ESG Rating Disagreement and Stock Returns

Abstract

Using environmental, social, and governance (ESG) ratings from seven different data providers for a sample of firms in the S&P 500 Index between 2010 and 2017, we studied the relationship between ESG rating disagreement and stock returns. We found that stock returns are positively related to ESG rating disagreement, suggesting a risk premium for firms with higher ESG rating disagreement. The relationship is primarily driven by disagreement about the environmental dimension. We discuss the practical implications of our findings for firms’ equity cost of capital as well as for investment managers and asset owners who use ESG investment strategies.

Declaration of Interests

Disclosure: The authors report no conflicts of interest.

Editor’s Note

Submitted 6 February 2021

Accepted 29 July 2021 by William N. Goetzmann

Notes

¹ Valuation mistake = [Equity(no disagreement) – Equity(disagreement)]/Equity(disagreement) – 1 = [FCFE/(r_E – g)]/[FCFE/(r_E + 0.0092 – g)] = [100/(0.05)]/[100/(0.05 + 0.0092)] – 1 = 0.184 (or 18.4%).

² See, for instance, Hong and Kostovetsky (2012); Krüger (2015); Lins, Servaes, and Tamayo (2017); Liang and Renneboog (2017); Gibson, Krueger, and Mitali (2020); Dyck, Lins, Roth, and Wagner (2019).

³ See, for instance, Chatterji, Levine, and Toffel (2009); Bouten, Cho, Michelon, and Roberts (2017); Delmas, Etzion, and Nairn-Birch (2013).

⁴ The concept of a common theorization refers to the idea that raters (or information intermediaries) agree on a common definition of CSR. Absence of commensurability captures the idea that different raters do not use the same measures when quantifying the same feature.

⁵ In a recent study, Lopez-de-Silanes, McCahery, and Pudschedl (2019) found evidence that firms with good ESG scores may disclose more information.

⁶ Asset4 was acquired by Thomson Reuters in 2009, but the ESG data were made available under the old name of Asset4. After the acquisition, the name changed to Thomson Reuters ESG Scores. Because the name Asset4 is widely known, however, we used the old name for simplicity. Note that as of 2018, the ESG ratings data of Thomson Reuters are part of Refinitiv and now also known as Refinitiv ESG.

⁷ After acquiring about a 40% stake in Sustainalytics in 2017, Morningstar purchased the remaining approximately 60% of Sustainalytics equity in 2020 (see https://bit.ly/3oXCgxM).

⁸ The data from KLD originate from Kinder, Lydenberg, Domini (KLD) & Co., Inc., which was acquired by RiskMetrics in 2009. In 2010, MSCI acquired RiskMetrics. Eccles, Lee, and Stroehle (2019) have provided details on the history of KLD. We refer to these data as either KLD or MSCI KLD.

⁹ The MSCI IVA dataset was initially created by Innovest Strategic Value Advisors, which was also acquired by RiskMetrics in 2009 before RiskMetrics was taken over by MSCI (see Eccles et al. 2019 for details).

¹⁰ Each provider, with the exception of Bloomberg, had a rather constant number of observations for the different scores they issue. Bloomberg had substantially lower coverage for environmental ratings.

¹¹ Available at http://www.ghgprotocol.org.

¹² In addition, evidence from both academics and practitioners indicates that environmental risk matters to investors. Bolton and Kacperczyk (forthcoming) found a sizable risk premium for firms with high carbon emissions, suggesting that investors do care about carbon risk. In a recent survey by BlackRock (2020), 88% of the respondents placed climate risk at the top of their portfolio concerns.

¹³ The 12-industry classification may be found at https://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library/det_12_ind_port.html.

¹⁴ The reader might wonder why any S&P 500 firms would not have a credit rating. In general, firms without a credit rating do not seem to be exceptional. For example, in a sample of 12,312 firms, Avramov, Chordia, Jostova, and Philipov (2009) reported that 9,051 firms did not have a credit rating. In our sample, 194 out of a total of 553 firms did not have a credit rating for at least one month.

¹⁵ We included industry-month fixed effects and controlled for standard characteristics (not reported) that have been found to explain volatility (see, for example, Dennis and Strickland 2004)—namely, market capitalization (Banz 1981), leverage, business segment (a dummy variable was 1 for multisegment firms), percentage of institutional ownership, the ratio of mutual fund ownership to total institutional ownership, and turnover.

¹⁶ If available, we used the redemption value as preferred stock. Otherwise, we used the liquidating value or, if the liquidation value was also unavailable, the carrying value.

¹⁷ The MSCI KLD dataset seems to have some issues with the CUSIP code. The codes do not always have the same number of characters, and leading zeros are often truncated. Therefore, we filled in leading zeros if the number of characters was less than eight. Then, we added the self-computed check digit to the code if the eighth number was not the would-be check digit if there had been an additional leading zero (in that case, we added a leading zero) or the last two characters consisted of commonly used issue codes.

¹⁸ For US stocks, the ISIN number consists of the country code (first two characters), the CUSIP code (characters 3 to 11), and a check digit.

¹⁹ Because the providers change their ratings at different points in time, we argue that for our purposes, using a monthly frequency makes sense.