Shiller’s Research Shows Why Stock Forecasting’s Record Is So Poor

Published on

The Buy-and-Holders got a lot of things right. I believe that they made a terrible mistake in thinking that valuation-based market timing is not required. I believe that Robert Shiller’s Nobel-prize-winning research showing that valuations affect long-term returns is legitimate research and that therefore all investors seeking to keep their risk profile constant over time absolutely must engage in valuation-based market timing. But, aside from that one, everything else that the Buy-and-Holders say about stock investing checks out, in my assessment. In particular, I believe that they are right that it is not possible to predict how prices will shift in the short term.

That’s an important insight. Investors don’t like it that price drops cause them to lose money and they want to be able to participate to the greatest extent possible in price jumps. So there is a great and widespread desire for informed takes on how prices will move in the next month or in the next six months or over the course of the next year. Because the demand for this type of information is so great, there are many articles published in which experts offer speculation as to where stock prices are headed and back up their claims with all sorts of rationales.

The Buy-and-Holders urge investors to ignore all that. They say that it is not possible to know where prices are headed. So all speculation in this area is “noise” that should be tuned out. I believe that the Buy-and-Holders are right about this. I believe that  they do us all a service by making this point so frequently and so forcefully.

But I do not believe that they possess a complete understanding of WHY it is not possible to forecast returns effectively. I believe that investors who do possess a complete understanding will be more likely to take the advice to heart.

Why stock forecasting do not work

There are two reasons why price forecasts do not work, according to the thinking of the Buy-and-Holders. One, all information affecting stock prices is quickly incorporated into the price. It’s not possible for an individual investor to know more than the market as a whole. And, two, price changes are caused by economic developments. Since it is not possible for an investor to know about economic developments in advance, it is not possible for him to know about price changes in advance.

It is of course possible for investors to anticipate economic developments that have not yet taken place. But such anticipation, to the extent that it is credible, would be incorporated into the price at the time that it appears in the minds of investors. For an investor to profit from such anticipation, he would need to act on the anticipation before the community of investors did so. At that time, the anticipation would not be sufficiently credible to justify action. The idea here is that the individual inverter cannot know more than the entire community of investors.

Shiller’s research changed our understanding of how stock prices are set in a fundamental way. Shiller agrees that short-term timing does not work and that most of what is known about stocks is incorporated into the price quickly once it becomes known. However, Shiller’s research points to an important exception to the general rule. When investors take economic developments into consideration in setting a new price for stocks, they are acting rationally. But there is one area in which investors are not capable of acting rationally – when they fail to take valuations into consideration in determining the value of their portfolio. That’s why it is called irrational exuberance. If they took valuations into consideration, they would lower their stock allocation to maintain the proper risk profile and the stock sales that resulted would pull prices back to their proper level.

So it is not quite true that stock forecasts do not work because all factors affecting price are incorporated into the price. One factor is never incorporated into the prices – mispricing on either the high side or the low side. If mispricing were considered, it would disappear!

Does this reality make it easier to forecast stock price changes? It does not! It makes it harder! The failure to consider valuations is irrational and irrational phenomena are much less predictable than rational phenomena. Shiller’s research shows why it is truly impossible to predict short-term price shifts. Such price changes are caused by shifts in investor emotion. How could anyone know in advance in which direction crazy emotion swings are going to go?

Shiller’s research shows that predictions of long-term price shifts can be made with a good degree of confidence (but certainly not with perfect confidence). Prices always must return to fair-value levels since it is a market’s core purpose to get prices right. So, anytime prices move far from their fair-value level, the long-term direction of prices is going to be in the direction of the fair-value prices.

So, yes, Shiller showed that some stock-return predictions are possible, a finding that conflicts with a Buy-and-Hold dogma. But Shiller’s model for understanding how stock investing works actually supports the Buy-and-Hold claim that short-term price forecasts possess little or no value.

Rob’s bio is here.