Rule Breaker - Strategies
Trying To Figure Out Valuation

Format for Printing

Format for printing

Request Reprints


By TMFBuster
March 7, 2001

Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light. How are these posts selected? Click here to find out and nominate a post yourself!


You're lucky you got howardroark and not me (assuming that I am the Buster to whom you referred, although I'm not sure, since I'm not a master at much beyond knowing exactly where to tickle my daughter). Anyway hr's the man you want on this topic and he captured it all better than I ever could.

Nonetheless, I can never resist adding my opinions, especially when they're unnecessary...

In my relatively brief time as an investor, here's what I've learned about Discounted Cash Flow or really, any attempt to value a company's stock:

1) It's always worth attempting. The exercise of running the numbers forces me to consider -- and make guesses about -- some elements of the business that I may not have considered as carefully before attempting the model.

Even if the final answer isn't of much use -- as often happens to me -- the process can't help but give you a better feel for the big picture. More information is always better than less, as long as you don't fall prey to the dark side -- seeing your educated guesses -- or anyone's, for that matter -- as fact.

2) As a statistician, I was trained to loathe one-number summaries, and this habit has served me well in investing. When someone says, say, "XYZ is fairly valued at $35," this kind of statement is next to useless. Just as no sensible person should digest a point estimate for an average without knowing anything about how the data were sampled, how many data there are, etc., a one number summary of value should never be accepted by itself.

Think of valuation intervals in terms of statistical confidence intervals. Confidence intervals are widely used, and valuable, even though they are nearly always based on only roughly true assumptions. The act of adding probability to the final statement, however crudely, just adds an important second dimension to the picture.

With stock investing, there is the added uncertainty that comes from the impossibility of collecting representative samples, since nobody can "fairly" draw business results from the future. In the same way, though, adding intervals around your answer allows you to press forward to a better understanding (as opposed to just throwing up your hands), without misleading yourself as to the precision of your knowledge.

It's pretty easy to setup a spreadsheet with key variables like earnings growth and discount rate entered at the top, as spreadsheet variables, so you can look at ranges of inputs and the associated ranges of outputs. I can send you mine if you're interested (as long as you recognize fairly the imprecision of my knowledge :)

3) The hardest thing about DCF, in my humble opinion, is not the math, but understanding exactly what you're doing when you adjust the discount rate. The purists will tell you that it should be adjusted according to risk in the investment, but assessing risk, in my experience is rarely a "closed form" exercise. Whatever your personal belief about risk -- whether it be price volatility or the odds of business failure or some confidence interval around net present value -- make sure the discount factor you choose is in line with this broader view.

4) Estimating the value of stocks is inherently imprecise. Because market pricing evolves, naturally, to eliminate consistent bias, it's almost axiomatic that precise valuation models will never emerge. There is no shortage of smart minds out there with awesome computing power at their behest, creating complex models -- from traditional forward looking, business-focused DCF-type exercises to backward looking, often market focused, mechanical studies. Most of these people, I'm willing to guess (with some imprecision) don't do any better, on average, than schmos like me and you. :)

5) Complexity is no necessary indicator of model value. Anyone that's created mathematical models for a living gets to the point of going round and round and round. Eventually, the best models chose some subset of the complex whole -- one that jives with the modeler's intuition, (improved as it usually is by the exercise).

Like chess, good models usually take a short time to learn and a lifetime to really understand. If somebody tries to browbeat you with their model, and make you feel stupid by dressing it up in a lot of fancy mathematical symbols, chances are it's not worth much.

6) I don't think the TMF position has ever been "ignore value." In fact, most of the writers and editors around here are "value hawks," at least from my perspective :)

Admittedly, Tom and Dave have coached a strong de-emphasis of value, but I don't think the idea has ever been to demean it as much as it's been to unmask it for what it is -- an inherently imprecise exercise -- that is often passed off on new investors as some kind of magic formula that only the true gurus can manipulate. Ultimately, valuation is done best when it's informed by a deep understanding of company fundamentals, so I don't think the Tom/Dave approach is really all that much at odds with the traditional valuation approach.

It all comes down to how you see the game. One extreme is to focus exclusively on "bargain hunting." The other extreme is to ignore the price entirely and focus only on business results. I think most sensible investment strategies lie somewhere in the great middle.

Tom and Dave are surely further to the business results side than most, and they've trumpeted this position loudly, I think, because it's traditionally been under-represented in the mainstream financial media. But I don't think they are the non-value extremists that they are often painted as (or even paint themselves as).

7) My own personal compromise, for what it's worth, has been to focus most of my research on the business side of investing, looking for great companies before getting bogged down with stock price and valuation exercises. And especially in the case of Rule Breakers, which move with so much volatility, often in the absence of any meaningful business information, I don't know that it's possible to consistently pick lows points for getting in and high points for getting out, without getting buried in trading costs and/or missed opportunities.

But, ultimately, I think you have to do some kind of valuation exercise before you buy a stock. If the price is beyond the bounds you generate using the most optimistic of scenarios, according to your research and opinions, I can't see buying it, regardless of how good the business plan looks. Conversely, if a great company is, in your estimation, going for a song, relative to the bounds you generate, I think this has to give you incentive to buy or buy more of the stock.

David may not use the "v" word much, but he does like the 10x/5y criterion, which is a value criterion, any way you slice it. To say he ignores value, then, is I think unfair. What's more fair, I think, is to say that he considers it a secondary, rather than a primary investing tool.

In particular, I think we, as a portfolio, are most unusual in that we make no attempt at all to play macro-economic cycles. I'm not convinced, personally, that these cycles can't be played profitably. I'm just not sure that they can be either, and I don't see any mention of them in the Rule Breaker Strategy, as it is written.

To note that Breaker stocks have tumbled in a bear market is to say that Minneapolis is colder than Baton Rouge in February. Since the strategy has no contingency for getting out before a bear market, it will always suffer in them. That much is predictable.

What remains to be seen is whether the folks who try to get in during the bear and take profits just before it returns can do better, or whether they'll just end up missing the peaks and valleys and getting buried in trading costs. I don't pretend to have the answer, nor am I sure that anyone ever will, but this strikes me as the real question.

With that, I'm waaaay off topic.

Read howardroark's answer. It's all there.