David Barron, head of index equity and smart beta at Legal & General Investment Management
Can environmental, social and governance concerns (ESG) fit within a factor-based portfolio? In this article, we tackle ways to approach this developing field.
As growing numbers of investors seek to adopt the principles of responsible investing, the world of factor-based strategies is far from an exception: 55% of European asset owners surveyed in 2018 stated they were looking to integrate ESG considerations within their factor portfolios.
But no clear picture has emerged so far about how best to achieve this goal, despite extensive research in recent years. Should the ‘responsible’ portion of portfolios be managed separately or is it more effective to integrate ESG considerations at the factor level?
Integrating ESG considerations in portfolios
Option 1: Bottom up with ESG as another ‘factor’
The bottom-up approach creates portfolios with strong factor exposures by assigning a multi-factor score to each security, which is then reconciled with the investment’s ESG score. This ensures that there is less dilution of target exposures, which is a risk with a top-down approach. It generally relies on a market capitalisation weights or tracking error constraints to minimise tracking error and concentration risk. In a ‘tilting’ approach, every security is assessed on ESG and multiple factor scores. A relatively simple ‘tilt’ is executed by multiplying the individual security scores by the market capitalisation weight. The result is an increased allocation to companies that score well on ESG and chosen factor scores and a reduced weight for that do not.
A second approach seeks the optimal way to combine ESG scores with factors. Usually, providers aim to achieve a target objective using an optimisation model, creating a composite score capturing desired portfolio attributes. The company with the highest aggregate score receives the largest weight, subject to constraints.
A drawback to these approaches is the difficulty of deriving attribution by decomposing security weights based on individual factor exposures. Also, as more factors are added, the universe narrows, increasing concentration. It becomes difficult to find stocks that outperform on all factors as well as the chosen ESG metrics. Therefore, time should be spent on determining the relative weight, or importance, each ESG factor has
Option 2: Exclude, then run the factor portfolio as normal
This option can be applied to a bottom-up or top-down portfolio construction process. The difference here from the methods above is that it does not allocate additional weight to companies with strong ESG scores. Instead, the laggards are removed, set at a pre-determined threshold. An important benefit of this approach is its simplicity. A fairly easy comparison can be made between pure factor and ESGexcluded portfolios. A concern with this approach is that it is difficult to engage with excluded companies to encourage them to improve their behaviour.
However, the tilting approach also eliminates significant numbers of poorly rated companies, owing to their attempts to identify and allocate to factor ESG champions. The key problem with this option is setting the exclusion level. Should it be a fixed quantity or a percentage, by region or sector? This depends on the portfolio’s objectives.
Option 3: Top down: tilt at the end or combining portfolios
This method keeps ESG considerations and factor scores separate until the final stage of the portfolio construction process. A factor portfolio is created, then the factor weighting of each stock is adjusted according to its ESG score.
The second option is to create an independent ESG-integrated portfolio and combining this with the factor-level portfolio afterwards. Attribution for the pure factor portfolio is possible in both, but extremely easy in the latter option.
This approach has the additional benefit of being able to control the exposure to factors and ESG considerations independently of each other simply using an allocation of capital. The downside of this approach is that it likely has the highest factor cancellation effect, as the two portfolio segments are managed separately. This will result in inconsistent exposures over time as the factor portfolio rebalances will not consider whether or not they are allocating to strong or weak ESG-scoring companies – and vice versa.
A fair-weather friend?
Investors are faced with a number of choices: if embracing the drawbacks of competing investments is worth the simplicity (when managing a separate ESG-integrated portfolio); if excluding poor performers (and enjoying the purity of a factor construction) compensates the work required to define the ‘laggards’, and if the bottom-up approach is worth the potential concentration risk.
If current trends are any indication, we can expect to see a lot of all three.