What is Multi-Factor and how does it work in long-only equities?
In recent years, multi-factor equity strategies have become more commonplace. Yet we are no closer to agreeing how best to exploit the underlying factors. Many have argued a case for why certain approaches should be favoured by investors and why others should be dismissed.
I’m going to try and unravel the myriad of options and arguments.
The idea of looking at investment through the lens of factors is nothing new. Focusing on stocks with certain characteristics to exploit prevalent behavioural biases has been widespread in financial markets throughout living memory.
Recently, however, evidence has mounted that a lot of active management “skill” is explained by exposure to these factors. Investors and asset owners have looked at various ways of using this to help them manage portfolios. A vast array of different factors have been “discovered” (with a significant amount being of questionable efficacy). And with this, assets have flowed into equity products that define themselves as “multi-factor”.
Broadly, these are any investment strategies which aim to create a portfolio with exposure to more than one specific factor.
These strategies can take many different forms. These include:
– relatively simple smart beta
– more complex quantitative strategies
– the traditional discretionary approach
Even within the separate categories returns can differ widely. This is due to which factors are included, how factors are defined, how portfolios are constructed and how decisions are made.
The below chart shows that even if you narrowed down your focus to just quantitative multi-factor equity strategies – your manager selection would still have played a huge part in your ultimate returns!
With the myriad of options, investors are asking – is there an ideal approach that gives equity investors multi-factor exposure?
1) Smart Beta
There is no strict definition as to what makes a multi-factor smart beta but broadly they tend to:
- Be implemented systematically
- Follow a relatively simple, rules based approach
- Be applied broadly across different sectors and geographies with little adjustment
- Use the different factors when organising the weights of a given index
The simplicity means they are easy to understand and monitor. Investors get exposure to these factors at a low cost due to the scalability and limited oversight required. They are also highly customisable and can be effectively used as “completion portfolios” to gain exposure to factors the overall portfolio may be lacking.
A common criticism is that these processes lack oversight. That the rules are strictly defined at the outset and there are no discretionary decisions taken from then on. However, if one is more sceptical on the value of traditional active decision making and wants to remove some of the potential behavioural biases, then removing this element from the process could be viewed positively.
This does not mean, however, that investors are alleviated from making an “active decision”. These are not “passive” investments in the sense a market cap benchmark is. Therefore, investors need to be sure what they’ve bought is right for them. You need to ensure products are thoughtfully constructed, well researched and have rigorous evidential backing. Thorough due diligence is key – otherwise returns may not fall in line with what investors expect.
2) Quant Multi-Factor
Complex, quantitative multi-factor approaches encompass all manner of strategies but typically will:
- Be higher turnover
- Use more data points
- Have more complex ways of creating factors
- Factor selection is more precise – customising metrics for a given sector or geography
- Tend to see more ongoing innovation/research
- Use sophisticated portfolio optimisation techniques
- Sometimes be dynamic in terms of changing the factor exposure
Like smart beta, investors are drawn to these. They believe the systematic implementation helps mitigate behavioural biases – thus giving comfort on the consistency of approach (rendering past due diligence more “useful”). They also tend to have more oversight than smart beta. This is because there is usually a dedicated team monitoring the model’s efficacy.
There is also potential for a better risk/return profile due to the use of differentiated factors, more thoughtful portfolio construction and potential “research skill”. Fees also tend to only be marginally higher than smart beta, however, the increased complexity can use up an investor’s governance budget.
An issue often cited with the above approaches is that they are backward looking. If you subscribe to the existence of factor premiums, you would have to agree that there is useful information in history. However, for some investors looking only in the rear-view mirror is uncomfortable. It is very difficult for backward looking systematic approaches to identify regime changes in markets.
So, here is where one could look to more traditional discretionary approaches.
3) Discretionary Multi-Factor
These discretionary approaches are essentially traditional stock pickers. Here, a fund manager will use positive factor exposure as the basis upon which they conduct fundamental analysis. For instance, this could involve using a quantitative screen that looks for stocks with exposures to these factors. It would then conduct investment analysis only on companies that meet this criteria.
The general benefit of this approach would be two things:
– the oversight from a portfolio manager gives investors comfort
– it incorporates a more forward-looking aspect to the process.
They also tend to have lower turnover and be more concentrated. This reduces trading costs and gives a higher tracking and therefore, assuming a similar information ratio, more alpha potential.
Environment, Social and Governance (ESG) factors are also better incorporated in discretionary approaches. Quantitative strategies are making progress here, but given ESG data is still in its nascent stages and of questionable reliability, ESG risks/opportunities are generally better assessed in a more traditional way.
The issues would be that these are less scalable and as such will typically come at the cost of higher fees. There will also be an increased chance of “key person risk”. This is because decisions normally rest in the hands of one portfolio manager or a small team, rather than a systematic, repeatable quantitative model. The forward-looking aspect is also only useful if the manager’s analysis is good and it reintroduces the human behavioural biases that systematic strategies look to nullify. Therefore, as with all the multi-factor approaches, carrying out robust due diligence, ensuring your beliefs are aligned with this type of strategy and getting comfort in the manager’s skill is crucial.
So, is there a silver bullet?
Overall, there are numerous avenues an investor can go down if they believe exposure to factors can increase the return potential of their equity portfolio. There are clear benefits and drawbacks no matter which avenue is chosen, and advocates for each, often with financial incentives to advocate, will always argue theirs is the correct way to go. In the ever-changing environment of financial markets, there is no perfect solution.
The only silver bullet is that investor’s must be sure that their decision aligns with their beliefs and that they fully understand what it is they’re buying by conducting thorough due diligence. They must also make sure that these decisions are aligned with their governance and fee budgets. If investors buy into a strategy without doing this, instead of a silver bullet, they’re more likely to find they’ve just shot themselves in the foot.