With the stock market having produced a total cumulative return of -13% over the last ten years, many investors are understandably discouraged. A common reaction is to fear that the market is likely to produce little or no return for the next 5 to 10 years. While we’re not willing to forecast market returns, and so won’t rule out any outcome, the history of the stock market suggests that paltry returns over the next 5 to 10 years are in fact an unlikely outcome.
Over the last 83 years, the 10 calendar years just ended represent the worst such period ever, including the Great Depression (-13.0% versus -7.6% in 1929-1938)! And that is one of the most compelling arguments for far better returns in the future. The reason is that free markets have a strong tendency toward mean-reversion, i.e. returning to their long term average performance. So after long stretches of unusually strong performance, subsequent returns tend to be weaker, and after unusually weak periods, subsequent returns tend to bounce back and be stronger. Until the 10 year period that just ended, the previous worst 10 calendar year stretch of returns ended in 1938, and in the 10 years that followed, the market returned 102% (7.3% per year). Furthermore, that bounce back in returns was the smallest rebound of any of the eight different 10 year periods of low or negative returns (less than 3% per year). If one looks at what has happened after all of these decade long return droughts, on average the market has bounced back to return 225% (12.5% per year) over the following 10 years.
However, it’s not enough just to demonstrate that 10 year back to back stretches of poor returns haven’t occurred in stock market history. To have confidence that this bounce back isn’t a quirk of history, but will persist in the future, one needs to understand why this reversion to the mean occurs. The three most important reasons for this bounce back phenomenon are: the long term nature of stock value drivers; the resilience of corporate earnings; and the importance of valuation as a driver of future returns. Stock ownership represents a share of all of a company’s future cash flow (earnings, dividends, etc.), not just this year’s or next year’s. Therefore, when the economy enters a recession and corporate cash flows fall sharply for a year or two, stock prices should also fall, but the stock price fall should be much smaller than the cash flow fall because recession impacts are transitory, lasting at most a few years. The graph below shows the long term path of earnings for the S&P 500 over the last 58 years.
It’s worth noting that the long term trend is remarkably steady at a 6% growth rate even though short term earnings fluctuate a great deal. Also note that, while periodic sharp drops occur in recessions, earnings typically reach new highs in just a few years. Despite this long term consistency of earnings growth and resiliency from setbacks, stock prices tend to overreact to short term earnings changes. These price changes create dramatic swings in market valuations and lay the groundwork for the reversion to the mean. The extremely poor returns of the last 10 years had a starting valuation in 1999 of over 25 times earnings with a dividend yield of 1.3%! That was a record high valuation and it led to record low returns. It’s no coincidence that the previous record for low returns over 10 years had a starting point at the record high valuations of 1929. By comparison, the market now sells for 12 times earnings with a dividend yield of 3.2%. The valuation is now lower than average, which portends higher than average returns going forward rather than a repeat of the last 10 years.
So for those who are thinking we’re about to see a repeat of the stock market from the Great Depression, our response is that we’ve already seen it! Ironically, although we usually caution against trying to predict the future based on the recent past, in this case the last 10 years can be a useful guide to the future as long as one recognizes that the appropriate conclusion is not a repeat of those years, but a rebound to long term averages.
