The Unpredictability of Short-Term Return Sequences Masks the Predictability of Long-Term Returns

Updated on

Valuation-Informed Indexing #262

by Rob Bennett

I wrote last week about how short-term return sequences, unlike long-term returns, are unpredictable. This reality causes a lot of confusion for investors seeking to make sense out of how the market works.

Say that it’s 1982 and you have just learned about Robert Shiller’s research showing that long-term returns are highly predictable. You have been advising people for years to follow a Buy-and-Hold strategy and you are questioning whether that is a good idea on a going-forward basis. It’s easy to get caught up in the short-term realty that the P/E10 level is 8. You think to yourself: “We are unlikely to ever again see stocks selling at fair-value prices much less at prices where valuations could create any dangers for long-term investors.” You silence those doubts and continue advocating Buy-and-Hold because your mind focuses on the unpredictable short term rather than the predictable log term.

Or say that it’s 1987 and stock prices have recently crashed dramatically and then recovered quickly. You jump to the conclusion that the Buy-and-Hold advice to always “stay the course” is good advice because you are more impressed by the short-term recovery than by the fact that it was not high valuations that caused the crash and that thus there was never any reason to believe that the crash would be consequential in any event.

Or say that you are Jack Bogle and it is the early 1990s and, after you warned investors about high valuations, you saw prices shoot even higher. You conclude that valuations don’t matter much not because this has been proven over a long period of time but because one short-term prediction did not come through. Your focus on the short term, which is the result of an inclination that is common to all of us, fooled you into drawing too big a conclusion from the playing out of a single return sequence.

Or say that you are an investor who listened to Robert Shiller’s advice in early 2009 that stocks would not be safe again until the P/E10 level dropped below 10 and you have not been out of stocks for six years of high returns and feel that you “missed out” on a nice run-up. Are you right to feel that way?

I don’t think so. I said in a podcast recorded in early 2009 that Shiller was being a bit incautious in suggesting that stocks were not a good buy when the P/E10 level dropped to 13. That’s below fair value. Stocks offered a very strong long-term value proposition at that time. So the idea of buying stocks or even moving up to a moderately high stock allocation made a lot of sense at that time.

However, it’s not quite right to say that investors who stuck with low stock allocations “missed out” because the return sequence that followed sent prices skyrocketing. Shiller was not wrong to believe that we will see a P/E10 of 10 or lower before this secular bear market comes to an end. There’s never been a secular bear market that ended before the P/E10 level dropped below 8. So most of the gains that have been experienced over the past  six years are likely temporary gains. You see them reflected on your portfolio statement today but you will likely see them disappear sometime over the next year or two. You might end up wishing that you had “missed out” on the stock crash that is probably going to scare millions of investors near to death.

We can predict some things. We cannot predict others. We need to take advantage of the things that we can predict while being careful not to get so caught up in the predicting game that we put our money at risk trying to make the sorts of prediction that cannot be made successfully.

We knew is early 2009 that stocks represented a strong long-term value proposition. And we knew that it was possible that a return sequence would show up in which the march downward to a P/E10 of 8 would play out quickly and that we would thus be happier remaining out of stocks. But we also knew that it was possible that a return sequence would show up in which the P/E10 level would return to the mid-20s and only fall back to 8 six or seven years later. Any strategic choice that was based on a belief as to which of those return sequences would play out in real life was a strategic choice based on a guess.

But so what?

It is always the long-term predictions (which can be made with a good bit of success) which matter. The short-term predictions (which cannot be made with much success) don’t matter much. It all works out nicely.

We could not say in early 2009 whether prices were going to continue their march downward or temporarily return to the mid-20s and only continue their march downward a number of years later. It’s not such a terrible thing that we did not know. We knew that stocks were at the time a far more risky asset class than is usually the case because we knew that we were in the early years of a secular bear market. Stocks are always a very dangerous asset class in the early years of a secular bear market. So we knew what mattered. Guessing either way on the less important thing was fine for investors who understood that getting the important thing right was what mattered.

Short-term predictions rarely work. Long-term predictions always work. The trick is to combine these two insights in building an effective and low-stress strategy for long-term investing success.

Rob Bennett’s bio is here.



Leave a Comment