Hi Mike - hmmm... there really isn't much to understand.
I prefer to base the price of a slice on something related directly to the slice itself. Maybe I don't belong on this thread, but I am not out merely to bag trophy gorillas and hang them in my parlor to gather dust like some 19'th century English adventurer.
I am more like a modern farmer, planting and tending and harvesting. My crop is capital. It's a cash crop <g>: I pay cash, I want cash. In my simple farmer's view of the world "How much does it cost for how much do I get" is the ONLY operative question. I am also a LT investor, so by implication I am looking 5-10 years over the horizon.
So beyond paring down the universe of stocks into a handful of worthwhile candidates, I am keenly interested in a mechanism that helps us allocate capital between different investments. This is not an academic exercise.
Let's forget for the time being that I prefer to use DCF.
But let's not forget what I derived. I showed how to relate the PE to DCF directly through a set of assumptions. I also showed that PEG does not relate to DCF. If you do not feel like understanding the math, then please credit those of us who obviously do for some intelligence and accept our assertions. Otherwise we will be tempted to bludgeon you with reams of equations, which isn't fun for anybody <vbg>.
But that is all I showed. This does not mean that using PE is grounded in reality. You might also note that PEG is even less firmly grounded than PE!
Perhaps if I state it another way. When you say "a company is fairly priced at PE=40", you will be closer or farther from an objective truth depending on the degree to which a certain set of assumptions hold true. I even articulated the assumptions and gave them names. You can even compute sensitivity analysis around the deviation. So PE can be a useful (blunt) instrument.
However, when you say "a company is fairly valued when PEG is 1.2", you are randomly< closer to or farther from this same objective value. I say randomly because there is no computable relationship between E*G and value... except (as math wizards will point out) one that would factor the G out again to get you back to my P = E times some function of (k,c,C,G,D,n)!
I happen to believe that neither are appropriate tools for all cases, except in very specific circumstances. I chuckle silently when folks debate whether it should be a "leading" or "trailing" PE... and can barely stifle my guffaws when folks say "I think it will be worth $30 in two years because with earnings of x and growth at y it will have a PEG of 1".
Now, you may be correct that your guess of DCF is going to be so far off base that you might as well use a random pricing methodology.
But if you are going to use PEG, then you are UNWITTINGLY forecasting something loosely unrelated to future cash flows. So you are hardly going to be worse off guessing at the cash directly (it can lead to surprising insights).
Now, maybe I'm from the old school - where the 10K and 10Q are very important documents to read and understand. So I'm of the opinion that anyone who doesn't feel comfortable staking a guess at the next 10 years cash flow really shouldn't be calling the shots on their portfolio at all!
It's easy. Open an excel spreadsheet. Label the first row "revenue", the second row "gross margin", the third row earnings, the fourth row "cash flow".
Populate with this year's data. Extrapolate in a first pass by using whatever growth rate you think is useful for the next 20 years. Convince yourself that each columns numbers "make sense". If not, adjust. Presto. You have the basis for a cash flow estimate.
Next, all you need to do is discount this back to present value and bingo, you have a DCF analysis. It's simple really. It is eye-opening when compared to some of the prices of tech stocks. Go ahead, give it a try. Maybe compare the results to PEG.
John |
