that the paradigm has changed. Those who are expecting Intel's earnings to grow at the same 32% compounded annual growth rate of the last ten years are going to be severely disappointed.
The days of the $2500 PC are gone forever. The double edged sword of Moore's Law will now become INTC's worst nightmare. Certainly, killer applications such as 3D, virtual reality, voice recognition, and artificial intelligence will drive the need for greater microprocessor power. But when these applications become mainstream, Moore's Law will have evolved to the extent that most retail and corporate customers will be upgrading to sub $1000 PC's to run them. We are already seeing evidence of that in terms of the recent shortage of 266 MHz and 300 MHz chips. This shortage provides the best indication that Intels business model needs to change. In the past, Intel could easily phase out production of low end chips and shift the market to the high end. The insatiable demand for sub $1000 PC's is now making that impossible. Moreover, the fact that Advanced Micro Devices and Cyrix combined now account for 60% of all sub $1000 PC's (and 38% of all retail desktop sales)is a concern which did not exist under the old paradigm. Being one generation ahead of AMD is no longer an advantage in the retail channel, and there is a growing possibility that AMD may be able to leverage their growing credibility in the corporate space. AMD has already secured design wins with all the major box makers except for Dell. And the probability that the sub $1000 PC will pervade the corporate space is also increasing. Does this mean that AMD is going to make any money in this business? Certainly not, but it does increase the possibility of AMD being a takeover candidate. Does this mean that Intel will not continue to dominate the business? Of course not, but the scenario of Intel evolving as a flat revenue cash cow becomes more likely.
With the Internet rapidly becoming the next killer application, it is causing a shift away from the PC centric world. Consider this as an example. By the end of this year, there will be PC's on the market in the $500 range that can do anything the majority of users want to do. However, household penetration is still languishing around 44%. How do you reach the remainder of those households, which is where the mass market and concomitant unit volume is?
The answer is twofold. First, you don't, at least not with the PC. It is becoming more apparent that demand for PC's is relatively inelastic. Second, when e-commerce, e-mail, and the integration of voice, video and data evolve to the extent that Joe/Jane Six Pack have a need to access the Internet, they will probably not be doing it on a PC.
In five years, more people will be accessing the Internet via Web TV, Web-enabled telephones, Web-enabled Personal Digital Assistants (PDA's), Web-enabled video game consoles, and enterprise network computers. MSFT is the one major player who appears right now to be the most focused on this coming paradigm shift.
Not only do the odds favor MSFT pushing Windows CE into many of these Internet appliances, but it is also probable that they will develop another platform to ensure
that they have a stake in all electronic appliances with Internet access capabilities. Conversely, INTC is still betting the future of the company on a PC centric world, and right now the StrongARM chip provides the only means for them to hedge their bets against the reduced importance of the PC. Intel has recently shown a greater interest in building this business, but it does not offer the high margins that Intel has become accustomed to.
Intel is counting on growth of the Internet to increase demand for servers to the extent that it more than offsets the decline in Asp's as well as unit volume in the PC business. Their strategic plan calls for 1 billion users globally to be connected to the Internet in five years, which should drive demand for 200-300 million servers. While this plan is definitely in the realm of reality, it remains to be seen whether Moore's Law, declining
Asp's, the commoditization of the CPU, the growing dominance of embedded microprocessors, and the shift to Internet appliances will largely negate this incremental volume and improved product mix.
Despite the total cost of ownership issues being addressed by INTC and MSFT right now, the fact is that corporate IT managers are becoming less receptive to the constant upgrade cycles. It is likely that upgrade cycles will move from 3 years to 5 years, thus increasing the cyclical nature of INTC's business.
Thus, the delay of Merced becomes even more significant, as INTC's best bet right now to offset declining margins is their quest to dominate enterprise computing in the same manner as they have the desktop. Furthermore, it is becoming more likely each day that Merced will not even make it as a viable product, but will be used simply as a development vehicle for the next generation, code named McKinley. There is no doubt that Intel will receive tremendous support for their IA-64 architecture, whether it be Merced or McKinley. However, the continued delays serve to make the earnings visibility picture cloudier over a much longer period of time.
Does all this mean that Intel should no longer be included as a core holding in a diversified, long term portfolio. No, but it does mean that investors should lower their expectations, at least over the next two years.
**********************************************************************************************
SEMICONDUCTOR UPDATE
As I had indicated several weeks ago, the rally in the SOX was nothing more than a fake out break out, with sector rotation simply driving the beaten down semi stocks up for a dead cat bounce. I continue to assert that we are at a bottom in the industry, but it is probable that the bottom will last a bit longer. Witness the recent semiconductor equipment book to bill numbers reported for July, which came in at 0.69. This dismal number is another indication of the blood that is currently in the streets, as it marks the lowest point in the current downturn. Another indication that we are at the low point is the incredible level of debt being reported by the major semiconductor producers in Asia. Consider that LG Semicon recently reported that it's total debt level equaled more than 500% of it's equity, and that it's foreign currency debt was $3billion. Hyundai also reported that their debt is $8.4 billion, which is equal to 930% of equity. However, there are signs that at least a tradable rally should occur soon in the semiconductor and semiconductor equipment stocks. DRAM prices have begun to stabilize, although this is no indication of how they will trend the remainder of the year. It is significant to note this stabilization, however, since prices have been in a decline since the beginning of the year. In addition, year to year comparisons could very well be favorable starting with October, since these 1997 numbers are very weak. This could really provide the impetus for a strong rally in the SOX that could be sustained if evidence of an upturn continues to materialize.
***************************************************************************************************
TRADING UPDATE
With the market down big today, the CSCO aug 90 puts that were shorted at 2 and 1 5/8 are set to expire worthless at the close today. With CSCO showing excellent relative strength (with the anticipation of the stock split, deal with AOL, and another acquisition today), and support in the 95 area, I am waiting to short the 95 sept puts on Monday. If we open down on Monday, and CSCO stays above 100, will take at least a half position.
Continue holding the MSFT 01 100 calls, and the CSCO 01 85 calls, and continue short the TXN 00 45 puts, with premiums invested in the CSCO 01 85 calls.
DISCLAIMER: THIS E-MAIL DOES NOT CONSTITUTE A RECOMMENDATION TO BUY OR SELL SECURITIES, AND IS FOR INFORMATIONAL PURPOSES ONLY.
HOUSEKEEPING ISSUES: Note that I am now calling my newsletter The High Tech Arena. Please feel free to provide me with any names of your associates who would be interested in receiving my newsletter, as I am looking to build my readership. Thanks to all for your continued support and feedback. Note to long time readers: Some of the points made in the INTC article you have seen on a prior post to the INTC board, but I have updated and elaborated upon them in this e-mail.
Joe Arena
|
