The September Effect
The market is full of quotes, effects and algorithms that some people live by and others roll their eyes at.
For example, you’ve heard about the old adage to ‘sell in May and go away’. The full version reads ‘sell in May and go away, don’t come back until St Leger’s Day’.
The principle behind this effect is that from St Leger’s Day (in the middle/end of Sept) through to April, the stock market is supposed to offer the best period for returns.
Yet in reality, this supposed effect isn’t really one to live by. Below shows the years when such a strategy would have worked:
What about the December effect, otherwise known as the Santa Rally?
This theory supports buying stocks during the last week of December and the first two trading days of January, due to the potential for a strong holiday season rally.
According to the Stock Trader’s Almanac, the Santa Rally does occur more often than not. Below shows the years from 1950 when we did get a rally (shown in the SC Rally).
In the past 60 years, it has occurred more than two thirds of the time.
So this effect is more valid to be taken seriously, although the fundamental drivers are still somewhat murky. Institutional buyers reopening their books at the start of the year makes sense to support buying, and positive retail flow to end the year due to the holiday season can also be pointed to. Yet neither are concrete enough to build a trading strategy around.
You can become a premium subscriber with us for just $20 a month. This provides you full access to our weekly articles without a paywall, including our flagship Monday trade ideas and full access to what we are buying and selling in our Global Asset Portfolio.
This brings us to the September effect, which is one of the lesser spoken about theories that some look to. Yet given that we’ve just started the month, and the fact that it promises to be an important one, we thought we’d take a closer look.
What is the September effect?
Simply put, the September effect refers to the historical tendency for stock markets to underperform during the month of September. This phenomenon has been observed in various global stock indices over many decades.
So let’s start with looking at whether the observations are valid enough from the data we have:
This is quite a telling chart. The negative reading for September is not only an outlier (February is the only other negative month but this is marginal) but also a significant one. At negative 51bps on average, it’s significantly below any other calendar month.
However, there’s a word of caution when using averages, in that the data can be skewed by some exceptionally poor outliers. According to CME Group, the S&P 500 has lost ground in 55% of Septembers over the last century. So this tells us that September is really more of a 50:50 coin toss as to whether or not it’ll be positive or not. Yet the skew tells us that if it is a negative month, the magnitude of the loss will be greater than the gain from a positive month.
It’s true that September has seen some of the most significant market declines, such as the 29.6% drop in 1931 during the Great Depression. The Global Financial Crisis from 2008/09 also saw a circa 10% drawdown during September. Let’s also not forget the sad events of 9/11, which triggered a sharp fall in the stock market too.
Outside of the large declines, there are some other factors that have historically added to providing a weak September.