I have long suspected that the momentum anomaly is driven by earnings momentum. Chordia and Shivakumar (2005) have studied this extensively:

This paper narrows the search for an explanation to the anomalies by documenting that the price momentum anomaly is a manifestation of the earnings momentum anomaly.

furthermore:

The finding that price momentum is subsumed by a common factor related to the macroeconomy is significant since it does not rely on capital market frictions to explain the price momentum effect.

i.e. market inefficiency is not required to explain the momentum anomaly!

The paper uses a standardized unexpected earnings (SUE) metric: standard deviation of 12 month earnings change:

SUE for month t = (Eit– Eit-4)/sit , where Eit is the most recently announced earnings and sit is the standard deviation of (Eit-Eit-4) over the prior 8 quarters.

Decile SUE portfolios are formed (PMN = P10-P1):

from January 1972 through December 1999, the monthly holding period returns increase monotonically from 0.79% for the lowest SUE portfolio, P1, to 1.68% for the highest SUE portfolio, P10.

For comparison, the price momentum returns are similar:

Over the entire sample period from January 1972 through December 1999, the monthly holding period returns increase from 0.84% for the lowest past-return portfolio, P1, to 1.60% for the highest past-return portfolio, P10.

However, price momentum (WML) has a much larger deviation:

the monthly standard deviation of returns of PMN is 2.21% while that for WML is 5.62%. These results are consistent with WML being a noisy measure of PMN.

PWM is also successful at predicting 12 month ahead economic activity, for example Industrial Production Growth (IPG):

Note the negligible coefficients for the Fama-French size and value factors. See the paper for more details.