JMOM - QMOM: Second-Best In The Momentum Crash Of January 2023

Summary

• January 2023 was brutal for momentum because many of last year's losers suddenly outperformed - a momentum crash par excellence.
• The Dow Jones US Market Neutral Momentum Index, a simple and transparent implementation of the long-short momentum factor, lost 19% YTD (as of February 3, 2023).
• Most of the losses came from the short-side. The Dow Jones US Low Momentum Index, the portfolio of past losers, returned 25.68% YTD.
• The Alpha Architect U.S. Quantitative Momentum ETF returned -0.83% YTD and underperformed the US market by >10% points. Although painful, this is still second-best in a peer-group of other momentum ETFs.
• This speaks for the differentiated momentum process of Alpha Architect. Nobody likes bad months, but momentum crashes are actually a plausible reason why momentum worked historically and probably continues to do so in the future.

In my first article on momentum and the Alpha Architect U.S. Quantitative Momentum ETF ( QMOM ), I have written about the following reason why momentum historically worked and probably continues to do so in the future:

My personal favorite risk-based explanation are momentum crashes ( Daniel & Moskowitz, 2016 ). The idea is straight-forward. If you look at the return-profile of momentum, you will recognize strong excess returns but also brutal crashes (for example after the financial crisis 2008/09). This is not too surprising as momentum is a form of trend-following (you bet that past winners continue to win in the future). Whenever a trend breaks down (past winners suddenly reverse and start losing), the strategy cannot adapt immediately and crashes. Why is this a risk-based explanation for excess returns? Because you capture long-term outperformance as a compensation for exposing yourself to a higher risk for few but brutal crashes.

Source: "QMOM: A Close Look At The Methodology" on Seeking Alpha

Now, when I have written this in late November 2022, I obviously didn't expect that we will experience such a momentum crash that early. But here it is. January 2023 was brutal for momentum because many of last year's losers suddenly outperformed - a momentum crash par excellence.

The January 2023 momentum crash

Before we come to the long-only QMOM, let's take a look at the Dow Jones US Market Neutral Momentum Index to see the full extent of last month's momentum-pain. The index is a transparent and simple way to track the performance of a long-short momentum strategy, the "purest" form of momentum a.k.a. the momentum factor (Kenneth French only uploads his data with some delay).

The index is long (short) the 200 US stocks with the highest (lowest) 12-month price momentum. The result is an equal-weighted portfolio of 400 stocks that is rebalanced quarterly. So overall, this is pretty close to common scientific implementations. Moreover, the quarterly rebalancing makes the index even better than some competing offers like the "slower" MSCI momentum style indices which rebalance only semi-annually. For the sake of analyzing the performance of the momentum factor in January 2023, the index should therefore definitely suffice.

The following chart shows the YTD returns of the market neutral long-short index (purple line) as well as the long and short sides separately. The first observation is that the long-short momentum index lost 19% since the start of the year. In my opinion, a 19% loss within just one month is quite painful and definitely deserves the label "crash" (the magnitude is comparable to the COVID crash of the overall market in March 2020).

The second observation, and this is also a common feature of momentum crashes, is that most of the losses came from the short side. The portfolio of the 200 "low momentum" stocks returned 25.68% YTD and being short something that makes almost 26% is obviously the wrong position.

To see that January 2023 really was the prime example of a momentum crash, take a look at the following quote from a paper by Daniel & Moskowitz (2016) that examines the issue in great detail (the paper is even called "Momentum crashes"...):

When poor market conditions ameliorate and the market starts to rebound , the losers experience strong gains , resulting in a momentum crash as momentum strategies short these assets. We find that, in bear market states , and in particular when market volatility is high, the down-market betas of the past losers are low, but the up-market betas are very large. This optionality does not appear to generally be reflected in the prices of the past losers. Consequently, the expected returns of the past losers are very high, and the momentum effect is reversed during these times.

Source: Daniel & Moskowitz (2016, p.242), "Momentum crashes", annotations by the author.

With the benefit of hindsight, all of these points fit to the last month. Markets started to rebound (the Russell 3000 returned 9.81% YTD after a 19% loss in 2022), losers experienced strong gains (the "low momentum" portfolio returned 25.68%), at least some periods within 2022 certainly qualify as "bear market states", and ultimately, momentum reversed/crashed in January 2023.

Now, what should we do with this information? I understand that it is somewhat unsatisfying to precisely describe the momentum crash after the fact, especially since Daniel & Moskowitz argue in their paper that...

[...] these crash periods are predictable. We use bear market indicators and ex-ante volatility estimates to forecast the conditional mean and variance of momentum strategies. Armed with these estimates, we create a simple dynamically weighted version of the momentum portfolio that approximately doubles the Sharpe ratio of the static momentum strategy and is not spanned by constant volatility momentum strategies or other factors, and we do so consistently in every market, asset class, and time period we study.

Source: Daniel & Moskowitz (2016, p.242), "Momentum crashes".

I think there are two important implications. First, as I repeated in the beginning, momentum crashes are to some extent part of the deal. Momentum is not an arbitrage strategy and the existence of such crashes could even justify the long-term premium. This is nothing different than for the overall stock market. The US market also crashed occasionally, but it also delivered an attractive equity risk premium as compensation for living through such periods.

Second, knowing that such momentum crashes occasionally happen and where they come from is helpful to stick with momentum over the long-term. Daniel & Moskowitz even show empirically that such crashes are predictable and manageable. Against this background, it may be worth the effort to look for differentiated momentum managers whom's processes attempt to mitigate the problem of momentum crashes at least to some extent. This brings us back to QMOM...

QMOM and other momentum ETFs during the recent momentum crash

The following chart shows the YTD performance of QMOM, my admittedly somewhat arbitrary momentum peer-group, and the Russell 3000 as a benchmark for the US market. First observation: all long-only ETFs underperformed the aggregate market which is consistent with the momentum crash. If the long-short momentum factor loses money in absolute terms, a long-only momentum strategy should lose against its benchmark. In the long-only world, "crash" therefore means massive underperformance with respect to the benchmark (which may be somewhat less painful but still not pleasant).

Second observation: except for the J.P. Morgan US Momentum Factor ETF ( JMOM ), all ETFs posted negative returns in a very strong market (Russell 3000 +9.8%). So they not only underperformed but also lost money when the market was heavily positive. In the long-only world, this is probably the worst possible thing you can have and the label "crash" seems again justified.

To be honest, I don't fully understand how JMOM achieved this performance but they definitely deserve credit for mitigating this momentum crash. I skimmed the underlying index methodology for phrases regarding momentum crashes but couldn't find something explicit. However, the fund employs various layers of risk management in all aspects of the investment process. My best guess is that it was the sum of those actions that helped in January. Note, however, that the fund still underperformed the market by about 3%-points. So although being much better than the pack, it couldn't fully detach itself from the momentum crash.

Turning to the focus of this article, QMOM returned -0.83% YTD. Although this is much better than the well-known iShares MSCI USA Momentum Factor ETF ( MTUM ) and Invesco S&P 500 Momentum ETF ( SPMO ), the result is still only second-best and considerably behind the +6.85% of JMOM. Compared to the 9.8% of the overall market it is a disaster, and it is also trailing the 4.28% return of the Dow Jones High Momentum Index from the very first chart. So what happened?

To the best of my knowledge, Alpha Architect has not yet publicly commented the performance of QMOM on its website or elsewhere. So all I am doing here is to give you my personal view on the issues and make some (hopefully) educated guesses (I welcome other views and challenging arguments in the comments).

The first thing is diversification and "activeness". QMOM is a much more concentrated bet on momentum than the other ETFs. For example, QMOM holds 50 stocks while JMOM currently stands at 285. This automatically leads to a higher tracking error and active share which by definition comes with higher risk of deviating from the benchmark for prolonged periods.

To treat QMOM fairly, however, note that this arguments also applies to the even worse performing MTUM and SPMO. Both ETFs are more diversified, but still lost more than QMOM. In my view, this speaks for the differentiated momentum process of QMOM which also considers the quality of momentum (see my first article for details). If you just invest in 50 stocks and still perform better than more diversified peers, I think you deserve credit for picking the better ones.

Another issue that potentially explains the lagging performance is QMOM's "outlier adjustment". As a reminder, before they apply any momentum ranking, Alpha Architect removes (among other things) the 10% of stocks with the highest betas. As I mentioned in the last article, this is not necessarily a problem because high-beta stocks historically underperformed low-beta stocks. But if you apply the beta filter before the momentum ranking, you don't even have the chance to catch high-beta stocks with good momentum. Given that high-beta stocks often outperform during momentum crashes this could be a problem and may explains some of the disappointing performance in January 2023 and during momentum crashes in general.

Conclusion

January 2023 wasn't fun for momentum investors and I am feeling the pain myself as I also have some fraction of my equities to momentum (although in a combination with value that makes it less painful overall). However, I believe it is still important to understand that momentum crashes are (at least to some extent) a feature not a bug. Even if you successfully mitigate them, you will probably not change the overall return profile of a momentum strategy. In fact, if you believe in momentum you should be actually happy about the crash. It shows us a plausible reason why momentum profits exist in the first place, and all else equal, it is better to buy a long-term attractive strategy after it crashed than before.

Turning to QMOM, I still believe the ETF offers outstanding momentum exposure for investors who can live with more concentration, higher tracking error, and some periods of underperformance. Next to the outstandingly robust performance of JMOM, -0.8% YTD of course looks bad. Note however, that despite a much more concentrated portfolio, QMOM was still considerably better than the -2.98% and -3.2% of SPMO and MTUM, respectively.

In my opinion, this suggests that the differentiated momentum process of Alpha Architect still delivered on its promise. Living through a month with >10% underperformance to the overall markets is of course painful, but seems to be the price for earning a long-term premium. Not all investors are ready to do that and this is one reason why momentum probably continues to work in the future. Nobody said it is going to be easy...

For further details see: