"You know the saying, 'Red sky at night means fair weather tomorrow; red sky in the morning means foul weather all day.' You know how to interpret the weather signs in the sky, but you don't know how to interpret the signs of the times!" - Jesus Christ, quoted in Matthew 16:2-3 (NLT)
"Come writers and critics who prophesize with your pen. And keep your eyes wide, the chance won't come again. And don't speak too soon for the wheel's still in spin. And there's no tellin' who that it's namin'. For the loser now will be later to win; for the times they are a-changin.'" - Bob Dylan, The Times They Are A-Changin'
You may be wondering what Jesus and Bob Dylan have to do with investing. The answer lies in where we, as investors, focus our attention.
The last four years -- with their cliff-hanging global crises, last-second government rescues, renegade flash-crashing robots and villainous "vampire squids" -- have been enough to turn investors into either frazzled wrecks or the adrenaline-charged investing equivalents of the guy who jumps off mountains in a flying squirrel suit. The past month has been no exception. Global equity markets were first buffeted by slowing in major economies and the reemergence of the eurozone debt crisis. Now, the initial period of correction has given way to news-driven intra-day volatility, with the S&P 500 experiencing swings of more than 1% in five out of the last eight days, though searching for overall direction in front of the Greek election.
Both of the quotes referenced above encouraged people -- 2000 years apart -- to lift their gaze from the turbulence in front of them and try to assess the underlying, long-term reality of the situation. This article attempts to help that process for investors by walking quickly through eight indicators of expected equity returns in the areas of valuation, interest rates and economic activity and analyzing through basic regression analysis what those indicators are saying about annual inflation-adjusted (real) equity returns over the next ten years.
The idea behind this exercise is that returns on equities -- and all risk assets and strategies -- are driven by risk premiums which provide compensation for underlying risk factors such as liquidity and tail risk (negative performance in "bad times"). Also, risk premiums are not static, but instead vary over time, with higher risk premiums related to periods of economic stress, low current returns and higher future expected returns. (For an excellent, detailed survey of these concepts from a practitioner's perspective, see Antti Ilmanen, "Expected Returns.")
Indicators such as those used in this brief study are thermometers that help to gauge where we are in the risk premium cycle. They produce a range of estimates and none have correlations greater than 0.42 (or less than -0.42) with future expected returns, illustrating the difficulty in trying to predict future returns. However, by looking at a range of different indicators, we can develop a picture of where we are in the cycle and whether we have a margin of safety for making long-term equity investments.
(Note: I have done my best to do this analysis in a way that anyone could replicate, using publicly available data and conducting the analysis in MS Excel. While I have not had time in this article to give an extremely detailed discussion of each indicator or the data sources used, I would be happy to go into more detail in response to comments or messages.)
10-Year Real Returns - A Historical Perspective: Before examining how various indicators relate to future, 10-year real returns on the S&P 500, let's look at how 10-year real returns have fluctuated through several cycles.
From January 1871 to January 2012, compound, inflation-adjusted annual returns on the S&P 500 averaged 1.99% per year. The following chart shows real, annual compounded returns over rolling 10-year periods starting in 1890 (chart starts in 1900).
So approximately where are we in the cycle? Over the past 112 years, historical 10-year returns have moved in four major cycles, peaking in 1906 (6.6% per year), 1930 (13.7% per year), 1959 (13.0% per year) and 2000 (13.5% per year). Corresponding major bottoms were in 1921 (-10.3% per year), 1939 (-7.1% per year), 1982 (-8.2% per year) and - hopefully - 2009 (-7.5% per year). The peak-to-trough period of the three previous cycles ranged from 9.3 years (1930-39) to 23.2 years (1959-82) and the distance between peak and trough annualized real returns ranged from 17% (1906-21) to 21% (1959-82).
In the current cycle, roughly 8.5 years elapsed between the top in September 2000 and what may have been the bottom in March 2009 and 10-year average annual returns fell by 21%, both broadly consistent with previous cyclical lows. However, during past cycles, 10-year annualized returns tended to dip below -5% two or three times as part of a bottoming process, suggesting that at least one major leg down may be in the cards before the end of the current secular bear market.
Five year real return cycles are much more volatile than the 10-year return cycles, though paint a similar picture (see chart below).
Now that we have looked at the progress of long-run returns over time, we can start to evaluate future 10-year real returns using various indicators to see whether we can shed some light on whether future returns will be relatively high or low. We will start with valuation and then work our way through interest rates and indicators or economic activity.
Valuation - P/E Ratios and Dividend Yield
Two traditional measures of expected equity returns are the price-to-earnings ratio (P/E) and the dividend yield. Unlike interest rate or economic activity variables, which correlate to future equity returns by reflecting elements of risk factors that affect equity prices, both P/E ratios and dividend yields relate equity prices directly to earnings or dividends accruing to investors, providing a more direct indicator of valuation.
The price-to-earnings ratio can be calculated using net income or operating income, which is less volatile than net income, and also by using either trailing-twelve-month (TTM) earnings, projected earnings for the next year or a longer-term average of earnings. For assessing long-term expected returns, I prefer to look at Dr. Robert Schiller's Cyclically Adjusted P/E (CAPE) ratio which looks at a trailing 10-year sum of net income to try to remove business-cycle variation.I like the CAPE over the TTM P/E ratio because I find that a P/E ratio using the trailing-twelve-months of reported net income can give strange signals during times of stress. For instance, because of the collapse of earnings during the financial crisis, the TTM P/E reached over 110 in March 2009, suggesting substantially negative long-term expected returns at a time when other measures of expected returns were suggesting higher expected returns.
The dividend yield can be calculated either using the trailing twelve months of dividends or projecting the most recent dividend out over the next twelve months. For this exercise, I use the trailing twelve month dividend yield.
So, what do these indicators suggest about long-term expected returns? The current Cyclically Adjusted P/E ratio is sitting very close to its long-run trend level and - at 21.3 - is still high relative to history, implying 10-year returns of only about 0.03% per year before inflation (see chart).
Dividend yields also suggest lackluster returns over the next 10 years. Despite a rising trend since 1999 and a spike during the acute phase of the financial crisis, the current TTM dividend yield of about 2.1% is still are slightly below the long-term rolling trend and low relative to history, implying -0.28% annual compound real returns over the next ten years (see chart).
On the whole, regression analysis on the valuation indicators suggests future 10-year real returns of about -0.06% per year, plus or minus 8.6% (see table below).
Some may argue, with respect to dividend yield, that the comparison to past periods is unfair because share buy-backs have grown considerably in importance since the late 1970s. However, even if we add an estimated net-buyback yield (share buybacks minus share issuance) of roughly 1% -- bringing the dividend + buyback yield to 3.07% -- it would still only imply ten-year compound real returns of only 0.63% per year.
Interest Rates: Term Premium and Fed Model
The yields on benchmark Treasury maturities can provide information on expected equity returns when compared to each other (eg. shape of the yield curve) or to other investment yields. For instance, the 10-year Treasury note yield can be compared to the earnings yield on the S&P 500 (Earnings / Price, the inverse of the P/E ratio), a comparison known as the "Fed Model", or dividend yields on the S&P 500 or sector indexes such as those for REITs or Utilities. I personally compare the 10-year benchmark Treasury yield to the 1-year Treasury yield - an indicator known as the "Term Premium" - and to the cyclically adjusted earnings yield on the S&P 500 (eg. inverse of the CAPE).
The Term Premium is a measure of expected future interest rates and a measure of the risk premium that investors demand to hold interest rate risk. The relationship between the term premium and equities is mainly based on the future interest rate expectations component, as markets expect that short-rates will fall during periods of economic stress and rise during periods of strength. Indeed, negative term premiums (also known as "inverted yield curves," because the 1-year rate is higher than the 10-year rate), have been good predictors of future recessions.
The Term Premium has come down quite a bit since early 2011 as 10-year yields have fallen a good deal faster than already low 1-year yields (see chart below). Historically, elevated Term Premium levels have pointed to relatively strong future equity returns; the current average level of about 163 basis points in May implies future 10-year returns of about 3.5% per year.
The comparison between the S&P 500 earnings yield and the 10-year Treasury yield is widely referred to as the "Fed Model" because a chart of the S&P 500 earnings yield versus the 10-year Treasury yield is first thought to have appeared in the July 1997 Humphrey Hawkins report to Congress. Proponents of the model argue that earnings yield should theoretically be positively correlated to Treasury yields because they both reflect future expectations for inflation, with P/E ratios suffering during periods of high inflation. Critics of the model argue that it is inappropriate to compare a "real" concept like earnings yield with a nominal measure like the 10-year Treasury and that "long-term expected real stock returns are low regardless of the starting point of interest rates," making the Fed Model a poor forecasting tool. I tend to side with the critics, as P/E ratios have historically been low during both periods of very high and very low inflation, though include the CAPE Fed Model indicators in this analysis for sake of reference.
The CAPE earnings yield is currently high relative to today's extremely low 10-year Treasury rates (see chart below), implying future 10-year real expected returns of 2.11%.
Overall, interest rate indicators suggest long-term expected returns of 2.86%, plus or minus 9.2% (see table).
This expected return estimate is higher than the one produced by the valuation indicators, though the average correlation between these indicators and expected returns are lower than for the valuation indicators. Also, Barry Knapp of Barclays Capital argued in his 2012 U.S. Equity Market Outlook that "valuations relative to fixed income … might be flawed due to Fed policy distorting the interest rate curve." Knapp writes: "In other words, equities may not be cheap relative to Treasuries; the case for Treasuries being rich may be stronger than the case for equities being cheap." He cites previous instances of financial repression - or the excessive downward management of interest rates - in the U.S., when increasing inflation volatility led to persistently lower P/E ratios / higher earnings yields (see page 11 of the linked document).
Economic Activity: Unemployment, ISM, Consumption / Wealth, Corporate Profits / GDP
Finally, economic activity indicators offer clues as to whether the economy is currently experiencing "good times," when future expected returns should on-balance be lower, or "bad times," when expected returns should be higher. I look at indicators related to manufacturing activity (ISM Manufacturing Purchasing Manager's index), corporate profits (ratio of corporate profits to GDP), the labor market (unemployment rate) and the consumer (ratio of household consumption to household wealth).
Based on history, the continuing high unemployment rate - well above its long-term trend - seems to be the most positive for stocks, as future expected returns have been high when unemployment has been this high in the past. However, the ISM Manufacturing PMI and Consumption-to-Wealth ratio are both close to their long-term trend or average levels. And, corporate profits are actually at multi-decade highs relative to GDP. The following charts give a view of how these economic indicators have been related to expected returns in the past.
Overall, the four indicators of economic activity that I analyzed produce a weighted average prediction of 10-year real expected equity returns of 3.61% per year, plus or minus 8.94%, with individual estimates ranging from 1.04% per year to 6.04% per year (see table below).
Putting It All Together - Neutral Expected Returns and Downside Risks
10-year Real Expected Returns of 2.1% Per Year: The eight indicators studied had absolute value correlations with future expected returns between 0.18 and 0.42 and produced real return predictions of between -0.28% per year and 6.04% per year for the next ten years. By taking a weighted average of these estimates, with weights based on the explanatory power of the regressions between the indicators and future 10-year real returns, I arrive at a 10-year real return prediction of 2.1% per year (see table below).
A 2.1% real equity return over ten years, or a 5.1% per year nominal return if we add 3.0% annual CPI inflation, is nothing to sneeze at and suggests that investors with long (more than 10 year) horizons will likely not experience significant valuation headwinds over their holding period. Indeed, 2.1% real returns compare very favorably to the 10-year Treasury Inflation Protected Securities (TIPS) - one measure of real fixed income returns - which have been selling for an average yield of -0.44% over the past month!
Medium-Term Downside Risks: However, looking at the expected return calculation over time, it becomes apparent that current expected returns are rather middle of the road and that more attractive valuation levels may be forthcoming (see chart).
Indeed, Eric Parnell wrote on Friday, May 30th about the case building in real S&P 500 chart patterns for one more major correction in the current secular bear market, potentially occurring within the next 1-3 years. Patterns in the cyclically adjusted P/E support this scenario as well.
During the two secular bear markets in the 20th century, the CAPE declined to the 5x-10x range before launching into a new secular bull market (see CAPE chart above). If we de-trend the CAPE using its long-term rolling OLS trend -- to mitigate issues related to permanent increases in valuations -- we see that secular bull markets were launched in the 20th century after two conditions were achieved: 1. Three major declines in the CAPE; 2. The de-trended CAPE declining to around 0.5 or below one or more times (see chart).
In the current secular bear market, we have experienced two major decreases in valuation with the most recent reaching a de-trended CAPE of 0.62, suggesting that we could have one more large correction before launching another secular bull market.
How To Position? We recommend that longer-term investors, as appropriate for their objectives, position for neutral expected returns and potential down-side risks by maintaining equity exposure through higher-dividend yielding shares and more defensive sectors. Funds we like for this environment include the SPDR S&P Dividend ETF (SDY) and the Vanguard Dividend Growth Fund (VDIGX). SDY is overweight consumer defensive, utilities and industrials sectors and has a 3.23% trailing-twelve-month dividend yield (3.77% on a forward looking basis). VDIGX is overweight Consumer Defensive, Healthcare and Industrials sectors and has yielded 1.99% over the last 12 months (2.54% on a forward looking basis). Both are less volatile than the S&P 500 more broadly, with SDY achieving a standard deviation of 13.7% and VDIGX a standard deviation of 12.8% over the past three years, versus 16.0% for the S&P 500 as a whole.
More aggressive investors might consider tactical allocation between stocks and bonds using a moving average overlay (see Faber & Richardson, The Ivy Portfolio for one such overlay strategy) or a sector momentum strategy. Sectors with increasing relative strength versus the S&P 500 over the past 200 trading days include Utilities (XLU), Healthcare (XLV) and Consumer Staples (XLP).
The first sentence in the last paragraph of the sub-section titled "10-year Real Expected Returns of 2.1% Per Year" should read: "A 2.1% real equity return over ten years, or a 5.1% per year nominal return if we add 3.0% annual CPI inflation...". I had originally put 2.5%, which is roughly the rate implied by the 10-year TIPS, though I decided before publication to bump that up a bit to 3.0%.
Also, the first sentence in the last paragraph of the sub-section titled "Medium-Term Downside Risks" should read: "In the current secular bear market, we have experienced two major decreases in valuation with the most recent reaching a de-trended CAPE of 0.62...".
Disclaimer: Bard Luippold is Corporate Finance Manager for Meracord LLC ("Meracord"). This article is prepared by Mr. Luippold as an outside business activity. As such, Meracord does not review or approve materials presented herein. The opinions and any recommendations expressed in this article are those of the author and do not reflect the opinions or recommendations of Meracord.
None of the information or opinions expressed in this blog constitutes a solicitation for the purchase or sale of any security or other instrument. Nothing in this article constitutes investment advice and any recommendations that may be contained herein have not been based upon a consideration of the investment objectives, financial situation or particular needs of any specific recipient. Any purchase or sale activity in any securities or other instrument should be based upon your own analysis and conclusions. Past performance is not indicative of future results.