Is ESG a systematic risk factor for US equity mutual funds?

ABSTRACT

On the outperformance of responsible investing (RI) which incorporates environmental, social, and governance (ESG) into investment decisions, the empirical evidence to date is inconsistent from the viewpoint of ex-post performance. This paper tries to explain the nature of return differential between RI and conventional investing within the well-known risk-return paradigm. From the viewpoint of ex-ante equity risk premium, the five factor model of Fama and French [2015. “A Five-factor Asset Pricing Model.” Journal of Financial Economics 116: 1–22] combined with a ESG-related factor applies to returns on 1,425 US open-end equity funds for the period from April 2009 to December 2016. Empirical findings include that US open-end equity funds tend to hedge the ESG-related systematic risk, and that the exposure to ESG-related systematic risk is significantly priced in the market. The result implies that RI provides the downside protection against ESG-related systematic risk which is not reduced even through extensive diversification.

KEYWORDS:

• ESG
• CSP
• responsible investing
• downside protection
• systematic risk

JEL:

• C15
• G11
• G12
• G13
• G22
• G33

Acknowledgements

I would like to show my gratitude to the Morningstar Korea for generously sharing data with me during the course of this research. I am immensely grateful to Andy Jung, Chris Park, and Jungtae Chae for their insights, although any errors are my own and should not tarnish the reputations of these esteemed persons. I also thank two anonymous reviewers for their comments on an earlier version of the manuscript. The analysis and conclusions set forth are those of the author and do not indicate concurrence by other members of the research staff or the National Assembly Budget Office.

Disclosure statement

No potential conflict of interest was reported by the author.

ORCID

Ick Jin http://orcid.org/0000-0001-9671-1488

Notes

1. Hoepner and McMillan (2009) defines RI as investment in assets based on screening and selection processes which is developed and practiced on the basis of environmental, social or governance criteria. This paper uses the term RI to indicate investment in assets based on weighting rather than screening and selection.

2. The US SIF Foundation first measured the size of RI in the US market at 1995.

3. Modern portfolio theory (MPT) refers to principles underlying analysis and evaluation of rational portfolio choices based on risk-return trade-offs and efficient diversification. For details, see Bodie, Kane, and Marcus (2014).

4. In Clark, Feiner, and Viehs (2015), the term of ‘aggregate sustainability scores’ refers to the score combining three dimensions of ESG: environmental performance of sustainability (E), social performance of sustainability (S), and governance performance of sustainability (G). They do not discuss explicitly how such scores are conducted.

5. Models in the same category tend to give similar estimates for the ERP. See (Duarte and Rosa 2015, 5).

6. Following the conventional financial theory, we estimate a fund’s systematic risk as its sensitivity to changes in ESG-related factor that is relevant to all funds.

7. A factor portfolio is a well-diversified portfolio constructed to have a beta of 1 on one of the factors and a beta of zero on any other factor. For more detailed discussion, see p.338 of Bodie, Kane, and Marcus (2014).

8. Such an approach requires that all securities in universe available to investors have long-term ESG-scores. Until now, the number of stocks covered through a corporate sustainability assessment for ESG-score is relatively small. For instance, the 11th-consecutive Corporate Sustainability Assessment carried out by RobecoSAM in 2009 covered only about 1,200 companies.

9. Note that components of ESG-weighted portfolio are overlapped with components of unweighted market portfolio. It contrasts to other risk factors (SMB, HML and WML) which are constructed using the difference between returns on two portfolios containing mutually exclusive stocks.

10. Using rolling windows is useful in a couple of ways. Most of all, it allows factor betas to vary. Factor betas for fund of interest are constantly monitored, and changes in factor loadings become readily identifiable.

11. It is the bias in the measurement of active asset management performance arising from the fact that funds that become obsolete are usually the poor-performing funds. Therefore, excluding the obsolete funds from the analysis will tend to overestimate the overall performance of active managers as a group.

12. For instance, the US stock market peaked in October 2007, when the S&P 500 exceeded 1,549 points, but it then entered a pronounced decline, which accelerated markedly in October 2008. By March 2009, the S&P had reached a trough of around 735.

13. Our analysis is based on the return streams of the share class which has the longest performance history. Considering that different share classes typically have different performance history lengths, it would be ideal to calculate asset-weighted average performance by incorporating all the available data across various asset classes. However, Morningstar Direct does not contain the historical market capitalization of the various share classes in a particular fund. Thus, it would be the next best thing to base our analysis on the return streams of the oldest share class.

14. SAP-ESG was launched on April 18, 2016. SAP-ESG is one of the first global ESG index series that serves the growing market of smart beta indices. See (S&P Dow Jones Indices 2016).

15. For more details about SAP-ESG including ESG characteristics, top 10 constituents by index weight, and sector breakdown, see S&P Dow Jones Indices (2017).

16. It began on April 20, 2010, in the Gulf of Mexico on the BP-operated Macondo Prospect, and thus is referred to as the BP oil spill, the Gulf of Mexico oil spill, and the Macondo blowout.

17. A VIF of higher than 10 suggests severe multi-collinearity problems.

18. With a series of extended models, most return anomalies are interpreted as proxies for various forms of risk (Fama and French 1993; Pastor and Stambaugh 2003; Vassalou and Xing 2004). Others attribute the observed effects to market inefficiencies (Lakonishok, Shleifer, and Vishny 1994; Haugen and Baker 1996).