00Thaler_FM i-xxvi.qxd

360 JEGADEESH AND TITMAN

Table 10.3

Momentum Portfolio Returns in January and Outside January

This table presents the average monthly momentum portfolio returns. The sample
includes all stocks traded on the NYSE, AMEX, or NASDAQ excluding stocks
priced less than $5 at the beginning of the holding period and stocks in smallest
market cap decile (NYSE size decile cut off). The momentum portfolios are
formed based on past six-month returns and held for six months. P1 is the equal-
weighted portfolio of ten percent of the stocks with the highest past six-month
returns and P10 is the equal-weighted portfolio of the ten percent of the stocks
with the lowest past six-month returns. The sample period is January 1965 to
December 1998.

Percent

P1 P10 P1−P10 t-statistic Positive

Jan 3.40 4.95 −1.55 −1.87 29
Feb–Dec 1.49 0.01 1.48 7.89 69
All 1.65 0.42 1.23 6.46 66

Source: Jegadeesh and Titman (2001).

have found a January seasonality for other anomalies such as the size effect
(see Keim, 1983) and long-horizon return reversals (DeBondt and Thaler,
1985). In fact, these anomalies are almost entirely concentrated in January,
and are insignificant outside January. In marked contrast, the momentum
effect is entirely a non-January phenomenon and January returns hurt the
momentum strategy.

2 .Potential Sources of Momentum Profits

A natural interpretation of the fact that past winner and losers continue the
trend is that stock prices underreact to information. For example, suppose
a firm releases good news but the stock prices only react partially to the
news. Then, stock prices would rise during the news release but will con-
tinue to rise even after the news becomes public, as the market fully adjusts
to the new information. Therefore, underreaction is one possible source of
momentum profits. However, this is not the only possible reason why past
winners outperform past losers. Another possible reason is that past winners
may be riskier than past losers, and the difference between winner and loser
portfolio returns could simply be compensation for risk. Also, if the premi-
ums for bearing certain types of risk vary across time in a serially correlated
fashion, momentum strategies will be profitable. We can use the following