The U.S. stock market’s “Halloween Indicator” will be more trick than treat over the next six months.
Unlike many other alleged seasonal patterns, this one certainly appears to be the real deal: The average winter-versus-summer difference in the U.S. market does satisfy traditional standards of statistical significance. This is well illustrated by the chart above. Notice that in midterm election years there’s a huge difference between average winter and summer returns — 10.4% versus 0.8%. In non-midterm election years, the difference is 3.6% versus 2.6%, which is not significant at the 95% confidence level that statisticians often use when assessing whether a pattern is genuine.
The professors reached this conclusion after analyzing historical patterns back to 1900 in the Economic Policy Uncertainty index that was created by Scott Baker of Northwestern University, Nicholas Bloom of Stanford University, and Steven Davis of the University of Chicago. Chan and Marsh found a distinct pattern of the EPU reaching well-above-average levels before the midterms and well-below-average levels thereafter.