Considerations before investing in multifactor funds
As we saw in my last post, multifactor investing has provided investors with a way to harness the benefits of factor investing in a more diversified way, with lower risk than investing in any single factor.
The risk of any one factor underperforming is reduced by using a multifactor approach, but there are several other drawbacks to multifactor investing that still need to be considered. The list below isn’t meant to be a list of reasons not to invest in factors, but more a list of things to think about before making the decision to allocate to a multifactor strategy. These points are covered in more detail in this post, but I’ve provided summaries of potential problems with factor/smart beta investing below:
- Defining a factor is hard. One person’s idea of the value factor may be very different to another person’s. There are a huge number of “value” indices, all of which are constructed slightly differently, using different characteristics to define value, measuring them differently, and weighting them differently. As an example, the quilt chart below shows the performance of 8 different “value” strategies. All are valid ways to measure value, but all have performed surprisingly differently over the same 5 year periods.
Source: Research Affiliates
- Factors can change. The traditional, academic definition of value was the P/B ratio, but many people now think that P/B is an outdated measure of value. Book value excludes brands, intellectual property, customer loyalty, data, code, and business models which is what most of today’s service economy is built on. Factors might also be refined, such as the traditional size factor being discovered to not be as robust as originally thought, before being revived by AQR when applying a quality screen to the small-cap stocks. Factors changing may cause problems for funds who might be slow in updating their own factor definitions in light of new evidence.
- Factors decay over time. There’s considerable evidence showing that factor returns reduce after the factor becomes widely known. The more people that know about a factor, the more people try to exploit it, and the lower the future returns become. The chart below shows how that the actual post-publication returns from a portfolio of 46 factors (black line), haven’t been as good as was predicted by how the factors performed pre-publication (red line). After the factors became known, the actual returns were much lower than the research would’ve predicted.
Source: Research Affiliates
- It’s impossible to know when a factor breaks down. Nobody knows exactly why factor premia exist – most people believe it’s a combination of risk-based and behavioural-based explanations, but without knowing exactly why factors exist, we can’t know exactly when they cease to exist. Secondly, thanks to the noisiness of markets, it would be very difficult to tell for certain when a factor broke down as it would still occasionally behave as if it worked. To make matters worse, it’s not in the interests of the majority of financial institutions to admit that a factor has stopped working – it’s not be in their best interests to disprove the efficacy of a strategy that their clients are invested in. All of this could lead to investors maintaining positions in factors which no longer work.
- Multifactor exposure can approximate beta. By combining too many factors, or by combining a few watered-down factors, the end result can end up looking an awful lot like beta. The chart below shows the performance of 3 major multifactor indices (MSCI, Goldman Sachs, and JP Morgan) vs the S&P. Over the last 3 years of mostly out-of-sample performance, the performance of the multifactor indices is almost identical to the S&P 500.
Source: Bloomberg
ETF.com research has shown that achieving beta returns has been the best that multifactor investors can have hoped for since multifactor ETFs started being launched. For those multifactor ETFs with performance going back 10 years, the longest-lasting and often the most trusted names in the space, they underperformed SPY by almost 3% over the past 12 months, by more than 6% over the past three years, and over the past 10 years, have underperformed the S&P 500 ETF ‘SPY’ by 5.24% to 8.62%. Newer funds with shorter track records haven’t been beating SPY either.
- Academia is not real life. Academic factor investing is not the same as real life ‘smart beta’ investing. Academia uses long/short factor portfolios, mainly within small caps, with no trading costs – whereas smart beta uses long only factor exposure, usually within large caps, with real-world trading costs. It’s therefore sensible not to expect the same results in real life than in academia. Most factor funds (and multifactor funds) provide watered-down factor exposure for a couple of reasons. 1) the stocks with the highest factor exposure tend to also be the smallest stocks. In order for a strategy to be able to scale to large AUM, it needs to operate in the higher-cap spectrum, which gives lower factor exposure. 2) Pure factors (and multifactor strategies) can have long periods of underperformance. As many investors will abandon a strategy if it underperforms for too long, it’s in the provider’s best interest to water the factor exposure down and reduce the likelihood of redemptions.
- Data mining. Data mining involves finding patterns in data where none exist. For example, the weight of turkeys in the US happens to have a strong correlation with the performance of the MSCI World (R2 of 0.96). Yet nobody would try to predict future stock market returns based on the weight of turkeys. It’s possible that historic factor performance has been a result of thousands of academics poring over the same sets of data and finding patterns where none exist.
- Front-running. For those investing in public, transparent factor index funds where the rules around how stocks are included in the index are free for all to see, there is evidence that this information is being used to profit at the expense of the index investor. Given that the majority of factor exposure is via index tracking products, this is a problem that applies to a lot of factor investors. One study estimates the cost of being front-run for factor index investors to be in the region of 0.165% per year. This problem won’t affect traditional active factor funds whose rules aren’t as transparent, so this raises the question of whether the higher price of the active fund is worth the extra cost and lack of transparency?
- Multifactor funds aren’t as intuitive as single factor funds. It’s easy to understand why the growth factor outperformed in the tech bubble, but it becomes much more difficult to form intuitive explanations for a diversified factor portfolio’s performance. This was recently discussed by AQR, who acknowledge that when multifactor funds start to underperform, they become more difficult to stay invested in because it’s less obvious why they’ve gone down, and so less obvious why they should eventually go back up. Diversification is obviously something we should strive for in portfolios, but with multifactor investing it comes at the cost of being able to easily explain results, and introduces the risk of performance chasing.
Conclusion – multifactor funds
Probably the biggest advantage for using multifactor funds over single factor funds is their higher, more consistent performance, combined with lower risk. They’re a one-stop-shop for diversifying between factors (and potentially geographies too), which provides a more palatable set of returns for most people.
However, portfolio construction still matters. Do you want active or passive exposure? An integrated or mixed fund? What do you think is the best definition of each factor? Do you want seat-of-your-pants academic factor exposure, or a more watered-down, but easier to stick with factor exposure? Are you happy to stick with the fund when it underperforms?
No two multifactor funds are alike, and investors still need to do their homework before investing.
Factors have a huge amount of supporting evidence (see this post) for their outperformance and explanations for their existence, but there are also a number of things to consider before making a bet on factor investing in general.
Overall, multifactor funds look like a good bet for investors interested in trying to generate long term alpha, but investors need to understand what they’re buying. Investors should be fully aware of the drawbacks of a multifactor strategy, and be willing and able to stick with the strategy through the inevitable periods of underperformance.
Multifactor investing is one of the major arguments for using active management over passive. The next article looking at the arguments for active management will focus on the ability of active managers to change their asset allocation based on long term valuations.