Can We Improve Momentum Factor Investing via Salience Theory?

This research examines the potential of enhancing a standard momentum strategy using signals derived from Salience Theory (ST).  The strategy presented here is to exclude stocks with extreme salience scores and then analyze the risk and return properties of the ST strategy.

Salience theory and enhancing momentum profits

• Myounghwa Sim, Hee-Eun Kim
• Finance Research Letters
• A version of this paper can be found here
• Want to read our summaries of academic finance papers? Check out our Academic Research Insight category.

What are the research questions?

Salience Theory (ST) is one variety of behavioral theories of asset pricing, where investors are thought to make investment choices in the context of alternative payoffs.  Within that framework, investors are predicted to focus on stocks with salient payoffs, ultimately contributing to their mispricing.  By definition, the most salient payoff for any one stock is the one that stands out relative to the payoffs of all other stocks.  Since investors pay attention to salient payoffs, they are drawn to and create demand for stocks with salient “upsides.”  Given the excess demand that results, those stocks become overvalued and are expected to produce lower future returns.  Similarly, stocks with salient “downsides” are characterized by lower demand and, as a result become undervalued and are expected to produce higher future returns.

What are the Academic Insights?

The predictions of Salience Theory were successfully used to enhance a standard momentum strategy. The sample included stocks listed on the NYSE, AMEX, and Nasdaq observed from 1965 to 2021. Momentum portfolios were formed consistent with Jegadeesh and Titman (1993), using cumulative returns from month t-12 to t-2. Momentum profits consisted of the winner vs. loser portfolio spreads. The ST measure was constructed consistent with Cosemans and Frehen (2021), basically a distance metric. Recall that the typical momentum strategy involves buying past winners and selling past losers.

However, in this application, the extremely salient stocks in momentum portfolios are more likely to have subsequent reversals and are therefore excluded from the portfolio. Specifically, past winners with the highest ST and past losers with the lowest ST exhibit the largest differences from the market and are likely to experience return reversals. The results of eliminating the highest and lowest ST momentum stocks are presented in the chart below. The profitability of the ST momentum strategy is expected to outperform the standard momentum version. The cumulative profits and drawdowns from a 5% and 10% salience screen are presented in Panels A and B and compared to the no screen or standard momentum strategy. The results were quite good.  For the 5% screen the five-factor alpha increased from 1.641% (no screen) to 1.749%. The Sharpe ratio also increased from 0.852 to 0.927. 

Why does it matter?

Salience theory’s contribution to asset pricing theory turns out to be valuable insight for momentum strategies. Excluding stocks with extreme salience values enhances the basic return effect established in the literature on momentum. Focusing on the extremity of returns and how investors perceive it clarifies the fundamentals that drive short-term return reversals. 

The most important chart from the paper

Abstract

Salience theory predicts that stocks with salient upsides are overvalued and earn lower subsequent returns, whereas those with salient downsides are undervalued and generate higher future returns. This study investigates an enhanced momentum strategy excluding stocks with extremely salient payoffs which attenuate the profitability of the momentum strategy. We find that this approach generates a significantly higher return and Sharpe ratio than the traditional momentum strategy. We further find that the performance improvement is more prominent for loser portfolios than for winner portfolios.