You Can't Predict Future Markets After Flat Years
U.S. markets essentially made no gains in 2015. The major indexes are more or less flat for the year to date. The Nasdaq-100 Index has done best, gaining about 8 percent, driven by a handful of big winners such as Amazon, which is up 116 percent for the year. Yet the flat returns belie a big increase in volatility, including a selloff of almost 20 percent at the end of the summer that investors had made up by late fall.
There have been two ways of predicting what flat years mean for future returns. Some bearish traders assert that the flattening of indexes is a sign that the bull market, which began almost seven years ago, in March 2009, has grown old and tired. In this view, flattening indicates that markets are setting up for a major correction or worse.
The opposing argument has been that markets have had a great run, up over 200 percent, after the 57 percent fall during the financial crisis. These big gains need to be digested, and a sideways year is a "pause that refreshes." Allowing earnings to catch up with prices also allows markets to become more attractively priced.
Both of these narratives can be compelling, depending on the subconscious bias you may be seeking to confirm. Rather than be taken in by a good story, or accepting the one that agrees with your views, look at the market data. You might be surprised.
Consider the rolling 12 months of Standard & Poor's 500 Index returns whose changes in price only were within a plus or minus 3 percent range (using calendar years only generated a sample set of four, hence the rolling 12 months).
The results show that both the pro and con arguments are pretty much dead wrong. There is no very strong conclusion to be drawn from the underlying data.
After a flat 12-month period -- defined as one in which price changes in the S&P 500 were between plus 3 percent and minus 3 percent -- the average return 12 months later was 7.39 percent. Now compare this to the universe of all returns (1928 to today; all data via Bloomberg): Twelve months later, the average return was 7.60 percent. In other words, there was little or no discernible difference in the following year's returns.
There was one notable difference in the periods following flat years: They tended to be less extreme, with smaller swings in either direction. After a flat 12-month period, the worst returns are a loss of 47 percent and the maximum gain is plus 50 percent; the universe of all years is much wider -- a 70.13 percent possible downside and a maximum gain of 146.28 percent. You could conclude that the most extreme years have actually followed years that were not flat.
One note: I have warned about trying to draw big conclusions from reviewing a single variable (see When Correlations Lie and Single vs. Multiple Variable Analysis in Market Forecasts). Ideally, we could run a flat year against a variety of other factors -- inflation, P/E ratio, recent returns, for example -- and any number of permutations among those variables.
The key takeaway is that by itself, a flat market does not tell us very much of anything about the following years' subsequent returns.
This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.
To contact the author of this story:
Barry L Ritholtz at firstname.lastname@example.org
To contact the editor responsible for this story:
Max Berley at email@example.com