Robert Shiller has a free online database of historical stock market data that I’ve been using for years.
Shiller has data on historical interest rates, dividends, earnings, inflation, and valuations dating back to 1871.
His preferred valuation measure is the cyclically adjusted price-to-earnings ratio (CAPE).
The average CAPE ratio since 1871 is 17.4 times the past 10-year inflation-adjusted gains for the US stock market:
We’re talking over 150 years of data here, so when it comes to averages, it’s a very long-term period.
Looking back to 1990, the CAPE rate from just over 400 monthly observations has only been below the long-term average for 22 months. That’s about 5% of the time.
However, these are not valuations at screaming buy levels, but only just below average.
Between 1990 and 1991 there was a 12 month period of below average multiples. Only in the period from 2008 to 2009 did the ratings fall below the long-term average again.
So if you waited until valuations were reasonable before buying stocks, you’ve had exactly two chances over the past three decades or more.
And since 2010, there hasn’t been a single monthly reading that’s been below average. In fact, since the end of 2009, there hasn’t been a single monthly reading below 19.6x.
After 2009, if the ratings were below average, you didn’t have a single chance to buy.
Already in 2010 the alarm bells were ringing because of too high ratings:
Here is Henry Blodget at the time:
As Professor Robert Shiller’s latest cyclically-adjusted P/E update shows, US stocks are now more than 30% overvalued, on a gain of 21. That’s more reasonable than the over 100% overvaluation in 2000, but is approaching levels of the three other bubble peaks of the 20th century: 1901, 1929 and 1966.
He wasn’t alone.
It’s funny looking back to the low-interest-rate period of the 2010s because it looks like stocks would thrive in such a scenario. But back then, people were saying that those low interest rates would be the cause of low yields (because everything was designed for the 10-year term).
And the Fed would trigger hyperinflation, not a bull market in stocks.
Remember PIMCO’s new normal of low interest rates, low growth and low financial market yields?
Well, they got two of the three right.
At the beginning of the last decade, I’ve listened to countless presentations from professional investors who have told me that valuations for US stocks are in the 97th percentile, or close to historical norms, and that we should expect much lower returns going forward. 1
Heck, I wrote about the psychology of lower returns back in 2014.2
The US stock market has been overvalued 95% of the time since 1990, but during that time it has risen more than 10% annually:
In the 2010s, the S&P 500 returned nearly 14% per year, even though people hollered about how overvalued it was throughout its rise:
And in the 2020s, a decade in which we’ve seen a pandemic, a 40-year high in inflation, two bear markets, and one of the most aggressive Fed rate-hiking cycles in history, the S&P 500 is up more than 11% a year gone up:
I know what you’re thinking – Ben, you’re crazy! Haven’t you read this 40-page Financial Analysts Journal research report showing the importance of valuations?!
Yes, I probably read it. I know the dates I’ve written about it many times (here, here and here).
I am not saying that this will continue. I’m not naive.
At some point, above-average returns lead to below-average returns. This is how long-term averages work on the stock market.
My point here is that in the investment community, myself included, we probably pay far too much attention to valuations.
Understanding the history of the financial market is a prerequisite for investment success.
But becoming a slave to data can become a liability if you don’t put it in context.
The funny thing is that the historical averages we now use for comparison purposes were completely unknown to 99% of the investors who were in the stock market before us.
They either didn’t have the data or the knowledge, or they didn’t want to understand these basics. Knowledge of reviews has probably lost people more money over the years than it has made them.
I’m not saying ratings don’t matter at all. They are probably more important for individual stocks than for the market as a whole, but in extreme cases (e.g. 1999) valuations do matter.
It’s just that markets rarely go to extremes. Most of the time we are somewhere in the middle between insanely cheap and insanely expensive.
People pay far too much attention to stock-level valuations.
There are many other factors that are more important than the ratings. Things like demographics, money allocation decisions, investor risk appetite, the proliferation of tax-advantaged retirement vehicles, the trillions of dollars controlled by financial advisors, the positioning of institutions, and more.
US stock market returns over the past 15 years are a wonderful example of how difficult it is to predict what will happen next.
Certainly no one could have guessed that the Fed would keep interest rates at 0% for so long. No one expected tech stocks to reach staggering heights. And no one would have guessed that a pandemic would see governments around the world spending trillions of dollars.
But maybe that’s the point.
Predicting the future is difficult, especially when it comes to the markets.
The stock market, in most cases, doesn’t care that much about historical averages.
Are reviews important?
The majority of investors would probably be better off ignoring them most of the time.
Michael and I talked about stock market valuations and more in this week’s Animal Spirits video:
Subscribe to The Compound so you never miss an episode.
Further reading:
Is that the top?
Here’s what I’ve read lately:
Books:
1To be fair, most of these people were trying to sell a hedge fund or an alpha-like strategy that didn’t rely on rising stock markets.
2Since I wrote this article, the S&P 500 is up 11.9% annually.
Comments are closed.