Date-driven 📅

Investors have an almost religious devotion to the calendar, seeking patterns in the seasons, holiday bounces, and quarterly rituals.

January favours small caps. Summers are sluggish. Holidays mean rallies. None of these are rules per se, more like market folklore. Nonetheless, these predictive strategies shepherd money in and out of the market. What’s weird is that many seem to have some efficacy.

Perhaps seasonal strategies’ endurance reflects the fact they offer a framework to hang beliefs on. Humans also have an uncanny ability to fit the data to a narrative. Many of these seasonal myths are based on patterns that persist for decades or more. 

It helps they often align with structural flows. Pension contributions arrive monthly, companies report quarterly, tax deadlines in April, holidays compress trading. These are real forces driving volume and momentum. Other strategies survive because they’re memorable, shareable, and occasionally self-fulfilling. 

It’s calendar as a kind of meta-indicator. Navigating uncertainty, anchoring to dates. Humans, like all of God’s creatures, are beholden to seasonal shifts. Why should markets be immune?

‘Tis the season

Seasonality, the differences in asset returns across calendar periods, has been studied for more than a century. The broader category is known as calendar effects, covering day-of-week effects, monthly patterns, election cycles, and more. 

The grandaddy of seasonal strategies is Sell in May and go away AKA the Halloween strategy. It’s the belief the six-month stretch from May through October tends to underperform relative to the November-April interval. 

Studies like that of Bouman & Jacobsen have found this effect in many countries, stretching back centuries in some cases. The effect, strongest often in Europe, suggests that returns in the ‘winter’ half-year tend to be higher (on average) than during the ‘summer’ half-year. 

The Santa Claus rally is another seasonal mantra. Yale Hirsch coined the term in the 1973 Stock Trader’s Almanac. It reflects the tendency for equity markets to rise in the last days of December and the first few trading days of January. Since 1950, during this seven-day trading window, the S&P 500 has gained an average of 1.3% and been positive 79% of the time.

US prescient

In the US, one of the most enduring seasonal stories is the presidential cycle. Introduced by the Stock Trader's Almanac back in 1967, it suggests the third year of a president’s four-year term tends to deliver stronger equity returns – as incumbents look to juice the economy ahead of re-election campaigns. It’s the economy, stupid. 

‍Studies have indeed shown, looking across the twentieth century, the average stock market performance in year three was often higher than in years one, two, or four. But the record since the late 1990s has been patchier. Globalised capital, partisan gridlock, and more assertive central banks mean the link between White House and Wall Street has weakened. 

Almanacs – compendiums of seasonal knowledge, omens, and prognostications – go back way further than 1967. During the sixteenth century almanacs and calendars were the most frequently printed books besides the Bible. Historians such as Alexandra Walsham have shown how cultures once relied on almanacs to interpret both nature and politics. Modern market almanacs echo that same instinct.

‍Sell Rosh Hashanah, buy Yom Kippur dates to a time when Jewish holidays had a greater impact on NYC trading desks. Liquidity would briefly thin around these pauses. But in today’s bigger, electronic markets the effect has no statistical heft.

Then there are the summer doldrums. For much of the twentieth century this lull was a tangible phenomenon. Back in the day, traders made like Plissken and escaped from New York for the holidays. But in our always-on world, the case for a seasonal slowdown is weaker. Investors still talk of listless summers, when volatility dips and news flow feels anaemic. But this one is now much more vibe than law.

I want to believe

The pull of the calendar is real. But the record of seasonal strategies is uneven. Many effects shrink or scram once subjected to the rigour of testing, and especially once costs, taxes, and behaviours are factored in.

In some markets or periods, May-October returns are positive or even strong. But the big dog of seasonal strategies is losing its bite. There’s an argument that Sell in May may partly arise from data mining. A few extreme seasonal crashes or structural shifts may dominate the averages. The effect tends to weaken as it becomes more widely known, as more investors try to front-run it. 

The Halloween effect shows up in many national markets but some indices or years diverge sharply from the pattern. One study reviewing 62,962 observations across global indices observed that November-April returns average around 4% higher than May-October, but with large dispersion.

The Santa Claus rally is also unreliable. Surprise, surprise. The lack of a strong December-January pop is sometimes deemed a warning sign, but it is far from a consistent lead indicator.

The biggest challenge of all, major macro events, are often uniquely unexpected. Even small fry like a central bank pivot, a geopolitical shock, or an earnings surprise can swamp calendar effects entirely. When fundamentals assert themselves, the primacy of the calendar diminishes.

Human nature

As we know, markets are not purely mechanical systems. The persistence of seasonality in cross-country studies, across decades, suggests there’s something more than just randomness at play. The fact that Bouman & Jacobsen found the Sell-in-May effect in 36 of 37 countries has to be evidence there is something structural underpinning it.

A missing Santa rally may not predict recession, but it colours the way investors talk about December. A strong summer run may be viewed with suspicion precisely because it defies the Sell in May refrain. These patterns frame sentiment and highlight portfolio positions, even as the statistical case has softened.

When enough people believe in a pattern – that October is cursed, that November brings relief, that January restores risk appetite – the belief itself can nudge behaviour in that direction. When it works, the why rarely matters.

‍

Important Information

‍