Momentum: A Sword Without a Hilt

Momentum Performance – A First Look

Of the 4 most commonly used investment factors – market, size, value, and momentum¹ – we’ve seen momentum produce the highest alpha by an absurdly wide margin. However, it has also proven to be the most dangerous factor.

The chart and table below summarize some key performance statistics for those big 4 financial factors². There are immediately several noteworthy takeaways:

• Momentum outperforms the other factors by multiple orders of magnitude.
• Momentum is the most volatile factor by a substantial margin.
• Momentum’s alpha – at 19.87% a year – feels ludicrously high.

Long-Short Factor Portfolios: Growth of $1

Table 1. Factor Performance Statistics

The Dark Side of Momentum

After the quick analysis above, one might conclude that momentum is the most dominant factor: sure, it appears to have a bit higher volatility, but with returns that high, a little extra volatility could be an acceptable price to pay. Unfortunately, looking just at volatility alone doesn’t really convey the true dangers that momentum investors may face.

The momentum factor occasionally experiences extremely severe and prolonged drawdowns – devastating to such an extent that they’ve earned the nickname of “momentum crashes³”. The two most notable instances stand out for their severity and length: during the Great Depression, momentum suffered a drawdown exceeding 95%, and following the Great Financial Crisis, it experienced a drawdown of 81%.

These are life-altering crashes for investors who withdraw their money near the bottom and career-ruining for those individuals managing other people’s money. We know that risk and return are intrinsically related, so momentum’s high long-term return potential now seems more appropriate given the price one pays to get there: extended periods of devastating losses.

Momentum as a standalone investment strategy could result in catastrophic outcomes, but what if it was used as a tool within a broader investment strategy: would you then be able to harness some of its benefits and mitigate the likelihood of experiencing the devastations seen above? To do this, we need to first understand why it crashes in the first place.

So why does momentum crash?

Taking a look under the hood to see how the momentum factor is constructed allows us to better understand why momentum tends to crash following periods of prolonged market downturns.

The momentum factor is just a dollar-neutral long/short portfolio: the long portion is a basket of securities that have had the strongest returns over the prior 12 months (excluding the most recent month) and the short component is a basket of securities with the worst returns in the prior 12 months (excluding the most recent month)⁴. A consequence of this methodology is that the momentum factor tends to have much higher turnover than its peers, and this high turnover results in portfolio characteristics that are quite dynamic – most notably, its beta.

When you plot the rolling betas for the components of the momentum factor during any of the excessive draw-down periods, the crash narrative is clear:

1. The overall market experiences a prolonged drawdown.
2. The prolonged market downturn causes a mechanical shift in momentum portfolios:
• The long side buys stocks that have held up relatively okay during the downturn (low-beta stocks).
• The short side sells stocks that have been especially beaten down during the downturn (high-beta stocks).
• Net Result: A portfolio with a significantly negative beta.
3. The market – which has severely overreacted to the downside – shoots upward rapidly: the short-side of the portfolio is absolutely killed.

Below we depict this evolution during the GFC, but a similar pattern emerges if you look at the post-depression crash as well. Our primary observation is that the worst periods for momentum are during sharp melt-ups, which have been preceded by a prolonged market downturn.

Rolling Betas: Post-GFC Period

Managing Momentum: Practical Tips for a Balanced Portfolio

What does all of this tell us: is momentum too dangerous a factor to use in managing portfolios? We think not. You just need to be thoughtful about how you incorporate momentum into the portfolio construction process. Some general tips for the safe use of momentum:

• Risk management is key. Even more so than with other factors, keeping a close eye on how portfolio risk is evolving is crucial:
• Be mindful of the portfolio’s beta relative to its benchmark, especially during periods of prolonged market drawdowns.
• Scale down momentum exposure in periods of heightened volatility, and scale it up during periods of lower volatility.
• Use multiple signals for momentum. The traditional factor is based on 12-1 returns for relative stock ranking, but incorporating elements of time-series momentum and a variety of different lookback windows can result in a more robust signal.
• Combining momentum with other, low-correlated factors like value or profitability is also likely to reduce the portfolio’s crash risk.
• Avoid shorting stocks based on momentum signals. Most of the pain experienced in crashes is driven by the short side of the momentum portfolio. While you are giving up long-term return potential by forgoing this leg of the premium, the diminished chance of devastating losses is likely an acceptable tradeoff for most.

Ultimately, we believe momentum is a powerful tool for enhancing portfolio returns, but it must be carefully applied and can cause devastating damage if used without due care and respect.

 
[1] When we say momentum, we are referring to “cross-sectional momentum”, which ranks stocks relative to each other, in contrast to “time-series momentum” which ranks stocks relative to their own history. Additionally, the momentum factor we depict is constructed by subtracting the bottom decile of loser stocks from the top decile of winner stocks. Some constructions of the momentum factor opt for less extreme breakpoints such as 30/70 percentiles.
[2] Note that the Y-Axis is log-scaled in the chart. This adjustment is necessary because the magnitude of momentum’s outperformance is so ridiculously large the chart looks silly on a normal scale.
[3] For an in-depth treatment of the entire momentum crash phenomenon and how it can be handled see Momentum Crashes. This entire blogpost draws heavily on the insights from this paper.
[4] This is a minor simplification, for the complete definition see Ken French Data Library Momentum Factor.
 

IMPORTANT NOTES

This material is for informational purposes and is intended to be used for educational and illustrative purposes only. It is not designed to cover every aspect of the relevant markets and is not intended to be used as a general guide to investing or as a source of any specific investment recommendation. It is not intended as an offer or solicitation for the purchase or sale of any financial instrument, investment product or service. This material does not constitute investment advice, nor is it a substitute for such professional advice or services, nor should it be used as a basis for any decision or action that may affect your business. Innealta does not provide tax, legal, insurance, or accounting advice. Before making any decision or taking any action that may affect your business, you should consult a qualified professional adviser. In preparing this material we have relied upon data supplied to us by third parties. The information has been compiled from sources believed to be reliable, but no representation or warranty, express or implied, is made by Innealta Capital, LLC as to its accuracy, completeness or correctness. Material contains opinions, estimates and projections of Innealta and are subject to change without notice. Any projections, forecasts, estimates, and forward-looking statements contained herein are necessarily speculative in nature and are based upon certain assumptions. It can be expected that some or all of such assumptions will not materialize or will vary significantly from actual results. Accordingly, any projections are only estimates and actual results will differ and may vary substantially from the projections or estimates shown.

Factor premiums are measured as follows:

Market: Market return over the risk-free rate. Size: Smaller market capitalization companies over larger market capitalization companies. Value: Lower relative price companies over higher relative price companies. Profitability: More profitable companies over less profitable companies. Investment: Conservative company investment over aggressive company investment. Momentum: Higher prior return companies over lower prior return companies. For more details on the factor indices and premiums represented on this brochure, please refer to the Kenneth R. French Data Library.

3429-INN-12/19/2023