In last month’s article, we rang in the new year as we do every year, by updating our annual reminder of the importance of avoiding surprise in our investing efforts. This month, we unpack the fallacy of the markets being “due” to crash.
The Fallacy of Being “Due”
In games of chance or skill, it is natural for the human mind to believe that a certain outcome is “due” to occur. When a professional baseball batter goes into a slump and gets zero hits for 20 at-bats, we say he is due for a hit. When a coin toss comes up heads five times in a row, we say that the coin is due to come up tails on the next flip.
Despite our natural tendency to try to make predictions based on past experience, when it comes to probability, nothing is ever due. Every single time I flip a coin, it has a 50% chance of landing on tails, no matter how many times in a row it has come up heads. Past experience has no bearing on this probability.
Investors can easily fall prey to this same idea. For example, many investors today have convinced themselves that a bear market is bound to come soon, thanks to the stellar performance of U.S. equities over the last ten years. We all know that what goes up must come down. But, is that really true when it comes to the markets?
Reverting to the Mean
The idea of an unpredictable outcome being due comes from the natural human instinct to believe that all things must eventually “revert to the mean.” If a baseball hitter has been a .300 hitter for his entire career, and he has a streak where he bats .500 for a month straight, we expect that he will eventually go into a slump and revert back to being a .300 hitter. Likewise, because the stock market has achieved an annual rate of return of just about 10% per year for the last 100 years, but has returned more like 13.6% per year for the last ten years, we expect that eventually it must revert to the mean and experience some below-average years.
While “mean reversion” may be a helpful way to think about expected investment returns, it is not a magic formula for making predictions about the future. Investment trends can last much longer than we expect, and although returns generally do eventually revert to the mean, it can take a very long time for that to happen. While it may be accurate to expect that the stock market is due for a big upward or downward movement based on past results, it is very difficult to predict exactly when that movement can be expected.
For example, in the current popular thinking, the S&P is due for below-average returns over the next ten years because it did so well over the last ten years. The problem with this thinking is that the historical data seems to imply that the S&P 500 return over the prior ten-year period has very little value in predicting the next ten years. In fact, since 1936, there has been little-to-no relationship between prior ten-year returns and growth over the next ten years. The correlation between prior ten-year growth and future ten-year growth has been very low, coming in just slightly negative at -.19.
Although equity markets can remain random and unpredictable for much longer than we expect, and may take a long time to revert to the mean, it does appear that most 30-year periods in history have ended up looking fairly similar, and that S&P 500 returns do tend to revert to the mean over 30 years or more. History shows if the markets did badly for a period of 20 years, they generally did well for the next ten years, and vice versa. You can see this relationship clearly in this chart, where the future ten-year returns tend to cluster together for any given level of performance over the prior 20 years:
What is Really Due?
This data argues that the markets might be poised to perform the opposite of how common wisdom suggests. Because the S&P 500 has produced an average annual return of 13.6% per year over the last ten years, compared to a long-term average of 10% per year, some people may believe that we are due for a serious correction.
However, if we extend our time horizon just a little, we find that equity returns over the last 20 years were much worse than usual because the period between 2000-2010 was one of the worst on record. Despite the stellar performance of the last ten years, when combined with the prior ten years, the S&P 500 compounded at a rate of about 6.3% (with dividends reinvested) between January of 2000 and December of 2019, which is one of the worst 20-year periods on record for equity returns.
Historically, when equities have struggled this badly over a 20-year period, they have generally performed quite well over the next ten years, with relatively little variance. Historically, looking at rolling 20-year periods, markets compounding in the 6%-6.5% range over the prior 20 years went on to grow between four to five-and-a-half times over the next ten years, which is a fairly tight range of actual outcomes.
Of course, history doesn’t have to repeat itself in any meaningful way for 2020 and beyond, but if the S&P 500 “only” doubled over the next decade, here is where that point would lie on this plot:
Historically speaking, if the S&P 500 “just doubled” over the next ten years, the period from 2000-2030 would be a significant negative outlier. For those expecting anything less than a doubling of the S&P 500 by 2030, the red star on the chart above will be even lower than its current position, which would be unlike anything we have ever seen before in terms of 30-year equity returns. Alternatively, if history were to repeat itself in some meaningful way, the S&P 500 would be four times higher by 2030 than where it is today.
Of course, there is no law forcing U.S. markets to follow this trend indefinitely, and we are in no way predicting that the S&P 500 will quadruple in the next ten years. However, if you are forecasting an awful next decade for U.S. stocks, the evidence is heavily against you.
Back when I started my career as a wealth advisor in the early 1990s, the S&P 500 was in a similar position as it is today and appeared to be due for a major correction. After a long bear market in stocks in the 1970s, a new bull market had begun in the early 80s, and the S&P 500 had risen from a low of 290 in August of 1982, all the way up to 718 by May of 1990. Not including dividends, the market had risen by almost two-and-a-half times in a short eight years. Surely, stocks were “due” to correct this torrid pace.
To the great dismay of pessimists everywhere, equities did not crash in the 1990s, but continued climbing at an even faster pace, moving from 718 in May of 1990, all the way to 2,249 by March of 2000, more than tripling again in just a decade! (Again, excluding dividends.)
Then, thanks to the irrational exuberance of investors all over the world, investors finally got their “due”, and had to endure two consecutive horrible bear markets between 2000 and 2010. The S&P 500 stood at a paltry 1,273 by the time the decade was over, and the damage was done.
Finally, the last ten years have been very kind to investors, as the S&P 500 has recovered back to a level of 3,379 at the time of this writing.
All in, from the time equities seemed due for a correction back in 1990 until today, the S&P 500 has traveled from 718 in May of 1990, to 3,379 in early 2020, excluding dividends. Not bad for a 30-year period of time, during which markets experienced two of the worst bear markets in the history of our great nation.
So What About Those Dividends?
Above, we took care to repeatedly note that the numbers quoted excluded dividends and only reflected the price appreciation in the S&P 500. So, what about those dividends? How did they do over that time period?
Let’s forget about capital appreciation for a moment and just assume that your equity capital stayed exactly the same over the last 30 years. But, consider the impact of the growth of the cash dividends that an investor was paid over this time period. In 1990, the 500 companies in the S&P paid a cash dividend of $11 per share. Today, the cash dividend from the S&P 500 companies is just about $58. Over that same period of time, the Consumer Price Index rose from 126 to 257. While the cost of living just about doubled, the cash flow from equities quintupled.
Serious long-term investors, who are funding goals like a worry-free retirement, should be much more concerned about the income stream that their investments produce than they are about the fluctuating capital value of those investments. As author Nick Murray puts it, “You don’t take your brokerage statement to the grocery store with you.” Indeed, you don’t; you take the income that is produced by your investment portfolio with you when you go to the grocery store, your favorite restaurant, or for coffee at Starbucks.
Investors with a long-term time horizon, who wish to fund a lifetime of income needs (that will relentlessly rise every year due to inflation), should not fund that income plan with investments that will produce a static income stream (bonds). They must use investments, which have an income stream that rises over time, just like equities have done consistently for years.
An income that we can’t outlive is one that rises at a premium to the rate our cost of living rises. Thus, the key test of retirement income is not its level (its “yield” on a fixed sum of invested capital) but its trajectory. That being the case, we hold that all fixed-income investment strategies are bound to fail in the long run. Most often, the flat line of fixed income is surmounted, during the investor’s lifetime, by an upward sloping curve of living costs. When that happens, principal capital must be withdrawn to supplement the income shortfall, creating a form of a death spiral. The best the investor can hope for is to die before the capital is exhausted.
In our view, sitting around hoping to die before your money runs out is no way to live.