Valuation-Informed Indexing #331

by Rob Bennett

Just about everyone agrees that valuations matter. But how? That one has just about everybody stumped.

My view is that we should look to the historical return data for guidance. We of course cannot know the future perfectly. But we have good data for 145 years of U.S. stock market history. Why not use that data to form general opinions as to what role valuations will play in the future? In all other fields of human endeavor we look at how things have always worked in the past to form assessments of how things may work in the future. It makes sense to proceed in a similar market when forming stock investing strategies.

Unfortunately, taking this approach seriously leads to some shocking findings. Every time we have seen the P/E10 value go above 25, we have seen an economic crisis arise in the wake of that out-of-control bull market. On the one occasion when we permitted the P/E10 value to rise to 33, we saw a Great Depression. We went far beyond 33 in the bull market that ended in early 2000; this time we saw the P/E10 level touch 44, uncharted territory. If the stock market really does continue to perform in the future at least somewhat as it always has in the past, investors are in for a world of hurt. And of course the loss of the spending power that disappears from our economic system when the market crashes always causes devastating consequences even for those who either don’t have assets to invest or who move to other asset classes when stock prices reach insanely dangerous levels.

I can count seven others ways of thinking about valuations among people who cannot bear to come to the conclusion that I have come to, that profits attributable to overvaluation rather than true economic growth are always reduced to cotton-candy nothingness in the long term.

One, there is a tiny group of people who truly believe that the market is efficient, that is, that the idea that the market can ever be overvalued in a meaningful sense is a logical impossibility. I don’t agree with these people but I admire them for their intellectual integrity. If overvaluation exists, the market is not efficient. If the market is efficient, overvaluation does not exist. For those who truly believe that the market is efficient, Buy-and-Hold really is the ideal strategy and I of course wish those in this group the best of luck with the strategy that makes perfect sense for them.

Two, there is a large group (probably the largest group of those listed here) that believes that valuations matter but that it is not possible to profit from that reality by changing one’s stock allocation in response to big valuation shifts. Investors in this group stick with the Buy-and-Hold strategy not because it possesses theoretical purity but because they believe that as a practical matter it is the best possible choice. I see the thought here to be similar to the thought behind Winston Churchill’s observation that democracy is the worst possible political system except for all the others that have been tried. Members of this group appreciate that valuations MUST matter since price obviously matters when buying anything other than stocks that is available for purchase but for reasons that generally are left vague (even in their own minds) they don’t make an effort to think through why stock purchases are the one exception to the otherwise universal rule.

Three, there are a good number who believe that valuations really have always affected long-term returns in the past but that it is not safe to use historical data to predict future returns because the economic realities have changed too much over the years for old data to remain relevant. The concern here is a legitimate one. The United States is obviously a very different place from what it was in 1870. However, I have a big problem with this take on things. If we throw out the idea of using historical data as a guide re the valuations matter, shouldn’t we throw it out for all other purposes too? There goes the idea of using academic research for guidance on any investing strategy question, no? Which means that Buy-and-Hold goes too, no? I hate the idea of giving up on the idea of using research as a guide to effective long-term investing, largely because I so much admire the Buy-and-Holders for the hundreds of powerful insights they have developed on non-valuation questions.

Some believe that valuations matter and that the effect of valuations on returns can be effectively measured but still resist doing so re their own portfolios because it’s not the popular thing to do. It feels funny going against the crowd re something as important as planning one’s retirement. This isn’t an intellectually defensible position. But it is one held by a large number of smart and good people. I don’t believe that it would be proper to ignore it.

A fifth take is that no one can know how valuations affect long-term returns. This strikes me as a defeatist view. I wonder whether investors in this camp sense the weakness of other positions and move into this camp so that they won’t have to defend their views in spirited debate. We all have a Get Rich Quick urge residing within. Once we find a way to rationalize ignoring valuations, we can give it free rein without having to acknowledge the contradiction of also maintaining that we believe in using research to guide our investing choices and that we have adopted Buy-and-Hold strategies as a result.

Some believe that valuations could have been used in the past to effectively predict long-term returns but that our technological advances have put us in a new era in which the old rules no longer apply. There is of course no way to disprove this. It’s a “this time it’s different” solution to the puzzle.

Finally, there are investors who believe that valuations predict long-term returns but that lots of other factors also predict long-term returns and that it would be a complicated business trying to sort out the various factors. My objection to this seventh way of thinking about valuations (other than my own, which would bring the total count to eight if it were included in the listing) is that it rejects the key Buy-and-Hold belief that all of these factors are incorporated into the market price that applies at any given time. I believe that the Buy-and-Holders are right that we don’t need to look at factors other than valuations because those factors are baked in but that the valuations factor is the one exception to the general rule because mispricing is caused by irrationality and irrationality disappears once those guilty of the irrationality become aware of it.

Rob’s bio is here.