A Different Look at Seasonality
By Adam Thurgood on July 13, 2017
Yesterday brought yet another good showing for stocks. The major market averages continue to defy Washington drama and surprise to the upside, marking new all-time highs. With seasonal weakness just a few weeks away, one has to wonder if this year’s late summer period will be a nonevent or one filled with commotion.
The concept of seasonal weakness is nothing new. We’ve probably all heard the “sell in May and go away” adage. Perhaps what is fresh perspective is the hit rate of positive returns during the late summer months. While August and September both have negative average returns over the past 20 years, the market has gone up 55% of the time in August and 60% of the time in September. That tells us that when the market experiences a negative outcome during this period, it’s usually far from welcome. In fact, when the market is down in August and September, the average decline is -5.2% and -5.7%, respectively. This compares to an average decline of 3.2% for all other months.
With positive outcomes occurring more than 50% of the time in August and September, one might draw the conclusion that seasonal weakness isn’t the norm. Well, that would be true if the two months’ returns were highly correlated, but they’re not. The correlation of returns between the two months is negative (-0.19), which means some sort of pullback occurs a lot more frequently than initially thought. Over the past 20 years, the market has experienced a negative August or September 70% of the time, which means seasonal weakness is the norm and a summer drawdown shouldn’t come as a surprise.
Summer drawdowns (meaning the peak to trough performance from the end of May to the end of September) have ranged from marginal to gut-wrenching. The median summer drawdown is just over 7%. While early, this summer we are tracking to have the least significant drawdown of the past 20 years. Justified? Perhaps, but I believe the odds favor some commotion in the weeks to come.
