Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Micron Only Forum -- Ignore unavailable to you. Want to Upgrade?

To: Carl Held who wrote (41106)	11/28/1998 6:05:00 AM
From: Carl R.	Read Replies (2) \| Respond to of 53903

For some companies looking at the past will help establish trends that allow you to predict the future. With high-tech cyclicals the immediate past is not predictive of the immediate future. Things change quickly, and stock prices react violently to shifts in direction. Stock prices reflect a consensus as to what the picture will be in a year, not where they are right now. Investing in (or shorting) stocks in this sector either requires you to be very patient so that you can ride out the cycles, or very agile because things change so suddenly. Let's go back to 1995 for example. Everything was rosy. MU was making money hand over fist. AMAT and the other semi equipment makers were selling equipment as fast as they could make it and had a huge order backlog. But suddenly far sighted investors recognized that too many fabs were being built and that that would ultimately lead to oversupply, price cuts, reduction in equipment purchases, etc. Stock prices started falling starting in about September 1995 and continuing until July 1996. Earnings for some equipment makers were setting records in mid-1996, but their stocks were hitting lows. Why? Because investors were looking ahead to the future not to the past or present. Is it rational for these stocks to anticipate the future to such an extent and not take into account the present at all? Maybe not, but it doesn't matter because that is how it works, so you have to live with it. To trade semis keep in mind the title of some SI threads "buy when blood is running in the streets" and the converse "sell when champagne is in the glasses". Blood was certainly running in 1996, and champagne was in the glasses in 1995 and 1997. More recently blood was running in October. Business sucked so badly that older fabs were being closed left and right. Asia was going to suck the world into depression. No one would ever make money in DRAM nor order equipment. So obviously that was time to buy DRAM and semi-cap stocks. <G> Now we are at a critical juncture where we need to decide if the glasses are full of champagne. Certainly there is a taste of champagne, but is it full yet? That is the question. In other words, if these moments of profit will once again be brief and ignite yet another round of shrinks and fab upgrades, then now is the time to sell. If on the other hand the growth in DRAM demand has now caught up with supply to the extent that by a year from now DRAM makers will be able to all produce full bore without creating oversupply then things will get better for a long time to come, and the glasses will eventually have a lot more champagne in it than right now. Thus you can argue that the stock should not trade based on uncertain events 1 or two years in the future, but the whole semiconductor sector does, and there isn't much you can do about it. You either have to learn to enjoy the cycles (both up and down) or find another area to invest that is not so violently cyclical. Has MU been wounded in the past year or two? Yes. Have they been wounded badly enough to prevent them from being competitive in the future? No. So forget the past and let the future be your guide. Remember in cyclicals the time to go long is when things are bad and about to get better, and the time to short is when things are great and about to get worse. So lets turn the debate from the past to the future. There was a chip glut, and MU has been selling below cost and losing money. There is a shortage of chips now due to strong demand. The Koreans are ramping production back up to meet that shortage. This is all past, the question is, what will 1999 bring? Prices will probably start down again after Chistmas build-up season unless Y2K hardware sales are strong. As demand softens, will the Koreans go back to partial shutdowns? How far will prices drop in the short run? How soon will demand rise enough to catch up with supply? Will non-PC applications such as digital TV show sudden growth and increase use of DRAM? Will most worldwide production of DRAM switch to RDRAM, thus decreasing production 30% because of the additional transistors and leading to a DRAM shortage? How long will it be before DRAM becomes sufficiently profitable that anyone ever builds another DRAM fab? These are the questions that will move the stock in the future. Carl