A community member named “Chipmunk” makes an important, common-sense point in Community Comment #84 in the article entitled Investing Discussion Boards Ban Honest Posting on Valuations!. Here is what Chip tells us:
“I’ll venture a regular guy opinion. I will say that Rob’s explanation does make sense. However, no one can predict the future. Just when someone uses history to try to backtest, create a model, and then predict the future, the factors change or new factors emerge, thus making the model useless for predicting the future.”
Is he right?
Yes, to a large degree he is right. There is no backtest that can tell you what stocks are going to do in the future. It’s dangerous to think that there could be such a thing. Chip’s common sense has generated for us a warning that we all need to keep in mind.
But what about the Return Predictor? Isn’t that the product of backtesting? When I agree with Chip, aren’t I rejecting my own tool for learning how to invest successfully?
I don’t think so.
Chip is absolutely right that we are going to see new factors emerge in coming years that will affect how stocks perform. The Return Predictor looks only at how stocks have always performed in the past. It cannot look into the future. It cannot anticipate factors that have not emerged yet.
Where Chip goes wrong is in saying that, when new factors emerge, it will render the Rational Investing model (the Rational Investing model posits that valuations will always affect long-term returns) “useless.” When new factors emerge, it may be that we will need to make adjustments to the Predictor. It may someday even mean that we will need to develop an entirely new tool. It will not mean that we will need to reject the model by which today’s tool was developed. If it is true today that valuations affect long-term returns (I strongly believe this to be the case), then this will always remain true.
The question of whether valuations affect long-term returns is fundamental. We need to be able to answer that question correctly to come to possess any reasonable understanding at all of how stocks work. This always has been and always will be a critical question. Lots of new factors have come to our attention in the past and our process of learning about these new factors did not affect the importance of the critical factor. The emergence of new factors in the future will not affect this core matter either.
The Predictor is like a weather report. A weather report does not tell you for certain whether it is going to rain tomorrow or not. It tells you the odds. It lets you know whether you need to carry an umbrella or not. A weather report does not tell you everything you could possibly want to know about the weather. But it does offer valuable information bits. The same is true of the Predictor. It informs you as to how stocks work. It does not tell you all there is to know and it cannot tell you all there is to know because we fallen humans are not capable of knowing all there is to know.
Can we predict tomorrow’s weather? Not perfectly. But we are capable of providing valuable information bits about tomorrow’s weather. Those planning an outdoor event for tomorrow afternoon would be foolish not to take a look at the weather report. Those investing in stocks to finance their retirements would be foolish not to look at the Predictor as one tool for use in determining whether their current stock allocation is appropriate or not.
I don’t know what goes into the development of a weather report. But I am confident that there have been changes over time in the process by which weather reports are developed. The people who study the weather learn new things over time. As they learn things, they need to incorporate what they learn into their models. We’re always making small changes in how we go about predicting the weather and over time the effect of the accumulated changes can be significant.
It is going to be the same with our development of tools to predict stock returns. There is today no other tool on the internet that does what the Predictor does (my guess is that other such tools have been developed that are not publicly available, but that is just a guess). I believe that there will come a day when there will be hundreds of web sites that will offer such tools. There will be grand debates about how such tools should best be constructed. Over time we will learn new things and it will become possible to predict stock returns more effectively than we can today.
The Return Predictor is state of the art today. It is not always going to be state of the art. Chip is right that there are going to be new factors or that we are going to discover factors that were always there but that we do not know about today. That’s just the way life is. I don’t think we have to see it as a bad thing. It does indeed follow from a belief that life gets better that life is not perfect today but the positive way to look at it is that it is today’s imperfections that make it possible for life to get better.
Chip is living in the black-and-white world. He is thinking that, if the best tool available to us today is not perfect, we should not make use of any tool, we should give up on the idea of predicting returns. I see that as a dangerous mistake.
The reality is that we all predict returns. Those who never look at the historical data have some idea of what return they are going to obtain from their stock investments. If they didn’t have any idea whatsoever what to expect, they wouldn’t be able to bring themselves to put money down on stocks.
The difference is that those who use what the historical data tells us about how stocks have always performed in the past are basing their predictions on something rational, something objective, something solid. They are not just taking guesses, they are not permitting the emotions of the moment to exert too powerful an influence over their investing choices.
No one can predict the future perfectly. That’s so. But we all can make an effort. We all can learn what is available for us to learn. We all can try to invest more successfully in the future than we have in the past. The reality that we can never do this perfectly is no excuse for abandoning reason. To fail even to take into consideration what the historical data says is to abandon reason. That’s Passive Investing.
Today’s Passion: A recent moderator’s decision at the Early Retirement Forum shows how the ban on honest posting on SWRs and other valuation-related topics has remained in place despite the widespread community support for a lifting of the ban. The moderator closed a thread on “Tomorrow’s Market” that showed promise of helping lots of early retirees get a better fix on the realities of investing in stocks at today’s prices. Why did he do this? Because two Goons showed up on the scene and made every community member present uncomfortable with their ugliness. Why weren’t the two Goons removed so that the rest of us could have the discussions which we intended the forum to facilitate when we built it? Only the moderator who made the improper decision knows the answer to that one. What the two Goons learned is that they can get any promising discussion shut down by engaging in highly abusive behavior. They won’t forget the lesson taught them by the board moderator anytime real soon. And neither will the community members who put time and effort into taking the thread in a postive direction. Boo, baby!