Sell-Side Analysts’ Assessment of ESG Risk

Sell-Side Analysts' Assessment of ESG Risk

Min Park University of Kansas

min.park@ku.edu Aaron Yoon

Northwestern University aaron.yoon@kellogg.northwestern.edu

Tzachi Zach Ohio State University

zach.7@osu.edu

October 2022

Abstract Financial analysts closely follow a firm's operations and assess the risks that it faces. In this paper, we examine whether analysts incorporate ESG risks into their stock recommendations and target prices. Specifically, we use a unique firm-day level dataset on negative ESG risk incidents to proxy for unobservable risk assessments of analysts. We find that analyst outputs predict future ESG incidents, suggesting that analysts incorporate ESG risks into their models. Our results are robust to controlling for ESG incidents that firms experienced in the past, and are stronger in more transparent information environments, and in the presence of more guidance on ESG issues from the Sustainability Accounting Standards Board. Importantly, we find that analysts incorporate ESG risks through adjusting discount rates rather than cash flow estimates. Overall, our results highlight the ability of financial analysts to synthesize and integrate ESG risks into their research.

*We thank Gus De Franco, Emmanuel De George, Michael Jung, Paul Healy, Peter Joos, Leonardo Madureira, Stan Markov, Seungjoo Ro, Rong Wang, Eric Weisbrod, and workshop participants at Boston University, KAIST, Texas Christian University, Seoul National University, University of Notre Dame for helpful discussions and comments. We thank RepRisk for sharing raw ESG incidents data. Aaron Yoon is a member at RepRisk's academic advisory board. All errors are of our own.

1. Introduction In this paper, we assess whether and how financial analysts incorporate information about

Environmental, Social and Governance (hereafter - ESG) risks. Specifically, we evaluate whether analysts' outputs, such as stock recommendations and price targets, have predictive ability over future negative ESG risk events. Such predictive ability should reflect analysts' incorporation of ESG risks into their analyses, as long as analysts' ex-ante risk assessments are positively correlated with ex-post realizations of ESG risk in the form of negative ESG incidents.

Evaluation of ESG risks is becoming increasingly important for stock market participants. The total assets under management of United Nations Principles for Responsible Investment (UN PRI) signatories that committed to incorporating ESG information to their portfolio decisions has grown remarkably from just a few hundred billion dollars in 2006 to $120 trillion by 2021, representing about three times the capitalization in U.S. markets (PRI 2021). As a result of this growing commitment of capital, obtaining and processing ESG information relevant for investment decisions are proving central to financial decision makers. Over the past decade, ESG raters (e.g., MSCI, Sustainalytics, Bloomberg, and etc.) have been the main information intermediaries in the ESG area, by producing ESG ratings that assess ESG risks for companies (Eccles et al. 2020). However, recent literature questions their validity by pointing out a substantial disagreement in ESG ratings by different ESG raters (Berg et al. 2022). Against this backdrop, sell-side financial analysts have recently claimed significant interest and expertise in evaluating ESG risks (High Meadows Institute 2017). Therefore, understanding financial analysts' capabilities in the area is important.

We study ESG incidents that are related to the operations of the firm (e.g., workplace health and safety, and hazardous waste and water management). It is reasonable to expect that financial

1

analysts would exert effort in collecting information about these risks and incorporate them into research outputs. Generally, analysts perform extensive research to evaluate firms' operating performance. They have deep knowledge about a firm's management, operations, and its business environment and industry, and insights about the risks associated with the stocks that they cover (Ramnath et al. 2008; Kadan et al. 2012; Brown et al. 2015). At least some ESG risks are financially material and would have important value, reputation, and regulatory implications to firms.1 Therefore, we expect some ESG risks to be reflected in analysts' models and outputs. Indeed, analysts often discuss ESG risks in their reports. Appendix 1 provides several examples. In one, a Morgan Stanley analyst explicitly incorporates ESG risks into his Nike's discounted cash flow (DCF) valuation model and cuts the cost of equity by 25 basis points (compared to peers) due to Nike's leadership on ESG issues and lower ESG-related operational risk. In another example, a UBS analyst justifies his positive investment thesis on Chevron by pointing out the company's superior ability to tightly manage oil field operations. In a third example, a Cowen analyst discusses Chevron's investment risks and explicitly deems risks such as oil spills, accidents, strikes, and weather-related events as important operational risks. In sum, analysts' efforts to incorporate ESG risks into their outputs are expected to get reflected in their overall assessment of firms' risks and should result in some predictive ability over future negative ESG incidents.

However, there are reasons to expect some frictions in how analysts process ESG risks. First, financial analysts are late entrants to the ESG space (Katz et al. 2020). Their interest in ESG issues has increased in recent years, likely in response to increase in investors' demand (Amel-Zadeh and Serafeim 2018). Thus, analysts may not have effectively incorporated ESG risks into the valuation

1 Bank of America found that 24 major ESG risks related to firms' operations and supply chain such as mistreating employees, waste management, and oil spills among S&P 500 companies during 2014-2019 had a total market value loss of $534 billion. See Bank of America Merrill Lynch: "10 Reasons You Should Care about ESG." Sep 23, 2019.

2

process throughout our sample period because their expertise in the area has been in the developing stage. Second, analysts argue that it is difficult for ESG information to be incorporated into traditional valuation frameworks because ESG information is non-financial in nature. Our conversations with financial analysts confirm this challenge. Third, given the disagreement about what constitutes ESG activities, unstandardized ESG disclosures, and challenges to measuring and quantifying the costs, benefits and risks associated with ESG performance, analysts may need time to understand ESG (Christensen et al. 2021). Therefore, it is conceivable that analysts might not efficiently incorporate information about ESG risks.

Using RepRisk data on realizations of firm-day level negative ESG risk incidents that are related to firms' day-to-day operations between 2007-2021, we find that analyst recommendation revisions and target price changes are negatively related to the frequency of negative ESG incidents that occur in the subsequent twelve months. Specifically, we find that financial analysts' downward (upward) revisions of their outputs predict more (less) frequent future negative ESG incidents. We focus on stock recommendations and target prices, rather than earnings forecasts, because analysts claim to incorporate ESG risks into their models through the adjustments of discount rates, perpetuity growth, and terminal value, rather than through adjustments to earnings forecasts (CFA Institute 2015, 2018). We also find that our main results hold even after controlling for changes in ESG ratings that are available from independent parties, suggesting that analysts predict future ESG risk incidents at least as well as the ESG raters.

Next, we conduct two tests to investigate the components in analysts' valuation model that could drive the negative relation between their outputs and future negative ESG incidents. In the context of a DCF valuation framework, the driver could stem either from analysts' ability to predict future cash flows or from their ability to incorporate ESG risks by adjusting discount rates. In the

3

first test, we find that analysts' predictive ability is not affected by future cash flow shocks. This evidence is consistent with the notion that analysts incorporate ESG risk into their estimates of firms' discount rates and also with the literature that points out firms' ESG profile is related to tail risks (Hoepner et al. 2022). In the second test, we decompose target price revisions into two components -- changes in EPS and changes in discount rates -- and examine which component drives the predictive ability. We find that the predictive ability only exists in the target price revisions that are not accompanied by concurrent EPS revisions by the same analyst, consistent with analysts embedding ESG risk into discount rates rather than into cash flow estimates. Taken together, these results suggest that the predictive ability of analysts' outputs originates from adjustments to expected returns rather than from revisions of future cash flow estimates.

Finally, we study the potential sources of analysts' ability to assess ESG risks. First, we exploit the staggered release of the Sustainability Accounting Standards Board (SASB) standards that provide guidance on financially material ESG issues. We find that analyst outputs' predictive ability significantly improves after SASB issued its guidance, which suggests that analysts better assess ESG risks as the information environment around ESG issues improves. Second, we find that analysts' predictive ability is concentrated in firms that publish separate ESG reports. This suggests that ESG disclosures contribute to analysts' ability to process ESG information and better assess firms' ESG risks.

Our study contributes to two literature streams: (1) the fast-growing literature on ESG, and (2) the large body of knowledge on financial analysts. We contribute to the emerging ESG literature that evaluates the predictive ability of information produced by intermediaries. Recently, it has been suggested that different ESG raters disagree on how to define, measure, and weigh ESG issues (Chatterji et al. 2016; Berg et al. 2022). In light of this debate, Serafeim and Yoon (2022)

4

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download