Valuation-Informed Indexing #135
by Rob Bennett
Chuck Carnevale recently posted an article to the Seeking Alpha titled Shiller P/E Continues to Mislead Investors. He’s not a fan of Valuation-Informed Indexing! Yikes!
Actually, this is good news.
I don’t agree with most of the points made in the article. But the appearance of the article is encouraging. I believe that we need to convert the entire world to Valuation-Informed Indexing. That won’t happen as a result of me posting articles praising the concept. It can only happen as the result of a national debate in which people who support the concept argue to the best of their ability in support of it and in which people who see flaws in the concept argue to the best of their ability to opposition to it. Canavale is advancing the debate by trying to shoot the concept down. For that we all owe him our gratitude.
A second reason why we should all be happy to see this article is that it shows that the Buy-and-Holders are getting nervous. I was around in the day when the Buy-and-Holders were so sure of themselves that they wouldn’t go to the bother of shooting down Shiller’s ideas. They just ignored them. The appearance of this article is evidence that at least some Buy-and-Holders are opening up their minds a wee bit. Arguing against Valuation-Informed Indexing is the first step in a long journey that ends with people like Chuck Carnevale becoming big-time advocates. He (and lots of others) has to got to work through his objections before he will permit his mind to be won over by the new ideas. By airing his reservations he is giving those of us who support the concept an opening to try to win him over.
Let’s grab it!
Chuck begins by making a point that I know from years of personal experience is on the mind of just about every Buy-and-Holder who has ever heard the terms “P/E10” and “Valuation-Informed Indexing.” He says: “Allow me to start this article by emphatically stating that I have a real problem with forecasting stock markets in the general sense.”
This is why Buy-and-Holders are turned off by Shiller and his return-prediction tool. To them it is a bunch of mystical, magical, mumbo-jumbo. It has been my experience that, once a Buy-and-Holder is able to get past this deep-in-the-bones objection, he soon experiences his epiphany and all the other objections fade into nothingness. So we want to listen closely to see if we can get a fix on WHY it is that Chuck (and many, many other Buy-and-Holders) find it so hard to believe that predicting long-term returns is really possible and in fact a prerequisite for long-term investing success.
He tells us: “I prefer to forecast the intrinsic values of individual businesses based on their earnings-justified fundamental values.”
My reaction is that Chuck is seeing a conflict where none exists. Valuation-Informed Indexers believe that long-term returns of a broad index can be effectively predicted. We certainly don’t say that earnings-justified fundamental values of individual companies should not be explored. The suggestion here is that there is something “non-fundamental” about valuations. Again, the not fully spoken idea is that anything relating to predictions must be some sort of mumbo jumbo. My take is that valuations are one of the fundamentals that investors always must be taking into consideration.
Carnevale evidences a big lack of understanding (in my assessment!) of why Valuation-Informed Indexing works in his second paragraph. He says that, while predictions re individual companies “can be done within a reasonable range of rational probabilities...trying to estimate the collective results of a large group of companies such as the S&P 500 is a very daunting task.”
No! It’s just the opposite.
Predictions re individual companies are indeed possible. Warren Buffett and those who follow his Value Indexing approach make them all the time. But making those predictions is hard word (hard work that often results in a big payoff, to be sure). Making predictions that apply for a broad index are a piece of cake in comparison. Carnevale worries that “there are just too many variables and too many data points to contemplate” when looking at a broad index. If that were so, I couldn’t have any involvement with this stuff. I am a simple man with simple dreams. I follow Valuation-Informed Indexing strategies because Value Indexing is too much work for a Bear of Little Brain to contemplate.
The magic of P/E10 is that it is the one bit of information that is not “priced in” to the current S&P value. None of those “many variables” matter! They have been taken into consideration by other investors and the Valuation-Informed Indexer need not worry his head about them. P/E10 is the exception. P/E10 tells us to what extent the index is mispriced. Mispricing by definition can never be priced in. So that one number carries a huge punch. That one number tells you by itself whether the index at the time offers a strong long-term value proposition or not.
Carnevale goes on to argue that “mathematically speaking the 10-year average of an advancing number will most often calculate earnings to be lower than they actually are.” Thus, he believes that “for the great majority of the time, the Shiller-calculated P/E ratio will generally indicate that the market is overvalued. Consequently, investors who buy into this thesis will generally tend to avoid investing in stocks.”
I don’t see it.
My sense is that Chuck is being overly swayed by the high stock valuations that have applied since 1996. It is true that the Shiller P/E10 has argued against a high stock allocation for 17 years running now. For that time period, it is true that investors who bought into the thesis generally tended to avoid investing in stocks. But stocks have never before in history been as overpriced (according to the P/E10 metric!) as they have been for the past 17 years.
The argument here is circular. Carnevale is saying that P/E10 has scared its advocates away from stocks for 17 years. That’s a solid statement. But he assumes that that has been a bad thing. Why? Because he doesn’t believe in the power of P/E10! If he did, he would be expecting a 65 percent price crash sometime over the next few years (we have never seen a secular bear market come to an end without first seeing stock prices fall to levels 65 percent below where they stand today.). If we see a 65 percent price crash, no one will be saying that P/E10 steered investors wrong over the past 17 years.
Finally, the article cites a number of time-periods over that 17-year time-period in which P/E10 argued for a low stock allocation and yet stocks performed well in the following years. The facts are of course stated accurately. The problem here is that the argument being made is rooted in a misunderstanding of what P/E10 is supposed to do. P/E10 is not supposed to predict stock returns one year out or three years out or even five years out. P/E10 only permits effective predictions of what will happen to stock prices ten or more years out.
Go 10 or more years out and you see that P/E10 ALWAYS works. For 140