>Fact 1. IT spending in the US is at a record high. >Spending on IT must be paid for. As the dot com bubble has >burst and no one will lend to shite companies anymore. IT >spending (ie semiconductors) must be paid for by profits. >Only problem is....there aren't any.
True, the majority of dot coms will die. There will be massive consolidation and few survivors. But those that do survive will have huge volumes of online sales of goods and services. Do you really think 5 years from now fewer $$$ will flow thru online transactions? To keep up with increased order flow, the survivors will have to invest heavily in servers and network infrastructure.
>Fact 2. Taiwan, Korea and UK have seen massive >disappointment in cell phone sales. Korea saw a 770,000 >REDUCTION in subscribers in June. Forget growth. BT >Cellnet in the UK got 125,000 WAP subs in 2Q00 but it was >forecasting 500,000.
Korean cell phone rates have fallen off recently because subsidies by the government have dried up. For now WAP is a boondoggle because of slow access speeds, limited web sites, and first generation equipment. Are you really serious, though, in suggesting that cell phone sales have peaked and will now dwindle to nothing? Even in countries with almost universal adoption, users trade in their models every 3 years or less. What about China, Africa, and South America where cell ownership rates are currently 3%? Do you think these people will ignore cellular technology and use tin cans with string?
>Fact 3. Volume growth in semiconductors HAS PEAKED!! >Anyone investing on the assumption that volume growth will >go higher than 30-35% is dreaming. It has never beaten >that figure before. The only way semi revenues are going >higher is by price inflation.
I hope you're not seriously arguing that semi revenues have hit an all time high. They will undoubtedly reach higher levels at some point, perhaps not in this cycle, but in the not too distant future. Unless some catastrophe occurs that profoundly affects human civilization, the world will continue to produce more computational, communication, and memory devices. Guess what building-block they all use?
Still skeptical? Draw a line starting at zero in the 1950's that corresponds to the $$ value of semis in the world. Make a plot of each year up to the present and connect the dots. You will see the familiar hockey stick pattern marketers love to talk about. Now ask yourself what technological trends will stop this progression. Answer: none that we have to worry about for at least 5-10 years.
A more rational argument could be made that semi revenue has peaked for this cycle. If so, just wait a while and AMAT will come back. Over the last 12 years, there has only been one 36 month period where investors have had a negative return on equity. If you stretch that to 42 months, no one would have lost money over the last 12 years. The best 36 month return was 706%--the average was 271%. Not such a bad deal if you got in at the peak of a cycle and had to wait 3 years for a turnaround.
>Fact 4. Everyone is bullish. Just like they were at the >peak of the last cycle. Utilisation is high, everyone is >adding capacity. Anyone see the problem?
See above point. No one can predict when the cycle has peaked but even in a worse case situation, historical data demonstrates that the next wave will be higher and AMAT's price will rise correspondingly. |
