I got this from the ATML board. ATML is another one of my holdings that I have been suffering with the the last couple of weeks. It's dated 6/13/00, but still very relevent today.
Stick with semiconductors My pick for the sector to overweight in your portfolio? Semiconductors. Chip stocks. San Jose sand. I think these stocks are likely to move up in any technology rally that we get after the Federal Reserve stops raising interest rates. And these stocks are likely to get a boost from a "flight to earnings growth," assuming that signs of a slowdown in economic growth get clearer and more nontechnology companies go down the trail that Procter & Gamble (PG, news, msgs) has blazed with its recent warnings of no growth.
Why semiconductors in particular? I like the evidence of accelerating growth that's come in lately from multiple, independent sources.
First, there's the news from the Semiconductor Industry Association. Monthly sales in April (based on a three-month moving average) were 1.4% higher than monthly sales in March. That confirms the forecasts from market watchers Dataquest, of 31% growth for 2000 (over 1999), and from IC Insights, of 32% growth. Those were both solid hikes from forecasts earlier in 2000 that growth would be in the vicinity of 18% to 25%.
Second, there's the news from the semiconductor firms themselves. Motorola (MOT, news, msgs), a company with products in almost every chip segment, has reported that it sees continued strong growth for the rest of the year. Texas Instruments (TXN, news, msgs), a bellwether in communications and wireless chips, recently told analysts to raise their growth forecasts for that company's current quarter.
But what semiconductor stocks in particular? Just as growth in the economy as a whole is an average of hot and cold sectors, so some segments of the chip market are hotter than others. For example, sales for the chips that go into computers are forecast to grow by somewhere around 15% to 20% this year. That's a rate of growth that most of the economy would kill for. But it's significantly lower than the 70% growth forecast for programmable communications processors -- a category that includes the chips that go into networks. Or the 40% growth forecast for the chips that go into wireless-phone handsets. Or the 35% growth forecast for embedded processors. Or the 25% growth forecast for digital signal processors.