Investment Research database over the 1981 to 1984 sample period. He
considers various measures of Up and Down revisions based on both indi-
vidual analyst’s forecast revisions and consensus forecast revisions. Stickel’s
Up and Down revision portfolios comprise 5 percent of stocks with the
highest and lowest forecast revisions. He finds that the Up minus Down re-
vision portfolio earns 7.07 percent using consensus forecast revisions, and
6.36 percent using individual analyst’s forecast revisions.
Chan, Jegadeesh, and Lakonishok (1996) use the sample of firms cov-
ered by I/B/E/S over the 1977 to 1993 sample period. They define forecast
revision as a six-month moving average of the ratio of consensus earnings
forecast revision to the stock price. The Up and Down revision portfolios
in Chan et al. comprise the decile of stocks with the largest and smallest
forecast revisions, respectively. Chan et al. find that the Up minus Down
revision portfolio earns 7.7 percent six months after portfolio formation,
and 8.7 percent after twelve months. The profitability of analyst forecast
revision strategy is also relatively short-lived, which is similar to the SUE
strategy.
The collective evidence in the literature indicates that the analyst forecast
revision strategy is remarkably robust. The profitability of this strategy is
not sensitive to the specific definition of forecast revisions, nor is it sensitive
to the data source for analyst forecasts. Also, both the SUE strategy and the
forecast revision strategy continue to be profitable although the original ev-
idence was published over twenty years ago.
A. Relation between Earnings and Return Momentum StrategiesChan et al. (1996) present a detailed analysis of the interactions among var-
ious momentum strategies and this subsection closely follows that paper. As
they point out, it is possible that a price momentum strategy is profitable
mainly because price momentum and earnings momentum are correlated,
and earnings momentum may be the dominant source of return predictabil-
ity. Alternatively, strategies based on price momentum and earnings mo-
mentum may be profitable because they exploit market underreaction to
different pieces of information. For instance, earnings momentum strategies
may exploit underreaction to information about the short-term prospects
of companies that will ultimately be manifested in near-term earnings. Price
momentum strategies may exploit slow reaction to a broader set of value-
relevant information, including the long-term prospects of companies that
are not fully captured by near-term earnings forecasts or past earnings
growth. If both these explanations were true, then a strategy based on past
returns and on earnings momentum in combination should lead to higher
profits than either strategy individually.
Chan et al. (1996) present the correlation between price and earnings
momentum variables, and their results are reproduced in table 10.12. Not
MOMENTUM 381