Hunting for Signs of Progress in Cisco, EMC and Intel
By Jim Jubak
7/13/01 10:10 AM ET
thestreet.com
In my last column, I said long-term investors thinking about buying one of yesterday's must-own technology superstars should measure the stock's progress in three cycles -- the business cycle, the technology cycle and the competitive cycle.
Here, I apply those general rules to three specific, widely held technology stocks: Cisco (CSCO:Nasdaq - news - commentary), Intel (INTC:Nasdaq - news - commentary) and EMC (EMC:NYSE - news - commentary). My conclusion? True long-term investors should start dollar-cost averaging into Intel at current prices. There's no need to chase the shares into rallies, I'd say, because I don't see the stock running away. But I think the stock represents solid long-term value right now. In the case of Cisco, I think it's still too early for even long-term investors to build a position. And I'd rate EMC a tossup for long-term investors.
I've built my conclusion for each stock on a three-step analysis. In the first step, I look at where the company -- and its stock -- falls in each of the three cycles. I'm looking for strengths and weaknesses in the company's current business. In the second step, I turn that analysis of strengths and weaknesses into a projected 2005 price for the stock by extrapolating business trends in a straight line.
In the third step, I look at how likely it is that the company will be able to improve on its position. I'm pretty sure the smart management at each of these three companies is out to fix the problems I've identified. Any success on this front will enable the company to do better than the straight-line extrapolation I've calculated in step two -- and that will translate into a better rate of return.
EMC Miscalculations
I'll start with EMC. In the column Find Out What Your Tech Stocks Are Really Worth, I calculated a March 2002 target price of $39 for shares of EMC and concluded that the stock, then trading at $41 a share, was fully valued. I based that on numbers that were slightly more pessimistic than the Wall Street consensus in early May: 2002 earnings would be 98 cents, I figured, and earnings growth after that would be 26% annually.
How quaint those numbers seem today, with shares selling for just $22 or so after EMC's huge July 5 earnings and revenue warning. Analysts are now busy cutting their earnings projections for 2002 from $1.05 to a level that looks likely to settle near 70 cents to 75 cents. That's well below my May estimate, but the big questions, of course, are whether even that number is low enough and what the company's earnings growth rate will be after 2002. I think investors can get the numbers they need if they adjust EMC's most recent figures to take account of the business, technology and competitive cycles.
The business cycle continues to look grim for EMC. In its warning, the company said it saw weakness in corporate spending on information technology in all regions of the globe, with the strong U.S. dollar making international sales especially tough. Major customers continue to cut spending plans. It looks as if EMC and its competitors really can't expect an upturn in their business until the beginning of 2002. Nothing is certain, of course, but counting on little or no sequential growth for the December quarter seems more reasonable than looking for the 20% jump in revenue analysts are still modeling.
I think analysts, looking back at the huge gains in revenue at the end of 2000, are still overstating the long-term growth rate in the storage market. Look at the pattern of year-over-year increases in revenue at EMC in 2000 to see why expectations are taking a while to come back to earth. In the first quarter of 2000, revenue at EMC grew by 23% vs. the same period of 1999. In the second quarter of 2000, the increase over 1999 was 30%. Third quarter, 34%; fourth quarter, almost 40%. No wonder that, until recently, analysts were pegging annual revenue growth rates for the next five years at well over 25%.
Projections are now down to 20% to 25%, and I think they're headed to something like 15% to 20%.
Over the next year or two, EMC is also facing the toughest competitive cycle it has confronted in a long time. Both IBM (IBM:NYSE - news - commentary) and Hitachi (HIT:NYSE - news - commentary) seem determined to win back part of the market share EMC grabbed during the past five years. Fielding their most technologically competitive products in years, both companies are competing fiercely on price. That has contributed to a quick deterioration in gross margins at EMC to the mid-40s in the recently completed June quarter, from nearly 60% in the fourth quarter of 2000. Margins are likely to drag on at these levels, even after business conditions improve, as this competitive dogfight plays out.
I can now use what I know about these three cycles to figure out a target price for the stock. Instead of Wall Street's projection of 70 cents to 75 cents in earnings per share in 2002, I estimate 60 cents a share based on my pessimistic reading of all three cycles. And instead of 28% annual earnings growth after that or even my earlier 22%, I think 20% is more reasonable given my lower revenue growth projections for the storage industry as a whole and the competitive pressures I see. That gives me earnings per share of $1.04 by the end of 2005. Figuring a multiple of 35 on a stock growing earnings at 20% a year -- a guess, I'll grant you, but not an unreasonable one -- I arrive at a 2005 price of $36.40. From a recent price of $22, that works out to an annual return of about 12% a year.
Cisco's Woes
The business cycle has hit Cisco even harder than a company like EMC. In an effort to break into the telecommunications market dominated by companies like Nortel (NT:NYSE - news - commentary), Cisco targeted start-ups building new telecom networks. Some of those are already in bankruptcy; others seem to be headed there. Whatever their exact fate, Cisco has had to write off loans it made to finance equipment purchases by these customers and to cope with a steady stream of order cancellations.
This business cycle problem makes it extremely difficult to tell exactly where Cisco is in the technology cycle. By financing sales and focusing almost exclusively on its own revenue growth goals (at the expense of profits), Cisco pushed its revenue growth rate -- 68% year over year in the October 2000 quarter -- up to unsustainable levels. But I don't think analysts have yet taken all that excess out of their numbers. The Wall Street consensus still projects 25% annual earnings growth for Cisco over the next five years. As with EMC, I think this figure is still headed down -- earnings growth for Cisco is likely to be closer to 15% to 20%. Cisco's position in the third cycle, the competitive cycle, argues for the lower part of that 15% to 20% range.
The collapse of the next-generation telecom companies has cut off Cisco's easiest route to break into a market the company believes is critical for its future growth. Selling against entrenched providers, such as Nortel and Lucent (LU:NYSE - news - commentary), at carriers such as AT&T (T:NYSE - news - commentary) and Verizon (VZ:NYSE - news - commentary), has been difficult for Cisco. All the telecom-equipment makers are suffering. But Cisco is the company I believe is suffering the most strategic damage from the telecom collapse.
Using my estimate of earnings per share of 18 cents for the 2002 calendar year and then 20% annual earnings growth after that, I calculate Cisco will earn 31 cents a share in 2005. Estimating that a stock growing earnings at 20% a year would command a multiple of 35 times earnings, Cisco would be selling at about $11 a share in 2005. That's a loss of a bit more than 30% from the current price of $16.
Intel Upside?
Analyzing Intel is the most straightforward exercise of these for two reasons. Intel's business cycle is relatively closer to turning -- the third quarter is an optimist's schedule, but the fourth quarter is a real possibility. Second, since Intel's revenue is still dominated by sales in the mature PC processor market, the revenue growth rates didn't spike so far above the long-term trend at the peak of the cycle in 1999 and 2000. I think this relative certainty on two of three cycles is one reason Intel has been essentially bound in a range between $24 and $34 since the beginning of 2001.
It's the third cycle that's likely to give investors problems. The company has been losing share in the PC processor market to Advanced Micro Devices (AMD:NYSE - news - commentary) over most of the past year and has decided to fight back by aggressively cutting prices.
That has worked for Intel in the past. But with PC makers squeezed and consumers less willing to pay for the Intel name, the price cuts have so far done little except drive down prices for all processors. So while analysts and investors know pretty accurately what the growth rate in PC unit sales will be, the growth rate of revenue from those sales, which depends on the selling price for each processor, is a huge question mark. Right now, with AMD's momentum showing no signs of slowing, my bet is that we'll see intensifying price wars throughout the remainder of the year. That won't do anything to improve margins at Intel, that's for sure.
Right now, analysts are expecting Intel to earn 78 cents a share in 2002 -- a 44% increase from the 54 cents a share they're projecting for 2001. That seems a bit high given the competitive challenges Intel faces over the next 12 months. My own projection comes in at 70 cents. If Intel grows earnings per share by 15% annually in 2003, 2004 and 2005, investors would be looking at $1.06 a share by the end of that year. Using a multiple of 30 times earnings, Intel would sell for $32 a share at the end of 2005. That represents a 3% annual rate of return from the recent price of $28 a share.
Buy, Sell or Hold?
Now for step three -- turning those 2005 price projections into buy, sell or hold decisions. The key is figuring out how likely it is that a specific company will do better than the straight-line extrapolation from current conditions created in steps one and two.
For example, Intel will certainly fix some of its processor business problems. Within six months to a year, Intel will have ramped up production of its Pentium 4 chips, the new chipsets to go with that processor and the associated motherboards. It will have worked out the kinks in its new 0.13-micron and 300-millimeter wafer production processes. That will lead to lower costs and rising margins, even if the price war with Advanced Micro Devices is still being fought with today's violence. And at some point next year, Intel's non-PC processor business -- its flash memory and communications-chip businesses -- will start to recover, and these businesses will go from being a drain on the company's income statement to making a positive contribution.
I think all of the problems Intel now faces explain why the stock sells at $28 a share, but I also think the odds that Intel will fix a significant percentage of those problems within the next 12 months are very good. Intel has a solid chance of doing better than the straight-line extrapolation from current trends that steps one and two suggest. The long-term potential return from Intel through 2005, in that case, is closer to 14% annually than to 3%.
That's not the case with Cisco. Its strategic problems are far harder to fix than the competitive problems bedeviling Intel. And Cisco is also just at the beginning of an awkward transition with analysts and money managers. Cisco's stock used to be valued on revenue growth; now investors are asking, "Where's the profit from all that growth?" Cisco's return on invested capital has collapsed in the past year from an annualized 43% in the April 2000 quarter to just 1.4% annualized in the April 2001 period, according to Robertson Stephens. In other words, Cisco got a return on its invested capital of less than it would have earned by sticking cash into a money market fund. Until CEO John Chambers can show some progress on the strategic and return-on-capital problems, it's too early to bet Cisco can better the negative results of the straight-line projection created above.
EMC falls between Intel and Cisco. The straight-line scenario is a pretty good projection of how the stock will behave -- as improvements in the economy are balanced against continued competitive challenges. A long-term investor might indeed be looking at a return of 12% annually, but with the strong possibility of another dip in the price of EMC's stock if the company warns in September because of slower-than-expected economic growth.
Buying EMC now depends on how an investor feels about the potential combination of 12% return and substantial short-term volatility. One way to figure that out is to compare the risk/reward profile for a stock like EMC with that of a benchmark stock.
In my next column, I'll do just that by seeing how the stocks of what used to be called the regional Bell phone companies -- in my opinion, the big winners in the Comcast (CMCSK:Nasdaq - news - commentary) bid for AT&T -- stack up against more volatile technology stocks like EMC. |
