Is there a way for long-term investors to navigate the technology sector right now without falling victim to either greed or panic? Certainly

Federal Reserve

Chairman Alan Greenspan didn't help any with his "recession yes/recession no" testimony to the

Senate Banking Committee

Feb. 13. Investors are excused if they feel more at sea than ever after that performance.

Here's the problem: In the long run -- i.e., over a holding period of 12 months or longer -- current prices promise some mouth-watering gains. A modest revival in the technology sector could produce gains of 100% in

Extreme Networks

(EXTR) - Get Report

, 80% in

Ciena

(CIEN) - Get Report

, 75% in

Brocade

(BRCD)

and 60% in

Juniper

(JNPR) - Get Report

within the next 12 months. Potential gains from more seasoned stocks like

Oracle

(ORCL) - Get Report

,

Nortel

(NT)

and

Nokia

(NOK) - Get Report

could easily top 30%. Greed shouts that this is the time to load up.

But in the short run, many of these stocks are still locked into painful declines. Every time investors think they've found a bottom, they discover to their loss that this market has the ability to further punish stocks that are already trading at 50% or more off their highs. Brocade, for example, clearly had major support at $80, $72 and $66 a share, but that didn't stop the stock from falling to $53 on Feb.13 on a warning from competitor

Emulex

(EMLX)

and a few negative comments from an analyst at

Goldman Sachs

. The stock now has major support at $48 and $40, but there are obviously no guarantees that those prices will hold, either. The same pattern holds for less volatile technology names. A reasonable investor might have figured that Oracle would hold support at $35, $30, $28 or $25. But on Feb. 13, it closed at $22.56. Panic yells that this is no time to buy anything.

Falling prey to either emotion will cost a long-term investor money. But fortunately, there are tried-and-true methods for keeping a balance between the two extremes. In my opinion, combining three of these methods -- waiting for the trend, anticipating the turn and analyzing risk/reward -- in a single strategy built around

dollar-cost averaging will give the long-term investor the best shot at market-beating returns when the turn finally comes in the sector.

Waiting for the Trend

Don't try to catch a stock on the way down, the conventional advice goes; wait until you see evidence that the trend has turned. For once, I think the conventional wisdom is exactly right. There's no point to buy a bargain stock that just gets cheaper every day. Wait until that stock establishes a pattern of heading up.

A stock's chart is the best indicator of this kind of reversal. First, what you'd like to see is some signs of a bottom. A stock that's putting in a base will trade in a narrow range for some time on relatively light volume. You can see this kind of pattern in the three-month chart of a nontechnology stock like

McDonald's

(MCD) - Get Report

, which built a base near $29 after a big fall on high volume from $34.70.

Second, you'd like to see signs that the stock is moving up from this base. The stock's price will begin to rise, and volume should at least remain steady, or better yet, gradually increase. Indicators that measure the money flowing into and out of a stock, such as money flow and on-balance volume, should turn up as well. Again, look at a chart for McDonald's to see what the beginnings of this pattern look like.

I'd buy a stock like McDonald's that is coming off a 52-week low and shows signs of building a sustainable upward trend.

Unfortunately, I can't find any technology stocks that are flashing this kind of positive sign.

Intel

(INTC) - Get Report

,

Dell

(DELL) - Get Report

and

Microsoft

(MSFT) - Get Report

all were building a solid upward trend out of decent bases -- until they turned down again last week. I'd like to see them re-establish those positive trends before I put money into the stocks.

I can, however, discriminate between technology stocks that look as if they're building a base and those that have yet to begin the process. I can't tell you whether the bottom on

Inktomi

(INKT)

is $12 or $11 or $10, but I do know that the stock has spent the past month stuck in a trading range between $12 and $19, and that there's a reasonably strong likelihood that the bottom is somewhere near the lower end of that range.

Contrast that to a Juniper Networks or an Oracle, whose stocks have broken below every recent support level without putting in a base at any level. Investors are now hoping that the stocks will bounce off the lows established in the spring and summer of 2000. In the short term, I'd argue that a stock like Juniper is likely to bounce off those support levels -- and a speculative buy right now might even produce a very nice short-term profit. But longer term, I think the odds that we've seen the bottom on these stocks is extremely remote.

Anticipating the Turn

The pattern traced by a chart falls into the category of trend-following indicators. You can see the pattern of an upward move in a stock only after the stock has moved upward for a while. (And the longer you wait, the stronger the pattern gets and the less likely it is to turn out to be a false signal.) Waiting for the trend to show itself means that you'll certainly miss the beginning of a stock's upward move.

The alternative is to attempt to anticipate the bottom in a company's business. The stock market buys expectations of earnings and other financial results. A stock will start to move up in price when investors begin to anticipate that the worst news is past and they start to look forward to an improvement in a company's fortunes.

A reasonable rule of thumb that is used by some very experienced technology investors is to start putting money into a stock about six months before you expect the company's business to turn. That way, you'll get in at a bargain price and receive the entire benefit of the run-up that will start as investors gradually begin to anticipate better times for the company.

So, for example, after listening on Feb. 13 to

Applied Materials

(AMAT) - Get Report

CEO James Morgan predict that the downturn will last for at least six months, you might decide that the downturn in the semiconductor-equipment industry will last for nine months. (CEOs, even great CEOs like Morgan, tend toward optimism about their companies.) That would put the turn in the industry in October, and a long-term investor might start building a position in the stock six months before that, in April.

The problem with anticipating the turn, however, is that you'll get left high and dry if the turn fails to arrive on time. In January of this year, the technology sector rallied, in part, because companies like Nortel said that they expected the turn in their markets sometime in the second quarter as the Federal Reserve's interest rate cuts created supplies of cheaper capital. If the turn was that close, investors reasonably concluded, the time to buy was

now

.

That rally came to a grinding halt in February when investors began to question that timetable and the market went into full-scale reverse when

Cisco Systems

(CSCO) - Get Report

reported earnings Feb. 6. With Cisco predicting that it will have worked out its inventory problems in two to three quarters, investors were suddenly looking at a turn, not in April or May, but in September or October. No need to buy now; March or April would be soon enough. And with individual stocks continuing to blow up, waiting instead of buying looked very attractive indeed.

Do we now know when the turn will take place? No way. Cisco's prediction, which roughly coincides with the guidance from Applied Materials, is just the wisdom of the moment and could well be superseded by later and better data. In his Feb. 13 testimony to the Senate Banking Committee, Greenspan noted that we, meaning I suppose the Fed, "don't know how far the adjustment of the stocks of consumer durables and business capital equipment has come." The economy could be, Greenspan was implying, near the beginning of the inventory correction or someplace further along with much of the excess worked out of the system. With Wall Street awash with rumors that distributors and contract assemblers are telling high-tech manufacturers that they can't absorb any more inventory -- which would have the effect of pushing the inventory back on the original manufacturer where it will then show up on next quarter's books instead of being hidden in the sales channel -- I doubt that we're close to the end of this work. But no one really knows.

What

is

a good bet is that the Federal Reserve's interest rate cuts will be felt first at technology businesses that don't keep much inventory and are leveraged more or less directly to the consumer. That category would include electronic brokerages such as

E*Trade,

(EGRP)

advertising-supported Internet media companies such as

AOL Time Warner

(AOL)

and

Yahoo!

(YHOO)

, as well as technology companies that sell primarily to consumers, such as Nokia and

eBay

(EBAY) - Get Report

. We don't know precisely when the turn in these businesses will be, but it is a good bet that it will be earlier than for inventory-heavy, capital-spending-dependent technology companies.

Risk/Reward Analysis

In my

last column, I used this method to handicap stocks such as Extreme Networks, Juniper Networks and Ciena. In essence, the method is very simple: Ballpark the upside of a stock by estimating 12-month forward earnings per share and times that by a reasonable multiple of the stock's earnings growth rate. Then ballpark the downside for the stock by looking not at the closest technical support, but at the strongest support level you can find one or two steps below that.

Let me run through one more example here to give you a rough idea of how it works. Brocade is projected to earn 62 cents a share for fiscal 2001. That would be up 121% from fiscal 2000 earnings per share. Figure that a stock growing at 121% a year would earn a multiple higher than its growth rate. I'll use a price-to-earnings ratio of 150, which is I think conservative and considerably below the stock's recent 225 P/E ratio. That gives me a rough upside price of $93 a share. Next, downside support on this stock is $47.50. After that, it's at $40. Finally, that gives me a potential upside of 75% from the stock's Feb. 13 close of $53 and a downside of 25%. Pretty good ratio, I'd say.

So how do I combine these three methods into a single strategy for long-term technology investing right now? By using dollar-cost averaging to buy the stocks that come out best using my three methods.

For example, among the 50 stocks in the Future 50, E*Trade and Yahoo! score fairly high on the "anticipating trend" scale (because both are consumer-oriented technology stocks), they score decently on the "wait for the trend" measure (because both show signs of having put in a bottom), and they score in the midrange on the risk/reward scale. They're not ideal buy candidates, but there aren't many ideal buy candidates in the technology sector right now. Long term, I think investors want to own these stocks.

What other stocks score well? Besides E*Trade and Yahoo!, holdovers from among the current new-money buy stocks include AOL Time Warner, Applied Materials, Exodus,

JDS Uniphase

(JDSU)

,

Metromedia Fiber Network

(MFNX)

and

RF Micro Devices

(RFMD)

. Stocks moving up to new buy ratings (for new money) include Intel and

Openwave Systems

(OPWV)

.

Drops from the buy group include Inktomi and

PMC-Sierra

(PMCS)

, which are likely to see their turns late in this current cycle. I think long-term investors can wait on both of these.

I wouldn't buy these or any other technology stocks for a long-term portfolio all at once. There's too much uncertainty in the market and in the economy. Instead I'd decide how big a position -- in dollar terms -- I want of each or any of these stocks, and then use dollar-cost averaging during the next six or nine months, depending on your take on the economy, to build that position.

If you don't feel experienced enough to call the dips and rallies of this market with some accuracy, simply buy an equal dollar amount of each stock (one-sixth or one-ninth of your allocation to this stock) on a fixed date each month. That way you'll buy more shares when the share price is lower, and fewer shares when the price is higher.

If you have some experience with calling dips and rallies, but not enough to feel you qualify as an expert market timer, you can still apply this method by buying an equal dollar amount of each stock whenever you feel it has dropped reasonably close to a bottom. I think you'll get enough volatility over the next six months to put this method to work again and again.

And of course, if you're an expert at calling tops and bottoms, hold your powder until you know the stocks you want have hit their absolute bottoms.

At the time of publication, Jim Jubak owned or controlled shares in the following equities mentioned in this column: AOL Time Warner, Applied Materials, Ciena, Cisco Systems, E*Trade, Extreme Networks, Inktomi, Intel, Mercury Interactive, Metromedia Fiber Network, Microsoft, Nokia, Nortel Networks, Oracle, PMC-Sierra and RF Micro Devices.