Improving momentum strategies

I continue to argue that the classical assumption in finance and economics of a representative agent is deeply flawed. In my book, I have shown how dropping this assumption and adopting a framework of many different investors with different goals and strategies can explain the swings in markets between value and momentum.

Now, I have come across a study by Iván Blanco, Miguel de Jesus and Alvaro Remesal that shows that this view of markets can help improve the performance of momentum strategies. Traditionally, momentum strategies are based on 12-month trailing price momentum (typically calculated as 12-month minus 1-month return to remove short-term reversals) and we know for decades, that momentum is one of the best ways to systematically outperform the market.

But different investors are trying to get an edge by using different momentum formation periods. Some investors use 6-month momentum to get into the market before the 12-month momentum investors do, others use even shorter formation periods. This creates an entire collection of different momentum strategies active in the market. 

If a stock has both good 6-month and 12-month momentum, the demand of the 12-month momentum investors and the 6-month momentum investors accumulates, and we should expect even better performance for these stocks than the stocks that only show up on a 12-month momentum screen. So, what the Blanco and his colleagues did was a simple exercise of looking at the 10% best and worst stocks based on both 12-month momentum and 6-month momentum. Then they looked at the stocks that showed up in both lists and the stocks that showed up only on the 12-month momentum list. The results of going long the best 10% and short the worst 10% for the traditional 12-month momentum strategy are already impressive, but if one selects only the stocks that overlap and appear in both the 12-month and 6-month momentum strategies, it gets even better. The additional demand for these stocks from a different group of momentum investors leads to outperformance, while the stocks in the non-overlapping group lag the traditional momentum approach. 

Excess return of momentum strategies over the market

Source: Blanco et al. (2020)

This picture stays the same if one looks at the Fama-French 5-factor alphas of these strategies and interestingly, the overlapping momentum approach does not introduce excess volatility in the returns either. In other words, it is a simple technique that many investors can easily implement to improve their returns without taking on additional volatility in their portfolios.

5-factor alpha of momentum strategies

Source: Blanco et al. (2020)