Past is Not Prologue - Are Valuations Irrelevant?

As markets rise, bullish market pundits are very prone to offer explanations for why fundamental valuation metrics shouldn't matter. But valuations have proven time and time again to be one of the strongest predictors of future returns.  We've covered this topic several times in past and have highlighted relevant work from Hussman Funds, GMO. as well as Research Affiliates.   At a recent investment symposium in London hosted by Research Affiliates, Rob Arnott dissected for a large group of institutional investors what today's valuations are telling us about future expected returns. 

 

Rob is the founder and chairman of the board of Research Affiliates, a global asset manager dedicated to impacting the global investment community through its insights and products. Please click here to view Rob's full video presentation and/or a copy of his Powerpoint presentation where full size versions of the images contained in this article can be accessed.  The content is free but you will need to become a register user with Research Affiliates to gain access. 

 

The following summarizes the key takeaways from Rob's presentation in bullet point form with accompanying charts:  

 

• The link between starting valuations and subsequent returns is very powerful.  The chart below highlights the high correlation of starting cyclically adjusted price earnings and subsequent 10 year returns.  The cyclically adjusted price-to-earnings ratio, commonly known as CAPE, Shiller P/E, or P/E 10 ratio, is a valuation measure usually applied to the U.S. S&P 500 stock market index. It is defined as price divided by the average of ten years of earnings (moving average), adjusted for inflation. 

As the chart highlights, a high CAPE correlates with lower subsequent investment returns and a low CAPE is indicative of higher future returns.  The average correlation between CAPE and subsequent 10 year returns has been 75% across all of the major developed markets, including the U.S.  As John Hussman has repeatedly emphasized in his market commentary,  valuation levels are not useful, in and of themselves, for timing market tops and bottoms.  The experience of the later part of this bull market cycle is certainly a testament to this fact as valuations have continued to rise beyond previous valuation extremes despite obvious economic, trade, and political stresses.  Hussman believes that technical measures of investor risk appetite (market divergences, credit spreads, etc.) provide far better clues as to market timing than pure valuation metrics.  The current period of extraordinary Fed monetary policy of quantitative easing and test of the zero rate interest rate bound have fueled investor risk taking and severely limited market timing signals/warnings derived from traditional valuation limits.

 

 

 

• Even after making multiple adjustments, U.S. equities face potentially severe headwinds from valuation contraction as valuations revert to their historical mean over time.  As depicted in the chart below, late October's headline CAPE for the U.S. of 29.4 was in the 95th percentile historically.  This compares to 1) the simple historical average of CAPE of 18.2 from 1926 through June 30, 2019, 2) the simple historical average CAPE of 21.2 since 1977 (modern market era), 3) Research Affiliates' adjusted CAPE for changes in business cycle and macro volatility of 21.6, 4) CAPE of 27.7 excluding the highest and lowest earnings years in the past 10 years, and 5) 28.7 if the lowest earnings year of the past 10 years is excluded.
   
  Research Affiliates estimates the current CAPE valuation differential implies an approximate 3% annual degradation in nominal S&P 500 total returns over the next 10 years as the market CAPE ratio reverts to more normal levels.

• The largest stocks by capitalization are often the most expensive and have historically under-performed after reaching a top 10 valuation milestone.  Today's top ten are dominated by the six names of FANMAG (Facebook, Apple, Netflix, Microsoft, Amazon, & Google; third blue bar in chart below).  FANMAG's combined market valuation of $4.0 trillion is larger than only 2 out of 61 countries in the Morningstar Global Markets Index.  U.S. market excluding FANMAG valuation of $25.7 trillion and Japan $5.5 trillion are the two markets that exceed the valuation of FANMAG. Wow!

 

 

• This next chart below walks through the valuation components of Research Affiliates 10 year expected nominal return of 2.6% for the S&P 500 - dividend yield of 1.9%, plus 1.2% of real growth and 2.1% from inflation, less (2.7%) valuation change as prices revert to CAPE historical norms.
  

 

• Across asset classes higher return potential exists in international and diversifying markets based upon Research Affiliates CAPE based valuation analysis.  The chart below depicts Research Affiliates expected 10 year nominal returns of various international markets compared to its 2.6% annual return expectations for the S&P 500.   Research Affiliates expects international developed markets (EAFE) to return 7.4% annually and emerging markets as a whole to return 9.4% with certain high risk markets like Turkey and Russia in the upper right expected to return more than 12%.  Arnott highlighted the United Kingdom as particularly attractive developed International market from a valuation standpoint with a 10 year expected return of close to 9%.  The market implied Brexit risk premium has made the U.K. market much more attractive than domestic U.S. equities.  
   
  Arnott suggested that investors balance their return maximization goals with risk relative to more conventional benchmarks (U.S home bias) as they explore the attractive valuations of developed and emerging international markets as a whole and individual country market allocations as risk appetites allow.

 

• As the extreme FANMAG valuation might suggest, Arnott sees pockets of value within U.S. and global equities based on various value factors that they monitor (value factors flag cheap stocks rather than momentum factors that chase growth) that they believe offer higher potential forward looking returns.  Based upon Research Affiliates' analysis, the three factors highlighted in green below offer interesting relative value with expected annual returns in excess of 5% - value composite developed markets, value composite U.S., and value composite emerging markets. 
• Growth factor have been outperforming value factors for a number of consecutive years much like they did prior to the 2000 Dot-com bust and 2008/09 Global Financial Crisis.  Research Affiliates analysis of market history suggest that markets will revert to the mean with value factors outperforming growth factors at some point in the future.  True to form, the domestic markets witnessed a large rotation in fund flows to value factors in September with value outperforming growth by 3.5% for the month.  The market rotation into value and out of momentum and growth has continued through October and into November.