Investing 102: The Cyclically Adjusted PE Ratio

Six years into a bull market for U.S. stocks, you may be wondering: Is the run over? The financial news is peppered with references to the stock market being overpriced and ripe for a correction, so it would be natural for an observer to worry.

I thought I would take a look at some of the data used in these predictions to see what I could learn. One of the measures often cited by those claiming the market is overvalued is the cyclically adjusted price-to-earnings ratio. This is also called the CAPE or P/E10. I call it the PE ratio’s big brother.

 
What is the CAPE?

In a previous post, I discussed the price-to-earnings, or PE, ratio. As a refresher, here is the basic equation:

PE = Market price/Net income

Robert Shiller, a Nobel Prize-winning professor of economics at Yale University and author of the book “Irrational Exuberance,” created the CAPE ratio. CAPE attempts to address one of the major flaws of the PE ratio: A company’s earnings (its net income) can be highly variable in the short term. To correct for this variability, the CAPE uses a 10-year average for earnings, with each year’s number adjusted for inflation. Theoretically, this reduces the volatility in the year-to-year numbers and captures at least a full business cycle’s worth of reported earnings.

Here’s the CAPE formula:

CAPE = Market price/10-year average earnings

Since the earnings figure used in this ratio covers a greater period of time, CAPE has a tendency to be much less volatile than the traditional PE ratio. It also has a higher correlation to 10-year investment returns than the standard PE does.
One of the reasons so many people view the U.S. market as overvalued is that the current CAPE is 26.7, compared with a historical average of 16.6 (since 1871). This puts the CAPE in the top 10% of readings ever recorded. This certainly seems to indicate that the U.S. market is expensive.

What do the data tell us?

I like to review source data when I read articles about these kinds of things to make a judgment as to the strength of the claim. Fortunately, Shiller is kind enough to post his data online so we can look at the information ourselves.

First, I decided to see whether elevated CAPE levels were followed by poor long-term returns. I identified every month in which the CAPE was above the current 26.7. Then I looked at the average returns 10 years later. Since 1871, the average 10-year return following a month with a CAPE above 26.7 was negative 14.75%, or about negative 1.5% per year. This does not include dividends paid over that time frame.
It certainly seems like the negative outlook for long-term returns is justified in the data. However, critics of using CAPE as a forecasting tool point to a few things:

• Industry in the United States has changed dramatically since 1871 — or even 1971, for that matter. Early industries such as mining, oil, railroads and manufacturing were heavily capital-intensive. Over time, the U.S. economy has shifted to much less capital-intensive industries such as services and finance.
• Accounting standards are much different today — and much more strictly enforced — than they were decades or a century ago. As a result, the earnings measurement isn’t consistent.
• Because of the preceding arguments and others, modern companies are more profitable and less capital-intensive. As such, they are less risky and may deserve to trade at a higher multiple of earnings.

To test these theories, I took a look at returns only since 1985. In that case, the average CAPE is 23.5. Subsequent 10-year returns for stocks above our current CAPE of 26.7 were negative 9.5% on average, or about negative 1% per year. With dividends reinvested, you probably would have made a small profit over that time frame.

Another thing that jumps out at me is just what a small data set this actually is. Although 144 years seems like a long time, it gives us only 1,733 monthly data points to look at. Since 1985, that number is more like 360. I’m sure these results have some statistical significance, but it is a relatively small data set to make conclusions on.

What does it all mean?

Here’s what I take away from this analysis:

1. The CAPE data indicate that returns in the U.S. market have been low historically following highs like we’re seeing now. Unfortunately, history does not give us any great guide. These types of valuations have resolved themselves in the past with big market drops, as well as with back-and-forth “sideways” markets for extended periods of time. There have even been times when the 10-year returns were significantly positive.
2. The amount of data under analysis is much smaller that I would be comfortable using to make firm future predictions.
3. Data from the past 30 years indicate a much higher average CAPE than the long-term historical average.
4. The amount of data available is unlikely to prove that CAPE will remain permanently elevated because of structural changes in the U.S. economy. In other words, we don’t have enough data to conclude that “it’s different this time.”

As you can see, this is a much more nuanced view than the standard headline of “The Market is Expensive — When Will it Crash?” Even so, I think this can still give some useful guidance:

• Since 10-year returns have historically been low following CAPE levels as high as they are now, we may want to dial back our expectations. It may be prudent to bump up your savings rate a little to help compensate.
• This highlights the need for diversification among different asset classes in addition to U.S. stocks. The good news is that on the CAPE measure, many non-U.S. markets are not nearly as expensive.
• If you have not made any adjustments to your portfolio lately, you may want to check to make sure that your allocation across your different asset classes reflects your tolerance for risk — including a potential drop in your stocks.

Rather than pay attention to the oversimplified headlines calling the market over- or undervalued, I find it useful to dive into the details and try to draw my own conclusions. I hope you have found this valuable as well.

This article was originally published on Nerdwallet and Nasdaq.