Before delving into the different ESG integration approaches, let’s look at the construction of traditional factor indices – to do so I focus on the quality and value factors.

The starting point is selection of the appropriate metrics capturing the factor:

  • Value criteria: Price-to-Book Value, Price-to-Earnings, Price-to-Sales, Price-to-Cash Earnings
  • Quality criteria: Return-on-Equity, Debt-to-Equity, Earnings-Variability

Using these metrics, we can assess the factor exposure of each company in the investment universe (e.g. MSCI USA index). To ensure comparability, it is typical to standardize these metrics by creating z-scores. These can be further aggregated into composite z-scores corresponding to the specific factor (value, quality), which can be interpreted as factor loadings. The higher the value of the z-score, the higher the factor exposure.

In an index-based approach, a factor portfolio can be constructed by selecting only those companies from the parent universe which have relatively high factor loadings (e.g. including 25-30% by market capitalization). Furthermore, the weighting of portfolio constituents may be linked to market capitalization as well as factor exposures (z-scores) to further tilt towards the targeted factor.

Constructing a standard factor index is about defining the relevant factor metrics, which are then used for stock selection and tilting weights to increase factor exposure.

ESG integration

In our view, a comprehensive ESG overlay should include several components such as business activity exclusions, promotion of higher ESG rated companies, and a carbon footprint reduction. In particular, sustainability ambitions can be defined as follows (based on MSCI data).

  1. Exclusions: UNGC violations, tobacco, weapons (controversial, nuclear, military, civilian), thermal coal mining, unconventional oil and gas extraction, thermal coal power generation
  2. Minimum ESG standards: minimum MSCI rating of BB and minimum controversy score of 1
  3. ESG score improvement target of +20%
  4. Carbon footprint reduction by 30%, potential emissions (fossil fuel reserves) reduction by 30%

Applying business activity exclusions1 and minimum ESG standards2 is straightforward and typically reduces the eligible universe by single digit percentages as measured by market capitalization.

Interplay between factor exposure and ESG considerations

The key challenge lies in balancing at the portfolio level between:

  1. factor exposure
  2. an improvement in ESG score3
  3. a reduction in carbon footprint4.

In essence, it is a multidimensional optimization problem. To help unpick things, I examine the relationship between the value factor exposure and ESG score for the MSCI USA universe (Figure 1), using MSCI ESG data. The most desirable companies are characterized by above average ESG scores and high value factor loadings (z-scores).

Figure 1. MSCI USA universe: ESG scores vs. value factor z-scores

The bubble area represents the weights of companies in MSCI USA Index. The horizontal red line represents the average (industry-adjusted) ESG score of 6.3, while the vertical red line represents the average value z-score by construction equal to 0.0.

All companies in the MSCI USA Index are shown as bubbles. Companies placed high in the chart have better ESG scores and vice versa – companies on the right side of the chart have higher value characteristics than those on the left.

Source: MSCI, ÃÛ¶¹ÊÓƵ Asset Management. Data as of December 2022 index rebalancing. For illustrative purposes only.

Each bubble represents the weight of a company in the MSCI USA Index. The horizontal red line represents the average (industry-adjusted) ESG score, while the vertical red line represents the average value z-score. The higher a company is placed on the chart, the better are its ESG scores – and companies on the right side of the chart have higher value characteristics than those on the left. With an optimization process it’s possible to identify companies that meet the required ESG and value criteria.

Two methods to create ESG factor portfolios

MSCI USA Prime Value ESG Low Carbon Select index

The MSCI USA Prime Value ESG LCS Index portfolio is indicated in Figure 1 with dark grey color; it includes 104 companies with an aggregate market cap weight of 17.0% versus its parent – the MSCI USA Index. We can see the optimizer makes trade-offs between factor exposures and ESG scores: at times a company with a lower factor loading enters the portfolio because it has a high ESG score and vice-versa. Clearly, companies with below average ESG scores and below average value factor loadings are excluded (bottom left quadrant in Figure 1).

The portfolio fulfils all objectives; MSCI USA Prime Value Index has a value exposure that is significantly higher compared to MSCI USA (z-score of 0.34 vs -0.13 , while it meets the ESG target improvement (+20%) and exceeds the carbon reduction target (-49%).

Table 1. Comparison of selected metrics (averages weighted by market capitalization)

Title

Title

Value z-score

Value z-score

ESG Score (Industry adj.)

ESG Score (Industry adj.)

Carbon intensity

Carbon intensity

Title

MSCI USA Prime Value ESG LCS

Value z-score

0.34

ESG Score (Industry adj.)

7.91

Carbon intensity

73.71

Title

MSCI USA

Value z-score

-0.13

ESG Score (Industry adj.)

6.59

Carbon intensity

144.41

ESG metrics based on MSCI data. Carbon intensity is measures by tCO2/$M sales.

Source: MSCI, ÃÛ¶¹ÊÓƵ Asset Management. Data as of December 2022 index rebalancing. For illustrative purposes only.

Performance

The performance of the mentioned factor ESG indices has been similar to standard factor indices (Figure 2). Since 2013, the Value ESG Index underperformed its standard Value equivalent by 47 bps per annum, while the Quality ESG Index outperformed its standard factor equivalent by 72 bps per annum . Also, correlations in daily excess returns between the two Value and Quality indices has been at 0.86 and 0.79, respectively, which further demonstrates similarity.

Figure 2. Historical performance of selected factor indices relative to MSCI USA

Each of the four lines on the chart represents the performance of an index over the past ten years. The ESG versions of the MSCI Quality and Value indices perform in line with their parent indexes.

Source: MSCI, ÃÛ¶¹ÊÓƵ Asset Management. Data as of 31 July 2023.
Index level data for factor indices contains live and back-tested data. Past performance, whether simulated or actual, is not a reliable indicator of future results. For illustrative purposes only.

This chart shows the performance of four indexes over a period of 10 years until end of July 2023. These are the MSCI USA Prime Value, the MSCI USA Prime Value ESG Low Carbon Select, the MSCI USA Quality and the MSCI USA Quality ESG Low Carbon Select. A simulation with historical data shows that the performance of the mentioned factor ESG indices has been similar to standard factor indices.

Optimizing appears optimal

Ultimately, historical simulations demonstrate that factor ESG solutions can be considered a good alternative to their traditional counterparts. And integrating ESG into factor indices can be achieved with an optimized approach that can efficiently balance between achieving factor exposures and sustainability objectives.

1 For a more detailed discussion on this topic, we refer to MSCI publication ‘The MSCI Factor ESG Target Indexes’ (September 2017).
2This portfolio average value z-score is weighted by market capitalization.
3 The value is -0.13 because it is the market cap-weighted average. For arithmetic mean, the value is 0.0.
4 The factor ESG indices are primarily based on backtested data.
Ìý

Related insights

Contact us

Make an inquiry

Fill in an inquiry form and leave your details – we’ll be back in touch.

Introducing our leadership team

Meet the members of the team responsible for ÃÛ¶¹ÊÓƵ Asset Management’s strategic direction.

Meet the team

Find our offices

We’re closer than you think, find out here.

ÃÛ¶¹ÊÓƵ locations