1. How to use GuruFocus - Tutorials
  2. What Is in the GuruFocus Premium Membership?
  3. A DIY Guide on How to Invest Using Guru Strategies
Holly LaFon
Holly LaFon
Articles (9180)  | Author's Website |

Brandes Investments Commentary - The CAPE Ratio and Future Returns: A Note on Market Timing

By Wim Antoons

December 04, 2018 | About:

Abstract

The cyclically adjusted price earnings (CAPE) ratio is often used to express the valuation of an equity market. This paper focuses on the CAPE of the U.S. equity market and finds it was very high in June 2018. Although a high CAPE has tended to suggest lower returns in the future, such forecasts only make sense for longer periods. CAPE has not been a reliable forecasting tool for short-term movements of the market and thus it is difficult to use as a market-timing indicator. Similar results were found for other developed countries. See Table 4 later in this report.

Explanation of CAPE

The CAPE ratio was developed by Yale University economics professor and Nobel Prize winner Robert Shiller. The difference between cyclically adjusted price/earnings and ordinary price to earnings (P/E) is that corporate earnings are inflation adjusted with CAPE. The CAPE ratio has proven to be much more stable than the classic P/E ratio. CAPE reflects the current market price relative to average inflation-adjusted profits over the preceding 10 years. I use the figures from Robert Shiller’s website2 that show the CAPE of the U.S. equity market from January 1881 through June 2018 (monthly figures). See Figure 1. The average CAPE over this period was 16.88. Figure 1 shows some outliers.

CAPE and Long-Term Future Returns: A Brief Literature Overview

Investors have used the CAPE ratio to get a sense of potential future returns. Before applying my analysis of the CAPE ratio and returns, I reviewed select, recently published articles addressing this topic. Here I summarize results from a sampling of my work. A January 2018 research paper, “CAPE Fear: Why CAPE Naysayers Are Wrong”3 discusses the many issues contemporary academics have against using CAPE to predict long-term returns. The issues include structural changes to earnings per share (EPS) growth, demographics, real interest rates and volatility. The paper concludes that all issues (aside from an aging workforce) are temporary in nature and that CAPE is still a good measure for predicting the long-term outlook for the market.

Another paper, “King of the Mountain: The Shiller P/E and Macroeconomic Conditions,”4 published in the July 2018 issue of The Journal of Portfolio Management studies the effect real interest rates and inflation have on the predictability of the CAPE ratio. According to this paper, adjusting CAPE for current interest rates and inflation can lead to an improvement in its predicatability of near-term capital market returns, which could lead to further research identifying other macroeconomic and market measurements that improve near-term capital market returns.

While history is not a guide for future returns and no one can predict future performance, I still believe a study of historical CAPE ratios and subsequent returns can be instructive.5 I believe a high CAPE ratio is most likely a time to expect lower returns going forward. In subsequent sections of this report, I will show why.

CAPE and Long-Term Future Returns: My Analysis

In his calculations of stock market performance, Shiller used the S&P Composite Price Index (USD). I did, too. I calculated annualized returns (both preceding and subsequent) for the CAPE ratio’s three highest and three lowest levels between 1881 and June 2018. I used 1-, 3-, 5- and 10-year periods for each of the extreme CAPE ratios studied. See the results in Table 1 and Table 2. Of course, I recognize that no investor could have known at these points in history that the CAPE ratio was at a historical high or low. I share them here as interesting extreme points in history and believe the returns associated with these moments can be instructive for investors today.

Table 1 shows the preceding returns (on 1, 3, 5 and 10 years) for six CAPE outliers. Not surprisingly, the very low CAPE ratios of 1920 (4.78), 1932 (5.57) and 1982 (6.64) showed preceding returns were mostly negative and sharply negative the year just preceding these CAPE bottoms. The high CAPE of 1929 (32.56), 1999

(44.20) and 2018 (32.11) showed preceding returns for all periods were more pleasant (and positive) and definitely very high in 1929 and 1999.

Table 2 shows the subsequent retuns for low and high CAPE extremes. Looking at the same dates of low CAPE ratios (1920, 1932 and 1982), it shows that subsequent returns for all periods were very strong. The one-year returns were spectacular for 1932 and 1982, showing that a recovery went very quickly. Subsequent returns in periods of high CAPE ratios (1929 and 1999) show the opposite. Even long-term returns over the next 10 years were negative.

While I’m highlighting six periods of extreme CAPE ratios in Tables 1 and 2, I also grouped all CAPE ratios into deciles and show the average preceding returns (on 1, 3, 5 and 10 years) and average subsequent returns (for 1, 3, 5 and 10 years) for each decile in Figures A1 and A2 in the Appendix. The pattern across deciles is clear: years with lower CAPE ratios reflected lower preceding returns but higher subsequent returns; years with higher CAPE ratios reflected higher preceding returns but lower subsequent returns.

Figures 2 and 3 show linear regressions for the relationship between historical CAPE ratios (monthly) and 5-and 10-year subsequent returns for the U.S. stock market (again measured by the S&P Composite Price Index). Each figure features a logarithmic trend line (Log. CAPE) designed to illustrate the “best fit” between these variables.

Although the R² of both graphs is not very high, the plot points clearly are more negative as the CAPE increases in both figures. It indicates that, historically, high CAPE ratios generally have been followed by low returns

in the subsequent 5- and 10-year periods and that low CAPE ratios generally have been followed by strong returns over subsequent 5- and 10-year periods. I believe the explanation for this effect stems from two influences: the value premium and mean reversion of returns. The value premium has been documented and has shown that inexpensive assets tend to offer higher future returns. The equity market has also shown a tendency to mean revert after periods of weak or strong performance.

Based on history, a CAPE of 31.79 (as of June 2018) suggests a future annualized 5-year return around 0.0%. Figure 4 shows the subsequent 5-year annualized return for the U.S. market when the CAPE ratio was above 32 (again, from 1881 to 2013). There have been only 51 months out of 1,644 observations where the market reached this valuation level, so the market was only more expensive in 3.1% of the monthy observations. The subsequent returns for five years were centrered around 0.0%. At an extreme, after September 1929, the subsequent 5-year annualized return was -22.27%. (Note: the R2 in Figure 4 is only 0.04, but it would jump to 0.6 if that Sept. ’29 outlier were removed.)

CAPE as a Market Timing Tool?

Can CAPE be used as a market timing tool? My previously published research shows that market timing is nearly impossible and is a losing strategy in the long run.6 Figure 5 shows the relationship between CAPE and 1-year subsequent returns. The picture is indeed different from the long-term future returns. No matter whether CAPE is low or high, the return of the next subsequent year (unlike the longer-term returns) was less dependent on CAPE. The R² of the regression analysis is near zero, also indicating no relationship between CAPE and subsequent 1-year returns.

Read more here.

About the author:

Holly LaFon
I'm a financial journalist with a master of science in journalism from Medill at Northwestern University.

Visit Holly LaFon's Website


Rating: 0.0/5 (0 votes)

Comments

Please leave your comment:


Performances of the stocks mentioned by Holly LaFon


User Generated Screeners


wilhelmstone10yr div gainer
DBrizanall 2018dec09 1048p
yankees96Income +++
pinalpatel2121blackberry
yankees96Income ++
yankees96Income +
DANGORDON12-9-18 REV INC PEG
merknt84russia1
ramfusPeter Lynch + 1 Guru Lax
Vincent ChiaDividend Stocks USA
Get WordPress Plugins for easy affiliate links on Stock Tickers and Guru Names | Earn affiliate commissions by embedding GuruFocus Charts
GuruFocus Affiliate Program: Earn up to $400 per referral. ( Learn More)

GF Chat

{{numOfNotice}}
FEEDBACK