The next 20 years?

"Looking at current 10-year returns gives me a sinking feeling in the pit of my stomach which I have experienced twice before, during the tech bubble and the financial crisis. Is there any data regarding significantly higher 10-year returns being negatively correlated with lower subsequent 10-year returns (like the tech boom followed by the 'lost decade')?"

My initial answer is that there is a loose inverse relationship between past and future 10-year returns, but so loose that one cannot make money off that information. This column provides the figures to defend--or refute--that assertion. In addition to showing the results of 10-year returns, I also examine five- and 20-year periods.

...

One could not hope for a clearer picture than what occurs with the 20-year chart. When blue is up, red is down, and the converse. There are only brief stretches when the two lines appear on the same side of the median, and in those instances the returns are close to normal, so the signal doesn’t flash brightly. The graph looks so clean that one suspects that its author erred. (The thought did initially cross my mind.)

This portrait is dominated by five events:

1. The depression/war years (weak results)
2. The 1950s/1960s (strong)
3. The 1970s oil crisis (weak)
4. The 1980s/1990s (strong)
5. The 2000s (weak, in fact considerably worse than the 1930s)

Consequently, these results not only lack statistical significance, because the 648 monthly observations so thoroughly overlap, but they also fail common sense. That buoyant economies generate optimistic stock valuations, which eventually decline as the economic news worsens, makes sense. But that such events have occurred on a seemingly regular cycle is surely accidental. The pattern’s apparent inevitability is a mirage, based on a tiny sample size. 

That said, I suspect the 20-year numbers offer a fair guide to the future, if not as accurately as their negative 0.84 correlation suggests. (Now that’s a correlation!) Secular economic changes do tend to occur gradually, and investor emotions can overshoot the mark. Although the letter of this finding need not be observed, its spirit deserves some respect.

Which implies good news for the U.S. stock market, as entering 2020 the real 20-year return on equities was a modest 3.85%, well below the historic norm. These days, it has become commonplace to bemoan high stock prices. Perhaps the skeptics will prove to be correct. But 20-year return measure foresees a happier outcome.

-- John Rekenthaler