PEGing Qualcomm
Brian, Keith and everyone,
It was no accident that I didn't provide the numbers I would use to calculate a PEG. Because it is so difficult to arrive at the estimated growth, I think it's something each of us should do for ourselves.
The one point I will stress is that, though I almost always use analysts' estimates, in the case of Qualcomm that process makes no sense to me. The analysts' estimates are barely more than half the trailing actuals. Unless you think the company is going to experience dramatic negative growth, there's no point in looking at the published estimates.
Having said all that, I'll walk through the steps. Anyone can change the assumptions or, better yet, develop a range of assumptions as they see fit.
Trailing Earnings
I always use trailing earnings excluding one-time charges and revenues. Because of the dramatic revelations about the infrastructure business that was sold, in the case of Qualcomm I also use the pro forma numbers that exclude that business. The numbers:
1998 Q4: $.27
1999 Q1: $.32
1999 Q2: $.60
1999 Q3: $.86
Trailing EPS = $2.05
PE
Having established the trailing EPS at $2.05, using a stock price of $163 the PE is 80.
Estimated Growth
This is the tricky part because there are a lot of assumptions that can be made, both with the magnitude of the growth and the period of time you want to include.
If I remember correctly, First Call's PEG uses estimated 5-year growth. I don't like to use that because it's so difficult to know what's going to happen five years out. Also, there are a lot of estimates that, when looked at closely, show glaring contradictions between the near-term estimates and the five-year estimates, so much so that they render the five-year estimates entirely beyond comprehension. I won't bore you with examples because I've already bored you with this diatribe.
So I always start with estimates as far out as the published annual estimates. In the case of Qualcomm, I might use that date but I'd rather not. Right now analysts are only projecting FY2000 which is only five quarters away. Such a short period of time can skew the numbers, so I would instead assume growth through FY01.
For now, let's assume 40% annualized growth. (Change that parameter as you see fit.)
PEG Ratio
We established that the PE is 80. Divide that by 40 (the estimated growth) and the resulting PEG ratio is 2.00.
Revised PEG
One assumption some people are making around here is that the company might achieve $5.00 EPS in FY01. It's optimistic and a nice round number that might be appropriate as a high end of a range.
TO recap, we're using a trailing EPS of $2.05, an ending point of FY01 which is 2.25 years forward, and an estimated EPS of $5.00. Using those parameters, the estimated annualized growth is 49%.
Divide the trailing PE of 80 by that 49 and the PEG is 1.63, far different from 2.00.
As you can see, changing the parameters can make a huge difference in the resulting PEG. That's why I hopefully came on strong enough without being critical in a personal way in my previous post that we have to be very certain that our methodology makes sense. Even with a viable methodology, the paramaters can widely change the results.
The PEG in Context of Gorillas
For true gorillas as defined in The Gorilla Game, I've got anecdotal evidence (not empirical evidence) that a PEG of 2.00 is reasonably normal. Using traditional metrics, 1.00 is fairly valued. But I tend to think of 2.00 as fair value for gorillas. On the other hand, because the market has always undervalued gorillas, I don't dwell on it.
Comments both pro and con are appreciated.
--Mike Buckley |
