Home / Wiki / S / Seasonality

Seasonality

Trading and SpeculationUpdated 16 Nov 2026

Seasonality, in markets and economics, is a regular pattern in a price or a data series that repeats over a fixed calendar period, such as a year, a quarter, a month or a week. Retail sales rise before Christmas, heating fuel demand peaks in winter, and grain prices often reflect the timing of planting and harvest. In stock markets the term refers to recurring calendar patterns in returns, such as the tendency for certain months or days to perform better than others on average.

Seasonal patterns that come from physical or institutional causes, such as weather, holidays or tax deadlines, are generally reliable and are removed from economic statistics through seasonal adjustment. Seasonal patterns in asset returns are more controversial. Under the efficient market hypothesis, a predictable pattern in prices should be traded away once it is widely known, and many apparent calendar effects have weakened or vanished after they were published.

01

Seasonality in economic data

Many economic series swing for reasons that have nothing to do with the underlying trend. Construction slows in cold months, hiring jumps before the holiday shopping season, and factory output falls during summer shutdowns. Statistical agencies such as the US Census Bureau and the Bureau of Labor Statistics publish seasonally adjusted figures so that one month can be compared with the next. A simple way to see the idea is a seasonal index for each month:

Adjusted value=Raw valueSIm\text{Adjusted value} = \frac{\text{Raw value}}{\text{SI}_m}

Here SIm\text{SI}_m is the seasonal index for month mm, equal to that month's typical level divided by the average level across all months. If December retail sales usually run 30 percent above the monthly average, the December index is 1.30. A raw December figure of 143 then becomes an adjusted 143 divided by 1.30, or 110, showing an underlying level 10 percent above an average month. Official agencies use more elaborate methods that also allow for trading days, moving holidays and changing patterns, but the principle is the same.

02

Seasonality in commodities

Commodity markets have the clearest seasonal patterns because supply and demand follow the weather and the farming calendar:

  • Energy. Natural gas demand peaks in winter for heating, measured by heating degree days, and electricity demand often peaks in summer for air conditioning, measured by cooling degree days. Gasoline demand rises in the northern summer driving season.
  • Agriculture. Grain prices are influenced by planting, weather during the growing season and the arrival of the harvest, which tends to increase supply in the autumn in the northern hemisphere.
  • Futures curves. These patterns appear in the prices of futures for different months, and traders use a calendar spread between winter and summer contracts to trade them.

Because the seasons are largely known in advance, much of the effect is already reflected in futures prices. The trading risk lies in departures from normal, such as an unusually warm winter or a drought.

03

Calendar effects in stock returns

Researchers and practitioners have catalogued many calendar patterns in equity returns. The best known include:

  • The January effect. Small company shares, as measured by indices such as the Russell 2000, historically outperformed in January, an effect often attributed to selling for tax losses in December and repurchases in the new year.
  • Sell in May. Returns from November to April have historically been stronger on average than from May to October in many markets, a pattern also called the Halloween indicator.
  • Turn of the month. Returns have tended to cluster in the last trading day of a month and the first few days of the next, possibly linked to the timing of salary and pension flows.
  • Year-end effects. Fund managers' window dressing before reporting dates and the so-called Santa Claus rally in late December are also cited.

September has been the weakest month on average for US stocks over long periods. Market calendars and the economic calendar of scheduled data releases and central bank meetings add their own patterns of volatility.

04

Evidence and criticism

Most calendar effects in stock returns are small compared with the normal month-to-month variation in prices, so long samples are needed to detect them, and they vary greatly between periods and countries. Several, including the January effect in large US stocks, have weakened since they became widely known, which is consistent with investors trading them away.

A broader criticism is data mining. When researchers test many possible calendar patterns on the same history, some will appear significant by chance. Patterns without a clear economic cause are the most likely to be statistical accidents, while those tied to tax rules, cash flows or reporting dates are easier to explain. Transaction costs, taxes and the risk of a pattern failing also reduce or remove any advantage from trading on them. Seasonality is therefore widely used to interpret data and commodity prices, but it offers no assurance of future returns in any market.

Last updated 16 Nov 2026