Claremont Colleges

Scholarship @ Claremont

CMC Senior Theses CMC Student Scholarship

2024

Is Green the new Gold? ESG’s Influence on the Financial

Performance of the Energy and Mining Industries
Alexa de Vegvar

Follow this and additional works at: https://scholarship.claremont.edu/cmc_theses

Part of the Environmental Studies Commons, and the Finance Commons

Recommended Citation
de Vegvar, Alexa, "Is Green the new Gold? ESG’s Influence on the Financial Performance of the Energy and
Mining Industries" (2024). CMC Senior Theses. 3450.
https://scholarship.claremont.edu/cmc_theses/3450

This Open Access Senior Thesis is brought to you by Scholarship@Claremont. It has been accepted for inclusion in
this collection by an authorized administrator. For more information, please contact [email protected].
 Claremont McKenna College

Is Green the new Gold?

ESG’s Influence on the Financial Performance of the Energy and Mining Industries

Submitted to

Professor Nishant Dass

by

Alexa de Vegvar

for

Senior Thesis

Fall 2023

December 4th, 2023

Acknowledgments

I would like to express my gratitude to my thesis reader Professor Nishant Dass, the Charles

M. Stone Professor of Finance, for his supportive and valuable guidance during the writing and

analysis process. Your dedication and extensive knowledge as an economist have significantly

enriched my work and development as a researcher. I am truly grateful for the time you invested

in the success of this thesis. I would also like to thank Professor Magilke and my thesis seminar

class for their continuous support throughout the semester. The feedback and encouragement

expressed were invaluable and I am fortunate to have shared the experience with you all. Lastly, I

want to thank my family for their abundant love and endless support for me and my academic

career; I could not have done it without you!

2
Abstract

The demand for Environment, Social, and Governance (ESG) reporting has increased in

recent years. As the concern for climate change has moved to the forefront of the investment

process, companies have needed to prioritize ESG in their operations. Previous literature addresses

how ESG has benefited companies’ financial performance and increased shareholder value. The

pandemic has proven to have positively impacted ESG reporting as social priorities shifted.

However, studies have also shown that ESG reporting is too ambiguous to concretely display any

relationship with financial performance. This thesis will analyze the impact of ESG and

environmental data reporting on the Return on Assets (ROA) and Return on Equity (ROE) for the

energy and mining industries over the past 5 years (2018-2023). The regressions were split into

pre- and post-COVID in hopes of seeing a shift in ESG reporting impacts on ROA and ROE. The

results reinforced the unclear relationship between the variables with most regressions being

statistically insignificant. Potential limitations could be due to the paucity of ESG research.

3
Table of Contents

I. Introduction ............................................................................................................................. 5

II. Literature Review .................................................................................................................... 8

a. The History of ESG ............................................................................................................. 8

b. Positive Relationship between ESG Scores and Financial Performance ........................... 10
c. The Role of COVID-19 in ESG ......................................................................................... 11
d. Negative Relationship between ESG and Financial Performance ..................................... 12
e. The Ambiguity of ESG Ratings ......................................................................................... 13

III. Hypothesis ............................................................................................................................. 15

IV. Methodology.......................................................................................................................... 16

a. Dependent Variables .......................................................................................................... 16

b. Independent Variables ....................................................................................................... 17
c. Control Variables ............................................................................................................... 18
d. Data Cleaning..................................................................................................................... 18

V. Data Analysis ......................................................................................................................... 20

a. Summary Statistics............................................................................................................. 20
b. Energy Industry Regressions ............................................................................................. 22
c. Mining Industry Regressions ............................................................................................. 28

VI. Analysis ................................................................................................................................. 34

VII. Conclusion ............................................................................................................................. 37

VIII. Bibliography......................................................................................................................... 39

4
I. Introduction
In this thesis, I analyze the energy and mining industries’ reactions to the increasing

environmental regulation frameworks and the clean energy transition. The 2015 Paris Agreement

is a legally binding international treaty to limit global warming and reduce greenhouse gas

emissions.1 The 194 parties involved in this agreement are required to communicate impactful

climate plans to support the goal of the Paris Agreement.2 This forced the oil and gas industry to

develop and offer alternative, renewable energy sources through governmental subsidy incentives.3

University divestment campaigns and changes in revenue streams are examples of key incentives

for the industry to consider incorporating ESG data into decisions and financial performance. In

2021, Harvard University announced its divestment from oil and gas firms, to “support the

development of a “green economy” and accelerate the reduction of carbon emissions in myriad

ways.”4 The tailwind of sustainable investment strategies will trigger an internal operational shift,

hence my interest in analyzing how environmental development has impacted industry

performance.

Most major corporate decisions run through metrics and data to ensure that they result in

the best outcome. Weighing up the pros and cons has never been a trivial task, and now the role of

Environment, Social, and Governance (ESG) reporting adds another dimension. The data behind

1
“The Paris Agreement,” Unfccc.int, 2023, https://unfccc.int/process-and-meetings/the-paris-
agreement#:~:text=The%20%20Paris%20%20Agreement%20is%20a.

2
United Nations, “The Paris Agreement,” United Nations, 2021, https://www.un.org/en/climatechange/paris-
agreement#:~:text=Today%2C%20194%20Parties%20(193%20States.

3
Brent Potts, “SAP BrandVoice: How the Oil and Gas Industry Is Building a Sustainable Future,” Forbes, October
23, 2021, https://www.forbes.com/sites/sap/2021/10/23/how-the-oil-and-gas-industry-is-building-a-sustainable-
future/?sh=4bc26dac72ce.
4
Nick Anderson and Michael Birnbaum, “Harvard, America’s Richest University, Will Divest from Fossil
Fuels,” Washington Post, September 10, 2021, https://www.washingtonpost.com/education/2021/09/10/harvard-
divest-fossil-fuels/.

5
a firm’s environmental impact has evolved to become a powerful tool for investors, stakeholders,

and policymakers to analyze their alignment with a particular company and understand its role in

society. ESG has also developed into a metric, not for morality, but for the financial benefit of the

company. Increasingly, investors evaluate investment decisions based on values, so it is merely

conducive for a company to promote high-achieving ESG scores as an integral part of their

investor, sales, and employment pitches. McKinsey surveyed CIOs, finding “85 percent of the

chief investment officers we surveyed state that ESG is an important factor in their investment

decisions” and “significant majority are prepared to pay a premium for companies that show a

clear link between their ESG efforts and financial performance.”5 The motivation to incorporate

ESG data into investment decisions is prominent, so it is in the best interest of companies to

emphasize ESG initiatives to win over investors. However, at this stage, ESG reporting’s

weighting is industry-dependent. This thesis’s focal points are specifically the energy and mining

industries and their relationship with ESG.

The expectation of the mining industry to incorporate ESG into their operations is

prominent among investors. Accenture’s Global Institutional Investor Study of ESG in Mining

discusses the rising importance of mining companies’ decarbonization strategies. The study

concluded that “59% of investors want miners to aggressively pursue decarbonization and be

market leaders in that effort” and “approximately 63% of investors responded that they would be

willing to divest or avoid investing in mining companies that fail to meet their decarbonization

5
Jay Gelb and Rob McCarthy, “Investors Want to Hear from Companies about the Value of Sustainability |
McKinsey,” www.mckinsey.com, September 15, 2023, https://www.mckinsey.com/capabilities/strategy-and-
corporate-finance/our-insights/investors-want-to-hear-from-companies-about-the-value-of-sustainability.

6
targets.”6 Investors’ passion for reducing environmental degradation in the mining industry is

apparent, but there is still an unwavering expectation lying on the companies’ financial

performance as a successful investment metric. The study also found that 83% of investors

surveyed said they view improving financial performance and strengthening the balance sheet as

“important” or “very important,”7 which was at the top of “the importance of criteria in driving

the allocation of capital” table. This highlights how the disparity between environmental and

financial values remains prominent for investors; they want improved environmental practices, yet

are not willing to sacrifice the monetary value.

This thesis explores how ESG and environmental variables affect financial performance. I

used the Refinitiv database to gather industry-wide data and ran regressions for each metric against

the Return on Assets (ROA) and Return on Equity (ROA). The results showed that pre- or post-

COVID environmental data logging has little but no impact on the ROA and ROE in the energy

and mining industries.

6
Kathryn Jacobs, Sean Keenan, and Fay Cranmer, “How Investors View Mining’s New Role as a Champion of
Decarbonization,” June 21, 2022, https://www.accenture.com/content/dam/accenture/final/a-com-migration/r3-
3/pdf/pdf-173/accenture-mining-role-champion-of-decarbonization.pdf#zoom=40.
7
Kathryn Jacobs, Sean Keenan, and Fay Cranmer, “How Investors View Mining’s New Role as a Champion of
Decarbonization,” June 21, 2022, https://www.accenture.com/content/dam/accenture/final/a-com-migration/r3-
3/pdf/pdf-173/accenture-mining-role-champion-of-decarbonization.pdf#zoom=40.

7
II. Literature Review
ESG is a criterion used to screen investment opportunities. The term can be broken down

into more specific sections. First, the “E” pillar touches on the environmental factors that need to

be considered. These cover everything from greenhouse gas emissions to water recycling and

biodiversity impact. A company’s impact on the environment is often the go-to assessment of

ESG.8 The “S” stands for social, highlights the considerations of employees and society. Metrics

include gender and racial diversity, human rights, and data protection. 9 Finally, the governance

pillar, “G”. It refers to data points such as decision-making opportunities on the board, shareholder

rights, and corporate performance measurements.10 The scope of ESG is often overwhelming for

investors, yet has become a growing consideration when making investment decisions.

a. The History of ESG

The ESG phenomenon has grown in prominence over the past decades, picking up speed

in the 1960s during the Vietnam War. Protesters “put significant pressure on corporations as the

anti-establishment era unfolded”11 to take on the public interest in a time of political and economic

unease. The Corporate Social Responsibility (CSR) notion grew through the late 20th Century,

incorporating a pyramid of priorities necessary for the fundamental success of a company. This

initial, simplified version of reporting was divided into four distinct sections. The base “economic”

8
“Sustainable Investing - ESG Definition,” Robeco.com - The investment engineers,
n.d., https://www.robeco.com/en-int/glossary/sustainable-investing/esg-definition.

9
“ESG Investing and Analysis,” Www.cfainstitute.org, n.d., https://www.cfainstitute.org/en/rpc-overview/esg-
investing#:~:text=ESG%20stands%20for%20Environmental%2C%20Social.

10
S&P Global, “What Is the ‘G’ in ESG?,” S&P Global, February 24, 2020, https://www.spglobal.com/en/research-
insights/articles/what-is-the-g-in-esg.

11
Cristina Dolan and Diana Barrero Zalles, Transparency in ESG and the Circular Economy (Business Expert
Press, 2021).

8
highlights the importance of offering positive returns to shareholders and upholding a productive

work environment. Thereafter, comes “legal,” another critical aspect of a company. A well-

functioning corporation needs to fit into distinct regulations and requirements, adhering to

employment and health and safety laws. The two smallest portions of the pyramid are the ethical

and philanthropic layers. The ambiguity behind what defines acting with good moral intent is

telling of how complex quantifying corporate ethical behavior can be. Whilst that set an impactful

precedent, the 1990s accelerated the globalization of CSR with the 1992 founding of the United

Nations Framework Convention on Climate Change (UNFCCC) and the 1997 Kyoto Protocol.

Countries that signed onto this effective framework joined “the only climate change agreement

imposing emission reductions on developed countries.”12 For large companies, this protocol

strengthened CSR and how it “represented international efforts for setting higher standards with

regards to climate-related issues and, indirectly to corporate behavior.”13 Adhering to “weak

regulatory frameworks” triggered the start of more concrete ESG reporting.

Corporate ESG has always been considered broad and vague. It covers a vast set of

important issues, from environmental footprint to the retention rate of minority groups, that much

ambiguity remains prominent. Seemingly, metrics, frameworks, and scores vary depending on the

context and tool used, making it challenging to get concise and measurable data to measure and

benchmark progress and reporting. Analyzing corporate ESG data alongside financial and

operating metrics will help us understand how beneficial ESG ratings have been/are across

different industries when making financial decisions depending on the year. It is an important

12
admin, “What Is the Kyoto Protocol and Why Is It Important?,” ESG | the Report, January 19,
2022, https://www.esgthereport.com/what-is-the-kyoto-protocol-and-why-is-it-important/.
13
Mauricio Andrés Latapí Agudelo, Lára Jóhannsdóttir, and Brynhildur Davídsdóttir, “A Literature Review of the
History and Evolution of Corporate Social Responsibility,” International Journal of Corporate Social
Responsibility 4, no. 1 (January 22, 2019): 1–23, https://doi.org/10.1186/s40991-018-0039-y.

9
analysis to conduct because the importance of ESG ratings is prominently growing. Social

responsibility and regulatory compliance are examples of reasons for companies to enhance the

significance of their ESG data in their investment and operation decisions. Looking at the publicly

listed companies allows for deeper analysis because it “requires social coordination, consensual

validation and stakeholder management.”14 This thesis aims to shed light on the extent to which

these ratings can accurately anticipate a company's actions through correlating ESG data with

financial performance.

Public opinion on the environmental impact a company can induce has gained momentum

over the past years. The “ESG Investing” strategy can be praised and ridiculed throughout time,

and the studies reflect this; there is no concrete proof to show its necessity in the investment

decision.

b. Positive Relationship between ESG Scores and Financial Performance

Studies have looked at factors influencing investors’ attitudes towards ESG stocks in hope

to reach a conclusion on the importance of environmental and economic values in ESG-backed

stocks. Raut et al (2023)15 found that although some investors are legally required to add ESG-

integrated investments to their portfolios, they wanted to study the altruistic (environmental

concerns) and egoistic (economic concerns) values that determine investor decisions. They found

that the altruistic value of caring about the environmental impact was weighted more heavily

compared to the egotistical, economic, valuation. This finding goes against the conventional view

that financial gains outweigh moral values, but Raut et al (2023) correlate these results with the

14
Poul Lykkesfeldt and Laurits Louis Kjaergaard, Investor Relations and ESG Reporting in a Regulatory
Perspective, Springer EBooks (Springer Link, 2022), https://doi.org/10.1007/978-3-031-05800-4.

15
Rajdeep Kumar Raut et al., “Investor’s Values and Investment Decision towards ESG Stocks,” Emerald 22, no. 4
(May 21, 2023).

10
international attention sustainability is receiving, making the longer-term return on investment

greater for ESG-backed stocks than from the traditional investment strategy. McKinsey’s 2019

research on how ESG creates values touches on the exponential drive towards sustainable

investing, highlighting the benefits to investment and asset optimization. The constantly changing

nature of regulatory compliance can bear significant costs on companies not addressing their

environmental impact. Carbon taxes for supply chains have been slowly rolled out as a measure to

eliminate energy-intensive manufacturing.16 McKinsey’s study found that the long-term value

creation of ESG is grounded in “building a strong connection with broad elements of society

creates value, not least because it builds resilience into the business model.”17 Society is

transitioning towards a world of green technology, energy neutrality, and light carbon footprints.

ESG reporting and analysis lay the groundwork for long-term business cycles, by mitigating the

impacts of climate change, adapting to the values of investors, and adhering to government policies

and regulations.

c. The Role of COVID-19 in ESG

The COVID-19 pandemic also played a role in corporate ESG priorities. Safraz et al (2021)

conducted interviews with heads of sustainability at European companies to model how the

pandemic impacted those corporations’ relationship with ESG reporting. They found “COVID-19

has increased this attention [to sustainability and corporate social responsibility] and next reporting

exercise will have to present even better what is truly material for the company and how resilience

16
Center for Climate and Energy Solutions, “Carbon Tax Basics,” Center for Climate and Energy Solutions,
February 16, 2018, https://www.c2es.org/content/carbon-tax-basics/.
17
Witold Henisz, Tim Koller, and Robin Nuttall, “Five Ways That ESG Creates Value Getting Your Environmental,
Social, and Governance (ESG) Proposition Right Links to Higher Value Creation. Here’s Why,” November
2019, http://dln.jaipuria.ac.in:8080/jspui/bitstream/123456789/2319/1/Five-ways-that-ESG-creates-value.pdf.

11
is the business to any type of ESG-related crisis that—as it happened for the COVID-19

outbreak—can have strong financial and non-financial impacts for the company”.18 The focus on

materiality highlights how companies are prioritizing how ESG metrics are impacting their

financial success. The pandemic drew in critical public opinions on how corporations were

addressing the crisis from environmental, heath, and financial lenses.

d. Negative Relationship between ESG and Financial Performance

Negative associations of environmental regulations on financial performance are also

common in literature. Ahmend et al (2022) used Russia’s invasion of Ukraine to assess how

beneficial ESG ratings are for corporate responsibility abroad. Deconstructing the Stoxx 600 firms’

ESG scores pre- and mid-invasion helped them calculate the extent of disclosure about

corporations’ Russian operations impacted their ESG scores. As the war progressed, companies

were quick to close business in Russia in support of Ukraine. Ahmend et al found that there “more

highly rated ESG firms were not less likely to operate in Russia nor more likely to meaningfully

inform investors about such activities,”19 indicating that companies that are diligent and well

ranked in ESG frameworks do not necessarily feel more morally obligated to do good. Interlinking

financial performance, CSR, and political alignment shows evidence for correlation of CSR reports

and negative future stock returns. Di Gioli and Kostovetsky (2013) test the hypothesis of

18
Muddassar Sarfraz et al., “Modeling the Relationship between Carbon Emissions and Environmental
Sustainability during COVID-19: A New Evidence from Asymmetric ARDL Cointegration
Approach,” Environment, Development and Sustainability 23 (March 24, 2021), https://doi.org/10.1007/s10668-
021-01324-0.
19
Daniyal Ahmed et al., “Are ESG Ratings Informative about Companies’ Socially Responsible Behaviors Abroad?
Evidence from the Russian Invasion of Ukraine,” SSRN Electronic Journal,
2022, https://doi.org/10.2139/ssrn.4151996.

12
publishing CSR reports and stock prices.20 The democratic party places stronger emphasis on

environmental protection and employee benefit laws, whereas right leaning Republicans place a

stronger emphasis on capitalism and monetary gains. The study researched the relation between

changes in CSR and future operating performance by regressing three-year changes in ROA and

future three-year growth on lagged changes in KLD scores. They found insignificant correlations

for future firm performance, though concluded increases in firm CSR result in a negative change

in financial performance.

e. The Ambiguity of ESG Ratings

Opinions on ESG and corporate responsibility are not warranted without a standardized

ESG framework. Confusion around reporting standards and metrics can halt ESG’s weighting in

investment decisions. Billio et al (2021) shed light on the lack of consistency across ESG rating

agencies and how the performance of ESG-reporting companies compares to those who do not.21

The study highlights the inconsistencies in ESG scores across agencies, due to methodology and

interpretation of factors. Moreover, they found no clear consensus whether high ESG ratings have

influence on a firm’s financial performance. Ultimately, the paper concludes diligence and

benchmarking need to improve to create a cleaner, standardized reporting for investors. Dimson

et al (2020) concluded similar themes, underlining the importance of standardization of ratings

20
Alberta Di Giuli and Leonard Kostovetsky, “Are Red or Blue Companies More Likely to Go Green? Politics and
Corporate Social Responsibility,” SSRN Electronic Journal, October 11,
2013, https://doi.org/10.2139/ssrn.2084900.
21
Monica Billio et al., “Inside the ESG Ratings: (Dis)Agreement and Performance,” Corporate Social
Responsibility and Environmental Management 28, no. 5 (September 1, 2021): 1426–
45, https://doi.org/10.1002/csr.2177.

13
and consistency as it impacts an investment decision.22 The push to prioritize ESG metrics in

investments cannot advance without clarity across the frameworks.

22
Dimson, Elroy, Paul Marsh, and Mike Staunton. 2020. “Divergent ESG Ratings.” The Journal of Portfolio
Management 47 (1): jpm.2020.1.175. https://doi.org/10.3905/jpm.2020.1.175.

14
III. Hypothesis
This thesis will focus on the energy and mining industries. The clean energy transition has

impacted the business development of energy companies, so it would be interesting to analyze how

their environmental scores have impacted their financial performance. In a similar vein, the mining

industry’s priorities have shifted in favor of ESG. Companies are juggling issues like “water

stewardship to ethical supply chains and mine closure — all while trying to navigate what

respondents [to a survey] describe as an “alphabet soup” of regulations and with ongoing data

integrity challenges.”23 As scrutiny around the impact of mining on resource scarcity increases,

the industry has been forced into the implementation of more environmentally-conscious

strategies.

When looking at the literature written about ESG data reporting and its relationship with

investor decision making and company value creation, it is clear there is a positive correlation

between ESG and financial performance.24 Beyond that, the United States Environmental

Protection Agency (EPA) implemented regulatory actions and guidelines to “help combat climate

change and reduce air pollution that harms public health.”25 The enforcement of environmental

policies forces the energy and mining industries to transition to better practices and report ESG

data. For this study, the hypothesis states there is a correlation between good environmental

reporting data and two financial metrics: ROA and ROE.

23
Paul Mitchell, “Top 10 Business Risks and Opportunities for Mining and Metals in 2023,” www.ey.com,
September 26, 2022, https://www.ey.com/en_gl/mining-metals/risks-opportunities.
24
Poursoleyman, Ehsan, Gholamreza Mansourfar, Saeid Homayoun, and Zabihollah Rezaee. 2022. “Business
Sustainability Performance and Corporate Financial Performance: The Mediating Role of Optimal Investment.”
Managerial Finance 48 (2): 348–69. https://doi.org/10.1108/mf-01-2021-0040.

25
EPA, “Controlling Air Pollution from the Oil and Natural Gas Industry,” www.epa.gov, August 10,
2016, https://www.epa.gov/controlling-air-pollution-oil-and-natural-gas-industry#:~:text=EPA.

15
 H1 = There is a positive correlation between environmental scores and ROA/ROE pre-

COVID

H2 = There is a positive correlation between environmental scores and ROA/ROE post-

COVID

IV. Methodology
The ESG and financial data of the energy and mining industries was collected from the

Refinitiv database.

a. Dependent Variables
Corporate Financial Performance can be evaluated by any metric on an income statement.

For this study the ROA and ROE will be the dependent variables to allow for a standardized

financial success metric for both industries.

The first metric, ROA, measures a business’ profitability relative to its total assets.26 The

net income generated, and the capital invested into its assets are compared to output a ratio metric.

Debt is included, so the higher the ROA of a company, the more efficient and successful the

generation of profit is. Similarly, ROE measures the annual net income divided by the

shareholders' equity. An increase in ROE over years indicates how successful the company is at

generating value from shareholders and reinvesting it into profits.27 For this study, both will be

used as dependent variables, because they will indicate the relationship strength between financial

performance and ESG data. ROA considers debt leverage, which would increase ROE and

26
Jeff Schmidt, “Return on Assets & ROA Formula,” Corporate Finance Institute, November 27,
2022, https://corporatefinanceinstitute.com/resources/accounting/return-on-assets-roa-formula/.

27
CFI Team, “Return on Equity (ROE),” Corporate Finance Institute, February 29,
2020, https://corporatefinanceinstitute.com/resources/accounting/what-is-return-on-equity-roe/.

16
decrease ROA, making them both key metrics for assessing how ESG has changed their decision

making.28 For the oil and gas industry, they are more likely to have a lower ROA due to the large

volume of assets and debt relative to their net income.

b. Independent Variables
The independent variables will be the companies’ ESG scores from the last 5 years and

select environmental data points with the “E” pillar. Refintiv “captures and calculates over 630

company-level ESG measures, of which a subset of 186 of the most comparable and material per

industry, power the overall company assessment and scoring process.”29 For this thesis, only the

environmental scores will be used; it allows for more in-depth regression analysis about how

change to green and sustainable investments translates into better financial performance. In

particular, the analysis will include total carbon dioxide emissions, renewable energy purchased,

and renewable energy used. Refinitiv’s definitions of the metrics are as follows:

1. The total CO2 emissions used is calculated through total CO2 and CO2 equivalents

emission in tonnes divided by net sales or revenue in US dollars in million.

2. The renewable energy purchased is calculated as the “total energy produced from

primary renewable energy sources in gigajoules.”

3. The renewable energy used is calculated as “renewable energy produced/purchased for

its own use only – waste that is converted to energy and is used by the company for

their own use.”

28
Ben McClure, “How ROA and ROE Give a Clear Picture of Corporate Health,” Investopedia, April 17,
2023, https://www.investopedia.com/investing/roa-and-roe-give-clear-picture-corporate-health/.
29
Refinitiv, “Environmental, Social and Governance (ESG) Scores from Refinitiv,” Refinitiv, May
2022, https://www.refinitiv.com/content/dam/marketing/en_us/documents/methodology/refinitiv-esg-scores-
methodology.pdf.

17
c. Control Variables
To improve the reliability of the tests and control for the influence of covariates, the

regression analyses will include three control variables. The first control variable I selected was

Total Assets because it accounts for the size of the firm. Due to the expansive nature of the firms

in the data set, the Total Assets variable will limit the economics of scale factor on ROA and ROE.

Debt to Equity Ratio is the second control variable as it accounts for the variation in capital

structure, use of debt, and financial risks from leverage when specifically examining how ROA is

impacted by ESG data. The last control variable chosen was Capital Expenditures. This metric

accounts for a company’s investment in operational improvements and efficiencies. It is important

to account for this when looking at ESG data, because a higher CapEx could imply growth, which

can influence ROE and ROA.

d. Data Cleaning
Firstly, I addressed the energy industry data. I selected all the companies under the

Refinitiv TRBC Energy Sector classification, which covers 2,361 companies. To get the most

accurate statistical analyses, the whole dataset covered the last 5 years of ESG and financial

metrics, which resulted in a total of 12,495 data points. In the same way, the Refinitiv Mining

sector covers 4,140 companies, which resulted in 20,545 observations across 5 years. There were

significant gaps due to incomplete data inputs, however due to the small sample, I kept the

observations in but with no data. I took the natural log of Total Size and renamed it Size. Next, I

scaled Capital Expenditures by Total Assets to create CapEx_Size. I performed the regression on

the log of ROA and ROE. This was done so that the analysis would focus on the relationship

between the ESG data and the percentage change in ROE and ROE.

18
 Due to the disparity in data across each year, I grouped my data into “preCOVID” (<2020)

and “postCOVID” (>2020). This allowed me to analyze the change in impact of ESG and

environmental data reporting on the change in ROA and ROE before and after the pandemic.

Initially, I was hoping to run a cross sectional test to analyze each year independently. The benefit

of a cross-sectional data analysis is that industry-wide comparisons are made which can help us

understand how environmental improvements have affected financial performance. 30 However,

due to the nature of the data inconsistency, I decided to run a time series regression instead.

Refinitiv only provides data spanning the past 5 fiscal years for my analyses, meaning it would

have been difficult to uncover key statistical changes if I had run each year separately.

30
Dheeraj Vaidya, “Cross-Sectional Data Analysis,” Wall Street Mojo, n.d., https://www.wallstreetmojo.com/cross-
sectional-data-analysis/.

19
V. Data Analysis
The large volume of observations from the database allowed me to regress the individual

independent variables on lnROA and lnROE separately. Using STATA, I ran a total of 16

regressions, 8 for each industry. The regressions went as followed:

1. ESG Score on lnROA

2. ESG Score on lnROE

3. CO2 Emissions on lnROA

4. CO2 Emissions on lnROE

5. Renewable Energy Purchased on lnROA

6. Renewable Energy Purchased on lnROE

7. Renewable Energy Used on lnROA

8. Renewable Energy Used on lnROA

The regressions were repeated for the energy and mining industry separately.

a. Summary Statistics
Below are the summary statistics for both industries. It includes both dependent variables,

lnROA and lnROE, the four independent variables, ESG Score, CO2 Emissions, Renewable

Energy Purchased, and Renewable Energy Used, and the control variables, Size, Debt to Equity

Ratio, and Capital Expenditures scaled by Total Assets.

The varied number of observations for each variable is due to the incomplete nature of the

data reporting. The sample size in the regressions only considered the observations with data

points.

20
 1. Energy

Table 1

2. Mining

Table 2

21
b. Energy Industry Regressions
To start, I regressed the overall pre- and post-COVID ESG score on lnROA. Both time

frames exhibited p-values greater than the 5% significance level, 0.115 and 0.332 respectively.

The coefficients imply that as 1 unit of ESG scores increase, there is no significant percentage

point change in ROA. Due to the statistical insignificance, I was not able to reject the null

hypothesis.

Table 3

I repeated the same regression, but with lnROE. In the same way, the ESG score p-values,

0.118 and 0.228, showed no significant relationship between ESG scores and lnROE. The

coefficients were even smaller than the lnROA, highlighting an overall weak and insignificant

relationship, giving no reason to reject the null.

22
Table 4

After the ESG scores resulted in no significant results, I began to break down the

environmental metrics. I started with the total CO2 emissions on lnROA. The p-values for pre- and

post-COVID were statistically very insignificant, 0.844 and 0.946 respectively. Along with the

very small coefficients, there was no justification for rejecting the null hypothesis.

Table 5

23
 Both variables outputted p-values that exhibited statistical significance, 0.00 and 0.018

(<5% significance). However, the small coefficient suggests that an increase in 1 unit of CO2

emissions would not make a substantial change to the lnROE, which only changes the financial

performance marginally.

Table 7
Table 6

Next, I regressed renewable energy purchased on the change in ROA. The pre-COVID

regression resulted in a 0.025 p-value but a 0.000 coefficient. This implies for each additional

unit of renewable energy purchased, there was an increase in ROA by 0.000 percentage points,

which doesn’t make a difference. Conversely, the post-COVID regression resulted in an

insignificant p-value of 0.470 and a 0.000 coefficient. This result concludes that the null

hypothesis cannot be rejected.

24
 Similarly, the renewable energy purchased pre-COVID was significant (p-value 0.05), but

the coefficient was 0.000, no impact on the lnROE. As 1 unit of renewable energy was purchased,

there was a 0.000 percentage point change to ROE, therefore the results don’t give evidence to

reject the null. The post-COVID variable came back minorly insignificant (0.065), reinforcing the

decision to not reject the null.

Table 8

25
 I started by analyzing the relationship between the volume of renewable energy used and

the change in ROA. Both pre- and post-COVID variables were extremely significant, with p-values

of 0.004 and 0.006 respectively. The pre-COVID regression highlights that an increase in 1 unit

of renewable energy used decreased the change in ROA by 0.024 percentage points. Similarly, a

1 unit increase in renewable energy used post-COVID was associated with a decrease in ROA by

0.026 percentage points. The small and inverse relationship between the variables does not give

enough evidence to reject the null.

Table 9

26
 The results were different when running the renewable energy used regression against

lnROE. Pre-COVID, the relationship had no significant significance, with a p-value of 0.220.

However, post-COVID, the p-value was 0.021, which is less than the 5% significance level. This

implies that a 1 unit increase in renewable energy used is associated with a decrease in ROE by

0.071 percentage points. This does not agree with the alternative hypothesis, so I cannot reject the

null.

Table 10

27
c. Mining Industry Regressions
I started with the overall ESG score variable to understand how the pandemic impacted its

effect on ROA and ROE in the mining industry. As shown in the results, the p-values for the ESG

scores pre and post covid had no statistical significance on lnROA, 0.688 and 0.075 respectively.

Alongside the weak coefficients, it was enough evidence not to reject the null hypothesis.

Table 11

When looking at the ESG score and lnROE relationship, pre-covid proved to have no

statistical significance on the lnROE (p-value was 0.131), however, the post-covid ESG score

had a p-value of 0.003, which makes it extremely statistically significant. Unfortunately, the

coefficient of 0.000 does not highlight a strong percentage point increase or decrease in ROE.

28
Table 12

After regressing the overall ESG score, I started analyzing the environmental variables

within the ESG framework. The first regression was the relationship between Total CO2 Emissions

and lnROA. Uniformly to the ESG score, the pre- and post-COVID CO2 emissions had no

significant impact on ROA. Both p-values, 0.845 and 0.856, are extremely close to 1, highlighting

the lack of relationship between CO2 emissions and ROA. The coefficients are also

inconsequential, so I cannot reject the null.

29
Table 13

Like lnROA, the relationship between CO2 emissions and lnROE fails in significance. The

pandemic’s impact on CO2 emissions proved to not make a difference on the ROE, as 0.242 and

0.134 p-values are not within the 5% significance level which would make a relationship

significant. These results are not able to justify the null hypothesis rejection.

Table 14

30
 Renewable energy purchased was my second environmental regression. There were only

290 observations in the regression on lnROA. The volume of renewable energy purchased pre-

and post-COVID had no statistical significance on lnROA, as the p-values were 0.741 and 0.248

respectively. Along with the extremely low coefficients, there is no reason to reject the null

hypothesis.

Table 15

Likewise, to lnROA, the relationship between renewable energy purchased and lnROE

failed to show significance. Though the post-COVID volume of renewable energy purchased was

0.069, close to the 5% significance, it remained statically insignificant. The coefficients were also

inconsequential, so I cannot reject the null.

31
Table 17

The final environmental metric I regressed was renewable energy used. The pre-COVID

p-value is 0.906, which is statistically insignificant. However, the pandemic did affect the

relationship between the volume of renewable energy used and the lnROA. The post-COVID p-

value is 0.045, which lands within the 5% significance level. The coefficient is 0.038, which

implies that a 1 unit increase in renewable energy used post-COVID is associated with a 0.038

percentage point increase in ROA.

Table 16

32
 I repeated the same metric regression on lnROE. The renewable energy used pre-COVID

lacked a significant p-value, with a result of 0.677 and a -0.007 coefficient. In comparison, the

post-COVID outputted a p-value of 0.034 (<5% significance) with a small coefficient of 0.041.

The low p-value indicates strong evidence against the null hypothesis because the relationship is

statistically significant. However, the small coefficient suggests that an increase in 1 unit of

renewable energy used is associated with a 0.04 percentage point increase in ROE units, which

doesn’t change the financial performance that much. Due to the low coefficients, I could not justify

a rejection of the null.

Table 18

33
VI. Analysis
Across 16 regressions for 2 industries, there were only a few statistically significant

relationships. The ESG score only displayed a significant relationship with the change in ROE in

the mining industry, post-COVID (p-value of 0.003). This could be due to increased compliance

regulations whereby companies with stronger ESG scores would be more attentive to the risks

associated with mining, which could positively impact a change in ROE. However, due to the

small coefficient of 0.001, we cannot assume that increasing ESG reporting substantially impacts

financial performance. Consequently, the insignificant relationship between ESG scoring and

ROA and ROE speak to how vague the metric can be when analyzing a company’s performance.

ESG data collection requires standardization and quality of reporting, which is difficult to execute

when needing to calculate metrics for three very different topics (Environment, Social, and

Governance).

The more concentrated environmental data also proved to show similar inconsistencies in

data analysis. The only statistically significant relationships were the pre- and post-COVID total

CO2 emissions and lnROE in the energy industry, with p-values of 0.000 and 0.018 respectively.

Pre-COVID, the coefficient was 0.000, highlighting that as CO2 emissions increased, there was no

change in ROE. This statistical significance could be attributed to the marketing and brand image

associated with large volumes of carbon emissions. As ROE emphasizes shareholder’s equity, the

inverse relationship could be attributed to shareholders’ expectations for environmentally

conscious companies. However, the post-COVID regression outputs a direct relationship between

CO2 emissions and lnROE. A one-unit increase in CO2 emissions is associated with a 0.000

percentage change in ROE. Though there is no strong coefficient, it could be connected to the

increased energy demand resulting from economic recovery. As businesses recover from a

recession and limited supply chains, economic growth could be associated with increased

34
consumption of carbon-intense energy. Though the relationships are significant, the coefficients

are so small that there is no concrete justification to reject the null hypothesis.

The renewable energy purchased was only statistically significant for the energy industry

pre-COVID. The relationship with lnROA was 0.025 with a 0.000 coefficient. It is possible

government subsidies for energy alternatives influence the volume of renewable energy purchased.

The financial benefits from investing in new energy sources could increase ROA, but the 0.000

percentage change would not be sufficient reasoning for the relationship. With lnROE however,

the relationship was inverse. An increase in one unit of renewable energy purchased is associated

with a 0.000 percentage point decrease in ROE. Once again, there is no strong coefficient, but the

statistical significance could be attributed to the high initial costs when investing in renewable

energy. Financing a clean energy project often requires large expenses, which could lead to an

inverse relationship.

Finally, the relationship between renewable energy used and ROA and ROE resulted in the

highest number of significant relationships. Firstly, the energy industry. Both pre- and post-

COVID variables displayed very low p-values (0.004 and 0.006) when regressed against lnROA.

The coefficients were also both negative (-0.024 and -0.026), showing the inverse relationship. As

the volume of renewable energy used increases, the lnROA decreases. Reasons to support the low

lnROA could be either the debt burden from investing in renewable energy or the operational

challenges associated with utilizing new clean technology. Similarly, only the renewable energy

used post-COVID had a statistically significant (p value = 0.021), yet inverse (-0.071) relationship

with lnROE. The economic downturn of the pandemic could be a potential reason behind the

inverse relationship. The energy sector’s reduced budgets and financial constraints could have led

to their inability to profit from their renewable energy investments. The mining sector only

35
displayed a statistically significant relationship with lnROE post-COVID (p-value of 0.034). The

coefficient was positive, 0.041, highlighting that a one-unit increase in renewable energy used is

associated with a 0.041 percentage point increase in lnROE. The pressure from society and

stakeholders for the mining industry to use renewable energy could be a possible reason behind

this result. Moreover, government incentives to use renewable energy may also be a cause for the

direct relationship.

36
VII. Conclusion

This thesis explored the complicated relationship between ESG and environmental scores

and the financial performance of energy and mining companies, measured by ROA and ROE over

five fiscal years. With the tightening of environmental policies, there was reason to believe that

stronger ESG scores and more diligent CO2 and renewable energy use would lead to significant

results.

After 16 regressions that isolated specific metrics to analyze their effect on ROA and ROE,

the results still display weak and ambiguous relationships. The overall ESG scores displayed little

significance on financial performance, highlighting the broad nature of the metric. It covers

Environment, Social, and Governance reports which all assess differently and should not be

encapsulated into one analytical tool for companies. Thereafter, specific environmental metrics

were run against both ROA and ROE. The results were sparse, as most of the p-values were

statistically insignificant and the coefficients were too small for any strong relationship.

Renewable energy used in the energy industry outputted significant but inverse relationships with

lnROA and lnROE. This implies the pandemic had no effect on how renewable energy used has

changed the financial performance of energy companies. On the contrary, the mining industry

shows a change in coefficient after the pandemic. Before 2020, the coefficient was -0.071 with

lnROE. This suggests a 1 unit increase in renewable energy used was associated with a decrease

in lnROE. However, after 2020, the coefficient was 0.041, presenting a positive impact of

renewable energy used on lnROE.

Since most of the regressions outputted minimal statistical significance, there is no strong

reason to reject the null hypothesis. ESG scores, CO2 emissions, renewable energy purchased and

used are not strongly correlated with ROA and ROE. This result reinforces the prior literature on

the topic. ESG and environmental data may have an influence on prospective investors, but

37
ultimately does not impact the operational and financial success of a company. This thesis

contributes insight into the energy and mining industries, two heavily criticized sectors from an

environmental standpoint. It emphasizes how these resource-heavily industries have yet to

incorporate environmental metrics into their decision making. These results speak to the lack of

movement towards clean energy in these sectors.

It is relevant to note this research has limitations that have impacted the results of the

analysis. All the data was pulled from the Refinitiv database, which cannot be assumed the most

accurate data. ESG scoring differs between frameworks making it difficult to get a succinct relation

analysis between ESG and financial performance. This database was chosen due to the limited

available ESG and financial data on one consolidated platform. Moreover, Refinitiv only offered

data from the past 5 years. Even though the total number of observations for the energy and mining

companies were 12,495 and 20,545 respectively, the large gaps in reporting lead to paucity in each

regression. Future studies would benefit from analyzing this relationship on a larger timespan

because of the lagged impact of ESG investments on financial performance.

38
VIII. Bibliography

admin. “What Is the Kyoto Protocol and Why Is It Important?” ESG | the Report, January 19,

2022. https://www.esgthereport.com/what-is-the-kyoto-protocol-and-why-is-it-

important/.

Agudelo, Mauricio Andrés Latapí, Lára Jóhannsdóttir, and Brynhildur Davídsdóttir. “A

Literature Review of the History and Evolution of Corporate Social

Responsibility.” International Journal of Corporate Social Responsibility 4, no. 1

(January 22, 2019): 1–23. https://doi.org/10.1186/s40991-018-0039-y.

Ahmed, Daniyal, Elizabeth Demers, Jurian Hendrikse, Philip Joos, and Baruch Itamar Lev. “Are

ESG Ratings Informative about Companies’ Socially Responsible Behaviors Abroad?

Evidence from the Russian Invasion of Ukraine.” SSRN Electronic Journal, 2022.

https://doi.org/10.2139/ssrn.4151996.

Anderson, Nick, and Michael Birnbaum. “Harvard, America’s Richest University, Will Divest

from Fossil Fuels.” Washington Post, September 10, 2021.

https://www.washingtonpost.com/education/2021/09/10/harvard-divest-fossil-fuels/.

Billio, Monica, Michele Costola, Iva Hristova, Carmelo Latino, and Loriana Pelizzon. “Inside

the ESG Ratings: (Dis)Agreement and Performance.” Corporate Social Responsibility

and Environmental Management 28, no. 5 (September 1, 2021): 1426–45.

https://doi.org/10.1002/csr.2177.

Center for Climate and Energy Solutions. “Carbon Tax Basics.” Center for Climate and Energy

Solutions, February 16, 2018. https://www.c2es.org/content/carbon-tax-basics/.

CFI Team. “Return on Equity (ROE).” Corporate Finance Institute, February 29, 2020.

https://corporatefinanceinstitute.com/resources/accounting/what-is-return-on-equity-roe/.

39
Di Giuli, Alberta, and Leonard Kostovetsky. “Are Red or Blue Companies More Likely to Go

Green? Politics and Corporate Social Responsibility.” SSRN Electronic Journal, October

11, 2013. https://doi.org/10.2139/ssrn.2084900.

Dimson, Elroy, Paul Marsh, and Mike Staunton. 2020. “Divergent ESG Ratings.” The Journal of

Portfolio Management 47 (1): jpm.2020.1.175. https://doi.org/10.3905/jpm.2020.1.175.

Dolan, Cristina, and Diana Barrero Zalles. Transparency in ESG and the Circular Economy.

Business Expert Press, 2021.

EPA. “Controlling Air Pollution from the Oil and Natural Gas Industry.” www.epa.gov, August

10, 2016. https://www.epa.gov/controlling-air-pollution-oil-and-natural-gas-

industry#:~:text=EPA.

“ESG Investing and Analysis.” Www.cfainstitute.org, n.d. https://www.cfainstitute.org/en/rpc-

overview/esg-

investing#:~:text=ESG%20stands%20for%20Environmental%2C%20Social.

Gelb, Jay, and Rob McCarthy. “Investors Want to Hear from Companies about the Value of

Sustainability | McKinsey.” www.mckinsey.com, September 15, 2023.

https://www.mckinsey.com/capabilities/strategy-and-corporate-finance/our-

insights/investors-want-to-hear-from-companies-about-the-value-of-sustainability.

Henisz, Witold, Tim Koller, and Robin Nuttall. “Five Ways That ESG Creates Value Getting

Your Environmental, Social, and Governance (ESG) Proposition Right Links to Higher

Value Creation. Here’s Why,” November 2019.

http://dln.jaipuria.ac.in:8080/jspui/bitstream/123456789/2319/1/Five-ways-that-ESG-

creates-value.pdf.

40
Jacobs, Kathryn, Sean Keenan, and Fay Cranmer. “How Investors View Mining’s New Role as a

Champion of Decarbonization,” June 21, 2022.

https://www.accenture.com/content/dam/accenture/final/a-com-migration/r3-3/pdf/pdf-

173/accenture-mining-role-champion-of-decarbonization.pdf#zoom=40.

Kumar Raut, Rajdeep, Niranjan Shastri, Akshay Kumar Mishra, and Aviral Kumar Tiwari.

“Investor’s Values and Investment Decision towards ESG Stocks.” Emerald 22, no. 4

(May 21, 2023).

Lykkesfeldt, Poul, and Laurits Louis Kjaergaard. Investor Relations and ESG Reporting in a

Regulatory Perspective. Springer EBooks. Springer Link, 2022.

https://doi.org/10.1007/978-3-031-05800-4.

McClure, Ben. “How ROA and ROE Give a Clear Picture of Corporate Health.” Investopedia,

April 17, 2023. https://www.investopedia.com/investing/roa-and-roe-give-clear-picture-

corporate-health/.

Mitchell, Paul. “Top 10 Business Risks and Opportunities for Mining and Metals in 2023.”

www.ey.com, September 26, 2022. https://www.ey.com/en_gl/mining-metals/risks-

opportunities.

Nations, United. “The Paris Agreement.” United Nations, 2021.

https://www.un.org/en/climatechange/paris-

agreement#:~:text=Today%2C%20194%20Parties%20(193%20States.

Potts, Brent. “SAP BrandVoice: How the Oil and Gas Industry Is Building a Sustainable Future.”

Forbes, October 23, 2021. https://www.forbes.com/sites/sap/2021/10/23/how-the-oil-and-

gas-industry-is-building-a-sustainable-future/?sh=4bc26dac72ce.

Poursoleyman, Ehsan, Gholamreza Mansourfar, Saeid Homayoun, and Zabihollah Rezaee. 2022.
“Business Sustainability Performance and Corporate Financial Performance:

41
 The Mediating Role of Optimal Investment.” Managerial Finance 48 (2): 348–69.
https://doi.org/10.1108/mf-01-2021-0040.
Refinitiv. “Environmental, Social and Governance (ESG) Scores from Refinitiv.” Refinitiv, May

2022.

https://www.refinitiv.com/content/dam/marketing/en_us/documents/methodology/refiniti

v-esg-scores-methodology.pdf.

Robeco.com - The investment engineers. “Sustainable Investing - ESG Definition,” n.d.

https://www.robeco.com/en-int/glossary/sustainable-investing/esg-definition.

S&P Global. “What Is the ‘G’ in ESG?” S&P Global, February 24, 2020.

https://www.spglobal.com/en/research-insights/articles/what-is-the-g-in-esg.

Sarfraz, Muddassar, Muhammad Mohsin, Sobia Naseem, and Amit Kumar. “Modeling the

Relationship between Carbon Emissions and Environmental Sustainability during

COVID-19: A New Evidence from Asymmetric ARDL Cointegration

Approach.” Environment, Development and Sustainability 23 (March 24, 2021).

https://doi.org/10.1007/s10668-021-01324-0.

Schmidt, Jeff. “Return on Assets & ROA Formula.” Corporate Finance Institute, November 27,

2022. https://corporatefinanceinstitute.com/resources/accounting/return-on-assets-roa-

formula/.

Unfccc.int. “The Paris Agreement,” 2023. https://unfccc.int/process-and-meetings/the-paris-

agreement#:~:text=The%20%20Paris%20%20Agreement%20is%20a.

Vaidya, Dheeraj. “Cross-Sectional Data Analysis.” Wall Street Mojo, n.d.

https://www.wallstreetmojo.com/cross-sectional-data-analysis/.

42