Twain was on the right track. Some market followers believe there is a “January Effect” leading to greater stock price increases in January than in other months. In fact, there are a number of calendar effect theories that are closely followed by some.

As much as economists would have you believe that all economic theories are grounded in empirical evidence, that is not always the case. Too often, with so much data available, theories can be tested and “proven” by compiling, or mining, the right data. Yet there is strong evidence of seasonal regularities in stock returns. Richard Thaler, writing in the inaugural issue of Journal of Economic Perspectives (1987), first identified these anomalies. He also put his work in plain English in his book, The Winner’s Curse.

The efficient market hypothesis states that stock prices should follow a random pattern and not be subject to seasonal influences. However, with data on daily prices going back to the 1920s, researchers have found evidence that supports the theory that the returns in January tend to be higher than in other months. Work by Rozeff and Kinney (1976) showed that the average returns in January exceeded the average returns of all other months by about 3%, with almost one-third of the annual returns occurring in January.

Some who have studied the January Effect have attempted to attribute it to year-end tax-loss selling. By January, the pressure to sell is over and buyers begin accumulating shares, bidding the prices up. To test this, Gultekin and Gultekin (1983) looked at patterns in 16 countries and found that January returns were exceptionally large in 15 of them, suggesting that tax-loss selling is not relevant.

Studies similar to those noted above seem to confirm the existence of other “calendar pricing” anomalies. For instance, stock prices seem to go up on Fridays and down on Mondays, and according to Ariel, the day before a holiday appears to have a greater percentage of advances over declines (1985). Ariel also tested data from 1963–1981 for intramonth trading and found that the first four days of each month plus the last day of the prior month averaged .473%, a number greater than the average total monthly return of .35%! In other words, aside from the four days around the turn of every month, the DJIA falls!

The “why” behind these anomalies remains a puzzle to economists. The fact that they do occur, and continue to occur even after significant documentation, would imply that the pricing differentials are not significant enough for trader speculation.

Thaler offers three possible explanations for these calendar effects:

First, price movements may be influenced by the timing of funds into and out of the market, such as pension deposits that are routinely made at certain times of the month or year.

Second, intuitive managers routinely engage in “window dressing” their portfolios in anticipation of reporting deadlines as they get rid of poorly performing assets. Since reporting deadlines typically occur at month and year-end, this may explain some of the seasonal price movements.

Finally, the systematic timing of good and bad news may have something to do with price movements. If bad news is systematically posted after the close of Friday’s markets, Monday is usually not a good day.

In any case, the fact that these calendar effects seem to exist provides welcome fodder for the financial journalists seeking to explain market movements they might not otherwise understand.