Your emissions or mine? Examining how emissions management strategies, ESG performance, and targets impact investor perceptions

ABSTRACT

Efforts to mitigate greenhouse gas emissions and curb climate change have recently become significant areas of concern to policymakers. We examine how management's focus on mitigating its direct versus indirect emissions influences the ability to attract capital from investors, and how this ability is moderated by the firm's environmental, social, and corporate governance (ESG) performance combined with adoption of an external emissions target. Using an experiment, we find that investors perceive a firm with a relatively poor ESG performance record as more socially responsible and are therefore more willing to invest when management focuses on mitigating direct versus indirect emissions. We also find that, regardless of ESG performance, adopting an external industry-based emissions target diminishes willingness to invest when management focuses on mitigating indirect emissions, but not when they focus on mitigating direct emissions. Our results provide insights for policymakers as to one impact of disaggregating direct (i.e. Scope 1) and indirect (i.e. Scope 2) emissions in ESG reporting.

KEYWORDS:

• ESG
• ESG reporting
• emissions management strategies
• emissions targets
• firm value

Acknowledgement

We thank Bryan Church, Elizabeth Demers, Krista Fiolleau, Jeff Hales, Khim Kelly, Kristina Rennekamp, workshop participants at Georgia Institute of Technology, University of Duisburg-Essen, University of Southern Denmark, Indiana University, the East Coast Behavioral Accounting Workshop, and brownbag participants at Emory University for their helpful comments and suggestions. We are also grateful to the participants for their time and the Scheller College of Business and the Center for International Business Education Research (CIBER) for financial support.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 As defined by the GHG Protocol Corporate Standard (GHG Protocol 2019; GHG is the acronym for greenhouse gas), direct emissions are characterized as Scope 1 emissions, which are produced from sources companies own or control. Indirect emissions include Scope 2 and Scope 3 emissions. Scope 2 emissions are ‘indirect emissions from the generation of purchased energy.’ Scope 3 emissions are all other indirect emissions that do not fall under the definition of Scope 2 but that occur up and down the value chain of a firm (see http://www.ghgprotocol.org/corporate-standard). We focus on Scope 2 emissions to operationalize indirect emissions as it is more commonly disclosed by companies than Scope 3 emissions.

2 In one example, Johnson et al. (2020) provide experimental evidence that investors prefer a firm mitigates its own direct emissions compared to purchasing carbon offsets. However, an important conceptual distinction between our paper and Johnson et al. (2020) is that purchasing carbon offsets does not reduce a firm's direct or indirect emissions as these emissions are not generated within the firm's value chain. Thus, Johnson et al. (2020) examine investor reactions to reduced emissions that are attributable (via direct emissions) versus unattributable (via carbon offsets) to the firm. As a result, it is unclear whether investors distinguish between types of emissions that are attributable to a firm, which is the focus of this paper.

3 Investor reactions to ESG may be associated with expectations of future financial performance, affective feelings about social and environmental impacts, or perceptions of the firm and its management. Thus, we use the term ‘socially responsible’ to broadly encompass all these dimensions.

4 Companies often utilize targets, which are an important management control tool (Journeault 2016; Feichter, Grabner, and Moers 2018), to meet their emissions-related goals. In fact, almost half of Fortune 500 companies used emissions targets in 2016 (WWF et al. 2017) as did more than 80 percent of globally surveyed companies (CDP 2016). Prior research provides evidence that, in some situations, targets can improve emissions management efforts. For instance, companies that set more ambitious targets, targets that are long-term in nature, or targets that focus on absolute emissions reductions are more likely to achieve those targets than other companies (Dahlmann, Branicki, and Brammer 2019; Ioannou, Li, and Serafeim 2016).

5 Investors are a key driver of ESG initiatives and reporting (Dyck et al. 2019; Eccles and Klimenko 2019), and the demand for investment screening based on ESG performance is particularly growing among retail investors (Fonda 2019; The Asset 2020). For example, sixty-seven percent of financial advisors state they have clients that have expressed interest in ESG factors (BAML Global Wealth and Investment Management survey; BAML 2019). While institutional investors are usually seen as having greater influence, e.g., on share price development, firm decision making, and governance than retail investors, investigating retail investors’ judgments and decisions is still highly important. Retail investors directly hold a significant proportion of equities (e.g., more than 1/3 of US equity ownership, measured in Dollar value; Forbes 2015). Furthermore, many regulatory initiatives in connection with capital markets aim at retail investor protection (MacIntosh 1993). We selected Mechanical Turk workers as our participants as they provide a reasonable proxy for retail investors (Owens and Hawkins 2019).

6 Controlling for investment experience, or any other demographic variable, does not change the results of our tests. The demographics of our study are comparable to other studies that have used Mechanical Turk workers as proxies for retail investors (e.g., Rennekamp 2012; Doxey et al. 2020; Johnson et al. 2020).

7 An experiment is well-suited to examine our research question for two reasons. First, an experiment enables us to randomly assign participants to conditions which only differ based on our manipulated variables of interest. Consequently, we can draw conclusions about causality from our results while avoiding potential endogeneity issues associated with real-world emissions data. Second, disclosure of emissions (including disaggregated direct and indirect emissions) as well as emissions targets can be generally inconsistent while also not generally required. Thus, the empirical data to examine our research question is limited.

8 We chose the amount by which to deviate from the industry average when establishing a firm as being a leader versus a laggard based on extensive pretesting. Specifically, in pretests, we collected participant assessments for our key variables for many different levels of firm ESG performance. We set operational firm level renewable energy production percentages to most integer values on the scale from 0–100 percent and collected multiple datapoints per integer value. The pretesting results were consistent with the economic theory of diminishing marginal utility and suggested a steadily concave association between ESG performance and all participant assessments. In other words, our pretests suggested that moving from 10 percent (ESG Laggard) to 25 percent (industry average) renewable energy production would increase participants’ assessments by about the same magnitude as moving from 25 percent to 70 percent (ESG Leader). From a theory perspective, our chosen manipulation levels capture above- and below-average ESG performance for industry leaders and laggards, respectively. While we acknowledge that variations in the level of these manipulations may affect the magnitude of our results, they would likely not interact with our results or change their directional interpretation.

9 All p-values are two-tailed unless otherwise noted.

10 Because the ESG performance and external target manipulations occurred before participants’ initial willingness-to-invest judgments, one concern with using the change in willingness to invest as our dependent variable is that doing so may introduce estimation bias in our model. To eliminate this potential bias, we follow Yzerbyt, Muller, and Judd (2004) who demonstrate that such bias can be mitigated using an ANCOVA approach by interacting the covariate (i.e., initial willingness to invest) with the uncorrelated manipulated variable (i.e., emissions management strategy) and including this interaction term as an additional covariate. In untabulated analyses, we find that using this approach does not change any of the results or inferences from our hypotheses tests.

11 To create our Socially Responsible mediation variable, we create a single factor which explains 90.24 percent of the variance (eigenvalue = 2.60) based on difference values for five statements about Firm Y taken after participants make their pre- and post-willingness to invest judgments related to perceptions of the firm being: environmentally responsible (0.85), financially responsible (0.29), socially responsible (0.84), honest (0.67), and ethical (0.81). These statements correspond with prior research about ways that ESG may influence investors’ perceptions of companies (Chava 2014; Cheng, Ioannou, and Serafeim 2014; El Ghoul et al. 2018; Elliott et al. 2014; Flammer 2015, 2018; Hartzmark and Sussman 2019; Lins, Servaes, and Tamayo 2017). Fits for both model results reported are good based on generally accepted measures (e.g., traditional chi-square test, Comparative Fit Index, and Root Mean Square Error of Approximation).

Additional information

Funding

This work was supported by Center for International Business Education Research (CIBER); Georgia Institute of Technology Scheller College of Business.