- Factors help investors understand where performance of their portfolios comes from. We examined whether ESG can help explain performance, using a factor model that incorporates ESG as a factor.
- We found that, over the last two years, the explanatory power for ESG has increased — perhaps due to more investment by companies in sustainability-related measures, and growing awareness and greater regulatory scrutiny of ESG and sustainability.
- We measured ESG exposure across a sample of global equity funds and found that exposure to ESG was not dependent on the funds’ use of MSCI ESG Ratings.
Environmental, social and governance (ESG) integration, or the inclusion of ESG into the investment process, has evolved over time. For example, some investment managers have pursued strategies that exclude stocks with poor ESG characteristics while others have focused on selecting stocks with superior ESG characteristics. While there is a robust set of ESG data and metrics to analyze portfolios, many investors want to know how much ESG contributed to overall portfolio risk and return. They want to know how ESG performed historically and whether ESG was influenced by other factors or helped explain performance as a factor in its own right.
While ESG has, at times, been described as a combination of quality, large size or low volatility exposures, we did not see this in our analysis using the MSCI Global Equity Factor Model + ESG (GEMLTESG).1 The figure below shows the ESG pure factor return for the sample period January 2007 to December 2020.
ESG Pure Factor Return
Looking beyond returns, the exhibit below presents a statistical summary for the period January 2007 to December 2018 and, separately, for the last two years. In the latter period, ESG became more important in explaining performance. While this recent increase in ESG explanatory power is beyond the scope of this blog, prior MSCI research has noted that there has been: 1) increased investment by companies in sustainability-related technology, 2) growing awareness of ESG and sustainability by investors and 3) greater regulatory scrutiny around ESG and sustainability.2
We estimated the ESG factor return each month and the exhibit summarizes t-statistics from these regressions. Cross-validated R2 shows the average gain in explanatory power provided by the ESG factor compared to all factors in the GEMLT model. The t-stat test helps determine the statistical significance of ESG.3
ESG’s Explanatory Power Has Increased
| January 2007 – December 2018 | |||
| |t|>2, % | Mean(|t|) | Annualized Volatility, % | Cross-Validated R2 Gain (bps) |
| 32.6 | 1.64 | 0.77 | 0.40 |
| December 2018 – December 2020 | |||
| |t|>2, % | Mean(|t|) | Annualized Volatility, % | Cross-Validated R2 Gain (bps) |
| 37.5 | 1.86 | 0.76 | 1.11 |
Properly Attributing ESG
To understand the impact that ESG had on risk and return for ESG-oriented strategies, we ran performance attributions using both the GEMLT and the GEMLTESG risk model using the MSCI ACWI ESG Focus Index as a sample ESG portfolio from November 2012 to December 2020.4 We use the ESG Focus Index in our analysis because its methodology is solely focused on maximizing the ESG score subject to a tracking-error constraint, thus minimizing all other effects (industries, countries, currencies and style factors).
The ESG Focus Index outperformed the MSCI ACWI Index by 93 basis points (bps) per year. The stock-specific return component (not attributable to other factors), however, was 75 bps using GEMLT and dropped to 38 bps using the GEMLTESG risk model — indicating that 37 bps previously attributed to stock-specific effects was identified as performance stemming from ESG.
Performance and Performance Attribution for the MSCI ACWI ESG Focus Index
Performance Attribution using Different Risk Models
| GEMLT | GEMLTESG | |
| Currencies | -0.02% | -0.02% |
| Stock-specific | 0.75% | 0.38% |
| Countries | 0.01% | 0.01% |
| Industries | -0.02% | -0.02% |
| Quality | 0.08% | 0.08% |
| Volatility | 0.10% | 0.10% |
| Size | 0.01% | 0.01% |
| Other Styles | 0.02% | 0.02% |
| ESG | 0.00% | 0.37% |
| Active Return: | 0.93% | |
Identifying ESG
We also tested the GEMLTESG model with ESG funds that do not employ MSCI ESG Ratings. As with other factors, MSCI’s ESG factor should measure exposure even when ESG is integrated using sources other than MSCI ESG ratings. Using MSCI ESG Manager,5 we selected global equity mutual funds with assets under management greater than USD 25 million and funds with ESG in the investment process. ESG fund criteria is described as those funds having broad, thematic or exclusionary types of ESG intent or any combination of the three. Our screen resulted in a total of 304 funds: 164 subscribed to our ESG ratings and 140 did not. We found that the two groups of funds showed similar active ESG exposures, with the median approximately 0.30 as can be seen in the exhibit below. We consider factor exposures of +/- 0.20 standard deviations as meaningful since they are likely to be intentional.
Active Exposures of ESG Funds
Data as of Sept. 30, 2020. In each box, the central mark indicates the median, and the bottom and top edges of the box indicate the 25th and 75th percentiles, respectively. The whiskers extend to the most extreme data points not considered outliers, and the outliers are plotted individually using the '+' symbol.
The Relevance of ESG
We used the GEMLTESG risk model to quantify risk and return directly attributable to ESG, and found that ESG became more relevant as an explanatory variable over the last two years, and accounted for a larger share of active return than shown by previous measures. Finally, our exposure analysis of global ESG funds shows the universality of MSCI’s ESG factor as it is able to measure ESG exposure of funds that do not subscribe to MSCI ESG ratings.
1In April 2020, MSCI launched a version of the MSCI Global Equity Factor Model (GEMLT) that added ESG as the 17th style factor. The GEMLTESG risk model was constructed using the industry-adjusted scores from MSCI ESG Ratings as the raw ESG exposures.
2Giese, G., Lee, L.-E., Melas, D., Nagy, Z., and Nishikawa, L. 2019. “Foundations of ESG Investing: How ESG Affects Equity Valuation, Risk and Performance.” Journal of Portfolio Management.
3In statistical analysis, it is common to use an absolute t-stat value of two or greater as a threshold to establish a level of significance.
4The ACWI ESG Focus Index history begins in November 2012.
5ESG Manager is an integrated tool that provides ESG research and analytics at the stock and fund level.
Further Reading
What ESG Ratings Tell Us About Corporate Bonds
Quantifying ESG fund performance
Factors and ESG: the truth behind three myths
Is ESG Investing a Price Bubble? Probably Not.
Creating a common language for factor investing
Introducing MSCI FaCS: A New Factor Classification Standard for Equity Portfolios