You've seen the short-term reaction.
warned that investors should expect not 50% or 60% revenue growth next quarter, but 30%. And in the quarter after that, the numbers would be even worse, with revenue projected to grow by just 12% from the same quarter of 2000.
The next day, Cisco shares dropped 13%. Shares of suppliers such as
Applied Micro Circuits
fell 10%, 10% and 12% respectively. Competitors such as
tumbled 8% and 12% each.
But what if you take a slightly longer view than, say, a week?
Is Cisco Systems a screaming buy here, or would you be better off waiting a quarter or even two?
Is it time to snap up shares of the communications chip makers that were the stars of 1999 now that they're really, really depressed?
Has the market unfairly punished Cisco's competitors until they're bargains, or is it smart to avoid the entire networking sector?
Tough questions, all three. I don't have any definitive answers -- the uncertainties in this slowing economy are just too great for that. But here's my best stab at answering the first two queries. The third question -- what it means for Cisco's competitors -- will be the topic of my next column.
Let's start with some analysis of the numbers that Cisco reported, and the company's guidance for the next two quarters. Then I'll give you my take on the implications of those numbers for Cisco, its suppliers and its competitors.
Cisco CEO John Chambers has spent the last month telling Wall Street that his company found business in January more "challenging" than expected. So I don't think anyone was terribly surprised when Cisco reported earnings of 18 cents a share, a penny below the analyst consensus, or revenue of $6.75 billion, about $500 million, or 7%, below expectations.
What did shock many investors, myself included, was how quickly Cisco now expects business to deteriorate over the next
quarters. Revenue in the next quarter -- the company's fiscal third quarter, which ends in April -- will be flat or even 5% lower than revenue in the just-reported quarter. That means revenue growth, which had hit a high of 68% in the quarter that ended in October 2000 (compared with the same period in 1999), would drop to just 30% in the April 2001 quarter. The relatively modest drop in year-over-year growth to 55% in the just-reported quarter had already broken Cisco's string of 11 straight quarters with accelerating earnings growth. But a drop to 30% would mean that revenue growth had slowed by more than half in just two quarters.
To my mind, though, that wasn't the worst of Cisco's news. Unlike so many technology companies that reported earlier in 2001, Cisco told analysts that the slowdown in growth wouldn't be limited to just one quarter. The company's fiscal fourth quarter, the one that ends in July, would be flat with the third quarter, Cisco estimated. That would put year-to-year revenue growth in that quarter at just 12% or so.
Earnings are likely to take just as hard a hit as revenue going forward -- maybe even harder. The costs of carrying the currently very high levels of inventory, up $575 million or almost 60% from the October 2000 quarter, took a full percentage point out of gross margins this quarter. I'd expect inventory costs to continue to punish margins over the next two quarters as Cisco works down its backlog of components and work in progress. Weak pricing at the lower end of Cisco's product line and a shift in the sales mix toward lower margin products also hurt Cisco this quarter and will do so over the next two periods. All that means that earnings per share, which had been projected to grow by 40% next quarter before the announcement, would actually be flat for the period, and that earnings in the July quarter would show a 12% decline instead of the previously projected 30% increase.
What's Cisco Worth?
The big question for investors today is why is Cisco's growth slowing? If you believe that the slump over the next two quarters is just a result of the slowdown in the general economy -- the view that Cisco pushed in its conference call -- then a reasonable, two-month target price for the stock is, by my calculations, about $49. If, however, a major part of the slump in Cisco's growth is a result of competition from Juniper, Extreme, and
-- a view that's supported by the numbers reported by these companies -- then a $35 12-month target for Cisco seems much more probable.
Here are the assumptions behind both those cases, which I'll call "It's the economy, stupid" and "The market-share blues."
To get to my target for the "It's the economy, stupid" scenario, I start with Cisco's own guidance. The company projected flat or slightly declining revenue for the next two quarters and then revenue growth of anywhere from 30% to 50% for fiscal 2002, the period that ends in July 2002. Flat growth for the next quarter would give the company $6.7 billion in revenue for the period that ends in April 2001 and for the quarter that ends in July. Then let's say that the economy recovers and growth at Cisco immediately rebounds to the midpoint of the company's projected year-to-year rate of growth in fiscal 2002 of 40%. That gives me revenue of $9.1 billion and $9.4 billion in the next two quarters. Add it all up and, a year from now, Cisco would be reporting $31.9 billion in revenue.
How much is Cisco worth if revenue growth rebounds to 40%? We can go back to a point in Cisco's past, when the company was growing revenue at about this rate. From July 1997 to July 1998, revenue at Cisco climbed by 31% as the company came off a slump that included the slowest sequential growth in the company's history. From July 1998 through July 1999, revenue grew by 44%. Now in July 1998, the stock traded at a price-to-sales ratio of 11.8. Shave that a bit to take account of the difference between 44% growth then and the projected 40% growth now and I think it's fair to use a price-to-sales ratio of 11 for the purpose of my calculations. That gives me a market capitalization of $351 billion and a target price of roughly $49 a share for July 2002 (after dividing by 7.2 billion shares outstanding.)
That's a 60% gain from the recent price of $30.44. At that level, Cisco shares would still be trading well below the $70 they hit in July 2000 and even further below the stock's 52-week high of $82. But a 60% gain from here certainly wouldn't be too shabby.
But remember, that's the pure bounce-back price that assumes Cisco comes back to all its glory after the economy ends its slump, and continues to dominate all its markets as it did last year.
But what if Cisco continues to lose market share to its competitors? Juniper Networks, for example, came on to take 30% of the router business by the end of the third quarter of 2000, according to
. Not bad for a company that had no market share in January 1999. To take another example, in the layer 2 and layer 3 switching market, Cisco is losing share to Extreme Networks, which is growing nearly twice as fast.
What if, instead of 40% growth, Cisco rebounded to just 25% revenue growth as competitors continued to take share and forced Cisco to fight back by lowering margins and cutting prices? Instead of $31.9 billion in revenue a year from now, Cisco would then be looking at $30 billion in revenue -- and a much lower price-to-sales multiple in the market. Shaving the 8.3 multiple that the stock showed in July 1997, when the company was looking at 31% growth, to 8, that gives me a market capitalization of $240 billion and a 12-month target price of $33.33. That would be a 10% gain in a year.
Which one of these scenarios -- and prices -- is most likely? I'd argue that right now we really don't and can't know. The range of the recent guidance given out by Cisco -- the company expects growth in fiscal 2002 to be somewhere between 30% and 50% -- suggests that the company doesn't know, either. At the moment, I'd handicap the stock this way for long-term investors: Up to a purchase price of, say, $35, I think you'll at worst get your money back in the course of a year. If the stock drops further, to say, $25, then the return for the worse of these two scenarios rises to 40% and, taking the risk/reward tradeoff, becomes very attractive. On the other hand, I certainly wouldn't care for the stock much above $35 right here. The returns from the better-case scenario are certainly attractive, but I'd argue that the true uncertainties in this situation are so great that I'd like the potential for at least a 50% return in compensation. The potential gain from $35 to $49 in a year comes to just that 50% figure.
What Did the News Mean for Cisco's Suppliers?
It's pretty simple: The huge inventories at Cisco are a potential disaster for the companies that sell chips and other components to Cisco. I think the market was perfectly justified in punishing Broadcom, PMC-Sierra, Applied Micro Circuits and
under these conditions. After all, we're talking about a huge buildup in inventory -- almost 60% from the previous quarter and almost 200% from the same quarter 12 months earlier. It's a big deal when a major customer says that working through past orders could take two to three quarters.
Unfortunately, for two reasons, it's very difficult to tell exactly how big a deal it is at any individual company. First, it's clear that the buildup of inventory to these levels resulted in huge orders to suppliers that inflated the real rate of growth in their business. Vitesse, for example, showed 20% revenue growth in its most recent quarter over the prior quarter. Revenue from Cisco, one of Vitesse's two biggest customers, climbed by 15% quarter to quarter. But, thanks to the inventory buildup at Cisco, investors really don't know -- I'm not sure that company CEOs know either -- how much of that 15% growth rate was simply a result of the surge in inventory and how much is real sustainable growth. This problem makes all the projected revenue growth rates for Cisco's suppliers suspect.
Second, while we can get a pretty good handle on how much business any supplier does with Cisco --in the third quarter, Cisco accounted for 11% of communications revenue at Applied Micro Circuits -- we don't have a precise breakdown of exactly how much inventory has built up for individual components. We know that Cisco's revenue from routers fell about 16% quarter to quarter in the most recent period and that switching revenue grew 18% quarter to quarter, but we don't know how much backlog Cisco has in the components that go into each product line, so we don't have a clear idea of how badly orders will fall off to suppliers of specific kinds of products.
This makes it extremely difficult to say whether the stocks of Cisco suppliers are bargains even after their recent nosedive. In my Jan. 31 column, Why Vitesse leads the 4 horsemen of chip stocks, I gave the nod to Vitesse and Applied Micro Circuits over Broadcom and PMC-Sierra and then said I preferred Vitesse to Applied Micro Circuits largely on valuation. That, however, was before Cisco reported -- and before Applied Micro Circuits issued what Wall Street took as a semi-warning.. Since Jan. 30, Applied Micro Circuits has tumbled almost 40% and Vitesse a much more modest 17%. That's enough to reverse my relative preference for the two stocks.
But that doesn't mean I'd buy either of them here. I think that Applied Micro Circuits is edging toward a warning, and I find it difficult to believe that Vitesse will be able to escape the fallout from Cisco's inventory buildup completely. Yes, the stocks are cheaper than they were, but at 63 and 47 times projected 2001 earnings per share, the stocks still aren't cheap. In this environment, I just can't justify recommending either of them in Jubak's Picks. They are simply too risky for most investors at the moment. I'll wait for either a warning from one of these companies or some evidence that they've dodged the bullet.
That doesn't mean there aren't opportunities in this market, however. In my next column, I'm going to look at if there's a silver lining in the Cisco news for its competitors. Stay tuned.
At the time of publication, Jim Jubak owned or controlled shares in the following equities mentioned in this column: Applied Micro Circuits, Cisco Systems, Extreme Networks, LSI Logic and PMC-Sierra.