Mark, I don't think we are disagreeing much about these things, but there are subtleties that I think are important.
Big stocks are in a completely separate market. A small company with the financial record of Microsoft would never receive a P/E of 62.
Agreed, at least in the current market environment. My only point in mentioning the likes of MSFT as an example, is to show how much the market will pay up for credible earnings. The S&P and WIND have the same trailing PE, but (according to Everen) the S&P is expected to grow earnings by a tiny 7%, which not only is miniscule compared to anyone's expectation of WIND's performance, but only a fraction of the 20% number that started this exchange.
I believe there are two primary market considerations in play here. The first is classic economics; the second is something that has more to do with what might be called “contrained optimization.” Let me describe these, the first being really easy.
1. Classic Economics
See Erwin Sanders (Post #4505) for a fine example of this kind of calculation and thinking. The objective simply is to find stocks that have high present values of future streams of per share free cash flow (or EPS as an approximation). Assume for the moment that most investors would concede WIND is on an inevitable growth path of at least x%, where x is significant. Then Erwin's argument make a compelling case for buying the shares, irrespective of PEG values.
Erwin is correct that PEG's are horrible measures, because they make no economic sense. A low PEG is just the market's way of saying, “I think this stock is likely to stumble.” Such a statement is inconsistent with the notion of the relatively high growth rate that generated the low PEG. The inconsistency spells danger, rendering the PEG totally useless.
Note to Erwin: I believe the 3 to 5-year growth rate is key to your analysis, but don't use analysts estimates, because they don't bother to think them through adequately. You can drop all your out-year (greater than 5 or so years) discounting by ending your projected series with the application of a conservative PE ratio to the final EPS. The length of the period in which earnings are projected should be no longer (or shorter) than the period about which you feel comfortable projecting. I agree with you about using comparable risk-free rates for discounting, because you want to calibrate the value the market would put on the earnings stream IF IT BELIEVED the stream would obtain as projected (and was otherwise unconstrained). Afterward, drop the number to account for risk, or simulate with various scenarios to capture risk.
2. Constrained Optimization
This is a factor that classic economic valuation models like Erwin's totally ignore, and I think it explains the “small stock” conundrum. The problem plaguing small stocks is not just that they often disappoint with a resulting stock crash, but that they are likely to crash whether or not they disappoint. WIND crashed in October and January on the basis of rumors, both of which proved untrue. (A couple of analysts have suggested that the second “slow down” fear had a smidgen of truth in it, but not enough to keep the company from beating its numbers once again.)
What this means from the point of view of a professional investor (institutions, funds, etc.) is that, irrespective of the credibility of future earnings, there is risk in owning the stock if it can crash 50% in one day, and afterward the market can choose to ignore classic economics for an indefinite period of time. Almost all professional investors not only seek under-valued stocks (by whatever technique they determine value), but their results will be compared to benchmarks on a daily, quarterly and annual basis. This comparison, which is much more important to them than hitting a multi-year home run, mathematically constrains their pursuit of optimal stock selections to combinations that preclude falling much behind their chosen benchmarks. Small stocks that can crash on a whim become poor choices for them. And small stocks that are caught in the maelstrom of even totally undeserved market mistreatment remain unattractive, almost no matter what classic economics suggests is appropriate pricing.
The good news is that the individual investor able to withstand the stress of a stock out of favor, and who is not constrained by benchmark comparisons, stands to rake in tremendous rewards by ignoring these optimization constraints and relying strictly on classic economic valuation.
This is why the comparison to MSFT, KO and the S&P do make sense. The P/E accorded those titans provides an understanding of the market's appreciation of future earnings flow when the short-term performance constraints are lifted and projected earnings are believable. If the short-term performance constraints do not affect the company's business performance, then they can be ignored when valuing a prospective company to buy.
The remaining issue concerns the possibility that short-term performance constraints do affect the company's business performance. That is, does treating a stock like a dog makes the company into a dog? This, by the way, is my only concern about the market's treatment of WIND over the last couple of years. So far, there have been no signs that the company is suffering from the market's disrespect. However, should management detect significant problems, with customers or obtaining and/or retaining employees, then the situation will have to be corrected.
Allen |
