ssrn-5067558

Clients, employees and institutional owners:
Who influences corporate decarbonisation commitments?

Andreas G. F. Hoepnera [1], Ifigenia Paliampeloub & Frank Schiemannc
aORCID: 0000-0002-4311-2403 Smurfit Graduate Business School University College Dublin, Dublin, Republic of
Ireland & Platform on Sustainable Finance, DG FISMA, European Union
bORCID: 0000-0001-5641-1181 Faculty of Social Sciences, Economics, and Business Administration University of
Bamberg, Bamberg, Germany
cORCID: 0000-0002-7817-7792 Faculty of Social Sciences, Economics, and Business Administration University of
Bamberg, Bamberg, Germany
Abstract: This study examines the determinants of corporate decarbonization commitments,

focusing on stakeholder pressures from institutional owners, employees with corporate social
responsibility (CSR) concerns, and clients and the effect of green revenue generation. Using
survival analysis with Weibull and Cox models, the findings highlight that institutional ownership
(IO) exerts the strongest influence on decarbonization decisions, followed by employees with CSR
concerns (SGA). Additionally, companies with higher green revenues are more likely to set
decarbonization commitments, often at the subsidiary level, driven by specific customer pressures.
While responsive to stakeholder needs, the subsidiary-level commitment dynamic raises questions
about the legitimacy and alignment of parent-level climate strategies. The research extends
stakeholder theory by linking CSR integration and stakeholder expectations to decarbonization
commitments, underscoring the role of green revenue as a key driver.
Keywords: Decarbonisation Commitments, Green Revenues, Employees with CSR concerns,

Institutional Ownership, Related Climate Actions, Parent Entity, Stakeholder Theory and Subsidiary Entity
JEL Codes: D12, E24, G33, Q54, Q56
Ackowledgements: We are very grateful to Lee Clements and Indrani Dee of FTSE for data provision. The views
expressed in this paper are not necessarily shared by DG FISMA or the European Commission. All remaining errors
are our own.
1
Corresponding author. University College Dublin, School of Business. E-mail address: [email protected]
1. Introduction
Science explains that to reduce carbon emissions, both the public and private sectors should
employ systematic and progressive measures towards climate change mitigation and adaptation.
Consequently, there is an urgent need to align climate commitments and related climate actions of
Party2 and non-Party3 stakeholders with the 1.5 C Paris goal (UNFCCC, 2023). From a climate
change mitigation perspective, non-Party stakeholders can contribute to global temperature
reduction by integrating decarbonisation commitments with related climate actions as part of their
corporate social responsibility (CSR). As non-Party stakeholders assume a greater role in climate
negotiations, understanding the mechanisms driving corporate emissions reduction is essential,
particularly in response to rising CSR expectations from clients, employees, and institutional
owners. This empirical analysis examines the role of stakeholder effect in UNFCCC
decarbonisation commitments and related climate actions, focusing specifically on clients,
employees, and institutional owners.
Previous research emphasizes connecting firm-level commitments with the Paris Agreement by
highlighting how science-based targets (SBTi) shield businesses from monetary losses during
crises (Ben-Amar et al., 2024). Jiang et al. (2023) highlight the lack of accountability and
transparency in emissions reduction targets, demonstrating how these shortcomings might erode
stakeholder trust and the legitimacy of business practices. Lemma et al. (2021) further examine
the impact of climate commitments on financial strategy, highlighting the advantages to a
company's reputation and easier access to long-term debt financing for companies that meet
climate action standards.
This study broadens the application of stakeholder theory, expanding its scope in CSR and
stakeholder effect in UNFCCC decarbonisation commitments and related climate actions.
Determinants comprise customer pressure, employees with CSR concerns, institutional ownership,
and parent-subsidiary entity-level commitment. The last hypothesis assesses the likelihood of a
company signing via its parent or subsidiary entity influenced by factors including customer
pressure, employees with CSR concerns, and institutional owners' interests. We employ survival
2
Parties are countries have ratified or acceded Kyoto Protocol, or the Paris Agreement. The Convention categorises countries in
three groups according to their commitments: Annex-I, Annex-II and non-Anex-I (UNFCCC, 2024).
3
Non-Party stakeholders include a range of entities that are not national governments.The Convention categorises Non-party
stakeholders in four groups: United Nations System and its Specialized Agencies, Intergovernmental Organizations (IGOs)
admitted by the COP, Non-governmental organizations (NGOs) admitted by the COP, other non-Party stakeholders (UNFCCC,
2024Parties
1
analysis, a technique from medical research, to study the likelihood of companies setting
decarbonisation commitments and undertaking related climate actions using financial and
environmental variables. We apply a Weibull parametric model to examine the temporal evolution
of decarbonisation commitments and related climate actions and further apply the Cox semi-
parametric model to evaluate robustness. The findings provide information about the temporal and
contextual factors affecting decarbonisation commitments and related climate actions, highlighting
their dynamic nature and reliance on essential organizational factors.
Our results show that institutional owners and employees with CSR concerns determine
decarbonisation commitments and related climate actions, with institutional ownership
demonstrating a stronger effect. As a result, companies are more likely to commit to
decarbonisation commitments or take related climate actions primarily due to institutional
ownership pressures and secondarily because of employee concerns over corporate social
responsibility, which aligns with the sensitivity analysis findings. We extend Cohen, Kadach, and
Ormazabal’s (2023) research by acknowledging institutional ownership improves corporate
disclosure of climate risk information and demonstrating that institutional ownership affects
decarbonisation commitments and related climate actions. Additionally, based on Kurruppu and
Milne’s (2010) study, employee participation in corporate strategic objectives often links to
internal sustainability reporting. Externally, this alignment boosts a firm’s reputation due to its
commitment to environmental and social performance. Moreover, we contribute to the literature
by demonstrating that firms relying on employees with CSR concerns are more inclined to
establish decarbonization commitments and related climate actions.
Further, we contribute to the literature by deriving that companies with green revenues are more
likely to sign decarbonisation commitments and related climate actions with a subsidiary rather
than a parent entity in line with sensitivity analysis results. This finding calls into question the
parent-subsidiary level commitment condition since a parent company usually has a sustainability
strategy for both the parent company and its subsidiaries. This undermines the legitimacy of the
parent entity's decarbonisation commitments and climate-related actions and, by extension,
challenges the legitimacy of the subsidiary company. A possible explanation is that the subsidiaries
most directly experience customer pressure. Therefore, in cases where the parent company does
not initiate a decarbonisation commitment or related climate action, the subsidiaries take the first
step.
2
3
2. Background
2.1 Institutional Setting

Without definitive reporting frameworks, there is the potential for increasingly 'fuzzy' reporting
practices. This involves companies taking advantage of the absence of frameworks as a one-way
communication channel to provide an overly positive view of their CSR practices in their
sustainability reports (de Freitas Netto et al., 2020).
In this direction, the Corporate Sustainability Reporting Directive (CSRD) was adopted in 2022 to
supersede the NFRD, increasing the number of companies subject to mandatory sustainability
reporting. This directive also implemented a requirement to report results following European
Sustainability Reporting Standards (ESRS), which necessitates the incorporation of sustainability
information into the management report and mandates that companies secure external assurance
of their figures. Specifically, the ESRS (E1) mandates companies subject to CSRD to disclose
information by 2024 on environmental impact and climate-related objectives (Hummel & Jobst,
2024). Although the CSRD is fundamental EU legislation regarding reporting non-financial
information, it does not require firms to adopt or report on net-zero targets.
Likewise, the International Sustainability Standards Board (ISSB) has implemented the
International Financial Reporting Standard (IFRS) S2, which mandates firms to alter their
behaviour and facilitate the transition to a net-zero economy, regardless of whether the individual
company has a specified net-zero target (Chan et al., 20224).
The Task Force on Climate-Related Financial Disclosures (TCFD) also provides a global, non-
mandatory framework for net zero reporting. This framework requires companies to disclose
climate change risks and makes it easier to track their progress toward net zero (Chen et al., 2024).
In the recent history of climate negotiations, there has been a significant effort to engage all levels
of society (Party and non-Party stakeholders) to set climate commitments and accelerate climate
action to tackle the climate crisis in one agreement (the Paris Agreement, COP20) (UNFCCC,
2023). Since COP 20, there has been increasing collaboration between governments and non-
party-stakeholders to recognise climate commitments from all actors (COP 21), establish a
partnership of collaboration to lower emissions (COP 22, the Marrakesh Partnership for Global
Climate Action) and set a plan with specific tools (COP 26, five-year plan of the improved
Marrakech partnership) to scale up climate action via the Marrakesh Partnership (UNFCCC,
4
2023). One of these tools is the ‘Yearbook of Global Climate Action’, the seventh of its series
(COP 27), which states the global climate action of non-Party stakeholders, listed in the Global
Climate Action Portal (GCAP), to encourage collaboration between Parties and non-Party
stakeholders (UNFCCC, 2023). Specifically, the UNFCCC secretariat and the governments of
Peru and France launched in 2014 the Global Climate Action Portal (GCAP). This platform
exhibits the current state of climate-related commitments of non-party stakeholders engaging in
climate action globally (UNFCCC, 2023). As of October 2023, 32,524 actors registered on the
GCAP portal (UNFCCC, 2023). Moreover, 12,000 actors focus on climate adaptation, and 1,654
investors engage in climate action via various commitments such as reducing emissions, issuing
green bonds, and re-financing investments with climate-friendly technologies or assets (UNFCCC,
2023).
In the last few years, there has been a growing number of decarbonisation pledges from corporate
actors as listed in many data initiatives, yet the pace and level of ambition are still not aligned with
the 1.5 C Paris goal (New Climate Institute, 2023; Net Zero Tracker 2023; Science Based Targets
Initiative; SBTi, 2023; UNFCCC, 2023; Utrecht University, Data-Driven EnviroLab, and CDP,
2023). Consequently, the UNFCCC secretariat phased down 2023 the 2015 Climate Neutral Now
initiative, as the climate-neutral recommendations differ from the updated net-zero pledges that
aim at the 1.5 C Paris goal (UN, 2022)4.
In this direction, the report ‘Integrity Matters: Net-Zero Commitments by Business, Financial
Institutions, Cities and Regions’ provides a baseline for accountability in net-zero emissions
pledges yet does not provide all the necessary tools to assess the progress of net-zero emissions
pledges (UN, 2022; UNFCCC, 2023). Consequently, the UNFCCC published the Recognition and
Accountability Framework for non-party stakeholders. This framework aims to enhance the
accountability of voluntary climate commitments and action initiatives by implementing principles
of engagement, governance, data management, and individual implementation plans (UNFCCC,
2023). Specifically, the implementation plans set objectives and timelines in line with the
framework to verify the integrity and transparency of voluntary commitments of non-party
stakeholders (UNFCCC, 2023). An ongoing consultation is needed to improve this framework to
analyse targets and assess their progress to enhance transparency and standardisation across
4
It is vital to calculate and report scope 1, 2, and 3 emissions across the value chain to enable all industry non-Party stakeholders
to transit to net-zero GHG emissions by 2050 (UN,2022; UNFCCC, 2023b).
5
commitments, delivering relevant recommendations around COP28 (UNFCCC, 2023). Further,
the UNFCCC secretariat aims to provide these recommendations to GCAP to facilitate the
recognition, progress reporting, and verification of voluntary climate action commitments by non-
Party stakeholders and assist non-Party stakeholders in registering commitments, plans, and
progress reports in the GCAP (UNFCCC, 2023)5.
Looking ahead to COP28, among other things, non-party stakeholders and key partners must
collaborate responsibly on climate action and set credible commitments that are systematically
validated in terms of their integrity and progress reporting (UNFCCC, 2023).
2.2 Stakeholder Theory
The term ‘stakeholder’ was first introduced by the Stanford Research Institute in 1963, stating that
shareholders and stakeholders should be considered. The theory was further developed by
Freeman’s book 1984, Strategic Management: A Stakeholder Approach, which established the
theory by arguing that companies have to provide value for all stakeholders, not just shareholders.
Thus, this represents a shift from the conventional shareholder-centric perspective of profit
maximisation for shareholders alone. Donaldson and Preston (1995) argued that firms have a moral
obligation to consider all stakeholders' interests, which can result in long-term profitability. At this
point, stakeholder theory emphasises the value of corporate responsibility embedding ethics and
strategic management.
In the early 2000s, stakeholder theory expanded to include sustainability, promoting businesses'
adoption of responsible environmental and social practices. This aligns with CSR principles, which
urge companies to meet their moral, social, and ethical obligations to benefit all stakeholders (Peng
and Yang, 2014). Relevant research links stakeholder engagement to CSR commitment and
corporate performance. For instance, Adomako and Tran (2022) conclude that stakeholder
integration positively influences a firm’s CSR commitment, which serves as a link between
stakeholder integration and corporate performance. Further research reveals that a firm’s CSR
commitment enhances social and environmental performance (Anser et al., 2020).
Linking stakeholder theory to specific stakeholder groups like clients and employees highlights its
5
In line with the accounting perspective, the International Sustainability Standards Board (ISSB), established the IFRS foundation
(COP 26) to set sustainability standards disclosure for corporate actors. In June 2023, the IFRS in collabordaiton with TCFD,
established, sustainability disclosure requirements for corporate risks and opportunities (IFRS S1) and climate related disclosure
(IFRS S2) to apply in conjunction.
6
practical implementation in business management. For instance, research proposes that the main
stakeholders are shareholders (primary stakeholders), employees (internal stakeholders),
customers, suppliers, communities, government, and the environment (Mahajan et al., 2023).
Another study explains that embedding CSR in corporate culture allows employees to align with
these values and reinforces a long-lasting competitive edge for the firm (Waheed & Zhang, 2022).
A different analysis suggests that customers prefer socially responsible suppliers as they are risk-
averse and align with their customers' CSR values, hence increasing their likelihood of being
selected for new business opportunities (Tao et al., 2023).
Empirical research shows that stakeholder pressures and engagement influence corporate
sustainability reporting (Gallego-Alvarez Ortas, 2027). It is valuable to examine if research
regarding specific stakeholder groups (clients, employees, and institutional owners) applies to real-
world company examples. First, Unilever’s business responsibility and sustainability report for
2023-2024 states that net-zero carbon technologies are integral to supplying goods to evolving
consumers and addressing environmental challenges (Unilever, 2024). This statement shows
customers can play an essential role in how a company applies CSR practices in production.
Second, in 2022, Apple stated in its environmental progress report that investing in forestry carbon
removal and other projects offers investors a financial return (Apple, 2022). This statement
illustrates how investors can create pressure to harmonise a profit perspective with sustainability
targets. Third, in 2020, Microsoft’s president, Brad Smith, apologised for using a carbon-neutral
definition in the past and revised Microsoft’s net-zero commitment to ‘carbon-negative by 2050.'
(Smith, 2020). This announcement clarifies that employees with CSR concerns may take the
proactive initiative to evaluate and implement efficient corporate decarbonisation commitments.
In conclusion, stakeholder theory evolved from a shareholder-centric approach to a more holistic
perspective that integrates CSR alongside stakeholders' interests to achieve sustainability and long-
term competitive advantage. Finally, connecting stakeholder theory research with specific
stakeholder groups' influence in real-world company examples is vital to comprehending how
clients, employees, and investors shape CSR today.
7
3. Hypotheses Development
3.1 Clients value alignment stakeholder theory hypothesis

Aligning one's sustainability reporting and decarbonisation commitments with client expectations
is a rational business decision from a stakeholder theory perspective, whereby the decarbonisation
commitment is an instrument expected to support client retention. Relevant research shows that
companies with green revenue generation either via regulatory initiatives (e.g., EU Taxonomy)
(Bassen et al., 2022) or technological advancements (e.g., green patents) (Klausmann et al., 2024)
can foster green transition and, hence, corporate decarbonisation.
Hypothesis H1 predicts that a company will commit to decarbonisation when its green revenue is
high, possibly due to a sustainable value-aligned client portfolio.
H1: Companies experiencing higher customer pressure are more likely to set decarbonisation
commitments.
3.2 Investor pressure (Equities) stakeholder theory hypothesis
Shareholder activism can induce voluntary corporate disclosure regarding climate risk information
(Flammer, Toffel, & Viswanathan, 2021). Research finds that such efforts are more effective when
initiated by institutional investors who value disclosure of climate risk information, resulting in
higher stock market post-disclosure (Flammer, Toffel, & Viswanathan, 2021). Companies with a
higher percentage of institutional investors are more likely to set decarbonisation commitments
due to increased pressure from these unique groups of stakeholders: corporate and institutional
owners. Institutional investors with significant ownership are positioned to demand transparency
and accountability, including long-term commitments to corporate emissions reduction. According
to stakeholder theory, such commitment by their investee firms puts institutional investors in a
better reputation with their stakeholders, such as pension members, bank depositors, or insurance
policyholders. Important research shows a positive association between institutional ownership
and enhanced corporate climate risk disclosure (Cohen, S., Kadach, I., & Ormazabal, 2023; Ilhan,
Krueger, Sautner and Starks, 2023). Essentially, institutional investors instrument corporations via
their share ownership to advance the investors' relationships with their stakeholders and improve
transparency in climate reporting.
In contrast to H1, the corporation is instrumented in this process rather than instrumenting
something else. Consequently, we expect a more substantial presence of institutional investors as
uniquely positioned stakeholders to incentivise companies to align with their expectations by
8
making decarbonisation commitments. Thus, the percentage of institutional ownership serves as a
proxy for shareholder pressure exerted on companies to set decarbonisation commitments.
H2: Companies with more institutional owners are more likely to set decarbonisation
commitments.
3.3 Employees influence stakeholder theory hypothesis
Research shows that employee engagement in CSR practices influences a firm's environmental
performance, thus rendering them a vital stakeholder group (Kucharska and Kowalczyk, 2019;
Testa, Boiral and Iraldo, 2018). Companies that rely more on employees with CSR concerns are
more likely to set decarbonisation commitments. Relevant research shows that companies that
prioritise employees interested in corporate strategic objectives often integrate sustainability
reporting into their strategic vision, using it to facilitate organisational transformation, improve
employee engagement, and foster innovation (Kurruppu and Milne, 2010). This strategy enhances
the company's reputation by committing to enhance its environmental and social performance
(Kurruppu and Milne, 2010). Consequently, companies that rely on employees with CSR concerns
should continuously enhance their capability to analyse and implement sustainability initiatives.
Decarbonisation commitments, as a logical extension of these efforts, align with stakeholder
theory.
H3: Companies that rely more on employees with CSR concerns are more likely to set
decarbonisation commitments.
3.4 Parent versus Subsidiary Level Commitment
A company can sign a decarbonisation commitment with a parent or subsidiary. Subsidiary entity
commitment may reflect limited operational greenhouse gas emissions (GHG) emissions scope
coverage of the parent company. For instance, a company that signs a commitment with a
subsidiary can generally cover distinct operations. Hence, a subsidiary commitment may cover
only a portion of the company’s operational activities and thus account for a limited scope of GHG
emissions. However, a parent’s entity commitment typically covers all subsidiaries, possibly the
entire operational GHG scope.
The decision to sign a decarbonisation commitment at the parent company vs. subsidiary level can
also be related to the stakeholder pressures experienced by the company. Customer pressure can
be specific to certain markets, products, or services. Therefore, it is more likely that customer
pressures are experienced more strongly at the subsidiary level. Accordingly, we hypothesise:
9
H4: Companies experiencing higher customer pressure are more likely to sign with a subsidiary
than a parent entity.
Pressure from employees and investors is more likely to address the company as a whole. Investors
are interested in the overall firm value, which directly relates to the whole company. If employees
seek to work for companies that align with their personal values, then such an alignment would be
most credibly signalled through commitments at the parent level. Accordingly, we hypothesize:
H5: Firms that have employees with CSR concerns are more likely to sign with a parent than with
a subsidiary entity.
H6: Firms that have institutional owners' interests are more likely to sign with a parent than with a
subsidiary entity.
10
4. Methods
4.1 Sample
This section defines our study sample, the data sources, and the collection and analysis of data.
For decarbonisation commitments, we used the UNFCCC Actors website. The website provides
information about ISIN, accounting year, organization name, the year companies set
decarbonisation commitments, and further information relating to decarbonisation commitments
and related climate actions (e.g., whether the company has a GHG inventory). This dataset has
been provided by SDGlabs.ai - Scientists for Sustainability. Based on the UNFCCC Actors
website, a fourth dataset was created in collaboration with the University of Bamberg, which
matched the legal names of companies listed on the UNFCCC Actor website with the exact legal
entities of the corporations as expressed on the database of the Global Legal Entity Identifier
Foundation (GLEIF). This process allowed us to determine if a company signed UNFCCC with
its ultimate parent or a subsidiary.
The data from the UNFCCC Actors website was merged with FTSE Russell's environmental and
financial data. Specifically, we distinguish vital environmental variables such as companies' green
revenues. FTSE Russell provides financial variables, including size and Tobin’s q—Shareholder’s
equity, total debt and market capital before investment.
A third merged dataset contains Refinitiv contextual financial variables, including beta, volatility,
revenue, Institutional owner’s percentage (IO), Selling General and Administrative percentage
(SGApercent), and Research and Development (R&D) expenditure. Descriptive statistics for all
these variables are provided in Table 2.
4.2 Dependent Variables
As we aim to explain the decarbonisation commitment decision based on stakeholder pressure, the
primary dependent variable of our study is commitments (UNFCCC, 2023), which is time-variant
and binary. The commitments variable equals one if a company published a decarbonisation
commitment in this or any previous year and zero otherwise.
Voluntary commitments are climate pledges that aim for transparency and accountability
(UNFCCC, 2023). The UNFCC introduced the Recognition and Accountability Framework and
its Implementation Plan for non-party stakeholders (businesses) to ensure that these commitments
11
adhere to engagement, governance, and data management (UNFCCC, 2023). The commitments
variable allows the tracking and reporting of the progress of voluntary commitments of non-party
stakeholders in this sample supported by the GCAP to verify the integrity and accountability of
the voluntary climate commitments (UNFCCC, 2023).
Emission inventories compile data on GHG emissions from various sources for accurate
assessment and reporting (UNFCCC, 2023). The UNFCCC initiative of GCAP allows non-party
stakeholders to track progress and enhance transparency in GHG reporting, validating climate
actions (UNFCCC, 2023). The GCAP embeds emission inventories into a broader climate
framework, which is decisive in monitoring progress towards the Paris Agreement goals, fostering
integrity and trust in global climate efforts (UNFCCC, 2023). It is important to note that according
to GCAP data, firms may report information on emission inventory independently of their
decarbonisation commitments. That is to say that a firm may maintain an emission inventory
without reporting a commitment to decarbonization. The emission inventory variable is time-
variant and binary. It is equal to one if a firm has an emission inventory in the current or previous
year and zero otherwise. For further analyses, we also test dependent variables that refer to related
climate actions: finance actions, mitigation actions, and impact (UNFCCC, 2023).
Finance actions: The company accounts for financial actions (e.g., carbon price reporting, green
bond issuance reporting). The finance actions variable is a time-variant, which equals one if a firm
reports finance actions in the current or previous year and zero otherwise.
Mitigation actions: Mitigation metrics focus on the significance of voluntary commitments for
climate mitigation, emphasising the importance of transparent frameworks (e.g. the GCAP) to
ensure the voluntary commitments effectively reduce GHG emissions (UNFCCC, 2023). Hence,
emission inventory metrics are essential for tracking and verifying climate actions (UNFCCC,
2023). The UNFCCC Actors website collects as mitigation actions the mitigation cause and
solution as reported in the 2023 CDP-ICLEI questionnaire for non-party stakeholders. The
mitigation actions variable is a time-variant equal to one if a firm reports mitigation actions in the
current or previous year and zero otherwise.
Impact: Impact metrics emphasise the credibility and transparency of non-party stakeholders'
voluntary commitments to climate action (UNFCCC, 2023). Accurate reporting and tracking of
emission inventories in the GCAP allow for monitoring climate action progress. For instance, an
organisation that displays a reduction in GHG emissions in the latest year compared with the
12
previous year shows a positive impact. The impact variable is time-variant and binary. The variable
equals one if a firm reports a positive impact for the current or a previous year and zero otherwise.
We present three company examples that report commitments, emission inventory, finance actions,
mitigations actions and impact in the GCAP platform (i.e. the variable is equal to one) in section
4.2 in Table 1.
4.3 Independent Variables of Interest
The Company Green Revenue Percentage (CompanyGR) quantifies the percentage of total revenue
derived from sustainable goods and practices (LSEG, 2024). FTSE Russell’s Green Revenues data
model evaluates this measure and categorises green revenues across many industries (LSEG,
2024). Research explains that green revenues is a new and precise environmental rating compared
to existing ESG ratings, as it provides specific metrics and thresholds per company activity,
maintaining rubric consistency (Bassen, Shu, and Tan, 2023).
Institutional ownership (IO) is the percentage of corporate outstanding shares owned by
institutional investors, including insurance companies, mutual funds, and pension funds (LSEG,
2024). This ownership grants institutional investors considerable influence over corporate
behaviour, particularly practices and policies concerning sustainability. Recent research concludes
that institutional investors’ demand for climate-related information increases climate disclosure
and reduces carbon emissions (Cohen et al., 2023).
Selling General and Administrative percentage (SGApercent) denotes the proportion of a
company’s revenue allocated to Selling General and Administrative (SGA) expenses, to which we
added research and development (R&D) costs. This indicator explains the quantity of resources
allocated to employees with CSR concerns.
Signwithsubsidiary describes circumstances in which a business enters into a commitment through
a subsidiary rather than the ultimate parent firm (LSEG, 2024a). The parent-subsidiary distinction
is essential when evaluating the extent and implementation of commitments and climate-related
actions inside business groupings (LSEG, 2024a).
4.4 Control Variables
To ensure the robustness of our analysis, we control for internal and external factors. Specifically,
we refer to prior literature to select control variables that account for heterogeneity, which could
13
influence companies' propensity to set decarbonisation commitments (Freiberg, Grewal and
Serafeim, 2021). We control for total Size, Volatility, and Profit as prior research suggests that
these variables relate to the difficulty of firms’ carbon emission reduction pledges (Ioanou et al.,
2016). Further, we control for profit, Tobinq and Beta_ftse as these factors impact decision-
making in a firm setting a decarbonisation commitment. Profit is calculated with FTSE Russell
data as the ratio of gross profit in the numerator and the sum of shareholders' equity and total debt
in the denominator in USD dollars. Similarly, Tobinq is calculated with FTSE Russell data as the
ratio of market capital before investment and the total debt in the numerator and the sum of
shareholders' equity and the total debt in the denominator in USD dollars. We control for Tobinq
as companies with high Tobinq are more intangible. We control for Beta as companies with a
higher beta are more sensitive to market fluctuations and most likely undergo investor and
stakeholder scrutiny in risk management. This also aligns with relevant empirical analysis that
indicates that companies with higher beta are more likely to be compelled to make firm
commitments (Bolton and Kacperczyk, 2021).
We provide an overview of all variables explained in sections 4.2 through 4.4 in Table 2.
[Insert Table 2]
4.5 Survival Analysis Model
This study aims to investigate the determinants that drive decarbonisation commitments or related
climate actions. In this analysis, we study the effect of corporate green revenues, institutional
ownership, and employees on decarbonisation commitments or related climate actions by applying
survival analysis. The statistical technique, survival analysis in comparison with logit regression,
considers the information on whether an event occurred and the length of time it took for an event
to occur. Survival analysis is derived from medical research that studies the survival of patients in
relation to an event that happens, such as death. Similarly, survival analysis can be applied in
scenarios where the occurring event is positive (Hoepner, Majoch and Zhou, 2021; George, Seals
and Aban, 2014; Hosmer and Lemeshow, 2008). Specifically, the event of interest in our analysis
represents a company setting a decarbonisation commitment or undertaking a related action. Thus,
when a company makes a decarbonisation commitment, it ‘dies’ in the sample (conditional
probability= 1) as opposed to a company with no commitment that is ‘surviving’ (conditional
14
probability= 0). The survival analysis runs yearly for 2019-2021; thus, once a company commits,
it disappears in the next year’s survival analysis sample, creating a ladder effect.
Here, the dependent variable is the hazard rate, the conditional probability that an event occurs at
a specific time interval (Hoepner, Majoch and Zhou, 2021; Hosmer and Lemeshow, 2008).
A parametric Weibull model as a function of time that covariates as follows eq (1):
H (x|t) = ho(t) r(x,β) (1)
The hazard function is the probability of two functions. The function ho(t) in equation (1) is
described as a ‘time function’, which characterises how the hazard function changes as a function
of survival time. The other function, r(x,β) in equation (1), described as a ‘characteristics function’,
explains how the hazard function changes as a function of our subject covariates. The Weibull
parametric model assumes the baseline time function and follows a Weibull distribution as in
equation (2):
ho(t) = ptp-1 (2), where p is the parameter to be estimated.
Here, p = 1, for any year a company sets a decarbonisation commitment or undertakes a related
action, the entire time function collapses to 1, and the overall hazard function turns into an
exponential regression eq (2). Following Cox (1972), we define the second characteristic function
as in equation (3):
r(x,β) = exp(β1χ1+....+βpxp) = exp (β1key variablesi + β2control variablesi) (3)
Integrating equations (2) and (3) into (1), the Weibull parametric regression model is specified as
in equation (4):
h(x|t) = ptp-1 exp (β1key variablesi + β2control variablesi) (4),
where h(x|t) is the hazard at time (t) for a given set of covariates x1+...+xp. In classic survival
analysis, a hazard ratio greater than one indicates that the company is more likely to commit to
decarbonisation or take a related action, and a hazard ratio less than one means it is less likely to
commit to decarbonisation or take a related action.
The Weibull parametric model is a specific case of the Cox hazard model, assuming a parametric
form on the baseline hazard function (ho(t)) (Clark et al., 2007). This model is appropriate for
modelling data with exponentially increasing or decreasing hazard rates over time. The Cox model
does not specify the baseline hazard function as Hosmer and Lemeshow (2008) explain but uses a
non-parametric Aalen-Breslow estimator to estimate the hazard function. Since net
decarbonisation commitments or related action signatories are likely exponential in function,
15
favouring the Weibull model but, theoretically speaking - the function is unknown, it represents
prudent research practice to use the semi-parametric Cox hazards model as a robustness test.
16
5. Results
5.1 Descriptive Statistics
The summary statistics table of the survival analysis data provides an overview of dependent,
independent, and control variables and their distribution across our sample of 5,402 firm-year
observations.
Dependent Variables
The variable Commitments shows a mean of 0.07441 (SD = 0.26247), signifying that merely
7.44% of companies in the sample set Commitments, with a median value of 0 and maximum of
1. Likewise, the variable Emission Inventory has a mean of 0.07487 (SD = 0.26307), indicating
that 7,49% of companies maintain their own emission inventories, with a median value of 0 and a
maximum of 1. However, these descriptives also indicate that firms in this sample are more likely
to maintain emission inventories than set Commitments. The Finance Actions variable
(mean = 0.04054, SD = 0.19724) shows that only 4,05% of companies take financial actions related
to climate change, while the Mitigation Actions variable (mean = 0.06687, SD = 0.24975), explains
that merely 6.69% of firms take mitigation actions related to climate change. Finally, the variable
Impact has a mean of 0.07460 (SD = 0.26272), indicating that 7.46% of enterprises make a positive
impact by reducing scope 1&2 GHG emissions.
Independent Variables of Interest
The variable CompanyGR has a mean of 6.5703 (measured as the percentage value out of 100) and
a substantial standard deviation of 17.7550, indicating significant variability in firms with green
revenue percentage, suggesting that while many companies exhibit no green revenue percentage
(Median = 0), for other companies we find all revenues to be classified as green (Max = 100). IO
exhibits an average of 17.4884 (i.e., average institutional ownership is 17.4884 %), with a
relatively wide dispersion of (SD = 13.2610), indicating that institutional ownership varies greatly
among companies, from almost none (Min = 0.0304) to complete ownership (Max = 100). Another
critical variable of this analysis, SGApercent has a mean of 0.2111 and a relatively low standard
deviation of 0.1634, implying that, on average, companies spend 21.11% of their revenue on
SGApercent expenses with some variability. The variable SignwithSusidiary has mean values close
to 0 and minimal variations (SD = 0.0560). This indicates that most companies in this sample do
not sign a commitment with a subsidiary or have a low probability of doing so.
17
Table 3 Descriptive statistics
Variables Observations Mean SD Min Median Max
(1) (2) (3) (4) (5) (6)
Commitments 5402 0.7744 0.2624 0 0 1
Emission Inventory 5402 0.0747 0.2630 0 0 1
Finance Actions 5402 0.0405 0.1972 0 0 1
Mitigations Actions 5402 0.0668 0.2497 0 0 1
Impact 5402 0.0746 0.2627 0 0 1
CompanyGR 5402 6.5703 17.7550 0 0 100
IO 5402 17.4884 13.2610 0.0304 14.7667 100
SGApercent 5402 0.2111 0.1634 0.0003 0.1690 0.9905
Signwithsubsidiary 5402 0.0031 0.0560 0 0 1
Leverage 5402 1.5512 1.4947 0 1.0759 13.6078
Profit 5402 0.2772 0.2791 -0.0871 0.2061 5.2033
Beta_ftse 5402 0.5834 0.3553 0.0011 0.5177 9.1432
Size 5402 22.4683 1.2895 15.3362 22.4016 27.2923
Vol 5402 2.2040 0.7806 0.4525 2.0991 7.5195
Tobin Q 5402 0.3639 0.2574 0 0.3409 1.0002
Table 3 represents the basic summary statistics of key and control variables for the whole sample period from 2011 to 2021. Columns 1-5 report
the total observations, mean, standard deviation, minimum, median, and maximum, respectively.
5.2 Results of Survival Analysis
5.2.1 Commitments
To test the association between stakeholder effect and decarbonisation commitments, we run a
Weibull regression and report the results of models 1-4 in Table 3. The results show that variable
IO is significant across models 1-4, with hazard ratios greater than one and p-values significant at
p<1%. This result supports H2, indicating that companies with a higher percentage of institutional
owners are likelier to set decarbonisation commitments. SGApercent has hazard ratios greater than
18
one and p-values significant at p<1% in models 2 and p<10% in models 1, 3 and 4. This result
supports H3, implying that companies relying more on employees with CSR concerns are likelier
to set decarbonisation commitments.
[Insert Table 3]
Further, testing the interaction coefficients6 in models 2-4, we conclude that H4 is supported, but
H5 and H6 are not. The coefficient CompanyGRSignwithsubsidiary shows a hazard ratio greater
than one and a p-value significant at p<1% in model 2. This finding shows that companies with a
higher percentage of green revenues are more likely to sign a decarbonisation commitment with a
subsidiary. This finding challenges the legitimacy of the parent company’s commitment, as a
parent company is generally expected to sign a decarbonisation commitment before extending it
to a subsidiary.
The firm characteristics show that size and profit are significant at <1% in models 1 through 4.
This confirms the expectation that more extensive and profitable firms are likelier to set a
commitment to decarbonisation. Further, consistently significant beta_ftse at p<1% in models 1-4
indicate that firms with more market volatility are compelled to commit to decarbonisation due to
stakeholder expectations.
In summary, these results suggest that for the dependent variable commitments, there is a more
substantial effect of institutional owners’ pressure and employees’ reliance on companies to set
decarbonisation commitments, supporting hypotheses H2 and H3. Moreover, we find that firms are
more likely to sign a commitment at the subsidiary level instead of the parent level, the higher the
customer pressure, in line with hypothesis H4.
6
The interaction coefficients of this study (CompanyGRprobnotsignwithparent, SGApercentprobnotsignwithparent,
IOprobnotsignwithparent) are the dependent variables of this study (CompanyGR, IO, SGApercent) multiplied by the likelihood
of signing with a subsidiary entity (Signwithsubsidiary). The coefficients explain the influence of the dependent variables in models
2-4 (Tables 3-5).
19
6. Sensitivity Analysis
For the sensitivity analysis, we group all the dependent variables (see section 4.2 Dependent
Variables) of this study except the main variable commitments as related climate action variables.
The climate action variables group includes the following dependent variables: emission inventory,
finance actions, mitigations actions and impact. The rationale is that the UNFCCC portal provides
per-company information on commitments and additional climate-related information on emission
inventory, finance actions, mitigations actions and impact (UNFCCC, 2023).
The first part of the sensitivity analysis uses Weibull regressions on related climate action variables
to test whether the stakeholder effects persist, providing a different perspective on robustness. The
second part of the sensitivity analysis uses Cox regressions on all dependent variables of this study
(commitments, emissions inventory, finance actions, impact) to assess the robustness of
stakeholder effects.
6.1. Related Climate Action Variables
To test the association between stakeholder pressure and the dependent variables: emission
inventory, finance actions, mitigations actions and impact (see section 4.2 Dependent Variables),
we run a Weibull regression and interpret models 1-4 as shown in Table 4. Panels A-D. We focus
on the effect of the independent variables (see section 4.3 Independent Variables): CompanyGR,
IO, SGApercent, CompanyGRSignwithsubsidiary (interaction coefficient) on the related climate
action variables.
Emission Inventory
The variable IO is significant in all models, with hazard ratios greater than one and p-values
significant at p<1%, indicating that companies with higher institutional ownership are more likely
to maintain emission inventory. The variable SGApercent demonstrates significance as well with
hazard ratios greater than one and significant p-values (p<1% in models 2 and 3 and p<5% in
models 1 and 4), suggesting that corporate reliance on employees with CSR concerns increases
the likelihood to maintain an emission inventory. Furthermore, the interaction coefficient
CompanyGRSignwithsubsidiary is significant in model 2 with hazard ratio greater than one and a
p-value significant at p<5%. This result indicates that companies with a higher percentage of green
20
revenues are abnormally likely to maintain an emission inventory with a subsidiary rather than
with a parent entity, raising questions about the legitimacy of the parent’s firms to maintain an
emission inventory.
[Insert Table 4. Panel A]
Finance Actions
The variable CompanyGR is significant in all models, with hazard ratios greater than one and p-
values significant at p<5%, suggesting that companies have a higher percentage of green revenues
and are more likely to take finance actions. Further, the variable IO is significant in all models,
with hazard ratios greater than one and p-values significant at p<1%, suggesting that the companies
with a higher percentage of institutional owners are more likely to take related finance actions.
Additionally, SGApercent has hazard ratios greater than one and significant p-values (p<5% in
model 2 and p<10% in models 3 and 4), implying that companies that rely employees with CSR
concerns are more likely to take finance actions. The interaction coefficient is significant in model
2, with a hazard ratio greater than one and a p-value significant at p<1%. This result suggests that
companies with a higher percentage of green revenues are abnormally more likely to finance
actions with a subsidiary than with a parent entity. This finding highlights the preference for
financial actions at the subsidiary level, challenging the legitimacy of the parent company to
undertake finance actions.
[Insert Table 4. Panel B]
Mitigations Actions
The variable CompanyGR has hazard ratios greater than one and significant p-values (p<10% in
models 3 and 4), indicating that companies with a higher percentage of green revenues are more
likely to take mitigation actions. The variable IO is significant in all models, with hazard ratios
greater than one and p-values significant at p<1%, verifying the effect of institutional ownership
in undertaking mitigations actions.
21
SGApercent is significant across all, with hazard ratios greater than one and p-values significant
at p<1%, highlighting the influence of employees with CSR concerns in undertaking mitigation
actions. Notably, CompanyGRSignwithsubsidiary is significant in model 2, with a hazard ratio
greater than one and p-value significant at p<5%, repeating the pattern where subsidiaries rather
than parent companies, disproportionately implement mitigations actions. This result challenges
legitimacy expectations of parent entity leadership in undertaking mitigation actions.
[Insert Table 4. Panel C]
Impact
The variable IO is significant in all models, with hazard ratios greater than one and p-values
significant at p<1%, suggesting that companies with a higher percentage of institutional owners
are more likely to make a positive impact by reducing scope 1&2 GHG emissions. SGApercent is
similarly significant in all models, implying that companies that rely more on employees with CSR
concerns are more likely to make a positive impact by reducing scope 1&2 GHG emissions.
CompanyGRSignwithsubsidiary is significant in model 2, with a hazard ratio greater than one and
p-value significant at p<1%, suggesting that companies with a higher percentage of green revenues
are abnormally more likely to make a positive impact by reducing scope 1&2 GHG with a
subsidiary than with a parent entity. Once more, this presents an ambiguous finding as a parent
company is generally expected to have the capacity to reduce its scope 1&2 GHG emissions, thus
contributing to a positive impact prior to a subsidiary. As a consequence, signing with a subsidiary
before the parent entity challenges the legitimacy of a firm contributing with a positive impact.
[Insert Table 4. Panel D]
Ultimately, to assess the robustness of these results, we also run Weibull regressions for further
dependent variables: initiative participations and actions undertaken. We find consistent results
with the related climate actions variables: emission inventory, finance actions, mitigations actions,
impact. IO and SGApercent are statistically significant in all dependent variables, explaining the
effect of institutional ownership and employees with CSR concerns in undertaking related climate
22
actions, initiative participations and actions undertaken. Finally, the interaction coefficient
CompanyGRSignwithsubsidiary is statistically significant in both related climate action variables
and further dependent variables, confirming a recurring pattern where companies with a higher
percentage of green revenues are abnormally more likely to undertake related climate actions,
initiative participations and actions undertaken with a subsidiary than with a parent entity.
6.2 Cox Regressions
We use a semi-parametric Cox model (see section 4.5 Survival Analysis Model) as a robustness
test, selecting this study's key variables and firm characteristics as specified in Model 1, as shown
in Table 5. We then use related climate action variables to test whether the stakeholder effects still
hold in model 1.
In line with our base analysis for the Weibull regressions, we find that IO is significant for all
related climate action variables, with hazard ratios greater than one and p-values significant at
<1%. We also find that the variable SGApercent is significant for all related climate action
variables and particularly significant at <1% for the related climate action variables mitigation
actions and impact in model 1. Additionally, CompanyGR is significant at <5% for the related
climate action variables, finance actions and mitigations actions. Thus, the Cox results show
robustness for the main independent variables of this study (CompanyGR, IO and SGApercent),
indicating that a higher proportion of CompanyGR, IO, SGApercent is positively associated with
the likelihood of signing a decarbonisation commitment.
In detail, the variable IO is significant at <1% for all related climate action variables in model 1,
indicating that companies with a higher percentage of institutional owners are more likely to set
decarbonisation commitments and undertake related climate actions due to institutional ownership
pressure. This positive relationship also exists between the variables that measure commitments
and actions related to climate change in models 1-4 (Tables 3–4). This proves that institutional
ownership affects decarbonization commitments and actions related to climate change.
Further, the variable SGApercent is significant at <1% for the related climate action variables
mitigation actions and impact in model 1. This finding indicates that companies with educated
employee reliance are likelier to set decarbonisation commitments and undertake related climate
actions. This result also aligns with our foundational analysis for the related climate action
23
variables mitigations actions and impact in models 1-4 (Tables 4), thus verifying the effect of
employees with CSR concern on decarbonisation commitments and related climate actions.
Finally, the CompanyGR is significant at <5% for the related climate action variables, finance
actions and mitigations actions in model 1. This result implies that companies with a higher
percentage of green revenues are more likely to set decarbonisation commitments and undertake
related climate actions. This finding is partially substantiated in our base analysis for the related
climate action variable finance actions in models 1-4 (Table 4), explaining the effect of customer
pressure on decarbonisation commitments and related climate actions.
All control variables are significant in model 1 for all the related climate action variables. The
variables size and profit are significant at 1% for all related climate action variables in Model 1.
This indicates that larger and more profitable companies are likelier to make decarbonisation
commitments and undertake related climate actions; this finding also aligns with our base analysis
for commitments and related climate action variables. Further, we derive that extensive and more
profitable firms with higher leverage, volatility, and a larger Tobinq are more likely to make
decarbonisation commitments and undertake related climate actions.
6.3 Logit Regression
We employ logit regression, selecting this study's key variables, firm characteristics, and fixed
effects, as shown in Table 6. Here, we use Signwithsubsidiary as the dependent variable and the
independent variables, which are the same as in model 1, plus industry fixed effects, to test whether
the stakeholder effects still hold.
CompanyGR is significant at <5% (p-value = 0.034), suggesting that companies with a high green
revenue percentage are more likely to sign a commitment or undertake a related action with a
subsidiary than with a parent entity. This result also aligns with our base analysis for commitments
and related climate action variables in models 1-4 (Tables 3-4). IO is significant at <10% with a
negative coefficient, implying that companies with higher institutional owners percentage are less
likely to sign a commitment or undertake a related climate action with a subsidiary than with a
parent entity. However, the variable is significant, confirming investor effect in signing with a
subsidiary entity also observed in the commitments and related climate action variables in models
1-4 (Tables 3-4). This finding does not necessarily support our foundational analysis with IO
24
consistently significant for commitments and related climate action variables in models 1-4
(Tables 3-4).
The control variables size, profit, tobinq, volatility, and beta_ftse are significant at<10%. In detail,
the variables size and profit have negative coefficients, indicating that smaller and less profitable
firms are less likely to sign a commitment or undertake a related climate action with a subsidiary
than with a parent entity. This outcome contradicts our base analysis, wherein larger and more
profitable companies set commitments and undertake related climate actions in models 1-4 (Tables
3-4). In conclusion, smaller and less profitable firms with lower volatility and beta_ftse and
reduced tobinq are less likely to sign commitments or undertake related climate actions. Though
the control coefficients are negative and significant, indicating a control variable effect in signing
with a subsidiary entity was also observed in the commitments and related climate action variables
in models 1-4 (Tables 3-4).
25
7. Discussion and Conclusion
With our study, we provide the first empirical evidence linking link specific stakeholder groups
(clients, employees, and institutional owners) to decarbonisation commitments and related climate
actions. We analyse determinants of these decarbonisation commitments by combining data from
various data providers (UNFCCC, FTSE, Refinitiv) and by employing Weibull regressions as well
as a range of sensitivity analyses to understand the role of related climate actions and to assess the
robustness of our results against other empirical approaches.
Our study offers several contributions to the existing literature. Notably, it identifies that
institutional ownership and employees with CSR concerns are substantial factors in prompting a
company to set decarbonisation commitments and undertake related climate actions, with
institutional ownership demonstrating a stronger influence. Building on Cohen, Kadach, and
Ormazabal’s (2023) findings that institutional ownership enhances corporate climate risk
disclosure, our study extends this consensus by revealing its fundamental role in shaping corporate
decisions to set decarbonization commitments and undertake related climate actions.
Relevant research explains that employee engagement in corporate strategic objectives often aligns
with sustainability reporting internally, while externally, this alignment a firm's reputation via its
commitment to environmental and social performance (Kurruppu and Milne, 2010). Our findings
add to the literature by demonstrating that companies relying on employees with CSR concerns
are more likely to set decarbonisation commitments and related climate actions.
We address a gap in the literature by analysing whether companies set decarbonisation
commitments or undertake related climate actions at the parent or subsidiary level in response to
customer pressure, employees with CSR concerns, and institutional owners. Thus, we find that
companies with higher percentages of green revenues (CompanyGRSignwithsubsidiary) are more
likely to set decarbonisation commitments or undertake related climate actions (emission
inventory, finance actions, mitigation actions, and impact) with a subsidiary rather than a parent
entity.
For instance, a company with high green revenues is likelier to maintain an emission inventory
(emission inventory) and reduce scope 1&2 GHG emissions (impact) through a subsidiary entity
than a parent entity. Subsidiary-level commitment may account for limited reduction in scope 1
and 2 GHG emissions, as a parent company is typically expected to maintain an emission inventory
26
and have greater capacity to reduce scope 1&2 GHG emissions on a broader scale than a subsidiary
entity.
Therefore, the recurring pattern of a company setting a commitment or undertaking a related
climate action at the subsidiary level rather than at the parent level challenges the legitimacy of
the parent entity. Further, we assume a possible explanation: customer pressure is most directly
experienced at the subsidiary level. Thus, it explains why a subsidiary may set a decarbonisation
commitment or undertake a related climate action without a parent-level initiative.
This study has two main implications for policy, practice, and research. The study highlights the
significance of stakeholder pressures in influencing environmental governance by identifying
institutional ownership and employees with CSR concerns as key motivators of corporate
decarbonisation commitments and related climate actions. Second, the observed pattern of
subsidiary-level commitments responding to customer pressures raises issues over the legitimacy
and coherence of parent-level environmental strategies, highlighting the necessity for regulatory
frameworks that assure alignment among corporate entities.
Although our findings are robust, they merit future research. That further explores the parent-
subsidiary level of commitment. Further, future research may study the type of institutional
ownership that drives decarbonisation commitments and/or related climate actions. Additionally,
subsequent studies could investigate how employees with CSR ethics stimulate decarbonisation
commitments and/or related climate actions.
Finally, it is important to note the following: While our results provide robust evidence of
stakeholder effect in shaping a firm's decisions to sign a decarbonisation commitment and to
engage in related climate actions to work towards reaching net zero, many companies fail to set
such commitments.
27
References
Adomako, S., & Tran, M. D. (2022). Stakeholder management, CSR commitment, corporate social performance: The
moderating role of uncertainty in CSR regulation. Corporate Social Responsibility and Environmental Management,
29(5), 1414-1423.
Anser, M. K., Yousaf, Z., Majid, A., & Yasir, M. (2020). Does corporate social responsibility commitment and
participation predict environmental and social performance?. Corporate Social Responsibility and Environmental
Management, 27(6), 2578-2587.
Apple. (2022). Environmental progress report 2022. Apple Inc. Available at:
https://www.apple.com/environment/pdf/Apple_Environmental_Progress_Report_2022.pdf
Bassen, A., Kordsachia, O., Lopatta, K., & Tan, W. (2022). Revenue alignment with the EU taxonomy regulation.
Available at SSRN 4100617.
Bassen, A., Shu, H., & Tan, W. (2023). Green revenues and stock returns: Cross-market evidence. Finance Research
Letters, 52, 103550.
Ben‐Amar, W., Comyns, B., & Martinez, I. (2024). Firm commitments on climate change: Effects of science‐based
targets on financial outcomes during the COVID‐19 crisis. Business Strategy and the Environment.
Bolton, P., & Kacperczyk, M. (2021). Firm commitments. National Bureau of Economic Research.
Clark, G. L., Hebb, T., & Wojcik, D. (2007). Institutional investors and the language of finance: The global metrics
of market per formance. Globalisation of Accounting Standards, pp. 15–33.
Chan, K. J. D., Cheung, B., & Shen, L. Y. (2024). An Economic Foundation for Assessing the Credibility of Corporate
Net Zero Transition Pathways. Available at SSRN 4504017.
Cox, D. R. (1972). Regression models and life-tables (with discus sion). Journal of the Royal Statistical Society, pp.
34, 187–220.
Cohen, S., Kadach, I., & Ormazabal, G. (2023). Institutional investors, climate disclosure, and carbon emissions.
Journal of Accounting and Economics, 76(2-3), 101640.
de Freitas Netto, S. V., Sobral, M. F. F., Ribeiro, A. R. B., & Soares, G. R. D. L. (2020). Concepts and forms of
greenwashing: A systematic review. Environmental Sciences Europe, 32, 1-12.
Donaldson, T., & Preston, L. E. (1995). The stakeholder theory of the corporation: Concepts, evidence and
implications. Academy of Management Review, 20(1), 65–91.
Dyreng, S. D., Hoopes, J. L., Langetieg, P., & Wilde, J. H. (2020). Strategic subsidiary disclosure. Journal of
Accounting Research, 58(3), 643-692.
Freeman, R.E. (1984). Strategic management: A stakeholder approach, Cambridge, UK: Cambridge University Press.
Freiberg, D., Grewal, J., & Serafeim, G. (2021). Science-based carbon emissions targets.
George, B., Seals, S. and Aban, I., (2014). Survival analysis and regression models. Journal of nuclear cardiology,
21(4), 686–694.
Flammer, C., Toffel, M. W., & Viswanathan, K. (2021). Shareholder activism and firms' voluntary disclosure of
climate change risks. Strategic Management Journal, 42(10), 1850-1879.
Gallego-Álvarez, I., & Ortas, E. (2017). Corporate environmental sustainability reporting in the context of national
cultures: A quantile regression approach. International Business Review, 26(2), 337-353.
28
Guenther, E., Guenther, T., Schiemann, F. and Weber, G. (2016). Stakeholder relevance for reporting: explanatory
factors of carbon disclosure. Business & Society, 55(3), pp.361-397.
Jiang, X., Kim, S., & Lu, S. (2023). Accountability of Corporate Emissions Reduction Targets. Available at SSRN
4676649.
Hoepner, A.G., Majoch, A.A. and Zhou, X.Y. (2021). Does an asset owner’s institutional setting influence its decision
to sign the principles for responsible investment?. Journal of Business Ethics, 168(2), pp.389-414.
Hosmer, D.W., & Lemeshow, S. (2008). Applied survival analysis: regression modeling of time to event
data.European Orthodontic Society, pp. 681, 561–562.
Hummel, K., & Jobst, D. (2024). An overview of corporate sustainability reporting legislation in the European Union.
Accounting in Europe, 1-36.
Ilhan, E., Krueger, P., Sautner, Z., & Starks, L. T. (2023). Climate risk disclosure and institutional investors. The
Review of Financial Studies, 36(7), 2617-2650.
Ioannou, I., S. Xin Li and G. Serafeim. (2016). The Effect of Target Difficulty on Target Completion: The Case of
Reducing Carbon Emissions. The Accounting Review 91 (5): 1467-1492.
Kucharska, W., & Kowalczyk, R. (2019). How to achieve sustainability?—Employee's point of view on company's
culture and CSR practice. Corporate Social Responsibility and Environmental Management, 26(2), 453-467.
Klausmann, J., Krueger, P., & Matos, P. (2024). The Green Transition: Evidence from Corporate Green Revenues.
Available at SSRN 4850449.
Mahajan, R., Lim, W. M., Sareen, M., Kumar, S., & Panwar, R. (2023). Stakeholder theory. Journal of Business
Research, 166. Available at: https://doi.org/10.1016/j.jbusres.2023.114104
Net Zero Traker. (2023). Net Zero Stockstake 2023. NewClimate Institute, Oxford Net Zero, Energy and Cllimate
Intelligence Unit and Data-Driven Envirolab. Available at: htpps://zerotracker.net/analysis/net-zero-stocktake-2023.
New Climate Institute (2023). Corporate Climate Responsibility Monitor 2023. Cologne: New Climate Institute.
Published on 13 February 2023. Available at: https://newclimate.org/resources/publications/corproate-climate-
responsibility-monitor-2023.
Science Based Targets Iniative (SBTi). (2023). SBTi Monitoring Report 2022. SBTi. Available at:
https://sciencebasedtargets.org/reports/sbti-monitoring-report-2022.
Kuruppu, S., & Milne, M. J. (2010, March). Dolphin deaths, organizational legitimacy and potential employees’
reactions to assured environmental disclosures. In Accounting Forum (Vol. 34, No. 1, pp. 1-19). No longer published
by Elsevier.
Laufer, W. S. (2003). ‘Social Accountability and Corporate Greenwashing’, Journal of Business Ethics, 43(3), pp.
253–261. doi: 10.1023/A:1022962719299.
The Official Microsoft Blog. 2020. Microsoft will be carbon negative by 2030 - The Official Microsoft Blog. [online]
Available at: <https://blogs.microsoft.com/blog/2020/01/16/microsoft-will-be-carbon-negative-by-
2030/#:~:text=By%202 030%20Microsoft%20will%20be,goal%20but%20a%20detailed%20plan.> [Accessed 5 July
2022].
Lemma, T. T., Lulseged, A., & Tavakolifar, M. (2021). Corporate commitment to climate change action, carbon risk
exposure, and a firm's debt financing policy. Business Strategy and the Environment, 30(8), 3919-3936.
29
LSEG. (2024) FTSE Russell Green Revenues. https://www.lseg.com/en/data-analytics/financial-data/company-
data/ftse-russell-green-revenues [Accessed 5 September 2024]
LSEG. (2024) Ownership and profiles data. https://www.lseg.com/en/data-analytics/financial-data/company-

data/ownership-and-profiles
UN (2022). Integrity Matters: net zero commitments by businesses, financial institutions, cities and regions. UN.
Available at: https://www.un.org/sites/un2.un.org/files/high-level_expert_group_n7b.pdf.
UNFCCC (2023). Yearbook of global Climate Action. Marrakesh Partnership for Global Climate Action.
UNFCCC (2023d). Fashion Industry Charter for Climate Action Progress Report 2023. Bonn: UNFCCC. Available
at: hhtps://unfccc.int/documents/627503 .
UNFCCC (2024). Parties & observers. United Nations Climate Change. Available at
https://unfccc.int/parties-observers
Unilever (2024). Business Responsibility and Sustainability Report. Available at:

https://www.hul.co.in/sustainability/sustainability-reporting-centre/sustainability-report/.
Peng, C.-W., & Yang, M.-L. (2014). The effect of corporate social performance on financial performance: The
moderating effect of ownership concentration. Journal of Business Ethics, 123, 171-182.
Smith, B. (2020b). Microsoft will be carbon negative by 2030.

https://blogs.microsoft.com/blog/2020/01/16/microsoft-will-be-carbon-negative-by-2030/ (Accessed: 15 November
2023).
Tao, R., Wu, J., & Zhao, H. (2023). Do corporate customers prefer socially responsible suppliers? An instrumental
stakeholder theory perspective. Journal of Business Ethics, 185(3), 689-712.
Waheed, A., & Zhang, Q. (2022). Effect of CSR and Ethical Practices Sustainable Competitive Performance: A Case
of Emerging Markets from Stakeholder Theory Perspective. Journal of Business Ethics, 175(4), 837–855. Available
at: https://doi.org/10.1007/s10551-020-04679-y.
Testa, F., Boiral, O., & Iraldo, F. (2018). Internalization of environmental practices and institutional complexity: Can
stakeholders pressures encourage greenwashing?. Journal of Business Ethics, 147, 287-307.
30
Tables
Table 1: UNFCC Dependent Variables Company reporting
Companies UNFCCC Dependent Variables
Commitments Emission inventory Finance actions Mitigation actions Impact
Apple Inc., United ‘Reduce CO2e ‘Scope 1 and 2 ‘USD 1.5 billion of Incomplete reporting ‘Latest emissions
States of America emissions from emissions 2022’ and green bonds issued in (2022) compared to
operations and selected ‘Scope 3 emissions 2016’ and ‘USD 1 previous inventory
upstream and 2022’ billion of green bonds (2021), increased’
downstream activities issued in 2017’
by 61% from 2019 to
2030.’
Delta Air Lines, United ‘Achieve net zero ‘Scope 1 and 2 ‘carbon price ‘Low-carbon energy ‘-17.8% CO2 reduction
States of America carbon emissions by emissions 2022’ and established’ consumption: liquid in 2022 from base year
2050 at the latest.’ ‘Scope 3 emissions biofuels’ 2019’
2022’
The Walt Disney ‘Reduce operational ‘Scope 1 and 2 ‘carbon price ‘Low-carbon energy ‘Changes in GHG
Company, United CO2e emissions by emissions 2022’ and established’ consumption: User emissions-scope 1 and
States of America 46% from 2019 to ‘Scope 3 emissions specified action type 2 combined’ and ‘-12%
2030.’ 2022’ reported’ CO2 reduction in 2022
from the base year
2019’
Table 1 presents company reporting on commitments and related climate action variables (commitments, emission inventory, finance
actions, mitigations actions, and impact). The companies selected are also included in the study sample.
31
Table 2: Summary of variables
Variables Definition Data Source
Dependent variables
Commitments (by year) The year the organization commits to a https://climateaction.unfccc.int/Actors

decarbonization commitment with detailed or non-
detailed reporting that refers to scope 1,2 & 3 scope
emission reduction percentage with target year.
Emission Inventory (by year) The organization maintains an emission inventory https://climateaction.unfccc.int/Actors
denoted as total GHG emissions or scope
breakdown emissions into Scope 1&2 and possibly
Scope 3 with verification.
Finance Actions (by year) The organization accounts for financial actions. For https://climateaction.unfccc.int/Actors
instance, carbon price reporting (CDP 2023 Climate
Change Disclosure) and/or green bonds issuance
reporting (Climate Bonds Initiative).
Mitigation actions (by year) The organization lists categories of actions to reduce https://climateaction.unfccc.int/Actors
GHG emissions as reported in the 2023 CDP-ICLEI
questionnaire as per company.
Impact (by year) The organization displays changes in greenhouse https://climateaction.unfccc.int/Actors

gas emissions (Scope 1 and 2 combined). Positive
impacts include reductions in greenhouse gas
emissions (e.g., Latest emissions (2021) compared
to base year emissions (2019)/Latest emissions
(2021) compared to previous inventory (2020).
Initiative Participations (by year) The organization accounts for initiative https://climateaction.unfccc.int/Actors
participations that aim at long-term decarbonisation
practices (e.g. Science Based Targets initiative
(SBTi), Climate Ambition Alliance, Race to Zero,
Climate Action 100+).
Actions Undertaken (by year) Actions undertaken embed mitigation actions https://climateaction.unfccc.int/Actors
providing the cause per sector and action (as a
solution) as reported in the 2023 CDP-ICLEI
questionnaire per company and finance actions.
Independent Variables of Interest
CompanyGR Company green revenue percentage. FTSE
IO Institutional owners percentage. Refinitiv
SGApercent The proportion of a company’s revenue allocated to Refinitiv

its selling, general, and administrative expenses
(SGA) plus research development (R&D) costs
indicating how much of the revenue is spent on
employees working in sales, marketing, business
administration, research or development.
Signwithsubsidiary The organization has not provided evidence that it GLEIF

signed the commitment with the ultimate parent's
legal entity, as it either signed with a subsidiary
entity or withheld its entity legal form (ELF) code
in signing.
Control variables
Profit A ratio that expresses a company’s profitability FTSE

relative to its total financial structure; by dividing
gross profit by the sum of shareholders’ equity and
total dept.
32
Leverage A ratio calculated as sharelaholder’s eequity relative FTSE
to total dept indicates that a higher value reflects a
greater dependence on enquiry financing than dept.
Beta_ftse A firm’s stock volatility relative to the overall Refinitiv

market, is benchmarked against FTSE.
Size A company’s financial size is expressed as the FTSE

logarithm of the sum of its shareholders’ equity and
total debt, that is a scaled measure of the company’s
capital structure in euros.
Volatility Company volatility shows the statistical variation in Refivitiv

a company’s stock price or returns over time, that is,
the level of risk uncertainty relative to market
valuation.
Tobin Q A ratio that compares the sum of its market FTSE

capitalisation and total dept (market values of the
company’s assets) to the sum of its shareholders’
equity and total dept (the book values of its assets),
indicating whether the firm’s market valuation
exceeds or falls below its recorded asset value.
Table 2 illustrates the summary of proxy variables to capture the accounting, financial and sustainability characteristics of the corporate setting of
the firms in our sample. Column (1) shows the variable’s name, column (2) explains the variable’s definition and column (3) reports the data used.
33
Table 3: Stakeholder effects on long-term commitments to corporate emissions reduction.
y=commitments (by year)
Model 1 Model 2 Model 3 Model 4
Key Variables
CompanyGR 1.00193 1.003004 1.002311 1.002122

0.420 0.241 0.338 0.380
IO 1.039385 1.027607 1.039195 1.039637

0.000*** 0.000*** 0.000*** 0.000***
SGApercent 2.517458 2.884543 2.623249 2.549555

0.002** 0.001*** 0.002** 0.002**
Interactions
CompanyGRSignwithsubsidiary 1.060712
0.001***
SGApercentSignwithsubsidiary 0.0012134
0.072
0.0049874
IOSignwithsubsidiary 0.078
Signwithsubsidiary 4.260806 28.11199 49.14842

0.000*** 0.000*** 0.000***
Firm Characteristics
Leverage 1.026787 1.056824 1.04994 1.052412

0.631 0.344 0.374 0.352
Size 1.469686 1.370658 1.46578 1.466236

0.000*** 0.000*** 0.000*** 0.000***
Profit 1.655473 1.567212 1.685682 1.681595

0.000*** 0.000*** 0.000*** 0.000***
Tobinq 0.8076713 0.8859142 0.9128102 0.9305868

0.509 0.712 0.780 0.826
Vol 0.1949796 0.110164 0.1817507 0.1845915

0.000 0.000 0.000 0.000
Beta_ftse 1.418709 8.580185 1.42559 1.422498

0.000*** 0.000*** 0.000*** 0.000***
N 5.484 4,383 5.402 5.402
Log Likelihood 2273.8001 2253.0715 2262.7942 2262.9706
Table 3 reports the Weibull regression results for stakeholder effects on the long-term commitments to corporate emissions
reduction. The dependent variable is commitments (by year) on a yearly basis from 2018 to 2021. The independent variables include
CompanyGR, IO, SGApercent, Signwithsubsidiary and firm characteristics. All independent variables are lagged by one year. Test
for coefficient significance 1%(***), 5%(**) and 10%(*) levels.
34
35
Table 4
PANEL A: Stakeholder effects on emission inventory to corporate emissions reduction.
y=emission inventory (by year)
Key Variables
CompanyGR 1.00245 0.9968048 1.002835 1.002654

0.299 0.321 0.233 0.265
IO 1.038956 1.01917 1.03877 1.039194

0.000*** 0.000*** 0.000*** 0.000***
SGApercent 2.585466 17.99824 2.695632 2.621334

0.002** 0.000*** 0.001*** 0.002**
Interactions
0.027**
0.070*
0.079 *

0.059* 0.000*** 0.000***
Controls YES YES YES YES
N 5.484 1.424 5.402 5.402
Log Likelihood 2286.8235 1165.329 2275.8008 2275.9173
36
Table 4
PANEL B: Stakeholder effects on finance actions to corporate emissions reduction.
y=finance actions (by year)
Key Variables
CompanyGR 1.006512 1.008308 1.006924 1.006668

0.030** 0.009** 0.023** 0.029**
IO 1.034655 1.019922 1.033627 1.034277

0.000*** 0.000*** 0.000*** 0.000***
SGApercent 1.983371 2.39854 2.17871 2.078424

0.114 0.042** 0.075* 0.095*
Interactions
0.001***
0.047**
0.0014074
IOSignwithsubsidiary 0.083*

0.000*** 0.000*** 0.000***
N 5.484 4,383 5.402 5,402
Log Likelihood 1129.1356 1163.1327 1139.5956 1139.3642
37
38
Table 4
PANEL C: Stakeholder effects on mitigations to corporate emissions reduction.
y=mitigation actions (by year)
Key Variables
CompanyGR 1.003941 1.0051 1.004362 1.004185

0.103 0.048** 0.074* 0.086*
IO 1.038826 1.026654 1.038531 1.038936

0.000*** 0.000*** 0.000*** 0.000***
SGApercent 2.946657 3.354362 3.035754 2.951495

0.001*** 0.000*** 0.001*** 0.001***
Interactions
0.002**
0.085*
0.0073335

0.000*** 0.000*** 0.000***
N 5.484 4,383 5.402 5.402
Log Likelihood 2016.9313 2012.9441 2010.4312
39
PANEL D: Stakeholder effects on impact to corporate emissions reduction.
y=impact (by year)
Key Variables
CompanyGR 1.002421 1.00358 1.002806 1.002626

0.306 0.157 0.239 0.271
IO 1.03897 1.027176 1.038791 1.039212

0.000*** 0.000*** 0.000*** 0.000***
SGApercent 2.595962 2.987763 2.707554 2.633185

0.001*** 0.000*** 0.001*** 0.001***
Interactions
0.001***
0.068*
0.0050558
IOSignwithsubsidiary 0.079*

0.001*** 0.000*** 0.000***
N 5.484 4,383 5.402 5.402
Log Likelihood 2282.7945 2262.2123 2271.9428 2272.0475
40
41
Table 5: Related climate actions variables estimated with Cox regression.
Model 1 Model 1 Model 1 Model 1 Model 1
y=commitments y=emission y=finance actions y=mitigations y=impact
inventory
Key variables
CompanyGR 1.003544 1.004054 1.0085 1.005674 1.00402

0.145 0.091* 0.005** 0.020** 0.095*
IO 1.035739 1.035307 1.029792 1.035106 1.035285

0.000*** 0.000*** 0.000*** 0.000*** 0.000***
SGApercent 2.338617 2.388649 2.115103 2.717107 2.39596

0.004** 0.003** 0.078* 0.001*** 0.003***
Leverage 0.883007 0.8819227 0.8224475 0.8528175 0.8818307

0.036** 0.034** 0.020** 0.014** 0.034**
Size 1.523773 1.521287 1.852216 1.520118 1.520409

0.000*** 0.000*** 0.000*** 0.000*** 0.000***
Profit 1.857299 1.859085 2.007336 1.886493 1.863349

0.000*** 0.000*** 0.001*** 0.000*** 0.000***
Tobinq 0.3472788 0.3502558 0.16466 0.2874322 0.3522397

0.001*** 0.001*** 0.000*** 0.000*** 0.001***
Vol 0.5552502 0.5477281 0.5433026 0.5258056 0.5416501

0.000*** 0.000*** 0.000*** 0.000*** 0.000***
Beta_ftse 1.194324 1.194603 1.234719 1.217974 1.194646

0.070* 0.069* 0.074* 0.044** 0.069*
N 5,484 5,484 5,484 5,484 5,484
Log Likelihood -2879.1263 -2893.8729 -1521.1016 -2573.7157 -2885.8353
Table 10 reports Cox regression results for related climate action variables estimated with model 1 of this study. The dependent
variables are the related climate action variables of this study (y=commitments, y=emission inventory,y=finance
actions,y=mitigations, y=impact) on a yearly basis from 2018 to 2021. The independent variables include CompanyGR, IO,
SGApercent, Signwithsubsidiary and firm characteristics. All independent variables are lagged by one year. Test for coefficient
significance 1%(***), 5%(**) and 10%(*) levels.
42
Table 6: Logit Regression
Model 5
y=Signwithsubsidiary
Key variables
CompanyGR 0.0954838
0.034**
IO -0.1885432
0.077*
SGApercent 4.942083
0.460
Leverage 0.5099395
0.654
Size -0.5663555
0.287
Profit -9.114982
0.083*
Tobinq -1.154599
0.789
Vol 2.476437
0.038*
Beta_ftse -7.018366
0.054*
Fixed Effects
SicCode1 0.0009395
0.070*
SicCode2 0.0003625
0.310
SicCode3 0.0006691
0.166
SicCode4 -0.0006126
0.131
SicCode5 -0.0011138
0.044**
SicCode6 0.0008282
0.162
SicCode7 -0.0005614
0.177
SicCode8 -0.0005389
0.134
N 503
Log Likelihood -17.273383
Table 11 reports Logit regression results for model 5 of this study with industry fixed effects.The dependent variable is
y=Signwithsubsidiary on a yearly basis from 2018 to 2021. The independent variables include CompanyGR, IO, SGApercent,
Signwithsubsidiary, firm characteristics and fixed effects. All independent variables are lagged by one year. Test for coefficient
significance 1%(***), 5%(**) and 10%(*) levels.
43

ssrn-5067558

Uploaded by

Copyright:

Available Formats

ssrn-5067558

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ssrn-5067558

Uploaded by

Copyright:

Available Formats

Clients, employees and institutional owners:

Who influences corporate decarbonisation commitments?

Abstract: This study examines the determinants of corporate decarbonization commitments,

Keywords: Decarbonisation Commitments, Green Revenues, Employees with CSR concerns,

2.1 Institutional Setting

2.2 Stakeholder Theory

3.1 Clients value alignment stakeholder theory hypothesis

4.2 Dependent Variables

4.3 Independent Variables of Interest

4.4 Control Variables

4.5 Survival Analysis Model

5.1 Descriptive Statistics

(1) (2) (3) (4) (5) (6)

Commitments 5402 0.7744 0.2624 0 0 1

Emission Inventory 5402 0.0747 0.2630 0 0 1

Finance Actions 5402 0.0405 0.1972 0 0 1

Mitigations Actions 5402 0.0668 0.2497 0 0 1

Impact 5402 0.0746 0.2627 0 0 1

CompanyGR 5402 6.5703 17.7550 0 0 100

IO 5402 17.4884 13.2610 0.0304 14.7667 100

SGApercent 5402 0.2111 0.1634 0.0003 0.1690 0.9905

Signwithsubsidiary 5402 0.0031 0.0560 0 0 1

Leverage 5402 1.5512 1.4947 0 1.0759 13.6078

Profit 5402 0.2772 0.2791 -0.0871 0.2061 5.2033

Beta_ftse 5402 0.5834 0.3553 0.0011 0.5177 9.1432

Size 5402 22.4683 1.2895 15.3362 22.4016 27.2923

Vol 5402 2.2040 0.7806 0.4525 2.0991 7.5195

Tobin Q 5402 0.3639 0.2574 0 0.3409 1.0002

5.2 Results of Survival Analysis

6.1. Related Climate Action Variables

[Insert Table 4. Panel A]

[Insert Table 4. Panel B]

[Insert Table 4. Panel C]

[Insert Table 4. Panel D]

6.2 Cox Regressions

6.3 Logit Regression

LSEG. (2024) Ownership and profiles data. https://www.lseg.com/en/data-analytics/financial-data/company-

Unilever (2024). Business Responsibility and Sustainability Report. Available at:

Smith, B. (2020b). Microsoft will be carbon negative by 2030.

Commitments Emission inventory Finance actions Mitigation actions Impact

Commitments (by year) The year the organization commits to a https://climateaction.unfccc.int/Actors

Impact (by year) The organization displays changes in greenhouse https://climateaction.unfccc.int/Actors

Independent Variables of Interest

CompanyGR Company green revenue percentage. FTSE

IO Institutional owners percentage. Refinitiv

SGApercent The proportion of a company’s revenue allocated to Refinitiv

Signwithsubsidiary The organization has not provided evidence that it GLEIF

Profit A ratio that expresses a company’s profitability FTSE

Beta_ftse A firm’s stock volatility relative to the overall Refinitiv

Size A company’s financial size is expressed as the FTSE

Volatility Company volatility shows the statistical variation in Refivitiv

Tobin Q A ratio that compares the sum of its market FTSE

Model 1 Model 2 Model 3 Model 4

CompanyGR 1.00193 1.003004 1.002311 1.002122

IO 1.039385 1.027607 1.039195 1.039637

SGApercent 2.517458 2.884543 2.623249 2.549555

Signwithsubsidiary 4.260806 28.11199 49.14842

Leverage 1.026787 1.056824 1.04994 1.052412

Size 1.469686 1.370658 1.46578 1.466236

Profit 1.655473 1.567212 1.685682 1.681595