There are many different ways to go about making projections for prospective returns of the stock market. On one side of the spectrum, at the beginning of each year Wall Street analysts give their projection for the price of the S&P 500 at the end of the upcoming year. Their sophisticated methodology usually consists of applying to the current price a mark-up that conveniently ranges between 8% and 12%. On the other side of the spectrum, financial planners build clients’ portfolios by projecting returns of different asset classes over time frames as long as their clients’ life expectancies. In this case, the very long-term average historical return of each asset class is usually the way to go. Somewhere in between these two extremes lies a methodology grounded in economic theory and valuation and which is used by some prominent portfolio managers (John Hussman and Jeremy Grantham, among others) to project stock market returns over periods of 5-10 years.
For the geeks, here’s a brief description of the methodology…
By definition, the total return of the S&P 500 over any time period can be split into two components, price appreciation and dividend income. The first component, price appreciation, can be split further into earnings growth and multiple (e.g. price-to-earnings ratio) expansion/compression. If no multiple expansion/compression is expected to happen during the selected time frame, then price appreciation will simply reflect earnings growth. However, if the current environment features unusually elevated/depressed valuations, then these should be expected to converge towards more normalized levels in the future. Multiple expansion will have a positive impact on price appreciation and total return, while multiple compression will have a negative impact. Following a similar rationale, the earnings growth component of price appreciation should be adjusted for expected changes in profit margins. While earnings tend to be higher when profit margins are higher and vice-versa, profit margins are known to be a mean-reverting series. Therefore, expectations for earnings growth following an environment of unusually elevated (compressed) profit margins should be adjusted downwards (upwards), reflecting the assumption that profit margins will eventually converge towards normalized levels.
The last adjustment we need to make involves the second component of total return, dividend income. Dividend yield is a function of the dividend payout ratio (what portion of the earnings is paid out as dividends) and the earnings yield (the inverse of the price-to-earnings ratio). If we assume the payout ratio to be constant, then the dividend yield becomes a pure function of, once again, multiple expansion/compression. However, in this case, multiple expansion will decrease the dividend yield and therefore reduce dividend income, while multiple compression will increase it. The equation below summarizes the adjustments we just described.
Total Return = [Price Appreciation] + [Dividend Income]
= [Adjusted Earnings Growth + Multiple Expansion/Compression] + [Dividend Yield +
= [(Unadjusted Earnings Growth + Profit Margin Expansion/Compression) + Multiple
Expansion/Compression] + [Dividend Yield + Multiple Compression/Expansion]
Using post-war data, we applied this methodology to project the ten-year annualized return of the S&P 500 following each month-end since January 1948. The chart below compares our projections with the actual S&P 500 ten-year annualized returns that followed (which for obvious reasons are only available up until April 2002, ten years ago). The resemblance of the two lines is impressive, as these are more than 80% correlated.
The methodology is currently projecting a ten-year annualized return of merely 3.17%, which is the result of the currently extended ten-year normalized price-to-earnings ratio (22.22 versus the post-war median of 17.74) and profit margins (12.97% versus the post-war median of 9.40%). The convergence of these two key valuation metrics towards more historically normalized levels should be expected to create significant headwinds for the stock market over the next ten years. Given that ten-year treasury bonds currently yield just below 2%, should you just buy stocks and declare victory? It depends on whether you think the additional 1.17% expected return is enough to compensate for the significantly higher volatility. In addition, while experiencing interim volatility, investing in ten-year treasuries today would allow you to lock in a virtually certain (unless the U.S. government defaults) 2% return over ten years; with stocks, there is no such thing as locking in a return: not only will you experience interim volatility, but the end result is also highly uncertain.
The good news for tactical investors is that while current market valuations suggest market returns will be subpar over the very long-term, anything can happen in the interim. For instance, in December 1999, with the S&P 500 approaching the heights of the tech bubble, this methodology projected a ten-year annualized return of -0.33%. Ten years later, the S&P 500 would have offered a ten-year annualized return of -0.71%. In the meantime, however, investors experience two bear markets and two bull markets of remarkable magnitudes (-42.44%, +108.27%, -50.92%, +103.06%). While we would certainly welcome more attractive valuations, as these would provide us with a higher margin of safety, we must also recognize that even from the currently extended levels the market can offer significant opportunities to make (and lose) money. Luckily, while valuation tends to work in the very long term, tactical investors have in their arsenal other tools that are geared towards reaping opportunities over short-to-intermediate time frames.
If however you are a buy-and-hold investor… well, good luck.
Copyright: blueximages / 123RF Stock Photo