Skip to main content

If this has been a disappointing summer for your portfolio, you're not alone. Market returns tend to follow seasonal patterns and summer is not usually the best time for strong gains.

In fact, a group of University of Miami professors recently found evidence that the old advice to "sell in May and go away" works in markets around the world.

My research has shown the same thing. A few years ago, I wrote several articles demonstrating that investors can prosper by selling in the spring and staying out of the market until October.

Story continues below advertisement

On two points, though, I disagree with the Miami researchers. They conclude that "the source of the sell in May effect remains a puzzle" and find that no similar effect holds true in bonds.

I beg to differ. First, though, some background.

The "sell in May and go away" rule refers to the historical tendency for stocks to do relatively poorly from May to October in relation to November to April. This is true on average, not every year. From 1981 to 2009, an equally weighted portfolio of Canadian stocks had an average six-month return of 15 per cent between November and April, whereas the average six-month return for May to October was minus 0.15 per cent.

Over the same period, a portfolio consisting of Government of Canada bonds had an average return of 4.6 per cent for November to April and 6.7 per cent for May to October. Had an investor put all her money in the stock portfolio from November to April, but gotten out of stocks from May to October and instead invested exclusively in Government of Canada bonds, she would have realized an average annual return of 22 per cent over the 1981 to 2009 period. Comparable results are obtained for different periods.

It seems that stocks and government bonds do move in the opposite direction. But why? I believe the sell in May pattern rests on human psychology and the conflicts of interest that affect professional portfolio managers.

These professionals have their own agendas and their efforts to maximize their own benefits lead them to rebalance their portfolios in a predictable way throughout the year.

My research shows that the high returns on risky securities around the turn of the year are caused by systematic shifts in the holdings of these managers, who rebalance their portfolios to boost their performance-based remuneration.

Story continues below advertisement

At the start of the year, institutional investors are net buyers of risky securities because they are trying to beat their benchmarks and are motivated to include less-known, high-risk securities in their portfolios.

Later on in the year, portfolio managers lock in returns by divesting lesser-known, risky stocks and replacing them with well-known and less risky stocks.

Such behaviour affects prices and security returns in a predictable way. Risky stocks and high-risk bonds are bid up early on in the year and down later on in the year, whereas low-risk stocks and risk-free bonds exhibit the opposite behaviour.

Meanwhile, other investors try to take advantage of this regular pattern in portfolio managers' behaviour by anticipating their buys and sells. Thus, the pressure on stock and government bond prices is spread over a few months, giving rise to relative strength in the stock market from November to April and relative weakness in May to October.

The seasonal pattern in security returns is difficult for the markets to eliminate for two reasons.

First, it is tied to the behaviour of professional portfolio managers who are attempting to maximize their own remuneration year in, year out. Second, the seasonality is not consistently observed every year. Unless we can anticipate seasonal behaviour on a consistent basis, market participants cannot fully arbitrage the seasonal behaviour of financial securities.

Story continues below advertisement

George Athanassakos is a professor of finance and Ben Graham Chair in Value Investing at the Richard Ivey School of Business of the University of Western Ontario.

Report an error Editorial code of conduct
As of December 20, 2017, we have temporarily removed commenting from our articles. We hope to have this resolved by the end of January 2018. Thank you for your patience. If you are looking to give feedback on our new site, please send it along to If you want to write a letter to the editor, please forward to