not contiguous than when they are contiguous. When the holding period
and the ranking period are contiguous, the profits to the momentum strat-
egy are attenuated by the negative serial correlation in returns induced by
bid-ask spreads, and by short horizon return reversals. Grundy and Martin
note that stocks in the random industry portfolio tend to be smaller than
the stocks in the real industry portfolio, and hence it is likely that the re-
turns to the random industry strategy are more adversely affected by the
bid-ask bounce. Also, a strategy that selects stocks based on their industry
returns would tend to suffer less from reversals due to bid-ask bounce than
a strategy that selects stocks based on their own returns.
In the case of industry momentum however, the profits entirely disappear
for the six-month ranking period when the ranking period and the holding
period are not contiguous. Therefore, industry momentum seems to benefit
from positive first-order serial correlation in industry returns while individ-
ual stock momentum is hurt by short-horizon return reversals. These results
indicate that the momentum strategies most commonly used in the literature
clearly benefit from the predictability of firm specific returns, although they
may also benefit from industry momentum.
A recent paper by Lewellen (2002) also finds that industry portfolios
generate significant momentum profits. Lewellyn, however, concludes that
industry momentum is driven primarily by a lead-lag effect within industry.
Specifically, his evidence suggests that industry portfolio returns tend to
move too much together. In addition, he finds that size and B/M portfolios
also exhibit momentum. The overall evidence suggests that neither industry
momentum nor firm-specific momentum subsume one another, but they are
both important.
3 .Behavioral ModelsAs we mentioned in the introduction, it is very difficult to explain observed
momentum profits with a risk-based model. Therefore, researchers have
turned to behavioral models to explain this phenomenon. Since these mod-
els are described in greater detail elsewhere in this book, we will provide
only a brief description of them in order to motivate some of the more re-
cent empirical work on momentum.
Most of these models assume that the momentum-effect is caused by se-
rial correlations of individual stock returns, which as we discussed above,
appears to be consistent with the evidence. However, they differ as to
whether the serial correlation is caused by underreaction or delayed overre-
action. If the serial correlation is caused by underreaction, then we expect
to see the positive abnormal returns during the holding period followed by
normal returns in the subsequent period. However, if the abnormal returns
are caused by delayed overreaction, then we expect that the abnormal
370 JEGADEESH AND TITMAN