In the paper Momentum Returns and the Role of Liquidity Improvements (Jeppe Bro, March 28, 2026, SSRN) argues that the momentum anomaly arises from cross-sectional liquidity dynamics rather than representing an independent risk premium. Past winners systematically improve in liquidity prior to the portfolio formation period, whereas past losers severely deteriorate (summary).
What if we used this LIQIM to setup a long-only portfolio of 20 Indian stocks?
It doesn’t quite work that way. So, we looked at the return stats of LIQIM quintiles and figured that we could use that to enhance momentum by filtering out stocks that had degraded liquidity (backtest).
Turns out, there is about 50 bps of performance that you can squeeze out of a typical long-only momentum strategy using this liquidity filter.
Code and charts are on github.