What the F@#%ctor?

What the Factor 361 Capital Blog2018 was a strange year, and not just because Baby Shark was playing on constant loop, or because everyone became mildly obsessed with a Netflix movie heavy on blindfolds during the holidays. As far as markets go, it was one of the only two years since 2009 that a multi-factor portfolio delivered a negative return. As you can see from the chart below, global multi-factor portfolios have been positive 8 out of 10 years, including 2011 and 2015 when the MSCI World was negative. 2018 was the only year in the last decade that multi factor portfolios and stocks broadly were down at the same time.

2009 2010 2011 2012 2013 2014 2015 2016 2017 2018
Value 28.2% (2.4%) (11.0%) (0.9%) (4.9%) 4.3% (10.9%) 13.9% (9.2%) (13.1%)
Size 16.1% 6.3% (3.6%) 2.9% 2.7% (3.2%) 4.7% 8.3% (0.5%) (3.4%)
Quality 6.0% 7.4% 10.6% 0.0% 9.7% 0.5% 12.4% 0.2% 15.5% 4.1%
Low Vol (15.2%) 6.5% 24.4% 17.5% 20.4% 29.4% 18.7% 5.2% 7.3% 11.7%
Momentum (41.4%) 3.4% 14.9% 11.0% 14.1% 0.1% 26.9% (10.2%) 10.5% (5.0%)
Multi Factor (0.8%) 4.2% 7.0% 5.6% 8.3% 5.9% 10.2% 3.1% 4.5% (1.1%)
MSCI World Index 29.99% 11.76% (5.54%) 15.83% 26.68% 4.94% (0.87%) 7.51% 22.40% (8.71%)

Source: FactorResearch

Much has been written about factor investing and how certain overweights (e.g., value, momentum, quality, volatility, size, etc.) have the potential to lead to outperformance over time. In any given year it can be argued that it wasn’t necessarily manager skill that led to outperformance, but unintended tilts towards certain factors that were being rewarded.

Another interesting feature of these factors is that they tend to have low correlation with one another over time. This means for instance, that if value is underperforming, perhaps quality and low volatility are doing well, and vice versa the next year. It is why many strategies employ a multi-factor approach—like traditional diversification for asset classes—but instead use individual factors.

So, what happened last year?

Interestingly, small caps were negative which was something we did not expect given all the rhetoric around trade policies. We thought that with a more protectionist stance, smaller companies (which aren’t as exposed to global markets) would have done better, but that didn’t happen. Part of the underperformance in the size premium could have also come from the fact that a handful of mega cap names drove most of the market last year.

Value continued to lag, yet again. In fact, this factor significantly underperformed long-only global equities. As you can see from the chart above, it has only been positive twice in the last 10 years and this current bout of underperformance is only the third time in history that it has happened, and it is by far the largest. (A recent piece by Schroders talks about this in more detail if you are interested in diving in.)

Momentum also had a tough year which was amplified in October during the equity sell off early in the month. A unique relationship between value and momentum over time is that they are negatively correlated, which means they often move in opposite directions (refer again to the chart to see how their performance versus one another varies over time). To see them both negative at the same time by this wide of a degree is rare, and perhaps why a multi-factor approach did not yield positive results in 2018.

What does it all mean?

I’d love to have the ability to see into the future (mainly to pick winning lotto numbers and see if the Broncos will ever win another Super Bowl) to know if, when and how things will be more favorable to a factor-based approach. Looking at the volumes of research on the topic, it could possibly work out more favorably in 2019 given a mean reversion to historical averages. However, there is nothing indicating it has to. The reality is we could see a repeat of 2018 this year and it would still be within the realm of expectations.

The frustrating thing about averages and historical observations is that they do not come in a linear and consistent pattern. Returns tend to be lumpy and the outperformance ebbs and flows. Unfortunately, the typical response after a bout of underperformance is to declare that it “doesn’t work anymore” and move on. That would be a completely valid emotional response to the last three years given the returns for a multi-factor portfolio vs. long only stocks.

It may create angst to see something you’ve vetted and have confidence in suddenly lag— especially when you expected the very opposite. However, it is only by sticking with an approach (whether asset allocation, manager selection, etc.) that you’ll be able to harvest these observed anomalies over time. 

We are big proponents of academic-based approaches like those employed in multi-factor portfolios. So, as tough as it is in a moment like this, we prefer to rely on the volumes of research spanning decades (which include periods of underperformance) and continue employing an approach that we believe will add value over the long term.

Read more in The Role of Factors in Portfolios >