It’s that time of year again! No, I’m not talking about the festive December. I’m actually referring to January. Although, known as a lousy month, filled with abandoned New Year’s resolutions and bad weather, it’s on average the best month for equities. So, from an asset pricing perspective, January isn’t that bad after all. For the past 90 years, the S&P 500 has yielded an average 1.2% return in January, making it one of the best performing months for investors. This empirical tendency has been dubbed, as you can guess, the “January effect” by many financial economists and Wall Street folks. This seasonal effect is a very common and discussed topic at the start of each year.
Why is the January effect so intriguing? Although the simple nature of this effect, it directly strikes and questions traditional asset pricing models. The efficient market hypothesis (EMH), which underlies key finance models such as the CAPM, predicts that asset prices follow a random walk. Therefore, it should be impossible to predict future returns based on publicly available information. It also implies that we should not be able to detect persistent seasonal patterns in asset returns. The January effect contrasts the EMH, because it simply implies seasonality in asset returns. It shows the limitations of the traditional models that we are taught and raises awareness to develop more empirically robust models.
The first one to document this interesting effect was Rozeff and Kinney (1976), who found that the average return in January, from 1904 through 1974 using a sample of NYSE prices, was 3.5% whereas the average return in the remaining months was 0.4%. Subsequently, the effect has gained much attention, for decades actually, from academics as well as practitioners. Other researchers have confirmed these findings in more recent and international samples. Also, the January effect seems to be in particular relevant for small firms (Roll, 1983). Despite the robust empirical evidence, many opponents of this seasonal anomaly assign this pattern to mere data-mining and data-snooping whereas proponents point out to behavioural explanations and limits to arbitrage.
So what exactly causes the January effect to persist over time? Almost 50 years after its documentation, we actually still do not have a full grasp. Two of the most prominent explanations of the January effect are the “tax-loss selling” hypothesis (Wachtel, 1942) and the “window dressing” hypothesis (Haugen and Lakonishok, 1988). Under the tax-loss selling hypothesis, retail investors sell stocks that have declined in price to realize the tax losses. These investors then wait until January to reinvest, as buying is augmented by cash infusions from year-end bonuses or from the sales of large caps on which capital gains have been realized. Essentially this is the case of “parking the proceeds” into small caps. However, this explanation can’t fully capture the January effect. Although tax-loss selling has been used for decades as a plausible explanation of the January effect in the United States, international evidence also suggests a January effect in equities for countries with different tax systems, and even in countries without capital gains tax. In addition, retail investors have increasingly allocated their wealth into tax-sheltered plans like 401K’s and IRA’s in recent years.
Multiple studies, however, argue that the anomaly stems from an end-of-year window dressings by institutional investors that try to hide their losers from their portfolios prior to the end of the reporting period. However, a reasonable assumption is that institutional ownership is predominantly concentrated in large caps rather than small caps. The January effect is rather a small cap phenomenon. So, the window-dressing hypothesis has limited power to explain the January effect, in my opinion. One other argument, which is more plausible, but harder to test, is the investor psychology. Many individual investors follow up on New Year’s resolution starting from January, and the momentum and energy associated with the need to start investing for the future may cause a rally in the stock market.
More naturally, perhaps for you, is to ask whether this asset pricing anomaly can result into a profitable trading strategy. This questions, however, is also a difficult one. As I mentioned before, the January effect is a small firm phenomena. Due to the illiquid nature of such equities, trading volumes are small and bid-ask spreads can be substantial for retail investors, giving rise to limits to arbitrage. A profitable trading strategy for retail investors is unlikely, but not impossible. This, however, does not render the January effect uninteresting.