"Sell in May and Go Away": Definition, Statistics, and Analysis

"Sell in May and go away" is a well-known saying in finance based on stocks' supposedly underperforming during the six months from May 1 to October 31. The Stock Trader's Almanac popularized the idea of the historical pattern, which found that investing in stocks as represented by the Dow Jones Industrial Average from November to April (we'll discuss this as the "winter" period) and switching to fixed-income investments the other six months (the "summer") would have "produced reliable returns with reduced risk since 1950."

What it didn't note is that if one used the S&P 500 index, which dates to 1927, you would have found the opposite: the summers almost always outperformed the winters. Our analysis shows returns were 11.23% and 4.51% higher in the summer months than the returns for the winter months for the S&P 500 index in the 1930s and 1940s, respectively. For this and other reasons, the adage has since been widely disputed.

Nevertheless, on average and over the long term, stocks have done worse over the warmer months in the decades since. For example, the S&P 500 has gained an average of about 3% from May to October from 1990 to 2023, compared with an average of about 6.3% from November to April from 1990 to 2024. These numbers change to about 3% and 4.5% for the summer and winter months, respectively, if we go back to 1930.

Why should there be such a difference? In addition, are the reduced summer returns reason enough to switch to bonds during that time? Even if there are differences between the summer and winter performance of equities, the summer returns for the S&P 500, once annualized, are still quite good—and better than other securities. Thus, as we'll see, it's a good idea for most investors to stay put during the summer months. We'll take you through the data below.

Why Is There Seasonality to Stock Prices?

The stock market's seasonal rhythms have long been yet another aspect of investing that traders have tried to predict. The underlying causes of these price movements remain a subject of heated debate among financial experts. However, summer vacations and the like leading to lower trading volumes in the summer is an easy one to make. But, the percentage change doesn't correlate with the changes in demand from lower trading volumes. Since there are seasonal divergences in stock prices, we need to examine the potential effects of human psychology, institutional behavior, and macroeconomic forces.

The summer doldrums are not the only supposed calendar effects at least some investors watch for. Others include the "January effect," which posits that stocks, particularly small caps, tend to rise at the beginning of the year. This could be attributed to tax-loss harvesting in December, followed by reinvestment in January, or to the psychological boost of starting a new year with fresh optimism. However, as markets have become more efficient and aware of this pattern, its predictability and magnitude have diminished.

A look at the SPDR S&P 500 ETF (SPY) for the decades from 1993 to 2023 shows there have been 17 winning January months (57%) and 13 losing January months (43%), making the odds of a gain slightly higher than the flip of a coin.

Institutional factors can also play a role in seasonal price differences. The end of the fiscal year for many mutual funds in October can lead to portfolio rebalancing and "window dressing"—the practice of selling underperforming stocks and buying high performers to improve the appearance of quarterly reports. This activity can create price pressure in certain sectors, contributing to seasonal patterns.

Why Not Sell in May and Go Away?

The drawback of using historical patterns for trading is not just that they don't reliably predict the future. It's also that, once recognized, they become a thing of the past as others in the market try to take advantage of them. If enough people were to become convinced that others are really selling on the idea of moving into bonds in May and coming back in November, there would no longer be an advantage to be gained. Early-bird sellers would sell in April and bid against each other to buy the stocks back in October.

Even if this seasonal divergence in the market remained, exiting equities in May and reentering in November would historically have led to many missed prospects. Using figures on the S&P 500, we see a general trend of winter months outperforming summer months. However, the picture is far more nuanced once we're talking about the trading circumstances for any particular year—let alone if we look at your particular circumstances, trading goals, and so on.

For instance, suppose we were given $300 to trade for the winter and summer of 1990, and we wished to try an experiment, investing $100 only in the winter months in the future (and taking our gains out of the market at the end of each April), investing $100 just in the summer months (and doing the same on Oct. 31 each year). The chart below depicts this scenario (leaving trading costs and other fees aside but including dividends as part of these returns):

The $100 you would have put in during summer would now be $249.33. While less impressive than the winter-only strategy, this growth is a significant positive return that investors would miss out on if they exited the market during the summer months. That's because these figures would compound with the other months to produce a higher overall set of annual average returns, including dividends.

In addition, the data shows many years when the summer performance was strong or even outpaced winter returns. Take 2020 as a striking example: summer returns were an impressive 24%, while winter returns were negative at -7.7%. Similarly, in 2009, summer returns were 21.9% compared with winter's -9.4%. These examples highlight that automatically following the "sell in May" strategy could result in missing out on substantial gains in specific years.

It's also worth noting that consistently timing the market by selling in May and buying back in November incurs transaction costs and potential tax implications that will eat into returns. This strategy ignores the benefits of dollar-cost averaging and dividend reinvestment, which are often significant contributors to long-term portfolio growth.

The seasonal tendency's averages also conceal big fluctuations from year to year. In any given year, the influence of seasonality is swamped by various other, often more pressing, causes of market changes.

Elections also tend to disrupt the May to October slump. Ned Davis Research's Ed Clissold has noted that since 1950, the S&P 500 has risen 78% of the time from April 30 to Oct. 31 in presidential election years. That compares with about 64% for nonpresidential election years in the same period.

Finally, it's unclear if the reasons are actually seasonal rather than being related to specific months or other calendar-based moments. Here are the average monthly returns for three of the major American stock indexes since their appearance:

Below are the monthly returns just for the S&P 500 for the decade from 2014 until the end of 2023. The data suggests less that the warmer period is a problem, as opposed to specific months, e.g., September.

Alternatives to "Sell in May and Go Away"

Instead of acting on the adage and exiting stocks, investors who believe the pattern exists could rotate from the higher-risk market sectors to those that tend to outperform in periods of market weakness.

A report by the Center for Financial Research and Analysis (CFRA) argues that cyclical sectors, such as the consumer discretionary, industrials, materials, and technology sectors, tend to outperform from November through April, and more defensive (i.e., generally recession-proof) consumer staples and healthcare sectors during the summer period. Since 1990 the S&P 500’s consumer staples and healthcare sectors rose 4.1% on average from May to October, higher than the broader market.

The CFRA's materials are included in the marketing materials for an exchange-traded fund (ETF) that's looked to take advantage of seasonal shifts in the market. The premise of Pacer CFRA-Stovall Equal Weight Seasonal Rotational ETF (SZNE) is that a custom index representing the strategy of rotating between healthcare and consumer staples stocks held from May to October and more economically sensitive market sectors from November to April would have significantly outperformed the S&P 500.

As of the third quarter of 2024, that has not been the case, looking at annual returns back to SZNE's launch in July 2018:

Even as it didn't match a buy-and-hold strategy for the S&P 500 index, its fees were nevertheless 12 times higher: a typical S&P 500 passive index fund charges a 0.05% annual fee; SZNE charges 0.60%.

Time in the Market vs. Timing the Market

Most investment professionals dislike the "sell in May and go away" adage because it encourages investors to keep chopping and changing their portfolios. In reality, most investors are better off adopting a buy-and-hold strategy, and we've seen how even professionals can't manage to do better than simply holding the S&P 500 for the long haul.

Hanging on to equities year-round, year after year, unless there's a change in fundamentals, helps cut fees, removes the risk of panicking and taking irrational action, and generally leads to better returns. There's abundant evidence to support this.

For example, Charles Schwab provides a helpful analysis of different ways of trading to show the potential returns. In their scenario, they suppose several investors were given $2,000 yearly for 20 years to invest, beginning in 2003. Here's how much each ended up with at the end of the period:

• An investor with perfect market timing (waiting to buy the S&P 500 at its low each year): $151,391
• An investor who immediately invested the $2,000 at the beginning of each year: $135,471
• An investor who uses dollar-cost averaging (putting away the amount periodically): $134,856
• An investor with bad market timing (buying the S&P 500 index at its annual high): $121,171
• An investor who left the accumulating money in cash: $44,438

The point of the analysis is to show how simply putting one's money into a broad market index pays. But it's also notable that, since your chances are pretty slim each year picking the exact low to buy more shares in an S&P 500-tracking ETF, your results are not that much worse for wear for it.

The Bottom Line

The "sell in May and go away" adage is a big talking point among investors. If you look at returns over some decades, stocks, on average, tend to perform worse during these six months. However, there are also many exceptions, and the average performance during the warmer months isn’t enough to warrant investors making significant changes to their portfolios.