The Strange Truth About Selling in May
You’ve probably heard the cliché phrase: “Sell in May and go away.” But do we really know if it has any merit?
To find out, I devised two strategies and put this market adage to the test.
The first strategy invested in the S&P 500 from the beginning of November until the end of April and held cash for the rest of the year.
The second strategy bucked conventional wisdom by investing in the S&P 500 from the beginning of May until the end of October and holding cash otherwise.
If returns were evenly distributed throughout the year, then over the long term, the total returns accruing to each six-month holding period should be approximately equal.
Yet in actuality, the results were amazingly lopsided…
As you can see above, if you’d actually sold in May and went to cash every year only to return to the market after Halloween, you would’ve done very well over the past 20 years.
By being invested in the S&P 500 solely during the November-to-April period, you would’ve captured the majority of the gains produced with a traditional buy-and-hold strategy. What’s more, you would’ve experienced far smaller drawdowns (peak-to-trough declines).
Clearly, stock market efficiency is defied by this anomaly, which stands up to more comprehensive analysis.
In “The Halloween Indicator, ‘Sell in May and Go Away’: An Even Bigger Puzzle,” researchers analyzed data from 109 stock markets around the world, starting with the United Kingdom in 1693. They concluded that the stock market risk-return dynamic has been more favorable in the winter than the summer by a statistically significant degree in the vast majority of countries.
Could it really be that we can systematically invest based on the Earth’s position in the solar system and earn superior risk-adjusted returns? It seems so.
An Easy Hedge Against Seasonality
The cause of stock market seasonality may be lower trading volumes and liquidity during the summer vacation months in the northern hemisphere, although, no one is sure.
Regardless of the reason, seasonality is actually present across asset classes, as well.
Wesley Gray and his team at Alpha Architect have shown that the risk-adjusted returns for bonds (as measured by the Sharpe ratio) are actually the most favorable from May to October.
This makes sense, as Treasuries typically benefit from a “flight to safety” bid when the equity markets are weak.
So, should investors be fully invested in stocks from November through April, and then switch into Treasuries from May to October?
Although this would likely be an extremely sound strategy over the long term, it’s not advisable.
That’s because no strategy works all the time, and lengthy periods of underperformance can frustrate even the most disciplined investors.
Imagine sitting in bonds during a great summer for stocks and then switching into stocks that subsequently perform poorly in the winter. You’re going to be mighty tempted to abandon the strategy (probably at the worst possible time).
Long-term timing strategies with binary equity exposures are virtually impossible to stick with because of a high element of regret.
That’s why we want to maintain a globally diversified portfolio of stock and bonds at all times.
We shouldn’t formulate a strategy strictly on the Gregorian calendar. Instead, we can adjust our exposures gradually based on market conditions.
Bottom line: The “Sell in May” effect is a bona fide market anomaly, and diversification is your hedge.
Safe (and high-yield) investing,
Alan Gula, CFA