The September Effect
There appears to be no convincing evidence why stocks have traditionally performed poorly during September (-1.09% on average vs. 0.91% for November, 1.4% for December, and 0.75% for all months except September). Despite this historical information compiled from 1896 through 2012 (source: Hulbert Financial Digest), investors and advisors alike should avoid any “system” or market timing program. The reality is causation is not correlation.
For example, United Nation data collected by David Leinweber from UC Berkeley found the indicator most highly correlated to the S&P 500 was butter production in Bangladesh. It is true that U.S. stocks tend to perform well in late December, early January, and in the winter months between Halloween and May first. Most of these theories are not backed by sound theoretical explanations. Unless the investor is able to discover something nobody else knows about, by the time we know of any kind of pattern, it is too late to profit from it.