At ExtractAlpha we were excited to read a recent article in Bloomberg describing new research from S&P Capital IQ which independently verified our research finding that stocks with increasing levels of liquidity tend to outperform. This liquidity change factor is built into ExtractAlpha’s TM1 model in the form of our Liquidity Shock component, which is available to TM1 clients broken out as a separate score.
We have a somewhat different take from S&P on the driver of the Liquidity Shock factor. Rather than speculating about the speed of the market’s reaction to a hypothetical illiquidity premium, we postulate a more mechanical explanation, from a practitioner’s perspective: that as a stock becomes more liquid, large institutional investors are able to add the name to their long only portfolio universe, whether they are active or passive managers, and the subsequent flows result in a strong positive short-term impact on the stock’s price. Conversely, a stock dropping below liquidity thresholds results in outflows and negative impact. This asymmetry – the fact that large, liquidity-sensitive investors are predominantly long rather than market neutral – explains why on balance an increase in liquidity is a good thing, as shown by the nearly monotonic returns across TM1’s Liquidity Shock deciles; top-ranked stocks outperformed bottom-ranked stocks by 14% per annum.
S&P’s researchers found some predictive power for liquidity shocks at relatively long horizons. Our findings indicate that the factor is actually much more powerful at relatively short horizons. Similarly to S&P, we find that the factor can be correlated to other risk factors such as momentum if not controlled properly, and we find that stripping out these risk exposures – as is done already in TM1 – makes the Liquidity Shock factor more powerful and consistent.
Unlike many quant factors, Liquidity Shock has not decayed in recent years andcontinues to perform well, as shown in the cumulative return graph below.
Finally, S&P’s report indicates that conditioning stock price reversal factors on liquidity shocks makes the reversal effect stronger. We find that this is because short-term stock moves tend to reverse less for stocks with sudden improvements in liquidity; in these cases, the recent stock moves are most likely to be driven by news and therefore more likely to be permanent rather than transitory. This conditional logic is also already built into TM1’s Reversal Component and is one reason TM1 consistently outperforms simpler reversal models.
To discuss how TM1 can improve your stock selection strategies, reach out to us!