Jubak's Journal
Figuring out what that stock is worth
Whether you want to sell at the top or buy on the dip, you'd better know how to value a stock. Here's why I think Tellabs and EFII are sells -- and a list of some promising stocks the market has slighted.
By Jim Jubak
It's useful to have some opinion about what a stock you own -- or are thinking about owning -- is worth. The market has administered a pretty good reminder of that in the last few weeks.
First, an excess of enthusiasm pushed many stocks, especially technology and Internet stocks, to rather heady levels. Investors thinking about selling wondered anxiously what their shares were really worth. And then, the stomach-churning plunge that included the 140-point, 6% drop in the Nasdaq Composite Index on April 19, sent investors scrambling to value their stocks again. Was the stock now trading at $117 a "buy" because it really was worth $175, where it had been only a few days before? Or was it a "sell" because it was headed for $60?
I had originally intended to use this column to run through the methods I've used to value two stocks, Tellabs (TLAB) and Electronics for Imaging (EFII). Both have announced great earnings numbers in the last couple of weeks and -- at least before the recent plunge -- had reached my target prices. I need to know whether to drop or hold Tellabs, a current Jubak's Pick. And I need to know whether I should re-recommend Electronics for Imaging, which I dropped from Jubak's Picks on April 16, before the recent pullback knocked $3 a share off the issue.
I'm still going to do that. And I'm still going to use my next two columns to take a whack at calculating valuations, first for telecommunications companies that have cash flow but no earnings and then for some Internet companies where even cash flow isn't very useful.
But I'm also going to throw in something extra at the end of this column. I'll name some stocks that the market has ignored during this volatile period, even as they posted great earnings. The stocks in that group are poised for at least a short-term pop, I believe.
Are Tellabs' numbers what they seem?
Let's start with Tellabs. On April 14, the company announced first-quarter earnings of 52 cents a share on $470 million in revenue. The earnings-per-share number was about 3 cents above analyst projections. Revenues were about in line with expectations. (All figures for Tellabs are before the 2-for-1 split announced April 21 to take effect May 17.)
So buy, sell or hold?
I start the process by making sure the numbers really mean what they seem to indicate. Are earnings inflated by one-time gains, a reduction in the corporate tax rate or extreme currency fluctuations? You can find the numbers you need for this in the company's press release, on its Web site and eventually in the quarterly documents (called Form 10Q) filed with the Securities and Exchange Commission. You can also get more detail by listening to the company's conference call with Wall Street analysts. (A phone number where you can hear the replay often is listed on the company's Web site, which can be found by clicking Company Report to the left.) Tellabs seems to pass this test: Its earnings are pretty clean.
Next, using the reported numbers and consensus analyst estimates, I put together my own estimate for earnings six months or a year from now. (Use whatever period you prefer.) Currently, the average estimate for Tellabs' 1999 earnings among the 27 analysts who follow the stock is $2.49, up 28% compared with 1998. Double-checking that number against the consensus EPS trend, which can be found by clicking to the left, I notice that analysts have raised their estimates in the last week from $2.43. That's a typical pattern after a company beats expectations.
Don't assume that this is the end of the upward revision, either. With 27 analysts in the group, it can take a while for all the upgrades to come in. I like to check the research alerts to see if I can find a pattern that indicates the projection might still be rising. None of the alerts listed in our news section is more recent than April 15, so I think it's safe to say that most of the enthusiasm generated by Tellabs' earnings report is now reflected in consensus estimates.
Details
--------------------------------------------------------------------------------
Company Report
Press Releases
Financial Statements
Earnings Estimates
Consensus EPS Trend
I'm willing to bump up my projected multiple to 48. That gives me a target price of $122.40 for the stock in January.
--------------------------------------------------------------------------------
Within the safety zone
Before taking any steps to turn that earnings forecast into a stock price, I have to decide whether I agree with the Wall Street experts. Is there any trend in Tellabs' numbers suggesting the company could be about to hit a rough patch? An unexplained slip in gross margins of the size that Tellabs reported this quarter -- from 63% in the first quarter of 1998 to 59% in 1999 -- would worry me, for example. But the company has explained in past reports that this decline is an expected result of the growth of service as a percentage of total revenue and some changes to the way the company accounts for costs for that service business. Other important indicators, such as accounts receivable and inventory, seem well within the safety zone.
It's just as important to figure out if analysts are being too conservative. Tellabs introduced two new products in the first quarter but didn't recognize any sales from either the AN2100, a voice/data switch already ordered by Sprint (FON), or the Titan 4500, an extension of the company's bread-and-butter Titan 5500 product for the international telecommunications market.
Tellabs told analysts to expect revenue from these products to kick in during the second and third quarter, respectively. Two other new products for the broadband communications market are expected to go into trials in 1999. This is all great news, but I don't see any potential big positive surprises here that would lead me to push up analyst estimates.
The one product that might contribute more than expected to the top and bottom lines is Cablespan, Tellabs' so-far disappointing technology that allows a cable TV operator to deliver phone service over cable lines. So far, Cablespan hasn't been much help to Tellabs earnings, even though the product is what got this stock into Jubak's Picks in the first place at $64.625, about 80% below the current price. (See my March 17, 1998, column, "Picking up on telecom mergers.")
But Cablespan should contribute about $60 million in revenue this year and analysts expect that figure to grow by better than 50% in 2000. If Tellabs could sign a big contract or two -- with AT&T (T), for example -- that might produce an extra $10 million or $20 million in revenue from this product in 1999. I'm willing to bump up the consensus estimate about 3 cents a share for Cablespan's prospects.
Tellabs also has a history of surprising analysts each quarter by about 3 cents a share. (Click Earnings Surprise below and to the left.) If you're willing to be aggressive -- and I am about this stock -- you can add that 3 cents in, too.
That gives me a total of $2.55 a share in earnings for all of 1999.
Details
--------------------------------------------------------------------------------
Tellabs Research
Earnings Surprise
Research Wizard
10-yr. Summary
Measuring enthusiasm
Now the harder part: figuring out the price-to-earnings multiple investors will be willing to pay in January 2000 for those earnings. You have a number of different choices, some of which are laid out in our Research Wizard's Price Target page. (Click to the left.) For example, you could use the stock's current P/E multiple of 46.2 and just project it to January. Or you could use the 58.7 multiple that the market awards to the stock's peers in the telecommunication's equipment group. (I'd take this number with a grain of salt -- it's distorted by the high P/E ratios of Lucent Technologies (LU) and Cisco Systems (CSCO), which reflect one-time charges against earnings for recent acquisitions.) Or you could use a historical multiple, either the highest that Tellabs has shown in the last five years (73) or the highest average in any recent year (48.4 in 1996). You can find this historical data in the section called Ten-Year Summary by clicking to the left.
All this gives you a lot of numbers, but not much guidance, I'd be among the first to admit. What multiple investors are willing to pay for future earnings is as much a question of market psychology as financial results. I try to ask myself a simple question: "Am I likely to be more or less enthusiastic about the company's prospects next January than I am now?"
In Tellabs' case, I don't see any reason to be less enthusiastic. The telecom equipment market will be just as hot as it is now. The company will still be well-run with the potential to move into the top rank next to Cisco and Lucent and Northern Telecom (NT). So I don't see any reason to think that Tellabs will have a lower P/E ratio than the current 46.2.
Do I see a reason to be significantly more enthusiastic? The company will have two important new products on the market and two even more important products -- a dense wave division multiplexing system for fiber-optic-based networks and its next generation cross-connect for broadband systems -- in field tests. So I think that by next January, Tellabs will have even more potential than it does now. On that basis, I'm willing to bump up my projected multiple to 48.
That gives me a target price of $122.40 for the stock in January -- a 48 multiple times $2.55 a share in projected earnings.
Is it worth the risk?
There's uncertainty, of course. The whole market could tank, or just the technology sector. Tellabs might flub a quarter or two, or a new product could fall behind schedule. The company could make another attempt at an acquisition like last year's bid for Ciena (CIEN) that wound up cutting the price of Tellabs' stock in half before it was abandoned.
Tellabs' 6% return over nine months falls short of that 25% annual hurdle. So for me, the stock is a sell at $116.
The difference between today's price -- $116 at the close on Thursday April 22 -- and that potential $122.40 is an investor's payment for taking on this risk. That potential $6.40 per share gain is about 6%. You now have to decide whether that potential 6% over nine months is worth the risk.
Two factors ought to enter into that decision. First, how certain to do you feel about your numbers? I've been reasonably aggressive in estimating future earnings; I have less confidence in the 6 cents I added to the consensus $2.49 than I do in that number itself.
As doubts go, that one's not large. It certainly doesn't begin to compare to the uncertainty I'd feel at projecting earnings for Compaq Computer (CPQ). So I'd have to say that I feel pretty good about the math for this specific stock.
Second, how risky does the market as a whole feel? Remember that the projected multiple for Tellabs rests on the assumption that the market next January will be about as healthy as it is now. The market, especially for technology stocks, is actually likely to be less risky in January 2000 than now, I feel, but with the market at historical highs, I certainly can't claim that there's no risk. So again, I say the numbers I've come up with for Tellabs come with about average risk.
I use a three-part system of risk hurdle rates to determine whether I want to invest in a stock. Looking out at next January and the definite risk in the current market, I'd say I'd need a 20% potential rate of return from a low-risk stock before I'd consider buying it. A stock with moderate risk should have the potential to return 25%. A high-risk stock should fall easily on the far side of 30%.
Tellabs' 6% return over nine months falls short of that 25% annual hurdle. So for me, the stock is a sell at $116.
Any comments on that article will be appreciated.
Rafael Silva.
|
